2013alonsobten PDF

Universidad de Oviedo
Departamento de Ingeniería Química y

Tecnología del Medio Ambiente
Obtención de ácidos orgánicos por

fermentación de subproductos lácteos
TESIS DOCTORAL
POR
Saúl Alonso Tuero
Febrero, 2013
AGRADECIMIENTOS
El largo y tortuoso camino parece que llega su fin pero alcanzar tal meta habría sido
imposible sin la inestimable e infinita ayuda de varias personas.
Quiero y debo agradecer en primer lugar al Prof. Mario Díaz y al Dr. Manuel Rendueles,
bajo cuya dirección he realizado el presente trabajo, por su infinita ayuda, orientación,
confianza y apoyo. Han contribuido decisivamente en la culminación de este trabajo, les
agradezco infinitamente el tiempo y el esfuerzo que me han dedicado.
Al Prof. Mario Díaz le agradezco asimismo la libertad científica de la que he gozado en

determinados momentos. Con mayor o menor acierto (quizás más real lo segundo) he hecho
y deshecho a mi libre albedrío, sin ataduras en relación a experimentos y demás
planteamientos. Creo que en mi caso ha sido un acicate real para mejorar día a día e intentar
corregir los errores en la manera de lo posible.
Tengo muy claro de que si aspiro a ser doctor es gracias al Dr. Manuel Rendueles, quien
ha sabido guiarme en el trabajo diario. Le agradezco además sus ánimos en los momentos
más difíciles, que han existido y que creo que fueron muchos. ¡Muchas gracias Manolo!. Aún
recuerdo el día cuando me dijiste que comenzaría a realizar ciencia de verdad!!. Confío al
menos en haberlo intentado.
Quisiera expresar mi enorme gratitud a las Dras. Mónica Herrero y Covadonga Quirós por
haberme enseñado e introducido una técnica desconocida para mí como la citometría de
flujo. Les agradezco enormemente la ayuda y sus magníficos consejos.
Quisiera asimismo agradecer al Prof. Luis A. García por sus valiosos consejos y su
cercanía. También por su inusitado entusiasmo por la ciencia en unos tiempos en los que
creer en el propio método científico se hace difícil. Al Prof. Luis A. García y al Dr. Antonio G.
Lavín les agradezco el haberme prestado la tarjeta para hacer fotocopias de artículos de la
biblia de Elmer Gaden y Douglas S. Clark llamada Biotechnology and Bioengineering.
Al Dr. Benjamín Paredes por su sapiencia, su curiosidad y su constante optimismo.
Igualmente quiero expresar mi gratitud a la Dra. Isabel González (Reny Picot, ILAS S.A.), a
quien no conozco personalmente pero que me ha proporcionado la materia prima con la
cual he realizado los experimentos de la presente memoria.
Llegado este punto tengo que agradecer a Ana Salas y Magdalena Choda (Servicio de
Citometría de los SCT’s, Universidad de Oviedo) por su amabilidad, su buen hacer, su
eficiencia y por la inmensa ayuda prestada en el análisis de las muestras mediante
citometría. Ciertamente, ha sido un verdadero placer contar con la ayuda de verdaderas
profesionales. Igualmente tengo que agradecer a la Dra. Marta Alonso (Área de Procesos de
Imágenes de los SCT’s, Universidad de Oviedo) por su amabilidad e inmensa ayuda en el
visionado de las muestras a través del microscopio confocal. Gracias de todo corazón a Ana,
Magdalena y Marta. Realmente ha sido una satisfacción constatar que no se ha cumplido el
principio de Peter con estas auténticas profesionales.
A todos los compañeros de laboratorio del Grupo TBR (Sergio, Carlos, Este, Paula,
Amanda, Irene, Rosana, Silvia, Fede, Lorea, Isma, Renata, Ainoa, ...) por su ayuda siempre
desinteresada y por ser responsables del agradable ambiente reinante. Agradecer al resto de
compañeros del Dpto. como Patri, Emilio, Rubén, Pepe, etc… que convirtieron el momento
del café en una ventana agradable de escape frente a los problemas del laboratorio.
Quiero también particularizar el agradecimiento en varias personas. En primer lugar

agradecer a la Dra. Lorea Legazpi, de la que intenté mimetizar al menos un poquito de su
meticulosidad científica cuando en su momento tomé contacto por primera vez con el grupo
TBR. Creo que he tenido el privilegio de coincidir con una persona con un talento científico
desbordante, la Dra. Estefanía Noriega, ¡ha sido un verdadero placer tomar nota de tus
consejos, Este!. Al Dr. Federico Gularte (un verdadero gentleman) por sus sabios y acertados
consejos. A Irene Rosas le debo multitud de agradables momentos pasados en el laboratorio
y porque siempre nos ha acompañado con buen humor y positivismo dentro del labo. ¡Ha
sido un verdadero placer trabajar contigo, Ire!.
Agradecer también a la Dra. Amanda Laca, una trabajadora infatigable y una frustrada
ecóloga afortunadamente creo yo. Al Dr. Carlos Álvarez por siempre estar dispuesto a echar
una mano, ayudar siempre que lo necesité, ¡muchísimas gracias Carlos, vales mucho chaval!.
Ha sido un privilegio trabajar con tan alto nivel de implicación y con ese compañerismo.
Quiero agradecer muy especialmente al clúster de oxidación húmeda. Del Dr. (o mejor
dicho autónomo) Sergio Collado poco se puede decir que no se sepa, un verdadero
profesional como la copa de un pino. ¡Vales millones de euros Jorge!. Del otro miembro de
pleno derecho, Paula (gran trabajadora y mejor persona), comentar que ha fascinado a toda
la sala con sus especialidades reposteras (¡qué gran repostera hemos perdido!, aunque a
decir verdad hemos ganado en una excelente científica-profesional). Agradecer asimismo a
Kike (futuro catedrático de Ingeniería Química o en su defecto Chemical Engineering) por su
alegría, motivación y por las conversaciones sobre su Barça. Ciertamente, agradezco
infinitamente el que siempre estuvieran ahí para ayudarme en todo momento.
Por último y no menos importante, quiero agradecer el apoyo constante y la ayuda de mis
padres y de mi hermano. Han sabido comprenderme y aguantarme en los malos tiempos,
que ciertamente han sido bastantes. Han tenido una infinita paciencia conmigo porque la
verdad es que cuando las cosas no salen en el laboratorio, uno se desespera y a veces se
convierte en inaguantable. A mi hermano le agradezco que me haya echado una mano en el
tramo final por la gran portada que me ha fabricado. Han sido un apoyo moral fundamental
para culminar este trabajo.
En definitiva, quiero agradecer a todo el mundo que haya estado directamente o

indirectamente implicado en la culminación de este viaje de casi cinco años. Espero no
haberme olvidado de nadie, si fuese el caso espero que me comprenda porque llegado este
punto, el que suscribe esta memoria ya está oxidado mentalmente y no fluyen correctamente
los recuerdos. Asimismo, confío en no haber tirado a la basura todo este tiempo, percepción
que en algún momento he tenido.
Índice I
ÍNDICE
RESUMEN............................................................................................................................................................................III
ABSTRACT .......................................................................................................................................................................... V
1. INTRODUCCIÓN ...........................................................................................................................................................3
1.1. Introducción ............................................................................................................................................................3
1.2. Objetivos ...................................................................................................................................................................7
1.3. Estructura de la memoria ..................................................................................................................................9
2. CONSIDERACIONES TEÓRICAS ......................................................................................................................... 13
2.1. Status actual de la producción biotecnológica de metabolitos microbianos ............................... 13
2.2. Producción biotecnológica de ácido láctico: status actual y aplicaciones .................................... 18
2.3. Producción biotecnológica de ácido lactobiónico: status actual y aplicaciones ......................... 23
3. MATERIALES Y MÉTODOS ................................................................................................................................... 81
3.1. Producción biotecnológica de ácido láctico ............................................................................................. 81
3.1.1. Microorganismo de cultivo ................................................................................................................... 81
3.1.2. Preparación del lactosuero de yogurt .............................................................................................. 81
3.1.3. Condiciones de cultivo ............................................................................................................................ 82
3.1.4. Métodos analíticos .................................................................................................................................... 82
3.1.5. Monitorización mediante citometría de flujo ................................................................................ 83
3.2. Producción biotecnológica de ácido lactobiónico .................................................................................. 86
3.2.1. Microorganismo de cultivo ................................................................................................................... 86
3.2.2. Preparación del suero lácteo ................................................................................................................ 86
3.2.3. Condiciones experimentales de cultivo ........................................................................................... 86
3.2.4. Métodos analíticos .................................................................................................................................... 87
3.2.5. Monitorización mediante citometría de flujo ................................................................................ 88
3.2.6. Separación celular .................................................................................................................................... 89

II Índice
3.2.7. Microscopía confocal de fluorescencia ............................................................................................. 90
4. RESULTADOS.............................................................................................................................................................. 93
4.1. Producción de ácido láctico por Lactobacillus casei ............................................................................. 93
4.2. Monitorización fisiológica de Lactobacillus casei ............................................................................... 103
4.3. Producción de ácido lactobiónico por Pseudomonas taetrolens.................................................... 121
4.4. Influencia de la disponibilidad de oxígeno disuelto sobre la producción de ácido lactobiónico

................................................................................................................................................................................. 131
4.5. Influencia de la heterogeneidad fisiológica sobre la producción de ácido lactobiónico ....... 141
4.6. Influencia del pH en la producción de ácido lactobiónico ................................................................ 155
4.7. Estrategias de alimentación para la mejora en la producción de ácido lactobiónico ........... 169
5. DISCUSIÓN GENERAL .......................................................................................................................................... 195
5.1. Producción biotecnológica de ácido láctico .......................................................................................... 195
5.2. Producción biotecnológica de ácido lactobiónico ............................................................................... 203
6. CONCLUSIONES ...................................................................................................................................................... 217
7. BIBLIOGRAFÍA ........................................................................................................................................................ 223
8. ANEXOS ...................................................................................................................................................................... 233
8.1. Difusión de la tesis doctoral ......................................................................................................................... 233
8.1.1. Artículos científicos ............................................................................................................................... 233
8.1.2. Comunicaciones a congresos ............................................................................................................. 234
8.2. Informe sobre el índice de impacto de los artículos de la tesis ....................................................... 235
Resumen III
RESUMEN
El creciente interés comercial experimentado en los últimos años por los ácidos
orgánicos ha estimulado la producción biotecnológica de los mismos como una solución
integral a la paulatina demanda suscitada en este segmento químico. De igual modo, la
sostenibilidad ha sido otro de los retos afrontados por la biotecnología industrial, la cual se
ha centrado tanto en el desarrollo de estrategias eficientes, como en el aprovechamiento
biotecnológico de residuos o subproductos industriales. En este contexto, el presente
trabajo ha abordado la producción biotecnológica de dos ácidos orgánicos con importante
relevancia comercial como son el ácido láctico y lactobiónico a partir de dos residuos
alimentarios, el suero lácteo de yogurt y de queso.
En primer lugar se ha estudiado la producción de ácido láctico por Lactobacillus casei a
partir de los azúcares fermentables presentes en el suero lácteo de yogures retirados de la
cadena de comercialización. Así, se ha evaluado la capacidad fermentativa en términos de
rendimiento, evolución y eficiencia de un sistema fermentativo caracterizado por la
presencia de glucosa, sacarosa y lactosa como substratos. Con el objetivo de profundizar en
el comportamiento fermentativo, se ha empleado asimismo la citometría de flujo
multiparamétrica como herramienta para la evaluación de la funcionalidad celular durante
el bioproceso. De hecho, la aplicación de esta técnica de cultivo independiente ha permitido
conocer el status fisiológico y la robustez celular del sistema a nivel poblacional.
Asimismo, se ha estudiado la bio-producción de ácido lactobiónico por Pseudomonas
taetrolens a partir de suero lácteo bajo diferentes condiciones operacionales. El ácido
lactobiónico ha irrumpido recientemente en el mercado con una pléyade de aplicaciones en
ámbitos tan diversos como el alimentario y el biomédico debido a sus excelentes
propiedades. Hasta la fecha su producción industrial está restringida a la catálisis vía
química, resultando en una metodología ineficiente e insostenible debido al uso de costosos
catalizadores y a la formación indeseable de subproductos laterales durante la reacción. La
producción biotecnológica puede resolver estas limitaciones así como posicionarse como la
alternativa productiva más adecuada ante el incremento en la demanda del ácido
lactobiónico. El bioproceso desarrollado en el presente trabajo ha sido llevado a cabo en
diferentes escalas con el fin de localizar y poner en evidencia las etapas limitantes,
optimizando el sistema en base a consideraciones operacionales lo más adecuadas tanto
para la fisiología como para el rendimiento productivo de P. taetrolens. Así, la
monitorización de la heterogeneidad fisiológica del sistema, a través de la citometría de flujo
IV Resumen
multiparamétrica, ha proporcionado el conocimiento preciso sobre el comportamiento

funcional de P. taetrolens asociado a la bio-producción de ácido lactobiónico. En el presente
trabajo se ha avanzado, por lo tanto, en el conocimiento de cómo diferentes factores, tales
como el pH operacional, la disponibilidad de oxígeno disuelto o la edad fisiológica, influyen
decisivamente sobre el rendimiento, productividad y eficiencia del sistema productivo,
constituyendo globalmente una valiosa información para una posible transferencia
industrial del sistema aquí desarrollado.
Abstract V
ABSTRACT
The recent growing commercial interest in the field of organic acids has stimulated their
biotechnological production as the solution to the increasing demand generated on this
chemical segment. Likewise, sustainability has been one of the challenges facing the
biotechnology industry, which has been focused on both the development of efficient
strategies and the biotechnological use of industrial wastes or byproducts. Within this
context, this thesis has addressed the biotechnological production of two commercial
organic acids such as lactic and lactobionic acid from two food wastes, cheese whey and
yoghurt whey.
Firstly, it was studied the lactic acid production by Lactobacillus casei from fermentable
sugars, which are available in the yoghurt whey removed from yoghurt derivates out of shelf
life. Thus, the fermentative capacity of L. casei has been evaluated in terms of performance
and efficiency within a fermentation featured by the presence of glucose, sucrose and
lactose as substrates. In order to gain insight into the fermentative behaviour,
multiparameter flow cytometry has been used as tool for the evaluation of cell functionality
during the bioprocess. In fact, the application of this technique has enabled to know the
cellular physiological status and system robustness at the population level.
Likewise, the bio-production lactobionic acid from whey by Pseudomonas taetrolens was
studied under different bioprocessing conditions. Lactobionic acid has recently emerged
into the market with a plethora of applications in areas such as food and biomedicine due to
its excellent properties. To date, its industrial production is restricted to chemical catalysis,
resulting in an inefficient and unsustainable methodology owing to costly catalysts as well
as the formation of side-reaction products. The biotechnological production may overcome
these drawbacks as the most appropriate alternative in order to meet the increasing
demand of lactobionic acid. The bioprocess developed in the present study has been carried
out at different scales in order to decipher the rate-limiting steps, optimizing the system
based on appropriate operational considerations both for physiology and yield of P.
taetrolens. Monitorization of the physiological heterogeneity through multiparameter flow
cytometry has thus provided precise knowledge about the functional behaviour of P.
taetrolens during the lactobionic acid bio-production. The present study has consequently
added further knowledge on how different bioprocessing factors such as the operational pH,
dissolved oxygen availability or physiological age may play a decisive role in the
VI Abstract
fermentation performance, productivity and efficiency, representing valuable information

for upscaling this system to the industrial sphere.
1. INTRODUCCIÓN
Introducción 3
1. INTRODUCCIÓN
1.1. Introducción
Los últimos años se han caracterizado por un indudable cambio de modelo productivo en
la industria química, pasando de sistemas basados en fuentes no renovables hacia la
implementación de procesos sostenibles y eficientes con una fuerte base biotecnológica.
Dicho estímulo ha implicado el empleo de microorganismos como factoría celular para la
producción de metabolitos de alto valor añadido, sirviendo como plataforma para la
elaboración posterior de productos con indudable interés comercial. Así, el concepto de los
microorganismos como factoría celular ha sido ampliamente estudiado desde el nacimiento
de la microbiología industrial tras la segunda guerra mundial, siendo objeto de estudio tanto
la producción microbiana de antibióticos, proteínas recombinantes o ácidos grasos como
ejemplos más significativos.
La actual coyuntura caracterizada por las economías de escala y el imperante desarrollo
económico de los países emergentes han incrementado sustancialmente la cantidad de
residuos agroalimentarios generados a nivel mundial. Asimismo, la toma de conciencia
medioambiental por parte del sector industrial y la implantación de medidas reguladoras ha
espoleado la búsqueda de nuevas alternativas relativas al aprovechamiento de los residuos
y subproductos generados. El tratamiento de estos residuos constituye una actividad que
implica un importante coste económico y medioambiental para la industria. Así, se han
venido mostrando esfuerzos en reducir la acumulación y generación de subproductos y
residuos alimentarios. Sin embargo, el potencial para la reducción de los residuos es
limitado y costoso. Bajo estas premisas se ha experimentado un notable desarrollo de
nuevos procesos biotecnológicos donde se emplean estos residuos procedentes de
actividades industriales, y carentes de valor comercial alguno, como sustratos con el fin de
obtener productos de valor añadido. Estas nuevas bioconversiones microbianas no sólo
juegan un indudable papel vertebrador de la sostenibilidad ambiental, sino que pueden ser
utilizadas también con el fin de aportar nuevas y atractivas sustancias químicas emergentes
con interés comercial.
Uno de estos subproductos industriales lo constituye el suero lácteo. Se trata de un
subproducto que tradicionalmente se ha venido empleando únicamente como fertilizante o
fuente de lactosa alimentaria. Múltiples han sido las aproximaciones realizadas con el fin de
lograr el aprovechamiento y utilización de un residuo industrial como el suero lácteo. Así,
4 Introducción
durante la última década, la utilización biotecnológica del suero lácteo ha surgido como una
posible opción a la valorización del mismo, centrándose en la producción de diversos
metabolitos microbianos como los ácidos carboxílicos, exopolisacáridos o más
recientemente los polihidroxialcanoatos.
El procesamiento de leche en forma de yogures y productos líquidos fermentados ha
seguido una tendencia claramente alcista en los últimos 10 años principalmente debido a la
importante demanda experimentada en los mercados internacionales asociados a países
emergentes. En concreto, el importante incremento del consumo asociado a la región
asiática ha influido decisivamente en el incremento en la producción de más de 33 millones
de toneladas en 2007 con respecto a 1997 (IDF, 2011; IDFA, 2008). Anualmente, el consumo
de yogures y derivados se distingue por un apreciable incremento en todas las áreas
comerciales, de hecho, el mayor aumento lo ha experimentado China, donde en 2007 la
producción se situó en 2,5 millones de toneladas. La aparición de nuevos formatos de
yogures que aportan beneficios extra, ha contribuido especialmente a que la tasa de
crecimiento anual de estos productos se sitúe de media en un 6,2%, representando el 11%
del mercado lácteo mundial (IDFA, 2008). En Europa, la elaboración de productos lácteos
fermentados y de yogures contribuye con el 1% de los residuos generados por la industria
alimentaria, mientras que el subsector lácteo fabricante de quesos en sus distintas variantes
llega a generar el 26% de los residuos totales.
Actualmente, el empleo directo del suero lácteo se restringe a su utilización como
complemento nutricional en piensos para animales, fertilizante o en último término se
gestiona como efluente residual (Figura 1.1). Es en éste último contexto donde residen sus
principales implicaciones medioambientales, debido a su notable potencial contaminante
(175 veces superior al de un agua residual común) por su alta demanda química de oxígeno
(DQO >60.000 ppm) y bioquímica de oxígeno (DBO >35.000 ppm), siendo considerado como
el residuo más contaminante dentro de la industria alimentaria. Sin embargo, este
subproducto contiene una importante cantidad de nutrientes que hacen atractivo su
aprovechamiento biotecnológico con el propósito de obtener nuevos productos o derivados
de lactosa que presentan un valor añadido superior y que también poseen nuevos nichos
comerciales. De hecho, en la última década se han desarrollado nuevas tecnologías
alrededor de técnicas cromatográficas, cristalización y membranas con el fin de aprovechar
los principales componentes del lactosuero. Sin embargo, el notable incremento en volumen
experimentado en el mercado tanto de lactosa de grado alimentario como de concentrados
proteicos ha provocado una caída en los precios de los mismos. Esta excesiva oferta ha
Introducción 5
implicado la búsqueda por parte de la industria de nuevos productos con un valor añadido
superior y con un potencial de mercado más amplio.
Figura 1.1. Utilización mundial del suero lácteo en 2006. De 177 millones de toneladas generadas durante 2006
sólo 124 (70% del suero lácteo generado) fueron utilizadas industrialmente como materia prima con el objeto
de obtener productos derivados del suero lácteo o bien de la lactosa (datos obtenidos a partir de Affertsholt,
2007).
Por otra parte, la industria alimentaria está intentando adoptar el concepto holístico de
producción alimentaria, integrando y actuando sobre todos los aspectos ambientales
relativos al desarrollo y fabricación de un producto alimentario. Por lo tanto, la adopción de
estos principios implica aportar un valor añadido en todos sus puntos a los residuos
generados a lo largo de la cadena alimentaria. De esta forma, los productos lácteos
fermentados no aptos para el consumo humano (incluyendo productos fuera de fecha de
caducidad, cambios de lineal o sobre-stocks) tienen que ser retirados del mercado,
representando un residuo para la industria láctea transformadora. De hecho, la industria
láctea estima que alrededor del 1% de la producción total de productos fermentados lácteos
tienen que ser tratados como residuos, con el consiguiente incremento en los costes de
operación. La naturaleza de dichos residuos les confiere una alta potencialidad
contaminante en base a su elevado contenido en azúcares añadidos. Además, se trata de
productos cuyo reciclado tiene que ser tratado directamente por la industria y cuya
posterior utilización se restringe únicamente al uso de la fracción sólida como pienso para
animales. La fracción líquida resultante sí que puede ser aprovechada como fuente de
carbohidratos fermentables en un proceso de fermentación ácido láctico. Los productos
lácteos fermentados, caso del yogurt, se caracterizan por presentar altos contenidos en
glucosa y sacarosa, azúcares que junto con la lactosa, constituyen una magnifica fuente de
6 Introducción
carbono y energía aprovechable por bacterias ácido-lácticas como L. casei. Ésta bacteria es
capaz de aprovechar estos carbohidratos mediante un proceso heteroláctico fermentativo
facultativo para dar lugar a ácido láctico. Por consiguiente, el uso de este residuo en
procesos fermentativos como sustrato reduce su potencialidad medioambiental, a la vez que
se enfatiza en el concepto sostenible dentro de la cadena alimentaria y se genera un
producto con un valor añadido superior.
La conversión oxidativa microbiana de lactosa presente en el suero lácteo puede
significar una alternativa atractiva para la producción del ácido lactobiónico, un derivado de
alto valor añadido que presenta importantes nichos de mercado. De hecho, este ácido
orgánico presenta propiedades únicas que le confieren un amplio espectro de aplicaciones
comerciales e industriales que van desde el ámbito cosmético al alimentario. La producción
biotecnológica del ácido lactobiónico se espera que sea, por lo tanto, una alternativa
industrial real que permita superar limitaciones como la formación de intermediarios o
subproductos asociados al proceso de producción, algo que por ejemplo ocurre en muchos
de los procesos oxidativos vía catálisis química.
Objetivos 7
1.2. Objetivos
El desarrollo de dos bioprocesos eficientes, efectivos, robustos y sostenibles para la

obtención vía microbiana de ácido láctico y lactobiónico se ha convertido en el principal
objetivo del presente trabajo. La presente memoria pretende aportar información relevante
que contribuya a la implantación de la vía biotecnológica como la metodología de referencia
para la producción industrial de ácido lactobiónico. En ambos bioprocesos se ha realizado
un particular énfasis en la importancia de la fisiología microbiana a nivel poblacional como
elemento clave en el desarrollo y optimización de los sistemas sometidos a estudio.
Específicamente, se han planteado los siguientes objetivos:
 Evaluar la capacidad de la fracción líquida residual procedente de productos lácteos

fermentados como fuente alternativa para la producción de ácido láctico.
 Conocer el comportamiento fermentativo de L. casei en una matriz natural y

compleja basada en una mezcla de azúcares fermentables, estableciendo las
condiciones operacionales más adecuadas.
 Caracterizar y monitorizar los diferentes estados fisiológicos de L. casei durante la

producción fermentativa de ácido láctico.
 Desarrollar un sistema biotecnológico eficiente para la obtención de ácido

lactobiónico por parte de P. taetrolens, estableciendo por lo tanto los parámetros
operacionales más adecuados para maximizar la producción del mismo a partir de
suero lácteo.
 Desarrollar protocolos de citometría de flujo multiparametrica como técnica que

permita evaluar la funcionalidad de P. taetrolens.
 Conocer la imbricación del status fisiológico con la bioconversión oxidativa por parte
de P. taetrolens. Revelar asimismo la heterogeneidad fisiológica inherente al sistema
productivo en sus diferentes escalas de cultivo.
8 Objetivos
 Proponer las condiciones óptimas de operación a escala de biorreactor con el fin de

alcanzar niveles industrialmente competitivos de ácido lactobiónico en un sistema de
cultivo en discontinuo con alimentación.
Estructura de la memoria 9
1.3. Estructura de la memoria
La actual memoria se presenta como compendio de publicaciones que se enmarcan en la

producción biotecnológica de los dos ácidos orgánicos que son objeto de estudio. Cada una
de las publicaciones se ha elaborado con arreglo al esquema tradicional, esto es, dentro de
cada artículo se puede distinguir específicamente una introducción, la descripción de la
metodología empleada y una exposición crítica de los resultados obtenidos. Dichas
publicaciones han sido aceptadas o están siendo evaluadas por revistas incluidas en el
Science Citation Index (Thomson Reuters). La estructura de la presente memoria se
encuentra dividida por lo tanto en ocho capítulos subdivididos cada uno de ellos en sus
respectivos apartados.
En el Capítulo 1, correspondiente a la parte introductoria de la memoria, se exponen las
causas reales que han motivado la realización del presente trabajo. Igualmente se presentan
los objetivos que se han pretendido alcanzar. En el Capítulo 2 se describe el estado del arte
en el que se encuentran los dos ácidos orgánicos sometidos a estudio en la presente
memoria. En primer lugar, se introduce el contexto actual en el que se encuentra la
producción vía fermentativa de determinadas sustancias dentro de la industria química. A
continuación, se detalla someramente el estado en el que se encuentra la producción
biotecnológica, así como las aplicaciones de un ácido orgánico maduro y ampliamente
estudiado como es el ácido láctico. La tercera parte de las consideraciones teóricas se
corresponde con un artículo conformado por una exhaustiva y documentada revisión
bibliográfica sobre la bio-producción de ácido lactobiónico. En el mismo se puede encontrar
tanto desde los orígenes del ácido, sus potenciales aplicaciones comerciales así como el
status en el que se encuentra actualmente. En dicha revisión bibliográfica, el lector podrá
advertir cómo el ácido lactobiónico ha llegado a convertirse, por ejemplo, en una sustancia
emergente dentro de la industria cosmética debido a sus únicas propiedades terapéuticas.
En el Capítulo 3 se describen, de forma global, las diferentes metodologías
experimentales utilizadas en la consecución de los resultados. Aunque la metodología
específica de cada parámetro evaluado se detalla en su correspondiente artículo, en este
capítulo se ha pretendido hacer hincapié en las técnicas comunes que han sido empleadas en
la elaboración de las diferentes partes experimentales que componen la presente memoria.
Los subapartados del Capítulo 4 constituyen el bloque de resultados que conforma el eje
central de la presente memoria. Cada subapartado se corresponde con su correspondiente
artículo publicado o en vistas de serlo. Dichos trabajos son independientes entre sí, aunque
10 Estructura de la memoria
como se podrá comprobar fácilmente todos comparten un mismo elemento central. Se ha

optado por comenzar con el aprovechamiento biotecnológico del suero de yogur
(subapartado 4.1) y con la monitorización fisiológica de los bioprocesos implementados
(subapartado 4.2). A continuación se exponen los restantes 5 artículos correspondientes a la
producción biotecnológica del ácido lactobiónico. En dichos artículos se ha estudiado la
influencia del tamaño de inóculo así como la descripción general del sistema a escala de
matraz Erlenmeyer y biorreactor (subapartado 4.3), la disponibilidad de oxígeno disuelto
(subapartado 4.4), la edad fisiológica (subapartado 4.5), el pH operacional (subapartado
4.6) y finalmente las estrategias de alimentación (subapartado 4.7).
En el Capítulo 5 se expone una discusión general sobre los aspectos más importantes
hallados en la presente memoria, resaltando la dimensión y la relevancia de los parámetros
operacionales que influyen sobre la producción microbiana de ambos ácidos orgánicos.
Finalmente, en el Capítulo 6 se han condensado las conclusiones más importantes que se
destilan del presente trabajo.
El Capítulo 7 está destinado a una bibliografía común a la memoria. Igualmente, se
recogen algunos de los trabajos que deben de ser de obligada referencia para cualquiera que
quiera conocer de primera mano las tendencias en microbiología industrial y abordar la
producción microbiana de metabolitos de alto valor añadido.
La última parte de la memoria, bajo el titulo de Anexos, recoge la difusión de la tesis
doctoral y el pertinente informe con el factor de impacto de las publicaciones.
2. CONSIDERACIONES TEÓRICAS
Consideraciones teóricas 13
2. CONSIDERACIONES TEÓRICAS
2.1. Status actual de la producción biotecnológica de metabolitos

microbianos
En los últimos años, la producción biotecnológica de determinados componentes

químicos se ha convertido en la alternativa viable, sostenible y efectiva ante el continuo
incremento en la demanda comercial mundial. Es indudable que la biotecnología ha jugado
un papel fundamental desde el descubrimiento de la penicilina en 1927 por Fleming,
momento que significó el nacimiento y posterior establecimiento de la fermentación
industrial como metodología productiva de referencia. No obstante, la producción industrial
de metabolitos microbianos por fermentación ha requerido el establecimiento de las bases
científicas necesarias para lograr incrementar la producción, así como la búsqueda de
nuevos organismos productores con características mejoradas (Demain, 2000, 2007). De
hecho, la biotecnología industrial se ha convertido en un potente y floreciente sector
económico con implicaciones directas en diversas industrias que van desde la automoción a
la farmacéutica (Erickson et al., 2012; Hatti-Kaul et al., 2007). En 2030, se espera que un
tercio de la producción mundial industrial esté relacionada con procesos biotecnológicos,
presentando un volumen de mercado de más de 300 mil millones de dólares (Koln Paper,
2007). Asimismo, la OCDE también señala a la biotecnología como la herramienta clave en el
desarrollo tanto de nuevas fuentes energéticas como de nuevos procesos industriales más
sostenibles (OECD, 2007). Como resultado de tales perspectivas, en Europa se viene
potenciando la denominada sociedad basada en el conocimiento biotecnológico, aglutinando
las tendencias del mercado, perspectivas y las visiones globales dentro de la industria. En
dicha alianza estratégica, que engloba varios programas marco, se ha fortalecido al sector
biotecnológico europeo, encaramando a Europa como el segundo mercado en producción
biotecnológica (EuropaBio y ESAB, 2006).
Los procesos biotecnológicos presentan la ventaja de ser sostenibles, reduciendo las
emisiones atmosféricas de gases efecto invernadero, así como también muestran una
reducida dependencia de los combustibles fósiles frente a los clásicos procesos químicos de
producción (Hermann et al., 2007). El desarrollo de la biología de sistemas, la biología
sintética o la ingeniería metabólica ha posibilitado asimismo la introducción de nuevas
funcionalidades en los microorganismos (Curran y Alper, 2012), incrementando el portfolio
14 Consideraciones teóricas
de bio-productos obtenidos mediante fermentación (Figura 2.1). Es indudable que la

biotecnología industrial ha sufrido un enorme impulso gracias al nacimiento de la ingeniería
metabólica hace ya más de 20 años por Jay Bailey (Bailey, 1991), una disciplina que ha
permitido alcanzar el fenotipo microbiano deseado a través de la modificación del
metabolismo celular (Curran y Alper, 2012; Yadav et al., 2012). Además, el desarrollo del
concepto de biorefinería, donde las diferentes corrientes generadas durante la utilización de
la biomasa son aprovechadas como materia prima, ha propiciado la irrupción de nuevos bio-
productos fermentativos que o bien han reemplazado o están en vías de hacerlo a
componentes químicos procedentes de combustibles fósiles.
Comercial
Estatinas (Merck) Ácido láctico (ADM; Cargill; BASF; Purac; Myriant) 1,3-Propanediol (DuPont/Tate
& Lyle; Metabolic Explorer;
PLA (Cargill; Purac; Reliance) Vitamina B12 (DSM; BASF; Sanofi-Aventis) Genencor)
Licopeno (DSM) Isosorbide (Roquette; ADM) Farneseno (Amyris) Ácido cítrico (Cargill; Roquette)
Omega-3 (DSM; Ocean; Lonza) Glutamato (Ajinomoto; Evonik) Bioetanol (ADM; Genencor; Abengoa)
Ácido succínico (Roquette; DSM; Myriant; BASF) Lisina (Ajinomoto; Evonik) Xylitol (Danisco; Roquette)
PHA (Metabolix; Meredian/DaniMer) Ácido glucónico (Roquette) Sorbitol (Global Sweeteners)
Fase planta piloto

3-Hidroxi propiónico (Cargill/Codexis; Microbia) Ácido itacónico (Microbia)
Ácido málico (Myriant) Coenzima Q10 (Ocean) Ácido levulínico (BioFine)
Butanol (DuPont/BP; Cathay; Gevo) Ácido adípico (Rennovia; Verdezyne)
1,4-Butanediol (Myriant) Isopreno (Genencor) Ácido acrílico (OPX Bio)
Ácido fumárico (Lonza; MBI; Merck) 1,2-Propanediol (GlycosBio)
Bajo desarrollo en laboratorio

Glucósidos (Evolva) Ácido glicólico (Roquette)
Ácido butírico (OXEA) α-santaleno (Firmenich)
Vanilina (Givaudan) Esfingolípidos (Evonik)
Metil-tetrahidrofurano (BioFine)
Ácido glucárico (Rivertop Renewables; Rennovia)
Figura 2.1. Portfolio de productos obtenidos por fermentación. Nótese que en la figura se obvian productos
ampliamente consolidados tales como las enzimas, antibióticos y proteínas recombinantes, productos todos
ellos obtenidos vía fermentativa a escala industrial.
En los últimos años han sido innumerables los lanzamientos al mercado de bio-productos
obtenidos a partir de procesos fermentativos (Figura 2.1). De hecho, esta proliferación de
bio-productos producidos vía fermentativa se ha correspondido con el desarrollo de un
robusto sector conformado por multinacionales que han diversificado su portfolio de
productos con el fin de sustituir muchos de los componentes químicos derivados del
petróleo. De hecho, en el mercado se encuentran productos afianzados comercialmente
como la lisina, el glutamato o el bioetanol. En el sector farmacológico son varios los bio-
fármacos producidos por fermentación, caso de las estatinas producidas por Merck o la
vitamina B12 por DSM (producida también con fines tecnológicos en el ámbito alimentario).
Reseñar que muchos de los bioprocesos ya implantados a nivel industrial se basan en el
empleo de materias primas renovables con bajo coste como residuos celulósicos o
subproductos de diferentes industrias. Ejemplo en este sentido lo constituye la producción
de farneseno por la norteamericana Amyris a partir de melazas de caña de azúcar o la
misma obtención de polihidroxialcanoatos por Metabolix a partir de los mismos residuos.
En fase de desarrollo a nivel de planta piloto se encuentran bio-productos tan atractivos
como el butanol (resaltando la joint-venture conformada entre DuPont y BP para su
desarrollo a nivel industrial), el ácido itacónico (en desarrollo por Microbia) o el isopreno
(en desarrollo por Genencor). Igualmente, en incipiente estado de desarrollo se encuentran
bio-productos tales como el ácido fumárico (en desarrollo por Lonza o Merck) o los
glucósidos (caso de la producción microbiana de stevia, que se encuentra en fase de
desarrollo por la compañía biotecnológica suiza Evolva) (Figura 2.1). Se estima que el
mercado mundial de productos de fermentación generará en 2013 un volumen de negocio
de 22000 millones de dólares dentro de un mercado altamente segmentado (Figura 2.2).
Particularizando dentro del sector químico, los ácidos orgánicos representan un
segmento floreciente y con un importante crecimiento económico (Demain, 2007; Sauer et
al., 2008). Se estima que el segmento químico conformado por los ácidos orgánicos
producidos por fermentación va a experimentar un importate incremento en su volumen de
negocio, pasando de los 2600 millones de dólares en 2009 a más de 4000 millones de
dólares en 2013 (IEA, 2012). De hecho, se trata de un segmento dinámico conformado por
ácidos maduros caso del cítrico, el láctico o el acético (reconocidos hoy en día casi como
commodities) (Papagianni, 2011; Sauer et al., 2008) y por ácidos emergentes con un alto
valor añadido que presentan nuevas y significativas aplicaciones gracias a sus singulares
propiedades. Este último grupo está representado por ácidos como el glucárico, glicérico,
fumárico, succínico, itacónico o el α-ceto glutárico, todos ellos en vista de ser producidos vía
fermentativa a nivel industrial (salvo el caso del succínico donde su producción

biotecnológica ya está implementada por Reverdia).
2000000
Producción mundial Ácido cítrico
1500000 Glutamato (MSG)

(t/año)
L-lisina
1000000
Ácido láctico
Vitamina C
Ácido glucónico Goma Xantano
500000 Biopolímeros
Ácido
L-hidroxialanina
lactobiónico
0
0 5 10 15
Precio unitario (€/Kg)
Figura 2.2. Mercado mundial de los principales bio-productos obtenidos por fermentación. El tamaño de la
burbuja es proporcional al valor de mercado o volumen de negocio generado. En la gráfica se han excluido el
bioetanol (con una producción mundial de 50 millones de t/año) y los antibióticos (35.000 t/año). Nótese que
la presente figura ha sido elaborada en base a datos encontrados en diferentes fuentes bibliográficas (Demain,
2000, 2007; Sauer et al., 2008; Soetaert y Vandamme, 2010) que necesariamente no tienen que ser los valores
reales, el lector tiene que poner en valor que la opacidad del mercado comercial sigue siendo todavía
imperante. Los valores estimados correspondientes al ácido lactobiónico son datos referidos a su producción
química puesto que las vías fermentativas o enzimáticas no han sido implementadas a nivel industrial.
La importancia de los ácidos orgánicos dentro del sector químico queda fuera de toda
duda con el papel que se presupone que van a jugar dentro del futuro contexto químico
industrial, cuyo ejemplo más clarividente en este sentido es la inclusión de hasta 9 ácidos
orgánicos (succínico, fumárico, málico, 3-hidroxi propiónico, glucárico, aspártico, glutámico,
itacónico y levulínico) dentro de los 15 futuros compuestos químicos con mayor potencial
(Werpy y Petersen, 2004). El papel de los ácidos orgánicos emergentes dentro de la
industria química se ha puesto asimismo de manifiesto con la transferencia e
implementación industrial de nuevas plataformas microbianas. Así, por ejemplo, la francesa
Roquette comercializa actualmente el ácido succínico producido por una Escherichia coli
recombinante desarrollada en los laboratorios de los profesores George Bennett y Ka-Yiu
San (Rice University, Houston, USA). De hecho, el sector industrial viene implementando
tales avances como se ha puesto de manifiesto con la puesta en marcha de la primera planta
industrial de producción fermentativa de ácido succínico por parte de Reverdia (joint
venture entre DSM y Roquette), o la planta de producción de ácido succínico que Succinity
(joint venture entre BASF y CSM) pondrá en marcha a finales de 2013 en Barcelona con una
capacidad de producción de 10.000 t/año.
Recientemente, un derivado del ácido succínico (en concreto el di-metil-succinato) ha
sido objeto de acuerdo entre la belga Proviron y Reverdia, propiciando el desarrollo e
implementación industrial de una nueva plataforma microbiana con objeto de producir
biotecnológicamente di-metil-succinato que pueda reemplazar a los tradicionales polímeros
derivados del petróleo. El año 2012 también ha sido clave en el nacimiento de alianzas
empresariales con el fin de desarrollar nuevos sistemas fermentativos para producir ácidos
orgánicos que hasta el momento sólo se producían vía química, tal es el caso de la alianza
entre Novozymes, Cargill y BASF con el fin de producir ácido acrílico. Estos movimientos en
este sector tan dinámico enfatizan el significativo papel que juegan actualmente los ácidos
orgánicos dentro de la industria química.
2.2. Producción biotecnológica de ácido láctico: status actual

y aplicaciones
El ácido láctico (ácido 2-hidroxipropanoico) representa a un ácido orgánico con un

estado de desarrollo comercial maduro y cuya producción biotecnológica fue implementada
a nivel industrial a principios de los años 90 del pasado siglo. El láctico es un ácido
carboxílico que presenta un amplio espectro de aplicaciones y que hoy en día se obtiene
industrialmente a través de procesos fermentativos llevados a cabo por bacterias ácido-
lácticas tales como Lactobacillus casei, L. helveticus o Lactococcus lactis. Se trata del quinto
producto fermentativo en volumen de producción (Figura 2.3), con una producción anual de
250.000 toneladas (Sauer et al., 2008). Actualmente es producido por compañías como
Cargill o Purac a partir de fermentación de almidón de maíz.
Figura 2.3. Ranking mundial de productos obtenidos por fermentación. Datos tomados de Soetaert y
Vandamme (2010).
El ácido láctico ha venido siendo empleado como conservante alimentario durante siglos,
sin embargo, no fue hasta el descubrimiento de la fermentación por Pasteur en 1857 cuando
se asoció que dicha actividad era provocada por microorganismos. En 1878, Joseph Lister
aisló el primer cultivo puro de bacterias ácido-lácticas productoras de ácido láctico. Se trata
de microorganismos microaerófilos Gram positivos que producen y secretan ácido láctico al
medio. Además de las bacterias ácido-lácticas, el hongo Rhizopus es conocido como
productor de ácido láctico. De hecho, en ocasiones es preferido como plataforma productora
puesto que a diferencia de las bacterias ácido-lácticas, el hongo no requiere suplementación
con nutrientes como el extracto de levadura o el licor de maíz fermentado. Además, R. oryzae
sintetiza el isómero L(+) del ácido láctico, mientras la mayoría de las bacterias ácido-lácticas
producen una mezcla racémica de ambos isómeros.
Las propiedades físico-químicas (véase Tabla 1) del ácido láctico le confieren una
importante relevancia en diferentes aplicaciones industriales, tal y como se muestra en la
Figura 2.4. Las aplicaciones de este ácido van desde el ámbito alimentario (empleado como
agente aromatizante y conservante), cosmético (agente hidratante) al farmacéutico. No
obstante, la producción de ácido láctico se espera que se vea significativamente
incrementada ante la irrupción del ácido poli-láctico (PLA) como nuevo polímero
biodegradable (Nampoothiri et al., 2010). Las perspectivas comerciales del PLA son
halagüeñas como demuestra la joint-venture entre Cargill y Dow formando NatureWorks
(propiedad actualmente en su totalidad de Cargill), la cual ya produce más de 150.000
toneladas anuales de éste biopolímero a partir de ácido láctico bajo la marca Ingeo. Otra
relevante aplicación del ácido láctico en la industria química se encuentra asociada a la
fabricación del solvente ecológico llamado etil-lactato.
Tabla 1. Propiedades y características del ácido láctico.
Propiedad Valores/características
Naturaleza Líquida
Fórmula molecular C3H6O3
Nombre sistemático Ácido 2-hidroxi-propanoico
Peso molecular 90,08
pKa 3,85
Punto de ebullición 122ºC
Solubilidad Alta
Precio unitario 1,55 €/kg
Producción mundial anual (t) 250.000
Producción biotecnológica anual (t) 250.000
Método de producción Fermentación
Principales fabricantes Purac, Cargill, Galactic

Excipiente en soluciones
parenterales
Prótesis
Preparaciones minerales
Soluciones
Suturas
Industria de diálisis
médicas
farmacéutica Regulador de pH
Agente hidratante
Agente humectante Conservante
Agente anti-acné Cosmética Ácido láctico Alimentación Aditivo
Agente queratinizante Aromatizante
Agente rejuvenecedor Acidulante
Industria química
Solvente Solventes
Ácido poliláctico
químico ecológicos
Agente
neutralizante Agente quelante
Figura 2.4. Aplicaciones y usos comerciales en diversos campos del ácido láctico y/o sus sales.
En los últimos tiempos, los esfuerzos relacionados con la producción biotecnológica del
ácido láctico se han centrado en tres ámbitos íntimamente interrelacionados, como son el
desarrollo de nuevas plataformas microbianas productoras, el empleo de nuevas materias
primas con bajo coste y la optimización del bioproceso (Figura 2.5). Estos dos últimos han
sido los aspectos desarrollados en la presente memoria, tal y como se mostrará en los
correspondientes subapartados (véanse subapartados 4.1 y 4.2).
Figura 2.5. Pasos claves en la producción biotecnológica de ácido láctico.

En la producción biotecnológica de ácido láctico, se ha mostrado como papel fundamental

la expresión de determinados genes involucrados en el metabolismo del carbono, el cual se
regula y se controla en función de la información procedente del estado metabólico celular
combinada con las señales externas que revelan la disponibilidad de determinados
carbohidratos. De hecho, las fuentes de carbono más habituales en las fermentaciones ácido-
lácticas lo representan sustratos con alta concentración en carbohidratos fermentables. De
esta forma, en los últimos años los esfuerzos se han centrado en la utilización de residuos
agro-industriales con altas concentraciones de carbohidratos en su composición. La
atención se ha centrado asimismo en la optimización de procesos fermentativos teniendo
como sustrato principal a materiales lignocelulósicos, los cuales pueden presentar altas
cantidades de carbohidratos. Sin embargo, la utilización de estos materiales tiene el
inconveniente de requerir pretratamientos previos a su utilización como sustrato,
precisando de una hidrólisis previa (bien sea química o enzimáticamente) y/o una
simultánea sacarificación de la celulosa y de la hemicelulosa.
Se estima que un 40-60% del coste total del proceso fermentativo lo representa el coste
del sustrato o materia prima empleada, motivo por el cual se ha estimulado la búsqueda de
nuevas materias primas renovables y con bajo coste que pudiesen ser utilizadas
industrialmente. Diferentes materias primas procedentes de subproductos de origen
alimentario han sido materia de estudio con el fin de ser sustrato fermentativo, tales como
almidón, residuos agrícolas procedentes del cultivo de la cebada o el maíz, así como el
propio salvado de arroz, la harina de trigo o el zumo de dátiles (Figura 2.6). Así, las últimas
investigaciones van encaminadas en la sacarificación (degradación de la biomasa compleja)
bien sea a través de hidrólisis ácida o enzimática y posterior fermentación (John et al.,
2009). Ambas etapas pueden ser llevadas a cabo a través de una hidrólisis y fermentación o
como sacarificación simultánea y posterior fermentación. De hecho, este proceso de
fermentación está íntimamente ligado a la producción de ácido láctico a partir de materias
con alto contenido en almidón (caso de la pulpa de mandioca, almidón de patata, salvado de
harina o almidón de maíz) y de materiales celulósicos (tales como la celulosa o residuos
madereros) utilizando microorganismos amilolíticos. No obstante, el empleo de esta
metodología encarece los procesos de obtención de ácido láctico al requerir la hidrólisis
previa de los sustratos. Recientemente, la fermentación de biomasa vegetal se ha centrado
en el empleo de microorganismos como L. amylovorus que son capaces de realizar
simultáneamente una primera sacarificación y posterior fermentación de los azúcares
resultantes (John et al., 2009).
En adecuación al empleo de nuevos sustratos, se han venido igualmente desarrollando

nuevas plataformas microbianas para la obtención de ácido láctico (Angermayr et al., 2012;
Zhou et al., 2003). Estos esfuerzos, se han centrado últimamente en dotar a determinados
microorganismos de nuevas funcionalidades, así como el desarrollo de cepas recombinantes
más robustas y sin tantos requerimientos nutricionales. De hecho, se han conseguido cepas
recombinantes de E. coli capaces de producir un nivel de ácido láctico cercano al
rendimiento teórico (Zhou et al., 2003).
Figura 2.6. Materias y residuos empleados como sustrato en la producción biotecnológica de ácido láctico.
2.3. Producción biotecnológica de ácido lactobiónico: status actual

y aplicaciones
En este subapartado relativo a las consideraciones teóricas se recogen la mayoría de los
hallazgos o hitos referentes al ácido lactobiónico, así como los hechos acaecidos durante su
más de un siglo de vida desde su descubrimiento por Fischer y Meyer en 1889. Se ha
realizado una exhaustiva documentación acerca de la historia y cronología del ácido
lactobiónico, enfatizando las actuales y potenciales aplicaciones de este ácido orgánico
emergente. En la revisión se reflejan todos los avances relacionados con el ácido
lactobiónico, que acompañados por una buena base documental permiten conocer las
tendencias asociadas al empleo del ácido lactobiónico dentro de la industria cosmética,
alimentaria o farmacológica. De hecho, se clasifica al ácido lactobiónico como una de las
moléculas emergentes en el campo de la ingeniería tisular, la nanofarmacología o la
biomedicina, circunstancia que no había sido advertida en ningún caso con anterioridad. Así,
se ha puesto en valor por primera vez la relevancia y el potencial del ácido lactobiónico
dentro de las disciplinas asociadas al ámbito biomédico, pasando desde la Solución Belzer al
empleo de la tecnología biónica en la industria cosmética por NeoStrata. El incremento de la
incidencia en el número de casos asociados al fallo hepático, que tras el cáncer es la 2ª causa
de mortalidad en los países occidentales, lleva a inferir el relevante papel que jugarán en un
futuro inmediato las terapias o metodologías que emplean al ácido lactobiónico como
molécula necesaria no sólo para la correcta liberación controlada del fármaco de interés
sino para la implantación exitosa de toda terapia celular hepática.
Se recoge igualmente el actual status del ácido dentro de la industria química o incluso su
papel como aditivo alimentario. En definitiva, se ha reflejado el significado del ácido dentro
de cada contexto comercial, enfatizando la importancia del ácido en las nuevas aplicaciones
comerciales. Además, en esta revisión bibliográfica se describe la producción vía
fermentativa y enzimática, subrayando por primera vez en la bibliografía las fortalezas y
debilidades de cada uno de las metodologías.
Publicación: Bio-production of lactobionic acid: current status, applications and future

prospects.
Situación: Enviada para su evaluación, bajo revisión.
Bio-production of lactobionic acid: current status, applications and

future prospects
Saúl Alonso, Manuel Rendueles, Mario Díaz *
Department of Chemical Engineering and Environmental Technology, University of Oviedo, Faculty

of Chemistry, C/ Julián Clavería s/n, 33071 Oviedo, Spain.
*Corresponding author. Tel.: +34 98 5103439; fax: +34 98 5103434.

E-mail address: [email protected]
Abstract
Lactobionic acid has appeared on the commercial scene as a versatile polyhydroxy acid with numerous
promising applications in the food, medicine, pharmaceutical, cosmetics and chemical industries. This
high value-added bio-product has recently received growing attention as a bioactive compound, providing
an excellent chemical platform for the synthesis of novel potentially biocompatible and biodegradable
drug delivery vehicles. Recent advances in tissue engineering and nanomedicine have also underlined the
increased importance of this organic acid as a key biofunctionalization agent. The growing commercial
relevance of lactobionic acid has therefore prompted the development of novel systems for its
biotechnological production that are both sustainable and efficient. The present review explores recent
advances and studies related to lactobionic acid bio-production, whether through microbial or enzymatic
approaches, highlighting the key bioprocessing conditions for enhanced bio-production. Detailed
overviews of the current microbial cell factories as well as downstream processing methodologies for
lactobionic acid production are also presented. Furthermore, the potential prospects and current
applications of this polyhydroxy acid are also discussed, with an emphasis on the role of lactobionic acid
as a key platform in the development of novel drugs, biomaterials, nanoparticles and biopolymer systems.
Keywords: lactobionic acid; microbial fermentation; organic acids; bioprocessing conditions;

bio-based production; drug delivery systems; galactosylated nanoparticles; galactosylated
scaffolds.
1. Introduction
The production of bulk organic acid chemicals by microbial fermentation has
undoubtedly undergone continuing growth over the last decade, progressively expanding its
market niche and portfolio (Jang et al., 2012; Sauer et al., 2008). In fact, bacteria as bio-
production platforms have become a reliable, cost-competitive, feasible alternative for
large-scale industrial production of many bulk and specialty organic acids (Demain, 2007).
Organic acids represent a growing chemical segment in which bio-based compounds such as
fumaric, glyceric, propionic, xylonic, itaconic, levulinic or α-ketoglutaric acids have also
emerged on the market as platform chemicals (Jang et al., 2012). This transition towards
bio-based industrial production has concomitantly involved the development of novel

sustainable bioprocesses focused on the use of cost-effective renewable resources, either
already implemented at an industrial level or still in the development pipeline (Willke and
Vorlop, 2004). Beyond traditional organic acids, the market has also shown substantial
interest in novel carboxylic acids like lactobionic acid due to its unique physicochemical
properties (Fig. 1). Lactobionic acid (LBA) is a high value-added lactose derivative (Gänzle et
al., 2008; Seki and Saito, 2012) which has recently emerged as a promising and versatile
substance with countless applications in the cosmetics (Green et al., 2009; Tasic-Kostov et
al., 2010; West, 2004a), pharmaceutical (Belzer et al., 1992) and food (Gerling, 1998;
Gutiérrez et al., 2012) industries. The recent market glut suffered by traditional lactose-
based products has additionally stimulated the dairy industry to seek new approaches for
lactose utilization which could overcome the traditional view of lactose as a commodity
(Affertsholt, 2007; Gänzle et al., 2008). As a result, novel lactose derivatives (such as lactitol,
lactulose and LBA) have recently come onto the commercial market with considerable
industrial applications (Playne and Crittenden, 2009; Seki and Saito, 2012).
Figure 1. Current trends in the organic acids market. The market is moving from commodity bulk carboxylic
acids towards value-added organic acids which display outstanding applications. LBA currently belongs to this
latter segment. The symbol * denotes an implemented industrial production by microbial fermentation.
In recent years, LBA has also received growing attention as a bioactive molecule since it
provides an excellent platform for the synthesis of biocompatible and biodegradable drug
delivery vehicles and biomaterials. In this respect, LBA will clearly play a major strategic
role in the treatment of hepatic disorders through nanomedicine, with a potential near-term
impact. Its prospect as a key biomolecule in the field of nanotechnology is thus of
outstanding significance. In view of this commercial relevance, both the development and
implementation of feasible LBA production systems emerge as crucial key challenges to
meet market demands. To date, LBA is manufactured by chemical synthesis in an energy-
intensive process which requires the use of costly metal catalysts (Kuusisto et al., 2008;
Mirescu and Prübe, 2007; Yang and Montgomery, 2005). However, this expensive
methodology may also involve the generation of undesirable side-reaction products (Chia et
al., 2008). Although this polyhydroxy acid has been available since the late 1940s (Stodola
and Lockwood, 1947; Stodola and Jackson, 1950), its production by biotechnological means
has not been developed so intensively up to now in comparison with other organic acids
such as lactic, succinic or citric acid (Papagianni, 2011). Nevertheless, bio-production of LBA
has emerged as both a promising and feasible approach to meet the growing demand for
this bio-product. Furthermore, environmentally-friendly and cost-effective LBA bio-
production can be accomplished by employing cheese whey as an inexpensive feedstock
(Alonso et al., 2011, 2012a). Despite being a traditional natural source for whey protein
isolate (Smithers, 2008) and lactose (Schaafsma, 2008), cheese whey upgrading and
treatment remain as two of the major challenges facing the dairy industry. Therefore, the
search for innovative solutions in the disposal and management of this high-strength waste
stream has become the driving force behind the development of novel sustainable
biotechnological processes (Guimarães et al., 2010).
Within this context, the present review explores recent advances in LBA bio-production,
either through enzymatic or microbial biosynthesis, as well as the current novel trends
addressing the application of LBA in the marketplace, with particular emphasis on those
emerging areas such as nanomedicine and tissue engineering. A detailed overview of
current microbial cell factories, further downstream processing methodologies for LBA
production and prospects are also provided.
2. Properties and current industrial status of lactobionic acid.

The structure and physicochemical properties of LBA confer on it a plethora of current
and potential commercial applications, as shown in Figure 2. This organic acid exhibits a
large number of newly discovered biological activities and great therapeutic potential due to
its excellent biocompatibility, biodegradability and nontoxicity, as well as its chelating,
amphiphilic and antioxidant properties. LBA belongs to the aldobionic family of acids, which
additionally comprises maltobionic and cellobionic acid (Pezzotti and Therisod, 2006). From
a nutritional point of view, this substance may be considered a low-calorie sweetener which
provides only 2 kcal/g (Schaafsma, 2008). Chemically, it comprises one galactose molecule
attached to one molecule of gluconic acid via an ether-like linkage (its molecular structure is
shown in Table 1). In particular, the presence of multifunctional groups (8 hydroxyl groups)
converts LBA molecule into a metal ion chelator which can sequester cations such as
calcium (Abbadi et al., 1999). This carboxylic acid presents a molecular weight of 358.3, with
a pKa of 3.6 (Armarego and Chai, 2009). Table 1 lists the main physicochemical properties of
LBA; its high water solubility being especially noteworthy. The calcium salt of LBA in fact
shows 40,000 times higher water solubility than calcium carbonate and 10 times that of
calcium lactate (Nakano et al., 2010).
Biomaterials
Nanoparticles
Anticoagulant Organ preservant
drugs
Contrast agent Drug carrier Gene carrier
Pharmaceutics Food additive
Moisturizer & Biomedicine
Cross-linker Gelling agent
Soluabgielinzitng agent
Peeling agent Cosmetics Lactobionic acid Food Solubilizing agent
Anti-aging agent Sweetener

Keratinization agent
Chemical industry Bioactive ingredient
Antioxidant
Chelating agent
Antioxidant Cobuilder in detergents
Functionalization agent
Sugar surfactants
Nanoparticles
Figure 2. Overview of LBA applications.
LBA was reported and synthesized for the first time by Fischer and Meyer (1889) as the
oxidation product of the lactose free aldehyde group after a chemical oxidation with
Bromine. In the industrial manufacturing context, LBA is currently produced by chemical
synthesis from refined lactose in an energy-intensive process (Kuusisto et al., 2007; Yang
and Montgomery, 2005). This manufacturing process also involves the use of harmful and
costly catalysts, as well as the generation of undesirable side-reaction products (Murzina et
al., 2008). Aside from bio-production methods, catalytic wet oxidation and electrochemical
catalysis of lactose have also been proposed as methodologies to obtain LBA. However,
these systems involve costly operational processes (Drulliole et al., 1995; Magariello, 1956)
or the generation of by-products during the course of the reaction (Chia et al., 2008).
Likewise, increasing environmental requirements will undoubtedly drive LBA industrial
production towards more sustainable production systems that, to date, otherwise not
complied with chemical catalytic oxidation or enzymatic methods.
Table 1. Physicochemical properties of LBA.
Structural formula
Systematic name 4-O-β-D-galactopyranosyl-D-gluconic acid
Molecular formula C12H22O12
Molecular weight (g/mol) 358.30
Physical status Solid
Appearance White powder
Melting point (ºC) 128-130ºC
Solubility Freely soluble in water, slightly soluble in anhydrous ethanol and methanol
[α]20D +53º to +22.6º
pKa a 3.6
a: 25ºC
The major industrial manufacturers of LBA are Solvay (Germany), FrieslandCampina

Domo (Netherlands), Sandoz (Germany), Reliable Biopharmaceutical Corporation (USA) and
the US Dairy Ingredient Company (USA) (Affertsholt, 2007; Playne and Crittenden, 2009).
Although no official data are available on its annual world production (Playne and
Crittenden, 2009), LBA has burst onto the commercial scene as a prominent emerging
substance with a significant improved commercial rate in recent years (Affertsholt, 2007).
In view of LBA market forecasts, Solvay in fact opened a plant in Germany in the late 1990s
with the aim of achieving an annual production of 1000 tons (Gerling, 1998).
3. Use and commercial applications of lactobionic acid.

3.1 Health and biomedical applications: from drug-delivery systems to preservative solutions
3.1.1 Drug-delivery systems
Recent trends in biomedicine have witnessed the appearance of novel techniques and
strategies dealing with nanotechnology, tissue engineering, drug-delivery systems or
biomaterials based on biodegradable and biocompatible chemicals which are coming
increasingly to the fore. Targeted delivery of therapeutic agents has thus emerged as a
promising approach in medicine not only due to its increased therapeutic efficacy, but also
to its lesser side effects. In fact, the lack of effective and practical pharmacological
approaches for targeting liver cells has become the driving force behind the development of
effective drug-delivery systems employing LBA as a ligand. In this context, LBA offers unique
properties such as biocompatibility, biodegradability, ion-chelating ability and self-
assembly, in addition to their synergistic combination, which confer on LBA a plethora of
applications as a vehicle for the innovative treatment of many potential diseases. Owing to
these unique properties, LBA provides an excellent platform for the synthesis of potentially
biocompatible and targetable drug delivery vehicles, from DNA to bioactive molecules. Table
2 lists recent research studies that have focused on the development and assessment of
drug-delivery systems containing LBA as a ligand.
The current growing prevalence of hepatic disorders has become the driving force behind
the development of novel therapeutic treatments based on hepatic-targeted drug delivery
systems (Chen et al., 2012a). Targeting hepatoma cells constitutes a promising approach for
clinical chemotherapy of liver cancer which is currently a subject of intensive research
efforts (Chen et al., 2012b; Duan et al., 2011; Huang et al., 2011; Na et al., 2000; Varshosaz et
al., 2012).
LBA-based drug delivery systems can successfully target hepatocytes due to the presence
of asialoglycoprotein receptors (ASGPR) located on their surface, since LBA works as a
ligand of these receptors (Kim et al., 2006; Peng et al., 2007). These receptors represent the
preferential targets for potential hepatic interventions, as the therapeutic effect of any drug
is often limited by the difficulty of accessing specific sites. This receptor additionally
provides a membrane-bound active site for cell-to-cell interactions (Stockert, 1995).
Considerable emphasis has been placed on the surface modification of polysaccharides by
galactosylated groups from LBA targeting ASGPR. The functionalization of LBA may thus
achieve enhanced binding to the cell surface on hepatocytes (Park et al., 2001).
Furthermore, successful receptor-mediated cell transfection in targeted specific tissues such
as epithelial cells has been achieved through the functionalization of PEI and PEI-PEG-
copolymers with LBA (Weiss et al, 2006).
Table 2. Comparison of research studies using LBA as the target ligand for drug delivery purposes.
Graft
System Applications Polysaccharide Drug material/ Targeted cells Reference
polymer
Nanoparticles Drug delivery Chitosan Oridonin n.u. Hepatocytes Zheng et al. (2012)
Jain and Jain

Nanoparticles Drug delivery Chitosan Doxorubicin n.u. HepG2
(2010)
Nanoparticles Drug delivery Chitosan n.u n.u HepG2 Mi et al. (2007)
Nanoparticles PET imaging Chitosan Fluorine-18 n.u. Hepatocytes Yang et al. (2010)
Nanoparticles Drug delivery Chitosan n.u MPEG Hepatocytes Lin et al. (2009)
Core-shell
Biotinylated Kim and Kim
polymeric Drug delivery ATRA n.u. Hepatocytes
poly(ethylene glycol) (2003)
nanoparticles
Core-shell
Diamine-terminated Kim and Kim
polymeric Drug delivery Clonazepam n.u. Hepatocytes
poly(ethylene glycol) (2002)
nanoparticles
Microcapsules Drug delivery Chitosan n.u. n.u. Hepatocytes Zhang et al. (2011)
Gene carrier DNA delivery Dextran n.u. PEI, PEG Hepatocytes Weiss et al. (2006)
Gene carrier DNA delivery Chitosan n.u. Spermine HepG2 Alex et al. (2011)
Gene carrier DNA delivery PEG n.u. Poly-L-lysine HepG2 Choi et al. (1998)
Gene carrier DNA delivery Chitosan n.u. n.u. HepG2 Gao et al. (2003)
Gene carrier DNA delivery Chitosan n.u n.u HepG2 Lin et al. (2011)
Gene carrier DNA delivery Chitosan n.u Dextran Hepatocytes Park et al. (2000)
Gene carrier DNA delivery Chitosan n.u. PEG Hepatocytes Park et al. (2001)
Gene carrier DNA delivery Chitosan n.u. PVP Hepatocytes Park et al. (2004)
Gene carrier DNA delivery Chitosan n.u. n.u. HepG2 Kim et al. (2004)
Gene carrier DNA delivery PEI n.u. PEG HepG2 Kim et al. (2005a)
Gene carrier DNA delivery Chitosan n.u. PEI HepG2 Kim et al. (2005b)
Gene carrier DNA delivery Chitosan n.u. PEI Hepatocytes Jiang et al. (2007)
Gene carrier DNA delivery Chitosan n.u. PEG Hepatocytes Jiang et al. (2008)
Table 2. (Continued)
Graft
System Applications Polysaccharide Drug material/ Targeted cells Reference
polymer
Photodynamic
Drug carrier n.u. Porphyrin n.u. Hepatocytes Li et al. (2007)
therapy
Liposomes DNA delivery Poly-L-lysine n.u. n.u HepG2 Wang et al. (2006a)
Wang et al.
Liposomes Drug delivery n.u. Doxorubicin n.u. Hepatocytes
(2006b)
Antigen
LDL nanoparticles a n.u. n.u. n.u. Macrophages Wu et al. (2009b)
delivery
Micelles Drug delivery PEO-PPO-PEO Silybin n.u. Hepatocytes Li et al. (2009)
Micelles Drug delivery Polycaprolactone Prednisone Dextran HepG2 Wu et al. (2009a)
Poly(Ɛ-
Micelles Drug delivery PEG Paclitaxel Hepatocytes Yang et al. (2011)
caprolactone)
Photodynamic
Micelles Polycaprolactone Porphyrin n.u. HepG2 Wu et al. (2010)
therapy
PET: positron emission tomography; PEG: poly(ethylene glycol); PVP: Poly(vinyl pyrrolidone); PEI: polyethylenimine; HepG2: liver
hepatocellular cells; ATRA: anti-cancer drug all-trans-retinoic acid; PEO-PPO-PEO: poly (ethylene oxide)-co-poly(propylene oxide)-
co-poly(ethylene oxide) (Pluronic triblock copolymer); MPEG: methoxy poly(ethylene glycol).
n.u.: not used
a Conjugated with fluorescein ovalbumin.
Recent developments in nanotechnology have also raised a great deal of interest in the
design and formulation of targeted non-viral delivery constructs for siRNA in vivo delivery
capable of overcoming barriers. In this regard, chemical modification of chitosan for the
delivery of DNA and siRNA has become a valuable tool for treating acquired diseases
through the transfection of malign cells using gene therapy (Alex et al., 2011). A
hepatocellular carcinoma cell-targeted gene carrier has also been approached by means of
poly-L-lysine grafted with LBA and poly(ethylene glycol) (Choi et al., 1998), low molecular
weight chitosan (Gao et al., 2003; Ill’ina and Varlamov et al., 2007) and chitosan-
graftpolyethylenimine (Jiang et al., 2007). LBA has also been employed for the specific
targeting of a rotavirus capsid VP6 to hepatocytes or hepatoma cells bearing ASGPR
receptors, achieving increased enhancement of the receptor-mediated endocytosis in
hepatoma cells (Zhao et al., 2011). Potential protection from fulminant hepatitis was
achieved by Dong et al. (2009) through targeting delivery of antisense oligonucleotides
containing LBA into Kupfer cells with the aim of reducing the expression of liver tumour
necrosis factor-α. In addition, complexes of DNA/polyethylenimine-graft-poly(ethylene
glycol) (Kim et al., 2005a) or poly(ethylenimine) (Kim et al., 2005b) have been shown to be
efficient gene carriers. LBA has been also successfully introduced into water-soluble
chitosan through an amide linkage in order to improve not only the hepatocyte specificity of
DNA carriers, but also the transfection efficiency of the complexes into hepatoma cells (Kim
et al., 2004). The transfection efficiency of DNA into hepatoma cells has been also enhanced
in synthesized PEGylated chitosan polyplexes (<200 nm) bearing LBA (Lin et al., 2011).
Chitosan nanoparticles functionalized with LBA have been proposed as alternative and
biocompatible cationic polymers for non-viral gene therapy (Jiang et al. 2008).
Nanoparticles comprising galactosylated chitosan (Kim and Kim, 2002; Kim et al., 2004; Kim
et al., 2005b), galactosylated poly(ethylene glycol) (Kim and Kim, 2003), galactosylated
polyethylenimine-graft-poly(ethylene glycol) (Jiang et al., 2008), galactosylated
poly(ethylene glycol)-chitosan-graft-polyethylenimine (Kim et al., 2005a) or galactosyl
conjugated N-succinyl-chitosan-graft-polyethylenimine (Lu et al., 2010) have likewise been
used as gene carriers with enhanced transfection efficiency and hepatocyte specificity both
in vitro and in vivo due to the presence of LBA as functional group (Table 2).
LBA can also serve as a chemical platform for the biosynthesis of antioxidant carriers
(Ortial et al., 2006). Amphiphilic copolymeric sub-micro particles containing LBA have
recently been devised as a carrier shielding from enzymatic attack for peptide and protein
delivery (Meng et al., 2012). Tian et al. (2008) have reported the synthesis, self-assembly
and in vitro evaluation of a novel amphiphilic glycopeptide triblock copolymer
(polytetrahydrofuran PTHF, poly(L-lysine)s PLLs and Ɛ-benzyloxycarbonyl-L-lysine N-
carboxyanhydride) containing LBA as a carrier for controlled drug release.
In addition to the above drug delivery systems containing LBA as a ligand molecule, a
novel and promising controlled-release system has also been recently devised. Interestingly,
Luo et al. (2011, 2012) have developed a biocompatible controlled-release system based on
mesoporous silica nanoparticles for intracellular pH-responsive delivery of anticancer drugs
using LBA (conjugated to BSA) as an end-cap. These authors devised an efficient
nanoreservoir for efficient intracellular anticancer drug delivery with high cell-specific
endocytosis and intracellular pH-responsive controlled drug release which induces effective
cell apoptosis. These nanoparticles may thus not only serve as intelligent nanoreservoirs of
drugs, but also control anticancer drug release in response to intracellular acidic pH (Luo et
al., 2011, 2012). Besides, pH-responsive nanogels using LBA as active targeting has been
employed for hepatoma-targeted delivery of oridonin, exhibiting a higher antitumor activity
than drug-loaded nanogels without LBA (Duan et al., 2011). A potential liver-targeting
prolonged-circulation polymeric prodrug of doxorubicin containing LBA was developed
likewise by Huang et al. (2010). This novel antitumor system showed better in vivo
antitumor efficacy and lower cytotoxicity, suggesting its great potential as highly efficient
polymeric antitumor prodrug (Huang et al., 2010).
3.1.2 Nanoparticle diagnosis

Recent advances in the synthesis and surface modification of nanoparticles using LBA
have emerged as promising candidates for novel diagnostic tools in biomedicine. Bio-
functionalization agents such as LBA have attracted increasing attention in recent years as
surface coating materials due to their enhanced properties. Specifically, bio-functionalized
nanoparticles exhibit several key advantages compared to traditional therapeutic tools, such
as enhanced biodegradability, biocompatibility, detection sensitivity, water-solubility,
longer stability and lower cytotoxicity (Knopp et al., 2009). As a result, these nanoparticles
are used for biomedical applications such as bio-detection, cancer therapies, magnetic
resonance imaging or bio-labelling. Table 3 summarizes the recent research studies on
nanoparticles containing LBA as a coating material.
Table 3. Summary of research studies dealing with nanoparticles containing LBA as a coating material.
Nature of Synthesis Average

Applications Coating effect Targeted cells Reference
nanoparticles method size (nm)
Manganese oxide
MRI contrast
surface doped Synthesis + Improved biocompatibility Choi et al.
agent, liver 4.6 Kidney
gadolinium oxide doping and water solubility (2010)
diagnosis
(Gd2O3@MnO)
Improved stability and Bahadur et al.

Magnetic iron oxide Co-precipitation Clinical diagnosis 8-10 Hepatocytes
dispersibility (2009)
Improved stability and

Selim et al.
Magnetic iron oxide Co-precipitation Liver diagnosis increased cellular 26 Hepatocytes
(2007)
internalization
Improved biocompatibility
Arrested Selim et al.
Cadmium sulfite Liver diagnosis and increased cellular 35 Hepatocytes
precipitation (2009)
internalization
Kekkonen et
Magnetic iron oxide Co-precipitation Liver diagnosis Improved stability 120 Hepatocytes
al. (2009)
MRI: Magnetic Resonance Imaging.
Surface coating of nanoparticles with LBA contributes to enhanced physicochemical

stability, as well as increased uptake by hepatocytes. Nanoparticles combining LBA (its
galactose moiety) and diamine-terminated poly(ethylene glycol) exhibited in vitro release of
A1 1-trans-retinoic acid over 1 month, underlining the key role of LBA as a surface coating
material (Kim and Kim, 2003).
In view of the increasing demand for novel magnetic resonance imaging contrast agents,
numerous approaches have employed LBA-functionalized nanoparticles for bioimaging
applications. These novel bioimaging techniques play a vital role in diagnosing diseases,
providing enhanced sensitivity and advanced functionalities (Kekkonen et al., 2009; Selim et
al., 2007; Richard et al., 2009). As a result, LBA conjugated to amino-modified silica
(fluorescent) nanoparticles has been shown to be an efficient and promising probe in the
detection and identification of liver cancer cells in cellular complex matrices such as blood
(Peng et al., 2007).
Another recently developed approach to targeting liver cancer cells involved the use of
synthesized glycoconjugated poly(amindoamine) dendrimers as diagnostic tools (Guo et al.,
2012). These dendrimers were synthesized by conjugating the fluorophore fluorescein
isothiocyanate and LBA to confirm their in vitro targeting capabilities as key elements for
the development of efficient strategies targeting carcinoma cell lines (Guo et al., 2012).
3.1.3 Tissue engineering

More recently, tissue engineering has emerged as a promising area for the development
of novel functional replacement tissues (scaffolds and biomaterials) aiming to substitute,
repair or regenerate injured tissues and organs. Liver tissue engineering using hepatocytes
and scaffolds has been developed as an alternative method for the treatment of liver failure.
In view of the growing incidence of hepatic disorders, new approaches are needed not only
to meet the liver shortage problem, but also to provide innovative solutions for regenerative
medicine (Kim et al., 2008). Although LBA does not contribute directly to the growth of any
tissue or biomaterial, this organic acid may facilitate the attachment and entrapment of
hepatocytes, as well as the establishment and construction of biocompatible scaffolds for
liver tissue engineering purposes. Table 4 summarizes the noteworthy recent research
studies focussing on the use of LBA as a bio-functionalization agent in biomaterials or
biocompatible scaffolds for tissue engineering.
Scaffolds generated from natural polymers such as alginates, chitosan or
glycosaminoglycans, as well as from synthetic polyesters such as poly(lactic acid) (PLA),
poly(glycolic acid) (PGA) or poly(caprolactone) (PCL), have been widely explored as tissue
engineering materials. However, their functional capacities and biocompatibility may be
enhanced through bio-functionalization with LBA. Novel lactone-based glycopolymers,
galactosylated chitosan (Fan et al., 2010), polycaprolactone scaffolds modified with
galactosylated chitosan (Qiu et al., 2012) and alginate/galactosylated chitosan sponge for
liver-tissue engineering have thus been developed in the last decade (Yang et al., 2001).
Specifically, the incorporation of functional groups such galactose moieties in the polymer
backbone enables the improvement of its functional properties, exhibiting higher levels of
liver-specific function maintenance in terms of albumin secretion or urea synthesis (Choi et
al., 2012; Feng et al., 2009). The interaction between ASGPRs and the galactose ligands
displayed by LBA is fundamental to enhance hepatocyte adhesion, hepatocyte aggregation,
maintenance of liver-specific functions and mechanical stability (Kim et al., 2011). In fact,
LBA-based scaffolds may be better recognized by hepatocytes through ASGPR receptors,
showing better cell viability and long-term maintenance of liver-specific functions (Feng et
al., 2009).
Table 4. Summary of noteworthy recent research studies on biomaterials using LBA.
Scaffold/polysaccharide Applications Conformation Reference
Polycaprolactone Hepatocyte attachment Film Qiu et al. (2012)
Alginate/chitosan Hepatocyte attachment Porous Yang et al. (2001)
Alginate Hepatocyte attachment Film Chung et al. (2002)
Alginate Hepatocyte entrapment Film Yang et al. (2002)
Chitosan Hepatocyte aggregation Nanofibrous Feng et al. (2009)
Chitosan Hepatocyte culture Film Fan et al. (2010)
The design of liver tissue-mimicking microenvironments with bioactive scaffolds

constitutes a key challenge for the field of tissue engineering. Without a doubt, synthetically
engineered matrices for cellular recognition and attachment constitute pivotal elements for
the success of any tissue engineering approach. Artificial extracellular matrices may play a
key role in tissue engineering since they provide cell-adhesive substrate, control of three-
dimensional tissue structure, and the presentation of growth factors or cell-adhesion signals
(Kim et al., 2011). In fact, galactosylated-matrices through LBA have been proved to be
effective in hepatocyte entrapment and attachment. In this regard, a LBA-functionalized
chitosan has been described as an effective extracellular matrix for hepatocyte attachment
(Mi et al., 2006; Park et al., 2003). Specifically, the incorporation of LBA in a synthetic
polymer such as poly(vinyl alcohol) may provide a better microenvironment for cell
aggregation along with nutrition and metabolite transfer (Guo et al., 2003). Photosensitive
poly(allylamine) containing LBA has been also suggested as effective extracellular matrix
(Kang et al., 2004; 2005). A biological adhesive comprising a photocrosslinkable chitosan
molecule formed by LBA and a photoreactive azide group has likewise been developed as an
innovative solution for hepatocyte attachment, which facilitates cell-cell as well as cell-
matrix interactions (Ono et al., 2000). Furthermore, a novel thermo-responsive biomaterial

containing LBA has been recently devised for biomedical and tissue engineering by Wang et
al. (2010). This synthesized copolymer was not only capable of improving the cellular
viability but also retained the activity of the detached cells (Wang et al., 2010).
3.1.4 Other health-related products

In addition to the aforementioned applications, LBA is already being widely used in
health-related products such as pharmaceuticals, antibiotics and preservative solutions for
organ transplantation. This polyhydroxy bionic acid has been used as an adjuvant in the
preparation of erythromycin for treating bacteremia (Hirakata et al., 1992) and in
formulations containing chlorhexidine as an antiseptic substance (Werle et al., 2002). LBA
may also work as a stabilizer of pharmaceutical formulations containing antibodies
(Agarkhed et al., 2010) or interferons (Goldstein and Thatcher, 1990), and as a solubilizer to
enhance the aqueous solubility of argatroban, an antithrombotic agent, in alcohol-free
pharmaceutical compositions (Palepu, 2008). Furthermore, novel potential applications
have emerged in recent years, such as the use of LBA-based derivatives to prevent and treat
malaria (Agustí et al., 2004; Giorgi et al., 2010). More recently, an ophthalmic composition
based on LBA as a therapeutic agent has been devised for the therapeutic treatment of
corneal edema and inflammation, overcoming the drawbacks displayed by conventional
hyperosmotic topical formulations (Cavallo et al., 2012). Stannard et al. (2010) have
demonstrated that LBA may act as a specific high affinity inhibitor compound of
carbohydrate binding proteins (galectin-blocking), which promotes vaccine-stimulated
immune responses against in vivo breast tumours that may lead to a decrease in tumour
progression, as well as an improved survival outcome.
LBA has also been present in the development of novel biopharmaceuticals such as
aprosulate and equol. The latter is the major intestinally-derived metabolite of the
phytoestrogen daidzein (one of the main isoflavones found abundantly in soybeans), which
displays high antioxidant and estrogen activities. Equol has already been revealed to be an
agent for treating osteoporosis or preventing breast cancer and vascular diseases, among
other pathologies (Oe and Kimura, 2009). Unitika (Japan) and the Osaka Municipal Technical
Research Institute (Japan) have developed an equol pharmaceutical composition containing
LBA as the main component with enhanced biological properties (Oe and Kimura, 2009;
Unitika, 2008). These authors found that the administration of LBA promotes plasma equol
production in animal laboratory tests, working as an anti-aging supplement in addition to
reducing menopause symptoms (Unitika, 2008).
Likewise, a LBA-based heparin analogue (Aprosulate) has been also devised in recent
years as a new amide-derived antithrombotic agent with lower risk of bleeding than
unfractionated heparin or low molecular weight heparins (Hoppensteadt et al., 1988; Ofosu
et al., 1992; Raake et al., 1989). This sulphated bis-lactobionic acid amide was found to be
25% as effective as unfractionated heparin in its ability to delay both the activation of factor
X and the onset of prothrombin activation (Béguin et al., 1991; Ofosu et al., 1992). Klöcking
et al. (1991) have additionally shown that the administration of bis-lactobionic acid amides
increases the release of the tissue-type plasminogen activator, thereby inducing fibrinolysis.
Unfortunately, however, further development of aprosulate as a novel anticoagulant drug
was abandoned after Phase II clinical trials, as it was associated with liver toxicity (Coombe
et al., 2012; Krylov et al., 2011), despite the reduced coagulation and platelet functions
achieved during Phase I clinical trials (Schenk et al. 1999).
Another major application for LBA involves its use as the core compound of preservation
solutions employed in organ transplantation protocols which suppress tissue damage
caused by free radicals during organ storage (D’Alessandro et al., 1999; Hart et al., 2002).
The University of Wisconsin solution (UW solution), also known as ViaSpan (commercial
name) or Belzer solution, was developed in the late 1980s by Belzer and colleagues as one of
the first intracellular-like preservation media for cold storage of organs before
transplantation (Belzer et al., 1992; Southard and Belzer, 1995). This preservation solution
has now been adopted worldwide as the gold standard clinical dissolution for organ
perfusion in transplantation medicine (Southard, 2002; Upadhya and Strasberg, 2000). UW
solution is known to effectively prevent ischemia and reperfusion injury, playing a key role
in organ cytoprotection and viability prior to transplantation (Hart et al., 2002). As the
central core component of UW solution, LBA works as a cell impermeant agent during the
cold storage of organs, as it provides osmotic support and prevents cell swelling
(D’Alessandro et al., 1999; Sumimoto and Kamada, 1990). Besides, LBA acts as an
antioxidant by scavenging free hydroxyl radicals generated in the medium (Charloux et al.,
1995; Hart et al., 2002), as well as a cryptic inhibitor of matrix metalloproteinases due to the
fact that it may bind to the divalent zinc atom in the core of these gelatinases and hence
inhibit their activity (Upadhya and Strasberg, 2000). These undisputable biological
properties offered by LBA have been recently employed for the development of
cryoprotectant suspensions containing LBA that are clinically used in cell therapies as
compatibles vehicles (Hope et al., 2012).
3.2 Cosmetics field

The cosmetics industry is currently employing LBA as the key active component of novel
anti-aging and regenerative skin-care products due to its therapeutic efficacy. The use of
LBA in cosmetics has grown three- to five-fold since it was launched commercially, with LBA
presenting a major market niche as a novel and essential protective component of skin-care
formulations. In fact, the cosmetics company NeoStrata (USA) has devised a broad portfolio
of skin-care products based on the unique features offered by LBA in a market which has
grown enormously over the past few years, generating US sales of about $1 billion (West,
2004a). As a cosmetic ingredient, LBA offers multiple benefits for the therapeutic treatment
of dermatological pathologies such as atopic dermatitis and rosacea (Briden and Green,
2006; Hatano et al., 2009) or can even be employed in anti-acne treatments (Decker and
Graber, 2012). Some thirty years ago, Van Scott and Yu found that topical application of
alpha-hydroxyacids showed a very specific effect on hyperkeratinization (Green et al., 2009;
Yu and Van Scott, 2004). Recently, polyhydroxy acids such as LBA have emerged as the new
generation of alpha-hydroxyacids which not only offer similar therapeutic effects, but also
reduce side effects like irritation and stinging derived from the use of traditional hydroxy
acids, suggesting that polyhydroxy acids may be employed as suitable molecules for the
treatment of sensitive skin (Decker and Graber, 2012; Green, 2005). Tasic-Kostov et al.
(2010) have clinically evidenced the lack of both skin irritation and skin barrier impairment
displayed by LBA compared to classical hydroxyacids such as glycolic acid.
As a versatile substance in cosmetics, LBA is employed as an anti-aging and
keratinization agent in cosmeceutical formulations, thus promoting the biosynthesis of
glycosaminoglycans or collagen in addition to improving skin thickness and firmness (Green
et al., 2008, 2009). Both barrier homeostasis and stratum corneum functionality are clearly
improved when LBA is applied topically (Hachem et al., 2010). LBA in fact constitutes an
inhibitor of the breakdown of matrix metalloproteinase enzymes due to metal chelation,
thus reducing the appearance of photoageing and wrinkles (Grimes et al., 2004). In addition
to LBA’s antioxidant role, it also exhibits strong moisturizing, exfoliative and humectant
properties (Yu and Van Scott, 2004).
3.3 Applications of LBA in the chemical industry

LBA is used in the chemical industry as a sugar-based surfactant or as a cobuilder in
biodegradable detergents (Gerling, 1998). Its iron-chelating and emulsifying properties
have suggested its potential use for many industrial purposes, including its use as an
important starting chemical for producing detergents (Gerling, 1998). Novel
alkylaminoamide (Oskarsson et al., 2007), nonionic (Rôzycka-Roszak et al., 2007),

catanionic (Bize et al., 2010; Rico-Lattes et al., 2005), lactobionamide (Lebaupain et al.,
2006), gemini (Yoshimura et al., 2005), polysiloxane (Zeng et al., 2012) and aldonamide
sugar surfactants (Wilk et al., 2002) have thus been recently developed as novel
biodegradable LBA-based surfactants which provide improved surface and performance
properties, as well as a reduced environmental impact. In particular, Gerling et al. (1996)
have developed LBA amide compositions (formed by reacting LBA with mixtures of primary
fatty acids) with good foam stabilizing, cleaning, emulsifying and softening properties in
aqueous systems for their application in detergents, cleaning agents and cosmetic
formulations. Environmentally-friendly compositions comprising LBA N-alkylamides have
likewise been proposed as corrosion prevention agents for their use in many metalworking
operations (Gerling et al., 1998). LBA is also being specifically used as an active ingredient in
a novel antibacterial agent composition which confers excellent preservation stability on
consumer goods as such as foods, raw materials, detergents, cosmetic products and
medicines (Araki et al., 2006). In another approach, LBA has been proposed as a building
block for the biocatalytic synthesis of novel polymers with possible industrial applications
(Kakasi-Zsurka et al., 2011).
Aside from the aforementioned applications, LBA can play a key role as a
functionalization agent for the synthesis of innovative industrial systems such as
functionalized carbon nanotubes with LBA amide amphiphile molecules which are capable
of adsorbing proteins (Feng et al., 2011). Interestingly, LBA-based derivatives could serve as
a potential protein cross-linker for both food and non-food applications (van Wijk et al.,
2006). Zhou et al. (2006) have reported the direct electrochemistry and electrocatalysis of
hemoglobin in a LBA film, which suggests that a new kind of unmediated third-generation
hydrogen peroxide biosensor might be developed.
3.4 Food applications

The use of LBA as a food additive has also received growing attention from the food
industry in recent years. It can serve as an anti-oxidant, stabilizer or gelling agent in bakery
products (Gerling, 1998; Gutiérrez et al., 2012), an acidifier agent in fermented milk
products (Faergemand et al., 2012), an aging inhibitor for bread (Oe and Kimura, 2011) and
as a mineral absorption enhancer in dairy desserts (Oe and Kimura, 2008). LBA has also
been proposed as a technological feed additive for laying hens with the aim of improving
eggshell qualities by boosting calcium absorption (Kimura, 2006). Equally importantly,
novel non-dairy beverages (Nielsen, 2007), milk-based beverages (Oe et al., 2007) as well as
cheeses (Nielsen and Hoeier, 2009) containing calcium lactobionate have recently been
developed to provide a valuable approach for calcium supplementation. Specifically, the
incorporation of LBA into foods may stimulate intestinal Ca2+ (Brommage et al., 1993) or
mineral absorption (Baldwin et al., 2007), thereby exerting a clearly health-promoting
influence. Experiments carried out by Oe et al. (2008) have shown that calcium absorption
in rats fed a diet supplemented with LBA was higher than in rats fed a control diet, which
underlines the potential role of LBA as an agent for preventing and/or treating health
disorders associated with calcium deficiency.
The dairy industry has been particularly involved in recent years in the development and
implementation of innovative manufacturing processes including LBA as a key ingredient in
novel dairy-making technologies (Koka et al., 2005; Merrill and Singh, 2011; Novozymes,
2009). In fact, adding LBA either directly (up to 10% of the mix) or indirectly (generated in
situ by an added lactose oxidase during the course of the process) enables enhanced
production yields, lower processing times and cost savings in manufacturing (Koka et al.,
2005). Coagulation may be induced during cheese-making without using culture starters
and/or rennet, thereby reducing the need for these costly additional ingredients.
Furthermore, the use of LBA as an ingredient in dairy-making and in other product
manufacturing processes also provides extra functional properties and sensory attributes
through the reduction of undesirable Maillard browning in cooking products (Merrill and
Singh, 2011). Above and beyond these applications, a novel LBA-based flavour enhancer for
foods or beverages has been also reported by Walter and Begli (2011), which particularly
underlines LBA’s versatility. Baldwin et al. (2004) have also devised an antioxidant
composition containing LBA and siderophores as key elements for retarding lipid oxidation
in food products. The role of LBA as a water-holding capacity agent in meat products
submitted to thawing and/or cooking processes has recently been reported for the first
time, resulting in higher industrial product yields and water contents after treating meat
products with LBA (Nielsen, 2009).
In addition to the aforementioned technological role, LBA may also exert potential
prebiotic effects as a bioactive ingredient in functional foods, as this organic acid, which is
resistant to digestive enzymes, is poorly absorbed in the small intestine and can be
subsequently fermented by the gastrointestinal microflora (Saarela et al., 2003; Schaafsma,
2008). In fact, Saarela et al. (2003) have evaluated the potential health benefits of LBA as a
prebiotic substance, although unfortunately the enhancement on its functional properties
could not be proved. Nonetheless, experiments carried out by Kontula et al. (2000) have
confirmed that LBA may promote both the growth and in vitro intestinal adhesion of
Lactobacillus cellobiosus (a lactic acid bacteria isolated from the human colonic intestine)
compared to other carbohydrate sources, thereby underlining its potential role as a
prebiotic substance. Although LBA rarely occurs in nature, it has been isolated in the upper
layer from “Caspian Sea yoghurt”, a yoghurt marketed in Japan (Kiryu et al., 2009). These
authors have estimated an individual intake of 0.5 to 1.0 g of LBA per year when consuming
100 g of the yoghurt per day, which suggests that the presence of this organic acid in dairy-
fermented products could be more common than previously reported (Kiryu et al., 2009).
Although the role of LBA as a food preservative is anticipated, its use as calcium
lactobionate (E-399) has already been approved in the USA by the FDA (FDA, 2011),
whereas its status in the EU is still pending its approval as a food additive (Playne and
Crittenden, 2009; Schaafsma, 2008). Further assessments regarding its long-term safety and
human exposure are thus necessary before its approval as a novel food by the European
Food Safety Agency (EFSA). Nevertheless, studies carried out by Van Dokkum et al. (1994)
have attempted to determine the effect on the human health of exposure to the continued
consumption of LBA for one week. These authors concluded that amounts of up to 24 g of
LBA per day were well tolerated without producing serious adverse effects on human
health. Evidence of LBA fermentation in the human colon was also found, suggesting a likely
role of LBA as a prebiotic substance (Schaafsma, 2008; Van Dokkum et al., 1994).
4. Lactobionic acid bio-production methods

4.1 Microbial fermentation strategies for lactobionic acid production.
The use of microbial cell factories for LBA bio-production has become a feasible way to
overcome certain drawbacks associated with chemical or enzymatic approaches. Despite the
apparent advantages offered, the industrial production of bionic acids by fermentation has
barely been explored to date compared to other conventional organic acids. From an
industrial point of view, the titer obtained through any biotechnological approach must be
at least 50-100 g/L in order to achieve product concentrations comparable to those
obtained by means of chemical processes (Pollard and Woodley, 2007).
The production of LBA by bacteria was first discovered in the late 1940s by Stodola and
Lockwood (1947) after a wide screening for LBA-producing ability in Pseudomonas species.
In fact, microbial production and consumption of bionic acids (maltobionic, aldonic and
lactobionic) have been explored since the 1950s (Stodola and Jackson, 1950). Research
studies by Kluyver et al. (1951) and Stodola and Lockwood (1947) both underlined the
importance of Pseudomonas species as the main bionic producer microorganisms. Strains
such as P. mucidolens, P. myxogenes or P. fluorescens were found to be able to oxidize lactose
directly without the need for prior hydrolysis or phosphorolysis. Pseudomonas taetrolens
(formerly known as P. graveolens) displayed the highest oxidative ability, with a yield of
75% in shake-flask culture after 165 h (Stodola and Lockwood, 1947). Furthermore,
experiments carried out by Sternberg and Lockwood (1969) demonstrated that P. taetrolens
was able to oxidize not only 1-4 disaccharides (i.e. lactose or maltose), but also isomaltose
(α 1-6 disaccharide) to their corresponding bionic acids without hydrolyzing the glycosidic
linkage. This latter feature was subsequently explored by Miyake and Sato (1975), who
reported the production of aldonic acids from starch sugars after a two-stage process,
including a first growth of the Pseudomonas species followed by the subsequent enzymatic
synthesis of the target compounds by the permeabilized cells. Pseudomonads are therefore
the ideal microbial platforms for lactose oxidation, which essentially means the production
of LBA, as they lack the genes encoding the β-galactosidase enzyme or lactose transporters.
A slow consumption phenomenon of LBA formed by Pseudomonas strains was described
at a later stage cultivation (>16 days) under prolonged nutrient starvation conditions
(Kluyver et al., 1951). Kluyver et al. (1951) showed that some LBA-producing Pseudomonas
species such as P. quercito-pyrogallica, P. calco-acetica and P. aromatica, which converted
lactose into LBA with yields of 85-90%, were able to consume bionic acids in media in which
these compounds were the sole carbon source. However, further studies by Bentley and
Slechta (1960) concluded that the rate of bionic formation by P. quercito-pyrogallica
appeared to greatly exceed that of utilization, a result which was not observed in short-term
growth experiments. These authors proposed a general mechanism for bionic acid
formation by Pseudomonas strains comprising the appearance of a lactone intermediate
prior to the formation of the aldonic acid, which was further confirmed by Nishizuka and
Hayaishi (1962).
In Pseudomonas species, LBA is formed via the lactose oxidation pathway, in which a
membrane-bound dehydrogenase system catalyzes this single biotransformation (Fig. 3A). A
lactose dehydrogenase enzyme first catalyzes the oxidation of the lactose to a lactone
intermediate (lactobiono-δ-lactone) which is subsequently hydrolyzed (its carbonyl group)
by a lactonase into LBA (Nishizuka et al., 1960; Nishizuka and Hayaishi, 1962). The lactose
dehydrogenase, biochemically characterized by Nishizuka and Hayaishi (1962) from P.
taetrolens cells, contains flavin adenine dinucleotide as a prosthetic group and is closely
associated with a hemoprotein electron transfer system. This flavoprotein, localized in the
particulate fraction of P. taetrolens cells, does not use oxygen as direct electron acceptor and
presents an optimum pH at 5.6, with a Km value for lactose of 11 mM. Although the purified
enzyme may oxidize several aldoses, such as maltose, glucose, mannose, arabinose,
cellobiose, fucose, xylose, ribose and galactose, the specific enzyme activities vary with
growth conditions. Conversely, lactonase presents an optimum pH at 6.5-6.7, with a Km
value for lactose of 20 mM (Nishizuka and Hayaishi, 1962). A similar mechanism is
responsible for the biosynthesis of LBA by purified fungal enzymes, e.g. the cellobiose
dehydrogenase from Sclerotium rolfsii, but in turn the intermediate lactobiono-δ-lactone is
spontaneously hydrolyzed here to LBA without the involvement of the lactonase enzyme
(Ludwig et al, 2004; Van Hecke et al., 2009a).
A
Lactose Lactobiono-δ-lactone Lactobionic acid
FAD FADH2 H2O
Periplasm
Lactose
Lactonase
oxidase
Cytoplasm
B
Sorbitol
Lactose NADP
NADP
NADP Lactobiono-δ-lactone Lactobionic acid

NADPH H H2O
Fructose
Glucose- Lactobiono-δ-lactonase
fructose
Cytoplasm
oxidoreductase
Figure 3. Oxidation pathway for LBA production in bacterial strains. Bioconversion of lactose to lactobionic acid
by Pseudomonas species is carried out through a membrane-located dehydrogenase system which mediates
lactose oxidation and subsequent hydrolysis by lactonase to yield LBA (A). Reaction schemes of a bi-substrate
enzymatic system carried out by glucose-fructose oxidoreductase in Z. mobilis for the production of LBA and
sorbitol (B).
Vakil and Shahani (1969) found that LBA also constitutes an intermediate in the lactose
metabolism of Streptococcus lactis. LBA is first produced from lactose and is then hydrolyzed
by S. lactis to gluconic acid and galactose via the β-galactosidase enzyme in order to be
further metabolized through both the Embden Meyerhof and hexose-monophosphate shunt
pathways (Vakil and Shahani, 1969).
Evidence of the production of LBA by filamentous fungi has also been reported by Bucek
et al. (1956). These authors studied the formation of LBA from lactose and its subsequent
consumption by Penicillium chrysogenum in shake-flask cultures with a LBA yield of only
50% after 120 h, which suggests the presence of residual lactose oxidase activity in this
fungus. Their preliminary results also concluded that this organic acid seems to be only an
intermediate within the lactose metabolism of this fungus. In addition, the lactose-oxidizing
ability has been also found in a red alga, Iridophycus flaccidum, which possesses a
carbohydrate oxidase with wide substrate specificity. The purified carbohydrate oxidase
from this photosynthetic organism was capable of oxidizing glucose, galactose, lactose,
maltose and cellobiose at an optimum pH of 5.0 (Bean and Hassid, 1956). Although there is
no established consensus regarding the role of sugar oxidation in living cells, the presence of
one consumption stage upon oxidation might suggest a key function of this phenomenon in
carbon metabolism under nutrient starvation conditions.
To date, most methodologies related to the microbial production of LBA are based on the
oxidation of refined lactose from media that is both costly and complex (Table 5). However,
LBA can be obtained from an inexpensive feedstock such as cheese whey through a
fermentation process carried out by P. taetrolens (Alonso et al., 2011, 2012a). The use of
cheese whey as an inexpensive source for LBA bio-production undoubtedly suggests a
promising strategy to achieve cost-effective whey upgrading. From a practical point of view,
this system also provides an environmentally-friendly and competitive alternative to costly
and laborious upstream processing steps for LBA production from chemical or enzymatic
methodologies based on refined lactose oxidation. In this system, LBA constitutes a non-
growth related metabolite formed as a consequence of lactose oxidation from whey by P.
taetrolens cells (Alonso et al., 2011, 2012a). P. taetrolens constitutes a non-pathogenic wild-
type microorganism, commonly found in spoiled foods as a natural niche (Levine and
Anderson, 1931; Tompkin and Shaparis, 1972; West, 2004b), which belongs taxonomically
to the P. chlororaphis group (Anzai et al., 2000). The enzymes from this microorganism are
even employed industrially for the synthesis of glutamic acid derivatives (Sugiyama et al.,
2005). Despite being a wild-type LBA-producing microorganism, efficient LBA production
yields, titers and productivity values can be achieved compared to LBA bio-production
approaches carried out by other bacteria such as Burkholderia cepacia or Zymomonas
mobilis which were performed under resting cell conditions involving high initial cell
densities (Malvessi et al., 2012; Murakami et al., 2002, 2003, 2006b; Pedruzzi et al., 2011)
(Table 5). In fact, increasing the inoculum size was shown to improve LBA productivity and
hence reduce the onset of the production phase (Alonso et al., 2011). The use of different
bioconversion strategies at the bioreactor scale such as pH-shift or high-cell density

conditions may undoubtedly overcome certain drawbacks which could limit LBA
productivity by P. taetrolens cells (Alonso et al., 2011, 2012b).
The deleterious influence of an over-supply of oxygen on bioconversion performance by
P. taetrolens has been studied by Alonso et al. (2012a), who concluded that LBA production
from whey was negatively influenced by both high aeration and agitation rates. High
agitation rates schemes stimulated cell growth, increased pH-shift values and the oxygen
uptake rate by P. taetrolens cells, whereas LBA production was negatively affected. Mild
agitation rates (350 rpm) were found to be the optimal agitation strategy during cultivation,
which increased LBA productivity 1.2-fold (0.58-0.70 g/L.h) compared to that achieved at
1000 rpm. Furthermore, aeration rates higher than 1.5 Lpm neither stimulated cell growth
nor LBA production (22% lower for an aeration rate of 2 Lpm), the best oxygen supply
conditions being found to be 0.5 Lpm. Both the supply and control of dissolved oxygen at the
appropriate level have thus been revealed as key bioprocess parameters for enhanced LBA
production, representing essential information for large scale implementation (Alonso et al.,
2012a).
Fermentation performance for LBA production is also strongly related to the
physiological status of the P. taetrolens cells used as inocula. Specifically, their central core
physiology may suffer from inappropriate bioprocessing conditions such as seed culture age
or microenvironmental pH (Alonso et al., 2012b, 2012c). Alonso et al. (2012b) established
the best criteria for harvesting cells for enhanced LBA production, besides revealing how
cells may respond to environmental stress due to prolonged exposure to extreme acidic
conditions, i.e. pH<3.5 (Alonso et al., 2012b). Fermentation performance, which essentially
results in a higher LBA titer, may thus be influenced by the physiological heterogeneity of
the seed culture, which is attributable to the time point of harvesting in the culture used as
inocula for bioreactor cultivations. Seed culture age, which essentially reflects the
physiological status displayed by P. taetrolens cells, may strongly influence not only overall
fermentation performance, but also LBA productivity.
Table 5. Comparison of research studies on the biotechnological production of LBA.
Specific
Cultivation strategy and Culture time Lactobionic acid Productivity Yield L/LBA
Microorganism Culture medium Inoculum mode productivity Reference
operating mode (h) titer (g/L) (g/L.h) (%)
(g/g.h)
Fed-batch in a 2-L Seed culture Miyamoto et al.

Pseudomonas sp. LS13-1 a Lactose, salts and peptone 155 290 - 1.87 90
bioreactor (10% v/v) (2000)
Lactose, salts, peptone, Fed-batch in a 2-L Seed culture Meiberg et al.

Burkholderia cepacia 50 178 - 3.56 85
yeast extract and glucose bioreactor (10% v/v) (1990)
Burkholderia cepacia No. Lactose, salts, peptone and Murakami et al.

Fed-batch in shake-flask Resting cells 240 400 - 1.67 100
24 a yeast extract (2003)
Burkholderia cepacia No. Lactose, salts, corn steep Murakami et al.

Batch in shake-flask Resting cells 27 150 - 5.55 ~100
24 a liquor and yeast extract (2006b)
Permeabilized Pedruzzi et al.

Zymomonas mobilis b Lactose Batch in a 0.15-L bioreactor 22 125 0.80 5.80 100
resting free-cells (2011)
Permeabilized Malvessi et al.

Zymomonas mobilis b Fructose, lactose Batch in a 0.6-L bioreactor 24 182 0.3 7.6 78
resting free-cells (2012)
Lactose, glucose, salts,

Kiryu et al.
Acetobacter orientalis complex nitrogen sources Batch in shake-flask Resting cells 91 49 - 0.54 98
(2012b)
and yeast extract
Seed culture Alonso et al.

Pseudomonas taetrolens Whey Batch in a 2-L bioreactor 58 42 0.56 0.70 100
(10% v/v) (2011)
Seed culture Alonso et al.

Pseudomonas taetrolens Whey Batch in a 2-L bioreactor 30 42 0.94 1.27 100
(30% v/v) (2012a)
Alonso et al.
Pseudomonas taetrolens Whey Batch in shake-flask Resting cells 60 42 0.42 0.70 100
(2012b)
-: not available; a mutant strain; b permeabilized cells

The lack of an adaptive response from the P. taetrolens cells, comprised in the seed
culture, clearly impaired cellular growth, as well as the onset of the production phase
(Alonso et al., 2012b). In fact, prolonged-time seed cultures (>12-h) employed as inocula
affected microbial adaptation, resulting in the presence of lag-phases along with reduced
fermentation efficiencies in terms of LBA productivity, being 1.55-, 2.3- and 3.5-fold (0.45,
0.30 and 0.2 g/L.h for 24-, 36- and 48-h seed cultures, respectively) lower than those
obtained using the youngest seed culture (0.70 g/L.h for a 12-h seed culture) (Alonso et al.,
2012b). Results also revealed that P. taetrolens cells suffered a progressive change in the
physiological status of the cells under the uncontrolled-pH conditions encountered at the
shake-flask scale (pH=3.35), suggesting that microenvironmental pH constitutes a key
bioprocessing factor in LBA microbial production (Alonso et al., 2012b). Production
schemes were thus influenced by the operational pH employed, with asynchronous LBA
production from damaged and metabolically active subpopulations at pH values lower than
6.0, suggesting a strong influence of the pH-control strategy on the fermentative LBA
production (Alonso et al., 2012d). Unlike cultivations performed at 6.5, both pH-shift and
pH-stat cultivation strategies performed at pH values lower than 6.0 resulted in decreased
LBA yields. Whereas the cellular response showed a stress-induced physiological response
under acidic conditions, healthy functional cells were predominant at medium operational
pH values (6.5-7.0) (Alonso et al., 2012d). P. taetrolens displayed accordingly a robust
physiological status at initial pH value of 6.5, resulting in enhanced bioconversion yield as
well as LBA productivity.
Inappropriate bioprocessing conditions undoubtedly play a key role in LBA production
by P. taetrolens cells, seeing as stress-induced physiological responses lead to the collapse of
cellular functions. The importance of physiological heterogeneity thus lies in variations in
cell capacities which may lead to undesired damaged and dead subpopulations of reduced
efficiency during microbial fermentation (Fig. 4). In fact, the biotechnological industry has
recognized the importance of detailed physiological knowledge and bioprocess
characterization during metabolite production at a population level (Díaz et al., 2010;
Lencastre Fernandes et al., 2011).
In this context, the aforementioned approach underlines the suitability of cheese whey as
an inexpensive source for LBA production by a wild-type strain (Alonso et al., 2011, 2012b),
unlike research studies addressing LBA production on synthetic culture media carried out
by mutant strains (Miyamoto et al., 2000; Murakami et al., 2002, 2003, 2006b). In terms of
productivity, high-level production of LBA from nutrient-rich media by Pseudomonas sp.
LS13-1 mutants was achieved under fed-batch fermentation supplemented with 15 g/L of
peptone as the nitrogen source, obtaining a LBA titer of 290 g/L after 155 h, with a
volumetric productivity of 1.87 g/L.h (Miyamoto et al., 2000). However, the use of mutant
strains on an industrial scale is discouraged due to their genetic instability if a rational
method such as metabolic engineering is not employed for their modification (Sauer and
Mattanovich, 2012).
A B C
Figure 4. Confocal fluorescence images of P. taetrolens cells during fermentative LBA production. The green-
fluorescent image represents the signal of metabolically active cells (A), the red-fluorescence image represents
dead cells (B), and the overlay of the two previous images (C). Metabolically active cells were stained green
(cFDA positive cells, green cells) whereas dead cells (PI positive cells, red cells) were stained red. Scale bars: 2
μm.
Complex media containing lactose, yeast extract and peptone with mineral
supplementation for LBA production have been also investigated by Murakami et al. (2002,
2003, 2006b). These authors used a conventional random mutagenesis and screening
methods to obtain bionic acid-producing microbial mutants from a soil sample (Murakami
et al., 2002). The resulting mutants of B. cepacia No. 24 were found to produce up to 400 g/L
of LBA on complex culture media over a period of 240 h, with a volumetric productivity of
up to 1.67 g/L.h (Murakami et al., 2003). In addition to their lactose-oxidizing activity,
aldonic acids have been prepared from galactose, mannose, xylose and arabinose by washed
cells of B. cepacia No. 24 (Murakami et al., 2006a). A subsequent study revealed that B.
cepacia No. 24 even exhibited higher lactose-oxidizing activities owing to the more
favourable pH conditions, which improved LBA productivity to a value of 5.55 g/L.h in
shake-flask cultivation (Murakami et al., 2006b). However, the exact gene cluster modified
by random mutagenesis so far remains undeciphered, unlike metabolic engineering
approaches that require a detailed knowledge and identification of the target genes directly
or indirectly involved in the formation of LBA. The use of complex and expensive media in
addition to the lack of genetic characterization may therefore result in a hurdle for upscaling
this system to the industrial sphere. B. cepacia cells were also found to produce a high LBA
titer (178 g/L) from a high nutrient-rich medium (containing yeast extract, peptone,
glucose, salts and lactose) after 50 h, with a volumetric productivity of 3.56 g/L.h and a yield
of 85% in 2-L fed-batch fermentation (Meiberg et al., 1990). However, the pathogenicity of
this strain discourages any industrial implementation, thus constituting a major
shortcoming for this system.
In view of the pathogenicity of B. cepacia, Kiryu et al. (2009, 2012b) have reported the
biotechnological production of edible LBA by an acetic acid microorganism, Acetobacter
orientalis, which was isolated from the upper layer of the Japanese “Caspian Sea yoghurt”. A.
orientalis was able to effectively oxidize 2-10% of lactose at a yield of 97 to 99% under
resting cell conditions in nutrient-rich media at the shake-flask scale (Kiryu et al., 2012b).
However, these authors reported the production of gluconic and acetic acids by A. orientalis
as undesirable metabolic by-products due to the presence of glucose in the medium. They
have further reported how the membrane-bound dehydrogenase contained in A. orientalis
may oxidize other monosaccharides (such as glucose, allose and xylose among others) as
well as disaccharides, although the highest oxidizing activity corresponded to glucose (Kiryu
et al., 2012a). In a preliminary assay, Oe et al. (2008) also investigated the oxidization of
lactose to LBA by acetic acid bacteria, concluding that Gluconobacter cerinus UTBC-427
showed the strongest lactose-oxidizing activity of the screened microorganisms.
In addition to the aforementioned whole-cell bioconversion approaches performed in
submerged fermentation systems, different systems involving permeabilized cells have been
evaluated in order to achieve high-level production of LBA (Table 5). Permeabilized free-
cells of Zymomonas mobilis under resting mode have been indeed employed to produce LBA
in a bi-substrate system in which the oxidation of lactose to LBA was coupled with the
reduction of fructose to sorbitol (Malvessi et al., 2012; Pedruzzi et al., 2011). As Figure 3B
shows, two products (LBA and sorbitol) result from the reaction carried out by the glucose-
fructose oxidoreductase enzymatic system contained in the permeabilized cells of Z. mobilis.
Malvessi et al. (2012) reported that the best enzymatic activities of glucose-fructose
oxidoreductase were obtained at a pH of around 6.4 and temperatures ranging from 39 to
45ºC. Results from Pedruzzi et al. (2011) concluded that higher initial concentrations of
lactose led to increased activity of this enzymatic system. However, a remaining practical
drawback of this enzymatic system for LBA production is the low affinity of the glucose-
fructose oxidoreductase enzyme for both lactose and fructose, requiring high levels of
substrate concentration to promote high productivities (Pedruzzi et al., 2011).
Comparatively, high LBA titers (125-182 g/L) and yields (78-100%) were achieved by using
permeabilized cells of Z. mobilis (Malvessi et al., 2012; Pedruzzi et al., 2011), although the
amount of LBA produced per g cell mass is worth noting (Table 5). In this regard, these
systems displayed impressively high volumetric productivity values in terms of LBA
production rates (5.8-7.6 g/L.h), although their specific values (0.3-0.8 g/g.h) were
significantly lower than those achieved by P. taetrolens cells (>0.94 g/g.h).
Bioconversion of fructose and lactose into sorbitol and LBA with immobilized cells of Z.
mobilis in calcium-alginate has been also approached (Malvessi et al., 2012; Pedruzzi et al.,
2011). Results from Pedruzzi et al. (2011) revealed that there were no marked mass
transfer limitations on the rate of LBA formation for calcium-alginate beads with an average
diameter of 1.2 mm below a productivity value of 14.9 g/L.h. These authors suggested that
additional studies should be performed to verify the long-term mechanical stability of
calcium-alginate beads (Pedruzzi et al., 2011). Similar results were obtained by Malvessi et
al. (2012), who showed that LBA titers, bioconversion yields and specific productivities
were similar to those obtained with permeabilized free-cells. The highest glucose-fructose
oxidoreductase activities were found at optimum pH values of 7.0-8.0 and temperatures of
47-50ºC (Malvessi et al., 2012).
As regards economic feasibility, the production of organic acids by microbial
fermentation requires productivities greater than 1 g/L.h and final product concentrations
above 50 g/L (Yang et al., 2007). Although most of the approaches listed in Table 5 meet
these criteria, many of them could not be implemented on a commercial scale due to the fact
that their high-level production values were achieved under resting cell conditions, which
are not usually employed in the fermentation industry.
4.2 Biocatalytic approaches: enzymatic synthesis of lactobionic acid.

Considerable research efforts have focused in recent years on the enzymatic production
of LBA (Table 6 summarizes these enzymatic approaches). In these enzymatic systems,
refined lactose is oxidized by specific biocatalytic cascades employing enzymes such as
oxidoreductoses (Nordkvist et al., 2007; Satory et al., 1997; Van Hecke et al., 2009).
However, these systems are costly processes, as they usually rely on the redox mediator or
cofactor regeneration, additionally associated with complex operational upstream steps
(Ludwig et al., 2004; Maischberger et al., 2008; Splechtna et al., 2001; Van Hecke et al.,
2009). Both enzyme production and further purification stages prior to the biocatalysis
process clearly constitute additional drawbacks seeing as these upstream steps are costly,
time-consuming and laborious. Figure 5B provides an overview of the different steps
involved in these biocatalytic approaches, from the cultivation of the source of the enzyme
to the purification of LBA. Comparatively, enzymatic catalysis usually displays higher LBA
titers, production yields and volumetric productivities (Table 5) than those obtained under
microbial fermentation (Table 6). Despite the high LBA productivity achieved, enzymatic
approaches also feature a lack of selectivity and long-term stability in complex media.
Another limitation is the loss in enzymatic activity throughout the biocatalysis process
which may impair not only LBA synthesis at later reaction times, but also its industrial
implementation.
As Table 6 shows, several approaches for LBA enzymatic synthesis involving different
carbohydrate oxidases have been reported to date. Dhariwal et al. (2006) studied the
production of LBA by cellobiose dehydrogenase (purified from S. rolfsii) in an integrated
electrochemical regeneration process using electricity instead oxygen as the final electron
acceptor, removing the need for a second enzyme (laccase) in the regeneration system. This
biocatalytic system achieved a LBA volumetric productivity of 1.8 g/L.h in a 25-mL
electrolysis cell using ABTS (2,2’-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid)) as the
redox mediator. However, the lack of information on commercial electrolysis flow cells as
well as the low LBA productivity value obtained in comparison to other biocatalytic
approaches constitute continuing practical drawbacks for the successful industrial
implementation of this promising biocatalytic system (Table 6).
Cellobiose dehydrogenase is an extracellular hemoflavoenzyme produced by several
wood-degrading fungi such as S. rolfsii (Baminger et al., 2001b) which utilizes oxygen as the
terminal electron acceptor, thus leading to the formation of hydrogen peroxide, a reaction
by-product that is detrimental to enzyme activity. Nevertheless, this latter drawback can be
overcome by the addition of redox mediators (electron acceptors) such as ABTS or DCIP
(2,6-dichloroindophenol), which are able to accept electrons from cellobiose
dehydrogenase. Likewise, the combination of a regenerating enzyme such as laccase enables
the re-oxidation of these redox mediators in a bi-enzymatic cascade (Fig. 6). Baminger et al.
(2001a) thus achieved the production of LBA from refined lactose by cellobiose
dehydrogenase in a biocatalytic approach in which the presence of an electron acceptor
(ABTS) was combined with a regenerating enzyme such as laccase (purified from Trametes
pubescens). In this biotransformation system, complete lactose bioconversion (100%) and
an impressive LBA volumetric productivity of 18 g/L.h were achieved in a batch reactor (50-
mL) using DCIP as the redox mediator (Baminger et al., 2001a).
Table 6. Summary of research studies on the enzymatic production of LBA.
Bioconversion strategy and Reaction time Lactobionic titer Yield L/LBA
Enzymes a Substrate Redox mediator Productivity (g/L.h) Reference
operating mode (h) (g/L) (%)
Carbohydrate oxidase (M. Pilot-scale batch reactor

Whey permeate n.u. 12 49 4 98 Hua et al. (2007)
nivale) + catalasec (600-L)
Carbohydrate oxidase (M.

Whey permeate n.u. - 5 - - 90 Budtz et al. (2007)
nivale) + catalasec
Carbohydrate oxidase (M. Nordkvist et al.

Lactose n.u. Batch reactor (1-L) 10 49 4.9 98
nivale) + catalasec (2007)
Cellobiose dehydrogenase (S.

Lactose DCIP Batch reactor (3-L) 10 72 7 100 Ludwig et al. (2004)
rolfsii) + laccase (T. pubescens)
Cellobiose dehydrogenase (S. Baminger et al.

Lactose DCIP Batch reactor (50-mL) ~4 72 18 100
rolfsii) + laccase (T. versicolor) (2001a)
Integrated process:
Cellobiose dehydrogenase
electrochemical enzyme Dhariwal et al.
(Trametes versicolor) + laccase Lactose ABTS - - 1.8 -
regeneration in an (2006)
(T. versicolor)
electrolysis cell (25-mL)
Cellobiose dehydrogenase (S. Lactose, Splechtna et al.

DCIP Batch reactor (0.1-L) 2.5 40 16 96
rolfsii) + laccase (T. pubescens) monosaccharides, GOS (2001)
Cellobiose dehydrogenase (S. Lactose, Maischberger et al.

DCIP Batch reactor (6-L) 9 55 6 100
rolfsii) + laccase (T. pubescens) monosaccharides, GOS (2008)
Table 6. (Continued).
Bioconversion strategy and Reaction time Lactobionic titer Yield L/LBA
Enzymes a Substrate Redox mediator Productivity (g/L.h) Reference
operating mode (h) (g/L) (%)
Glucose-fructose
Lactose n.u. Batch reactor (20-mL) 60 157 2.6 90 Satory et al. (1997)
oxidoreductase (Z. mobilis) b
Batch membrane reactor (0.2-

Cellobiose dehydrogenase (S. Van Hecke et al.
Lactose ABTS L) with bubble-free 6 18 3 100
rolfsii) + laccase (T. pubescens) (2009a)
oxygenation
Cellobiose dehydrogenase Batch membrane reactor (0.2

Van Hecke et al.
(Myriococcum thermophilum) Lactose ABTS L) with bubble-free 7 18 2.6 100
(2009b)
+ laccase (T. pubescens) oxygenation
Cellobiose dehydrogenase (S. Discontinuous membrane Van Hecke et al.

Lactose ABTS 12 37 3.1 97
rolfsii) + laccase (T. pubescens) reactor (20-L) (2011)
Lactose-oxidizing enzyme
Lactose n.u. Batch reactor (2-mL) 7 100 14 100 Kiryu et al. (2008)
(Paraconiothyrium sp. KD-3)
Lactose-oxidizing enzyme
Murakami et al.
(Paraconiothyrium sp. KD-3) + Lactose n.u. Batch reactor (2-L) 10-20 100-150 9-11 100
(2008)
catalase (Aspergillus niger)
-: not available; n.u.: not used; LBA: lactobionic acid; L: lactose; DCIP: 2,6-dichloroindophenol; ABTS: 2,2’-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid); GOS: galacto-oligosaccharides.
a The source of the enzyme is shown in parenthesis.
b Dithiothreitol and bovine serum albumin were additionally used as stabilizers.
c Commercial catalase enzyme: Catazyme® from Novozymes A/S.
A
Permeate LBA purification
Retentate
Microfiltration
Microbial
Product recovery
fermentation
Microbial Downstream processing
fermentation
B Enzyme
purification
Cellobiose
LBA purification
Permeate
dehydrogenase
Retentate
Ultrafiltration
Centrifugation
Fungal Product recovery
fermentation Laccase
Upstream processing Enzyme Downstream processing

catalysis
Figure 5. Schematic diagrams illustrating processes for the production of LBA via microbial production (A) and
enzymatic synthesis (B).
Flavo- Redox Laccase ox H20

enzyme ox mediator red
Lactose red
Flavo- Redox Laccase red 02

enzyme red mediator ox
Lactobionic acid ox
Figure 6. Continuous laccase-catalyzed regeneration of redox mediators that are employed in the oxidation of
lactose by a flavo-enzyme (i.e. cellobiose dehydrogenase).
Another approach involving LBA production by cellobiose dehydrogenase was studied by

Ludwig et al. (2004), who reported the inactivation of the enzymatic system under
increasing dissolved oxygen tension conditions, suggesting a possible deleterious
combination of aeration and shearing forces on the stability of cellobiose dehydrogenase
and laccase. Another drawback that emerged was the need to remove the redox mediators
from the system after biocatalysis by means of adsorption due to their toxicity (Ludwig et
al., 2004), a feature commonly found in most cell-free biocatalytic processes using purified
hemoflavoenzymes (Table 6). In terms of volumetric productivity, this biotransformation

system, featuring the continuous laccase-catalyzed regeneration of the electron acceptors,
displayed a LBA volumetric productivity of 7 g/L.h with complete lactose bioconversion
(100%) after 10 h (Ludwig et al., 2004). The inactivation of this bi-enzymatic system under
high aeration conditions was overcome using a bubble-free oxygenation strategy (Van
Hecke et al, 2009a, 2009b, 2011). These authors studied the discontinuous production of
LBA by cellobiose dehydrogenase in a 20-L membrane reactor, obtaining a volumetric
productivity of 3.1 g/L.h with a yield of 97% (Van Hecke et al., 2011). Moreover, the
modeling and further experimental corroboration of this biotransformation approach under
continuous conversion resulted in good enzyme stability (~80% of the initial activity) for 3
days, although long-term operation (>3 days) under non-sterile conditions was unsuccessful
(Van Hecke et al., 2011).
Residual lactose from a transgalactosylation process for galacto-oligosaccharide
production has also been submitted to a biotransformation process to yield LBA by
cellobiose dehydrogenase in an enzymatic regeneration method with productivities ranging
from 6 to 16 g/L.h, achieving 96 to 100% lactose bioconversion (Table 6) (Maischberger et
al., 2008; Splechtna et al., 2001).
The glucose-fructose oxidoreductase enzyme from Z. mobilis has been employed as a
biocatalyst for LBA synthesis under a continuous system (Satory et al., 1997). However, the
use of a glucose-fructose oxidoreductase to oxidize lactose and fructose results into two
products, LBA and sorbitol, respectively. Another disadvantage of this bi-substrate system is
its low affinity for lactose, as the Km value of the enzyme for lactose is 1.2 M, which is 80-fold
higher than that corresponding to its physiological substrate (glucose). Owing to the
presence of reaction stabilizers, the glucose-fructose oxidoreductase was able to retain its
biocatalytic activity for more than 150 h. This latter feature was used for the continuous
production of LBA in an ultrafiltration membrane reactor with a productivity of 4.6 g/L.h.,
which was 1.8-fold higher than that obtained under batch enzymatic conditions (2.6 g/L.h)
(Satory et al., 1997).
Nordkvist et al. (2007) studied the production of LBA from lactose by a carbohydrate
oxidase which was cloned from Microdochium nivale and expressed in Fusarium venenatum.
This carbohydrate oxidase is a flavoprotein which is able to oxidize several carbohydrates
and transfer the electrons to molecular oxygen, thus producing hydrogen peroxide that is
subsequently removed by an added catalase (Budtz et al., 2007; Hua et al., 2007; Nordkvist
et al., 2007). The addition of commercial catalase clearly constitutes a drawback for this
biocatalytic approach despite not requiring any redox mediator. A LBA volumetric
productivity 4.9 g/L.h with a yield of 98% was thus achieved in a batch enzymatic reaction
performed at 38ºC under 0.2 Lpm continuous gas (mixture of N2/O2) sparging (Nordkvist et
al., 2007). However, this biocatalytic system featured a decreasing production rate after 6 h
due to enzyme deactivation resulting from the slow addition of 2 M NaOH aimed at
maintaining an operational pH of 6.4 (Nordkvist et al., 2007; Villadsen, 2007). The additional
non-ideal mixing found by Nordkvist et al. (2007) in a 1-L stirred tank bioreactor was
overcome by Hua et al. (2007) employing a pilot-scale batch reactor (600-L) equipped with
a rotary jet head system which was able to provide the necessary mixing and mass transfer
conditions (Villadsen, 2007). This pilot-scale system yielded a LBA titer of up to 49 g/L from
whey permeate in 12 h, demonstrating the scalable potential of this biocatalytic process
(Hua et al., 2007).
Murakami et al. (2008) obtained complete lactose conversion (100%) employing a
purified extracellular lactose-oxidizing enzyme from Paraconiothyrium sp. KD-3 at 40ºC, in
which 100-150 g/L of lactose were completely converted to LBA within 10-20 h at a rate of
9-11 g/L.h (Murakami et al., 2008). Sugars such as arabinose, glucose, xylose, galactose and
xylobiose as well as cellooligosaccharides may also be oxidized by this lactose-oxidizing
enzyme which is a carbohydrate:acceptor oxidoreductase in terms of its biocatalytic
specificity (Kiryu et al., 2008). LBA production has been also approached employing an
immobilized purified lactose-oxidizing enzyme from Paraconiothyrium sp. KD-3 on a cation
exchange resin, which was able to oxidize up to 185 g/L of lactose (Nakano et al., 2010).
However the continuous repeated batch processing of the immobilized enzyme resulted in a
decrease in biocatalytic activity due to enzymatic inactivation by the hydrogen peroxide
produced during the reaction (Nakano et al., 2010).
In an industrial context, an enzymatic mixture called LactoYIELD was launched onto the
market in 2009 as a result of the strategic alliance achieved between the Danish companies
Chr. Hansen and Novozymes A/S in 2002 (Novozymes, 2009). This standardized enzymatic
mixture allows cheese manufacturers to convert lactose from cheese whey into LBA. The
resulting reaction product, calcium lactobionate, can thus be used as an ingredient in the
formulation of dairy products such as cheeses or desserts in order to increase adhesive
gelling properties, reduce water loss and replace skim milk powder. Furthermore, it can
become an interesting source for calcium supplementation of dairy products, providing
higher solubility and stability properties without the off-taste resulting from other calcium
sources (Novozymes, 2009).
5. Downstream processing of lactobionic acid.

The elucidation of a downstream processing methodology after the bioconversion
process that is both suitable and feasible could guarantee the successful implementation of
LBA bio-production on an industrial scale. Considering that the recovery process depends
primarily on the nature of the matrix employed for LBA production, media components and
metabolites from the bioconversion broth could constitute a drawback when purifying the
desired target compared to the lesser effort required for downstream processing of LBA
produced after catalytic approaches based on refined lactose.
The past few years have seen the development of new approaches for the downstream
processing of LBA; Table 7 summarizes different downstream processing approaches for
LBA purification. In particular, several attempts have been made to separate and purify LBA
after bioconversion processes. Subsequent to cell harvesting, recovery of LBA from the
bioconversion broth could thus be accomplished by conventional methods involving
precipitation, extraction or adsorption using ion-exchange resins (Pedruzzi et al., 2008).
Among the purification methods that have been explored, ion-exchange separation of LBA
has been chosen as the preferred downstream processing approach (Table 7). Higher
recovery yields were thus achieved under LBA titers of 40-50 g/L through ion-exchange,
whereas titers above 100 g/L might need innovative downstream processing approaches
such as simulated moving bed technology (Table 7).
Table 7. Summary of downstream approaches employed for the purification and recovery of LBA.
LBA concentration
Approach Matrix Source Recovery (%) Reference
(g/L)
Crystallization + ion
Synthetic Jones and Ho
exchange Synthetic 200 79
solution (2002)
chromatography
Ion exchange Synthetic Pedruzzi et al.

Synthetic 50 100
chromatography solution (2008)
Ion exchange Reaction Splechtna et al.

Enzymatic synthesis 40 100
chromatography solution (2001)
Bioconversion using
Simulated moving bed Bioconversio Borges da Silva
permeabilized cells of 125 100
technology n broth et al. (2011)
Z. mobilis
Bioconversion using
Bioconversio Peretti et al.
Electrodialysis permeabilized cells of 20 38.7
n broth (2009)
Z. mobilis
The first attempt to purify LBA after microbial fermentation was a reactive extraction
method proposed by Lockwood and Stodola (1950) through precipitation with ethanol.
Murakami et al. (2003) also added ethanol to the culture supernatant after fermentation,
resulting in 98% LBA recovery as precipitate. Crystallization via the addition of ethanol has
also been proposed as a purification step after LBA production by Armarego and Chai
(2009). Magariello (1956) reported the purification of LBA from an electrolytic reaction
system through a first evaporation process to obtain a thick syrup which was then
dehydrated by distillation with dioxane and toluene, finally obtaining a dry residue that was
further crystallized. Alternatively, the process proposed by Magariello (1956) comprised
passing the crude electrolytic reaction solution through acidic cation- and weakly basic
anion-exchange resins, recovering the aqueous LBA solution effluent and spray-drying it.
Jones and Ho (2002) proposed an optimized method for LBA separation combining a
prior evaporation stage (70ºC for 45 min), a crystallization step through a precipitation
method and a final ion-exchange process. The resultant precipitate in the precipitation stage
was passed through one or more ion-exchange resins before a final freeze-drying process,
resulting in a recovery yield of 79% (Table 7).
The selection of resins for ion-exchange separation of LBA (Pedruzzi et al., 2008) and the
use of simulated moving bed technology have also been assessed (Borges da Silva et al.,
2011). Specifically, Pedruzzi et al. (2007, 2008) studied different chromatographic systems
to perform the separation of quaternary mixture from an enzymatic process carried out by
the glucose-fructose oxidoreductase enzyme system. These authors selected a suitable
stationary phase to perform the separation of the quaternary mixture (lactose, fructose,
sorbitol and LBA) and established the valid parameters to predict not only the quaternary
separation by ion resins using water as the eluent, but also the dynamic behaviour of fixed
and moving adsorption beds (Pedruzzi et al., 2008). The separation of LBA from a glucose-
fructose oxidoreductase enzyme reaction through the use of electrodialysis technique was
additionally carried out by Peretti et al. (2009). However, the low yield obtained (38.7%)
may relegate this methodology as a LBA downstream processing methodology.
LBA recovery after production processes will undoubtedly be the subject of intensive
future research. Developing innovative solutions for cost-effective manufacturing of LBA
obviously involves integrated fermentation-separation processes. In general, integrated
fermentation-separation processes may improve system productivities as well as producing
a relatively pure product stream, thus further simplifying downstream processing (Pollard
and Woodley, 2007). In Situ Product Recovery (ISPR) techniques, already implemented for
numerous organic acids, may likewise constitute a promising strategy for LBA separation.
6. Biotechnological role and future prospects

In this review, several of the challenges and current commercial applications of LBA have
been discussed along with the potential perspectives regarding its particular role as an
emerging high value-added organic acid. So far, although LBA is not yet biotechnologically
produced on an industrial level, the bio-production of LBA from an inexpensive feedstock
such as cheese whey may suggest not only a sustainable, but also a cost-effective alternative
for obtaining this high value-added compound. Likewise, the disposal and treatment of
cheese whey continues to be a crucial challenge for the dairy industry, not only from the
environmental, but also from the economic point of view due to the growing costs of
disposal and waste treatment scenarios. Moreover, microbial production of LBA may
overcome the major drawbacks associated with chemical and enzymatic approaches such as
complex and time-consuming upstream steps. In the foreseeable future, biotechnological
production of LBA can be expected to make the transition to the industrial sphere. However,
metabolic engineering may constitute a key future issue to overcome rate-limiting steps for
the microbial production of LBA, introducing new functions into a host strain or improving
LBA production by wild-type strains through enzyme overexpression. In addition, more
straightforward downstream processes could also contribute to bioprocess development to
meet the foreseeable demand, as well as to reduce the costs associated with bio-based
production.
As to LBA applications, both functional foods and nutraceuticals constitute interesting
areas in which LBA may undoubtedly play a fundamental role over the coming years, given
that, as already mentioned, this polyhydroxy bionic acid provides unique health-promoting
functions. However, new data is needed to complete its comprehensive risk assessment
before drafting the opinion of panel members on food additives prior to the final adoption
by the EFSA of its standby status as a food additive in Europe.
Aside from its role in the food sector, LBA has a great potential as a useful compound in
the treatment of hepatic diseases. LBA may thus lead the way to novel therapies based on
hepatic targeting of ASGPR, seeing as the potential applications of LBA as a platform for
biomaterials or scaffolds in tissue engineering (Peng et al., 2007) or as bio-functionalization
agent of nanoparticles used for the treatment of recalcitrant diseases such as liver cancer
are really promising (Jain and Jain, 2010; Mi et al. 2007). Furthermore, bio-functionalization
of drug delivery systems through LBA affords a promising way to provide sustained release
of drugs targeting liver cancer cells (Duan et al., 2011; Huang et al., 2011; Varshosaz et al.,
2012). LBA will undoubtedly play a major role in these novel therapies as the core molecule,
expanding its usefulness beyond its technological role as an additive in food processing.
Acknowledgements
The authors are grateful for funding from the Spanish Ministry of Science and Innovation
through project MEC-CTQ2010-14918.
References
Abbadi A, Gotlieb KF, Meiberg JBM, Peters JA, Van Bekkum H. New Ca-sequestering materials
based on the oxidation of the hydrolysis products of lactose. Green Chem 1999;1:231-
235.
Affertsholt T. Market developments and industry challenges for lactose and lactose
derivatives. Presentation at IDF Symposium Lactose and its derivatives. 2007.
Available via: http://lactose.ru/present/1Tage_Affertsholt-Allen.pdf. Accessed 23 July
2012.
Agarkhed M, Srivastava A, Goldstein J. Lyophilized formulations of anti-egfr antibodies.
United States Patent Application Pub. No.: US 2010/0158925 A1; 2010.
Agustí R, París G, Ratier L, Frasch ACC, de Lederkremer RM. Lactose derivatives are
inhibitors of Trypanosoma cruzi trans-sialidase activity toward conventional
substrates in vitro and in vivo. Glycobiology 2004;14:659-70.
Alex SM, Rekha MR, Sharma CP. Spermine grafted galactosylated chitosan for improved
nanoparticles mediated gene delivery. Int J Pharm 2011;410:125-37.
Alonso S, Rendueles M, Díaz M. Efficient lactobionic acid production from whey by
Pseudomonas taetrolens under pH-shift conditions. Bioresour Technol 2011;102:9730-
6.
Alonso S, Rendueles M, Díaz M. Role of dissolved oxygen availability on lactobionic acid
production from whey by Pseudomonas taetrolens. Bioresour Technol 2012a;109:140-
7.
Alonso S, Rendueles M, Díaz M. Physiological heterogeneity of Pseudomonas taetrolens
during lactobionic acid production. Appl Microbiol Biotechnol 2012b;96:1465-77.
Alonso S, Rendueles M, Díaz M. Physiological states of Pseudomonas taetrolens cultured in
whey media under different oxygen and pH conditions. New Biotechnol
2012c;29,Supplement:89-90.
Alonso S, Rendueles M, Díaz M. Selection method of pH conditions to establish Pseudomonas
taetrolens physiological states and lactobionic acid production. Appl Microbiol
Biotechnol 2012d. doi: 10.1007/s00253-012-4607-x
Anzai Y, Kim H, Park JY, Wakabayashi H, Oyaizu H. Phylogenetic affiliation of the

pseudomonads based on 16s rRNA sequence. Int J Syst Evol Microbiol 2000;50:1563-
89.
Araki K, Fujii K, Ueno H, Fujii N, Takahane M. New antibacterial agent composition. Japan
Patent Application Pub. No.: JP2006104164; 2006.
Armarego WL, Chai CLL. Purification of biochemicals and related products. In: Armarego
WL, Chai CLL, editors. Purification of laboratory chemicals (Sixth edition). Oxford:
Elsevier Inc; 2009. p. 577-708.
Bahadur R, Lee SM, Yoo ES, Choi JH, Ghim HD. Glycoconjugated chitosan stabilized iron oxide
nanoparticles as a multifunctional nanoprobe. Mater Sci Eng C 2009;29:1668-73.
Baldwin C, Akashe A, Dinwoodie R, Koka R, West LG, Kortum O. Use of siderophores and
organic acid to retard lipid oxidation. United States Patent Application Pub. No.: US
2004/0170728 A1; 2004.
Baldwin C, Akashe A, Zeller AL, Marquez-Sanchez M. Mineral complexes of lactobionic acid
and method of using for mineral fortification of food products. United States Patent
Application Pub. No.: US 2007/0026110 A1; 2007.
Baminger U, Ludwig R, Galhaup C, Leitner C, Kulbe KD, Haltrich D. Continuous enzymatic
regeneration of redox mediators used in biotransformation reactions employing
flavoproteins. J Mol Catal B-Enzym 2001a;11:541-50.
Baminger U, Subramanian SS, Renganathan V, Haltrich D. Purification and characterization
of cellobiose dehydrogenase from the plant pathogen Sclerotium (Athelia) rolfsii. Appl
Environ Microbiol 2001b;67:1766-74.
Bean RC, Hassid WZ. Carbohydrate oxidase from a red alga, Iridophycus flaccidum. J Biol
Chem 1956;218:425-36.
Béguin SS, Dol F, Hemker HC. Influence of lactobionic acid on the kinetics of thrombin in
human plasma. Semin Thromb Hemost 1991;17:126-8.
Belzer FO, D’Alessandro AM, Hoffmann RM, Knechtle SJ, Reed A, Pirsch JD, Kalayoglu M,
Sollinger HW. The use of UW solution in clinical transplantation. A 4-year experience.
Ann Surg 1992;215:579-85.
Bentley R, Slechta L. Oxidation of mono- and disaccharides to aldonic acids by Pseudomonas
species. J Bacteriol 1960;79:346-55.
Bize C, Blanzat M, Rico-Lattes I. Self-assembled structures of catanionic associations: How to
optimize vesicle formation?. J Surfact Deterg 2010;13:465-73.
Borges da Silva E, Pedruzzi I, Rodrigues AE. Simulated moving bed technology to improve
the yield of the biotechnological production of lactobionic acid and sorbitol.
Adsorption 2011;17:145-58.
Briden ME, Green BA. Topical exfoliation - clinical effects and formulating considerations. In:
Draelos ZD, Thaman LA, editors. Cosmetic formulation of skin care products. New
York; Taylor and Francis Group; 2006. p. 237-50.
Brommage R, Binacua C, Antille S, Carrié AL. Intestinal calcium absorption in rats is
stimulated by dietary lactulose and other resistant sugars. J Nutr 1993;123:2186-94.
Bucek W, Connors WM, Cort WM, Roberts HR. Evidence for the formation and utilization of
lactobionic acid by Penicillium chrysogenum. Arch Biochem Biophys 1956;63:477-8.
Budtz P, Vindelov JT, Nielsen PM, Ashie I, Nordkvist M. Enzymatic process for obtaining
increased yield of lactobionic acid. United States Patent Application Pub. No.: US
20070154595; 2007.
Cavallo G, Stagni M, Sodo E. Ophthalmic composition based on lactobionic acid useful for
reducing corneal edema and inflammation. European Patent Application Pub. No.:
EP2494954A2; 2012.
Charloux C, Paul M, Loisance D, Astier A. Inhibition of hydroxyl radical production by
lactobionate, adenine, and tempol. Free Radical Bio Med 1995;19:699-04.
Chen H, Li M, Wan T, Zheng Q, Cheng M, Huang S, Wang Y. Design and synthesis of dual-
ligand modified chitosan as a liver targeting vector. J Mater Sci: Mater Med
2012a;23:431-41.
Chen Z, Xiao L, Liu W, Liu D., Xiao YY, Chen J, Liu X, Li WD, Li W, Cai BC. Novel materials
which possess the ability to target liver cells. Expert Opin. Drug Deliv. 2012b;9:649-
656.
Chia YN, Latusek MP, Holles JH. Catalytic wet oxidation of lactose. Ind Eng Chem Res
2008;47:4049-55.
Choi YH, Liu F, Park JS, Kim SW. Lactose-poly(ethylene glycol)-grafted poly-L-lysine as
hepatoma cell-targeted gene carrier. Bioconjugate Chem 1998;9:709-18.
Choi ES, Park JY, Baek MJ, Xu W, Kattel K, Kim JH, Lee JJ, Chang Y, Kim TJ, Bae JE, Chae KS, Suh
KJ, Lee GH. Water-soluble ultra-small manganese oxide surface doped gadolinium
oxide (Gd2O3@MnO) nanoparticles for MRI contrast agent. Eur J Inorg Chem
2010;28:4555-60.
Choi L, Kwak SJ, You SJ, Chun HJ, Kim HL, Shim YB, Kim MS, Park KD. Characterization and
hepatocytes adhesion of galactosylated poly(D,L-lactic-co-glycolic acid) surface.
Macromol Res 2012;20:93-100.
Chung TW, Yang J, Akaike T, Cho KY, Nah JW, Kim SI, Cho CS. Preparation of
alginate/galactosylated chitosan scaffold for hepatocyte attachment. Biomaterials
2002;23:2827-34.
Coombe DR, Kett WC. Heparin mimetics. In: Lever R, Mulloy B, Page CP, editors. Heparin – A
century of progress. Handbook of experimental pharmacology. Berlin: Springer-Verlag
Berlin Heidelberg; 2012. p. 361-83.
D’Alessandro AM, Hoffman RM, Southard JH. Solution development in organ preservation:
the University of Wisconsin perspective. Transplant Rev 1999;13:67-77.
Decker A, Graber EM. Over-the-counter acne treatments. J Clin Aesthet Dermatol 2012;5:32-
40.
Demain AL. The business of biotechnology. Ind Biotechnol 2007;3:269-83.
Dhariwal A, Mavrov V, Schroeder I. Production of lactobionic acid with process integrated
electrochemical enzyme regeneration and optimization of process variables using
response surface methods (RSM). J Mol Catal B-Enzym 2006;42:64-9.
Díaz M, Herrero M, García LA, Quirós C. Application of flow cytometry to industrial microbial
bioprocesses. Biochemical Eng J 2010;48:385-407.
Dong L, Zuo L, Xia S, Gao S, Zhang C, Chen J, Zhang J. Reduction of liver tumor necrosis factor-
α expression by targeting delivery of antisense oligonucleotides into Kupfer cells
protects rats from fulminant hepatitis. J Gene Med 2009;11:229-39.
Druliolle H, Kokoh KB, Beden B. Selective oxidation of lactose to lactobionic acid on lead-
adatoms modified platinum electrodes in Na2CO3 + NaHCO3 buffered medium. J
Electroanal Chem 1995;385:77-83.
Duan C, Gao J, Zhang D, Jia L, Liu Y, Zheng D, Liu G, Tian X, Wang F, Zhang Q. Galactose-
decorated pH-responsive nanogels for hepatoma-targeted delivery of oridonin.
Biomacromolecules 2011;12:4335-43.
Fan J, Shang Y, Yuan Y, Yang J. Preparation and characterization of chitosan/galactosylated
hyaluronic acid scaffolds for primary hepatocytes culture. J Mater Sci: Mater Med
2010;21:319-27.
Faergemand M, Gilleladen C, Qvist KB. Method for producing an acidified milk product.
United States Patent Application Pub. No.: US 20120045546 A1; 2012.
FDA. Code of Federal Regulations, Title 21, 21 CFR 172.720. US Food and Drug
Administration, 2011. Available via:
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?fr=172.72
0. Accessed 23 July 2012.
Feng ZQ, Chu X, Huang NP, Wang T, Wang Y, Shi X, Ding Y, Gu ZZ. The effect of nanofibrous
galactosylated chitosan scaffolds on the formation of rat primary hepatocyte
aggregates and the maintenance of liver function. Biomaterials 2009;30:2753-63.
Feng W, Xiao J, Li L, Ji P. Protein adsorption on functionalized carbon nanotubes with a
lactobionic amide amphiphile. Ind Eng Chem Res 2011;50:11608-3.
Fisher E, Meyer J. Oxydation des milchzuckers. Ber Dtsch Chem Ges 1889;22:361-4.
Gänzle MG, Haase G, Jelen P. Lactose: crystallization, hydrolysis and value-added derivatives.
Int Dairy J 2008;18:685-94.
Gao S, Chen J, Xu X, Ding Z, yang YH, Hua Z, Zhang J. Galactosylated low molecular weight
chitosan as DNA carrier for hepatocyte-targeting. Int J Pharm 2003;255:57-68.
Gerling KG. Large scale production of lactobionic acid-use and new applications. Int Dairy
Fed 1998;9804:251-6.
Gerling KG, Joisten S, Wendler K, Schreer C. Lactobionic acid amide compositions and their
use. United States Patent Application Pub. No.: 5525333; 1996.
Gerling KG, Rau H, Wendler K, Schwarz P, Uhlig K. Corrosion preventing compositions
comprising lactobionic acid amides. United States Patent Application Pub. No.: US
005779939 A; 1998.
Giorgi ME, Ratier L, Agustí R, Frasch ACC, de Lederkremer RM. Synthesis of PEGylated
lactose analogs for inhibition studies on T. cruzi trans-sialidase. Gycoconj J
2010;27:549-59.
Green B. After 30 years… The future of hydroxyacids. J Cosmet Dermatol 2005;4:44-5.
Green BA, Edison BL, Sigler ML. Antiaging effects of topical lactobionic acid: results of a
controlled usage study. Cosmet Dermatol 2008;21:76-82.
Green BA, Yu RJ, Van Scott EJ. Clinical and cosmeceutical uses of hydroxyacids. Clin Dermatol
2009;27:495-1.
Grimes PE, Green BA, Wildnauer RH, Edison BL. The use of polyhydroxy acids (PHAs) in
photoaged skin. Cutis 2004;73:3-13.
Goldstein JD, Thatcher DR. Stabilized formulations of gamma interferons. United States
Patent Application Pub. No.: 4895716;1990.
Guimarães PMR, Teixeira JA, Domingues L. Fermentation of lactose to bioethanol by yeasts
as part of integrated solutions for the valorisation of cheese whey. Biotechnol Adv
2010;28:375-84.
Gutiérrez LF, Hamoudi S, Belkacemi K. Lactobionic acid: a value-added lactose derivative for
food and pharmaceutical applications. Int Dairy J 2012;26:103-111.
Guo XL, Yang KS, Hyun JY, Kim WS, Lee DH, Min KE, Park LS, Seo KH, Kim YI, Cho CS, Kang IK.
Morphology and metabolism of Ba-alginate-encapsulated hepatocytes with
galactosylated chitosan and poly(vinyl alcohol) as extracellular matrices. J Biomater
Sci Polymer Edn 2003;14:551-65.
Guo R, Yao Y, Cheng G, Wang SH, Li Y, Shen M, Zhang Y, Baker Jr. JR, Wang J, Shi X. Synthesis
of glycoconjugated poly(amindoamine) dendrimers for targeting human liver cancer
cells. RSC Adv 2012;2:99-102.
Hachem JP, Roelandt T, Schürer N, Pu X, Fluhr J, Giddelo C, Man MQ, Crumrine D, Roseeuw D,
Feingold KR, Mauro T, Elias PM. Acute acidification of stratum corneum membrane
domains using polyhydroxyl acids improves lipid processing and inhibits degradation
of corneodesmosomes. J Invest Dermatol 2010;130:500-10.
Hart NA, Leuvenink HGD, Ploeg RJ. New solutions in organ preservation. Transplant Rev
2002;16:131-41.
Hirakata Y, Kaku M, Tomono K, Tateda K, Furuya N, Matsumoto T, Araki R, Yamaguchi K.
Efficacy of erythromycin lactobionate for treating Pseudomonas aeruginosa bacteremia
in mice. Antimicrob Agents Chemother 1992;36:1198-03.
Hope A, Miljan E, Sinden J. Cellular compositions for use in therapy. United States Patent
Application Pub. No.: US 2012/0076854 A1;2012.
Hua L, Nordkvist M, Nielsen PM, Villadsen J. Scale-up of enzymatic production of lactobionic
acid using the rotary jet head system. Biotechnol Bioeng 2007;97:842-9.
Huang J, Gao F, Tang X, Yu J, Wang D, Liu S, Li Y. Liver-targeting doxorubicin-conjugated
polymeric prodrug with pH-triggered drug release profile. Polym Int 2010;59:1390-6.
Huang FW, Yang J, Feng J, Zhuo RX, Zhang XZ. Design of hepatocyte-targeted gene transfer
vector and its in vitro transfer of tumor-suppressor p53 gene. J Mater Chem
2011;21:3585-96.
Hoppensteadt DA, Ahsan A, Walenga JM, Fareed J, Schumacher H. Activation of heparin
cofactor II by sulphated lactobionic acid. Blood 1988;72:198a.
Il’ina AV, Varlamov VP. Galactosylated derivatives of low-molecular-weight chitosan:
obtaining and properties. Appl Biochem Micro 2007;43:73-7.
Jain NK, Jain SK. Development and in vitro characterization of galactosylated low molecular
weight chitosan nanoparticles bearing doxorubicin. AAPS Pharm Sci Tech
2010;11:686-97.
Jang YS, Kim B, Shin JH, Choi YJ, Choi S, Song CW, Lee J, Park HG, Lee SY. Bio-based
production of C2-C6 platform chemicals. Biotechnol Bioeng 2012;10:2437-59.
Jiang HL, Kwon JT, Kim YK, Kim EM, Arote R, Jeong HJ, Nah JW, Choi YJ, Akaike T, Cho MH,
Cho CS. Galactosylated chitosan-graft-polyethylenimine as a gene carrier for
hepatocyte targeting. Gene Ther 2007;14:1389-98.
Jiang HL, Kwon JT, Kim EM, Kim YK, Arote R, Jere D, Jeong HJ, Jang MK, Nah JW, Xu CX, Park
IK, Cho MH, Cho CS. Galactosylated poly(ethylene glycol)-chitosan-graft-
polyethylenimine as a gene carrier for hepatocyte-targeting. J Control Release
2008;131:150-7.
Jones WE, Ho Y. Novel techniques for the preparation and crystallization of 4-o-Beta-D-
galactopyranosyl-D-gluconic acid. United States Patent Application Pub. No.: US
2002/0006884 A1; 2002.
Kakasi-Zsurka S, Todea A, But A, Paul C, Boeriu CG, Davidescu C, Nagy L, Kuki Á, Kéki S, Péter
F. Biocatalytic synthesis of new copolymers from 3-hydroxybutyric acid and a
carbohydrate lactone. J Mol Catal B-Enzym 2011;71:22-8.
Kang IK, Kim GJ, Kwon OH, Ito Y. Co-culture of hepatocytes and fibroblasts by
micropatterned immobilization of β-galactose derivatives. Biomaterials
2004;25:4225-32.
Kang IK, Moon JS, Jeon HM, Meng W, Kim YI, Hwang YJ, Kim S. Morphology and metabolites
of Ba-alginate encapsulated hepatocytes with galactosylated poly(allyl amine) and
poly(vinyl alcohol) as extracellular matrices. J Mater Sci-Mater M 2005;16:533-9.
Kekkonen V, Lafreniere N, Ebara M, Saito A, Sawa Y, Narain R. Synthesis and
characterization of biocompatible magnetic glyconanoparticles. J Magn Magn Mater
2009;321:1393-6.
Kim IS, Kim SH. Development of a polymeric nanoparticulate drug delivery system. In vitro
characterization of nanoparticles based on sugar-containing conjugates. Int J Pharm
2002;245:67-73.
Kim IS, Kim SH. Development of polymeric nanoparticles drug delivery systems: evaluation
of nanoparticles based on biotinylated poly(ethylene glycol) with sugar moiety. Int J
Pharm 2003;257:195-203.
Kim TH, Park IK, Nah JW, Choi YJ, Cho CS. Galactosylated chitosan/DNA nanoparticles
prepared using water-soluble chitosan as a gene carrier. Biomaterials 2004;25:3783-
92.
Kim EM, Jeong HJ, Park IK, Cho CS, Moon HB, Yu DY, Bom HS, Sohn MH, Oh IJ.
Asialoglycoprotein receptor targeted gene delivery using galactosylated
polyethylenimine-graft-poly(ethylene glycol): in vitro and in vivo studies. J Control
Release 2005a;108:557-67.
Kim TH, Kim SI, Akaike T, Cho CS. Synergistic effect of poly(ethylenimine) on the
transfection efficiency of galactosylated chitosan/DNA complexes. J Control Release
2005b;105:354-66.
Kim EM, Jeong HJ, Kim SL, Sohn MH, Nah JW, Bom HS, Park IK, Cho CS. Asialoglycoprotein-
receptor-targeted hepatocyte imaging using 99mTc galactosylated chitosan. Nucl Med
Biol 2006;33:529-34.
Kim IY, Seo SJ, Moon HS, Yoo MK, Park IY, Kim, BC, Cho CS. Chitosan and its derivatives for
tissue engineering applications. Biotechnol Adv 2008;26:1-21.
Kim BS, Park IK, Hoshiba T, Jiang HL, Coi YJ, Akaike T, Cho CS. Design of artificial
extracellular matrices for tissue engineering. Prog Polym Sci 2011;36:238-68.
Kimura T. Feed additive for laying hens and feed containing the additive. European Patent
Application Pub. No: EP 1731042 A1; 2006.
Kiryu T, Nakano H, Kiso T, Murakami H. Purification and characterization of a carbohydrate:
acceptor oxidoreductase from Paraconiothyrium sp. that produces lactobionic acid
efficiently. Biosci Biotech Bioch 2008:72:833-41.
Kiryu T, Kiso T, Nakano H, Ooe K, Kimura T, Murakami H. Involvement of Acetobacter
orientalis in the production of lactobionic acid in Caucasian yogurt (“Caspian Sea
yogurt”) in Japan. J Dairy Sci 2009;92:25-4.
Kiryu T, Ooe K, Kimura T, Kiso T, Nakano H, Murakami H. Sugar oxidation profiles of
membrane-bound dehydrogenase from Acetobacter orientalis. Biocatal Agric
Biotechnol 2012a;1:262-3.
Kiryu T, Yamauchi K, Masuyama A, Ooe K, Kimura T, Kiso T, Nakano H, Murakami H.
Optimization of lactobionic acid production by Acetobacter orientalis isolated from
Caucasian fermented milk, “Caspian Sea yogurt”. Biosci Biotech Bioch 2012b;76:361-3.
Knopp D, Tang D, Niessner R. Review: bioanalytical of biomolecule-functionalized
nanometer-sized doped silica particles. Anal Chim Acta 2009;647:14-30.
Klöcking HP, Hoffman A, Fareeed J. Influence of hypersulfated lactobionic acid amides on
tissue plasminogen activator release. Semin Thromb Hemost 1991;17:379-84.
Kluyver AJ, De Ley J, Rijven A. The formation and consumption of lactobionic and
maltobionic acids by Pseudomonas species. A Van Leeuw J Microb 1951;17:1-14.
Koka R, Mehnert DW, Fritsch RJ, Steffan W, Habermeier P, Bradbury AGW, Wolfschoon-
Pombo A, Rose M. Process for manufacturing cheeses and other dairy products and
products thereof. United States Patent Application Pub. No.: US 6916496; 2005.
Kontula P, Suihko ML, Suortti T, Tenkanen M, Mattila-Sandholm T, von Wright A. The
isolation of lactic acid bacteria from human colonic biopsies after enrichment on
lactose derivatives and rye arabinoxylo-oligosaccharides. Food Microbiol 2000;17:13-

22.
Krylov VB, Ustyuzhanina NE, Nifantiev NE. Synthesis of low-molecular-weight carbohydrate
mimetics of heparin. Russ J Bioorg Chem 2011;37:672-706.
Kuusisto J, Tokarev AV, Murzina EV, Roslund MU, Mikkola JP, Murzin DY, Salmi T. From
renewable raw materials to high value-added fine chemicals-catalytic hydrogenation
and oxidation of d-lactose. Catal Today 2007;121:92-9.
Lebaupain F, Salvay AG, Olivier B, Durand G, Fabiano AS, Michel N, Popot JL, Ébel C, Breyton
C, Pucci B. Lactobionamide surfactants with hydrogenated, perfluorinated or
hemifluorinated tails: physical-chemical and biochemical characterization. Langmuir
2006;22:8881-90.
Lencastre Fernandes R, Nierychlo M, Lundin L, Pedersen AE, Puentes Tellez PE, Dutta A,
Carlquist M, Bolic A, Schäpper D, Brunetti AC, Heins AL, Jensen AD, Nopens I, Rottwitt
K, Szita N, van Elsas JD, Nielsen PH, Martinussen J, Sørensen SJ, Lantz AE, Gernaey KV.
Experimental methods and modeling techniques for description of cell population
heterogeneity. Biotechnol Adv 2011;29:575-99.
Levine M, Anderson DQ. Two new species of bacteria causing mustiness in eggs. J Bacteriol
1931:337-47.
Li HP, Cao Z, Xiao HW. Synthesis of lactosylated piperazinyl porphyrins and their
hepatocyte-selective targeting. Med Chem Res 2007;16:28-38.
Li X, Huang Y, Chen X, Zhou Y, Zhang Y, Li P, Liu Y, Sun Y, Zhao J, Wang F. Self-assembly and
characterization of pluronic P105 micelles for liver-targeted delivery of silybin. J Drug
Target 2009;17:739-50.
Lin WJ, Chen TD, Liu CW. Synthesis and characterization of lactobionic acid grafted
pegylated chitosan and nanoparticle complex application. Polymer 2009;50:4166-74.
Lin WJ, Chen TD, Liu CW, Chen JL, Chang FH. Synthesis of lactobionic acid-grafted-pegylated-
chitosan with enhanced HepG2 cells transfection. Carbohydr Polym 2011;83:898-4.
Lockwood LB, Stodola FH. Process of culturing bacteria. United States Patent Application
Pub. No.: US2496297; 1950.
Lu B, Wy DQ, Zheng H, Quan CY, Zhang XZ, Zhuo RX. Galactosyl conjugated N-succinyl-
chitosan-graft-polyethylenimine for targeting gene transfer. Mol BioSyst 2010;6:2529-
38.
Ludwig R, Ozga M, Zámocky M, Peterbauer C, Kulbe K, Haltrich D. Continuous enzymatic
regeneration of electron acceptors used by flavoenzymes: cellobiose dehydrogenase-
catalyzed production of lactobionic acid as an example. Biocatal Biotransfor

2004;22:97-104.
Luo Z, Cai K, Hu Y, Zhao L, Liu P, Duan L, Yang W. Mesoporous silica nanoparticles end-
capped with collagen: redox-responsive nanoreservoirs for targeted drug delivery.
Angew Chem Int Ed 2011;50:640-3.
Luo Z, Cai K, Hu Y, Zhang B, Xu D. Cell-specific intracellular anticancer drug delivery from
mesoporous silica nanoparticles with pH sensitivity. Adv Healthcare Mater
2012;1:321-5.
Magariello ER. Production of lactobionic acid and its δ-lactone. United States Patent
Application Pub. No.: US 2746916; 1956.
Maischberger T, Nguyen TH, Sukyai P, Kittl R, Riva S, Ludwig R, Haltrich D. Production of
lactose-free galacto-oligosaccharide mixtures: comparison of two cellobiose
dehydrogenases for the selective oxidation of lactose to lactobionic acid. Carbohyd Res
2008;343:2140-7.
Malvessi E, Carra S, Pasquali FC, Kern DB, da Silveira MM, Ayub MAZ. Production of organic
acids by periplasmic enzymes present in free and immobilized cells of Zymomonas
mobilis. J Ind Microbiol Biotechnol 2012. doi: 10.1007/s10295-012-1198-6
Meiberg JBM, Bruinenberg PM, Sloots B. A process for the fermentative oxidation of
reducing disaccharides. European Patent Application Pub. No.: EP19900200393; 1990.
Meng Q, Tian L, Wang J. Random amphiphilic copolymeric sub-micro particles as a carrier
shielding from enzymatic attack for peptides and proteins delivery. J Mater Sci: Mater
Med 2012;23:991-8.
Merrill RK, Singh M. Food ingredients and food products treated with an oxidoreductase and
methods for preparing such food ingredients and food products. United States Patent
Application Pub. No.: US 8021704; 2011.
Mi FL, Yu SH, Peng CK, Sung HW, Shyu SS, Liang HF, Huang MF, Wang CC. Synthesis and
characterization of a novel glycoconjugated macromolecule. Polymer 2006;47:4348-
58.
Mi FL, Wu YY, Chiu YL, Chen MC, Sung HW, Yu SH, Shyu SS, Huang MF. Synthesis of a novel
glycoconjugated chitosan and preparation of its derived nanoparticles for targeting
HepG2 cells. Biomacromolecules 2007;8:892-8.
Mirescu A, Prübe U. A new environmental friendly method for the preparation of sugar acids
via catalytic oxidation on gold catalysts. Appl Catal B-Environ 2007;70:644-52.
Miyake T, Sato Y. Process for producing aldonic acids and starch sugars containing aldonic
acids. United States Patent Application Pub. No.: US 3862005; 1975.
Miyamoto Y, Ooi T, Kinoshita S. Production of lactobionic acid from whey by Pseudomonas

sp. LS13-1. Biotechnol Lett 2000;22:427-30.
Murakami H, Kawano J, Yoshizumi H, Nakano H, Kitahata S. Screening of lactobionic acid
producing microorganisms. J Appl Glycosci 2002;49:469-77.
Murakami H, Seko A, Azumi M, Ueshima N, Yoshizumi H, Nakano H, Kitahata S. Fermentative
production of lactobionic acid by Burkholderia cepacia. J Appl Glycosci 2003;50:117-
20.
Murakami H, Kiryu T, Kiso T, Nakano H. Production of aldonic acids from monosaccharides
by washed cells of Burkholderia cepacia and their calcium binding capability. J Appl
Glycosci 2006a;53:277-9.
Murakami H, Seko A, Azumi M, Kiso T, Kiryu T, Kitahata S, Shimada Y, Nakano H. Microbial
conversion of lactose to lactobionic acid by resting cells of Burkholderia cepacia No. 24.
J Appl Glycosci 2006b;53:7-11.
Murakami H, Kiryu T, Kiso T, Nakano H. Production of calcium lactobionate by a lactose-
oxidizing enzyme from Paraconiothyrium sp. KD-3. J Appl Glycosci 2008;55:127-32.
Murzina EV, Tokarev AV, Kordás K, Karhu H, Mikkola JP, Murzin DY. D-lactose oxidation over
gold catalysts. Catal Today 2008;131:385-92.
Na K, Park KH, Kim SW, Bae YH. Self-assembled hydrogel nanoparticles from curdlam
derivatives: characterization, anti-cancer drug release and interaction with hepatoma
cell line (HepG2). J Control Release 2000;69:225-36.
Nakano H, Kiryu T, Kiso T, Murakami H. Biocatalytic production of lactobionic acid. In: Hou
CT, Shaw JF, editors. Biocatalysis and biomolecular engineering. New Jersey: John
Wiley & Sons Inc; 2010. p. 391-404.
Nielsen PM. Non-dairy beverage product comprising calcium lactobionate. United States
Patent Application Pub. No.: US 2007/0281066 A1; 2007.
Nielsen PM. Meat based food product comprising lactobionic acid. United States Patent
Nielsen PM, Hoeier E. Food products containing aldobionic acid. International Patent
Application Pub. No.: WO 2009/007398 A1; 2009.
Nishizuka Y, Kuno S, Hayaishi O. Lactose dehydrogenase, a new flavoprotein. J Biol Chem
1960;235:PC13.
Nishizuka Y, Hayaishi O. Enzymic formation of lactobionic acid from lactose. J Biol Chem
1962;237:2721-8.
Nordkvist M, Nielsen PM, Villadsen J. Oxidation of lactose to lactobionic acid by a

Microdochium nivale carbohydrate oxidase: kinetics and operational stability.
Biotechnol Bioeng 2007;97:694-07.
Novozymes. Press release; 2009. Available via:
http://www.novozymes.com/en/news/news-archive/Pages/45275.aspx. Accessed
23 July 2012.
Oe K, Kimura T. Mineral absorption promoter. Japan Patent Application Pub. No.:
JP2008303208; 2008.
Oe K, Kimura T. Equol production accelerating composition. United States Patent
Oe K, Kimura T. Aging inhibitor for bread. Japan Patent Application Pub. No.: JP2011177121;
2011.
Oe K, Kimura T, Nakano H, Murakami H, Kiso T, Kiryu T. Method for producing lactobionic
acid-containing milk beverage. Japan Patent Application Pub. No.: JP200700092061;
2007.
Oe K, Nishikawa Y, Kimura T, Kiryu T, Kiso T, Murakami H, Nakano H. Oxidation of lactose to
lactobionic acid by acetic acid bacteria. Presentation at the 2nd International
Conference on Acetic Acid Bacteria. 2008.
Ofosu FA, Fareed J, Smith LM, Anvari N, Hoppensteadt DA, Blajchman MA. Inhibition of
factor X, factor V and prothrombin activation by the bis (lactobionic acid amide) LW
10082. Eur J Biochem 1992;203:121-5.
Ono K, Saito Y, Yura H, Ishikawa K, Kurita A, Akaike T, Ishihara M. Photocrosslinkable
chitosan as a biological adhesive. J Biomed Mater Res 2000;49:289-95.
Ortial S, Durand G, Poeggeler B, Polidori A, Pappolla MA, Böker J, Hardeland R, Pucci B.
Fluorinated amphiphilic acid derivatives as antioxidant carriers: a new class of
protective agents. J Med Chem 2006;49:2812-20.
Oskarsson H, Frankenberg M, Annerling A, Holmberg K. Adsorption of novel
alkylaminoamide sugar surfactants at tailor-made surfaces. J Surfact Deterg
2007;10:41-52.
Palepu NR. Alcohol free formulation of argatroban. United States Patent Application Pub.
No.: US 2008/0300272 A1; 2008.
Papagianni M. Organic acids. In: Moo-Young M, editor. Comprehensive Biotechnology, vol. 1.
Oxford: Elsevier Inc; 2011. p. 109-20.
Park YK, Park YH, Shin Bam Choi ES, Park YR, Akaike T, Cho CS. Galactosylated chitosan-
graft-dextran as hepatocyte-targeting DNA carrier. J Control Release 2000;69:97-08.
Park IK, Kim TH, Park YH, Shin BA, Choi ES, Chowdhury EH, Akaike T, Cho CS. Galactosylated
chitosan-graft-poly(ethylene glycol) as hepatocyte-targeting DNA carrier. J Control
Release 2001;76:349-62.
Park IK, Yang J, Jeong HJ, Bom HS, Harada I, Akaike T, Kim SI, Cho CS. Galactosylated chitosan
as a synthetic extracellular matrix for hepatocyte attachment. Biomaterials
2003;24:2331-7.
Park IK, Jiang HL, Cook SE, Cho MH, Kim SI, Jeong HJ, Akaike T, Cho CS. Galactosylated
chitosan (GC)-graft-Poly(vinyl pyrrolidone) (PVP) as hepatocyte-targeting DNA
carrier: in vitro transfection. Arch Pharm Res 2004;12:1284-9.
Pedruzzi I, Malvessi E, Mata VG, Silva EAB, Silveira MM, Rodrigues AE. Quantification of
lactobionic acid and sorbitol from enzymatic reaction of fructose and lactose by high-
performance liquid chromatography. J Chromatogr A 2007;1145:128-32.
Pedruzzi I, Borges da Silva EA, Rodrigues AE. Selection of resins, equilibrium and sorption
kinetics of lactobionic acid, fructose, lactose and sorbitol. Sep Purif Technol
2008;63:600-11.
Pedruzzi I, da Silva EAB, Rodrigues AE. Production of lactobionic acid from lactose/fructose
substrate using GFOR/GL enzymes from Zymomonas mobilis cells: A kinetic study.
Enzyme Microb Tech 2011;49:183-91.
Peng J, Wang K, Tan W, He X, He C, Wu P, Liu F. Identification of live liver cancer cells in a
mixed cell system using galactose-conjugated fluorescent nanoparticles. Talanta
2007;71:833-40.
Peretti FA, Silveira MM, Zeni M. Use of electrodialysis technique for the separation of
lactobionic acid produced by Zymomonas mobilis. Desalination 2009;246:253-7.
Pezzotti F, Therisod M. Enzymatic synthesis of aldonic acids. Carbohyd Res 2006;341:2290-
2.
Playne MJ, Crittenden RG. Galacto-oligosaccharides and other products derived from lactose.
In: McSweeney PLH, Fox PF, editors. Advanced dairy chemistry, volume 3: lactose,
water, salts and minor constituents. New York: Springer Science and Business Media,
LLC; 2009. p. 121-201.
Pollard DJ, Woodley JM. Biocatalysis for pharmaceutical intermediates: the future is now.
Trends Biotechnol 2007;25:66-73.
Qiu Y, Mao Z, Zhao Y, Zhang J, Guo Q, Gou Z, Gao C. Polycaprolactone scaffold modified with
galactosylated chitosan for hepatocyte culture. Macromol Res 2012;20:283-91.
Raake W, Klauser RJ, Elling H, Meinetsberger E. Anticoagulant and antithrombotic

properties of synthetic sulphated bis-lactobionic acid amides. Thromb Res
1989;56:719-30.
Richard C, Chaumet-Riffaud P, Belland A, Parat A, Contino-Pepin C, Bessodes M, Scherman D,
Pucci B, Mignet N. Amphiphilic perfluoroalkyl carbohydrates as new tools for liver
imaging. Int J Pharm 2009;379:301-8.
Rico-Lattes I, Blanzat M, Franceschi-Messant S, Perez E, Lattes A. Catanionic sugar derived
surfactants, polymers and dendrimers: from molecules to targeted self-organized
systems. CR Chimie 2005;8:807-14.
Rôzycka-Roszak B, Jurczak B, Wilk KA. Effects of nonionic sugar surfactants on the phase
transition of DPPC membranes. Thermochim Acta 2007;453:27-30.
Saarela M, Hallamaa K, Mattila-Sandholm T, Mättö J. The effect of lactose derivatives
lactulose, lactitol and lactobionic acid on the functional and technological properties of
potentially probiotic Lactobacillus strains. Int Dairy J 2003;13:291-02.
Satory M, Fürlinger M, Haltrich D, Kulbe KD, Pittner F, Nidetzky B. Continuous enzymatic
production of lactobionic acid using glucose-fructose oxidoreductase in an
ultrafiltration membrane reactor. Biotechnol Lett 1997;19:1205-8.
Sauer M, Mattanovich D. Construction of microbial cell factories for industrial bioprocesses. J
Chem Technol Biot 2012;87:445-50.
Sauer M, Porro D, Mattanovich D, Branduardi P. Microbial production of organic acids:
expanding the markets. Trends Biotechnol 2008;26:100-8.
Schaafsma G. Lactose and lactose derivatives as bioactive ingredients in human nutrition. Int
Dairy J 2008;18:458-65.
Schenk JF, Radziwon P, Mörsdorf S, Eckenberger P, Breddin HK. Effects of aprosulate, a novel
synthetic glycosaminoglycan, on coagulation and platelet function parameters: a
prospective, randomized phase I study. Clin Appl Thrombosis/Hemostasis
1999;5:192-7.
Seki N, Saito H. Lactose as a source for lactulose and other functional lactose derivatives. Int
Dairy J 2012;22:110-115.
Selim KMK, Ha YS, Kim SJ, Chang Y, Kim TJ, Lee GH, Kang IK. Surface modification of
magnetite nanoparticles using lactobionic acid and their interaction with hepatocytes.
Biomaterials 2007;28:710-6.
Selim KMK, Xing ZC, Guo H, Kang IK. Immobilization of lactobionic acid on the surface of
cadmium sulfite nanoparticles and their interaction with hepatocytes. J Mater Sc:
Mater Med 2009;20:1945-53.
Smithers GW. Whey and whey proteins - From “gutter-to-gold”. Int Dairy J 2008;18:695-04.
Southard JH. Coffee creamer, the bionic man, and organ preservation. Surgery
2002;131:228-9.
Southard JH, Belzer FO. Organ preservation. Annu Rev Med 1995;46:235-47.
Splechtna B, Petzelbauer I, Baminger U, Haltrich D, Kulbe KD, Nidetzky B. Production of a
lactose-free galacto-oligosaccharide mixture by using selective enzymatic oxidation of
lactose into lactobionic acid. Enzyme Microb Tech 2001;29:434-40.
Stannard KA, Collins PM, Ito K, Sullivan EM, Scott SA, Gabutero E, Grice ID, Low P, Nilsson UJ,
Leffler H, Blanchard H, Ralph SJ. Galectin inhibitory disaccharides promote tumour
immunity in a breast cancer model. Cancer Lett 2010;299:95-110.
Sternberg M, Lockwood LB. Oxidation of isomaltose by Pseudomonas taetrolens. J Bacteriol
1969;99:623.
Stockert RJ. The asialoglycoprotein receptor: relationships between structure, function, and
expression. Physiol Rev 1995:75:591-609.
Stodola FH, Lockwood BL. The oxidation of lactose and maltose to bionic acids by
Pseudomonas. J Biol Chem 1947;213-21.
Stodola FH, Jackson RW. Fermentation acids in industry. US Dept Agr. Yearbook 1950;85-92.
Sugiyama M, Watanabe K, Funakoshi N, Amino Y, Kawahara S, Takemoto T. Process of
producing glutamate derivatives. United States Patent Application Pub. No.: US
2005/0009153 A1; 2005.
Sumimoto R, Kamada N. Lactobionate as the most important component in UW solution for
liver preservation. Transplant Proc 1990;22:2198-9.
Tasic-Kostov M, Savic S, Lukic M, Tamburic S, Pavlovic M, Vuleta G. Lactobionic acid in a
natural alkylpolyglucoside-based vehicle: assessing safety and efficacy aspects in
comparison to glycolic acid. J Cosmet Dermatol 2010;9:3-10.
Tian Z, Wang M, Zhang AY, Feng ZG. Preparation and evaluation of novel amphiphilic
glycopeptide block copolymers as carriers for controlled drug release. Polymer
2008;49:446-54.
Tompkin RB, Shaparis AB. Potato aroma of lamb carcasses. Appl Microbiol 1972;24:1003-4.
Unitika. Press release; 2008. Available via: http://www.unitika.co.jp/e/news/news-
data/news080326.html. Accessed 3 August 2012.
Upadhya GA, Strasberg SM. Glutathione, lactobionate, and histidine: cryptic inhibitors of
matrix metalloproteinases contained in University of Wisconsin and
histidine/tryptophan/ketoglutarate liver preservation solutions. Hepatology
2000;31:1115-22.
Vakil JR, Shahani KM. Carbohydrate metabolism of lactic acid cultures. V. Lactobionate and
gluconate metabolism of Streptococcus lactis UN. J Dairy Sci 1969;52:1928-34.
Van Dokkum W, Wezendonk LJW, van Aken-Schneider P, Kistemaker C. Tolerance of
lactobionic acid in man. TNO Report. V94.115, TNO Nutrition and Food Research, Zeist,
The Netherlands, 1994.
Van Hecke W, Bhagwat A, Ludwig R, Dewulf J, Haltrich D, Van Langenhove H. Kinetic
modeling of a bi-enzymatic system for efficient conversion of lactose to lactobionic
acid. Biotechnol Bioeng 2009a;102:1475-82.
Van Hecke W, Ludwig R, Dewulf J, Auly M, Messiaen T, Haltrich D, Van Langenhove H.
Bubble-free oxygenation of a bi-enzymatic system: effect on biocatalyst stability.
Biotechnol Bioeng 2009b;102:122-31.
Van Hecke W, Haltrich D, Frahm B, Brod H, Dewulf J, Van Langenhove H, Ludwig R. A
biocatalytic cascade reaction sensitive to the gas-liquid interface: modeling and
upscaling in a dynamic membrane aeration reactor. J Mol Catal B-Enzym 2011;68:154-
61.
van Wijk A, Siebum A, Schoevaart R, Kieboom T. Enzymatically oxidized lactose and
derivatives thereof as potential protein cross-linkers. Carbohyd Res 2006;341:2921-6.
Varshosaz J, Hassanzadeh F, Sadeghi H, Khadem M. Galactosylated nanostructured lipid
carriers for delivery of 5-FU to hepatocellular carcinoma. J Liposome Res
2012;22:224-36.
Villadsen J. Innovative technology to meet the demands of the white biotechnology
revolution of chemical technology. Chem Eng Sci 2007;24:6957-68.
Walter T, Begli HA. Flavour enhancer. United States Patent Application Pub. No.: US
2011/0244080 A1; 2011.
Wang S, Cheng L, Yu F, Pan W, Zhang J. Delivery of different length poly(L-lysine)-conjugated
ODN to HepG2 cells using N-stearyllactobionamide-modified liposomes and their
enhanced cellular biological effects. Int J Pharm 2006a;311:82-8.
Wang S, Deng Y, Xu H, Wu H, Qiu Y, Chen D. Synthesis of a novel galactosylated lipid and its
application to the hepatocyte-selective targeting of liposomal doxorubicin. Eur J
Pharm Biopharm 2006b;62:32-8.
Wang J, Xiao F, Zhao Y, Chen L, Zhang R, Guo G. Cell proliferation and thermally induced cell
detachment of galactosylated thermo-responsive hydrogels. Carbohydr Polym
2010;82:578-84.
Weiss SI, Sieverling N, Niclasen M, Maucksch C, Thünemann AF, Möhwald H, Reinhardt D,
Rosenecker J, Rudolph C. Uronic acids functionalized polyethyleneimine (PEI)-
polyethyleneglycol (PEG)-graft-copolymers as novel synthetic gene carriers.

Biomaterials 2006;27:2302-12.
West D. AHA wrinkle creams come of age. Chem Week 2004a;166:20.
West TP. Regulation of pyrimidine nucleotide formation in Pseudomonas taetrolens ATCC
4683. Microbiol Res 2004b;159:29-33.
Werle P, Merz F, Jorgensen JN, Trageser M. Chlorhexidine formulations, new chlorhexidine
salts, solutions containing these and their use. United States Patent Application Pub.
No.: US 20020018814 A1; 2002.
Wilk KA, Syper L, Domagalska BW. Aldonamide-type gemini surfactants: synthesis,
structural analysis, and biological properties. J Surfactants Deterg 2002;5:235-44.
Willke T, Vorlop KD. Industrial bioconversion of renewable resources as an alternative to
conventional chemistry. Appl Microbiol Biotechnol 2004;66:131-42.
Wu DQ, Lu B, Chang C, Chen CS, Wang T, Zhang YY, Cheng SX, Jiang XJ, Zhang XZ, Zhuo RX.
Galactosylated fluorescent labeled micelles as a liver targeting drug carrier.
Biomaterials 2009a;30:1363-71.
Wu F, Wuensch SA, Azadniv M, Ebrahimkhani MR, Crispe IN. Galactosylated LDL
nanoparticles: a novel targeting delivery system to deliver antigen to macrophages
and enhance antigen specific T cell responses. Mol Pharmaceut 2009b;6:1506-17.
Wu DQ, Li ZY, Li C, Fan JJ, Lu B, Chang C, Cheng SX, Zhang XZ, Zhuo RX. Porphyrin and
galactosyl conjugated micelles for targeting photodynamic therapy. Pharmaceut Res
2010;27:187-99.
Yang BY, Montgomery R. Oxidation of lactose with bromine. Carbohyd Res 2005;340:2698-
05.
Yang J, Chung TW, Nagaoka M, Goto M, Cho CS, Akaike T. Hepatocyte-specific porous
polymer-scaffolds of alginate/galactosylated chitosan sponge for liver-tissue
engineering. Biotechnol Lett 2001;23:1385-9.
Yang J, Goto M, Ise H, Cho CS, Akaike T. Galactosylated alginate as a scaffold for hepatocytes
entrapment. Biomaterials 2002;23:471-9.
Yang ST, Huang H, Tay A, Qin W, De Guzman L, San Nicolas EC. Extractive fermentation for
the fermentation of carboxylic acids. In: Yang ST, editor. Bioprocessing for value-added
products from renewable resources. Oxford: Elsevier Inc; 2007. p. 421-46.
Yang W, Mou T, Guo W, Jing H, Peng C, Zhang X, MA Y, Liu B. Fluorine-18 labeled
galactosylated chitosan for asialoglycoprotein-receptor-mediated hepatocyte imaging.
Bioorg Med Chem Lett 2010;20:4840-4.
Yang R, Meng F, Ma S, Huang F, Liu H, Zhong Z. Galactose-decorated cross-linked

biodegradable poly(ethylene glycol)-b-poly(Ɛ-caprolactone) block copolymer micelles
for enhanced hepatoma-targeting delivery of paclitaxel. Biomacromolecules
2011;12:3047-55.
Yoshimura T, Ishihara K, Esumi K. Sugar-based gemini surfactants with peptide bonds-
synthesis, adsorption, micellization, and biodegradability. Langmuir 2005;21:10409-
15.
Yu RJ, Van Scott EJ. Alpha-hydroxyacids and carboxylic acids. J Cosmet Dermatol 2004;3:76-
87.
Zeng X, Lu Z, Liu Y. Synthesis and solution properties of novel sugar-based polysiloxane
surfactants. J Surfact Deterg 2012. doi: 10.1007/s11743-012-1369-5
Zhang J, Li C, Xue ZY, Cheng HW, Huang FW, Zhuo RX, Zhang XZ. Fabrication of lactobionic-
loaded chitosan microcapsules as potential drug carriers targeting the liver. Acta
Biomater 2011;7:1665-73.
Zhao Q, Chen W, Chen Y, Zhang L, Zhang J, Zhang Z. Self-assembled virus-like particles from
rotavirus structural protein VP6 for targeted drug delivery. Bioconjugate Chem
2011;22:346-352.
Zheng D, Duan C, Zhang D, Jia L, Liu G, Liu Y, Wang F, Li C, Guo H, Zhang Q. Galactosylated
chitosan nanoparticles for hepatocyte-targeted delivery of oridonin. Int J Pharm
2012;436:379-86.
Zhou H, Chen Z, Yang R, Shang L, Li G. Direct electrochemistry and electrocatalysis of
hemoglobin in lactobionic acid film. J Chem Technol Biot 2006;81:58-61.
3. Materiales y métodos
Materiales y métodos 81
3. MATERIALES Y MÉTODOS
3.1. Producción biotecnológica de ácido láctico
3.1.1. Microorganismo de cultivo
Lactobacillus casei ATCC 393, obtenida a partir de la Colección Americana de Cultivos

Tipo (Virginia, Estados Unidos) fue mantenida congelada (en solución con 40 % v/v de
glicerol a -20ºC) y a 4ºC en placas de Man Rogosa and Sharpe agar (MRS, Biokar Diagnostics,
Francia). La cepa fue cultivada en caldo MRS o en placas de agar (con un 2 % p/p de agar).
3.1.2. Preparación del lactosuero de yogurt
El lactosuero residual de yogurt fue obtenido a partir de yogures fuera de la fecha de

caducidad, los cuales fueron mezclados previamente con el fin de obtener el lactosuero
mediante coagulación térmica, eliminando de esta forma la mayoría de la grasa y las
proteínas. A continuación, el resto de lípidos y proteínas fueron retirados de esta fracción
líquida mediante centrifugación a 12.000 x g (Kubota, modelo 6500, Tokyo, Japón) durante
10 minutos a 4ºC. Posteriormente, el lactosuero se diluyó con agua destilada (en un ratio 1:1
v/v) y se ajustó el pH a 6,5 con NaOH 1 N. Una vez diluido, el lactosuero de yogurt fue
esterilizado a través de una microfiltración tangencial con un cassette de membranas de
PVDF (Millipore, Massachusetts, Estados Unidos) con un tamaño de poro de 0,22 μm. La
Tabla 3.1 recoge la comparación entre los valores físico-químicos determinados
experimentalmente de la fracción líquida de partida y del lactosuero utilizado en los
experimentos.
Tabla 3.1. Características físico-químicas del lactosuero de yogurt utilizado en las fermentaciones.
NTK [Proteína] Densidad DQO
ºBrix Humedad (%) pH
(mg/L) (g/L) (g/L) (mg/L)
Lactosuero de yogurt
580 3.6 1042 13 91.6 90 4.2
inicial
Lactosuero de yogurt
utilizado en las 309 1.9 1018 6.5 96.1 65 6.5
fermentaciones
82 Materiales y métodos
3.1.3. Condiciones de cultivo
Tras la reactivación de L. casei en MRS bajo condiciones microaerófilas, un 10 % (v/v) de

este cultivo (en fase de crecimiento exponencial) fue empleado para inocular tanto 90 mL de
lactosuero de yogurt como lactosuero de yogurt suplementado con 2,5 g/L de extracto de
levadura, ambos cultivos utilizados como preinóculos. Estos preinóculos fueron incubados a
37ºC durante 20 h a 100 rpm en un incubador orbital (Infors HT, modelo Flyer Aerotron,
Bottmingen, Suiza). A continuación, estos precultivos fueron empleados para inocular las
fermentaciones de lactosuero de yogurt. Las fermentaciones sin control de pH fueron
llevadas a cabo en un agitador orbital (100 rpm) con un volumen final de trabajo de 1 L a
37ºC. Mientras que las fermentaciones con control de pH se llevaron a cabo en un
biorreactor de 2 L (Bioflo 110, New Brunswick Scientific Co. Inc., Nueva Jersey, Estados
Unidos) con un volumen de trabajo de 1 L, equipado con un control digital de la unidad,
siendo la temperatura controlada a 37ºC y el pH fue fijado en 6,5 usando NaOH 3 N. La
agitación fue fijada en 50 rpm con el fin de mantener el medio de fermentación mezclado, y
así evitar la sedimentación tanto de L. casei como del ácido láctico formado. Las muestras
fueron tomadas a intervalos regulares de 2 h con el fin de determinar la densidad óptica y el
pH, mientras que los sobrenadantes libres de células (filtrados por 0,45 μm) fueron
almacenados bajo congelación hasta posterior análisis. Las fermentaciones se llevaron a
cabo en duplicado como experimentos independientes.
3.1.4. Métodos analíticos
Con objeto de determinar la biomasa celular, las muestras tomadas a lo largo de la

evolución de los cultivos fueron centrifugadas durante 5 minutos a 16.000 x g (Eppendorf,
modelo 5415 D, Hamburgo, Alemania), siendo resuspendido el correspondiente pellet
resultante en una solución de NaCl 0,7% (p/v). El crecimiento celular fue medido
espectrofotométricamente como densidad óptica a una longitud de onda de 660 nm
(Shimazdu, modelo UV 1203). La densidad óptica fue convertida en medidas de biomasa
(peso seco) usando la correspondiente curva de calibración previamente determinada,
expresando los resultados en gramos por litro (g/L).
El contenido proteico de los sobrenadantes fue determinado de acuerdo con el
procedimiento descrito por Bradford (Bradford, 1976) (reactivo Sigma-Aldrich, Steinheim,
Alemania) utilizando sero-albúmina bovina como patrón.
Mientras que la lactosa, glucosa, sacarosa y el ácido láctico, presentes en los

sobrenadantes tomados a lo largo de la evolución de los cultivos, fueron analizados
mediante Cromatografía de Líquidos de Alta Eficacia (HPLC) siguiendo el método descrito
por Pedruzzi et al. (2007) con pequeñas modificaciones. En síntesis, se han utilizado las
siguientes condiciones analíticas: 0,450 mM H2SO4 como fase móvil (pH 3,1), una
temperatura de la columna fijada en 75ºC y un flujo de 0,3 mL/min. En la determinación y
cuantificación analítica se ha empleado una columna ICSep ICE-ION-300 (Transgenomic Inc.,
San Jose, California, Estados Unidos), utilizando el índice de refracción como detector en un
cromatógrafo Agilent (modelo serie 1200, California, Estados Unidos) (Figura 3.1). Las
muestras fueron cuantificadas empleando standards externos de grado HPLC obtenidos de
las siguientes casas comerciales: lactosa, sacarosa y ácido láctico fueron obtenidos de Chem
Service (West Chester, Pennsylvania, Estados Unidos), glucosa de Merck (Darmstadt,
Alemania) y etanol de Panreac (Barcelona, España). La adquisición y análisis de datos fue
realizada con el software Agilent ChemStation.
Figura 3.1. Cromatógrafo de líquidos de alta eficacia (HPLC) empleado en las determinaciones analíticas.
3.1.5. Monitorización mediante citometría de flujo
La monitorización del estado fisiológico de L. casei, se ha efectuado a lo largo de dos

procesos fermentativos, con y sin control de pH. Un resumen de los pasos llevados a cabo
durante el proceso se muestra en la Figura 3.2. Diferentes muestras fueron tomadas con el
fin de determinar el estado fisiológico mediante citometría de flujo y densidad óptica a lo
largo de la evolución de ambos procesos fermentativos. Los sobrenadantes fueron
almacenados (a -20ºC) hasta el correspondiente análisis de los mismos.
Figura 3.2. Diagrama que esquematiza las etapas previas y fermentaciones llevadas a cabo en el seguimiento y
monitorización mediante citometría de flujo. Los precultivos en lactosuero de yogur con y sin extracto de
levadura fueron utilizados como inóculos en las fermentaciones finales con lactosuero de yogur.
El análisis de las diferentes muestras mediante citometría de flujo ha sido llevado a cabo
siguiendo una metodología previamente publicada con ligeras modificaciones (Quirós et al.,
2007). Una vez centrifugadas las muestras (12.000 x g durante 5 min.), la biomasa
resultante fue lavada dos veces en tampón fosfato salino (PBS) (pH 7.4, estéril y filtrado por
0,22 μm) y ajustado a una densidad celular de aproximadamente 4 x 105 unidades
formadoras de colonias por mL en el mismo tampón. Con el fin de teñir las células de L.
casei, a 200 μL de la suspensiones celulares se les adicionó las soluciones de tinción
previamente preparadas. Para los análisis, las muestras fueron teñidas con una mezcla de
dos fluorocromos, ChemChromeV6, CV6 (Chemunex, France) e Ioduro de Propidio, IP
(Molecular Probes). El CV6 es un éster fluorogénico que difunde al interior celular donde es
hidrolizado por las enzimas esterasas intracelulares dando lugar a un producto fluorescente
(Cao-Hoang et al., 2008; Parthuisot et al., 2000). La solución stock utilizada fue preparada
mediante dilución 1:10 en agua destilada estéril (filtrada por 0,22 μm), de la cual 8 μL
fueron añadidos a la suspensión celular y se incubó durante 15 minutos en oscuridad y a
temperatura ambiente.
Mientras que el fluorocromo CV6 tiñe las células viables caracterizadas por presentar
actividad enzimática, el ioduro de propidio sólo penetra en células que muestran una
membrana dañada o permeable. En estas células, el ioduro de propidio penetra uniéndose al
DNA formando un complejo fluorescente rojo. La solución stock de ioduro de propidio
empleada (1 mg/mL solución en agua, Molecular Probes) fue diluida en agua destilada
estéril y añadida a la suspensión celular a una concentración final de 5,25 μg/mL, siendo
incubada la mezcla durante 30 minutos en las mismas condiciones anteriores. Cada análisis
fue llevado a cabo en duplicado con el ajuste de compensación correspondiente. Una

disposición del densitograma generado (fluorescencia verde versus señal en rojo) al
analizar las muestras se muestra en la Figura 3.3.
Figura 3.3. Citograma generado durante la tinción dual CV6/IP.
Como controles, se emplearon suspensiones celulares tratadas térmicamente (90ºC

durante 30 minutos e inmediatamente enfriadas en hielo), células en fase exponencial y
mezclas de células en fase exponencial y tratadas térmicamente (1:1). Asimismo, muestras
sin teñir fueron empleadas como controles negativos. Los análisis fueron llevados a cabo en
un citómetro de flujo Cytomics FC 500 (Beckman Coulter) equipado con una fuente láser de
iones (488 nm) y con cuatro detectores de fluorescencia (Figura 3.4). La fluorescencia
procedente de células teñidas con CV6 fue recogida en el canal FL1 (525 nm), mientras que
la fluorescencia procedente del IP fue registrada en el canal FL3 (610 nm). Como standards
internos fueron utilizados 100 μL de microsferas fluorescentes (Perfect Count; Cytognos,
España), así en cada análisis fueron adquiridas 2000 microsferas empleando en la
adquisición el software Cytomics RXP (Beckman Coulter).
Figura 3.4. Citómetro analizador empleado para la adquisición de datos durante la monitorización fisiológica.
3.2. Producción biotecnológica de ácido lactobiónico
3.2.1. Microorganismo de cultivo
Para la producción de ácido lactobiónico, Pseudomonas taetrolens LMG 2336 (obtenida a

partir de la Colección Coordinada Belga de Microorganismos, BCCM) fue mantenida a 4ºC en
placas de Nutrient Broth Nº3 agar (Fluka), con la siguiente composición: extracto de carne
(1 g/L), extracto de levadura (2 g/L), peptona (5 g/L) y NaCl (5 g/L). A partir de estas placas
en medio sólido se procedió a revitalizar la bacteria en condiciones aeróbicas.
3.2.2. Preparación del suero lácteo
Se han utilizado dos concentraciones distintas de lactosuero dulce (suministrado por

ILAS S.A., Navia, Asturias), sin diluir y lactosuero diluido con agua destilada (en un ratio 1:1
v/v) para seguidamente ajustar el pH a 6,5 con NaOH 6 N. Una vez diluido y con el pH
ajustado, el lactosuero fue esterilizado a través de una microfiltración tangencial con un
cassette de membranas de PVDF (Millipore, Massachusetts, Estados Unidos) con un tamaño
de poro de 0,22 μm.
3.2.3. Condiciones experimentales de cultivo
Se han realizado varias estrategias de bioconversión oxidativa, de este modo, se han

realizado experimentos en matraz Erlenmeyer y en biorreactor con control de la agitación y
del pH. Así, células procedentes de un cultivo líquido en NB Nº3 fueron empleadas como
inóculo del lactosuero (tras centrifugación de la biomasa necesaria a 12.000 x g durante 10
minutos) en un matraz Erlenmeyer con capacidad de 500 mL (relación volumen de
trabajo:aire, 1:5). Estos cultivos fueron incubados a 30ºC durante 80 h con una agitación de
250 rpm en un incubador orbital (New Brunswick Scientific Co. Inc., Nueva Jersey, Estados
Unidos).
Los cultivos en biorreactor se llevaron a cabo en un biorreactor de 2 L (Bioflo 110, New
Brunswick Scientific Co. Inc., Nueva Jersey, Estados Unidos) con un volumen de trabajo de 1
L (Figura 3.5). Dicho biorreactor ha estado equipado con un control digital de la unidad,
siendo la temperatura fijada a 30ºC y el pH controlado a diversos valores empleando una
disolución concentrada de NaOH como agente regulador del pH. El valor del pH inicial, así
como la estrategia de control de pH empleada durante los cultivos han variado según la
variable estudiada. Así, se ha estudiado la influencia diferentes valores de pH inicial, control
extricto de pH o estrategias de control de pH en fase de producción, variando los valores
entre 4,5-7,5 en cada una de las estrategias estudiadas. Igualmente, las tasas de agitación y
aireación se fijaron en distintos valores (entre 150-1000 rpm y 0.5-2 Lpm, respectivamente)
en función de la variable objeto de estudio (véanse los diferentes subapartados relativos a
los resultados de la presente memoria para particularizar). Como agente antiespumante se
ha empleado la emulsión diluida (1:10) Y-30 (Sigma-Aldrich, Steinheim, Alemania). Como
inóculo de estas bioconversiones en biorreactor se han empleado cultivos en matraz
Erlemeyer conteniendo suero lácteo y cultivados con una agitación de 250 rpm y a una
temperatura de 30ºC (la duración de los cultivos ha dependido de nuevo en función de la
variable sometida a estudio, véase en concreto cada subapartado de resultados para
particularizar). Asmismo, la duración (12-48 h) así como el tamaño de inóculo (5-30%) han
sido parámetros igualmente estudiados.
Figura 3.5. Equipo de fermentación empleado en los experimentos.
3.2.4. Métodos analíticos
Con el fin de determinar la biomasa, las muestras tomadas a lo largo de los cultivos
fueron centrifugadas durante 5 minutos a 16.000 x g (Eppendorf, modelo 5415 D,
Hamburgo, Alemania), siendo el pellet resultante resuspendido en una solución de NaCl 0,7
% (p/v). El crecimiento celular fue medido espectrofotométricamente como densidad óptica
a una longitud de onda de 600 nm (Shimazdu, modelo UV 1203), siendo estos valores
convertidos en términos de biomasa (peso seco, expresado en términos de g/L) empleando
la correspondiente curva de calibración previamente determinada.
El contenido en lactosa y ácido lactobiónico procedente de las muestras fue cuantificado
siguiendo la misma metodología explicada con anterioridad en el subapartado 3.1.4. En este
caso, se emplearon la lactosa (Chem Service, West Chester, Pennsylvania, USA) y el ácido
lactobiónico (Sigma-Aldrich, Steinheim, Alemania) como estándares de alta pureza.
3.2.5. Monitorización mediante citometría de flujo
Las muestras tomadas a lo largo de los procesos fermentativos fueron centrifugadas

(16.000 x g durante 5 min.) con el fin de retirar el sobrenadante y separar la biomasa
celular. Previo al proceso de tinción, las células fueron lavadas dos veces en tampón fosfato
salino (PBS, pH 7.4, estéril y filtrado por 0,22 μm), para ser sometidas posteriormente a un
tratamiento en ultrasonidos durante 2 segundos con el fin de evitar cualquier posible
agregación celular previa al análisis multi-paramétrico (Hewitt y Nebe-von-Caron, 2004).
Ioduro de propidio (PI, Invitrogen), bis-(1,3-dibutylbarbituric acid) trimethine oxonol (bis-
oxonol, DiBAC4(3), Invitrogen) y carboxifluoresceína (cFDA, Invitrogen) fueron usados como
fluorocromos en un protocolo de tinción doble y dual (DiBAC4(3)/PI y cFDA/PI) con el
objeto de evaluar el status fisiológico celular (la actividad metabólica, integridad de
membrana y polarización de membrana fueron evaluados a través de la tinción con cFDA, PI
y DiBAC4(3), respectivamente). Las soluciones stock fueron preparadas de la siguiente
forma: PI fue preparado a una concentración de 1 mg/mL en agua destilada (filtrada por
0.22 μm) y almacenada a 4ºC, mientras que tanto las soluciones de DiBAC4(3) como las de
cFDA fueron preparadas en dimetil sulfóxido (DMSO, Sigma-Aldrich) a una concentración de
1 mM y posteriormente almacenadas a -20ºC. La disolución de tinción de PI fue preparada
mediante la dilución de la solución stock en agua destilada estéril, añadiendo dicha solución
a la suspensión celular a una concentración final de 5 μg/mL. Esta mezcla fue a continuación
incubada bajo condiciones de oscuridad a temperatura ambiente. Las soluciones de trabajo
de DiBAC4(3) y cFDA fueron realizadas en concentraciones de 5 y 10 μM en PBS conteniendo
1 mM EDTA, respectivamente. Las células teñidas con fueron posteriormente incubadas con
1 μM DiBAC4(3) o 0.1 μM cFDA durante 15 min en oscuridad y a temperatura ambiente.
La adquisición de los correspondientes datos multi-paramétricos fue llevada a cabo en
un citómetro analizador Cytomics FC 500 (Beckman Coulter) equipado con una fuente de
excitación laser de 488 nm y 633 nm procedente de un laser de argón. La fluorescencia
verde de las muestras (correspondiente a células teñidas con DiBAC4(3) y cFDA) fue
recogida en el canal FL1 (530 nm), mientras que la fluorescencia de PI fue registrada en el
canal FL3 (610 nm). Una disposición del citograma generado se muestra en la Figura 3.6.
Cada análisis fue realizado por duplicado con una velocidad de adquisición de 4000
eventos/seg. La adquisición se llevó a cabo con el software Cytomics RXP software
(Beckman Coulter). Tanto las ventanas como cuadrantes fueron establecidas en base
controles de tinción. En cada análisis dual-paramétrico DiBAC4(3)/PI y cFDA/PI, 150.000 y
100.000 eventos fueron recogidos, respectivamente. Siendo analizados los datos a través del
software Summit v4.3 (DakoCytomation, Colorado, USA).
Figura 3.6. Densitograma o citograma donde se muestran la diferentes subpoblaciones generadas con la
aplicación del protocolo de tuición cFDA/IP
3.2.6. Separación celular
La separación celular (Fluorescence-Activated Cell Sorting, FACS) fue realizada en un

MoFlo XDP cell sorter (DakoCytomation) equipado con un laser de argón que emite luz azul
a longitudes de onda de 488 y 630 nm (200 mW) (Figura 3.7). La fluorescencia verde
procedente de las células teñidas con cFDA fue detectada en el canal FL1 (530 nm), mientras
que la fluorescencia roja de la señal de PI fue recogida en el canal FL3 (610 nm).
Microesferas fluorescentes (Flow-check Pro Fluorospheres, Beckman Coulter) con tamaños
de 10, 6 y 3 μm fueron usadas para evaluar la hidrodinámica del sistema y el alineamiento
óptico del citómetro (coeficiente de variación, valor CV, alrededor de 2%). Las muestras
fueron separadas en modo purify con una velocidad alrededor de 200 células/seg.
empleando Isoton II (Beckman Coulter) como solución fluídica de transporte. El sistema
operó con una presión de 54 psi y fue equipado con un tamaño de nozzle de 70 μm. Las
ventanas de separación fueron establecidas en base a la fluorescencia del canal FL1 (células
cFDA(+)/PI(-)) frente a la fluorescencia del canal FL3 (células cFDA(-)/PI(+)). La
adquisición de datos fue realizada usando el software Summit v5.0 (DakoCytomation). Las
células procedentes de las diferentes subpoblaciones discriminadas en función de los
cuadrantes (véase Figura 3.6) fueron separadas (96 spots) en placas de agar conteniendo
lactosa (10 g/L), peptona (2 g/L), extracto de levadura (1 g/L), K2HPO4 (0.2 g/L), azul de
bromofenol (0.08 g/L) y agar (15 g/L) con el objetivo de comprobar tanto la capacidad
oxidativa como la viabilidad de las mismas. Dichas placas fueron incubadas a 30ºC durante
96 h.
A B
Figura 3.7. Citómetro separador empleado en los experimentos (A). Detalle de la disposición de la bandeja de
separación (contiene una placa de agar sobre la que se han separado las células en 96 spots) con las placas
deflectoras dentro de la cámara hermética (B).
3.2.7. Microscopía confocal de fluorescencia
Las muestras teñidas fueron también examinadas en un microscopio laser confocal Leica
TCS-SP2-AOBS (Leica Microsystems Inc., Heidelberg, Alemania) a unas longitudes de
excitación de 488 y 568 nm, con una longitud de emisión de 530 (fluorescencia verde) o 630
nm (fluorescencia roja).
4. RESULTADOS
Resultados 93
4. RESULTADOS
4.1. Producción de ácido láctico por Lactobacillus casei
En este subapartado se recogen los resultados referentes a la obtención biotecnológica de

ácido láctico a partir de lactosuero de yogurt empleando L. casei como plataforma
microbiana. Como se ha venido señalando en anteriores subapartados (2.2), el ácido láctico
es un ácido orgánico maduro cuya producción biotecnológica está más que asentada en el
ámbito industrial. Sin embargo, la industria y el mundo académico continúan aunando
esfuerzos en la búsqueda de nuevas fuentes que sirvan como materia prima o sustrato de la
fermentación ácido-láctica. Ante las enormes perspectivas de crecimiento del ácido
poliláctico, las actuales tendencias en investigación relacionadas con el ácido láctico se han
focalizado en la búsqueda y el empleo de nuevos sustratos ricos en carbohidratos
fermentables. En la presente memoria se ha propuesto al lactosuero de yogurt, procedente
de yogures caducados o retirados de la cadena comercial, como una matriz idónea para la
obtención de ácido láctico. La industria láctea estima que alrededor del 1% del volumen
total de yogures comercializados (categoría en la que se incluyen todos los postres
fermentados) han de ser retirados y ser tratados como un residuo, con el consiguiente coste
económico. En el artículo de investigación que a continuación se presenta, se ha hecho
hincapié en la importancia de la naturaleza de los azúcares y de cómo la presencia de estos
carbohidratos lleva consigo la modulación del catabolismo de los mismos por parte de L.
casei. Asimismo, se ha estudiado la importancia de cómo diferentes parámetros
operacionales influyen sobre el rendimiento y la productividad del sistema fermentativo.
En suma, se ha propuesto al lactosuero de yogurt como sustrato dentro de un sistema
caracterizado por no precisar de pretratamiento ni sacarificación alguna a diferencia de
otros residuos empleados como sustratos, caso de los residuos lignocelulósicos o agro-
industriales.
Publicación: Residual yoghurt whey for lactic acid production.

Situación: Artículo publicado en Biomass and Bioenergy (Elsevier).
biomass and bioenergy 34 (2010) 931–938
Available at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Residual yoghurt whey for lactic acid production
Saúl Alonso, Mónica Herrero, Manuel Rendueles, Mario Dı́az*

Department of Chemical Engineering and Environmental Technology, University of Oviedo, Faculty of Chemistry, C/Julián Claverı́a s/n,
33071 Oviedo, Spain
article info abstract
Article history: The operation in global yoghurt market provokes frequently the reject and withdrawn of
Received 27 July 2009 yoghurt derivates out of shelf life. Yoghurts that have past their ‘best before’ date consti-
Received in revised form tute a waste that has to be environmentally treated. Its use as a source for lactic acid
24 January 2010 production by Lactobacillus casei is studied in this work, being also a proposal that can
Accepted 25 January 2010 reduce the end-of-cycle impact of the residue. Production of yoghurt generates a residue
Available online 19 February 2010 highly supplemented with sugar and fruits syrups that can be metabolized to lactic acid,
but no data are available in the literature about this potential transformation into
Keywords: a commercial valuable lactic acid. Bioconversion of total sugars around 44% was achieved
Yoghurt residue in pH controlled batch fermentation after 34 h. Control of pH and yeast extract supple-
Yoghurt whey mentation in the preculture stage have shown a remarkable enhancement of fermentation
Sugars performance due to the reduction in the toxicity and the increase in nitrogen content.
Lactic acid Successive bioconversions of lactose, glucose and sucrose have been shown, with a sugar
Yeast extract consumption profile giving a hierarchical mode consumption in the order
Lactobacillus casei glucose > sucrose > lactose.
ª 2010 Elsevier Ltd. All rights reserved.
1. Introduction animal feed [2]. Indeed, dairy industry estimates that around
1% of total yoghurt production including not only rejected and
Over the past years, yoghurt and fermented dairy products expired yoghurts but also yoghurts generated from over-
have been converted in the fastest-growing dairy segment, stocks and product line changeover are wasted off. To date,
contributing more and more with higher rates to global sales. yoghurt residuals are employed in animal feeding with other
Recent forecasts from consulting companies even consider rest-streams generated by food industry. However, their high
the increase of global sales over the coming years, especially water content limits their use as animal feed, making conve-
in emerging markets as Asiatic area. This product has grown nient to remove liquid whey. Whereas dry fraction is used for
at a rate of double-digit for the last years, so world market of animal feed purposes, yoghurt whey is proposed here to be
liquid yoghurt had a rate of growth between 13 and 18% from used as source for lactic acid fermentation, the main chal-
2000 to 2006 [1]. lenge of this work.
The future of yoghurt production, considering the envi- Yoghurt is a dairy product sweetened with high levels of
ronmental implications of increasing manufacture, involves conventional added sugars such as sucrose and glucose [3].
the treatment of the dairy fermented products out of shelf life. For this reason their environmental impact is higher than
The withdrawn and rejects from market of damaged yoghurts cheese whey and some additional efforts must be carried out
and drinking yoghurts or those over their sell-by date, create to avoid its waste. The most noteworthy feature of this residue
an important amount of human foodstuff waste which is is the high content on fermentable sugars such as lactose,
unsuitable for sale, although in some cases it could be used for glucose and sucrose. Therefore, one of its possible uses is to be
* Corresponding author. Tel.: þ34 98 5103439; fax: þ34 98 5103434.

E-mail address: [email protected] (M. Dı́az).
0961-9534/$ – see front matter ª 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biombioe.2010.01.041
932 biomass and bioenergy 34 (2010) 931–938
considered as raw material in fermentations in order to

reduce its waste with substantially lower environmental 2. Materials and methods
impacts emphasizing the sustainable concept in the food
industry. Nevertheless, to date no studies have been focused 2.1. Microorganism
on the biotechnological use of residual yoghurt whey.
Lactic acid and its derivates are widely used in several The L. casei ATCC 393 strain, obtained from the American Type
sectors as food or pharmaceutical industries. Moreover, Culture Collection (Virginia, USA), was maintained frozen
recently there is an additional and increasing interest on lactic (in 40% v/v glycerol at 20 C). The strain was cultured on MRS
acid to be used for polylactic acid (PLA) production. broth (de Man Rogosa and Sharpe, Biokar Diagnostics, France)
This sustainable polymer made from renewable resources or in solid plates (containing 2% w/w agar).
is a substitute for petrochemical-based polymers, an
environmental-friendly biodegradable plastic material [4]. 2.2. Yoghurt whey preparation
Different renewable resources and agroindustrial wastes have
been used as fermentative substrate for lactic acid production, Yoghurts expired date were previously mixed in order to
such as corn [4], starch [5], rice bran [6], wheat flour [7] or date obtain yoghurt whey by heat treatment (thermal coagulation)
juice [8]. However, their potential use is limited by costly and removing fat and caseins. Next, the remaining fat was
laborious steps such as enzymatic hydrolysis or simultaneous withdrawn from yoghurt whey by centrifugation at
saccharification [9]. Furthermore, the use of yoghurt whey as 12,000 gravity for 10 min at 4 C. Dilution was performed
substrate for fermentation reduces the wasteful and costly with distilled water (1:1) and adjusted to pH 6.5 (by adding
disposal of unsuitable yoghurts for human consumption, NaOH 1 mol dm3). Once diluted, yoghurt whey was sterilized
advancing in its chain value. using a tangential microfiltration device equipped with a PVDF
Yeast extract adds suitable nutrients like minerals, vita- membrane-cassette (Millipore, Massachusetts, USA) 0.22 mm
mins and peptides that promote optimum growth of lactic pore size. Table 1 summarizes the features of yoghurt whey.
acid bacteria in low nitrogen nutrient media. An efficient
bacterial growth is promoted by the presence of growth 2.3. Culture and fermentation conditions
factors in yeast extract, although it is strongly influenced not
only by the kind of autolysate but also by culture medium The methodology employed on fermentation experiments is
composition [10]. Yeast extract are usually obtained from summarized in Fig. 1. After reactivation on MRS under
baker’s or spent brewer’s yeast by autolysis [11]. However, this microaerophilic conditions (without shaking), 10% (v/v) of
complex nutrient presents a high cost, which becomes this culture (exponential growth phase) was used to inocu-
a major shortcoming in fermentation processes. Moreover, late either 90 mL of yoghurt whey or yoghurt whey supple-
some food by-products are able to be considered as appro- mented with 2.5 g L1 of yeast extract as precultures.
priate supplements such as fish hydrolysates, spent brewer’s Precultures were incubated at 37 C for 20 h, at 100 r.p.m in
or cider’s yeast lees, an attractive replacement of yeast extract an orbital shaker (Infors HT, model Flyer Aerotron,
in fermentation processes [8,12]. Bottmingen, Switzerland). Subsequently, these precultures
The aim of this research work was to evaluate the were employed to inoculate yoghurt whey batch fermenta-
potential use of yoghurt whey as source for lactic acid tions. Uncontrolled pH fermentations were performed in
production by Lactobacillus casei through an eco-friendly a final working volume of 1 L at 37 C in an orbital shaker
process. The yoghurt whey was used as a feedstock in (100 r.p.m). Whereas, batch experiments with pH controlled
fermentations, so different strategies were performed either were performed in a 2 L bioreactor (Bioflo 110, New Bruns-
with or without pH control. In order to enhance fermenta- wick Scientific Co. Inc., NJ, USA) with a final working volume
tion, preculture supplementation with yeast extract was of 1 L, equipped with a digital control unit; temperature was
also tested. Likewise, another aim of this study was to find controlled at 37 C and the pH was set at 6.5 using NaOH
out the sugars consumption profile by L. casei throughout 3 mol dm3, agitation was set at 50 r.p.m to keep fermen-
the fermentation processes in order to determine the effi- tation medium mixed. Samples were collected at regular
cacy and the sequence of substrates consumption. By this intervals of 2 h to determine optical density and pH;
means, lactic acid production and sugar consumption was cell-free supernatants (0.45 mm filtered) were stored frozen
monitored by HPLC throughout the fermentation experi- (20 C) until analysis. Fermentations were carried out in
ments carried out. duplicates as two independent experiments.
Table 1 – Characterization of yoghurt whey used in fermentation experiments.

TKN Protein Density Brix Moisture COD pH
(mg L1) concentration (kg m3) (%) (g L1) (units)
(g L1)
Raw yoghurt whey 580 3.6 1042 13 91.6 90 4.2

Yoghurt whey medium for 309 1.9 1018 6.5 96.1 65 6.5
fermentation experiments
biomass and bioenergy 34 (2010) 931–938 933
Expired date yoghurts
Protein and fat

Thermal coagulation
removal
Centrifugation Protein and fat

removal
Concentrated yoghurt whey
Dilution and pH
adjustment to 6.5 with
NaOH
Sterilization by tangential Sterilization and

flow microfiltration protein removal
Inoculum
10 % (v/v)
Batch fermentation in yoghurt whey
Growth on yoghurt
10 % (v/v)
whey (20 h, 37ºC, 100
r.p.m.)
pH-controlled batch fermentation in
yoghurt whey
Inoculum
Lb. casei in MRS (at 10 % (v/v)
exponential growth phase)
Inoculum
10 % (v/v)
Batch fermentation in yoghurt whey
Growth on yoghurt
10 % (v/v)
whey supplemented with
yeast extract pH-controlled batch fermentation in
yoghurt whey
Inoculum
10 % (v/v)
Fig. 1 – Diagram that summarizes the processes to use residual yoghurt whey for lactic acid production as carried out in this
paper.
2.4. Bacterial growth Chem. Service (West Chester, Pennsylvania, USA), glucose
from Merck (Darmstadt, Germany) and ethanol from Panreac
Culture samples were centrifuged for 10 min at (Barcelona, Spain). Ethanol was not detected since L. casei is
16,000 gravity and subsequently the pellet was resuspended a facultative heterolactic fermentative microorganism,
in NaCl 0.7% (w/v). Bacterial growth was measured by optical although lactic acid is the dominant end product under typical
density at 660 nm (Shimazdu, UV 1203 model). Optical density fermentation conditions. Data acquisition and analysis were
data was converted to cell dry weight using the corresponding performed with Agilent ChemStation software.
calibration curve previously obtained, and results were
expressed as g L1.
2.5. Analytical methods
Protein content was determined according to Bradford [13]

(Sigma–Aldrich reagent, Steinheim, Germany) using bovine
serum albumin as standard.
Lactose, glucose, sucrose and lactic acid concentrations
from fermentation broth were measured by HPLC according to
a method described by Preduzzi et al. [14] with minor modi-
fications. The setup used for the analysis was a chromato-
graph Agilent, 1200 Series model (California, USA) with an
ICSep ICE-ION-300 column (Transgenomic Inc., San Jose,
California, USA). The following analytical conditions were
used: 450 mM H2SO4 as mobile phase (pH 3.1), column
temperature set at 75 C and flow rate of 0.3 mL min1 with the
Refractive Index as detector. Samples were quantified Fig. 2 – Lactobacillus casei growth evolution in residual
according to analytical external standards with HPLC grade; in yoghurt whey cultures with (6) and without yeast extract
that way, lactose, sucrose and lactic acid were obtained from supplementation (B).
Fig. 3 – (a) Lactose (A), sucrose (B) and glucose content (,), as well as lactic acid formation (6) through a pH uncontrolled
fermentation without yeast extract supplementation in preculture stage. (b) Biomass (expressed as dry cell weight) (:) and
protein (C) concentration of pH uncontrolled fermentation without yeast extract.
growth evolution in residual whey cultures, with or without

3. Results and discussion yeast extract supplementation. In the conditions tested, cells
reached exponential phase after 20 h of incubation. In the
Batch fermentation is usually used for lactic acid production supplemented medium, higher cell densities were obtained
at industrial scale. Previous to the fermentation process, the (around 1.8 g L1 versus 1.2 g L1 without supplementation).
raw yoghurt whey and the yoghurt whey sterilized (by filtra- The influence of yeast extract on lactic acid bacteria
tion) were characterized. The main parameters are summa- growth and on lactic acid yield has been verified by several
rized in Table 1. Since yoghurt whey presents a high content in works [10,11,15]. The peptide content from yeast extract may
sugars, it is a suitable substrate to obtain a value-added be positive since peptides and peptones stimulate all param-
product such as lactic acid. eters involved in fermentation [9]. The possibility of using
hydrolysed whey protein as supplementation has also been
3.1. Effect of yeast extract supplementation in residual investigated obtaining good lactic acid productivity, yield and
yogurt whey on bacterial growth high lactose consumption [15].
At laboratory scale yeast extract is usually utilized in
It is well known in fermentation studies that supplementation fermentation processes as supplementation [16]. Usually the
of proteins and nitrogen in general improves the bioconver- problem to perform medium supplementation is the high cost
sion of the substrates. In order to evaluate its effect on the of the protein supplement that makes the process economi-
growth of the lactic acid bacteria strain in the residual yogurt cally unfavourable. From an industrial practical approach, the
whey, commercial yeast extract was used. Fig. 2 shows L. casei culture medium should be easy to prepare and at low cost.
Fig. 4 – (a) Lactose (A), sucrose (B) and glucose content (,), as well as lactic acid formation (6) through a pH uncontrolled
fermentation with yeast extract supplementation in preculture stage. (b) Biomass (expressed as dry cell weight) (:) and
protein (C) concentration of pH uncontrolled fermentation with yeast extract supplementation in preculture.
After 34 h of fermentation, the quantity of lactic acid

obtained from both experiments was 14.5 and 12.25 g L1 for
pH uncontrolled batch culture with and without yeast extract
supplementation in preculture stage respectively. Sugar
consumption was more efficient in the case of fermentation
with yeast extract supplementation in precultures since it
achieved remaining final concentrations of 15.5 g L1 of
lactose, 13.47 g L1 of sucrose and 9.32 g L1 of glucose. On the
other hand, the remaining total sugar content without yeast
extract was higher, 42.54 g L1 versus 38.28 g L1 at the end of
the fermentation processes.
Differences in the pH values were observed in uncontrolled
and controlled fermentations, as shows Fig. 5. Whereas
fermentation without yeast extract reached a pH of 4.2 at 12 h,
the fermentation with yeast extract in preculture stage fell
Fig. 5 – pH evolution in batch pH uncontrolled under pH 4, maintaining the pH near 3.5. The different
fermentations, with (6) and without yeast extract evolution suggests a favourable and positive effect of yeast
supplementation in preculture stage (B). extract supplementation in the preculture stage, although the
behaviour is very similar within the first 12 h of the process.
Differences in the initial pH values were in accordance with
the acidity reached in the preculture. Hence, as it is shown in
Moreover, addition of supplements would increase the
Fig. 5 the pH inhibition and the low cell growth provoked the
impurity level in the fermentation broth which has to be
inability of the strain in using the remaining sugars. As the
separated for further product purification. For these reasons
fermentation continued, the rate slowed down because of
and based on the results obtained (Fig. 2) the potential
accumulation of lactic acid so pH values fell under 4.2 causing
capacity of the residual yogurt whey for lactic acid production
metabolic inhibition.
was investigated without the use of costly nutrients during
The lactic acid production of uncontrolled pH fermentation
batch fermentation. In this work, the advantage of using
was around 14 g L1. Other authors have found productions of
supplements as yeast extract only in the preculture stage was
10 g L1 of lactic acid from fermentation of cheese whey
tested, in order to compare the results obtained.
without pH control using Lactobacillus helveticus [17]. The pH
inhibition and the final lactic acid concentration were
3.2. Fermentation without pH control suggested to have a strong inhibitory effect.
Regarding to total protein content in fermentation broth, it
In accordance with the results shown in Fig. 3, the pH can be concluded that this parameter had not influence in the
uncontrolled fermentation proceeded slowly. A relatively high global fermentation profile since the initial concentration was
content of residual sugars were left on the fermentation broth maintained.
leading to a low yield.
Yeast extract supplementation in the preculture stage
(Figs. 3 and 4) had a positive effect on fermentation perfor- 3.3. Fermentation with pH control
mance. The lag phase was clearly reduced and at the same
time the lactic acid production associated to bacterial growth Figs. 6 and 7 show that L. casei consumes sugars at similar
was increased. rates, but in this case the higher lactic acid production, around
Fig. 6 – (a) Lactose (A), sucrose (B) and glucose content (,) and lactic acid formation (6) through a pH controlled
fermentation in bioreactor (37 8C). (b) Biomass (:) and protein (C) evolution throughout this fermentation.
Fig. 7 – (a) Lactose (A), sucrose (B) and glucose content (,) and lactic acid formation (6) through a pH controlled
fermentation with yeast extract supplementation in preculture stage. (b) Biomass (:) and protein (C) evolution throughout
this fermentation.
23 g L1 (Fig. 7), was linked to a better consumption of glucose processes from yoghurt whey. Carbohydrate consumption
and sucrose versus lactose. Protein content has not changed curves followed nearly the same pattern, but in each case the
throughout fermentations, both in pH-free and controlled amount of lactic acid produced was diverse due to the
batch processes. As mentioned, the possible effect observed different sugar consumption. Moreover, fermentation of
for yeast extract supplementation in preculture stage may be carbohydrates is achieved sequentially; this suggests hierar-
probably related to its complex composition. The undefined chical simultaneous sugar utilization with a control accom-
components of yeast extract, also rich in vitamins (specially plished by a regulatory mechanism.
the B group) showed growth-promoting properties on lactic Consequently, lactose is transported through the cyto-
acid bacteria as reported [10]. plasmic membrane by means of a permease as lactose phos-
The results (Figs. 6 and 7) show how the glucose and phate, followed by cleavage by b-galactosidase to yield glucose
sucrose were metabolized quickly in comparison to lactose. and galactose. Thereby, glucose and galactose are phosphor-
Nevertheless, remaining sugars were still present in fermen- ylated by glucokinase and metabolized through the glycolytic
tation broth at the end of the experiments; approximately pathway. Sucrose fermentation is mediated by a sucrose
15 g L1 of lactose remains in all experiments. The lactic acid phosphotransferase system, which is simultaneously cleaved
production suffered a significant increase at late stationary by a sucrose hydrolase to yield glucose and fructose [18,19].
growth phase and the clearly growth and productivity A conversion rate of 44% from initial total sugar to lactic
enhancement surged as result of yeast extract supplementa- acid was obtained in the best fermentation experiment, which
tion in the preculture stage, and therefore it meant a better
efficiency of the biomass reached (0.97 g L1) over the process
yield. The content of residual sugar varied from 15.2 g L1 of
lactose to 8.5 g L1 of sucrose, whereas in the same fermen-
tation procedure but without yeast extract supplementation
the final sugar values were 15.3 g L1 of lactose, 7 g L1 of
glucose and 10.6 g L1 of sucrose. Besides, total protein
content has been demonstrated not to be a key parameter of
the process since their concentration did not suffer any
modification.
The carbohydrate consumption by L. casei reflects its
ability to efficiently use carbon sources available in yoghurt
whey. Figs. 6a and 7a contain meaningful information
concerning the performance of carbohydrate metabolism by
L. casei. These results suggest a co-metabolic biotransforma-
tion by L. casei, since this strain is able to metabolize actively
glucose, lactose and sucrose. However, there is a preference
of carbohydrate consumption, thus, glucose is the first sugar Fig. 8 – Experimental values obtained corresponding to pH
metabolized, whereas sucrose and lactose content slightly controlled fermentation without yeast extract in preculture
decreased over time. Experimental results demonstrated that stage. Evolution of biomass (:), lactose (A), sucrose (B)
L. casei prefers glucose and sucrose over lactose as primary and glucose content (,), as well as lactic acid formation (6)
carbon and energy source throughout lactic acid production throughout batch culture.
Table 2 – Summary of values obtained in the yoghurt whey fermentation experiments.

Fermentation Fermentation Fermentation Fermentation with
without pH without pH with pH pH control and
control control and with control with yeast extract
yeast extract in in preculture stage
preculture stage
Initial lactose (g L1) 21.5 21.5 21.5 21.5

Initial glucose (g L1) 18.4 18.1 18.4 18.1
Initial sucrose (g L1) 19.5 19.1 19.5 19.0
Final lactose (g L1) 15 15.5 15.3 15.2
Final glucose (g L1) 11.78 9.32 7 6.4
Final sucrose (g L1) 15.76 13.47 10.6 8.5
Lactic acid produced (g L1) 12.25 14.5 20.2 25.9
Lactic acid yield on substrate (g g1) 0.72 0.71 0.76 0.9
Lactic acid productivity (g L1 h1) 0.36 0.426 0.59 0.76
proceeded without any nutrient adding in batch fermentation. and with a lactic acid yield on substrate of 0.72 g g1.
Although experiments performed showed an acceptable lactic Comparing with the uncontrolled pH fermentation, the pH
acid production, a remarkable enhancement in conversion control showed 64.8% improvement in lactic acid production.
yield was obtained with yeast extract supplementation in the Yeast extract supplementation in preculture stage showed an
preculture stage. In comparison with the same fermentation improvement of 28% in lactic acid production in pH controlled
procedure but without yeast supplementation in precultures, cultures.
the bioconversion diminished until 34%. Thereby, yoghurt Yeast extract favoured to enhance lactic acid maximum
whey has been shown as a favourable and suitable medium production, increasing the process yield. The lack of nitrogen
for lactic acid production. Another important aspect is the supplementation (either yeast extract or peptones) in the
L. casei ability to consume nearly all remaining sugars left in medium at the preculture stage induces an increase in lag
broth after 34 h of culture. It can be inferred observing the phase length of the culture as a result of the cellular adapta-
carbohydrate consumption patterns that L. casei presents tion to the whey [20].
different yields of lactic acid on sugar substrate, being the Finally, with the aim to determine the environmental
higher related with glucose and sucrose. A higher perfor- impact, chemical oxygen demand (COD) can be used as an
mance and fermentation efficiency could be achieved if indirect measurement of soluble and insoluble organic
fermentation time is increased up to 48 h obtaining a biocon- matter. The organic waste reduction can be achieved as result
version rate over 75% as can be observed in Fig. 8 for pH of microbial pure and mixed cultures in food by-products such
controlled fermentation and without yeast extract supple- as cheese whey [21]. In this work, COD was analyzed before
mentation in preculture stage. and after fermentation experiments in order to determine the
Table 2 summarizes the fermentation experiments carried theoretical reduction of the pollution. As it seen in Table 3, it
out in this work. As it is shown, supplementation with yeast can be concluded that fermentation process reduced the
extract in preculture stage and controlled pH promote lactic environmental impact (COD values dropped from 90.4 to
acid production. Moreover, it can be seen that highest lactic 72.5 g L1) although the increase on initial values were result
acid is reached with pH controlled conditions, whereas of yeast extract supplementation and the necessary use of
fermentation with yeast extract in preculture stage rendered MRS on previous preculture steps, maybe explained by the use
better productivity. The maximum lactic acid concentration of high organic content in precultures used as inoculum.
obtained was 25.9 g L1 corresponding to batch fermentation
with pH control and yeast extract supplementation in pre-
culture stage. Whereas, fermentations without pH control and
4. Conclusions
without yeast extract supplementation in preculture stage
were performed with a final value of 12.25 g L1 in lactic acid
Yoghurt whey, a waste material containing among other
constituents important concentrations of lactose, glucose and
sucrose, added in the yoghurt production process, has been
revealed as a suitable and direct substrate to lactic acid
production by L. casei ATCC 393. An adequate pH control has
Table 3 – Chemical oxygen demand values of yoghurt
shown to produce an important enhancement in carbohy-
whey throughout process.
drate bioconversion efficiency. In addition, microbial carbo-
Medium COD (g L1)
hydrate uptake and lactic acid production have been
Initial yoghurt whey 90 improved by the presence of surplus growth factors from
Yoghurt whey after tangential filtration 65 yeast extract in preculture stage. The evolution of the sugar
and before fermentation experiments
content has shown how glucose is metabolized more
Yoghurt whey after fermentation 72
efficiently by L. casei, and then how lactose and sucrose are
experiments without cells
consumed at similar rates. All sugars are metabolized at
different stages, already from the initial fermentation time. juice by Lactobacillus casei subsp. rhamnosus. Bioresource
Results suggests that L. casei was able to metabolize most Technology 2001;78:149–53.
glucose and sucrose within 34 h to produce 25.9 g L1 lactic [9] Mussatto SI, Fernandes M, Dragone G, Mancilha I, Roberto I.
Brewer’s spent grain as raw material for lactic acid
acid, but the lactose content left was 15.2 g L1. In spite of the
production by Lactobacillus delbrueckii. Biotechnology Letters
lactic acid levels achieved, residual sugar content could be 2007;29:1973–6.
further reduced. Alternative fermentation strategies could be [10] Gaudreau H, Champagne CP, Conway J, Degré R. Effect of
used to improve the efficiency compared with the traditional ultrafiltration of yeast extract on their ability to promote
batch culture avoiding inhibition effects. The results obtained lactic acid bacteria growth. Canadian Journal of Microbiology
here supported the interest in the biotechnological re-use of 1999;45:891–7.
[11] Chae HJ, Joo H, In MJ. Utilization of brewer’s yeast cells for
yoghurt-expired date whey, otherwise treated as a waste. In
the production of food-grade yeast extract. Part 1: effects of
brief, lactic acid production from a raw material as yoghurt
different enzymatic treatments on solid and protein recovery
proved to be effective through glucose, sucrose and lactose and flavor characteristics. Bioresource Technology 2001;76:
bioconversions. Thereby, it constitutes a sustainable and 253–8.
interesting bioprocess to obtain a value-added industrial [12] Herrero M, Garcı́a LA, Dı́az M. Malolactic bioconversion using
product from an accepted worthless feedstock until now. a Oenococcus oeni strain for cider production: effect of yeast
extract supplementation. Journal of Industrial Microbiology
and Biotechnology 2003;30:699–704.
[13] Bradford MM. A rapid and sensitive method for the
Acknowledgements quantification of microgram quantities of protein utilizing
the principle of dye-binding. Analytical Biochemistry 1976;
Authors thank the Spanish Science and Innovation Minister 72:248–54.
for the economical support with the project MEC07-CTQ62182. [14] Preduzzi I, Malvessi E, Mata VG, Silva EAB, Silveira MM,
Rodrigues AE. Quantification of lactobionic acid and sorbitol
from enzymatic reaction of fructose and lactose by high-
references performance liquid chromatography. Journal of
Chromatography A 2007;1145:128–32.
[15] Mussatto SI, Fernandes M, Mancilha IM, Roberto IC. Effects of
[1] ACNielsen. What is hot around the Globe: insights on growth medium supplementation and pH control on lactic acid
in food and beverages. New York: ACNielsen; 2006. production from brewer’s spent grain. Biochemical
[2] Sholten RHJ, van der Peet-Schwering CMC, Verstegen MWA, Engineering Journal 2008;40:437–44.
den Hartog LA, Schrama JW, Vesseur PC. Fermented [16] Arasaratnam V, Senthuran A, Balasubramaniam K.
co-products and fermented compound diets for pigs: Supplementation of whey with glucose and different
a review. Animal Feed Science and Technology 1999;82:1–19. nitrogen sources for lactic acid production by Lactobacillus
[3] Tiamine AY, Robinson RK. Yoghurt: science and technology. delbrueckrii. Enzyme and Microbial Technology 1996;19:482–6.
2nd ed. Cambridge, England: Woodhead Publishing; 1999. [17] Tango MSA, Ghaly AE. Effect of temperature on lactic acid
[4] Vink ETH, Rábago KR, Glassner DA, Gruber PR. Applications of production from cheese whey using Lactobacillus helveticus
life cycle assessment to NatureWorks polylactide (PLA) under batch conditions. Biomass and Bioenergy 1999;16:
production. Polymer Degradation and Stability 2003;80:403–19. 61–78.
[5] Altaf M, Naveena BJ, Reddy G. Use of inexpensive nitrogen [18] Salminen S, Wright A. Lactic acid bacteria, microbiology and
sources and starch for L(þ) lactic acid production in functional aspects. 2nd ed. New York: Marcel Dekker; 1998.
anaerobic submerged fermentation. Bioresource Technology [19] Barrangou R, Azcarate-Peril A, Duong T, Conners S, Kelly R,
2007;98:498–503. Klaenhammer T. Global analysis of carbohydrate utilization
[6] Gao M-T, Kaneko M, Hirata M, Toorisaka E, Hano T. Utilization by Lactobacillus acidophilus using cDNA microarrays. PNAS
of rice bran as nutrient source for fermentative lactic acid 2006;103:3816–21.
production. Bioresource Technology 2008;99:3659–64. [20] Amrane A. Seed culture and its effect on the growth and
[7] Oh H, Wee YJ, Yun JS, Han SH, Jung S, Ryu HW. Lactic acid lactic acid production of Lactobacillus helveticus. Journal of
production from agricultural resources as cheap raw General and Applied Microbiology 2003;49:21–7.
materials. Bioresource Technology 2005;96:1492–8. [21] Ustáriz F, Laca A, Garcı́a LA, Dı́az M. Mixed cultures of
[8] Nancib N, Nancib A, Boudjelal A, Benslimane C, Blanchard F, Serratia marcescens and Kluyveromyces fragilis for
Boudrant J. The effect of supplementation by different simultaneous protease production and COD removal of
nitrogen sources on the production of lactic acid from date whey. Journal of Applied Microbiology 2007;103:864–70.
102 Resultados
Resultados 103
4.2. Monitorización fisiológica de Lactobacillus casei
En este subapartado se ha monitorizado fisiológicamente la fermentación ácido-láctica

llevada a cabo por L. casei con el objetivo de producir ácido láctico a partir de lactosuero de
yoghurt. Así, se ha evaluado la influencia de diferentes condiciones operacionales y del
propio microambiente sobre la fisiología de L. casei. En este artículo se describe el
comportamiento del sistema bajo condiciones de presencia o ausencia en el control del pH y
de cómo la respuesta fisiológica de L. casei se ve influenciada por estas condiciones. A
diferencia de los sistemas fermentativos clásicos donde la acidificación puede significar la
pérdida de viabilidad, L. casei desarrolla en este caso una robustez fisiológica ante
condiciones desfavorables en contraposición con la respuesta fisiológica exhibida por el
microorganismo ante condiciones a priori mucho más favorables (bajo control de pH),
donde el lactato parece jugar un papel clave en la respuesta fisiológica. Bajo estas últimas
condiciones, L. casei experimenta una fuerte pérdida de la viabilidad celular, siendo las
subpoblaciones de células dañadas y muertas las mayoritarias en tiempos de fermentación
relativamente prolongados. En el trabajo se ha utilizado a la citometría de flujo multi-
paramétrica como herramienta en la monitorización y control de dichas respuestas
fisiológicas. Tanto la actividad metabólica como la integridad de la membrana han sido las
propiedades físiológicas evaluadas a lo largo de los sistemas fermentativos llevados a cabo,
con el fin de proponer las condiciones operacionales más adecaudas desde el punto de vista
fisiológico de L. casei. Estos resultados han demostrado la importancia de las condiciones
operacionales de cultivo sobre la viabilidad celular de L. casei, circunstancia que puede
llevar a un desaprovechamiento de la capacidad metabólica del microorganismo en
fermentaciones prolongadas en el tiempo.
Publicación: Physiological states of Lactobacillus casei fermentation in a mixed natural

substrate with lactose, glucose and sucrose.
Situación: En elaboración.
104 Resultados
Physiological states of Lactobacillus casei fermentation in a mixed

natural substrate with lactose, glucose and sucrose
Abstract
Although Lactobacillus casei is well-known as a potential lactic acid-producer microorganism with
substantial interest on industrial sphere, the influence of different bioprocessing conditions on its cellular
functionality and physiological status is not yet studied. In the present study, physiological states of L.
casei during a sugars-to-lactic acid bioprocess using residual yoghurt whey as inexpensive source were
monitored by multiparameter flow cytometry. L. casei displayed a robust physiological status under
uncontrolled-pH conditions despite the acid microenvironment encountered (pH<3.6), suggesting the
development of an acid tolerance response. Conversely, a stress-induced physiological response took
place when a controlled-pH cultivation strategy was adopted. The decline in metabolically active
subpopulation revealed that L. casei cells entered into a state of cellular damaged due to stressful
bioprocessing conditions encountered at stationary growth phase. Such conditions were featured by a
high lactate titer (41.5 g/L) which was deleterious to L. casei cells. Regardless the physiological response,
a lactic acid titer of 41.5 g/L with a yield of 0.91 g/g on sugars and a volumetric productivity of 0.86 g/L.h
were achieved under controlled-pH conditions. The sugar pattern metabolism with simultaneous co-
fermentation of glucose, sucrose and lactose was additionally maintained despite either the stress-
induced physiological or the acid tolerance responses developed by L. casei cells. These results provide
further understanding on the influence of acid and non-acid bioprocessing conditions on the physiological
responses of L. casei, suggesting additionally the feasibility of glucose, sucrose and lactose co-fermentation
as substrates for lactic acid production.
Keywords: multiparameter flow cytometry; lactic acid; Lactobacillus casei; acid tolerance
response; bioprocessing conditions; residual yoghurt whey.
1. Introduction
Fermentative production of lactic acid has steadily placed within the chemical industry as
a cost-effective and environmentally benign path for its large-scale manufacture (John et al.,
2009; Wee et al., 2006). However, the next step on this successful market consolidation
involves the use of cheap feedstocks for a sustainable bio-production through microbial
fermentation. Multiple raw materials such as molasses, starchy or lignocellulosic wastes
from agricultural and agro-industrial residues have been thus proposed as carbohydrate
rich substrates for lactic acid microbial production (Wee et al., 2006). In fact, residual
yoghurt whey has been recently employed as inexpensive source for lactic acid production,
reducing therefore the end-of-cycle environmental impact of this food processing waste
(Alonso et al., 2010). Unlike fermentable carbohydrate rich materials such as agricultural-
based residues, residual yoghurt whey may be converted into lactic acid with no pre-
Resultados 105
treatment in a single sugars-to-lactic acid fermentation process carried out by Lactobacillus

casei (Alonso et al., 2010). However, microbial fermentation systems are strongly reliant on
the cellular functionality and viability, which may impair the metabolic ability of the
microorganism to produce the desired target metabolite (Hewitt and Nebe-von-Caron,
2001; Lopes da Silva et al., 2009).
The characterization of healthy or nonviable microorganisms is therefore essential to
determine the real health status of the bioprocess, since changes in the underlying
physiology strongly influences not only on the cellular robustness but also on the biological
performance, leading consequently to a reduced bioprocess yield (Amanullah et al., 2002;
Díaz et al., 2010; Hewitt and Nebe-von-Caron, 2004). Multi-parameter flow cytometry has
recently emerged as a powerful tool in industrial biotechnology since it enables to monitor
physiological heterogeneity within industrially relevant bioprocesses (Alonso et al., 2012a;
Amanullah et al., 2002; Herrero et al., 2006; Quirós et al., 2007). Furthermore, bioprocess
monitoring through flow cytometry may provide insights into the right fermentation
strategy, enabling a better knowledge of the deleterious bioprocessing conditions on the
fermentation system otherwise not detected by conventional culture-dependent or biomass
estimation approaches (Hewitt et al., 2001; Lopes da Silva et al., 2009). The use of this
technique has been indeed helpful to monitor microbial responses at the population level
under hostile conditions such as freezing (Rault et al., 2007), starvation (Lopes da Silva et
al., 2005), pH- (Baatout et al., 2007) or bile salt-stress conditions (Ben Amor et al., 2002).
Environmental perturbations or deleterious bioprocessing conditions may undoubtedly
induce different patterns of physiological changes, reducing not only the microbial fitness
but also biological performance (Alonso et al., 2012a, 2012b). In fact, lactobacilli face several
stressing challenges during batch cultivation such as pH oscillation (Rault et al., 2009;
Wallenius et al., 2012), which could become a presumed bottleneck during lactic acid bio-
production under large-scale conditions. However, the impact of different bioprocessing
conditions on the physiological responses of L. casei during lactic acid bio-production
remains largely unexplored to date.
Within this context, the present study aims to evaluate the physiological responses of L.
casei during a sugars-to-lactic acid bioprocess using the mixed sugar composition of residual
yoghurt whey as inexpensive substrate. Physiological states of L. casei were therefore
evaluated and monitored through a multi-parameter flow cytometric assessment during
uncontrolled- and controlled-pH batch cultivations, with the aim of elucidating the
fermentation performance, cellular behaviour and robustness of L. casei under acid or non-
acid conditions. These results may provide valuable guidelines on the bioprocessing
106 Resultados
conditions for enhanced carbohydrate co-fermentation of L. casei under changing

environmental transitions at an industrial scale.
2. Material and methods

2.1. Microorganism
Lactobacillus casei ATCC 393, obtained from the American Type Culture Collection
(Virginia, USA), was maintained frozen (in 40 % [v/v] glycerol at -20ºC). This strain was
subsequently subcultured on MRS (de Man Rogosa and Sharpe, Biokar Diagnostic, France)
agar plates, incubated for 48 h at 30ºC and then preserved at 4ºC.
2.2. Yoghurt whey preparation
Whey from expired-date yoghurts was obtained by heat treatment (thermal coagulation)
in order to remove fat and major proteins. The remaining fat and proteins were
subsequently withdrawn from this liquid fraction by centrifugation at 12,000g for 10 min at
4ºC. The resulting yoghurt whey was 1-fold diluted with distilled water (1:1) and adjusted
to pH 6.5 (by adding NaOH 1 M) prior to sterilization using a tangential microfiltration
device equipped with a PVDF membrane-cassette of 0.22 μm pore size (Millipore,
Massachusetts, USA).
2.3. Culture conditions and fermentation experiments
After reactivation on MRS under microaerophilic conditions at 30ºC for 16 h, 10% (v/v)
of this culture was used to inoculate 90 mL of yoghurt whey. This preculture was incubated
at 37ºC for 20 h and placed at 100 rpm in an orbital shaker (Flyer Aerotron, Infors HT,
Bottmingen, Switzerland) and then it was used to seed batch fermentations. Controlled-pH
fermentation was performed in a 2-L bioreactor (Bioflo 110, New Brunswick Scientific Co.
Inc., NJ, USA) with a final working volume of 1 L. pH was strictly controlled at 6.5 by means
of computer-controlled peristaltic pumps via automatic addition of 3 M NaOH. Bioreactor
cultivation was conducted with an inoculation level of 10% (v/v), an agitation rate of 50
rpm, and at 37ºC.
Uncontrolled-pH batch fermentation was performed with a final working volume of 1 L at
37ºC in an orbital shaker at 100 rpm. In this case, the preculture stage was supplemented
with 2.5 g/L of yeast extract in order to promote acidic conditions during final batch
cultivation (Alonso et al., 2010). Samples from fermentation broths were withdrawn
periodically to determine bacterial growth and physiological status, whilst cell-free
supernatants were stored (at -20ºC) until further analysis. Cultivations were carried out in
duplicate as independent experiments.
2.4. Staining procedures
Resultados 107
Samples from cultures were harvested by centrifugation at 16,000g for 5 min. Before
staining, cells were washed twice in phosphate-buffered saline (PBS, pH 7.4, sterile and
filtered at 0.22 μm), and adjusted to cellular densities corresponding approximately to 4 x
105 cells/mL in the same buffer. For viability assessment, 200 μL from this cellular
suspension was then added to the different staining solutions previously prepared. Samples
were therefore stained with a mixture of two fluorescent dyes, ChemChrome V6 (CV6,
Chemunex, France) and Propidium Iodide (PI, Molecular Probes). CV6 is a fluorogenic ester
which only diffuses into viable cells and it is hydrolyzed by intracellular esterase enzymes to
give fluorescein, a fluorescent product (Parthuisot et al., 2000). Stock solution was prepared
by dilution 1:10 in sterile distilled water (0.22 μm filtered) and 8 μL from this solution was
added to the cell suspension which was incubated for 15 min in the dark at room
temperature. Whilst CV6 stains viable cells which are featured by enzymatic activity, PI is a
fluorescent nucleic acid dye which stains damaged and dead cells. PI binds to DNA to form a
red fluorescent DNA-complex which enables the assessment of the membrane integrity
since this staining dye can only cross the plasmatic membrane if it is permeabilized,
corresponding to those cells whose membranes are compromised. The stock solution of PI
(1 mg/mL solution in water, Molecular Probes) was both diluted in sterile distilled water
and added afterwards to the cell suspension at a final concentration of 5.25 μg/mL. This
mixture was incubated for 30 min at the same conditions as CV6 staining. Gates in the flow
cytometric dot plots were established according to control samples. Heat-killed cells
(treated at 90ºC for 30 min and immediately cooled on ice), early exponentially growing
cells and mixtures containing exponential and heat-killed cells (1:1) were therefore
analyzed. An unstained sample was additionally used as negative control.
2.5. Multiparameter flow cytometry
Flow cytometry measurements were performed using a Cytomics FC 500 flow cytometer
(Beckman Coulter) equipped with 488- and 633-nm excitation light source from an argon
ion laser. Green fluorescence from CV6-stained cells was collected on the FL1 channel (530
nm), whereas PI fluorescence was registered on the FL3 channel (610 nm). Each analysis
was performed in duplicate at a low flow rate setting (4,000 events/s). Fluorescent
microspheres (Perfect Count, Cytognos, Spain) were used as internal standards in each
sample. For every analysis 2,000 microspheres were acquired and performed at a low flow
rate setting. Both data acquisition and analysis were carried out using Cytomics RXP
software (Beckman Coulter).
2.6. Analytical determinations
108 Resultados
Bacterial growth was measured spectrophotometrically as optical density at 660 nm

(Shimazdu, UV 1203 model) after centrifugation of culture samples at 16,000g for 5 min.
Optical density data was converted to cell dry weight (expressed in grams per litre) using
the corresponding previously obtained calibration curve.
The lactic acid, lactose, glucose and sucrose content of cell-free culture samples were
measured by high performance liquid chromatography as reported previously by Alonso et
al. (2011). Samples were quantified according to analytical standards with HPLC grade:
lactose, sucrose and lactic acid were purchased from Chem Service (Pennsylvania, USA),
whereas glucose was obtained from Merck (Darmstadt, Germany). The liquid
chromatographic system used for analysis (Agilent 1200, Agilent Technologies Inc.,
California, USA) was equipped with an ICSep ICE-ION-300 column (Transgenomic Inc.,
California, USA) coupled to a refractive index detector. Sulphuric acid (0.450 mM, pH 3.1)
was employed as the mobile phase at a flow rate of 0.3 mL/min with the column
temperature set at 75ºC. Data acquisition and analysis were performed with ChemStation
software (Agilent).
2.7. Calculations
The yields of lactic acid on substrate (YLA/S in g/g) and on biomass (YX/S in g/g) were
calculated by dividing the amount of lactic acid or biomass produced, respectively, by the
total amount of sugars consumed. Specific sugar uptake rates (qS in g/g DCW h) and specific
lactic acid production rate (qLA in g/g DCW h) were calculated by linear regression of the
concentrations divided by the average biomass concentration at each interval.
3. Results
3.1. Assessment of L. casei physiological status
Figure 1 shows the flow cytometric dot plot disposition of L. casei cells, combining CV6
and PI as staining dyes. This dual parameter flow cytometric assessment enabled to
distinguish up to three subpopulations: metabolically active, damaged and dead cells (Fig.1).
Whereas healthy cells were not stained by PI due to the impermeability of the intact
membrane to this nucleic acid dye, damaged cells (double stained cells) exhibited both
enzymatic activity and compromised membranes. These sub-lethally injured cells are
featured for still maintaining metabolic activity as result of multiple changes in the gene
expression pattern which involves either low growth or reduced survival (Díaz et al., 2010;
Hewitt and Nebe-von-Caron, 2004). Likewise, healthy cells undergoing physiological stress
may be lead to either dead or damaged physiological status since membrane damage
involves energy depletion due to the lack of proton motive force (Díaz et al., 2010).
Resultados 109
A B
ChemChrome V6
C D
Propidium Iodide
Figure 1. Distribution of L. casei cells within a flow cytometric dot plot. Panels show double stained (CV6/PI
staining) samples used as flow cytometric controls: unstained cells (A), cells at early-exponential growth phase
(B), heat-killed cells (C) and mixture containing early-exponential phase and heat-killed cells (1:1) (D). Gate P in
dot plots shows healthy viable cells (CV6(+)/PI(-) cells), while double stained cells (CV6(+)/PI(+) cells)
representing sub-lethally damaged L. casei cells are shown in gate Y. Cells located in gate Z correspond to dead
cells (CV6(-)/PI(+) cells) that exhibited irreversible damages in plasmatic membranes.
3.2. Description of L. casei physiological states during uncontrolled-pH batch cultivation.

As can be seen in Fig. 2, three main subpopulations were clearly distinguishable
(metabolically active, damaged and dead cells) during uncontrolled-pH batch cultivation.
Bioprocess parameters reveal the existence of carbohydrate co-metabolism by L. casei on
the mixed carbohydrate composition of residual yoghurt whey, as previously reported
(Alonso et al., 2010). Both glucose and sucrose were consumed as the preferential sugars to
support L. casei growth and metabolism (Fig. 3a). Although batch cultivation was featured
by a predominant co-consumption of glucose and sucrose since the beginning, lactose
uptake was not lapsed at any time as shown in Fig. 3a. In fact, marked differences in the
specific carbohydrate consumption rates were found, with values of 0.5, 0.36 and 0.32
g/gDCW.h for glucose, sucrose and lactose, respectively. This uncontrolled-pH fermentation
cultivation was consequently not influenced by the unavailability of preferential
carbohydrates since 30.6 g/L of residual sugars were left after 48 h (Fig. 3a).
110 Resultados
0h 16 h 24 h
ChemChrome V6
40 h 48 h
Propidium Iodide
Figure 2. Cytograms representing physiological subpopulations of L. casei during uncontrolled-pH fermentation
for lactic acid bio-production from yoghurt whey. Cultivation was dominated by three main functional
subpopulations: metabolically active, damaged and dead cells.
In terms of lactic acid production, a titer of 18.8 g/L with a yield of 0.67 g/g on sugars and
a volumetric productivity of 0.39 g/L.h were achieved after 48 h (Table 1). Obviously, the
absence of pH-control caused end-product inhibition and subsequently metabolic feedback
repression, leading therefore to a slow hierarchical bioconversion of carbohydrates into
lactic acid (Fig. 3a).
As regards physiological heterogeneity, cultivation displayed a robust physiological
status, remaining metabolically active most of L. casei cells (~70%) throughout cultivation
(Fig. 3b). The onset of the stationary growth phase was thus accompanied by the
maintenance on viability values notwithstanding of the acidity broth reached (pH<3.6). Both
damaged and dead subpopulations remained therefore at values of 8 and 24%, respectively,
throughout uncontrolled-pH batch cultivation.
Resultados 111
Glucose (○); Sucrose (▼); Lactose (●)
Figure 3. Time-course profile of bioprocess parameters obtained during uncontrolled-pH cultivation (A).
Percentages of L. casei cell subpopulations throughout the uncontrolled-pH fermentation (B).
Table 1. Comparison of bioprocess parameters obtained in each fermentation process.

Fermentation process
Uncontrolled-pH Controlled-pH
Lactic acid production, PLA (g/L) 18.8 41.5
Yield on substrate, YLA/S (g/g) 0.67 0.91
Yield on biomass, YX/S (g/g) 0.02 0.02
Volumetric productivity, QLA (g/L h) 0.39 0.86
Residual total sugars (g/L) 30.6 13.7
Specific glucose uptake rate, qGlc (g/g DCW h) 0.50 0.79
Specific sucrose uptake rate, qSuc (g/g DCW h) 0.36 0.51
Specific lactose uptake rate, qLac (g/g DCW h) 0.32 0.40
Specific lactic acid production rate (g/g DCW h) 0.76 1.20

112 Resultados
3.3. Controlled-pH batch cultivation: physiological heterogeneity and description of L. casei

physiological states.
In most lactobacilli fermentations, lactic acid yield can be ameliorated avoiding end-
product inhibition by pH-control cultivation strategies (Schepers et al., 2002; Wallenius et
al., 2012). In order to further clarify the role of pH-control strategy on L. casei behaviour,
batch cultivation was carried out at 6.5. As shown in Fig. 4, cytograms revealed that
controlled-pH cultivation suffered from a progressively increase in damaged and dead
subpopulations. Conversely, controlled-pH cultivation proceeded at higher lactic acid
production and sugar consumption rates than uncontrolled-pH batch culture (Fig. 5a). Lactic
acid volumetric productivity was thus remarkably enhanced, resulting in a 1.2-fold (0.86
g/L.h) higher than that achieved under acidic conditions (0.36 g/L.h). Table 1 shows a
comparison between bioprocess values obtained under uncontrolled- and controlled-pH
conditions.
0h 16 h 24 h
ChemChrome V6
40 h 48 h
Propidium Iodide
Figure 4. Dot plots representing CV6 fluorescence versus PI fluorescence of L. casei cells during controlled-pH
batch cultivation at 6.5 on residual yoghurt whey.
L. casei cells also displayed simultaneous carbohydrate co-consumption as Fig. 5a shows,

although meaningful higher sugar uptake (0.79, 0.51 and 0.4 g/gDCW.h for glucose, sucrose
and lactose, respectively) and lactic production rates (1.2 g/gDCW.h) were attained in
comparison to those achieved under uncontrolled-pH conditions (Table 1). Interestingly, the
Resultados 113
culture was featured by the depletion in the content of preferential sugars at later stages,
leaving only 2.2 and 2.7 g/L of glucose and sucrose after 48 h, respectively (Fig. 5a). As Fig.
5a shows, the pattern in sugar co-utilization was not modified, being glucose and sucrose
the preferred carbon sources. Depending on the mixed sugars, microorganisms may exhibit
sequential, simultaneous or mixed consumption patterns. In fact, Adler et al. (2012) have
demonstrated that the availability of preferred sugars during lactic acid production from
lignocellulose may induce changes in the consumption behaviour of Pediococcus
pentosaceus.
As can be seen in Fig. 5b, L. casei displayed a stress-induced physiological response
during controlled-pH cultivation. Physiological-state dynamics of L. casei showed therefore
that dead and damaged subpopulations increased concomitantly to a decline in
metabolically active subpopulation. The progressive decrease in cellular viability (from 70
to 40%) after the onset of the stationary growth phase suggests that the amount of lactate
produced (41.5 g/L) led to stressful bioprocessing conditions for L. casei cells.
Glucose (○); Sucrose (▼); Lactose (●)
Figure 5. Time-course profile of bioprocess parameters obtained during controlled-pH batch cultivation (A).
Percentages of L. casei cell subpopulations throughout the controlled-pH fermentation (B).
114 Resultados
4. Discussion
While the current lactic acid bio-production market is based on the use of fermentable
sugars derived from food crops such as sugarcane and corn, recent efforts have focused on
the conversion of inexpensive feedstocks such as lignocellulose and agro-industrial wastes
to lactic acid, which can emphasize the cost-effectiveness and sustainability of fermentation
systems (Alonso et al., 2010; John et al., 2009; Wee et al., 2006). L. casei is a well-known
microbial platform for industrial-scale bioproduction of lactic acid not only due to its high
metabolic efficiency but also its ability for producing L-lactic acid. However, significant
stress physiological responses may hinder biological performance of L. casei. In the present
study, physiological-state dynamics of L. casei during an efficient sugars-to-lactic acid
fermentation system was monitored through multiparameter flow cytometry.
Lactobacilli face with multiple and diverse environmental changes during bioprocessing
which may trigger relevant physiological responses on a population level. Nutrient
starvation, inappropriate pH or osmotic strength may indeed lead to an adaptive
physiological response by lactobacilli to these stressing conditions (van de Guchte et al.,
2002). Among those bioprocessing challenges, pH may a play a fundamental role in the
cellular homeostasis of microorganisms (Nicolau et al., 2010). In fact, stress-induced
physiological responses have been found in industrially-relevant fermentation systems due
to acid bioprocessing conditions, reducing therefore fermentation performance (Alonso et
al., 2012a, 2012b).
Despite the acid environment, L. casei displayed a robust physiological status under
uncontrolled-pH conditions, suggesting the development of an acid tolerance response (Fig.
3b). These mild acidic conditions (pH<3.7 from 16 h onwards) were not detrimental on the
metabolically active subpopulation, which remained at values around 70% during
stationary growth phase. Whereas pH stress may result in a loss of cellular functionality and
viability in diverse microorganisms (Alonso et al., 2012a, 2012b; Baatout et al., 2007),
Lactobacilli are capable of inducing an acid tolerance response in response to mild acid
conditions, including pH homeostasis, protection and repair mechanisms (van de Guchte et
al., 2002). In contrast, the synergistic negative influence of respiration and pH stress may
involve the formation of radicals and loss of membrane integrity in Lactobacillus plantarum
(Watanabe et al., 2012). Specifically, the acid tolerance response in L. casei involves a
diverse transcriptional response for enhanced acidurance (Broadbent et al., 2010). In fact, L.
casei combats acid stress by maintaining cell membrane functionality through alteration in
membrane fluidity and fatty acid distribution (Wu et al., 2012). Besides, the maintenance of
Resultados 115
intracellular pH during acidification of the environment requires the modification of cellular

metabolism (Rault et al., 2009). Genes related to H+-ATPase are therefore upregulated in L.
casei in order to maintain cytoplasmic pH at values close to neutral (Chen et al., 2009).
Maintaining the extracellular pH at a high value stabilizes intracellular pH of lactobacilli,
decreasing therefore the inhibiting effect of lactic acid (van de Guchte et al., 2002).
Conversely, L. casei cells could not withstand stressful bioprocessing conditions imposed
by dissociated lactate accumulation (up to 41.5 g/L) under controlled-pH conditions (Fig. 5).
These stressful bioprocessing conditions encountered during lactate-producing phase lead
therefore to a progressive drop in metabolically active cells due to the damage in membrane
integrity. Lactobacilli viability may be particular sensitive to lactate, including an inhibitory
growth effect of dissociated form (Yáñez et al., 2008). Rault et al. (2009) have addressed the
effect of dissociated lactate level on Lactobacillus bulgaricus physiological status. These
authors have found that lactate may be important chemical stressor for the lactic acid
bacteria during controlled-pH cultivations carried out at pH values of 6. The elevated
content of the ionic lactate form at high pH values damaged thus membrane integrity that
led to a viability decrease (Rault et al., 2009). Viability of Lactobacillus helveticus cultures
was indeed better at pH 4.2 than cultivations performed at pH 5.5 (Schepers et al., 2002).
Overall, these results suggest that strategies involving continuous cultivation of L. casei at
high pH control values may result not only in a long-term loss of metabolic efficiency but
also in a loss of cellular functionality due to the lack of membrane integrity.
Both fermentative ability and metabolic efficiency of L. casei were not altered due to the
physiological deleterious bioprocessing conditions encountered during pH-controlled
cultivation (Fig. 5a). In terms of lactic acid bio-production, titer was 2.2-fold (41.5 g/L)
higher in controlled-pH cultivation than that achieved under acidic conditions (18.8 g/L).
Interestingly, sugar consumption profile did not change under these circumstances. The
simultaneous co-fermentation of glucose, sucrose and lactose was therefore maintained
despite the stress-induced physiological response or acid tolerance response developed by
L. casei cells under controlled- and uncontrolled-pH conditions, respectively (Figs. 3 and 5).
Furthermore, the revealed preferential consumption pattern (glucose>sucrose>lactose) was
maintained over the course of the cultivations, suggesting that availability of preferred
sugars governed and adjusted the specific carbohydrate palate of L. casei regardless the
physiological status of the culture.
These results provided a link between the physiological status and lactic acid production
by L. casei, representing a potential target for cellular robustness efforts designed to
produce higher titer of lactic acid from carbohydrate rich sources. Information from
116 Resultados
multiparameter flow cytometry assessment provided additionally valuable guidelines on the

bioprocessing conditions for enhanced co-fermentation of L. casei under changing
environmental transitions. Therefore, a lower pH control value may allow the L. casei cells
not only to maintain a more robust physiological state but also to avoid the end-product
inhibition of uncontrolled-pH cultivation strategies. Whereas lactic acid affects pH
homeostasis of L. casei leading to an inhibition of cellular reactions, lactate concentration
may strongly affect membrane integrity and cellular viability.
5. Conclusions
In the present study, results have revealed for the first time how the bioprocessing
conditions during an industrially relevant sugars-to-lactic acid bioprocess can lead to
cellular stress responses with substantially deleterious effects on the physiological status of
L. casei. Whereas poor co-fermentation performance was obtained under acidic
bioprocessing conditions (obtaining a lactic acid titer of 18.8 g/L with a yield of 0.67 g/g on
sugars and a volumetric productivity of 0.39 g/L.h), L. casei cells displayed an acid tolerance
response featured by a robust physiological status despite the acidic conditions encountered
(pH<3.6). Conversely, controlled-pH cultivation was featured by a remarkably drop in the
viability associated to the presence of high dissociated lactate titer (41.5 g/L). Addressing
therefore the deleterious influence of such bioprocessing conditions constitutes valuable
information for improving fermentation performance since this flow cytometry approach
may provide guidelines on L. casei behaviour otherwise not detected by conventional
methodologies.
Acknowledgements
Authors thank the financial support from the Spanish Ministry of Science and Innovation by
the project MEC07-CTQ62182. Authors also wish to thank Covadonga Quirós for her
invaluable advices in setting up the flow cytometric protocols. Ana Salas is gratefully
acknowledged for her technical assistance (Flow Cytometry Area, Scientific-Technical
Services, University of Oviedo).
References
Adler, P., Song, H.S., Kästner, K., Ramkrishna, D., Kunz, B., 2012. Prediction of dynamic
behavior of Pediococcus pentosaceus producing lactic acid from lignocellulosic sugars.
Biotechnol. Prog. 28, 623-635.
Resultados 117
Alonso, S., Herrero, M., Rendueles, M., Díaz, M., 2010. Residual yoghurt whey for lactic acid
production. Biomass Bioenerg. 34, 931-938.
Alonso, S., Rendueles, M., Díaz, M., 2011. Efficient lactobionic acid production from whey by
Pseudomonas taetrolens under pH-shift conditions. Bioresour. Technol. 102, 9730-
9736.
Alonso, S., Rendueles, M., Díaz, M., 2012a. Physiological heterogeneity of Pseudomonas
taetrolens during lactobionic acid production. Appl. Microbiol. Biotechnol. 96, 1465-
1477.
Alonso, S., Rendueles, M., Díaz, M., 2012b. Selection method of pH conditions to establish
Pseudomonas taetrolens physiological states and lactobionic acid production. Appl.
Microbiol. Biotechnol. In press. doi: 10.1007/s00253-012-4607-x
Amanullah, A., Hewitt, C.J., Nienow, A.W., Lee, C., Chartrain, M., Buckland, B.C., Drew, W.,
Woodley, J.M., 2002. Fed-batch bioconversion of indene to cis-indandiol. Enzyme
Microb. Technol. 31, 954-967.
Baatout, S., Leys, N., Hendrickx, L., Dams, A., Mergeay, M., 2007. Physiological changes
induced in bacteria following pH stress as a model for space research. Acta Astronaut.
60, 451-459.
Ben Amor, K., Breeuwer, P., Verbaarschot, P., Rombouts, F.M., Akkermans A.D.L., De Vos,
W.M., Abee, T., 2002. Multiparametric flow cytometry and cell sorting for the
assessment of viable, injured, and dead Bifidobacterium cells during bile salt stress.
Appl. Environ. Microbiol. 68, 5209-5216.
Broadbent, J.R., Larsen, R.L., Deibel, V., Steele, J.L., 2010. Physiological and transcriptional
response of Lactobacillus casei ATCC 343 to acid stress. J. Bacteriol. 192, 2445-2458.
Chen, X., Sun, Z., Meng, H., Zhang, H., 2009. The acid tolerance association with expression of
H+-ATPase in Lactobacillus casei. Int. J. Dairy Technol. 62, 272-276.
Díaz, M., Herrero, M., García, L.A., Quirós, C., 2010. Application of flow cytometry to
industrial microbial bioprocesses. Biochemical Eng. J. 48, 385-407.
Herrero, M., Quirós, C., García, L.A., Díaz, M., 2006. Use of flow cytometry to follow the
physiological states of microorganisms in cider fermentation processes. Appl. Environ.
Microbiol. 72, 6725-6733.
Hewitt, C.J., Nebe-von-Caron, G., 2001. An industrial application of multiparameter flow
cytometry: assessment of cell physiological state and its application to the study of
microbial fermentations. Cytometry. 44, 179-187.
118 Resultados
Hewitt, C.J., Nebe-von-Caron, G., 2004. The application of multi-parameter flow cytometry to
monitor individual microbial cell physiological state. Adv. Biochem. Eng. Biot. 89,197-
223.
John, R.P., Anisha, G.S., Nampoothiri, K.M., Pandey, A., 2009. Direct lactic acid fermentation:
focus on simultaneous saccharification and lactic acid production. Biotechnol. Adv. 27,
145-152.
Lopes da Silva, T., Piekova, L., Mileu, J., Roseiro, J.C., 2009. A comparative study using the
dual staining flow cytometric protocol applied to Lactobacillus rhamnosus and Bacillus
licheniformis batch cultures. Enzyme Microb. Technol. 45, 134-138.
Lopes da Silva, T., Reis, A., Kent, C.A., Kosseva, M., Roseiro, J.C., Hewitt, C.J., 2005. Stress-
induced physiological responses to starvation periods as well as glucose and lactose
pulses in Bacillus licheniformis CCMI 1034 continuous aerobic fermentation processes
as measured by multi-parameter flow cytometry. Biochemical Eng. J. 24, 31-41.
Nicolau, S.A., Gaida, S.M., Papoutsakis, E.T., 2010. A comparative view of metabolite and
substrate stress and tolerance in microbial bioprocessing: from biofuels and
chemicals, to biocatalysis and bioremediation. Metab. Eng. 12, 307-331.
Parthuisot, N., Catala, P., Lemarchand, K., Baudart, J., Lebaron, P., 2000. Evaluation of
ChemChrome V6 for bacterial viability assessment in waters. J. Appl. Microbiol. 89,
370-380.
Quirós, C., Herrero, M., García, L.A., Díaz, M., 2007. Application of flow cytometry to
segregated kinetic modelling based on physiological states of microorganisms. Appl.
Environ. Microbiol. 73, 3993-4000.
Rault, A., Béal, C., Ghorbal, S., Ogier, J.C., Bouix, M., 2007. Multiparametric flow cytometry
allows rapid assessment and comparison of lactic acid bacteria viability after freezing
and during frozen storage. Cryobiology. 55, 35-43.
Rault, A., Bouix, M., Béal, C., 2009. Fermentation pH influences the physiological-state
dynamics of Lactobacillus bulgaricus CFL1 during pH-controlled culture. Appl. Environ.
Microbiol. 75, 4374-4381.
Schepers, A. W., Thibault, J., Lacroix, C., 2002. Lactobacillus helveticus growth and lactic acid
production during pH-controlled batch cultures in whey permeate/yeast extract
medium. Part I, Multiple factor kinetic analysis. Enzyme Microb. Technol. 30, 176-186.
van de Guchte, M., Serror, P., Chervaux, C., Smokvina, T., Ehrlich, S.D., Maguin, E., 2002. Stress
responses in lactic acid bacteria. Antonie Leeuwenhoek 82, 187-216.
Watanabe, M., van der Veen, S., Abee, T., 2012. Impact of respiration on acid stress resistance
of Lactobacillus plantarum WCFS1. Appl. Environ. Microbiol. 78, 4062-4064.
Resultados 119
Wallenius, J., Barth, D., Eerikäinen, T., 2012. The effects of pH oscillation on Lactobacillus
rhamnosus batch cultivation. Appl. Microbiol. Biotechnol. 95, 1265-1273.
Wee, Y.J., Kim, J.N., Ryu, H.W., 2006. Biotechnological production of lactic acid and its recent
applications. Food Technol. Biotechnol. 44, 163-172.
Wu, C., Zhang, J., Wang, M., Du, G., Chen, J., 2012. Lactobacillus casei combats acid stress by
maintaining cell membrane functionality. J. Ind. Microbiol. Biotechnol. 39, 1031-1039.
Yáñez, R., Marques, S., Gírio, F.M., Roseiro, J.C., 2008. The effect of acid stress on lactate
production and growth kinetics in Lactobacillus rhamnosus cultures. Process Biochem.
43, 356-361.
120 Resultados
Resultados 121
4.3. Producción de ácido lactobiónico por Pseudomonas taetrolens
En este trabajo se ha descrito la utilización de suero lácteo para la obtención de ácido

lactobiónico por fermentación llevada a cabo por Pseudomonas taetrolens. El presente
trabajo ha contribuido significativamente en el avance del conocimiento asociado a la
producción biotecnológica de ácido lactobiónico puesto que introduce conceptos nuevos, no
descritos hasta ese momento. Así, en el mismo se ha definido por primera vez cómo existe
una fase de proliferación celular y otra etapa de producción asociada a la formación de ácido
lactobiónico por P. taetrolens (se trata, por lo tanto, de un compuesto que en su génesis es
similar a un metabolito secundario). Igualmente, se ha desarrollado un sistema basado en el
control de pH en la fase de producción (pH-shift strategy) con el fin de reducir la posible
inhibición causada por la acidez generada en un proceso sin control de pH. También se ha
enfatizado en los siguientes conceptos como son: el empleo del suero lácteo como materia
prima, el incremento del pH asociado a la fase de proliferación celular (salto en el pH o pH-
shift value), la entrada en fase de producción asociada al incremento en los valores de
oxígeno disuelto, el efecto positivo del tamaño de inóculo tanto a nivel de matraz como de
biorreactor sobre el rendimiento y la productividad del sistema (variable habitualmente
infravalorada en otros sistemas), el desarrollo del proceso sin control de pH a escala de
matraz, la influencia del tamaño de inóculo sobre la tasa específica de crecimiento o el
desarrollo de una nueva metodología de screening oxidativo en placa de agar no descrita
hasta la fecha. Todos ellos han sido conceptos que se recogen por primera vez en el artículo
científico que a continuación se muestra.
En suma, en este trabajo se ha definido por primera vez un sistema de producción
biotecnológica de ácido lactobiónico caracterizado por el empleo de un sustrato con bajo
coste. Asimismo, en este trabajo se sentaron las bases para posteriores implementaciones
del bioproceso, que servirían para incrementar los valores asociados a la producción
biotecnológica.
Publicación: Efficient lactobionic acid production from whey by Pseudomonas taetrolens

under different pH-shift conditions.
Situación: Artículo publicado en Bioresource Technology (Elsevier).
Bioresource Technology 102 (2011) 9730–9736
Contents lists available at SciVerse ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
Efficient lactobionic acid production from whey by Pseudomonas taetrolens

under pH-shift conditions
Saúl Alonso, Manuel Rendueles, Mario Díaz ⇑
Department of Chemical Engineering and Environmental Technology, University of Oviedo, Faculty of Chemistry, C/Julián Clavería s/n, 33071 Oviedo, Spain
a r t i c l e i n f o a b s t r a c t
Article history: Lactobionic acid finds applications in the fields of pharmaceuticals, cosmetics and medicine. The produc-
Received 7 May 2011 tion of lactobionic acid from whey by Pseudomonas taetrolens was studied in shake-flasks and in a biore-
Received in revised form 20 July 2011 actor. Shake-flask experiments showed that lactobionic acid was a non-growth associated product.
Accepted 22 July 2011
A two-stage pH-shift bioconversion strategy with a pH-uncontrolled above 6.5 during the growth phase
Available online 29 July 2011
and maintained at 6.5 during cumulative production was adopted in bioreactor batch cultures. An inoc-
ulation level of 30% promoted high cell culture densities that triggered lactobionic acid production at a
Keywords:
rate of 1.12 g/L h. This methodology displayed efficient bioconversion with cheese whey as an inexpen-
Whey
Lactobionic acid
sive substrate for lactobionic acid production.
Pseudomonas taetrolens Ó 2011 Elsevier Ltd. All rights reserved.
Biotechnological upgrading
Oxidative bioconversion
1. Introduction stabilizer-gelling agent (Gerling, 1998) and has been approved as

a food additive by the US FDA. This substance is also considered
Currently, about 30% of annual world cheese whey production to confer potential prebiotic effects in functional foods (Saarela
(177 million tonnes in 2006) remains underutilized, ending up as et al., 2003; Schaafsma, 2008).
waste or being employed as animal feed (Affertsholt, 2007). Envi- At present, lactobionic acid is obtained via chemical synthesis,
ronmental and economic concerns demand that more of the whey but this methodology involves harmful catalysts in an energy-
should be converted to value-added products. Whey has been intensive process and generates undesirable side-reaction products
explored as a source for producing polyhydroxyalkanoates (Koller (Chia et al., 2008; Kuusisto et al., 2007; Murzina et al., 2008). Enzy-
et al., 2008; Povolo et al., 2010), ethanol (Guimarães et al., 2010) matic synthesis by oxidation of lactose is possible (Satory et al.,
or hydrogen (Venetsaneas et al., 2009), and the conversion of lac- 1997; Nordkvist et al., 2007; Van Hecke et al., 2009), but the pro-
tose into lactobionic acid has been proposed as a novel promising cess is costly and requires redox mediators or cofactor regenera-
solution (Gänzle et al., 2008). tion (Splechtna et al., 2001; Ludwig et al., 2004; Maischberger
Lactobionic acid displays antioxidant, biodegradable, biocom- et al., 2008; Van Hecke et al., 2009).
patible and chelating properties that make it useful in biodegrad- Whole-cell bioconversion for lactobionic acid production carried
able detergents (Gerling, 1998), anti-corrosive coatings (Gerling, out by Pseudomonas sp. proceeds slowly and yields lower biocon-
1998) as well as sugar-based surfactants (Oskarsson et al., 2007). version rates (Stodola and Lockwood, 1947) than enzymatic
Lactobionic acid is used in the manufacture of pharmaceuticals, approaches (Pedruzzi et al., 2011), but improvements in yield and
antibiotics and preservative solutions for organ transplants that costs could be achieved by developing efficient production strate-
suppress tissue damage during organ storage (D’Alessandro et al., gies and utilizing cheap raw materials such as whey. Therefore, in
1999; Hart et al., 2002). It is also employed as an anti-aging and the present study, a pH-shift cultivation strategy was devised to
keratinizing agent in cosmeceutical formulations (Green et al., increase bioconversion efficiency by varying inoculum age and size.
2009), and it has attracted attention as part of drug delivery sys-
tems (Lin et al., 2009; Wu et al., 2009), biocompatible scaffolds 2. Methods
for tissue engineering (Chung et al., 2002) and nanoplatforms for
new bioimaging applications (Selim et al., 2007; Kekkonen et al., 2.1. Microorganism and inoculum preparation
2009). Lactobionic acid is used as a preservative, antioxidant and
2.1.1. Microorganism
⇑ Corresponding author. Tel.: +34 98 5103439; fax: +34 98 5103434. Pseudomonas taetrolens LMG 2336, obtained from the Belgian
E-mail address: [email protected] (M. Díaz). Coordinated Collection of Microorganisms (Gent, Belgium), was
0960-8524/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2011.07.089
S. Alonso et al. / Bioresource Technology 102 (2011) 9730–9736 9731
maintained frozen (in 40% [v/v] glycerol at 20 °C). This strain was 100 mL of sweet whey. Subsequently, cells were grown in cheese
subsequently subcultured on Nutrient Broth (NB, containing 1 g/L whey at 250 rpm and 30 °C in an orbital shaker for 12, 24 and
meat extract, 2 g/L yeast extract, 5 g/L peptone and 5 g/L NaCl) agar 36 h. Each culture was employed as seed culture in the bioreactor
plates, incubated for 48 h at 30 °C and then preserved at 4 °C. at 10% (v/v) containing sweet whey to study the effect of inoculum
age on culture parameters. The influence of inoculation level on
2.1.2. Inoculum preparation lactobionic acid production was studied with inoculation levels
A loopful of P. taetrolens from a fresh Nutrient Broth agar plate of 5%, 10%, 20% or 30% (v/v) of seed cultures cultivated for 12 h
was used to inoculate a 500 mL Erlenmeyer flask containing in sweet whey at 250 rpm and 30 °C. Cultivations were carried
100 mL of Nutrient Broth medium. This flask was incubated on out in duplicate as independent experiments.
an orbital shaker (Infors HT, model Flyer Aerotron, Bottmingen,
Switzerland) at 250 rpm and 30 °C for 10 h. Actively growing cells 2.5. Analytical methods
from this culture were then employed as inoculum for the produc-
tion of lactobionic acid in shake flasks and bioreactor seed cultures Bacterial growth was measured spectrophotometrically as opti-
containing sweet whey, as subsequently reported. cal density at the 600 nm (Shimazdu, UV 1203 model) after centri-
fugation of culture samples at 16,000g for 5 min. Optical density
2.2. Sweet whey preparation data was converted to cell dry weight (expressed in grams per li-
ter) using the corresponding calibration curve previously obtained.
Sweet whey (kindly provided by ILAS, Asturias, Spain) was Lactobionic acid and lactose contents of culture samples were
1-fold diluted with distilled water (1:1) and adjusted to pH 6.5 measured by high performance liquid chromatography following
(by adding NaOH 6 N) prior to sterilization using a tangential a modified version of a previously published method (Pedruzzi
microfiltration device equipped with a PVDF membrane-cassette et al., 2007). Analyses were carried out using a liquid chromatogra-
of 0.22 lm pore size (Millipore, Massachusetts, USA). phy system (Agilent 1200, Agilent Technologies Inc., California,
USA) equipped with an ICSep ICE-ION-300 column (Transgenomic
2.3. Shake-flask bioconversions Inc., California, USA) coupled to a refractive index detector. As the
mobile phase, 0.450 mM H2SO4 (pH 3.1) was employed at a flow
Shake-flask bioconversion experiments were conducted in rate of 0.3 mL/min with the column temperature set at 75 °C. Sam-
500 mL Erlenmeyer flasks containing 100 mL of sweet whey and ples were quantified according to HPLC-grade external analytical
inoculated with 0.035, 0.07, 0.14, or 0.64 g/L of wet biomass from standards, lactose was obtained from Chem Service (Pennsylvania,
NB precultures harvested by centrifugation at 11,000g for 10 min. USA), whereas lactobionic acid was purchased from Sigma–Aldrich
These cultures were incubated on an orbital shaker (New Bruns- (Steinheim, Germany). Data acquisition and analysis were per-
wick Sci., NJ, USA) at 250 rpm and 30 °C. formed with ChemStation software (Agilent).
Additionally, Erlenmeyer flasks containing 100 mL of sweet
whey were supplemented with yeast extract or peptone at concen-
2.6. Lactose-oxidizing capacity of P. taetrolens
trations of 0.5, 1, 2, 2.5 and 5 g/L. The flasks were inoculated with
0.14 g/L of wet biomass from NB inoculum cultures and incubated
A loopful of P. taetrolens cells was streaked onto agar plates con-
for 48 h.
taining 10 g/L lactose, 2 g/L peptone, 1 g/L yeast extract, 0.2 g/L
Samples were aseptically withdrawn periodically to determine
K2HPO4, 0.08 g/L bromophenol blue and 15 g/L agar as an screening
bacterial growth and pH. Biomass was removed by centrifugation
medium for oxidation and acid production and incubated at 30 °C
at 16,000g for 5 min, and the cell-free supernatants were stored
for 96 h. The same medium composition, though excluding lactose,
frozen ( 20 °C) until further analysis. Bioconversion experiments
was employed as a negative control of lactose-oxidizing capacity.
were carried out in triplicate as independent trials.
2.4. Batch cultures in a stirred tank bioreactor 3. Results and discussion
Lactobionic production from sweet whey was performed in a 3.1. Preliminary screening of oxidative bioconversion by P. taetrolens
2-L bioreactor (BioFlo 110, New Brunswick Scientific Co., NJ, USA) on agar plates
with a working volume of 1 l. The bioreactor was equipped with
a pH meter (Mettler Toledo, Switzerland) and a polarographic dis- The development of a preliminary plate assay has enabled the
solved oxygen electrode (InPro 6830, Mettler Toledo, Switzerland) screening of the oxidizing capacity of P. taetrolens as the presence
in order to measure and continuously monitor on-line pH and dis- of bromophenol blue allowed the appearance of clear zones sur-
solved oxygen tension (DOT) values, respectively. Batch experi- rounding the lactobionic-producing colonies. Supplementary
ments were conducted under the following conditions: Fig. 1A and C illustrate color changes in the agar plates. As a result,
temperature was maintained at 30 °C, the agitation rate was set an observable yellow coloring was formed due to lactobionic acid
at 350 rpm, employing an aeration rate of 1 v.v.m. through a ring production and subsequent decrease in pH, verifying the capacity
sparger located at the bottom of the bioreactor vessel. Excessive of P. taetrolens cells to oxidize lactose after 48 h incubation com-
foam formation was prevented by automatic addition of diluted pared to control plates without the presence of lactose (Supple-
(1:10) Y-30 emulsion (Sigma–Aldrich, Steinheim, Germany). mentary Fig. 1B). This approach has enabled the distinction
A two-stage pH-shifted bioconversion strategy was adopted in bio- between the cell growth phase during the first 24 h and the subse-
reactor experiments based on preliminary shake-flask results. pH quent lactobionic acid production phase on agar plates (Supple-
control was activated above 6.5 (pH was left uncontrolled above mentary Fig. 1A). These results also give support to the idea that
this value during the growth phase and maintained at 6.5 thereaf- P. taetrolens may oxidize lactose as originally reported for many
ter) by computer-controlled peristaltic pumps via automatic addi- Pseudomonas species by Stodola and Lockwood (1947).
tion of 2 M NaOH. The bioconversion features a single biotransformation to lactob-
The seed culture was obtained by harvesting biomass (after ionic acid catalyzed by a membrane-bound lactose oxidation sys-
centrifugation at 11,000g for 10 min) from a NB shake-flask culture tem that catalyzes lactose oxidation to a lactone intermediate
and re-suspending cells in a 500 mL Erlenmeyer flask containing (lactobiono-d-lactone) whose carbonyl group is subsequently
9732 S. Alonso et al. / Bioresource Technology 102 (2011) 9730–9736
hydrolyzed by a lactonase into lactobionic acid (Fig. 1) (Nishizuka growth phase, as shown in Tables 1 and 2. As can be seen from
and Hayaishi, 1962). Fig. 3B, nitrogen source supplementation with either yeast extract
or peptone increased the specific growth rate and it reached a max-
imum when using 5 g/L of nitrogen supplementation of 0.25 and
3.2. Shake-flask cultivations: effect of inoculum size
0.26 h 1 for yeast extract and peptone supplementation,
respectively.
The influence of inoculum size on lactobionic acid production
In terms of yield (defined as the percentage of lactose converted
from cheese whey is shown in Fig. 2. Lactobionic acid production
into lactobionic acid after 48 h), values varied from 74% to 88% and
followed a biphasic pattern with an initial phase of cell growth
from 72% to 84% for yeast extract and peptone supplementation
and a second production phase characterized by concomitant broth
experiments, respectively, suggesting improved bioconversion per-
acidification. Increased inoculum size triggered lactobionic acid
formance in the presence of yeast extract, although the delay in the
production by achieving higher cell culture densities of 1.18,
production phase was slightly higher than for the peptone experi-
1.24, 1.36 and 1.86 g/L for 0.035, 0.07, 0.14 and 0.64 g/L of initial
ments (Tables 1 and 2). However, these high cell density conditions
biomass, respectively (Fig. 2A). The residual lactose contents also
after 48 h did not lead to higher yields in comparison to those ob-
showed a clear influence of inoculum size on lactobionic acid pro-
tained without extra nitrogen sources (87%), indicating that P. tae-
duction from whey (Fig. 2C). Employing an initial inoculum of
trolens cells were induced towards greater cell proliferation rather
0.64 g/L of biomass, complete lactose bioconversion was achieved
than product formation in the presence of extra nutrients. Lactob-
after 60 h versus a yield (defined as the percentage of lactose con-
ionic acid production might be affected by the oxygen-limited cul-
verted into lactobionic acid after 80 h) of 81% obtained with
ture conditions encountered by P. taetrolens cells during a
0.035 g/L after 80 h (Fig. 2D).
prolonged cell proliferation phase.
The pH of the cultures remained within a pH range of 6.5–8 un-
til the onset of the stationary growth phase, sharply declining to
acidic pH values thereafter as a consequence of lactobionic acid 3.4. Effect of seed culture age on lactobionic acid production
production (Fig. 2B). The pH became slightly basic during the
growth phase owing to the appearance of nitrogen by-products As Fig. 4 shows, the lag phases increased significantly with the
from metabolized proteins which raised the broth pH from 6.5 to use of cells from inocula in the stationary growth phase. A lag
7.36 employing 0.64 g/L of biomass as inoculum. Thus, for the cul- phase of 4 and 8 h was respectively detected in bioreactor cultiva-
tures with a lower initial biomass, the pH-peaked values during the tions at 24- and 36-h, whereas an absence of the lag phase was ob-
growth phase were always higher than for those obtained with served when employing a 12-h seed culture (Fig. 4). The maximum
high initial biomass sizes. Clearly, the influence of initial higher cell dry cell weight varied as a function of inoculum age and was higher
densities may have contributed to the improved yield performance for the youngest inoculum (1.25 versus 1.01, 0.98 g/L of biomass
in shake-flasks (Fig. 2C and D). The use of greater inoculum sizes being obtained for cultivations seeded with 24 and 36-h cultures,
progressively shortened the onset of the production phase coincid- respectively). Due to the lower cell densities obtained in the 24
ing with pH-shift. Thus, with an inoculum of 0.64 g/L, the time be- and 36-h cultures, yields of 91% and 66%, respectively were ob-
fore the onset of the production phase was 6 h, compared to 16 h tained and the residual lactose concentrations were 3.9 and
for a biomass size of 0.035 g/L (Fig. 2D). Moreover, volumetric pro- 14.1 g/L after 82 h for cultivations seeded with 24- and 36-h cul-
ductivity was also clearly affected by inoculum size, achieving val- tures, respectively. Dissolved oxygen tension (DOT) and pH values
ues of 0.43, 0.45, 0.48 and 0.6 g/L h for 0.035, 0.07, 0.14 and 0.64 g/ remained unchanged during the lag phase, as Fig. 4B and C show.
L of biomass, respectively. Physiological ages at 24 and 36 h exerted major effects on lactob-
In contrast, the specific growth rate decreased with increasing ionic acid production from whey seeing as the length of the lag
inoculum sizes as a consequence of limiting substrate concentra- phase preceding growth increased in comparison to batch cultiva-
tions (Fig. 3A). In terms of bioprocess yield, the use of a higher ini- tion with a 12-h seed culture.
tial biomass concentration had a particularly positive effect on Overall, it may be concluded from these results that the use of
bioconversion efficiency, reaching a maximum above 0.37 g/L of time-prolonged seed cultures of P. taetrolens led to an increase in
initial biomass (Fig. 3A). the lag phase, as well as a reduction in lactobionic acid yield in cul-
Lactobionic acid may thus be described as a non-growth associ- tivations compared to late exponential growth phase inocula.
ated product whose whole-cell oxidative bioconversion from the Therefore, a 12-h seed culture in whey was selected as the optimal
lactose present in cheese whey was closely related to the cell den- inoculum for the experiments investigating inoculation level.
sity achieved after the growth phase. Therefore, shorter growth
phases, a higher yield and a significant reduction in bioprocess
3.5. Effect of inoculation level on lactobionic acid production at
duration were obtained under high cell-density conditions.
bioreactor scale
3.3. Effects of nitrogen source supplementation on cell growth and As can be seen in Fig. 5, high inoculum density favored the rapid
lactobionic acid production at shake-flask scale activation of P. taetrolens cells, affecting the onset of the production
phase. Thus, with an inoculation level of 30% (v/v), the culture
The presence of extra nitrogen sources prolonged the growth reached the maximum cell density of 1.35 g/L in 4 h with a specific
phase and higher cell densities were obtained in the stationary growth rate of 0.15 h 1. Figs. 4A and 5 show the time plots of bio-
CH2OH CH2OH CH2OH CH2OH CH2OH CH2OH

Lactose
HO O O oxidase HO O O Lactonase HO O OH O
O O O O C
OH OH OH OH OH OH OH
OH O2 H2O
OH OH OH OH OH OH
Lactose Lactobiono-δ-lactone Lactobionic acid
Fig. 1. Bioconversion of lactose to lactobionic acid by P. taetrolens is carried out through a membrane-located dehydrogenase system which mediates lactose oxidation and
subsequent hydrolysis by lactonase to yield lactobionic acid.
A 2 B 8
Dry cell weight (g/L)

1.5
0.035 g/L
6 0.07 g/L
pH
1 0.14 g/L
0.035 g/L 5 0.64 g/L
0.07 g/L
0.5
0.14 g/L 4
0.64 g/L
0 3
0 20 40 60 80 0 20 40 60 80
Time (h) Time (h)
C D
40 40
0.035 g/L
0.07 g/L
Lactobionic acid (g/L)

30 0.14 g/L 30
Lactose (g/L)
0.64 g/L
20 20 0.035 g/L
0.07 g/L
10 10 0.14 g/L
0.64 g/L
0 0
0 20 40 60 80 0 20 40 60 80
Time (h) Time (h)
Fig. 2. Effect of inoculum size on biomass (A), pH (B), lactose (C) and lactobionic acid production (D) profiles from whey by P. taetrolens in shake-flask cultures. Four initial
biomass concentrations (0.035, 0.07, 0.14 and 0.64 g/L) were employed as inoculum in the experiments.
A 0.25 100 B 0.3

Specific growth rate (h-1 )
Specific growth rate (h-1 )
0.2 80 0.25
0.2
Yield (%)
0.15 60
0.15
0.1 40
0.1
0.05 20
0.05
0 0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 1 2 3 4 5
Initial biomass (g/L) Concentration (g/L)
Fig. 3. Influence of initial biomass concentration on the specific growth rate (h) and P. taetrolens bioconversion efficiency (d), expressed as percentage of lactose converted
into lactobionic acid after 80 h, in sweet whey at shake-flask scale (A). Relationship between specific growth rate and yeast extract () or peptone (M) supplementation on
whey for lactobionic acid production (B).
process parameters for a 5%, 10%, 20% and 30% (v/v) inoculation le- of 1.12 g/L h was thus achieved with the highest inoculation level
vel, corresponding to an initial biomass of 0.09, 0.19, 0.36 and compared to 0.59 g/L h using the lowest inoculation level (Table 3).
0.56 g/L, respectively. Table 3 shows the overall bioconversion This maximum volumetric productivity obtained was thus 2.4-fold
parameters achieved with the four different inoculation levels higher than that attained by Miyamoto et al. (2000) who reported
employed. around 0.46 g/L h in batch mode after 96 h. Undoubtedly, whole-
The use of higher inoculation levels also shortened the duration cell bioconversions often proceed with lower efficiencies and long-
of the growth phase from 10 to 4 h for a 5% and 30% inoculation le- er operation times compared to those achieved with biocatalytic/
vel, respectively. Moreover, a similar trend to the specific growth enzymatic approaches like the one employed by Pedruzzi et al.
rates in shake-flask cultivations was found and the highest specific (2011), which can yield lactobionic acid titers of around 125.4 g/L
growth rate was observed for the lowest inoculation level, 5% from 252 g/L of chemically pure lactose solution.
(0.28 h 1). A significant reduction in bioconversion duration was The pH-shift values correlated with inoculation levels (Table 3)
obtained from 72 to 32 h with a 5% and 30% inoculation level, and increased levels led to a decrease in pH-shift values. The pH in-
respectively. Maximum volumetric lactobionic acid productivity creased to about 8.13 (at 11 h), 8.00 (at 9 h), 7.60 (at 6 h) and 7.30
Fig. 4. Effect of seed age on lactobionic acid production from whey by P. taetrolens. Time-course profiles of bioprocess parameters during batch bioconversion in a bioreactor
employing 12-h (A), 24-h (B) and 36-h (C) seed cultures.
Table 1 Table 2
Effect of yeast extract supplementation on growth and lactobionic acid production Effect of different concentrations of peptone on growth and lactobionic acid
from whey by P. taetrolens after 48 h. production from whey by P. taetrolens after 48 h.
YE Maximum DCW l Onset of production Yield Peptone Maximum DCW l Onset of Yield
1 1
concentration obtained (g/L) (h ) phase (h) (%) concentration (g/ obtained (g/L) (h ) production phase (%)
(g/L) L) (h)
0.5 1.36 ± 0.08 0.16 12 74 0.5 1.35 ± 0.07 0.18 12 72
1 1.38 ± 0.06 0.19 12 76 1 1.36 ± 0.05 0.20 12 74
2 1.54 ± 0.06 0.22 14 79 2 1.48 ± 0.03 0.24 14 77
2.5 1.78 ± 0.12 0.23 15 81 2.5 1.75 ± 0.1 0.25 14 78
5 2.24 ± 0.03 0.25 16 88 5 1.97 ± 0.02 0.26 15 84
YE = yeast extract; DCW = dry cell weight; l = specific growth rate. DCW = dry cell weight; l = specific growth rate.
(at 5 h) for a 5%, 10%, 20%, 30% inoculation level, respectively, sub- and 5). The batch bioconversion seeded with a 5% (v/v) of inoculum
sequently decreasing to 6.5 during the production phase (Figs. 4A volume showed the highest culture broth alkalinity (pH of 8.13)
Fig. 5. Time profiles of biomass, pH, DOT, lactose and lactobionic acid during pH-shift bioreactor cultivations with a 5 (A), 20 (B) and 30% (v/v) (C) inoculation level.
Table 3
oxygen concentration occurred after nutrient depletion, signaling
Summary of values obtained under the influence of inoculation level on batch the switch from growth to product formation (Fig. 5). High micro-
bioconversion at bioreactor scale. bial cell densities impacted on oxygen availability during the
Inoculation Maximum Duration pH- l Volumetric Time Yield
growth phase period (Fig. 5). Therefore, in order to avoid oxygen-
level% (v/v) dry cell of growth shift (h 1
) productivity (h) (%) limited cultivation conditions with an inoculation level of 30% (v/
weight (g/ phase (h) value (g/L h) v), the DOT was maintained above 20% (Fig. 5C) by automatic con-
L) trol of the agitation speed (up to 500 rpm).
5 1.08 10 8.13 0.28 0.59 72 100 In the bioreactor experiments, lactobionic acid was a non-
10 1.25 8 8.00 0.19 0.69 60 100 growth associated product within a bioprocess which proceeded
20 1.27 6 7.60 0.17 0.80 50 100
with a concomitant drop in dissolved oxygen concentration as well
30 1.35 4 7.30 0.15 1.12 32 100
as an increase in pH values during growth phase.
The elucidation of a suitable and feasible downstream process-
ing methodology after the bioconversion process could guarantee
(Fig. 5A), whereas the maximum pH value obtained in the the successful implementation of this system. Considering that
bioreactor when employing a 30% inoculation level was 7.30 the recovery process depends primarily on the nature of the matrix
(Fig. 5C). employed for lactobionic acid production, media components and
During the growth phase, a sharp drop in dissolved oxygen ten- metabolites from bioconversion broth could become a drawback
sion values was found, but a sharp rise in the saturated dissolved to purify the desired target compared to the lesser effort required
for downstream processing of lactobionic acid produced after cat- Koller, M., Bona, R., Chiellini, E., Grillo Fernandes, E., Horvat, P., Kutschera, C., Hesse,
P., Braunegg, G., 2008. Polyhydroxyalkanoate from whey by Pseudomonas
alytic approaches based on refined lactose. After cell harvesting,
hydrogenovora. Bioresour. Technol. 99, 4854–4863.
recovery of lactobionic acid from bioconversion broth could be Kuusisto, J., Tokarev, A.V., Murzina, E.V., Roslund, M.U., Mikkola, J.-P., Murzin, D.Y.,
accomplished by conventional methods involving precipitation, Salmi, T., 2007. From renewable raw materials to high value-added fine
extraction and adsorption using ion-exchange resins. Undoubtedly, chemicals-catalytic hydrogenation and oxidation of d-lactose. Catal. Today 121,
92–99.
lactobionic acid recovery after production processes will be subject Lin, W.L., Chen, T.D., Liu, C.-W., 2009. Synthesis and characterization of lactobionic
of intensive future research. acid grafted pegylated chitosan and nanoparticle complex application. Polymer
50, 4166–4174.
Ludwig, R., Ozga, M., Zámocky, M., Peterbauer, C., Kulbe, K., Haltrich, D., 2004.
4. Conclusions Continuous enzymatic regeneration of electron acceptors used by
flavoenzymes: cellobiose dehydrogenase-catalyzed production of lactobionic
acid as an example. Biocatal. Biotransfor. 22, 97–104.
Cheese whey was employed as a raw material for production of
Maischberger, T., Nguyen, T.-H., Sukyai, P., Kittl, R., Riva, S., Ludwig, R., Haltrich, D.,
lactobionic acid by P. taetrolens using a two-stage pH-shifted bio- 2008. Production of lactose-free galacto-oligosaccharide mixtures: comparison
conversion strategy, and lactobionic acid production of 42.4 g/L of two cellobiose dehydrogenases for the selective oxidation of lactose to
lactobionic acid. Carbohydr. Res. 343, 2140–2147.
with a 30% volume seed culture inoculum was obtained after
Miyamoto, Y., Ooi, T., Kinoshita, S., 2000. Production of lactobionic acid from whey
32 h. Whey offers an alternative means to costly synthetic media by Pseudomonas sp. LS13–1. Biotechnol. Lett. 22, 427–430.
with nutrient-mineral supplementation for lactobionic acid pro- Murzina, E.V., Tokarev, A.V., Kordás, K., Karhu, H., Mikkola, J.-P., Murzin, D.Y., 2008.
duction. Further improvements concerning the influence of physi- D-lactose oxidation over gold catalysts. Catal. Today 131, 385–392.
Nishizuka, Y., Hayaishi, O., 1962. Enzymic formation of lactobionic acid from
cal culture parameters on microbial behavior as well as lactose. J. Biol. Chem. 237, 2721–2728.
downstream processing are required for implementation of the Nordkvist, M., Nielsen, P.M., Villadsen, J., 2007. Oxidation of lactose to lactobionic
process on an industrial level. acid by a Microdochium nivale carbohydrate oxidase: kinetics and operational
stability. Biotechnol. Bioeng. 97, 694–707.
Oskarsson, H., Frankenberg, M., Annerling, A., Holmberg, K., 2007. Adsorption of
Acknowledgements novel alkylaminoamide sugar surfactants at tailor-made surfaces. J. Surfact.
Deterg. 10, 41–52.
Pedruzzi, I., da Silva, E.A.B., Rodrigues, A.E., 2011. Production of lactobionic acid
The authors are grateful for financial support from the Spanish from lactose/fructose substrate using GFOR/GL enzymes from Zymomonas
Ministry of Science and Innovation through project MEC-CTQ2010- mobilis cells: a kinetic study. Enzyme Microb. Tech. 49, 183–191.
14918. Pedruzzi, I., Malvessi, E., Mata, V.G., Silva, E.A.B., Silveira, M.M., Rodrigues, A.E.,
2007. Quantification of lactobionic acid and sorbitol from enzymatic reaction of
fructose and lactose by high-performance liquid chromatography. J.
Appendix A. Supplementary data Chromatogr. A 1145, 128–132.
Povolo, S., Toffano, P., Basaglia, M., Casella, S., 2010. Polyhydroxyalkanoates
production by engineered Cupriavidus necator from waste material containing
Supplementary data associated with this article can be found, in lactose. Bioresour. Technol. 101, 7902–7907.
the online version, at doi:10.1016/j.biortech.2011.07.089. Saarela, M., Hallamaa, K., Mattila-Sandholm, T., Mättö, J., 2003. The effect of lactose
derivatives lactulose, lactitol and lactobionic acid on the functional and
technological properties of potentially probiotic Lactobacillus strains. Int.
References Dairy J. 13, 291–302.
Satory, M., Fürlinger, M., Haltrich, D., Kulbe, K.D., Pittner, F., Nidetzky, B., 1997.
Affertsholt, T. 2007. Market developments and industry challenges for lactose and Continuous enzymatic production of lactobionic acid using glucose-fructose
lactose derivatives. Presentation at IDF Symposium Lactose and Its Derivatives, oxidoreductase in an ultrafiltration membrane reactor. Biotechnol. Lett. 19,
14–16 May 2007. 1205–1208.
Chia, Y.N., Latusek, M.P., Holles, J.H., 2008. Catalytic wet oxidation of lactose. Ind. Schaafsma, G., 2008. Lactose and lactose derivatives as bioactive ingredients in
Eng. Chem. Res. 47, 4049–4055. human nutrition. Int. Dairy J. 18, 458–465.
Chung, T.W., Yang, J., Akaike, T., Cho, K.Y., Nah, J.W., Kim, S.I., Cho, C.S., 2002. Selim, S.M.K., Ha, Y.-S., Kim, S.-J., Chang, Y., Kim, T.-J., Lee, G.H., Kang, I.-K., 2007.
Preparation of alginate/galactosylated chitosan scaffold for hepatocyte Surface modification of magnetite nanoparticles using lactobionic acid and their
attachment. Biomaterials 23, 2827–2834. interaction with hepatocytes. Biomaterials 28, 710–716.
D’Alessandro, A.M., Hoffman, R.M., Southard, J.H., 1999. Solution development in Splechtna, B., Petzelbauer, I., Baminger, U., Haltrich, D., Kulbe, K.D., Nidetzky, B.,
organ preservation: the University of Wisconsin perspective. Transplant. Rev. 2001. Production of a lactose-free galacto-oligosaccharide mixture by using
13, 67–77. selective enzymatic oxidation of lactose into lactobionic acid. Enzyme Microb.
Gänzle, M.G., Haase, G., Jelen, P., 2008. Lactose: crystallization, hydrolysis and Tech. 29, 434–440.
value-added derivatives. Int. Dairy J. 18, 685–694. Stodola, F.H., Lockwood, B.L., 1947. The oxidation of lactose and maltose to bionic
Gerling, K.G., 1998. Large scale production of lactobionic acid-use and new acids by Pseudomonas. J. Biol. Chem., 213–221.
applications. Int. Dairy Fed. 9804, 251–256. Van Hecke, W., Bhagwat, A., Ludwig, R., Dewulf, J., Haltrich, D., Van Langenhove, H.,
Green, B.A., Yu, R.J., Van Scott, E.J., 2009. Clinical and cosmeceutical uses of 2009. Kinetic modeling of a bi-enzymatic system for efficient conversion of
hydroxyacids. Clin. Dermatol. 27, 49–501. lactose to lactobionic acid. Biotechnol. Bioeng. 102, 1475–1482.
Guimarães, P.M.R., Teixeira, J.A., Domingues, L., 2010. Fermentation of lactose to Venetsaneas, N., Antonopoulou, G., Stamatelatou, K., Kornaros, M., Lyberatos, G.,
bioethanol by yeasts as part of integrated solutions for the valorisation of 2009. Using cheese whey for hydrogen and methane generation in a two-stage
cheese whey. Biotechnol. Adv. 28, 375–384. continuous process with alternative pH controlling approaches. Bioresour.
Hart, N.A., Leuvenink, H.G.D., Ploeg, R.J., 2002. New solutions in organ preservation. Technol. 100, 3713–3717.
Transplant. Rev. 16, 131–141. Wu, D.-Q., Lu, B., Chang, C., Chen, C.-S., Wang, T., Zhang, Y.-Y., Cheng, S.-X., Jiang, X.-
Kekkonen, V., Lafreniere, N., Ebara, M., Saito, A., Sawa, Y., Narain, R., 2009. Synthesis J., Zhang, X.-Z., Zhuo, R.-X., 2009. Galactosylated fluorescent labeled micelles as
and characterization of biocompatible magnetic glyconanoparticles. J. Magn. a liver targeting drug carrier. Biomaterials 30, 1363–1371.
Magn. Mater. 321, 1393–1396.
Resultados 129
Appendix A. Supplementary data
Figure 1 Supplementary data. Details of oxidative bioconversion carried out by Pseudomonas taetrolens on
agar plates containing lactose. Progressive temporal color changes in an agar plate incubated for 96 h (A).
Negative control plates without lactose after 24 and 96 h (B). Detail of yellow clearance zone formed around P.
taetrolens colonies (C).
130 Resultados
Resultados 131
4.4. Influencia de la disponibilidad de oxígeno disuelto sobre la

producción de ácido lactobiónico
La disponibilidad de oxígeno disuelto en el sistema fermentativo anteriormente

desarrollado debería en principio jugar un papel relevante al tratarse de un proceso
aeróbico. Por lo tanto, en este subapartado se ha evaluado la influencia de este parámetro
sobre el proceso de producción biotecnológica de ácido lactobiónico por P. taetrolens a
partir de suero lácteo. Es evidente que la producción de este ácido podría verse afectada en
el proceso de aireación y/o mezcla como consecuencia de la limitación en el aporte de
oxígeno, parámetro que ha de ser tenido en cuenta en todo sistema aeróbico. No obstante, el
presente sistema no sólo se caracteriza por la dependencia del oxígeno durante la
proliferación y mantenimiento celular (siendo utilizado el O2 como aceptor final de
electrones en la cadena respiratoria) sino que el O2 es un elemento clave en la oxidación de
la lactosa por parte de P. taetrolens para dar lugar al ácido lactobiónico ya que re-oxida al
cofactor rédox (dinucleótido de flavina-adenina, FAD) involucrado en la formación de dicho
ácido (véase el subapartado 2.3 para mayores detalles sobre la reacción). Se han evaluado
igualmente la transferencia de oxígeno y el consumo por parte de P. taetrolens. De esta
forma, se ha comprobado cómo una supresión del aporte de oxígeno puede conllevar un
irremediable efecto negativo sobre la producción de ácido lactobiónico. Asimismo, se ha
evaluado tanto el impacto de diferentes niveles de aireación (0-2 Lpm) como de agitación
(150-1000 rpm) sobre los diferentes valores asociados al sistema (proliferación celular, tasa
específica de crecimiento, valor de pH, etc…).
Se ha evaluado también el comportamiento del sistema ante diferentes niveles de
oxígeno disuelto en la fase de proliferación celular en condiciones de alta densidad celular
inicial (tamaño de inóculo del 30%, v/v). Para ello, se han variado los niveles de oxígeno
disuelto en el medio a través de una cascada de agitación, evaluando por lo tanto el
comportamiento del sistema bien sea bajo condiciones en exceso o limitadas de aporte de
oxígeno.
Artículo: Role of dissolved oxygen availability on lactobionic acid production from whey.
Situación: Artículo publicado en Bioresource Technology (Elsevier).
Bioresource Technology 109 (2012) 140–147
Contents lists available at SciVerse ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
Role of dissolved oxygen availability on lactobionic acid production from whey

by Pseudomonas taetrolens
Saúl Alonso, Manuel Rendueles, Mario Díaz ⇑
Department of Chemical Engineering and Environmental Technology, University of Oviedo, Faculty of Chemistry, C/ Julián Clavería s/n, 33071 Oviedo, Spain
a r t i c l e i n f o a b s t r a c t
Article history: The influence of dissolved oxygen availability on cell growth and lactobionic acid production from whey
Received 18 November 2011 by Pseudomonas taetrolens has been investigated for the first time. Results from pH-shift bioreactor cul-
Received in revised form 9 January 2012 tivations have shown that high agitation rate schemes stimulated cell growth, increased pH-shift values
Accepted 10 January 2012
and the oxygen uptake rate by cells, whereas lactobionic acid production was negatively affected. Con-
Available online 20 January 2012
versely, higher aeration rates than 1.5 Lpm neither stimulated cell growth nor lactobionic acid production
(22% lower for an aeration rate of 2 Lpm). Overall insights into bioprocess performance enabled the
Keywords:
implementation of 350 rpm as the optimal agitation strategy during cultivation, which increased lactob-
Lactobionic acid
Whey
ionic productivity 1.2-fold (0.58–0.7 g/L h) compared to that achieved at 1000 rpm. Oxygen supply has
Pseudomonas taetrolens been shown to be a key bioprocess parameter for enhanced overall efficiency of the system, representing
Dissolved oxygen essential information for the implementation of lactobionic acid production at a large scale.
Oxygen uptake rate Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction solved oxygen to microbial cells must accordingly be considered

(García-Ochoa and Gómez, 2009). Both monitoring and proper con-
Lactobionic acid has recently burst onto the cosmetic and phar- trol of dissolved oxygen levels represent decisive factors in aerobic
maceutical scene as a relevant polyhydroxy acid covering a broad systems seeing as insufficient oxygen supply may inhibit metabo-
spectrum of applications. In addition to its cosmeceutical use lism, cell growth and microbial biosynthesis. Owing to their intrin-
(Green et al., 2009), lactobionic acid is being increasingly employed sic nature, oxygen as well as agitation regimes have been shown to
as a potential drug carrier (Wu et al., 2009; Zhang et al., 2011), as play a major role in numerous submerged liquid systems involved
well as in the development of carbohydrate-functionalized nano- in gellam (Giavasis et al., 2006), hyaluronic acid (Huang et al.,
particles used as therapeutic agents for biomedical purposes (Lin 2006), ganoderic acid (Tang and Zhong, 2003), alginate (Peña
et al., 2009). Considering that its commercial relevance is undergo- et al., 2000), enzyme (Rahulan et al., 2011) or xantham gum syn-
ing stepwise growth (Affertsholt, 2007), both the development and thesis (Amanullah et al., 1998). Consequently, the elucidation of
implementation of feasible lactobionic production systems appear suitable aeration supply schemes enables the generation of a suit-
as crucial key challenges to meet market demands. able microenvironment for improved cellular proliferation pat-
Presently, emerging considerations regarding environmental terns, as well as higher metabolite production yields.
regulatory frameworks are moving the dairy industry towards In industrial practice, the mass transfer coefficient (KLa) as well
more sustainable practices. As a result, cheese whey has been the as oxygen requirements by microbial cells (measured as the
subject of valorization either to obtain different value-added bio- oxygen uptake rate, OUR) are used as scale-up criteria (Zou et al.,
products (Koutinas et al., 2009; Guimarães et al., 2010) or to reduce 2008; García-Ochoa and Gómez, 2009) given that KLa is essentially
its pollution potential (Ghaly and Kamal, 2004). a measure of how much oxygen can be supplied to microorganisms
Within this context, cheese whey has been proved to be a suit- growing in a bioreactor (Bandyopadhyay et al., 2009). Specifically,
able, inexpensive and attractive source for obtaining lactobionic oxygen mass transfer has been chosen as the controlling step rate
acid through a biotechnological process carried out by Pseudomo- for the overall behavior in oxygen-limited fermentation processes
nas taetrolens (Alonso et al., 2011). However, as submerged micro- (García-Ochoa et al., 2000; Lozano et al., 2011). In fact, a concise
bial bioprocesses involving aerobic platforms are often strongly understanding of the key oxygen parameters associated with any
influenced by mass transfer limitations, the availability of dis- submerged liquid culture is crucial to enhance microbial perfor-
mance (Wang et al., 2010; Xu and Zhong, 2011).
The aim of the present study was thus to investigate the rela-
⇑ Corresponding author. Tel.: +34 98 5103439; fax: +34 98 5103434. tionship between dissolved oxygen and biotechnological produc-
E-mail address: [email protected] (M. Díaz). tion of lactobionic acid from whey by P. taetrolens, exploring for
0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2012.01.045
the first time bioconversion performance under oxygen-limited 2.5. Influence of oxygen deprivation after the growth phase
and non-oxygen-limited conditions. To this end, bioreactor cultiva-
tions were performed under different aeration and agitation rates Process performance was analyzed by combining a two-stage
in order to examine whether such effects would have an impact batch cultivation carried out at a medium agitation rate
on lactobionic acid synthesis. Two bioprocess parameters were (350 rpm): an initial aerobic growth phase (1 Lpm during 0–10 h)
also evaluated, namely the oxygen uptake rate and the volumetric followed by an oxygen supply deprivation stage after the exponen-
oxygen transfer coefficient, thereby enabling improved knowledge tial growth phase (10–60 h).
of the role of oxygen in culture performance and providing insights
into the most suitable oxygen supply scheme for optimal lactob- 2.6. Effect of agitation rate on biomass and lactobionic acid production
ionic acid production from whey. from whey
In order to examine the influence of the agitation rate on bio-

2. Methods process parameters, bioreactor cultivations were carried out at
150, 350, 500, 700 and 1000 rpm with the aeration rate kept at
2.1. Microorganism 1 Lpm.
Pseudomonas taetrolens LMG 2336, obtained from the Belgian

2.7. Effect of aeration rate on bioprocess parameters
Coordinated Collection of Microorganisms (Gent, Belgium), was
maintained frozen (in 40% [v/v] glycerol at 20 °C). This strain
The effect of different aeration rates (0, 0.5, 1, 1.5 and 2 Lpm) on
was subsequently subcultured on Nutrient Broth (NB, containing
bioprocess parameters was studied in batch cultivations carried
1 g/L meat extract, 2 g/L yeast extract, 5 g/L peptone and 5 g/L
out at 350 rpm.
NaCl) agar plates, incubated for 48 h at 30 °C and then preserved
at 4 °C.
2.8. Effect of dissolved oxygen concentration under high-cell density
conditions
2.2. Inoculum and seed culture preparation
In order to gain insight into how the system and bioprocess
A loopful of P. taetrolens from a fresh Nutrient Broth agar plate parameters are affected by the availability of different dissolved
was used to inoculate a 500 mL Erlenmeyer flask containing oxygen concentration, high-cell density cultivations were per-
100 mL of NB medium. This flask was incubated on an orbital shaker formed under several DOT control levels during the exponential
(Infors HT, model Flyer Aerotron, Switzerland) at 250 rpm and 30 °C growth phase. The DOT level was left free or controlled at 20%,
for 10 h. Actively growing cells from this culture were then har- 40% and 60% (percentage of air saturation) by cascading the agita-
vested by centrifugation at 11,000g for 10 min and re-suspended tion speed (from 350 up to 500 rpm). For the cell density condi-
in a 500 mL Erlenmeyer flask containing 100 mL of whey. This flask tions, cultivations were conducted with an inoculation level of
was incubated at 250 rpm in an orbital shaker at 30 °C for 12 h and 30% (v/v) at 1 Lpm as aeration rate.
subsequently employed as seed culture in the bioreactor experi-
ments containing sweet whey. 2.9. Determination of the oxygen uptake rate (OUR) and volumetric
oxygen transfer coefficient (KLa)
2.3. Sweet whey preparation
The oxygen uptake rate (OUR) and volumetric oxygen transfer
coefficient (KLa) were determined via the dynamic method
Cheese sweet whey (provided by ILAS, Asturias, Spain) was 1-
(Bandyopadhyay et al., 2009; García-Ochoa and Gómez, 2009).
fold diluted with distilled water (1:1) and adjusted to pH 6.5 (by
The concentration of dissolved oxygen under steady state condi-
adding NaOH 6 N) prior to sterilization using a tangential microfil-
tions can thus be formulated as Eq. (1):
tration device equipped with a PVDF membrane-cassette of
0.22 lm pore size (Millipore, Massachusetts, USA). dC
¼ K L aðC C L Þ qO2 X ¼ 0 ð1Þ
dt
2.4. Batch cultivations in stirred tank bioreactor where qO2 is the specific oxygen consumption rate, X is the concen-
tration of biomass, while the first term corresponds to the mass
Batch cultivations were performed in a 2-L bioreactor (BioFlo transfer rate. Accordingly, the OUR was obtained from the slope
110, New Brunswick Scientific Co., New Jersey, USA) with 1 L of of the plot of dissolved oxygen concentration (CL) versus time fol-
whey as working volume. Bioreactor experiments with an inocula- lowing a momentary interruption of air supply to the bioreactor.
tion level of 10% (v/v) were conducted at 30 °C. The bioreactor was After aeration was re-established, KLa was calculated from the slope
equipped with a pH meter (Mettler Toledo, Switzerland) and a of Eq. (2) by plotting CL versus (qO2 X þ dC L =dt):
polarographic dissolved oxygen electrode (InPro 6830, Mettler To-
1 dC L
ledo, Switzerland) in order to measure pH values on line and con- CL ¼ C qO2 X þ ð2Þ
KLa dt
tinuously monitor dissolved oxygen tension (DOT), respectively.
Excessive foam formation was prevented by automatic addition
of diluted (1:10) Y-30 emulsion (Sigma–Aldrich, Steinheim, 2.10. Analytical methods
Germany). An efficient two-stage pH-shifted bioconversion strat-
egy was adopted as previously described (Alonso et al., 2011): Bacterial growth was measured spectrophotometrically as opti-
pH was controlled above 6.5 (pH was left uncontrolled above this cal density at 600 nm (Shimazdu, UV 1203 model) after centrifuga-
value during the growth phase and subsequently maintained at tion of culture samples at 16,000g for 5 min. Optical density data
6.5) by means of computer-controlled peristaltic pumps via auto- were converted to cell dry weight (expressed in grams per liter)
matic addition of 2 M NaOH. These prior conditions were applied using the corresponding calibration curve obtained previously.
to all cultivations unless otherwise specified. Cultivations were The lactobionic acid and lactose content of cell-free culture sam-
carried out in duplicate as independent experiments. ples were measured by high performance liquid chromatography
following a modified version of a previously published method Conversely, the use of the highest oxygen supply scheme (2 Lpm)
(Pedruzzi et al., 2007). The liquid chromatography system (Agilent only supported a cell density of 1.16 g/L after the growth phase.
1200, Agilent Technologies Inc., California, USA) used for analysis The oxygen-limited conditions encountered by cells under aeration
was equipped with an ICSep ICE-ION-300 column (Transgenomic supply deprivation (0 Lpm) clearly constrained cell growth, yield-
Inc., California, USA) coupled to a refractive index detector. Sulfuric ing a maximum biomass of only 0.57 g/L.
acid (0.450 mM, pH 3.1) was employed as the mobile phase at a flow Dissolved oxygen profiles were strongly related to the growth
rate of 0.3 mL/min with the column temperature set at 75 °C. Data phase, a sharp depletion in dissolved oxygen concentration being
acquisition and analysis were performed with ChemStation soft- found at the initial stage of the cell growth phase followed by a sud-
ware (Agilent). den rapid increase to saturated values indicating the onset of the
production phase (Fig. 2B). In fact, rapid exhaustion of the dissolved
oxygen was found under the absence of oxygen supply (0 Lpm),
3. Results and discussion DOT remaining at <1% throughout the batch cultivation process.
Furthermore, there was not a substantial difference in the mini-
3.1. Influence of oxygen deprivation in a two-stage cultivation process mum DOT levels obtained in the fermentation broth although the
dissolved oxygen availability during growth phase was slightly in-
As can be seen in Fig. 1, a lactobionic acid titer of 15.9 g/L with a creased with the aeration rate. Thus, a minimum DOT level of 5%
volumetric productivity of 0.26 g/L h was obtained by applying an was reached during growth phase at an aeration rate of 0.5 Lpm
oxygen deprivation following the growth phase (at 10 h). It can be in comparison to 33% attained at 2 Lpm. In many bioprocesses, oxy-
seen from the lactobionic acid synthesis profile that its titer re- gen supply plays a key role in cellular growth and metabolite pro-
mained practically unchanged from 24 h right up to the end of duction due to the fact that poor volumetric mass transfer (KLa)
the process, coinciding with the total absence of dissolved oxygen becomes the rate-limiting factor as fermentation progresses (Wang
in the fermentation broth. In addition, a poor bioconversion effi- et al., 2010; Lozano et al., 2011). However, as Fig. 2B shows, cultiva-
ciency of 37% was obtained, leaving 26.5 g/L of residual lactose tions with aeration rates higher than 0.5 Lpm were not limited by
after 60 h. The pH profile became un-asymmetric after switching dissolved oxygen availability. In the system under study, the high-
directly from aerobic to an imposed oxygen-depleted production est aeration rate (2 Lpm) provided a KLa (49.98 ± 2.12 h1) 2-fold
phase in which a rapid drop in the concentration of dissolved higher than the lowest aeration rate (0.5 Lpm), whereas medium
oxygen from 80% to 30% was observed within less than 40 min, aeration schemes such as 1 and 1.5 Lpm resulted in KLa values of
remaining at <1% from 20 h onwards. Accordingly, pH peaked at 36.88 ± 1.31 and 44.05 ± 2.37 h1, respectively (Fig. 3). Obviously,
10 h after completing the growth phase and its profile showed an KLa was affected by the different aeration rates employed, indicat-
unusual gradual drop associated with a delayed production rate ing the system capacity of oxygen supply.
of lactobionic acid under oxygen deprivation conditions. These As regards pH profiles, fermentation broths suffered the usual
results show that the oxidative bioconversion of lactose was inhib- sharp increase in pH values during the exponential growth phase
ited by restricting the supply of oxygen, despite having attained a (Alonso et al., 2011), as shown in Fig. 2C. However, as a result of
biomass concentration of 1.24 g/L after the aerobic phase. ineffective growth, cultivation under aeration supply deprivation
(0 Lpm) showed an absence of pH-shift at the same time as re-
3.2. Effect of the aeration rate on cell growth and lactobionic acid duced lactose bioconversion, remaining the pH profile unchanged
production from whey and stable at 6.5 throughout the batch culture process (Fig. 2C).
pH-shift values were correlated with aeration rates (Fig. 3), an in-
Fig. 2 shows the impact of different oxygen supply levels on bio- crease in aeration rates leading to a progressive rise in pH-shift
process parameters and lactobionic acid production from whey by values (7.76, 8.00, 8.20 and 8.32 for 0.5, 1, 1.5 and 2 Lpm, respec-
P. taetrolens. Lower aeration rates contributed to enhanced cell tively). Batch cultivation with an aeration rate of 2 Lpm showed
growth of P. taetrolens, the highest maximum biomass concentra- the highest culture broth basicity, with a substantial increase in
tion (1.36 g/L) being obtained under the influence of 0.5 Lpm its duration (pH of 8.32 at 16.4 h), whereas the maximum pH value
(Fig. 2A). Similar maximum biomass values of 1.25 and 1.22 g/L obtained when employing an aeration rate of 0.5 Lpm was only
were obtained under aeration rates of 1 and 1.5 Lpm, respectively. 7.76 at 9 h (Fig. 2C). Moreover, oxygen demand by P. taetrolens cells
Fig. 1. Time-course profile of bioprocess parameters during pH-shift bioconversion under dissolved oxygen deprivation upon cellular growth phase. The short dashed vertical
line marks the shut down of aeration at 10 h.
A 1.5 B 100

80
DOT %
60
0 Lpm
40
0.5 0.5 Lpm
0 Lpm 0.5 Lpm 1 Lpm
20 1.5 Lpm
1 Lpm 1.5 Lpm
2 Lpm
2 Lpm
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
C 8.5 D 140
120
8 0 Lpm
OUR (mg/L.h)
0.5 Lpm 100
1 Lpm 0.5 Lpm
7.5 80
pH
1.5 Lpm 1 Lpm

2 Lpm 1.5 Lpm
7 60
2 Lpm
40
6.5
20
6 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
E F
40 40 0 Lpm
0.5 Lpm
Lactose (g/L)
1 Lpm
30 30
1.5 Lpm
2 Lpm
20 0 Lpm 20
0.5 Lpm
10 1 Lpm 10
1.5 Lpm
2 Lpm
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
Fig. 2. Effect of aeration supply levels on P. taetrolens batch cultivations. Time-course profiles of cell growth (A), DOT (B), pH (C), OUR (D), lactose (E) and lactobionic acid (F)
are shown.
Fig. 3. Relationship between aeration rates and specific growth rate, maximum biomass, KLa and pH-shift values obtained during batch cultivations.
Table 1
Summary of values obtained in batch cultures under the influence of the aeration rate after cultivation.
Aeration rate Maximum dry cell weight Duration of growth phase pH-shift l Volumetric productivity Culture time Yield
(Lpm) (g/L) (h) value (h1) (g/L h) (h) (%)a
0 0.57 ± 0.06 12 – 0.08 0.04 ± 0.02 60 6 ± 2.5
0.5 1.36 ± 0.05 8 7.76 ± 0.02 0.18 0.80 ± 0.00 52 100 ± 0
1 1.25 ± 0.04 8 8.00 ± 0.05 0.19 0.70 ± 0.00 60 100 ± 0
1.5 1.22 ± 0.04 8 8.20 ± 0.00 0.20 0.70 ± 0.00 60 100 ± 0
2 1.16 ± 0.07 10 8.32 ± 0.03 0.20 0.54 ± 0.03 60 78 ± 3
–: absence of pH-shift value; l = specific growth rate.

a
Yield was defined as the percentage of lactose converted into lactobionic acid after culture time.
coincided with this increase in pH, higher maximum OUR values Results from specific growth rates also revealed that higher agita-
being obtained under increased oxygen supply conditions tion rates stimulated cell growth performance. The specific growth
(Fig. 2D). Cultivation employing the highest aeration rate (2 Lpm) rate thus increased with increasing agitation rate, although a sim-
presented a maximum OUR value of 115 mg/L h at 8 h, compared ilar rate of around 0.20 h1 was found for medium (350, 500 rpm)
to the 59.4 mg/L h obtained with 0.5 Lpm at 6 h. However, it was and high (700, 1000 rpm) agitation regimes (Fig. 5). Regarding the
observed that the OUR decreased with the onset of the production onset of the production phase, the same duration of growth phase
phase due to nutrient limitation. Such diverse oxygen demands (8 h) was found for cultivations with agitation regimes higher than
evidence that these cultivations were markedly influenced by the 350 rpm. The onset of the production phase at the lowest rate
different volumetric oxygen transfer rates provided as a function (150 rpm) was delayed, however, due to the oxygen-limited condi-
of oxygen supply (Fig. 3). Changes in the OUR as well as KLa pro- tions (DOT < 1% for 10 h) encountered by P. taetrolens cells, which
vided improved knowledge of this incipient biotechnological sys- clearly suffered a reduced proliferation (0.12 h1) during the
tem, elucidating the role of dissolved oxygen within oxidative growth phase compared to those obtained at higher rates
bioconversion. (>0.20 h1).
The use of high aeration rates was detrimental to lactobionic Parallel to this decrease in dissolved oxygen, pH profiles
acid production, as can be seen in Fig. 2F. Lactobionic acid produc- showed an increase during the growth phase, as shown in
tion decreased markedly with increasing aeration rates. In fact, Fig. 4C. The diverse dissolved oxygen availability explained the dif-
complete lactose bioconversion was achieved after 52 h at ferent pH-shift values observed during batch cultivations, being
0.5 Lpm compared to a 78% process yield obtained after 60 h under 7.88, 8.00, 8.10, 8.23 and 8.45 for 150, 350, 500, 700 and
an aeration rate of 2 Lpm (Fig. 2F). Accordingly, optimal aeration 1000 rpm, respectively. Temporal distribution of these pH-shifts,
rate for a complete lactose bioconversion ranged from 0.5 to with the exception of that from cultivation at 150 rpm, took place
1.5 Lpm as Fig. 2E shows. Table 1 summarizes the overall cultiva- at 8 h, coinciding with the end of the growth phase (Table 2).
tion parameters obtained with the five different aeration rates em- Microbial oxygen requirements and consumption also increased
ployed. A reduced specific growth rate (0.08 h1) as well as a low with the agitation rate, as Fig. 4D shows. The onset of the produc-
titer of lactobionic acid (2.5 g/L) were obtained under the absence tion phase featured a gradual decrease in the OUR, suggesting that
of aeration (0 Lpm), suggesting that the lack of oxygen supply the maximum metabolic activity of the cells was reached during
inhibited microbial cell proliferation and subsequent lactobionic the growth phase, subsequently remaining constant in the latter
acid synthesis (Fig. 2A and E). In fact, normal growth patterns as stage of fermentation. Moderate agitation regimes thus resulted
well as metabolite synthesis are negatively affected by the unbal- in a lower dissolved oxygen content being available for the micro-
anced redox state of cells generated under oxygen deprivation con- bial metabolism of P. taetrolens cells, leading to either a reduction
ditions in many bioprocesses involving aerobic microorganisms in pH-shift values or the OUR. Furthermore, vigorous agitation re-
(Huang et al., 2006). Nevertheless, P. taetrolens displayed a broad gimes (700 and 1000 rpm) provided batch cultivations with a sig-
tolerance to oxygen, as reflected by the fact that it grew under oxy- nificantly higher KLa, which resulted in 7- and 7.5-fold higher
gen supply deprivation conditions, although only a maximum bio- values compared to the lowest agitation rate (150 rpm, 10.63 ±
mass of 0.57 g/L was obtained after 60 h. Conversely, these 1.12 h1). However, KLa levels were fixed at 36.88 ± 1.31 and
microaerobic conditions barely influenced lactose oxidative bio- 56.80 ± 1.62 h1 for medium agitation rates (350 and 500 rpm),
conversion (Fig. 2E), seeing as the yield was reduced up to 6% as can be seen in Fig. 5. It is clear that increasing agitation facili-
(Table 1). tated cell proliferation by enhancing overall oxygen mass transfer.
Thus, the detrimental effect of oxygen depletion at 150 rpm lay in
3.3. Effect of the agitation rate on bioprocess parameters oxygen mass transfer limitations due to the fact that poor mixing
caused ineffective oxygen transfer and, as a result, the inefficient
The time courses of the bioprocess parameters under the influ- availability of dissolved oxygen for microbial cells.
ence of different agitation rates (at a constant aeration rate of Despite the fully aerobic conditions obtained at agitation rates
1 Lpm) are shown in Fig. 4. As can be seen, higher agitation rates above 500 rpm (Fig. 4B), lactobionic production was not related
stimulated growth of P. taetrolens cells, resulting in improved pro- to oxygen availability, as Fig. 4F shows. In terms of lactobionic acid
liferation patterns, maximum biomass values of 1.06, 1.25, 1.29, volumetric productivity, the best culture performance was ob-
1.40 and 1.48 g/L being progressively obtained for 150, 350, 500, tained at 350 rpm, with a productivity of 0.70 g/L h, compared to
700 and 1000 rpm, respectively (Fig. 4A). As can be seen in the 0.57, 0.56 and 0.58 g/L h obtained at 150, 700 and 1000 rpm,
Fig. 4B, culture broths were supplied with different dissolved respectively (Table 2). Similar process yields (of around 80%) were
oxygen levels by changing the agitation speed in order to meet obtained for these three batch cultivations, although cultivation at
the oxygen demands of microbial cells. Thus, full aeration condi- 150 rpm exhibited retarded cell growth proliferation (Fig. 4A) and
tions were encountered by P. taetrolens cells under agitation rates a lowered OUR (Fig. 4D) due to oxygen-limited conditions which
higher than 700 rpm (Fig. 4B). At 150 rpm, the oxygen supplied clearly inhibited subsequent lactobionic acid production synthesis
was completely consumed by P. taetrolens, thus leading to a micro- during the initial stage of the lactose oxidative process (Fig. 4E).
aerobic environment (DOT < 1%) in the fermentation broth for 10 h. Hence, medium agitation rates (350 and 500 rpm) were found to
A 1.5
B 100
80
DOT %
1 60
150 rpm 150 rpm

40
350 rpm 350 rpm
0.5
500 rpm 500 rpm
700 rpm 20 700 rpm
1000 rpm 1000 rpm
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
C 8.5 D 140
120
8 150 rpm
350 rpm 100
OUR (mg/L.h)
500 rpm 150 rpm
7.5 80 350 rpm
pH
700 rpm
1000 rpm 500 rpm
7 60 700 rpm
1000 rpm
40
6.5
20
6 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
E F
40 150 rpm 40
350 rpm
500 rpm
Lactose (g/L)
30 700 rpm 30
1000 rpm
20 20 150 rpm
350 rpm
500 rpm
10 10
700 rpm
1000 rpm
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
Fig. 4. Time-course profiles of biomass (A), DOT (B), pH (C), OUR (D), lactose (E) and lactobionic acid (F) at different agitation rates in pH-shift bioreactor cultivations with an
aeration rate of 1 Lpm.
be the best oxygen supply conditions for lactobionic acid produc- In response to increased oxygen availability, P. taetrolens cells dis-
tion despite obtaining lower cell densities than those at vigorous played better specific growth rates and cultivations presented
rates (700 and 1000 rpm) after the growth phase. higher pH-shift values. However, as can be seen in Fig. 6, dissolved
oxygen availability was detrimental for lactobionic acid produc-
3.4. Influence of the dissolved oxygen level under high-cell density tion, as the lactobionic acid yield was progressively reduced under
conditions high supply conditions, yielding complete lactose bioconversion
under non-controlled conditions (DOT < 5% during the growth
In order to further clarify the role of oxygen on lactobionic pro- phase) and at a level of 20%. The influence of dissolved oxygen
duction from whey by P. taetrolens cells, cultivations were carried availability shows that lactobionic acid synthesis by P. taetrolens
out under high-cell density conditions. The influence of different cells increased markedly when cultivations were submitted to a
dissolved oxygen levels on the bioprocess parameters is shown short oxygen-limited period (DOT between 0–20%). Fermentation
in Fig. 6 and summarized in Table 3. In terms of cell growth, culti- efficiencies in terms of volumetric productivity under oxygen-
vations carried out under high oxygen supply conditions (40% and limited conditions were also enhanced, being 1.7- and 1.5-fold
60%) showed slightly enhanced maximum cell densities (1.45 ver- (1.27 and 1.12 g/L h for non-controlled and the 20% level, respec-
sus 1.35 g/L). Likewise, a longer cell growth phase (6 h) was ob- tively) higher than under the highest oxygen supply scheme
tained at a level of 60% compared to 4 h obtained at lower levels. (0.75 g/L h for a level of 60%). In fact, oxygen limitation can induce
Fig. 5. Relationship between agitation rates and specific growth rate, maximum biomass (expressed as Dry Cell Weight, DCW), KLa and pH-shift values obtained during batch
cultivations.
Table 2
Summary of values obtained under the influence of the agitation rate on batch production of lactobionic acid from whey after 60 h.
Agitation rate (rpm) Maximum dry cell weight (g/L) Duration of growth phase (h) pH-shift value l (h1) Volumetric productivity (g/L h) Yield (%)a
150 1.06 ± 0.09 10 7.88 ± 0.04 0.12 0.57 ± 0.02 82 ± 3
350 1.25 ± 0.04 8 8.00 ± 0.05 0.19 0.70 ± 0.00 100 ± 0
500 1.29 ± 0.02 8 8.10 ± 0.02 0.20 0.65 ± 0.00 93 ± 0
700 1.40 ± 0.05 8 8.23 ± 0.02 0.21 0.56 ± 0.03 80 ± 4
1000 1.48 ± 0.07 8 8.45 ± 0.00 0.21 0.58 ± 0.03 82 ± 4
l = specific growth rate.

a
Yield was defined as the percentage of lactose converted into lactobionic acid after 60 h.
Fig. 6. Effects of different dissolved oxygen levels on bioprocess parameters under high-cell density conditions.
Table 3
Summary of values obtained in high-cell density batch cultivations in which DOT was controlled at different levels via an agitation cascade (from 350 to 500 rpm).
DOT level (%) Maximum dry cell weight Duration of growth phase pH-shift value l (h1) Volumetric productivity Culture time Yield
(g/L) (h) (g/L h) (h) (%)a
Not controlled 1.35 ± 0.03 4 6.95 ± 0.05 0.13 1.27 ± 0.00 30 100 ± 0
20 1.35 ± 0.06 4 7.30 ± 0.02 0.15 1.12 ± 0.00 32 100 ± 0
40 1.45 ± 0.08 4 7.50 ± 0.06 0.16 1.10 ± 0.02 32 93 ± 2
60 1.45 ± 0.05 6 7.64 ± 0.03 0.17 0.75 ± 0.04 32 58 ± 3
l = specific growth rate.

a
Yield was defined as the percentage of lactose converted into lactobionic acid after culture time.
secondary metabolite production (Clark et al., 1995), although, as Interestingly, lactobionic acid titers obtained were 2.9-fold low-
previously shown, prolonged oxygen depletion in this fermenta- er than those attained by Pedruzzi et al. (2011) who reported
tion system can constitute a limiting factor for cell growth and sub- around 125.4 g/L of lactobionic acid from 252 g/L of chemically
sequent lactobionic acid production from whey by P. taetrolens due pure lactose solution employing permeabilized resting cells of
to the delayed onset of the production phase. Zymomonas mobilis. Conversely, this later biotransformation sys-
tem was featured by a specific productivity of around 0.8 g/ Ghaly, A.E., Kamal, M.A., 2004. Submerged yeast fermentation of acid cheese whey
for protein production and pollution potential reduction. Water Res. 38, 631–
g DCW h which was substantially lower than that obtained in the
644.
present work under high-cell density conditions (0.94 g/g DCW h), Giavasis, I., Harvey, L.M., McNeil, B., 2006. The effect of agitation and aeration on the
representing thus an efficient rate of bioconversion ability by synthesis and molecular weight of gellan in batch cultures of Sphingomonas
P. taetrolens cells. paucimobilis. Enzyme Microb. Tech. 38, 101–108.
Guimarães, P.M.R., Teixeira, J.A., Domingues, L., 2010. Fermentation of lactose to
bioethanol by yeasts as part of integrated solutions for the valorisation of
cheese whey. Biotechnol. Adv. 28, 375–384.
4. Conclusions Green, B.A., Yu, R.J., Van Scott, E.J., 2009. Clinical and cosmeceutical uses of
hydroxyacids. Clin. Dermatol. 27, 495–501.
Results have revealed how dissolved oxygen availability plays a Huang, W.C., Chen, S.J., Chen, T.L., 2006. The role of dissolved oxygen and function of
agitation in hyaluronic acid fermentation. Biochem. Eng. J. 32, 239–243.
key role in the overall efficiency of the system under study. Poor Koutinas, A.A., Papapostolou, H., Dimitrellou, D., Kopsahelis, N., Katechaki, E.,
bioconversion performances were obtained under vigorous agita- Bekatorou, A., Bosnea, L.A., 2009. Whey valorisation: a complete and novel
tion rates, which did however trigger cell proliferation of P. taetro- technology for dairy industry starter culture production. Bioresour. Technol.
100, 3734–3739.
lens. Conversely, the use of high aeration rates neither stimulated
Lin, W.L., Chen, T.D., Liu, C.W., 2009. Synthesis and characterization of lactobionic
cell growth nor lactobionic acid production due to a prolonged pro- acid grafted pegylated chitosan and nanoparticle complex application. Polymer
liferation phase that delayed the onset of the production phase. 50, 4166–4174.
Therefore, addressing the deleterious influence of high oxygen sup- Lozano, E., Galindo, E., Peña, C.F., 2011. Oxygen transfer rate during the production
of alginate by Azotobacter vinelandii under oxygen-limited and non oxygen-
ply schemes constitutes valuable information for industrial-scale limited conditions. Microb. Cell Fact. 10, 13.
bioproduction of lactobionic acid from whey. Pedruzzi, I., da Silva, E.A.B., Rodrigues, A.E., 2011. Production of lactobionic acid
from lactose/fructose substrate using GFOR/GL enzymes from Zymomonas
mobilis cells: a kinetic study. Enzyme Microb. Tech. 49, 183–191.
Acknowledgements Pedruzzi, I., Malvessi, E., Mata, V.G., Silva, E.A.B., Silveira, M.M., Rodrigues, A.E.,
2007. Quantification of lactobionic acid and sorbitol from enzymatic reaction of
fructose and lactose by high-performance liquid chromatography. J.
The authors are grateful for financial support from the Spanish Chromatogr. A 1145, 128–132.
Ministry of Science and Innovation through project MEC-CTQ2010- Peña, C., Trujillo-Roldan, M.A., Galindo, E., 2000. Influence of dissolved oxygen
14918. tension and agitation speed on alginate production and its molecular weight in
cultures of Azotobacter vinelandii. Enzyme Microb. Tech. 27, 390–398.
Rahulan, R., Dhar, K.S., Nampoothiri, K.M., Pandey, A., 2011. Production of leucine
References amino peptidase in lab scale bioreactors using Streptomyces gedanensis.
Bioresour. Technol. 102, 8171–8178.
Tang, Y.J., Zhong, J.J., 2003. Role of oxygen supply in submerged fermentation of
Affertsholt, T., 2007. Market developments and industry challenges for lactose and
Ganoderma lucidum for production of Ganoderma polysaccharide and ganoderic
lactose derivatives. Presentation at IDF Symposium Lactose and Its Derivatives,
acid. Enzyme Microb. Tech. 32, 478–484.
14–16 May 2007.
Wang, Z.J., Wang, H.Y., Li, Y.L., Chu, J., Huang, M.Z., Zhuang, Y.P., Zhang, S.L., 2010.
Alonso, S., Rendueles, M., Díaz, M., 2011. Efficient lactobionic acid production from
Improved vitamin B12 production by step-wise reduction of oxygen uptake rate
whey by Pseudomonas taetrolens under pH-shift conditions. Bioresour. Technol.
under dissolved oxygen limiting level during fermentation process. Bioresour.
102, 9730–9736.
Technol. 101, 2845–2852.
Amanullah, A., Tuttiet, B., Nienow, A.W., 1998. Agitator speed and dissolved oxygen
Wu, D.-Q., Lu, B., Chang, C., Chen, C.-S., Wang, T., Zhang, Y.-Y., Cheng, S.-X., Jiang, X.-
effects in xanthan fermentations. Biotechnol. Bioeng. 57, 198–210.
J., Zhang, X.-Z., Zhuo, R.-X., 2009. Galactosylated fluorescent labeled micelles as
Bandyopadhyay, B., Humphrey, A.E., Taguchi, H., Rao, G., 2009. Dynamic
a liver targeting drug carrier. Biomaterials 30, 1363–1371.
measurement of the volumetric oxygen transfer coefficient in fermentation
Xu, Y., Zhong, J.-J., 2011. Significance of oxygen supply in production of a novel
systems. Biotechnol. Bioeng. 104, 841–853, Introduction by.
antibiotic by Pseudomonas sp. SJT25. Bioresour. Technol. 102, 9167–9174.
Clark, G.J., Langley, D., Bushell, M.E., 1995. Oxygen limitation can induce microbial
Zhang, J., Li, C., Xue, Z.-Y., Cheng, H.-W., Huang, F.-W., Zhuo, R.-X., Zhang, X.-Z., 2011.
secondary metabolite formation: investigations with miniature electrodes in
Fabrication of lactobionic-loaded chitosan microcapsules as potential drug
shaker and bioreactor culture. Microbiology 141, 663–669.
carriers targeting the liver. Acta Biomater. 7, 1665–1673.
García-Ochoa, F., Gómez, F., Santos, V.E., 2000. Oxygen transfer and uptake rates
Zou, X., Hang, H.J., Chu, J., Zhuang, Y.P., Zhang, S.L., 2008. Oxygen uptake rate
during xantham gum production. Enzyme Microb. Tech. 27, 680–690.
optimization with nitrogen regulation for erythromycin production and scale-
García-Ochoa, F., Gómez, F., 2009. Bioreactor scale-up and oxygen transfer rate in
up from 50 L to 372 m3 scale. Bioresour. Technol. 100, 1406–1412.
microbial processes: an overview. Biotechnol. Adv. 27, 153–176.
140 Resultados
Resultados 141
4.5. Influencia de la heterogeneidad fisiológica sobre la producción

de ácido lactobiónico
No sólo los gradientes de concentración (hidrodinámicos, de sustrato, etc…) influyen
decisivamente sobre el correcto devenir de un bioproceso en términos productivos (como
han recogido de forma excelente Lara et al. (2006)) sino como Enfors et al. (2001) han
señalado magnificamente, las respuestas fisiológicas de los microorganismos (como
consecuencia de la falta de homogeneidad) pueden jugar un papel clave en el desarrollo de
todo sistema fermentativo a escala industrial. Es indudable que la heterogeneidad fisiológica
juega un papel clave en todo bioproceso, incluso dicho concepto está en la génesis de la falta
de adaptación microbiana o en la propia aparición de fases lag. En el presente trabajo, se ha
estudiado por primera vez la heterogeneidad fisiológica subyacente como factor inherente o
implícito al desarrollo del proceso de producción de ácido lactobiónico por P. taetrolens a
partir de suero lácteo en condiciones naturales (sin control de pH y sin aporte de oxígeno,
existiendo por lo tanto una limitación en el nivel de oxígeno disuelto) en contraposición con
las condiciones generadas en un sistema bajo control de pH (pH-shift strategy). Para llevar a
cabo tales conclusiones se han desarrollado y optimizado dos protocolos dual-paramétricos
de citometría de flujo basados en la actividad metabólica, la integridad de la membrana y en
la polarización de la misma. Además, se ha desarrollado una estrategia de separación celular
(fluorescence-activated cell sorting, FACS) en placa de agar (con un screening oxidativo
simultáneo) llegando a poder discriminar y concluir que la gran mayoría de las células
dañadas (el 75%) se corresponde con células viables pero no cultivables (células viable but
not culturable, VBNC). Asimismo, se ha descrito por primera vez (hasta ahora tal enfoque no
había sido desarrollado en ningún sistema fermentativo o cultivo) cómo la edad del inóculo
(asociada a una heterogenidad fisiológica subyacente) conlleva la aparición de fases de
latencia, la falta de adaptación o de proliferación celular de un sistema fermentativo. Así, la
presencia de células muertas o dañadas como subpoblaciones dominantes ha conllevado
cambios más que significativos en los valores asociados al rendimiento y productividad. De
este trabajo se puede concluir que el enfoque clásico de optimización de todo proceso
fermentativo, basado en el seguimiento de determinadas variables macroscópicas, debe
considerar la información aportada por determinados parámetros fisiológicos con el fin de
concocer el estado de salud real (health status).
Artículo: Physiological heterogeneity of Pseudomonas taetrolens during lactobionic acid

production.
Situación: Artículo publicado en Applied Microbiology and Biotechnology (Springer).
Appl Microbiol Biotechnol (2012) 96:1465–1477
DOI 10.1007/s00253-012-4254-2
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Physiological heterogeneity of Pseudomonas taetrolens

during lactobionic acid production
Saúl Alonso & Manuel Rendueles &
Mario Díaz
Received: 16 May 2012 / Revised: 14 June 2012 / Accepted: 15 June 2012 / Published online: 10 July 2012
# Springer-Verlag 2012
Abstract Physiological heterogeneity constitutes a critical of P. taetrolens. It also constitutes a valuable, broad-ranging
parameter in biotechnological systems since both metabolite methodology for the enhancement of microbial bioprocesses
yield and productivity are often hampered by the presence involved in the production of secondary metabolites.
of undesired physiological cell subpopulations. In the pres-
ent study, the physiological status and functionality of Pseu- Keywords Flow cytometry . Physiological status .
domonas taetrolens cells were monitored by multiparameter Lactobionic acid . Whey . Pseudomonas taetrolens .
flow cytometry during fermentative lactobionic acid produc- Secondary metabolite
tion at the shake-flask and bioreactor scale. In shake-flask
fermentation, the onset of the lactobionic acid production
phase was accompanied by a progressive loss of cellular
Introduction
metabolic activity, membrane polarization, and membrane
integrity concomitantly to acidification. In fact, population
Both overall efficiency and metabolite productivity within a
dynamics has shown the prevalence of damaged and dead
bioprocess are strongly reliant on the physiological status
subpopulations when submitted to a pH<4 from 16 h on-
and functionality displayed by microbial cells. In fact, phys-
wards. Furthermore, fluorescence-activated cell sorting
iological heterogeneity constitutes a critical parameter in
revealed that these sublethally injured cells were noncultur-
biotechnological systems since variations in cell capacities
able. In contrast, P. taetrolens cells exhibited a robust phys-
may lead to undesired cell subpopulations of reduced effi-
iological status during bioreactor cultivations performed
ciency (Avery 2006; Müller et al. 2010). The segregation in
with a pH-shifted strategy at 6.5, remaining predominantly
the phenotype of an isogenic population due to unbalanced
healthy and metabolically active (>96 %) as well as main-
environmental or bioprocessing conditions (Amanullah et
taining bioconversion efficiency throughout the course of
al. 2003; Nicolau et al. 2010) can undoubtedly result in poor
the fermentation. Additionally, an assessment of the seed
culture fitness as well as reduced microbial performance
culture’s physiological robustness was carried out in order to
(Zhang et al. 2009).
determine the best seed culture age. Results showed that
In many industrially relevant biotechnological processes,
bioreactor culture performance, growth, and lactobionic acid
bioprocessing conditions often involve cellular stress
production efficiency were strongly dependent on the phys-
responses with substantially deleterious effects on physio-
iological heterogeneity displayed by the seed culture. This
logical status (Enfors et al. 2001; Zhu et al. 2011). Major
study provides the most suitable criteria for optimizing
cell responses have been found to be induced as a result of
lactobionic acid production efficiency through a novel flow
environmental perturbations such as lack of nutrient avail-
cytometric-based approach based on the physiological status
ability (Hewitt et al. 1999; Lopes da Silva et al. 2005),
stressing (Amanullah et al. 2003; Amor et al. 2002; Papa-
S. Alonso : M. Rendueles : M. Díaz (*) dimitriou et al. 2007), or hydromechanical conditions
Department of Chemical Engineering and Environmental (Lüders et al. 2011). Depending on the nature or degree of
Technology, Faculty of Chemistry, University of Oviedo,
these perturbations, loss of cellular functionalities could
C/ Julián Clavería s/n,
33071 Oviedo, Spain consequently result in irreversible cellular senescence
e-mail: [email protected] (Bogosian and Bourneuf 2001; Nyström 2005). The
1466 Appl Microbiol Biotechnol (2012) 96:1465–1477
complex responses and performances of bacteria in biotech- taetrolens during fermentative lactobionic acid production
nological systems have therefore led to the development of at the shake-flask and bioreactor scale, highlighting the
advanced monitoring tools to achieve a better understanding importance of the uncontrolled pH versus pH shift condi-
of microbial physiology at the population level (Díaz et al. tions encountered by the cells in each system. Cellular
2010; Hammes et al. 2011). While conventional bulk parameters such as membrane polarization, membrane in-
approaches for microbial biomass estimation reflect neither tegrity, and metabolic activity of P. taetrolens cells were
the functional information nor the physiological robustness of evaluated and monitored through a multiparameter flow
cells (Hewitt and Nebe-von-Caron 2001), multiparametric cytometry assessment in order to determine whether the
flow cytometry has emerged as a valuable technique to detect physiological age of cells could impair lactobionic acid
and monitor changes in these parameters. Such information on production from whey. The underlying physiology and pop-
the health status of cells may explain cell-to-cell variability in ulation behavior of P. taetrolens were thus deciphered for
behavior, robustness, shape, or size, since changes in these the first time, thereby providing additional key information
microbial parameters are intimately linked to the presence or for the implementation of strategies to prevent potential
absence of an intact fully polarized cytoplasmic membrane deleterious bioprocessing effects on fermentation perfor-
and the transport systems across it (Hewitt and Nebe-von- mance. From a practical point of view, information obtained
Caron 2004). Furthermore, assessment of the physiological from this study could also be of substantial interest for
status of cells using flow cytometry can reveal asynchronous industrially relevant microbial systems submitted to stress-
phenotypic subpopulations comprising non- and metabolite- ing conditions involved in the production of secondary
producing cells (David et al. 2011; Herrero et al. 2006) or even metabolites.
community dynamics within a bacterial consortium (Müller et
al. 2009). Undoubtedly, a concise understanding of the under-
lying microbial physiology provides insights into the most Materials and methods
suitable cultivation strategy for enhanced metabolite produc-
tion, as well as unraveling the deleterious effects on cellular Microorganism
fitness (Lopes da Silva et al. 2009).
Although numerous studies have focused on flow cyto- P. taetrolens LMG 2336, obtained from the Belgian Coor-
metric assessment of bacteria (Díaz et al. 2010; Hammes et dinated Collection of Microorganisms (Gent, Belgium), was
al. 2011), only a few approaches to bioprocess monitoring maintained frozen (in 40 % [v/v] glycerol at −20 °C). This
involving Pseudomonas species have been developed to strain was subsequently subcultured on nutrient broth (NB,
date (Amanullah et al. 2003; Nielsen et al. 2009). Consid- containing 1 g/L meat extract, 2 g/L yeast extract, 5 g/L
ering the relevance of the genus Pseudomonas as a cell peptone, and 5 g/L NaCl) agar plates, incubated for 48 h at
factory, the development and application of flow 30 °C, and then preserved at 4 °C.
cytometry-derived information could be a valuable tool for
enhanced biological performance of incipient bioprocesses. Inoculum preparation
Bioproduction of lactobionic acid from an inexpensive
source such as cheese whey through a fermentation-based A loopful of P. taetrolens from a fresh nutrient broth agar
process carried out by Pseudomonas taetrolens has thus plate was used to inoculate a 500-mL Erlenmeyer flask
become both a promising and attractive alternative to obtain containing 100-mL of nutrient broth medium. This flask
this value-added substance (Alonso et al. 2011, 2012). Lac- was incubated on an orbital shaker (Infors HT, model Flyer
tobionic acid has recently emerged as a relevant commercial Aerotron, Switzerland) at 250 rpm and 30 °C for 10 h.
polyhydroxy acid which displays interesting pharmaceutical Actively growing cells from this culture were then
properties with promising biomedical applications (Lin et al. employed as inoculum for the production of lactobionic acid
2011; Zhang et al. 2011). This versatile bioproduct also in shake-flask and bioreactor seed cultures containing sweet
finds a broad range of applications in the cosmetics industry, whey.
in which it is being currently employed as the key active
component in novel skin care products (Green et al. 2009). Sweet whey preparation
In view of the relevance of lactobionic acid bioproduction,
significant improvements in microbial performance could be Sweet cheese whey (provided by ILAS S.A., Asturias,
achieved after revealing the underlying microbial physiolo- Spain) was 1-fold diluted with distilled water (1:1) and
gy and understanding the complex interaction between P. adjusted to pH 6.5 (by adding NaOH 6 N) prior to steriliza-
taetrolens cells and their surrounding microenvironment. tion using a tangential microfiltration device equipped with
In this context, the aim of this study was to gain a deeper a 0.22-μm pore size PVDF membrane cassette (Millipore,
process understanding of the physiological status of P. MA, USA).
Appl Microbiol Biotechnol (2012) 96:1465–1477 1467
Shake-flask fermentations Staining procedures
Shake-flask fermentation experiments were conducted in Samples from cultures were harvested by centrifugation at
500 mL Erlenmeyer flasks containing 100 mL of sweet 16,000×g for 5 min. Before staining, cells were washed
whey inoculated with 0.64 g/L of wet biomass from NB twice in phosphate-buffered saline (PBS, pH 7.4, sterile
inoculum cultures harvested by centrifugation at 11,000×g and filtered at 0.22 μm), and then cells were held in the
for 10 min. These cultures were subsequently incubated on “hot spot” of a sonication bath for 2 s in order to prevent
an orbital shaker (New Brunswick Scientific, NJ, USA) at bacterial aggregation before flow cytometric analysis
250 rpm and 30 °C. Samples were aseptically withdrawn (Hewitt and Nebe-von-Caron 2004). Propidum iodide (PI,
periodically to determine bacterial growth and pH. Biomass Invitrogen), bis-(1,3-dibutylbarbituric acid) trimethine oxo-
was removed by centrifugation at 16,000×g for 5 min, and nol (bis-oxonol, DiBAC4(3), Invitrogen), and carboxyfluor-
the cell-free supernatants being stored frozen (−20 °C) until escein diacetate (cFDA, Invitrogen) were used as
further analysis. Bioconversion experiments were carried fluorescent dyes in a double dual-staining procedure
out in triplicate as independent trials. (DiBAC4(3)/PI and cFDA/PI) in order to evaluate cell phys-
iological status (metabolic activity, membrane integrity, and
Bioreactor cultivations membrane polarization were evaluated through cFDA, PI,
and DiBAC4(3) staining, respectively). Stock solutions were
Batch cultivations for lactobionic production from sweet prepared as follows: PI was made up to 1 mg/mL in distilled
whey were also performed in a 2-L bioreactor (BioFlo water (0.22 μm filtered) and maintained at 4 °C, whereas
110, New Brunswick Scientific, NJ, USA) with a work- both DiBAC4(3) and cFDA were prepared in dimethyl sulf-
ing volume of 1 L. Bioreactor experiments were con- oxide (Sigma-Aldrich) at a concentration of 1 mM and
ducted with an inoculation level of 10 % (v/v), an stored at −20 °C. PI staining was prepared by diluting the
agitation rate of 350 rpm, and at 30 °C. Moreover, the stock solution in sterile distilled water and adding this work
bioreactor was aerated (1.0 vvm) through a ring sparger solution to the cell suspension at a final concentration of
located at the bottom of the vessel. Excessive foam 5 μg/mL. This mixture was then incubated for 30 min in the
formation was prevented by automatic addition of dilut- dark at room temperature. Working solutions of DiBAC4(3)
ed (1:10) Y-30 emulsion (Sigma-Aldrich, Steinheim, and cFDA were made up to 25 and 10 μM in PBS contain-
Germany). The bioreactor was equipped with a pH ing 1 mM EDTA, respectively. PI-stained samples were
meter (Mettler Toledo, Switzerland) and a polarographic subsequently incubated with 1 μM DiBAC4(3) or 0.1 μM
dissolved oxygen electrode (InPro 6830, Mettler Toledo, cFDA for 15 min in the dark at room temperature.
Switzerland) for continuous on-line monitoring of pH
and dissolved oxygen tension (DOT) values, respective- Multiparameter flow cytometry
ly. A two-stage pH-shifted bioconversion strategy was
adopted in the bioreactor experiments as previously Flow cytometry measurements were performed using a
reported (Alonso et al. 2011). The strategy consisted Cytomics FC 500 flow cytometer (Beckman Coulter)
in controlling pH above 6.5 (pH was left uncontrolled equipped with a 488- and 633-nm excitation light source
above this value during the growth phase and then from an argon ion laser. Green fluorescence from samples
maintained at 6.5) by means of computer-controlled (corresponding to DiBAC4(3) and cFDA-stained cells) was
peristaltic pumps via automatic addition of 2 M NaOH. collected on the FL1 channel (530 nm), whereas PI fluores-
Cultivations were carried out in duplicate as indepen- cence was registered on the FL3 channel (610 nm). Each
dent experiments. analysis was performed in duplicate at a low flow rate
The seed culture was obtained by harvesting biomass setting (4,000 events/s). Data acquisition was carried out
(after centrifugation at 11,000×g for 10 min) from a NB using Cytomics RXP software (Beckman Coulter). Gates
inoculum culture and re-suspending cells in a 500-mL and quadrants were established according to staining con-
Erlenmeyer flask containing 100 mL of sweet whey. trols. For DiBAC4(3)/PI and cFDA/PI dual parameter flow
Subsequently, these cultures were grown at 250 rpm in cytometric analysis, data collected from 150,000 and
an orbital shaker at 30 °C for 12 (control experiment, 100,000 events, respectively, were analyzed using Summit
also used for the sake of comparison with shake-flask v4.3 software (DakoCytomation, CO, USA).
cultivation), 24, 36, or 48 h. Each culture was then
employed as seed culture at 10 % (v/v) in the bioreac- Fluorescence-activated cell sorting
tor containing sweet whey in order to study the influ-
ence of cellular physiological age on bioprocess Cell sorting was performed using a MoFlo XDP cell sorter
performance. (DakoCytomation) equipped with an air-cooled argon ion
laser emitting blue light at 488 and 630 nm (200 mW). time following a momentary interruption of air supply to the
The green fluorescence from cFDA-stained cells was bioreactor (Alonso et al. 2012). The specific consumption
detected through a 530-nm filter (FL1 channel), whereas rate of oxygen (qO2) was calculated by dividing the OUR
the red fluorescence from the PI signal was collected on value by the biomass concentration at each sampling point.
the FL3 channel (610 nm). Fluorescent beads (Flow-
check Pro Fluorospheres, Beckman Coulter) with sizes
of 10, 6, and 3 μm were used to check the fluidics Results
system and flow cytometer’s optical alignment (coeffi-
cient of variation, CV value, around 2 %). Samples Monitoring of the physiological status of P. taetrolens
were sorted in purify mode at a rate of 200 cells/s using during shake-flask fermentation
Isoton II (Beckman Coulter) as the sheath fluid. The
instrument also operated at a system pressure of 54 psi Figure 1 shows the changes in the physiological status of P.
and was equipped with a nozzle size of 70 μm. Sorting taetrolens in terms of metabolic activity, membrane integri-
gates were established on the basis of FL1 (cFDA(+)/PI ty, and polarization during shake-flask fermentation. While
(−) cells) versus FL3 fluorescence (cFDA(−)/PI(+) panels a and b of Fig. 1 both illustrate the dual flow cyto-
cells). Data acquisition was performed using Summit metric assessment of green fluorescence versus red fluores-
v5.0 software (DakoCytomation). Bacterial cells from cence (DiBAC4(3)/PI and cFDA/PI, respectively), Fig. 1c
gates were sorted onto agar plates containing lactose shows representative dot plot diagrams of the side scatter
(10 g/L), peptone (2 g/L), yeast extract (1 g/L), (SSC) versus forward side scatter (FSC) signal obtained
K2HPO4 (0.2 g/L), bromophenol blue (0.08 g/L), and during cultivation. The upper left quadrant of dot plots in
agar (15 g/L) with the aim of screening culturability and Fig. 1a shows depolarized cells (DiBAC 4(3)-positive;
lactose-oxidizing ability by P. taetrolens cells, as previ- DiBAC4(3) enters cells only when the membrane potential
ously reported (Alonso et al. 2011). Subpopulations of is lost), while cells with esterase activity (cFDA-positive
interest were thus sorted onto 96 spots on agar plates, cells; cFDA is cleaved by the esterase activity inside the
which were then incubated for 96 h at 30 °C. living cells, thus releasing a polar fluorescent portion which
is unable to pass through the intact membrane) are shown in
Fluorescence and confocal microscopy the upper left quadrant of Fig. 1b. Damaged cells
(corresponding to cFDA(+)/PI(+) cells or DiBAC4(3)(+)/PI
Stained samples were also examined under a Leica TCS- (−) cells) are shown in the upper left quadrant (Fig. 1a) and
SP2-AOBS confocal laser scanning microscope (Leica in the upper right quadrant (Fig. 1b) for the DiBAC4(3)/PI
Microsystems Inc., Heidelberg, Germany) at excitation and cFDA/PI staining, respectively. Likewise, dead cells
wavelengths of 488 and 568 nm with an emission wave- (PI-positive cells; PI can only cross the plasmatic membrane
length of 530 (green fluorescence) or 630 nm (red if it is permeabilized corresponding to those cells whose
fluorescence). membranes are compromised) are shown in the upper right
quadrant and in the lower right quadrant for the DiBAC4(3)/
Analytical methods PI (Fig. 1a) and cFDA/PI staining (Fig. 1b), respectively.
Examination of these cytograms reveals that shake-flask
Bacterial growth was measured spectrophotometrically as cultivation displayed marked physiological heterogeneity
optical density at 600 nm (Shimazdu, UV 1203 model) after in terms of metabolic status, membrane integrity, and cellu-
centrifugation of culture samples at 16,000×g for 5 min. lar size. Results based on esterase activity and membrane
Optical density data were converted to cell dry weight integrity thus suggest that loss of metabolic activity was
(expressed in grams per liter) using the corresponding pre- progressive, leading first to an intermediate “damaged” cell
viously obtained calibration curve. The lactobionic acid and state and then irreversibly to cell death (Fig. 1b). According
lactose content of cell-free culture samples were measured to Fig. 1a, cells also showed a poor health status in terms of
by high performance liquid chromatography as reported membrane polarization at later cultivation stages (36 and
previously by Alonso et al. (2011). 48 h) given that the presence of an intact, fully polarized
cytoplasmic membrane represents a signal of cellular ro-
Determination of the oxygen uptake rate bustness (Amor et al. 2002; Hewitt and Nebe-von-Caron
2004). As can be seen in Fig. 1c, an increase in the intensity
The oxygen uptake rate (OUR) from bioreactor cultivations of the FSC was evidenced in the flow cytometric dot plots.
was determined via the dynamic method (Bandyopadhyay et Cells of P. taetrolens exhibited both higher relative cell size
al. 2009). Consequently, the OUR was obtained from the (forward scatter, FSC) and cell complexity signals (side
slope of the plot of dissolved oxygen concentration versus scatter, 90°-angle scatter of blue laser light, SSC) with age.
A
0h 10 h 24 h 34 h 48 h
DiBAC4(3)
PI
B
0h 10 h 24 h 34 h 48 h
cFDA
PI
C
0h 10 h 24 h 34 h 48 h
SSC
FSC
Fig. 1 Dot plots representing DiBAC4(3) fluorescence versus PI fluorescence (a), cFDA fluorescence versus PI fluorescence (b), and side scatter
light (SSC) versus forward scatter light (FSC) signals (c) of P. taetrolens cells during shake-flask fermentation
Figure 2a shows the bioprocess parameters obtained production phase (Fig. 2a, b). As a result, the population
during shake-flask fermentation, which was featured by dynamics showed an increase in the percentage of positively
a specific lactobionic acid productivity of 0.38 g/g DiBAC4(3)-stained cells from 2 to 15 %. The onset of the
DCW h. These data provide evidence that lactobionic lactobionic acid production phase was accompanied by the
acid constitutes a secondary metabolite mostly produced passing of healthy, metabolically active cells into a damaged
by P. taetrolens during the stationary growth phase, as state, which was accompanied by a progressive loss of
previously reported (Alonso et al. 2011). Cultivation membrane integrity and the predominance of cellular death
reached the stationary growth phase after 10 h, coincid- from 36 h onwards (Fig. 2b), suggesting that the cells were
ing with the onset of the production phase (Fig. 2a), highly affected by prolonged exposure to extreme acidic
reaching a maximum biomass of 1.84 g/L. During the conditions (pH03.35). Obviously, cells with a permeabi-
growth phase, the pH first rose (with a pH shift value lized cytoplasmic membrane cannot sustain the electro-
of 7.36 at 6 h) and then decreased abruptly, thereby chemical gradient, leading irreversibly to cellular
signaling the onset of the production phase. A pH value senescence (Díaz et al. 2010). Consequently, the lactobionic
of 3.42 was reached after 24 h, remaining unchanged at acid production phase in shake-flask cultivation was high-
3.35 from 48 h onwards, which coincided with the lighted by a poor health status of P. taetrolens cells in terms
depletion of lactose content in the fermentation broth of metabolic activity concomitant to cellular depolarization
(Fig. 2a). and an increase in the FSC signal.
As can be seen in Fig. 2b, cells were predominantly
healthy and metabolically active (>90 %) during the cell Cell sorting at the single cell level
proliferation phase (0–10 h), simultaneously displaying an
intact and polarized membrane. However, from 12 to 16 h, Cell sorting can reveal the relationship between the
there was a sudden change in these parameters concomitant different subpopulations within a bacterial monoculture
to the abrupt drop in pH from 6.2 to 4 and coinciding with and their culturability (Papadimitriou et al. 2006; Want
the transition phase between growth and the lactobionic acid et al. 2011). Hence, with the aim of shedding more light
Fig. 2 Time-course profile of

bioprocess parameters obtained
during shake-flask cultivation
of P. taetrolens (a). Percentages
of P. taetrolens cell subpopula-
tions throughout the shake-flask
fermentation (b)
on the underlying physiology during shake-flask culti- Monitoring of the physiological status of P. taetrolens
vation, P. taetrolens cells were sorted individually onto during pH shift bioreactor cultivation
agar plates containing lactose. The observation of
stained samples under fluorescence microscopy revealed Flow cytometric dot plots obtained during the pH shift
that P. taetrolens cells were predominantly metabolically bioreactor cultivation are shown in Fig. 4. In accordance
active at 12 h, whereas cells at the late stationary phase with the data, cells displayed a robust physiological status
(36 and 48 h) were dead (Fig. 3a). As can be seen in throughout the course of the fermentation (Fig. 4a, b). As
Fig. 3a, aggregates of dead cells were detected and Fig. 4b shows, cells exhibited high metabolic activity and
verified during the later stages of cultivation (36 and retained their membrane integrity even at late stages of
48 h). Interestingly, the observed increase in light scat- cultivation. The stationary phase was also accompanied by
tering properties during flow cytometric analysis slight membrane depolarization, although the fluorescence
(Fig. 1c) could be ascribed to this phenomenon, as intensity was significantly lower than that obtained in the
previously reported by Kottmeier et al. (2009) for Han- shake-flask culture.
senula polymorpha cells. Figure 5a shows the bioprocess parameters obtained dur-
In accordance with the results obtained in Fig. 3b, cell ing pH shift bioreactor cultivation, which showed 1.5-fold
sorting on agar plates demonstrated that the majority of cells higher specific lactobionic acid productivity (0.56 g/g DCW
at 12 h were culturable (97.9 %), while damaged cells h) than that obtained in shake-flask cultivation. As can be
(corresponding to the cFDA(+)/PI(+) cells found at 24 h) seen, the growth phase was featured by an abrupt drop in the
displayed a culturability of only 24.8 % (Fig. 3c). There was DOT concomitant to a sharp increase in pH values (pH shift
an abrupt decrease in the culturability of sorted cells of 8 at 8 h). The biomass reached a maximum value of
(corresponding to cFDA(−)/PI(+) cells) at the late stationary 1.26 g/L at the end of the exponential growth phase, with
growth phase (36 and 48 h), as no cell was capable of a specific growth rate of 0.19 h−1. The onset of the produc-
growing on agar plates (Fig. 3b, c), verifying that these cells tion phase took place with a concurrent return of DOT to
were nonculturable and dead. values near to air saturation level (indicating the end of the
A
12 h 24 h 36 h 48 h
B 12 h 24 h 36 h 48 h
Fig. 3 Fluorescence micrographs of P. taetrolens cells from shake- from shake-flask cultivation at 12, 24, 36, and 48 h which were sorted
flask fermentation stained with cFDA (green fluorescence) and PI (red directly on agar plates containing lactose (b). Percentages of cells that
fluorescence) (a). Aggregates of dead P. taetrolens cells were detected formed or did not form colonies on agar plates after fluorescence-
at later stages of cultivation (36 and 48 h). Culturability of stained cells activated cell sorting (c). Scale bars05 μm
exponential cell growth phase). The profile of lactobionic acid Influence of the physiological age of cells on fermentation
production confirmed that lactobionic acid is a non-growth- performance at the bioreactor scale
related product, showing the typical production pattern from
whey by P. taetrolens (Alonso et al. 2011, 2012). A maximum Figure 6 shows the comparative profiles of bioprocess
specific oxygen consumption rate (qO2) of 156 mg/g h was parameters obtained in pH shift bioreactor cultivations
obtained at the beginning of the fermentation process, subse- employing 12-, 24-, 36-, and 48-h seed cultures with the
quently remaining around 1.75 mg/g h during the stationary aim of determining the influence of seed culture age. As can
growth phase (Fig. 5a). Obviously, the sudden decrease in the be seen in Fig. 6a, younger seed cultures contributed to
DOT level during the first 8 h may be attributed to the increase enhanced growth of P. taetrolens, obtaining a maximum
in oxygen consumption resulting from the active growth pro- biomass concentration of 1.25, 1.07, 0.98, and 0.85 g/L
liferation of P. taetrolens cells. for 12-, 24-, 36-, and 48-h seed cultures, respectively. Both
Analysis of the physiological status of the cells revealed cell growth and the specific growth rate were clearly im-
that 96 % of these remained healthy, metabolically active, paired by the increase in seed culture age, as can be seen in
and polarized during the cell proliferation phase (Fig. 5b). Table 1. Moreover, large lag phases preceded microbial
Upon shifting to the production phase, no significant growth in batch cultivations inoculated with seed cultures
changes were found in terms of metabolic activity (~96 % older than 12 h (Fig. 7). This lack of growth caused a
of the cells). prolonged delay in the onset of the production phase: the
A
0h 10 h 24 h 34 h 48 h
DiBAC4 (3)
PI
B
0h 10 h 24 h 34 h 48 h
cFDA
PI
C
0h 10 h 24 h 34 h 48 h
SSC
FSC
Fig. 4 Dot plots representing DiBAC4(3) fluorescence versus PI fluorescence (a), cFDA fluorescence versus PI fluorescence (b), and light
scattering properties (c) of P. taetrolens cells during pH shift bioreactor cultivation employing a seed culture age of 12 h
Fig. 5 Bioprocess parameters

obtained during pH shift
bioreactor cultivation of P.
taetrolens (a). Percentages of P.
taetrolens cell subpopulations
throughout the bioreactor
cultivation (b) employing a
seed culture age of 12 h
A 1.5 B
100
Dry cell weight (g/L) 80
DOT (%)
1
60
12-h 40 12-h
0.5 24-h
24-h
36-h 20 36-h
48-h 48-h
0 0
0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80
Time (h) Time (h)
C 8.5 D 100
8 12-h 80
OUR (mg/L.h)
24-h
12-h
7.5 60
pH
36-h
24-h
48-h
7 40 36-h
48-h
6.5 20
6 0
0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80
Time (h) Time (h)
E 50 F 50
12-h
40 40 24-h
Lactose (g/L)
36-h
30 30
48-h
20 12-h 20
24-h
10 36-h 10
48-h
0 0
0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80
Time (h) Time (h)
Fig. 6 Time-course profiles of cell growth (a), DOT (b), pH (c), OUR (d), lactose (e), and lactobionic acid (f) during pH shift bioreactor
cultivations employing a seed culture age of 12, 24, 36, or 48 h
lactobionic acid production phase began at 38 h when onset of the production phase. Furthermore, there was a
employing a 48-h seed culture compared to 10 h when a substantial difference in the minimum DOT levels obtained
12-h seed culture was inoculated. in the fermentation broth. A minimum DOT level of 20 %
Dissolved oxygen profiles were strongly related to the was reached during growth phase employing a 12-h seed
growth phase, as can be appreciated in Fig. 6b. A pro- culture compared to the 70 % obtained when using a 48-
nounced depletion in dissolved oxygen concentration was h seed culture (Fig. 6b).
found at the initial stage of the cell growth phase followed As regards pH profiles, fermentation broths suffered the
by a sudden rapid increase to saturated values, indicating the usual sharp increase in pH values during the exponential
Table 1 Comparison of values obtained in pH shift bioreactor cultivations under the influence of different seed culture ages
Inoculum Maximum dry cell Duration of the lag Onset of the pH shift μ Volumetric Culture Yield
age (h) weight (g/L) phase (h) production phase (h) value (h−1) productivity (g/L h)a time (h) (%)b
12 1.25±0.04 0 10 8.00±0.05 0.19 0.70±0.00 58 100±0

24 1.07±0.05 4 16 8.02±0.08 0.15 0.45±0.02 80 90±2
36 0.98±0.05 8 26 8.19±0.03 0.10 0.30±0.02 80 63±1
48 0.85±0.07 16 38 8.43±0.02 0.06 0.20±0.03 80 46±2
μ specific growth rate

a
Data were calculated considering each culture time
b
Yield was defined as the percentage of lactose converted into lactobionic acid after each culture time
growth phase (Alonso et al. 2011, 2012), as can be seen in main cell subpopulations in the 12- (Fig. 5b) and 24-
Fig. 6c. The increase in seed culture age led to a progressive h (Fig. 8b) seed cultures, respectively. As Fig. 8a shows,
rise in pH shift values (8.00, 8.02, 8.19, and 8.43 for 12, 24, cells from this latter culture showed a growth advantage
36, and 48 h, respectively). The maximum oxygen demand over cells from 36- and 48-h seed cultures, seeing as they
by P. taetrolens cells coincided with the proliferation phase, rapidly entered into a metabolically active status. Thus, once
as can be appreciated in Fig. 6d, whereas OUR values cells adapted their cellular machinery (identified by the
remained constant or slightly reduced at the onset of the disappearance of the lag phase), dead and damaged subpo-
production phase. The cultivation inoculated with the youn- pulations were partially replaced by cells that remained
gest seed culture (12 h) presented a maximum OUR value of metabolically active (Fig. 8b, c).
87.5 mg/L h at 4 h, compared to similar maximum OUR Thus, the use of an active healthy seed culture minimized
values of 64, 61, and 60 mg/L h obtained when employing the duration of the lag phase in the subsequent pH shift
seed culture ages of 24, 36, and 48 h, respectively. cultivation and also the timing thereof for switching from
As can be seen in Fig. 6f, an increase in seed culture age growth to product formation. Prolonged time cultures (lon-
resulted in a marked decrease in the lactobionic acid titers ger than 12 h) employed as seed cultures displayed poor
produced by P. taetrolens. Both volumetric productivities bioconversion efficiencies as well as poor fermentation per-
and bioprocess yields were severely affected when employ- formances (Fig. 7) due to the presence of damaged and dead
ing prolonged time seed cultures (Fig. 7), being 1.55-, 2.3- cells in these cultures (Fig. 8).
and 3.5-fold (0.45, 0.30, and 0.2 g/L h for 24-, 36-, and 48-
h seed cultures, respectively) lower than those obtained
using the youngest seed culture (0.70 g/L h for a 12- Discussion
h seed culture). These results provide evidence that senes-
cent cultures employed as inocula exhibited a poor physio- Identifying the complex population dynamics as well as the
logical status which impaired subsequent re-growth and underlying physiology currently remains one of the major
lactobionic acid production. In accordance with the results challenges for the industrial development of many biopro-
obtained (Fig. 8), prolonged time seed cultures (36 and 48 h) cesses in which microbial biomass would otherwise be
were predominantly composed of dead cells, whereas approached as a whole steady-state bulk parameter, conse-
healthy, metabolically active and damaged cells were the quently leading to misinterpretations. In contrast, assessment
Fig. 7 Relationship between

seed culture age and bioprocess
parameters obtained during
batch cultivations
status of cells as well as valuable guidelines on bioprocessing

conditions for enhanced microbial performance (Hewitt and
Nebe-von-Caron 2001; Lopes da Silva et al. 2009). In this
study, flow cytometric assessment of P. taetrolens cells has
revealed major functional and physiological changes during
lactobionic acid production from whey, in addition to deci-
phering the influence of this physiological heterogeneity on
culture performance.
Shake-flask cultivation showed that P. taetrolens cells
suffered a progressive loss of metabolic activity and mem-
brane integrity compared to bioreactor cultivations, in which
the imposed pH shift bioprocessing conditions supported a
robust physiological status (Figs. 2b and 5b). In shake-flask
fermentation, the transition from the growth to the stationary
growth phase was accompanied by an abrupt loss in cellular
parameters. Thus, physiologically and functionally compro-
mised cells were predominant during the later stages of
shake-flask fermentation, suggesting that prolonged expo-
sure to the acidic environment first led to a damaged state
and then to cellular death (Figs. 2b and 3a). Cellular param-
eters such as membrane potential, membrane integrity,
pump activity, or metabolic activity are identified as the
main indicators of microbial viability (Hewitt and Nebe-
von-Caron 2004; Hammes et al. 2011). In fact, cells with
damaged or compromised membranes are not able to main-
tain or generate the electrochemical gradient which is in-
volved in the generation of energy or nutrient transport
(Díaz et al. 2010). In this regard, loss of membrane potential
usually arises during the stationary phase due to higher
stressing levels or deleterious bioprocessing conditions
(David et al. 2011; Nielsen et al. 2009).
Undoubtedly, exposure to deleterious bioprocessing con-
ditions involves major physiological responses from bacte-
ria, as cellular homeostasis and the metabolic machinery of
the cells are highly influenced by the surrounding microen-
vironment (Nicolau et al. 2010). The onset of the stationary
phase in gram-negative bacteria also comprises major mor-
phological changes through an alteration of their size and
shape (Navarro et al. 2010). In fact, P. taetrolens cells
became smaller and spherical at the late stationary phase
compared to the rod-shaped form they showed at early
stages of shake-flask cultivation (Fig. 3a). According to
the data, the functional status of P. taetrolens was intimately
linked to the extreme environment encountered during
shake-flask cultivation (pH03.35), seeing as cells could
not maintain optimal homeostasis under prolonged exposure
to acidic conditions (Fig. 2a, b). These severe conditions
clearly led to P. taetrolens cells entering the death phase as
Fig. 8 Time-course profiles of cell subpopulations during pH shift biore- well as the formation of cellular aggregates (Fig. 3a). Con-
actor cultivations employing a seed culture age of 24 (a), 36 (b), or 48 h (c) versely, cells remained predominantly healthy and metabol-
ically active (>96 %) in addition to displaying an intact
of the physiological state of cells by means of multiparameter depolarized membrane throughout pH shift bioreactor culti-
flow cytometry can provide relevant information on the health vation (Fig. 5b), highlighting the suitability of this
cultivation strategy. Accordingly, pH was controlled above coefficient in fermentation systems. Biotechnol Bioeng 104
6.5 during the lactobionic acid production phase (Fig. 5a) in (5):841–853
Bogosian G, Bourneuf EV (2001) A matter of bacterial life and death.
order to prevent the deleterious bioprocessing conditions EMBO Rep 2(9):770–774
encountered in shake-flask cultivation (Fig. 2a, b). David F, Berger A, Hänsch R, Rohde M, Franco-Lara E (2011) Single
Physiological heterogeneity represents the main source of cell analysis applied to antibody fragment production with Bacil-
poor biological performance and yield in microbial cultures lus megaterium: development of advanced physiology and bio-
process state estimation tools. Microb Cell Fact 10:23
(Avery 2006; Müller et al. 2010). In accordance with the Díaz M, Herrero M, García LA, Quirós C (2010) Application of flow
results, fermentation performance was strongly affected by cytometry to industrial microbial bioprocesses. Biochemical Eng
the use of senescent seed cultures older than 12 h (Fig. 6), J 48(3):385–407
highlighting the importance of an appropriate time criteria Enfors SO, Jahic M, Rozkov A, Xu B, Hecker M, Jürgen B, Krüger E,
Schweder T, Hamer G, O’Beirne D, Noisommit-Rizzi N, Reuss M,
for harvesting the seed culture. Accordingly, a seed culture Boone L, Hewitt C, McFarlane C, Nienow A, Kovacs T, Trägårdh
of 12 h (mostly comprising healthy, metabolically active C, Fuchs L, Revstedt J, Friberg PC, Hjertager B, Blomstein H, Hjort
cells, as shown in Fig. 5b) may be established as the best S, Hoeks F, Lin HY, Neubauer P, van der Lans R, Luyben K, Vrabel
physiological age in terms of biological performance P, Manelius Å (2001) Physiological responses to mixing in large
scale bioreactors. J Biotechnol 85(2):175–185
(Fig. 6). Otherwise, the prevalence of sublethally injured Green BA, Yu RJ, Van Scott EJ (2009) Clinical and cosmeceutical uses
or dead cells in the seed culture could impair bioconversion of hydroxyacids. Clin Dermatol 27(5):495–501
efficiency as well as the adaptive response by P. taetrolens Hammes F, Berney M, Egli T (2011) Cultivation-independent assess-
cells once inoculated (Figs. 7 and 8). ment of bacterial viability. Adv Biochem Eng Biot 124:123–150
Herrero M, Quirós C, García LA, Díaz M (2006) Use of flow cytom-
Considered as a whole, this study has provided an ap- etry to follow the physiological states of microorganisms in cider
proach for optimizing lactobionic acid production from fermentation processes. Appl Environ Microbiol 72(10):6725–
whey based on the physiological status of P. taetrolens, 6733
deciphering for the first time the role played by this under- Hewitt CJ, Nebe-von-Caron G (2001) An industrial application of
multiparameter flow cytometry: assessment of cell physiological
lying physiology in fermentation performance and lacto- state and its application to the study of microbial fermentations.
bionic acid productivity. Furthermore, the use of this Cytometry 44(3):179–187
information could be of considerable interest for the devel- Hewitt CJ, Nebe-von-Caron G (2004) The application of multi-
opment and optimization of biotechnological systems target- parameter flow cytometry to monitor individual microbial cell
physiological state. Adv Biochem Eng Biot 89:197–223
ing the microbial production of secondary metabolites. Hewitt CJ, Nebe-von-Caron G, Nienow AW, McFarlane CM (1999)
The use of multi-parameter flow cytometry to compare the phys-
Acknowledgments The authors are grateful for the financial support iological response of Escherichia coli W3110 to glucose limita-
from the Spanish Ministry of Science and Innovation through project tion during batch, fed-batch and continuous culture cultivations. J
MEC-CTQ2010-14918. The technical assistance of Ana Salas, Mag- Biotechnol 75(2–3):251–264
dalena Choda (Flow Cytometry Area, Scientific-Technical Services, Kottmeier K, Weber J, Müller C, Bley T, Büchs J (2009) Asymmetric
University of Oviedo), and Marta Alonso (Process Image Area, division of Hansenula polymorpha reflected by a drop of light
Scientific-Technical Services, University of Oviedo) is gratefully scatter intensity measured in batch microtiter plate cultivations at
acknowledged. phosphate limitation. Biotechnol Bioeng 104(3):554–561
Lin WJ, Chen TD, Liu CW, Chen JL, Chang FH (2011) Synthesis of
lactobionic acid-grafted-pegylated-chitosan with enhanced
References HepG2 cells transfection. Carbohydr Polym 83(2):898–904
Lopes da Silva T, Reis A, Kent CA, Kosseva M, Roseiro JC, Hewitt CJ
(2005) Stress-induced physiological responses to starvation peri-
Alonso S, Rendueles M, Díaz M (2011) Efficient lactobionic acid ods as well as glucose and lactose pulses in Bacillus licheniformis
production from whey by Pseudomonas taetrolens under pH- CCMI 1034 continuous aerobic fermentation processes as mea-
shift conditions. Bioresour Technol 102(20):9730–9736 sured by multi-parameter flow cytometry. Biochemical Eng J 24
Alonso S, Rendueles M, Díaz M (2012) Role of dissolved oxygen (1):31–41
availability on lactobionic acid production from whey by Pseu- Lopes da Silva T, Piekova L, Mileu J, Roseiro JC (2009) A compar-
domonas taetrolens. Bioresour Technol 109:140–147 ative study using the dual staining flow cytometric protocol ap-
Amanullah A, Hewitt CJ, Nienow AW, Lee C, Chartrain M, Buckland plied to Lactobacillus rhamnosus and Bacillus licheniformis batch
BC, Drew SW, Woodley JM (2003) Measurement of strain- cultures. Enzyme Microb Tech 45(2):134–138
dependent toxicity in the indene bioconversion using multiparam- Lüders S, David F, Steinwand M, Jordan E, Hust M, Dübel S, Franco-
eter flow cytometry. Biotechnol Bioeng 81(4):405–420 Lara E (2011) Influence of the hydromechanical stress and tem-
Amor BK, Breeuwer P, Verbaarschot P, Rombouts FM, Akkermans perature on growth and antibody fragment production in Bacillus
AD, De Vos WM, Abee T (2002) Multiparametric flow cytometry megaterium. Appl Microbiol Biot 91(1):81–90
and cell sorting for the assessment of viable, injured, and dead Müller S, Vogt C, Laube M, Harms H, Kliensteuber S (2009) Com-
Bifidobacterium cells during bile salt stress. Appl Environ Microb munity dynamics within a bacterial consortium during growth on
68(11):5209–5216 toluene under sulfate-reducing conditions. FEMS Microbiol Rev
Avery SV (2006) Microbial cell individuality and the underlying 70(3):586–596
sources of heterogeneity. Nat Rev Microbiol 4(8):577–587 Müller S, Harms H, Bley T (2010) Origin and analysis of microbial
Bandyopadhyay B, Humphrey AE, Taguchi H. Introduction by Rao G population heterogeneity in bioprocesses. Curr Opin Biotech 21
(2009) Dynamic measurement of the volumetric oxygen transfer (1):100–113
Navarro JM, Tormo A, Martínez-García E (2010) Stationary phase in by flow cytometry and single-cell sorting. Appl Environ Microb 73
gram-negative bacteria. FEMS Microbiol Rev 34(4):476–495 (2):465–476
Nicolau SA, Gaida SM, Papoutsakis ET (2010) A comparative view of Want A, Hancocks H, Thomas CR, Stocks SM, Nebe-von-Caron
metabolite and substrate stress and tolerance in microbial biopro- G, Hewitt CJ (2011) Multi-parameter flow cytometry and
cessing: from biofuels and chemicals, to biocatalysis and biore- cell sorting reveal extensive physiological heterogeneity in
mediation. Metab Eng 12(4):307–331 Bacillus cereus batch cultures. Biotechnol Lett 33(7):1395–
Nielsen TH, Sjøholm OR, Sørensen J (2009) Multiple physiological 1405
states of a Pseudomonas fluorescens DR54 biocontrol inoculant Zhang Y, Zhu Y, Zhu Y, Li Y (2009) The importance of engineering
monitored by a new flow cytometry protocol. FEMS Microbiol physiological functionality into microbes. Trends Biotechnol 27
Ecol 67(3):479–490 (12):664–672
Nyström T (2005) Bacterial senescence, programmed death, and pre- Zhang J, Li C, Xue ZY, Cheng HW, Huang FW, Zhuo RX, Zhang XZ
meditated sterility. ASM News 71:363–369 (2011) Fabrication of lactobionic-loaded chitosan microcapsules
Papadimitriou K, Pratsinis H, Neve-von-Caron G, Kletsas D, Tsakalidou as potential drug carriers targeting the liver. Acta Biomater 7
E (2006) Rapid assessment of the physiological status of Strepto- (4):1665–1673
coccus macedonicus by flow cytometry and fluorescence probes. Int Zhu J, Thakker C, San KY, Bennet G (2011) Effect of culture operating
J Food Microbiol 111(3):197–205 conditions on succinate production in a multiphase fed-batch
Papadimitriou K, Pratsinis H, Neve-von-Caron G, Kletsas D, Tsakalidou bioreactor using an engineered Escherichia coli strain. Appl
E (2007) Acid tolerance of Streptococcus macedonicus as assessed Microbiol Biot 92(3):499–508
Resultados 155
4.6. Influencia del pH en la producción de ácido lactobiónico
En este trabajo se ha buscado descifrar la influencia del pH operacional sobre la

viabilidad celular de P. taetrolens y cómo la heterogeneidad fisiológica resultante está
directamente imbricada en la producción de ácido lactobiónico. Se han empleado estrategias
de control de pH no descritas hasta la fecha (estrategia pH-shift a diferentes valores,
diferentes valores de pH inicial e incluso valores de pH estrictamente controlado) con el
objetivo de evaluar la influencia de las mismas sobre la producción de ácido lactobiónico a
partir de suero lácteo.
Tradicionalmente, se ha venido enfocando el estudio de todo bioproceso obviando la
influencia del pH microambiental sobre la heterogeneidad fisiológica del sistema. En este
trabajo, se ha optimizado un sistema fermentativo por primera vez en base a conocimientos
fisiológicos aportados por la citometría de flujo multi-paramétrica. Así, una vez establecidas
las condiciones óptimas de aireación/agitación y la edad del inóculo (véanse anteriores
subapartados), se ha procedido a optimizar el valor óptimo de pH inicial, así como la
estrategia de control de pH más adecuada en términos productivos o en base a
consideraciones fisiológicas de P. taetrolens. Los resultados han revelado la indudable
influencia tanto del pH inicial como de la estrategia de control de pH adoptada. Se ha
perseguido en definitiva combinar ambos conocimientos con el fin de proponer la
metodología más adecuada desde ambas perspectivas, llegando a señalar a un pH inicial de
6,5 y una estrategia de control estricto de pH (estrategia pH-stat) o bien de control en fase
de producción (estrategia pH-shift) en 6,5 como las estrategias óptimas desde el punto de
vista productivo.
Artículo: Selection method of pH conditions to establish Pseudomonas taetrolens

physiological states and lactobionic acid production.
Situación: Publicación aceptada en Applied Microbiology and Biotechnology (Springer).
Appl Microbiol Biotechnol
DOI 10.1007/s00253-012-4607-x
BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING
Selection method of pH conditions to establish Pseudomonas

taetrolens physiological states and lactobionic acid production
Saúl Alonso & Manuel Rendueles & Mario Díaz
Received: 15 October 2012 / Revised: 18 November 2012 / Accepted: 20 November 2012

# Springer-Verlag Berlin Heidelberg 2012
Abstract Microbial physiological responses resulting from performance of P. taetrolens, providing key information
inappropriate bioprocessing conditions may have a marked for bio-production of lactobionic acid on an industrial scale.
impact on process performance within any fermentation
system. The influence of different pH-control strategies on Keywords Flow cytometry . pH-control strategy .
physiological status, microbial growth and lactobionic acid Lactobionic acid . Bioprocessing conditions . Pseudomonas
production from whey by Pseudomonas taetrolens during taetrolens . Physiological status
bioreactor cultivations has been investigated for the first
time in this work. Both cellular behaviour and bioconver-
sion efficiency from P. taetrolens were found to be nega-
tively influenced by pH-control modes carried out at values Introduction
lower than 6.0 and higher than 7.0. Production schemes
were also influenced by the operational pH employed, with Microbial stress responses can frequently be found associ-
asynchronous production from damaged and metabolically ated with inappropriate bioprocessing conditions in many
active subpopulations at pH values lower than 6.0. industrially relevant bioprocesses. Microorganisms con-
Moreover, P. taetrolens showed reduced cellular prolifera- stantly face environmental fluctuations which may induce
tion and a subsequent delay in the onset of the production the development of complex phenotypes for an enhanced
phase under acidic conditions (pH<6.0). Unlike cultivations tolerance against potentially adverse conditions (Hewitt et
performed at 6.5, both pH-shift and pH-stat cultivation al. 2007; Nicolau et al. 2010; Onyeaka et al. 2003). As a
strategies performed at pH values lower than 6.0 resulted result, clonal bacterial populations may develop a stress-
in decreased lactobionic acid production. Whereas the cel- induced physiological response which can lead to a hetero-
lular response showed a stress-induced physiological re- geneous culture comprising cells with different physiologi-
sponse under acidic conditions, healthy functional cells cal states (Lidstrom and Konopka 2010; Müller et al. 2010).
were predominant at medium operational pH values (6.5– Such knowledge regarding cell-to-cell heterogeneity within
7.0). P. taetrolens thus displayed a robust physiological a bioprocess can be suitably addressed with multi-parameter
status at initial pH value of 6.5, resulting in an enhanced flow cytometry (Díaz et al. 2010), which enables the detec-
bioconversion yield and lactobionic acid productivity (7- tion and monitoring of changes in cellular constituents and
and 4-fold higher compared to those attained at initial pH properties or the overall physiological status of microorgan-
values of 4.5 and 5.0, respectively). These results have isms otherwise unidentified by conventional culture-based
shown that pH-control modes strongly affected both the approaches (Hewitt and Nebe-von-Caron 2001, 2004). Flow
physiological response of cells and the biological cytometry has therefore become an outstanding tool for
bioprocess monitoring and optimization in view of the fact
S. Alonso : M. Rendueles : M. Díaz (*) that it can provide valuable guidelines on the most suitable
Department of Chemical Engineering and Environmental bioprocessing conditions (David et al. 2011; Delvigne et al.
Technology, Faculty of Chemistry, University of Oviedo,
2011; Silva et al. 2011). Information on functional status
C/ Julián Clavería s/n,
33071 Oviedo, Spain and physiological robustness accordingly enables the suc-
e-mail: [email protected] cessful implementation of non-deleterious fermentation
strategies, essentially resulting in improved microbial fitness fermentation performance. The information obtained from
and biological performance (Lopes da Silva et al. 2009). cellular parameters such as membrane polarization, mem-
Undoubtedly, microenvironmental pH constitutes a criti- brane integrity and metabolic activity may thus enable the
cal bioprocessing parameter that exerts a major influence on implementation of novel process strategies aimed at pre-
microbial physiology. Inappropriate bioprocessing condi- venting any potential detrimental influence.
tions may thus involve the development of asynchronous
cell subpopulations comprising metabolite producing and
non-producing cells, which has a profound impact on fer- Materials and methods
mentation performance (Nicolau et al. 2010; Want et al.
2009). Although numerous approaches have focused on Microorganism
deciphering the influence of pH in submerged liquid fer-
mentations, the physiological response and segregation of P. taetrolens LMG 2336, obtained from the Belgian
microbial subpopulations due to inappropriate operational Coordinated Collection of Microorganisms (Gent, Belgium),
pH conditions have scarcely been evaluated so far. was maintained frozen (in 40 % [v/v] glycerol at −20 °C). This
Determining the deleterious impact of such bioprocessing strain was subsequently subcultured on Nutrient Broth (NB;
conditions could be of substantial importance for the devel- containing 1 g/l meat extract, 2 g/l yeast extract, 5 g/l peptone
opment of incipient, industrially relevant bioprocesses tar- and 5 g/l NaCl) agar plates, incubated for 48 h at 30 °C and
geting novel bio-products such as lactobionic acid. The then preserved at 4 °C.
cosmetics and pharmaceutical industries have recently
shown considerable interest in lactobionic acid as the key Inoculum preparation
active component of novel therapeutic formulations (Green
et al. 2009) or as a chemical platform for the synthesis of A loopful of P. taetrolens from a fresh NB agar plate was used
bioactive molecules (Ortial et al. 2006). Lactobionic acid is to inoculate a 500-ml Erlenmeyer flask containing 100 ml of
a high value-added polyhydroxy acid which additionally NB medium. This flask was incubated on an orbital shaker
presents a growing niche market with applications in the (Infors HT, model Flyer Aerotron, Switzerland) at 250 rpm and
fields of foodstuffs (Gerling 1998) and medicine (Southard 30 °C for 10 h. Actively growing cells from this culture were
and Belzer 1995; Peng et al. 2007). To date, lactobionic acid then employed as inoculum for the bioreactor seed cultures.
is industrially produced by chemical synthesis in an energy-
intensive process using costly metal catalysts (Kuusisto et Sweet whey preparation
al. 2007). However, bio-based lactobionic acid production
from an inexpensive feedstock such as cheese whey can be Sweet cheese whey (provided by ILAS S.A., Asturias,
efficiently accomplished through a biotechnological process Spain) was 1-fold diluted with distilled water (1:1) and
(Alonso et al. 2011, 2012a) in which both fermentation adjusted to pH 6.5 (by adding 6 M NaOH) prior to sterili-
performance and productivity are strongly reliant on the zation using a tangential microfiltration device equipped
physiological status displayed by Pseudomonas taetrolens with a 0.22 μm pore size PVDF membrane-cassette
(Alonso et al. 2012b). Consequently, deciphering the detri- (Millipore, Massachusetts, USA).
mental influence of the operational pH mode on the cellular
behaviour of P. taetrolens constitutes a key issue in order to Bioreactor cultivations
establish the optimal bioprocessing conditions for enhanced
fermentation performance. Identifying both the cellular be- Batch cultivations were performed in a 2-l bioreactor
haviour and underlying physiology involved constitutes an (BioFlo 110; New Brunswick Scientific, New Jersey,
essential step for the industrial development of any biopro- USA) with a working volume of 1 l. Bioreactor experiments
cess in which the microbial biomass would be otherwise were conducted at 30 °C with an inoculation level of 10 %
approached erroneously as a whole steady-state bulk param- (v/v) and an agitation rate of 350 rpm. Besides, the biore-
eter (Hewitt and Nebe-von-Caron 2004). actor was aerated (1.0 vvm) via a ring sparger located at the
Hence, the aim of the present study was to assess the bottom of the vessel. Excessive foam formation was pre-
impact of different pH-control strategies (uncontrolled-pH, vented by automatic addition of diluted (1:10) Y-30 emul-
pH-stat, pH-shift and initial pH) on lactobionic acid produc- sion (Sigma-Aldrich, Steinheim, Germany). The bioreactor
tion as well as on the physiological status of P. taetrolens. was equipped with a pH meter (Mettler Toledo,
To this end, the physiological heterogeneity of P. taetrolens Switzerland) and a polarographic dissolved oxygen elec-
during batch bioreactor cultivations was monitored through trode (InPro 6830; Mettler Toledo) for continuous online
multi-parameter flow cytometry in order to determine the monitoring of pH and dissolved oxygen tension (DOT)
deleterious influence of these bioprocessing conditions on values, respectively. The seed culture was obtained by
harvesting biomass (after centrifugation at 11,000×g for Staining procedures

10 min) from a NB shake-flask culture and re-suspending
cells in a 500-ml Erlenmeyer flask containing 100 ml sweet Samples from cultures were harvested by centrifugation at
whey. Subsequently, these cultures were grown at 250 rpm 16,000×g for 5 min. Before staining, cells were washed
in an orbital shaker at 30 °C for 12 h and then employed as twice in phosphate-buffered saline (PBS; pH 7.4, sterile
seed culture in the bioreactor at 10 % (v/v) containing sweet and filtered at 0.22 μm) and were then held in the “hot spot”
whey. These prior conditions were applied to all cultivations of a sonication bath for 2 s to prevent bacterial aggregation
unless otherwise specified. Cultivations were carried out in before flow cytometric analysis (Hewitt and Nebe-von-
duplicate as independent experiments. Caron 2004). Propidum iodide (PI; Invitrogen), bis-(1,3-
dibutylbarbituric acid) trimethine oxonol (bis-oxonol,
Uncontrolled-pH bioreactor cultivation DiBAC4(3); Invitrogen) and carboxyfluorescein diacetate
(cFDA; Invitrogen) were used as fluorescent dyes in a
In order to gain insight into how the system and bio- double dual-staining procedure (DiBAC4(3)/PI and cFDA/
process parameters are affected by the lack of pH- PI) in order to evaluate cell physiological status (metabolic
control, bioreactor cultivation was performed under activity, membrane integrity and membrane polarization
uncontrolled-pH conditions. The pH was accordingly left were evaluated through cFDA, PI and DiBAC4(3) staining,
to vary freely after adjusting the initial pH to 6.5 (by respectively). Staining procedures were performed as previ-
adding 3 M NaOH). ously reported by Alonso et al. (2012b).
Multi-parameter flow cytometry

Influence of initial pH on bioreactor cultivations
Flow cytometry measurements were performed using a
The influence of different initial pH values (4.5, 5.0, 6.0,
Cytomics FC 500 flow cytometer (Beckman Coulter)
6.5, 7.0 and 7.5) on bioprocess parameters combined with a
equipped with a 488 and 633 nm excitation light source
pH-shift control strategy was evaluated. A two-stage pH-
from an argon ion laser. Green fluorescence from samples
shift bioconversion strategy was thus adopted in bioreactor
(corresponding to DiBAC4(3) and cFDA-stained cells) was
cultivations as previously detailed (Alonso et al. 2011).
collected on the FL1 channel (530 nm), whereas PI fluores-
After adjusting the initial pH to its corresponding value,
cence was registered on the FL3 channel (610 nm). Each
the pH was left uncontrolled during the growth phase and
analysis was performed in duplicate at a low flow rate
subsequently maintained at 6.5 during the production phase
setting (4,000 events/s). Data acquisition was carried out
by means of computer-controlled peristaltic pumps via au-
using Cytomics RXP software (Beckman Coulter). Gates
tomatic addition of 3 M NaOH.
and quadrants were established according to staining con-
trols. For DiBAC4(3)/PI and cFDA/PI dual-parameter flow
Bioreactor cultivations under different pH-shift control cytometric analysis, data collected from 150,000 and
values 100,000 events, respectively, were analyzed using Summit
v4.3 software (DakoCytomation, Colorado, USA).
The aforementioned pH-control strategy was adopted at
different pH-shift values after the onset of the produc- Analytical methods
tion phase. The pH was therefore left uncontrolled
above 6.5 during the growth phase and then maintained Bacterial growth was measured spectrophotometrically as
at different pH values (4.5, 5.0, 6.0, 6.5, 7.0 and 7.5) the optical density at 600 nm (Shimazdu, UV 1203 model)
during the production phase through automatic addition after centrifugation of culture samples at 16,000× g for
of 3 M NaOH. 5 min. Optical density data was converted to cell dry weight
(expressed in grams per litre) using the corresponding pre-
Bioreactor cultivations under different pH-stat conditions viously obtained calibration curve. The lactobionic acid and
lactose content of cell-free culture samples was measured by
In order to examine the influence of different pH-stat values high performance liquid chromatography as reported previ-
on the bioprocess parameters, bioreactor cultivations were ously by Alonso et al. (2011).
carried out with continuous pH control at 4.5, 5.0, 6.0, 6.5,
7.0 and 7.5 through automatic addition of either 3 M NaOH Determination of the oxygen uptake rate
or 3 M H2SO4. By applying this bioconversion strategy, the
culture broth basicity associated with the cellular growth The oxygen uptake rate (OUR) from bioreactor cultivations
phase is avoided (Alonso et al. 2011). was determined via the dynamic method as previously
reported (Alonso et al. 2012a). The specific consumption onwards), suggesting that the onset of lactobionic acid syn-
rate of oxygen (qO2) was calculated by dividing the OUR thesis was intimately linked to changes in the cytoplasmic
value by the biomass concentration at each sampling point. membrane polarization of P. taetrolens cells.
Effect of initial pH on bioprocess parameters

Results
Figure 2 shows the impact of initial pH values on bioprocess
Uncontrolled-pH bioreactor cultivation parameters during bioreactor cultivations performed with a
pH-shift strategy at 6.5. This two-stage pH-control strategy
Figure 1 shows the influence of uncontrolled-pH conditions was devised to avoid the negative influence exerted by the
on bioprocess parameters during bioreactor fermentation. As uncontrolled-pH conditions on lactose oxidative bioconver-
can be seen, the onset of the stationary growth phase featured sion (Fig. 1a). Whereas maximum biomass values were
the return of the DOT to values near to the air saturation level clearly impaired by initial pH values lower than 6.0, cell
concomitant to an initial sharp decrease (from 10 to 24 h) and growth was stimulated under cultivations performed at initial
subsequent progressive decay in pH (from 24 h onwards). The pH higher than 7.0 (Fig. 2a). Moreover, initial pH values
pattern of lactose oxidative bioconversion from whey by P. ranging from 6.0 to 6.5 contributed to enhanced cellular
taetrolens cells showed the typical non-growth associated proliferation, resulting in specific growth rates of 0.18 and
production, as previously reported (Alonso et al. 2011). 0.19 h−1 for an initial pH of 6.0 and 6.5, respectively
However, the synthesis of lactobionic acid ceased after 34 h (Table 1). As shown in Fig. 2a, cultivation at an initial pH
with a production rate of 0.35 g/lh, leaving 14 g/l of residual of 4.5 showed a lag phase of 8 h, thereby delaying the onset
lactose (Fig. 1a). Despite the fully aerobic conditions, cells of the production phase up to 34 h. The lag phase suggests
exhibited a low specific oxygen consumption rate (around poor physiological robustness and further biological perfor-
2 mg/gh) throughout the lactobionic acid formation phase mance since its presence is a signal of the lack of an adaptive
which also featured the predominance of damaged cells. response from P. taetrolens cells. The DOT profiles (Fig. 2b)
Thus, P. taetrolens cells were predominantly (>90 %) healthy showed that P. taetrolens cells were barely consuming oxy-
and metabolically active during the growth phase, whereas gen under initial pH values of 4.5 and 5.0, since DOT values
damaged cells constituted the main cell subpopulation were near the saturation level. This was consistent with the
(>90 %) from 34 h onwards (Fig. 1b). Additionally, the oxygen consumption rate from P. taetrolens cells under these
transition from the growth to the stationary phase featured circumstances, given that the OUR values decreased pro-
the prevalence of depolarized cells (>85 % from 24 h gressively as fermentation proceeded (Fig. 2d).
Fig. 1 Time-course profile of
bioprocess parameters obtained
during uncontrolled-pH biore-
actor cultivation of P. taetrolens
(a). Percentages of P. taetrolens
cell subpopulations throughout
bioreactor fermentation (b)
Fig. 2 Time-course profiles of

cell growth (a), DOT (b), pH a b
1.5 100
(c), OUR (d), lactose (e)

and lactobionic acid (f) during 80
pH-shift bioreactor cultivations
DOT (%)
1
at different initial pH values 60
4.5
5.0
40 6.0
0.5 6.5
4.5 5.0 7.0
20 7.5
6.0 6.5
7.0 7.5
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
c 9.5
d 100
8.5 80
OUR (mg/L.h)
4.5
7.5
60 5.0
pH
4.5
6.5 6.0
5.0
40 6.5
6.0
5.5 6.5 7.0
7.0 7.5
20
4.5 7.5
3.5 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
e 50
f 50
4.5
40 40 5.0
6.0
Lactose (g/L)
6.5
30 30 7.0
7.5
4.5
20 5.0 20
6.0
10 6.5 10
7.0
7.5
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
Table 1 Comparison of values obtained in bioreactor cultivations under the influence of different initial pH values
Initial pH Maximum dry Duration of the Onset of the pH-shift value μ (h−1) Volumetric Yield (%)a
cell weight (g/l) lag phase (h) production phase (h) productivity (g/lh)
4.5 0.55±0.04 8 34 – 0.02 0.1±0.00 14±1

5.0 0.55±0.02 2 28 – 0.10 0.18±0.02 25±1
6.0 1.24±0.03 0 12 8.09±0.08 0.18 0.42±0.04 58±2
6.5 1.25±0.04 0 10 8.00±0.05 0.19 0.70±0.00 100±0
7.0 1.40±0.03 0 34 8.47±0.03 0.15 0.22±0.02 31±3
7.5 1.33±0.06 0 48 9.22±0.02 0.13 0.15±0.02 20±2
– absence of a pH-shift value; μ0specific growth rate

a
Yield was defined as the percentage of lactose converted into lactobionic acid after 58 h
Lactobionic acid production schemes were highly influ- biological performance (Fig. 3a). This progressive increase
enced by the initial pH value, as shown in Fig. 2f. Initial pH in the damaged subpopulation thus involved asynchronous
values lower than 5.5 induced the onset of lactobionic acid lactobionic acid production comprising producing and non-
formation without the end of the cellular growth phase, producing cells. In contrast, cells were predominantly
whereas initial pH values above 5.5 induced the simulta- healthy and metabolically active (>90 %) throughout the
neous formation of lactobionic acid with cellular prolifera- cultivation at an initial pH of 6.5 and 7.5, although the onset
tion. As a result, both fermentation efficiency and of the lactobionic acid production phase involved a depo-
lactobionic acid yield were clearly reduced under pH values larization of the cellular membrane (Fig. 3c and d).
higher than 7.0, resulting in a 4-fold lower yield (25 %) at an
initial pH of 5.0 compared to that obtained at 6.5 (100 % Influence of pH-shift bioconversion values
yield). In terms of volumetric productivity, bioreactor culti-
vation at an initial pH of 6.5 (0.7 g/lh) resulted in a 7- and In order to elucidate the role of pH-shift control on lacto-
4.7-fold higher value compared to that achieved at 4.5 bionic acid production, bioreactor cultivations were carried
(0.1 g/lh) and 7.5 (0.15 g/lh), respectively. The presence out under a pH-shift bioconversion strategy (Fig. 4).
of lag phases at initial pH values lower than 6.0 (being 8 and Obviously, cellular proliferation patterns were similar dur-
2 h for 4.5 and 5.0, respectively) clearly delayed the onset of ing the growth phase (Fig. 4a), although cultivations per-
the production phase, which was prolonged up to 34 and formed under acidic conditions (pH-shift 04.5, 5.0)
28 h at initial pH values of 4.5 and 5.0, respectively supported higher cell densities during the stationary growth
(Table 1). Regardless of the absence of lag phases at initial phase compared to those performed at higher pH-shift val-
pH values higher than 6.5, the delay in the onset of the ues. Upon reaching their maximum biomass at 10 h, the
lactobionic acid production phase was longer than at acidic transition from the growth to the stationary growth phase
values (i.e., 4.5 or 5.0) (Fig. 2e and f), which was indicative was accompanied by an abrupt increase in DOT values
of overflow metabolism under these circumstances. (Fig. 4b). As regards microbial oxygen requirements and
Regarding physiological status, reduced initial pH values consumption, all OUR patterns peaked and coincided at 8 h
severely affected the physiology of P. taetrolens cells, as during the growth phase, as shown in Fig. 4d. The onset of
indicated by a decrease in the healthy, metabolically active the production phase featured a gradual decrease in the
subpopulation (Fig. 3). At an initial pH of 4.5, metabolic OUR, suggesting that the maximum catabolic activity of
activity decreased with two distinct subpopulations emerg- the cells was reached during the growth phase, subsequently
ing (damaged and dead), thereby leading to decreased remaining constant in the late stage of fermentation.

cell subpopulations during a b
pH-shift bioreactor cultivations
performed at an initial pH pH0 = 5.0
of 4.5 (a), 5.0 (b), 6.5
(c) or 7.5 (d) pH0 = 4.5
c d
pH0 = 6.5 pH0 = 7.5


cell growth (a), DOT (b), pH
a 1.5
b 100
(c), OUR (d), lactose (e) and

lactobionic acid (f) during 80
bioreactor cultivations carried
1
DOT (%)
out at a pH-shift bioconversion 60
value of 4.5, 5.0, 6.0, 6.5, 7.0 or 4.5 4.5
5.0 5.0
7.5 40 6.0
0.5 6.0
6.5 6.5
20 7.0
7.0
7.5
7.5
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
c 8.5
d
100
4.5 5.0
8 6.0 6.5
7.0 7.5 80
7.5
OUR (mg/L.h)
4.5
7 60 5.0
pH
6.5 6.0
40 6.5
6
7.0
5.5 7.5
20
5
4.5 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
e 50
f 50
4.5
40 40 5.0
Lactose (g/L)
6.0
6.5
30 30 7.0
4.5 7.5
20 5.0 20
6.0
6.5
10 7.0
10
7.5
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
As can be seen in Fig. 4f, lactobionic acid decreased phase (Fig. 5c and d). Upon shifting to the production phase,
markedly not only with the increase in pH-shift values no significant changes were found in terms of metabolic
higher than 7.0, but also with values lower than 6.0. In fact, activity (around 92 % of the cells remaining active). In
complete lactose bioconversion was achieved after 58 h at a contrast, two phases were clearly observed at cultivations
pH-shift value of 6.5 compared to a 65 % process yield performed with a pH-shift strategy at 4.5 and 5.0: an active
obtained at a pH-shift value of 4.5 (Table 2). Accordingly, proliferation phase, in which cells were metabolically active
the optimal pH-shift value for complete lactose bioconver- and polarized, followed by the onset of the production
sion ranged from 6.5 to 7.0, as can be seen in Fig. 4e. phase, in which a marked decrease in the polarization level
In accordance with the data shown in Fig. 5, both culti- was detected (Fig. 5a and b). As these cultivations have
vations under a pH-shift bioconversion at 6.5 and 7.5 dis- revealed, metabolically active cells have been shown to be
played a robust physiological status throughout the course more effective towards lactobionic acid production com-
of the fermentation. Cells exhibited high metabolic activity pared to cells from pH-shift bioconversion strategies per-
and retained their membrane integrity even at late stages of formed at 4.5 and 5.0, in which asynchronous production
cultivation. Analysis of the physiological status of the cells arose due to the presence of damaged cells (Fig. 5a and b).
revealed that 92 % of these cells remained healthy, meta- Therefore, responses of P. taetrolens cells to an acidic envi-
bolically active and polarized during the cell proliferation ronment during the lactobionic acid production phase
Table 2 Comparison of values

obtained in pH-shift bioreactor pH-shift Dry cell Onset of the Volumetric Yield
cultivations under different pH- strategy weight (g/l)a production phase (h) productivity (g/lh) (%)b
shift values
4.5 1.30±0.00 10 0.47±0.00 65±0
μ0specific growth rate 5.0 1.15±0.02 10 0.56±0.00 77±0
a
Maximum dry cell weight 6.0 1.11±0.08 10 0.65±0.04 90±2
reached at the production phase 6.5 1.13±0.04 10 0.70±0.00 100±0
b
Yield was defined as the percent- 7.0 1.04±0.05 10 0.68±0.02 94±2
age of lactose converted into lac- 7.5 0.95±0.06 10 0.59±0.02 82±2
tobionic acid after 58 h
suggested that healthy, metabolically active cells assumed a productivity (Table 3) compared to experiments carried out
damaged status, thereby displaying poor bioconversion per- at different initial pH values (Table 1).
formance. However, these pH-shift bioconversion condi- pH-stat cultivation strategies at medium pH values (6.0–
tions exerted lesser deleterious physiological effects than 6.5) contributed to enhanced cell growth of P. taetrolens, the
cultivation performed at different initial pH values, suggest- highest maximum biomass concentration (1.36 g/l) being
ing that changes in the microenvironmental pH were less obtained at 6.5. Results from specific growth rates also
detrimental during the lactobionic acid production phase. revealed that pH-stat strategies ranging from 6.0 to 6.5
stimulated cell growth performance. The specific growth
pH-stat cultivations at different values rate thus decreased on increasing the operational pH higher
than 6.5, being 0.19 and 0.17 h−1 for pH-stat strategies at 7.0
The potentially deleterious influence of inappropriate pH- and 7.5, respectively (Fig. 6a).
stat conditions on both bioconversion performance and bio- Dissolved oxygen profiles were strongly related to the
process parameters is shown in Fig. 6. As can be appreciated growth phase, a sharp depletion in the dissolved concentra-
in Figs. 6 and 7, both bioprocess performance and physio- tion being found at the initial stage of the cell growth phase
logical status were strongly dependent on the operational pH followed by a sudden rapid increase to saturated values
applied. The continuous pH-control strategy showed a pos- (Fig. 6b). However, it was observed that the dissolved
itive influence on lactobionic acid titer and volumetric oxygen profile remained unchanged at saturation values
Fig. 5 Percentages of cell

subpopulations during
a b
bioreactor cultivations carried
out at a pH-shift bioconversion
pH-shift = 4.5 pH-shift = 5.0
value of 4.5 (a), 5.0 (b),
6.5 (c) or 7.5 (d)
c d
pH-shift= 6.5
pH-shift = 7.5
Fig. 6 Influence of the pH-stat

strategy on cell growth (a),
a b
1.5 100
DOT (b), pH (c), OUR (d),

lactose (e) and lactobionic acid
80
(d) during bioreactor
DOT (%)
cultivations 1
60
4.5
5.0
40 6.0
0.5 6.5
7.0
4.5 5.0 20 7.5
6.0 6.5
7.0 7.5
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
c d
8.5 100
8
7.5 80
OUR (mg/L.h)
7 4.5
6.5 60 5.0
pH
6 6.0
5.5 40 6.5
7.0
5
7.5
4.5 20
4 4.5 5.0 6.0
6.5 7.0 7.5
3.5 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
e f
50 50
4.5
40 40 5.0
6.0
Lactose (g/L)
6.5
30 30 7.0
7.5
4.5
20 5.0 20
6.0
6.5
10 10
7.0
7.5
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time (h) Time (h)
throughout the cultivation performed under a pH-stat strat- impaired fermentation efficiency and therefore lactobionic
egy of 4.5, thereby illustrating poor cellular proliferation acid production.
under these conditions. Moreover, the temporal distribution During pH-stat bioreactor cultivations at 6.5 and 7.5, cells
of the maximum OUR values (with the exception of that remained predominantly healthy and functional throughout
from pH-stat cultivation at 4.5, which peaked at 28 h) took cultivation, as can be observed in Fig. 7c and d. In contrast,
place at 6–10 h, coinciding with the end of the growth phase cultivation under pH-stat conditions lower than 6.0 showed an
(Fig. 6d). As regards fermentation performance, both volu- increase in the damaged cell subpopulation, leading to a
metric productivities and bioprocess yields were severely stress-induced physiological response by P. taetrolens cells.
affected by the pH-stat cultivation strategy employed Therefore, when the pH was strictly controlled at 4.5 and 5.0,
(Fig. 6e and f), being 1.4- and 4-fold lower (0.50 and both fermentation performance and lactobionic acid produc-
0.18 g/lh for pH-stat strategies at 7.5 and 4.5) than those tivity were impaired due to the presence of non-producing
obtained under a pH-stat strategy at 6.5 (0.70 g/lh). These lactobionic acid cells (Fig. 7a and b).
results provide evidence that cultivations performed under Interestingly, results have shown that maximum lacto-
pH-stat strategies ranging from 4.5 to 6.0 and from 7.0 to bionic acid titers were obtained at medium pH values
7.5 exhibited a poor physiological status (Fig. 7) which (6.5–7.0), which differs substantially to the findings of

cell subpopulations during pH-
a b
stat bioreactor cultivations per- pH-stat = 4.5
formed at 4.5 (a), 5.0 (b), 6.5
(c) or 7.5 (d) pH-stat = 5.0
c d
pH-stat = 6.5 pH-stat = 7.5
Nishizuka and Hayaishi (1962), who established 5.6 as the fermentation system. Inappropriate bioprocessing conditions
optimal pH value for the highest enzymatic lactose oxidase could hence lead to major physiological responses with a
activity. In contrast to the present study, their experiments marked impact on fermentation performance (Onyeaka et al.
were performed under cell-free conditions, specifically with 2003; Want et al. 2009). In fact, stressful bioprocess conditions
the purified membrane-located dehydrogenase system from such as nutrient starvation, over-nutrient supply or hydrody-
P. taetrolens which mediates lactose oxidation and subse- namic conditions may induce a stress-induced physiological
quent hydrolysis by lactonase to yield lactobionic acid. response (Delvigne et al. 2011; Lopes da Silva et al. 2005; Silva
et al. 2011) in which the membrane integrity or the metabolic
activity of microorganisms is lost (Hewitt and Nebe-von-Caron
Discussion 2004). Unlike healthy functional cells, microorganisms with
compromised cytoplasmic membranes are not able to maintain
Environmental perturbations within bioprocesses often play or generate an electrochemical gradient and can hence neither
a major role in the cellular homeostasis and metabolic create membrane potential nor carry on metabolic activity
machinery of microorganisms, therefore affecting their over- (David et al. 2011; Díaz et al. 2010). The present study has
all physiology (Baatout et al. 2007; Nicolau et al. 2010). revealed for the first time how the cell growth, lactobionic acid
The functional status of microorganisms may thus be strong- production and fermentation efficiency of P. taetrolens have
ly affected by the bioprocessing conditions encountered in a been markedly affected by the pH-control strategy adopted.
Table 3 Comparison of values

obtained in bioreactor pH-stat Maximum Duration of Onset of the μ (h−1) Volumetric Yield
cultivations under different strategy dry cell the lag production productivity (%)b
pH-stat values weight (g/l) phase (h) phase (h) (g/lh)a
4.5 0.53±0.03 8 34 0.01 0.18±0.00 25±0

5.0 1.01±0.05 0 24 0.08 0.52±0.02 72±2
6.0 1.13±0.08 0 10 0.20 0.69±0.00 97±0
μ0specific growth rate 6.5 1.30±0.03 0 10 0.21 0.70±0.00 100±0
b
Yield was defined as the per- 7.0 1.17±0.04 0 10 0.19 0.64±0.02 88±2
centage of lactose converted into 7.5 0.98±0.06 0 24 0.17 0.51±0.02 70±2
lactobionic acid after 58 h
Specifically, this identified lack of biological performance from Furthermore, the initial pH value plays a key role in fermen-
P. taetrolens cells has been ascribed to major physiological tation performance, as can be seen in Fig. 2. Initial pH values
responses due to deleterious pH conditions. lower than 5.5 not only delayed the onset of the lactobionic acid
Despite the acidic conditions (pH 03.5) reached in production phase, but also inhibited the cellular proliferation
uncontrolled-pH bioreactor cultivation (Fig. 1), cells remained phase of P. taetrolens. Interestingly, the onset of lactobionic
damaged during the stationary growth phase, unlike in shake- acid production took place without reaching the end of the
flask fermentation, where prolonged exposure to an acidic as growth phase or obtaining meaningful cellular densities. In
well as an oxygen-limited environment led to a stress-induced contrast, pH-shift cultivations performed with an initial pH
physiological response by P. taetrolens cells (Alonso et al. value higher than 7.0 neither provided support for higher cell
2012b). Analysis of the pH-dependent physiological response densities in the stationary growth phase nor triggered lacto-
has provided essential information on the bioconversion effi- bionic acid production (Fig. 2a and f). The onset of the lacto-
ciency exhibited by P. taetrolens cells. Undoubtedly, micro- bionic acid phase was clearly inhibited at initial pH values
organisms submitted to stressful pH conditions may undergo higher than 7.0, suggesting that P. taetrolens cells responded
disruption of cellular homeostasis and functions, which can to these deleterious basic conditions with a longer length of the
concomitantly lead to major physiological changes (Baatout pH-shift (Fig. 2c). It was clear that the control of pH during
et al. 2007). Unlike fully functional cells, sub-lethally injured lactose bioconversion exerted a major influence on the physi-
and dead cells suffer from irreversible structural changes asso- ology of P. taetrolens, revealing that a pH of around 6.5 was
ciated with the collapse of cellular functions such as the proton optimal both for lactobionic acid production and physiological
motive force (Hewitt and Nebe-von-Caron 2004). In fact, status. pH values outside the range 6.0–7.0 thus displayed a
prolonged exposure to an acidic environment may involve the detrimental influence on fermentation efficiency, since these
development of a reduced microbial phenotype or a lower values involved the prevalence of sub-lethally injured or dead
degree of physiological robustness (Papadimitriou et al. cells during cultivation in contrast to healthy functional cells
2007), which has a profound impact on biological performance supported by pH-control strategies at 6.5.
and product metabolite yield. In conclusion, this study has revealed the physiological
Results have showed that optimization of the pH-control responses of P. taetrolens cells to different pH-control strate-
strategy led to enhanced lactobionic acid yields as well as gies, providing key information for identifying the main del-
improved fermentation performances (Tables 1, 2 and 3). eterious factors affecting bioconversion performance and
Undoubtedly, the poor biological performance from P. taetro- lactobionic acid production. Both lactobionic acid yield and
lens cells was attributable to the underlying physiological het- productivity were strongly affected by the pH-control strategy
erogeneity displayed by cultivations submitted to deleterious applied. This concise understanding of the functional and
microenvironmental pH conditions (Figs. 3, 5 and 7). Whereas physiological parameters involved has thus enabled the im-
pH-control strategies (both pH-shift and pH-stat control) at 6.5 plementation of suitable pH-control modes to prevent poten-
provided support for a robust physiological status, cultivations tially deleterious bioprocessing conditions for enhanced
performed at lower values (i.e., 4.5 or 5.0) exhibited a complex lactobionic acid production.
physiological heterogeneity (mainly comprised by damaged
and dead cells) that translated into reduced lactobionic acid
productivity (4-fold lower under pH-stat conditions at 4.5 Acknowledgements The authors are grateful for the financial sup-
port from the Spanish Ministry of Science and Innovation through
compared to that obtained at 6.5). These results demonstrate project MEC-CTQ2010-14918. The technical assistance of Ana Salas
that an appropriate pH-control strategy may enhance microbial and Magdalena Choda (Flow Cytometry Area, Scientific-Technical
culture performance and hence increase the lactobionic acid Services, University of Oviedo) is gratefully acknowledged.
titer. Lactobionic acid production phase may also be influenced
by the cellular phenotypic diversity on the expression of lactose
dehydrogenase system. However, experiments carried out at
different pH-shift values (Fig. 4) suggested that differences at References
this molecular level could play a minor role in the bioconver-
sion performance, assuming similar gene expression levels in
Alonso S, Rendueles M, Díaz M (2011) Efficient lactobionic acid
the cells submitted to the same pH conditions during growth production from whey by Pseudomonas taetrolens under pH-
phase. Undoubtedly, the induced-stress physiological shift conditions. Bioresour Technol 102(20):9730–9736
responses may impact on the bioconversion ability of P. tae- Alonso S, Rendueles M, Díaz M (2012a) Role of dissolved oxygen
trolens since the lactose dehydrogenase is a membrane-bound availability on lactobionic acid production from whey by
Pseudomonas taetrolens. Bioresour Technol 109:140–147
system which requires not only the membrane integrity but also Alonso S, Rendueles M, Díaz M (2012b) Physiological heterogeneity
cellular viability for cofactor regeneration (Nishizuka and of Pseudomonas taetrolens during lactobionic acid production.
Hayaishi 1962). Appl Microbiol Biotechnol 96(6):1465–1477
Baatout S, Leys N, Hendrickx L, Dams A, Mergeay M (2007) Lopes da Silva T, Piekova L, Mileu J, Roseiro JC (2009) A
Physiological changes induced in bacteria following pH stress as comparative study using the dual staining flow cytometric
a model for space research. Acta Astronaut 60(4–7):451–459 protocol applied to Lactobacillus rhamnosus and Bacillus
David F, Berger A, Hänsch R, Rohde M, Franco-Lara E (2011) Single licheniformis batch cultures. Enzyme Microb Technol 45
cell analysis applied to antibody fragment production with (2):134–138
Bacillus megaterium: development of advanced physiology and Müller S, Harms H, Bley T (2010) Origin and analysis of microbial
bioprocess state estimation tools. Microb Cell Factories 10:23 population heterogeneity in bioprocesses. Curr Opin Biotechnol
Delvigne F, Ingels S, Thonart P (2011) Evaluation of a set of E. coli 21(1):100–113
reporter strains as physiological tracer for estimation bioreactor Nicolau SA, Gaida SM, Papoutsakis ET (2010) A comparative view of
hydrodynamic efficiency. Process Biochem 45:1769–1778 metabolite and substrate stress and tolerance in microbial biopro-
Díaz M, Herrero M, García LA, Quirós C (2010) Application of flow cessing: from biofuels and chemicals, to biocatalysis and biore-
cytometry to industrial microbial bioprocesses. Biochem Eng J 48 mediation. Metab Eng 12(4):307–331
(3):385–407 Nishizuka Y, Hayaishi O (1962) Enzymic formation of lactobionic acid
Gerling KG (1998) Large scale production of lactobionic acid-use and from lactose. J Biol Chem 237:2721–2728
new applications. Int Dairy Fed 9804:251–256 Onyeaka H, Nienow AW, Hewitt CJ (2003) Further studies related to
Green BA, Yu RJ, Van Scott EJ (2009) Clinical and cosmeceutical uses the scale-up of high cell density Escherichia coli fed-batch fer-
of hydroxyacids. Clin Dermatol 27:495–501 mentations: the additional effect of a changing microenvironment
Hewitt CJ, Nebe-von-Caron G (2001) An industrial application of when using aqueous ammonia to control pH. Biotechnol Bioeng
multiparameter flow cytometry: assessment of cell physiological 84(4):474–484
state and its application to the study of microbial fermentations. Ortial S, Durand G, Poeggeler B, Polidori A, Pappolla MA, Böker J,
Cytometry 44(3):179–187 Hardeland R, Pucci B (2006) Fluorinated amphiphilic acid deriv-
Hewitt CJ, Nebe-von-Caron G (2004) The application of multi- atives as antioxidant carriers: a new class of protective agents. J
parameter flow cytometry to monitor individual microbial cell Med Chem 49:2812–2820
physiological state. Adv Biochem Eng Biotechnol 89:197–223 Papadimitriou K, Pratsinis H, Nebe-von-Caron G, Kletsas D,
Hewitt CJ, Onyeaka H, Lewis G, Taylor IW, Nienow AW (2007) A Tsakalidou E (2007) Acid tolerance of Streptococcus macedoni-
comparison of high cell density fed-batch fermentations involving cus as assessed by flow cytometry and single-cell sorting. Appl
both induced and non-induced recombinant Escherichia coli un- Environ Microbiol 73(2):465–476
der well-mixed small-scale and simulated poorly mixed large- Peng J, Wang K, Tan W, He X, He C, Wu P, Liu F (2007) Identification
scale conditions. Biotechnol Bioeng 96(3):495–505 of live liver cancer cells in a mixed cell system using galactose-
Kuusisto J, Tokarev AV, Murzina EV, Roslund MU, Mikkola JP, conjugated fluorescent nanoparticles. Talanta 71(2):833–840
Murzin DY, Salmi T (2007) From renewable raw materials to Silva F, Lourenço O, Maia C, Queiroz JA, Domingues FC (2011)
high value-added fine chemicals—catalytic hydrogenation and Impact of plasmid induction strategy on overall plasmid DNA
oxidation of D-lactose. Catal Today 121:92–99 yield and E. coli physiology using flow cytometry and real-time
Lidstrom ME, Konopka M (2010) The role of physiological heterogeneity PCR. Process Biochem 46:174–181
in microbial population behaviour. Nat Chem Biol 6(10):705–712 Southard JH, Belzer FO (1995) Organ preservation. Annu Rev Med
Lopes da Silva T, Reis A, Kent CA, Kosseva M, Roseiro JC, Hewitt CJ 46:235–247
(2005) Stress-induced physiological responses to starvation periods Want A, Thomas ORT, Kara B, Liddell J, Hewitt CJ (2009) Studies
as well as glucose and lactose pulses in Bacillus licheniformis CCMI related to antibody fragment (Fab) production in Escherichia coli
1034 continuous aerobic fermentation processes as measured by W3110 fed-batch fermentation processes using multiparameter
multi-parameter flow cytometry. Biochem Eng J 24(1):31–41 flow cytometry. Cytometry A 75:148–154
168 Resultados
Resultados 169
4.7. Estrategias de alimentación para la mejora en la producción de

ácido lactobiónico
y aplicaciones
En el presente subapartado se ha pretendido mejorar la producción de ácido lactobiónico
a partir de suero lácteo con P. taetrolens, empleando para ello estrategias sencillas de
alimentación en discontinuo. Se ha perseguido alcanzar un rendimiento de producción alto
(>90%) empleando técnicas simples de alimentación basadas en el incremento de la
disponibilidad de la lactosa en el medio (>100 g/L). Así, se ha evaluado la idoneidad de
diferentes estrategias como son la alimentación en contínuo, con multi-pulsos o bien una
estrategia de co-alimentacion conformada por un aporte constante de suero lácteo junto con
pulsos de una solución concentrada de lactosa. Igualmente, se ha evaluado el status
fisiológico de P. taetrolens vinculado a tales estrategias de alimentación, sugiriendo las
mejores condiciones con el fin de obtener unos niveles de productividad y de producción
relevantes a escala industrial. La concentración de ácido lactobiónico obtenida (180 g/L)
supera ampliamente los márgenes que Pollard y Woodley (2007) o Yang et al. (2007)
consideran como niveles base (80-100 g/L) para empezar a considerar a cualquier sistema
biocatalítico de producción industrial como alternativa factible a los sistemas basados en
catálisis química. Como se ha mencionado en el subapartado 2.3, es obvio mencionar que los
procesos fermentativos en general presentan valores de productividad inferiores a los
sistemas enzimáticos. Sin embargo, la complejidad asociada aguas arriba y el altísimo coste
de los sistemas enzimáticos disponibles en el mercado impide la total implementación
industrial de los mismos.
Asimismo, en este trabajo se realiza una comparación con todos los sistemas microbianos
existentes en la bibliografía, enfatizando las características más relevantes como son la
eficiencia, la sostenibilidad o el bajo coste del sustrato que conforman el sistema aquí
descrito en contraposición con el resto de sistemas fermentativos descritos hasta la fecha.
Artículo: Feeding strategies for enhanced lactobionic acid production from whey by
Pseudomonas taetrolens.
Situación: Publicación aceptada en Bioresource Technology (Elsevier).
170 Resultados
Feeding strategies for enhanced lactobionic acid production from

whey by Pseudomonas taetrolens
Saúl Alonso, Manuel Rendueles, Mario Díaz *
Department of Chemical Engineering and Environmental Technology, University of Oviedo, Faculty of
Chemistry, C/Julián Clavería s/n, 33071 Oviedo, Spain.
*Corresponding author. Tel.: +34 98 5103439; fax: +34 98 5103434.

e-mail address: [email protected] (M. Díaz)
Abstract
High-level production of lactobionic acid from whey by Pseudomonas taetrolens under fed-batch
fermentation was achieved in this study. Different feeding strategies were evaluated according to the
physiological status and fermentation performance of P. taetrolens. A lactobionic acid titer of 164 g/L was
obtained under co-feeding conditions affording specific and volumetric productivities of 1.4 g/g.h and
2.05 g/L.h, respectively. Flow cytometry assessment revealed that P. taetrolens cells exhibited a robust
physiological status, which makes them particularly well-suited for employing concentrated nutrient
solutions to further prolong the growth and production phases. Such detailed knowledge of the
physiological status has been revealed to be a key issue to further support the development of high-yield
lactobionic acid production processes under feeding strategies. The present study has demonstrated the
feasibility of P. taetrolens to achieve high-level bio-production of lactobionic acid from whey through fed-
batch cultivation, suggesting its major potential for industrial-scale implementation.
Keywords: lactobionic acid; fed-batch; feeding strategies; flow cytometry; Pseudomonas

taetrolens.
1. Introduction
Lactobionic acid has recently emerged as a high value-added polyhydroxy bionic acid
with a plethora of applications in the food (Gerling, 1998), cosmetics (Green et al., 2009) and
pharmaceutical industries (Southard and Belzer, 1995). As an emergent substance,
lactobionic acid is also gaining substantial relevance in tissue engineering (Park et al., 2003)
and nanomedicine (Peng et al., 2007; Selim et al., 2007) due to its unique biocompatible,
biodegradable, chelating, amphiphilic and antioxidant properties. Although lactobionic acid
is currently produced at an industrial scale by means of chemical catalysis, its production
through microbial fermentation may overcome drawbacks such as the use of polluting
catalysts (Kuusisto et al., 2007) or the generation of undesirable side-reaction products
associated with the chemical synthesis (Chia et al., 2008).
Unlike current chemically-based processes, bio-production of lactobionic acid through
microbial fermentation from an inexpensive feedstock such as cheese whey additionally
provides an environmentally-friendly and sustainable alternative (Alonso et al., 2011,
2012a). The use of this high-strength polluting waste as an inexpensive source undoubtedly
Resultados 171
opens up the path for cost-effective lactobionic acid manufacturing through microbial
fermentation. In this regard, lactobionic acid constitutes a non-growth-associated bio-
product resulting from lactose oxidation by Pseudomonas taetrolens (Alonso et al., 2011).
The deleterious influence of an over-supply of oxygen (Alonso et al., 2012a) as well as the
operational pH (Alonso et al., 2012c) have both been identified as key bioprocessing factors
affecting bioconversion performance. Likewise, both fermentation performance and
lactobionic acid productivity have been shown to be strongly reliant on the overall
physiological status displayed by P. taetrolens (Alonso et al., 2012b). In fact, physiological
heterogeneity within a fermentation system represents a critical parameter that reduces
product yield and productivity due to the presence of damaged and dead subpopulations
(Hewitt and Nebe-von-Caron, 2004; Müller et al., 2010). The presence of asynchronous cell
subpopulations comprising metabolite producing and non-producing cells may alter the
product formation pattern and consequently overall bioprocess efficiency (Want et al.,
2009). Furthermore, both imbalanced microenvironmental and bioprocessing conditions
may involve a stress-induced physiological response with cellular deterioration (Hewitt et
al., 1999; Lopes da Silva et al., 2011). Therefore, a precise understanding of the underlying
physiological responses to the microenvironmental limitations provides valuable
information for further successful implementation of non-deleterious fermentation
strategies (Lopes da Silva et al., 2009).
Fed-batch cultivation has become the standard operating mode adopted by industrial
biotechnology in order to achieve high-yield metabolite productivity (Ahn et al., 2000;
Koller et al., 2008; Yu et al., 2012). However, these cultivation approaches must take
physiological information into account, as microorganisms could be submitted to undesired
nutrient depletion or metabolic overflow, consequently leading to an irreversible stress-
induced physiological response (Amanullah et al., 2002; Hewitt et al., 1999; Want et al.,
2009). Such information can be thoroughly addressed by multi-parameter flow cytometry
(Díaz et al., 2010; Hewitt and Nebe-von-Caron, 2004), which enables the detection and
monitoring of changes in the physiological status of microorganisms. Physiological
robustness could thus be deciphered before industrial-scale implementation in order to
prevent any potential deleterious influence on the biological performance of microbial cells.
Undoubtedly, detailed knowledge about physiological status constitutes a key issue to
further support the development of bioprocesses targeting high-yield lactobionic acid
production.
The present study evaluates for the first time different feeding strategies, including co-
feeding, continuous and multi-pulse feeding conditions, for enhanced lactobionic acid
172 Resultados
production from whey by P. taetrolens. Fed-batch cultivations were accordingly carried out
in order to decipher and establish the most suitable operating approach for improved
fermentation performance. Furthermore, the underlying physiological heterogeneity of P.
taetrolens was also monitored through multi-parameter flow cytometry, thereby addressing
the potential deleterious influence of these novel feeding strategies on the physiological
status of these microorganisms. Overall, the results may lead to a deeper process
understanding for the bio-production of lactobionic acid from whey at an industrially
relevant titer.
2. Materials and methods

2.1. Microorganism
Pseudomonas taetrolens LMG 2336, obtained from the Belgian Coordinated Collection of
Microorganisms (Gent, Belgium), was maintained frozen (in 40 % [v/v] glycerol at -20ºC).
The strain was subsequently subcultured on Nutrient Broth (NB, containing 1 g/L meat
extract, 2 g/L yeast extract, 5 g/L peptone and 5 g/L NaCl) agar plates, incubated for 48 h at
30ºC, and then preserved at 4ºC.
2.2. Inoculum preparation
A loopful of P. taetrolens from a fresh Nutrient Broth agar plate was used to inoculate a
500 mL Erlenmeyer flask containing 100 mL of Nutrient Broth medium. This flask was
incubated on an orbital shaker (Infors HT, model Flyer Aerotron, Switzerland) at 250 rpm
and 30ºC for 10 h. Actively growing cells from this culture were then employed as inoculum
for the production of lactobionic acid in bioreactor cultures containing sweet whey, as
subsequently reported.
2.3. Sweet whey preparation
Diluted sweet whey was prepared by diluting sweet cheese whey (provided by ILAS S.A.,
Asturias, Spain) 1-fold with distilled water (1:1) and adjusted to pH 6.5 (by adding 6 M
NaOH) prior to sterilization using a tangential microfiltration device equipped with a 0.22
μm pore size PVDF membrane-cassette (Millipore, Massachusetts, USA). Concentrated
cheese whey was prepared as above by microfiltration of undiluted sweet cheese whey. A
highly concentrated whey solution was prepared as above by microfiltration of undiluted
sweet cheese whey supplemented with lactose (82 g/L), peptone (2.5 g/L) and yeast extract
(2.5 g/L).
2.4. Bioreactor cultivations
Bioreactor cultivations were performed in a 2-L bioreactor (BioFlo 110, New Brunswick
Scientific, NJ, USA) with an initial working volume of 1 L at 30ºC and an agitation rate of 350
Resultados 173
rpm. Besides, the bioreactor was aerated (1 Lpm) via a ring sparger located at the bottom of
the vessel. Excessive foam formation was prevented by automatic addition of diluted (1:10)
Y-30 emulsion (Sigma-Aldrich, Steinheim, Germany). The bioreactor was equipped with a
pH meter (Mettler Toledo, Switzerland) and a polarographic dissolved oxygen electrode
(InPro 6830, Mettler Toledo, Switzerland) for continuous on-line monitoring of pH and
dissolved oxygen tension (DOT) values, respectively. A two-stage pH-shifted bioconversion
strategy was adopted as previously reported (Alonso et al. 2011). The strategy consisted in
controlling pH above 6.5 (pH was left uncontrolled above this value during the growth phase
and then maintained at 6.5) by means of computer-controlled peristaltic pumps via
automatic addition of 4 M NaOH. These prior conditions were applied to all cultivations
unless otherwise specified. Cultivations were carried out in duplicate as independent
experiments.
Batch cultivation for lactobionic production from undiluted sweet whey was conducted
with an inoculation level of 30% (v/v). The seed culture was obtained by harvesting
biomass (after centrifugation at 11,000xg for 10 min) from a NB shake-flask culture and re-
suspending cells in a 500 mL Erlenmeyer flask containing 100 mL of diluted sweet whey.
These cultures were subsequently grown at 250 rpm in an orbital shaker at 30ºC for 12 h
and then employed as seed culture in the bioreactor at 30% (v/v) containing undiluted
sweet whey.
2.5. Influence of feeding rates on fermentation performance
The influence of different feeding rates (4.2, 8.4, 16.8, 33.6 and 67.2 mL/h) on bioprocess
parameters in combination with a pH-shift control strategy was evaluated. A total volume of
0.3 L of highly concentrated whey solution were thus continuously added at different flow
rates into bioreactor containing diluted sweet whey. Fed-batch cultivations were conducted
with an inoculation level of 10% (v/v), in which the DOT was maintained at 10% via an
agitation cascade (from 350 to 500 rpm) during the growth phase.
2.6. Influence of multi-pulse feeding conditions
In order to gain insight into how system and bioprocess parameters are affected by the
availability of different lactose levels, cultivation was submitted to multi-pulse feedings of
highly concentrated lactose solution (275 g/L). Pulses of 0.1 L were performed at either 10,
24 or 34 h. Fed-batch cultivation was conducted at an inoculation level of 30% (v/v),
maintaining the DOT level above 10% via an agitation cascade (from 350 to 500 rpm).
2.7. Fed-batch cultivations under a co-feeding strategy
Fed-batch cultivations were carried out under a co-feeding strategy which consisted in
combining multi-pulse feeding with lactose (every pulse of 0.1 L contained 275 g/L of
174 Resultados
lactose) with continuous feeding of highly concentrated whey (a total volume of 0.3 L
containing 160 g/L of lactose, 2.5 g/L of yeast extract and 2.5 g/L of peptone) at rates of 4.2
and 33.6 mL/h. The pulse feeding of the high-lactose solution (275 g/L) was carried out at
regular intervals (10, 24 and 34 h). Fed-batch cultivations were conducted at an inoculation
level of 30% (v/v), maintaining the DOT level above 10% via an agitation cascade (from 350
to 500 rpm).
2.8. Staining procedures and multi-parameter flow-cytometry
Samples from cultures were harvested by centrifugation at 16,000xg for 5 min. Before
staining, cells were washed twice in phosphate-buffered saline (PBS, pH 7.4, sterile and
filtered at 0.22 μm) and were then held in the “hot spot” of a sonication bath for 2 s to
prevent bacterial aggregation before flow cytometric analysis (Hewitt and Nebe-von-Caron,
2004). Propidium iodide (PI, Invitrogen), bis-(1,3-dibutylbarbituric acid) trimethine oxonol
(bis-oxonol, DiBAC4(3), Invitrogen) and carboxyfluorescein diacetate (cFDA, Invitrogen)
were used as fluorescent dyes in a double dual-staining procedure (DiBAC4(3)/PI and
cFDA/PI) to evaluate cell physiological status (metabolic activity, membrane integrity and
membrane polarization were evaluated through cFDA, PI and DiBAC4(3) staining,
respectively). Staining solutions were prepared and procedures were conducted as
previously reported by Alonso et al. (2012b).
Flow cytometry measurements were performed using a Cytomics FC 500 flow cytometer
(Beckman Coulter) equipped with a 488 and 633 nm excitation light source from an argon
ion laser. Green fluorescence from samples (corresponding to DiBAC4(3) and cFDA-stained
cells) was collected on the FL1 channel (530 nm), whereas PI fluorescence was registered
on the FL3 channel (610 nm). Each analysis was performed in duplicate at a low flow rate
setting (4000 events/s). Data acquisition was carried out using Cytomics RXP software
(Beckman Coulter). Gates and quadrants were established according to staining controls.
For DiBAC4(3)/PI and cFDA/PI dual-parameter flow cytometric analysis, data respectively
collected from 150,000 and 100,000 events were analyzed using Summit v4.3 software
(DakoCytomation, Colorado, USA).
2.9. Analytical methods
Bacterial growth was measured spectrophotometrically as optical density at 600 nm
(Shimazdu, UV 1203 model) after centrifugation of culture samples at 16,000xg for 5 min.
Optical density data was converted to cell dry weight (expressed in grams per litre) using
the corresponding previously obtained calibration curve. The lactobionic acid and lactose
content of cell-free culture samples were measured by high performance liquid
chromatography as reported previously by Pedruzzi et al. (2007). The liquid
Resultados 175
chromatography system used for analysis (Agilent 1200, Agilent Technologies Inc.,
California, USA) was equipped with an ICSep ICE-ION-300 column (Transgenomic Inc.,
California, USA) coupled to a refractive index detector. Sulphuric acid (0.450 mM, pH 3.1)
was employed as the mobile phase at a flow rate of 0.3 mL/min with the column
temperature set at 75ºC. Data acquisition and analysis were performed with ChemStation
software (Agilent).
2.10. Determination of the oxygen uptake rate
The oxygen uptake rate (OUR) from bioreactor cultivations was determined via the
dynamic method. The OUR was accordingly obtained from the slope of the plot of dissolved
oxygen concentration versus time following a momentary interruption of air supply to the
bioreactor (Alonso et al. 2012a). The specific consumption rate of oxygen (qO2) was
calculated by dividing the OUR value by the biomass concentration at each sampling point.
3. Results and discussion

3.1. pH-shift batch cultivation on concentrated whey
As can be seen in Fig. 1A, a lactobionic acid titer of 78.4 g/L with a volumetric
productivity of 1.63 g/L.h was obtained under batch cultivation using concentrated whey.
These levels respectively represent 1.9- and 2.3-fold enhancement compared to the titer and
productivity values obtained under batch cultivation with diluted whey medium (Alonso et
al., 2011), subsequently achieving complete lactose bioconversion (100%) after 48 h (Fig.
1A). The onset of the stationary growth phase featured the usual increase in DOT values
concomitantly to an abrupt drop in pH of the culture broth (Alonso et al., 2011).
Interestingly, P. taetrolens cells displayed a specific productivity value of 1.02 g/gDCW.h,
which is significantly higher than that obtained in diluted whey media under high-cell
density conditions, namely 0.92 g/gDCW.h (Alonso et al., 2012a). Upon the onset of the
lactobionic acid production phase, a return of dissolved oxygen to saturated values was
observed along with a decreased specific oxygen uptake rate at a late stage (~10
mg/gDCW.h).
Figs. 1B and 1C show the changes in the physiological status of P. taetrolens in terms of
metabolic activity, membrane integrity and polarization during the pH-shift batch
cultivation on concentrated whey. As previously reported, the physiological heterogeneity of
P. taetrolens plays a crucial role in lactobionic acid bio-production (Alonso et al., 2012b).
The presence of both an intact and polarized cytoplasmic membrane and functional nucleic
acids undoubtedly represent the key parameters of cellular viability (Hewitt and Nebe-von-
Caron, 2004). As can be seen in Figs. 1B and C, P. taetrolens cells were not induced towards a
176 Resultados
stress physiological response during the lactobionic acid production phase in batch
bioreactor cultivation, remaining predominantly metabolically active (~65%). This result
contrasts with shake-flask cultivation, in which prolonged exposure to stressful
bioprocessing conditions led to a loss in cellular functionality (Alonso et al., 2012b).
B
0h 4h 10 h 24 h 48 h
DiBAC4(3)
PI
C
0h 4h 10 h 24 h 48 h
cFDA
PI
Figure 1. Bioprocess parameters (A), dot plots representing DiBAC4(3) fluorescence versus PI fluorescence (B)
and cFDA fluorescence versus PI fluorescence signals (C) obtained during pH-shift bioreactor cultivation of P.
taetrolens on concentrated whey. The upper left quadrant of the dot plots in B shows depolarized cells
(DiBAC4(3)-positive; DiBAC4(3) enters cells only when the membrane potential is lost), while cells with esterase
activity (cFDA-positive cells; cFDA is cleaved by the esterase activity inside the living cells, thus releasing a polar
fluorescent portion which is unable to pass through the intact membrane) are shown in the upper left quadrant
of C. Damaged cells (corresponding to cFDA(+)/PI(+) cells or DiBAC4(3)(+)/PI(-) cells) are shown in the upper left
quadrant of B and in the upper right quadrant of C for the DiBAC4(3)/PI and cFDA/PI staining, respectively.
Likewise, dead cells (PI-positive cells; PI can only cross the plasmatic membrane if it is permeabilized
corresponding to those cells whose membranes are compromised) are shown in the upper right quadrant and
in the lower right quadrant for the DiBAC4(3)/PI (B) and cFDA/PI staining (C), respectively.
Resultados 177
Unlike healthy functional cells, damaged and dead cells suffer structural changes
associated with the collapse of cellular functions, including the proton motive force (Díaz et
al., 2010; Müller et al., 2010). In fact, stress-induced physiological responses due to
inappropriate bioprocessing conditions may directly result in reduced fermentation
efficiency and biological performance (Lopes da Silva et al., 2009). While the growth phase
feature a high percentage of metabolic activity (>90%), the transient phase between the
growth and production periods featured a loss in membrane polarization, as shown in Fig.
1B, remaining mainly depolarized (~70%) during the stationary phase. Moreover, flow
cytometric assessment revealed the presence of two distinct cell subpopulations, as the dot
plots in Fig. B show, although this fact did not lead to asynchronous lactobionic acid
production.
3.2. Impact of the feeding rate on lactobionic acid production

Fig. 2 shows the influence of different constant feeding rates on bioprocess parameters
aimed at improving the lactobionic acid titer and yield. Feeding was completed during the
exponential growth phase at the highest rate (67.2 mL/h for 4.5 h) in contrast with the
continuous feeding strategy adopted under the lowest rate (4.2 mL/h for 72 h). Table 1
summarizes the overall cultivation parameters obtained with the five different feeding rates
employed. As can be seen, higher feeding rates stimulated the growth of P. taetrolens cells,
resulting in improved proliferation patterns (Fig. 2A). The highest maximum biomass
concentration (2.15 g/L) and specific growth rate (0.16) were obtained under the influence
of the highest rates (33.6-67.2 mL/h) (Table 1), suggesting that increasing feeding rates
contributed not only to higher biomass, but also to higher specific growth rates. Dissolved
oxygen profiles showed similar trends in the fed-batch cultivations (Fig. 2B), in which DOT
levels were maintained at 10% during the growth phase by means of an agitation cascade
(from 350 to 500 rpm), as oxygen-limited conditions during this phase improve the
fermentation performance of P. taetrolens cells (Alonso et al., 2012a). pH-shift values were
correlated with feeding rates (Fig. 3), with increasing rates leading to a progressive rise in
pH-shift values (7.52, 7.64, 7.74, 7.90 and 7.92 for 4.2, 8.4, 16.8, 33.6 and 67.2 mL/h,
respectively) (Fig. 2C). Increased feeding rates reduced oxygen consumption by P. taetrolens
cells, as indicated by a decrease in OUR values during the lactobionic acid production phase
(Fig. 2D). Whereas lower OUR values (~7 mg/L.h) were found at higher rates (16.8-67.2
mL/h), a feeding rate of 4.2 mL/h supported values of even 25 mg/L.h at a late cultivation
stage. These results suggest that the higher nutrient availability found under the continuous
178 Resultados
feeding rate (4.2 mL/h) led to elevated metabolic activity of P. taetrolens cells during the
lactobionic acid production phase.
A B 100
2
80
DOT (%)
1.5
60
1 4.2 mL/h 4.2 mL/h

40
8.4 mL/h 8.4 mL/h
16.8 mL/h 16.8 mL/h
0.5 20 33.6 mL/h
33.6 mL/h
67.2 mL/h
67.2 mL/h
0 0
0 20 40 60 80 0 20 40 60 80
Time (h) Time (h)
C 8
D 100
4.2 mL/h 4.2 mL/h
8.4 mL/h 80 8.4 mL/h
OUR (mg/L.h)
7.5
16.8 mL/h 16.8 mL/h
33.6 mL/h 60
pH
33.6 mL/h
7 67.2 mL/h 67.2 mL/h
40
6.5
20
6 0
0 20 40 60 80 0 20 40 60 80
Time (h) Time (h)

E 100 F 100
4.2 mL/h
80 8.4 mL/h 80
Lactose (g/L)
16.8 mL/h
60 33.6 mL/h 60
67.2 mL/h 4.2 mL/h
40 40 8.4 mL/h
16.8 mL/h
20 20
33.6 mL/h
67.2mL/h
0 0
0 20 40 60 80 0 20 40 60 80
Time (h) Time (h)
Figure 2. Effect of different constant feeding rates (4.2, 8.4, 16.8, 33.6 and 67.2 mL/h) on P. taetrolens fed-
batch cultivations. Time-course profiles of cell growth (A), DOT (B), pH (C), OUR (D), lactose (E) and lactobionic
acid (F) are shown.
Resultados 179
Unlike cellular proliferation, lactobionic acid production was negatively affected by

higher feeding rates (Fig. 3). The use of feeding rates higher than 8.4 mL/h was therefore
detrimental to lactose oxidation, leaving residual lactose contents of 2, 13 and 17 g/L after
80 h for rates of 16.8, 33.6 and 67.2 mL/h, respectively (Fig. 2E). Despite the higher lactose
availability in the fermentation broth (around 90 g/L) at rates of 33.6 and 67.2 mL/h,
lactose oxidation into lactobionic acid was negatively affected, suggesting that 8.4 mL/h was
the optimal rate, since it met the specific requirements for lactose oxidation by P. taetrolens
cells, as can be seen in Figs. 2E and F. Note that a feeding rate of 4.2 mL/h led to nearly
complete exhaustion of the lactose content at 34 h which was detrimental for further
oxidation of the lactose added via this continuous feeding strategy (Fig. 2E), resulting in a
bioconversion yield of 90% compared to the complete bioconversion (100%) achieved at
8.4 mL/h (Table 1). This latter fact might suggest that induction of lactose oxidation in P.
taetrolens cells needs a lactose level higher than 5 g/L in the fermentation broth. The overall
fermentation performance in terms of lactobionic acid productivity was clearly reduced
under increased feeding rates, resulting in a volumetric productivity of 0.97 g/L.h obtained
at a rate of 67.2 mL/h versus a productivity of 1.25 g/L.h at a rate of 8.4 mL/h (Fig. 3).
Figure 3. Relationship between feeding rates and specific growth rate, maximum biomass (expressed as Dry
Cell Weight, DCW), volumetric productivity and pH-shift values obtained during fed-batch pH-shift cultivations
at 6.5.
180 Resultados
Table 1. Summary of values obtained in pH-shift fed-batch cultivations under the influence of different feeding
rates.
Duration of Onset of the Volumetric
Feeding rate Maximum dry cell
the growth production pH-shift value μ (h-1) productivity Yield (%) a
(mL/h) weight (g/L)
phase (h) phase (h) (g/L.h)
4.2 1.50 ± 0. 06 10 6 7.52 ± 0. 01 0.08 1.13 ± 0. 03 90 ± 2
8.4 1.70 ± 0. 04 10 6 7.64 ± 0. 04 0.09 1.25 ± 0. 00 100 ± 0
16.8 2.01 ± 0. 05 10 8 7.74 ± 0. 03 0.12 1.19 ± 0. 02 95 ± 2
33.6 2.12 ± 0. 05 10 10 7.90 ± 0. 05 0.16 1.02 ± 0. 00 82 ± 0
67.2 2.15 ± 0. 07 12 12 7.92 ± 0. 00 0.16 0.97 ± 0. 02 77 ± 2
a Yield was defined as the percentage of lactose converted into lactobionic acid after 72 h.
μ = Specific growth rate
3.3. Influence of multi-pulse feeding conditions on lactobionic acid production

As Fig. 4 shows, a multi-pulse feeding mode was conducted to further improve
lactobionic acid titer as well as to simultaneously examine the influence of this high-yield
production on the physiological status of P. taetrolens. A highly concentrated lactose
solution was accordingly fed into the bioreactor at different time points (10, 24 and 34 h)
before the initial content was depleted, thus maintaining a lactose level of around 80 g/L in
the fermentation broth (Fig 4A). This lactose pulse feeding strategy involved a momentary
depletion in dissolved oxygen values, as can be seen in the DOT profile, which quickly
decreased to lower levels after the pulse and then increased progressively to saturated
values. As can be seen, P. taetrolens cells retained their bioconversion ability throughout the
cultivation, leaving only 9 g/L of residual lactose after 80 h, thus enabling both high-yield
production (94%) as well as high-level productivity (1.8 g/L.h) from whey (Table 2). P.
taetrolens cells also displayed a relatively high specific oxygen consumption rate (~30
mg/gDCW.h) at the mid-stationary phase, although the late cultivation stage featured lower
consumption rates (~16 mg/gDCW.h) (Fig. 4A).
As regards the physiological status of P. taetrolens, the feed pulse did not exert a major
deleterious influence on the membrane integrity of the cells, as the dot plots in Figs 4B and C
show. Cells remained predominantly healthy and metabolically active (70%) from the end of
exponential growth phase onwards (Fig. 4C), despite the high lactose content of the pulse
feed (275 g/L), suggesting that cells did not suffer any resulting stress-induced physiological
response. The onset of the production phase featured progressive depolarization (85% from
34 onwards), although the cells retained their membrane integrity throughout the
Resultados 181
cultivation (Fig. 4B). This robust physiological status exhibited by P. taetrolens cells makes
them particularly well-suited for employing concentrated nutrient solutions to further
prolong the growth and/or production phases. The loss in the cellular functionalities of P.
taetrolens cells could lead to a decline in their lactobionic acid producing ability due to the
presence of undesired damaged and dead subpopulations (Alonso et al., 2012b).
B
0h 10 24 h 48 h 72 h
DiBAC4(3)
PI
C
0h 10 h 24 h 48 h 72 h
cFDA
PI
Figure 4. Time-course profile of bioprocess parameters (A) and dot plots representing DiBAC4(3)/PI (B) and
cFDA/PI fluorescence signals (C) during multi-pulse fed-batch cultivation. Arrows indicate the time point of
pulse feeding with 0.1 L of 275 g/L lactose.
3.4. Co-feeding strategy for enhanced lactobionic acid production

In order to achieve high-level production of lactobionic acid, cultivations were submitted
to a novel co-feeding strategy combining continuous feeding of concentrated whey
(performed at two different rates, 4.2 and 33.6 mL/h) with multi-pulse feedings of highly
concentrated lactose solutions. Specifically, highly concentrated whey solutions (Ahn et al.,
2000) in combination with nutrient supplementation (Koller et al., 2008; Nath et al., 2008)
182 Resultados
are usually employed to achieve enhanced metabolite production from whey in fed-batch
cultivations. Supplementary nitrogen sources such as yeast extract and peptone do in fact
promote P. taetrolens growth, since they provide vitamins, trace elements and amino acids
as carbon source which support cellular growth (Matsui et al., 2009). As can be seen in Fig.
5A, the co-feeding strategy at a rate of 4.2 mL/h displayed a higher specific growth rate
(0.21 h-1) compared to batch cultivation (0.12 h-1), although the maximum cell biomass
resulted in similar values (around 2.1 g/L). Furthermore, similar maximum cell biomass
values were comparatively achieved between the two feeding rates tested, 4.2 and 33.6
mL/h, although the former rate supported a higher biomass during the lactobionic acid
production phase due to the continuous feeding strategy (Fig. 6A).
B
0h 10 24 h 48 h 80 h
DiBAC4(3)
PI
C
0h 10 h 24 h 48 h 80 h
cFDA
PI
Figure 5. Bioprocess parameters (A) and physiological status of P. taetrolens from DiBAC4(3)/PI (B) and cFDA/PI
staining (C) during fed-batch cultivation performed under a co-feeding strategy at a rate of 4.2 mL/h. Arrows
indicate the time point of pulse feeding with highly concentrated lactose solution (0.1 L of 275 g/L).
Resultados 183
The operational life span of the fed-batch fermentations also increased in comparison
with batch cultivation using concentrated sweet whey (100 versus 48 h) (Fig. 1). However,
the growth phases were not prolonged under these co-feeding strategies, as Figs. 5A and 6A
show. The DOT profile in fed-batch cultivations performed at 33.6 mL/h returned to
saturated values after completing the feed (4.5 h) in contrast with the cultivation carried out
under a continuous feeding rate, in which the oxygen-period was prolonged for 17 h (Figs.
5A and 6A). Besides, an abrupt decline in the DOT profile was found at 4.2 mL/h in response
to the lactose feeding pulses, whereas the absence of DOT responses at 33.6 mL/h indicated
that the cultivation was operating under lower nutrient availability. As regards the oxygen
consumption, plateaus in the specific oxygen uptake rate were reached simultaneously in
both cases with the onset of lactobionic acid formation.
B
0h 10 24 h 48 h 80 h
DiBAC4(3)
PI
C
0h 10 h 24 h 48 h 80 h
cFDA
PI
Figure 6. Time-course profile of bioprocess parameters (A) and dot plots representing DiBAC4(3)/PI (B) and
cFDA/PI fluorescence signals (C) during fed-batch cultivation performed under a co-feeding strategy at a rate of
33.6 mL/h. Arrows indicate the time point of pulse feeding with highly concentrated lactose solution (0.1 L of
275 g/L).
184 Resultados
Fermentation performance was clearly enhanced under the co-feeding strategy at 4.2
mL/h, in which a lactobionic acid titer of 180 g/L and a yield of 90% were obtained after
100 h, compared to a titer of 162 g/L and a yield of 80% obtained under co-feeding at 33.6
mL/h. The adopted feeding strategy led to an uncoupling of lactobionic acid formation from
cell growth, as can be seen in the cellular growth profiles of P. taetrolens (Figs. 5A and 6A).
Moreover, the co-feeding strategy at 4.2 mL/h maintained the lactose level in the
fermentation broth above 100 g/L (Fig. 5A), in contrast with the 150 g/L maintained under
a co-feeding strategy at 33.6 mL/h (Fig. 6B), thus suggesting that higher lactose availability
levels in the fermentation broth did not stimulate lactobionic acid production.
Comparatively, the co-feeding strategy at 4.2 mL/h presented an enhancement in terms
of productivity (1.13- and 1.8-fold higher than the continuous and the multi-pulse feeding
strategy, respectively) and lactobionic acid titer (180 versus 81 or 143 g/L under
continuous and multi-pulse feeding, respectively). Whereas a yield of 94% was obtained
under a fed-batch process with multi-pulse lactose feeding, a yield of 82% was only obtained
under co-feeding conditions (Table 2). These results demonstrate that these fed-batch
cultivation approaches enhanced microbial culture performance as well as increasing the
lactobionic acid titer.
Table 2. Comparison of bioprocess parameters, lactobionic acid productivity and titer values obtained under
different feeding strategies.
Maximum dry Duration of Onset of the Volumetric
Feeding pH-shift Culture
cell weight the growth production μ (h-1) productivity Yield (%) a
strategy value time (h)
(g/L) phase (h) phase (h) (g/L.h)
Continuous b 1.50 ± 0. 06 10 8 7.52 ± 0. 01 0.08 1.13 ± 0. 03 72 90 ± 2
Multi-pulse 1.97 ± 0. 05 10 12 6.82 ± 0. 05 0.04 1.80 ± 0. 02 80 94 ± 1
Co-feeding c 2.05 ± 0. 08 10 12 7.08 ± 0. 05 0.21 2.05 ± 0. 02 72 82 ± 2
a Yield was defined as the percentage of lactose converted into lactobionic acid after each culture time.
b Feeding rate of 4.2 mL/h on diluted whey
c Feeding rate of 4.2 mL/h on concentrated whey
μ = Specific growth rate
Regarding the physiological status of P. taetrolens, cells remained polarized during the
growth phase, whereas a damaged subpopulation appeared at the late cultivation stage (Fig.
5C and 6C). Despite the high lactobionic acid productivity achieved under co-feeding
conditions, cells remained depolarized throughout fed-batch cultivation (85%). The feed
pulse neither induced major changes in the physiological status displayed by cells nor
influenced fermentation performance or lactobionic acid productivity (Fig. 5B and 6B).
Resultados 185
Comparatively, the main subpopulation (70%) was damaged at late cultivation stages under
the co-feeding mode at 33.6 mL/h (Fig. 6C) versus a higher metabolic status (50%) under
continuous co-feeding at 4.2 mL/h even at late stages due to higher nutrient availability (Fig.
5C). The co-feeding strategy thus led to a robust physiological status, suggesting that these
bioprocessing conditions did not exert any potential deleterious influence on the cellular
functionality of P. taetrolens. Undoubtedly, the co-feeding mode at the lowest rate (4.2
mL/h) was able to cope with the cellular responses and achieved a balance between cellular
maintenance and lactobionic acid production. The nutrient depletion suffered in the co-
feeding mode at the highest rate (33.6 mL/h) could induce considerably higher cell stress
and therefore less cellular robustness in terms of physiological status.
Table 3 shows the comparison of different microbial fermentation approaches carried
out for the bio-production of lactobionic acid, including whole-cell and permeabilized-cell
bioconversion systems. In terms of productivity, the values obtained in the present study
(2.05 g/L.h) were higher than those obtained by Miyamoto et al. (2000) (1.87 g/L.h). In this
latter study, Pseudomonas sp. LS13-1 mutants were able to produce 290 g/L of lactobionic
acid from complex nutrient media with a yield of 90% in a fed-batch cultivation
supplemented with 15 g/L of peptone as nitrogen source (Miyamoto et al., 2000). Mutants of
Burkholderia cepacia were additionally found to produce up to 150-400 g/L of lactobionic
acid with high volumetric productivities of 1.67-5.55 g/L.h on complex culture media over
periods of 27-240 h under resting-state conditions at a shake-flask scale (Murakami et al.,
2002, 2006). Despite incomplete lactose bioconversion (85%), a high-level lactobionic acid
titer of 178 g/L and a productivity of 3.56 g/L.h were also achieved by B. cepacia on a
complex culture medium (Meiberg et al., 1990). Nevertheless, the pathogenicity of B. cepacia
discourages any industrial-scale implementation of this microbial platform as a bionic acid-
producing microorganism.
Permeabilized resting free-cell systems of Zymomonas mobilis have also been used to
obtain high-level production of lactobionic acid from lactose and fructose solutions
(Malvessi et al., 2012; Pedruzzi et al., 2011). Comparatively, the volumetric productivities
obtained under resting-state conditions in these systems were 2.8- and 3.7-fold higher,
respectively, than that obtained in the present study under co-feeding conditions (2.05
g/L.h). Interestingly, the lactobionic acid titers obtained in the present study were 1.4-fold
higher than that obtained by Pedruzzi et al. (2011), who reported around 125.4 g/L of
lactobionic acid from 252 g/L of chemically pure lactose solution (Table 3). The specific
lactobionic acid productivity obtained in the present study (1.4 g/gDCW.h) is the highest
reported to date for submerged microbial bioprocesses targeting the production of this
186 Resultados
polyhydroxy bionic acid. In this regard, bioconversion systems employing permeabilized

resting cells of Z. mobilis featured specific productivities of 0.3 (Malvessi et al., 2012) and 0.8
g/gDCW.h (Pedruzzi et al., 2011). These results were substantially lower than that obtained
in the present study (1.4 g/gDCW.h), which thus represents an efficient rate of
bioconversion ability by P. taetrolens cells (Table 3).
As previously highlighted, a distinctive feature of the present fermentation approach is
the use of cheese whey as an inexpensive feedstock for lactobionic acid production. In this
study, a novel co-feeding strategy was developed to achieve high-level lactobionic acid
production from concentrated cheese whey without costly nutrient supplements.
Fermentation efficiency in terms of volumetric productivity was clearly enhanced, being 2.9-
fold higher than that obtained in batch cultivation at bioreactor (Alonso et al., 2011) or
shake-flask scale (Alonso et al., 2012b) (2.05 versus 0.70 g/L.h for fed-batch and batch
cultivations, respectively) (Table 3). Furthermore, the lactobionic acid titer (164 g/L) was
3.4-fold higher than that obtained in batch cultivation (42 g/L) under high-cell density
conditions on diluted whey (Alonso et al., 2012a). The present study also examined the
cellular functionality of P. taetrolens cells during fed-batch cultivations. In this regard, fed-
batch approaches are applied to obtain high production yields of target metabolites (Tang et
al., 2009; Yu et al., 2012). However, these bioprocessing conditions may induce undesired
microbial physiological responses which could have a marked impact on overall system
performance. Thus, the monitoring of cellular physiology may specifically provide valuable
information to improve biological performance, thereby achieving higher biological
efficiencies as well as enhanced lactobionic acid production yields by P. taetrolens (Alonso et
al., 2012b, 2012c). Such detailed knowledge of physiological heterogeneity may also provide
valuables guidelines for the design of improved formulations and cultivation strategies
(Amanullah et al., 2002; Lopes da Silva et al., 2011; Want et al., 2009). Results from the
present study have revealed that a co-feeding strategy with highly concentrated whey
solution did not exert a major deleterious influence on the physiological status of P.
taetrolens cells. Lactobionic acid-producing cells retained their membrane integrity and
functionality throughout fed-batch cultivations, although multi-parameter flow cytometry
assessment enabled the detection of an increase in the depolarization level which was
associated with nutrient exhaustion following the feeding pulse.
Resultados 187
Table 3. Comparison of research studies on the microbial production of lactobionic acid.

Cultivation
Lactobionic Specific
Culture strategy and Culture Productivity Yield
Microorganism acid titer productivity Reference
medium operational time (h) (g/L.h) (%) a
(g/L) (g/g.h)
mode
Pseudomonas sp. Lactose, salts Fed-batch in Miyamoto

155 290 - 1.87 90
LS13-1 b and peptone bioreactor et al. (2000)
Lactose, salts,
peptone, yeast Fed-batch in Meiberg et
Burkholderia cepacia 50 178 - 3.56 85
extract and bioreactor al. (1990)
glucose
Lactose, salts,
Burkholderia cepacia Fed-batch in Murakami
peptone and 240 400 - 1.67 100
No. 24 b shake-flask et al. (2003)
yeast extract
Lactose, salts,
Burkholderia cepacia corn steep Batch in shake- Murakami
27 150 - 5.55 ~100
No. 24 b liquor and flask et al. (2006)
yeast extract
Batch in Pedruzzi et
Zymomonas mobilis c Lactose 22 125 0.80 5.80 100
bioreactor al. (2011)
Lactose, Batch in Malvessi et

Zymomonas mobilis c 24 182 0.30 7.60 78
fructose bioreactor al. (2012)
Pseudomonas Batch in Alonso et al.

Whey 58 42 0.56 0.70 100
taetrolens bioreactor (2011)
Pseudomonas Batch in Alonso et al.

Whey 30 42 0.94 1.27 100
taetrolens bioreactor (2012a)
Pseudomonas Batch in shake- Alonso et al.

Whey 60 42 0.42 0.70 100
taetrolens flask (2012b)
Pseudomonas Concentrated Batch in

48 78 1.02 1.63 100 This study
taetrolens whey bioreactor
Pseudomonas Concentrated Fed-batch in

80 164 1.40 2.05 82 This study
taetrolens whey bioreactor
-: not available
a Yield was defined as the percentage of lactose converted into lactobionic acid after each cultivation time.
b Mutant strain.
c Permeabilized resting free-cells.
188 Resultados
4. Conclusions
The present study has demonstrated the feasibility of whey as an inexpensive source for
lactobionic acid bio-production by P. taetrolens at an industrially relevant titer. A high-level
titer of 180 g/L was obtained with a yield of 90% via fed-batch cultivation carried out under
co-feeding conditions. Moreover, the physiological responses of P. taetrolens cells were
monitored through flow cytometry in order to assess the impact of different feeding
strategies on bioprocess efficiency. High-yield bio-production of lactobionic acid was
directly linked to the fully functional status of P. taetrolens, thus providing relevant
information for successful industrial implementation.
Acknowledgements
The authors are grateful for funding from the Spanish Ministry of Science and Innovation
through project MEC-CTQ2010-14918. The technical assistance of Ana Salas and Magdalena
Choda (Flow Cytometry Area, Scientific-Technical Services, University of Oviedo) is
gratefully acknowledged.
References
Ahn, W.S., Park, S.J., Lee, S.Y., 2000. Production of poly(3-hydroxybutyrate) by fed-batch
culture of recombinant Escherichia coli with a highly concentrated whey solution.
Appl. Environ. Microb. 66, 3624-3627.
Alonso, S., Rendueles, M., Díaz, M., 2011. Efficient lactobionic acid production from whey by
Pseudomonas taetrolens under pH-shift conditions. Bioresour. Technol. 102, 9730-
9736.
Alonso, S., Rendueles, M., Díaz, M., 2012a. Role of dissolved oxygen availability on lactobionic
acid production from whey by Pseudomonas taetrolens. Bioresour. Technol. 109, 140-
147.
Alonso, S., Rendueles, M., Díaz, M., 2012b. Physiological heterogeneity of Pseudomonas
taetrolens during lactobionic acid production. Appl. Microbiol. Biotechnol. 96, 1465-
1477.
Alonso, S., Rendueles, M., Díaz, M., 2012c. Selection method of pH conditions to establish
Pseudomonas taetrolens physiological states and lactobionic acid production. Appl.
Microbiol. Biotechnol. In press. doi: 10.1007/s00253-012-4607-x
Amanullah, A., Hewitt, C.J., Nienow, A.W., Lee, C., Chartrain, M., Buckland, B.C., Drew, S.W.,
Woodley, J.M., 2002. Fed-batch bioconversion of indene to cis-indandiol. Enzyme
Microb. Tech. 31, 954-967.
Resultados 189
Chia, Y.N., Latusek, M.P., Holles, J.H., 2008. Catalytic wet oxidation of lactose. Ind. Eng. Chem.
Res. 47, 4049-4055.
Gerling, K.G., 1998. Large scale production of lactobionic acid-use and new applications. Int.
Dairy Fed. 9804, 251-256.
Green, B.A., Yu, R.J., Van Scott, E.J., 2009. Clinical and cosmeceutical uses of hydroxyacids.
Clin. Dermatol. 27, 49-501.
monitor individual microbial cell physiological state. Adv. Biochem. Eng. Biot. 89,197-
223.
Hewitt, C.J., Nebe-von-Caron, G., Nienow, A.W., McFarlane, C.M., 1999. The use of multi-
parameter flow cytometry to compare the physiological response of Escherichia coli
W3110 to glucose limitation during batch, fed-batch and continuous culture
cultivations. J. Biotechnol. 75, 251-264.
Koller, M., Bona, R., Chiellini, E., Grillo Fernandes, E., Horvat, P., Kutschera, C., Hesse, P.,
Braunegg, G., 2008. Polyhydroxyalkanoate production from whey by Pseudomonas
hydrogenovora. Bioresour. Technol. 99, 4854-4863.
Kuusisto, J., Tokarev, A.V., Murzina, E.V., Roslund, M.U., Mikkola, J.P, Murzin, D.Y. Salmi, T.,
2007. From renewable raw materials to high value-added fine chemicals-catalytic
hydrogenation and oxidation of d-lactose. Catal. Today. 121, 92-99.
Lopes da Silva, T., Piekova, L., Mileu, J., Roseiro, J.C., 2009. A comparative study using the
dual staining flow cytometric protocol applied to Lactobacillus rhamnosus and Bacillus
licheniformis batch cultures. Enzyme Microb. Tech. 45, 134-138.
Lopes da Silva, T., Feijão, D., Roseiro, J.C., Reis, A., 2011. Monitoring Rhodotorula glutinis
CCMI 145 physiological response and oil production growing on xylose and glucose
using multi-parameter flow cytometry. Bioresour. Technol. 102, 2998-3006.
Matsui, D., Oikawa, T., Arakawa, N., Osumi, S., Lausberg, F., Stäbler, N., Freudi, R., Eggeling, L.,
2009. A periplasmic, pyridoxal-5’-phosphate-dependent amino acid racemase in
Pseudomonas taetrolens. Appl. Microbiol. Biotechnol. 83, 1045-1054.
Malvessi, E., Carra, S., Pasquali, F.C., Kern, D.B., da Silveira, M.M., Ayub, M.A.Z., 2012.
Production of organic acids by periplasmic enzymes present in free and immobilized
cells of Zymomonas mobilis. J. Ind. Microbiol. Biotechnol. In press. doi:
10.1007/s10295-012-1198-6
190 Resultados
Meiberg, J.B.M., Bruinenberg, P.M., Sloots, B., 1990. A process for the fermentative oxidation
of reducing disaccharides. European Patent Application Pub. No.: EP19900200393.
Miyamoto, Y., Ooi, T, Kinoshita, S., 2000. Production of lactobionic acid from whey by
Pseudomonas sp. LS13-1. Biotechnol. Lett. 22, 427-430.
Müller, S., Harms, H., Bley, T., 2010. Origin and analysis of microbial population
heterogeneity in bioprocesses. Curr. Opin. Biotech. 21, 100-113.
Murakami, H., Seko, A., Azumi, M., Kiso, T., Kiryu, T., Kitahata, S., Shimada Y., Nakano, H.,
2006. Microbial conversion of lactose to lactobionic acid by resting cells of
Burkholderia cepacia No. 24. J. Appl. Glycosci. 53, 7-11.
Murakami, H., Seko, A., Azumi, M., Ueshima, N., Yoshizumi, H., Nakano, H., Kitahata, S., 2003.
Fermentative production of lactobionic acid by Burkholderia cepacia. J. Appl. Glycosci.
50, 117-120.
Nath, A., Dixit, M., Bandiya, A., Chavda, S., Desai, A.J., 2008. Enhanced PHB production and
scale up studies using cheese whey in fed batch culture of Methylobacterium sp. ZP24.
Bioresour. Technol. 99, 5749-5755.
Park, I.K., Yang, J., Jeong, H.J., Bom, H.S., Harada, I., Akaike, T., Kim, S.I., Cho, C.S., 2003.
Galactosylated chitosan as a synthetic extracellular matrix for hepatocyte attachment.
Biomaterials 24, 2331-2337.
Pedruzzi, I., da Silva, E.A.B., Rodrigues, A.E., 2011. Production of lactobionic acid from
lactose/fructose substrate using GFOR/GL enzymes from Zymomonas mobilis cells: A
kinetic study. Enzyme Microb. Tech. 49, 183-191.
Pedruzzi, I., Malvessi, E., Mata, V.G., Silva, E.A.B., Silveira, M.M., Rodrigues, A.E., 2007.
Quantification of lactobionic acid and sorbitol from enzymatic reaction of fructose and
lactose by high-performance liquid chromatography. J. Chromatogr. A. 1145, 128-132.
Peng, J., Wang, K., Tan, W., He, X., He, C., Wu, P., Liu, F., 2007. Identification of live liver cancer
cells in a mixed cell system using galactose-conjugated fluorescent nanoparticles.
Talanta 71, 833-840.
Selim, K.M.K., Ha, Y.S., Kim, S.J., Chang, Y., Kim, T.J., Lee, G.H., Kang, I.K., 2007. Surface
modification of magnetite nanoparticles using lactobionic acid and their interaction
with hepatocytes. Biomaterials 28, 710-716.
Southard, J.H., Belzer, F.O., 1995. Organ preservation. Annu. Rev. Med. 46, 235-247.
Tang, Y.J., Zhang, W., Zhong, J.J., 2009. Performance analyses of a pH-shift and DOT-shift
integrated fed-batch fermentation process for the production of ganoderic acid and
Ganoderma polysaccharides by medicinal mushroom Ganoderma lucidum. Bioresour.
Technol. 100, 1852-1859.
Resultados 191
Want, A., Thomas, O.R.T., Kara, B., Liddell, J., Hewitt, C.J., 2009. Studies related to antibody
fragment (Fab) production in Escherichia coli W3110 fed-batch fermentation
processes using multiparameter flow cytometry. Cytometry A 75, 148-154.
Yu, Z., Du, G., Zhou, J., Chen, J., 2012. Enhanced α-ketoglutaric acid production in Yarrowia
lipolitica WSH-Z06 by an improved integrated fed-batch strategy. Bioresour. Technol.
114, 597-602.
5. DISCUSIÓN GENERAL
Discusión general 195
5. DISCUSIÓN GENERAL
5.1. Producción biotecnológica de ácido láctico
Diferentes materias primas procedentes de subproductos de origen alimentario han sido

materia de estudio con el fin de ser utilizadas como sustrato fermentativo, desde cereales
caso de la cebada o el trigo (Hofvendahl y Hahn-Hägerdal, 2000; Oh et al., 2005), residuos
agrícolas como el salvado de arroz (Altaf et al., 2004), o incluso hasta zumo de dátiles
(Nancib et al., 2001). Así, las últimas tendencias en materia de investigación en este campo
van encaminadas en la sacarificación (degradación de la biomasa compleja) bien sea a través
de hidrólisis ácida o enzimática y posterior fermentación (John et al., 2009). Ambas etapas
pueden ser llevadas a cabo a través de una hidrólisis y fermentación o como sacarificación
simultánea y posterior fermentación. De hecho, este proceso de fermentación está
íntimamente ligado a la producción de ácido láctico a partir de materias con alto contenido
en almidón (caso de la pulpa de mandioca, almidón de patata, salvado de harina o almidón
de maíz) y de materiales lignocelulósicos (tales como la celulosa o residuos madereros)
utilizando microorganismos amilolíticos. No obstante, el empleo de esta metodología
encarece los procesos de obtención de ácido láctico. Recientemente, los esfuerzos están
centrados en la búsqueda de microorganismos capaces de realizar simultáneamente una
primera sacarificación y posterior fermentación de los carbohidratos resultantes. De hecho,
los esfuerzos relacionados con la producción biotecnológica de ácido láctico van
encaminados actualmente en dos direcciones: mejora en las capacidades funcionales de los
microorganismos a través del aislamiento de muevas cepas amilolíticas o bien en la
utilización de nuevas fuentes no costosas de carbohidratos fermentables. Recientemente
también se vienen sumando esfuerzos en consolidar sistemas fermentativos con el fin de
producir ácido láctico mediante la utilización de especies termofílicas como Bacillus sp.,
microorganismos que permiten el empleo de condiciones operacionales sin la necesidad de
esterilidad (Gao et al., 2012; Meng et al., 2012; Ouyang et al., 2012; Patel et al., 2006).
Así, el presente trabajo se ha centrado en este último ámbito, estudiando la eficiencia del
suero de yogurt como nueva y sostenible fuente de carbohidratos fermentables a través de
una bacteria ácido-láctica con el objetivo de producir ácido L(+)-láctico. L. casei es una
bacteria heterofermentativa facultativa que bajo las condiciones ensayadas ha metabolizado
los correspondientes carbohidratos sin la consiguiente formación de ácido acético o etanol.
Como se ha explicado grosso modo en el subapartado correspondiente (2.1), tanto los
196 Discusión general
requerimientos nutricionales como la productividad de un microorganismo acido-láctico en

un proceso fermentativo están fuertemente influenciados por la naturaleza de la fuente de
carbono presente en el medio. Así, las bacterias del ácido láctico utilizan los carbohidratos
siguiendo de forma estricta un control jerárquico para el cual han desarrollado una serie de
mecanismos de control que dirigen y coordinan la regulación específica del metabolismo
central.
El consumo de carbohidratos por L. casei ha reflejado la capacidad de este
microorganismo para metabolizar de forma eficiente las fuentes de carbono presentes en el
suero de yogurt. Las Figuras 5.1 y 5.2 muestran la capacidad metabólica de L. casei. Estos
resultados han sugerido un consumo co-metabólico llevado a cabo por L. casei, a partir del
cual se puede inferir que este microorganismo es capaz de metabolizar activamente la
glucosa, la sacarosa y la lactosa presentes en el medio. Sin embargo, hay un perfil de
preferencia de consumo de estos azúcares, así, la glucosa es el primer azúcar metabolizado,
mientras que el contenido en lactosa apenas se ve ligeramente reducido con el tiempo
(Figura 5.2). Dicho contenido residual es similar en todos los casos, lo que sugiere que la
mejora en el sistema operacional adoptado (control de pH y/o suplemento de extracto de
levadura) no estimula el consumo de lactosa. Sin embargo, el control del pH y/o la presencia
de extracto de levadura en precultivo sí han estimulado tanto el consumo de glucosa como
de sacarosa (Figura 5.2).
Figura 5.1. Rendimiento (porcentaje de carbohidratos consumidos sobre el contenido en azúcares totales) y
productividad volumétrica de los distintos procesos fermentativos llevados a cabo para obtener ácido láctico.
Asimismo, estos resultados han demostrado que L. casei prefiere la glucosa y sacarosa
sobre la lactosa como fuente primaria de carbono y energía a lo largo de un proceso ácido-
láctico. De hecho, el consumo de carbohidratos ha seguido un patrón muy similar en todos

los casos, pero en cada caso, la cantidad producida de ácido láctico ha sido distinta debido al
rendimiento diferencial de cada proceso fermentativo. La presencia de extracto de levadura
en el precultivo junto con el control del pH incrementó el rendimiento hasta 0.9 g/g frente a
valores similares entorno a 0.71 g/g encontrados sin control de pH. Como se muestra en la
Figura 5.1, la productividad volumétrica alcanzó valores de 0.76 g/L.h bajo condiciones de
control de pH y en presencia de extracto de levadura en el precultivo, siendo 2.1 veces
superior a la productividad lograda sin control de pH (0.36 g/L.h). El consumo metabólico
de estos carbohidratos fermentables ha seguido además una secuencia, algo que sugiere una
utilización jerárquica y simultánea, llevada a cabo por un estricto mecanismo de control
metabólico. A partir de los perfiles de consumo se puede inferir también que L.casei
presenta diferentes rendimientos de ácido láctico producido en base al sustrato empleado,
siendo más altas las relativas al consumo de glucosa y sacarosa (Figura 5.2).
Figura 5.2. Porcentaje de carbohidratos consumidos tras 34 h respecto al contenido inicial.
En concreto, en L. casei, la lactosa es translocada al interior celular como lactosa fosfato,

donde es hidrolizada en glucosa y galactosa-6-fosfato por una fosfo-β-D-galactosidasa. En
función del modo de transporte, los disacáridos (caso de la sacarosa o la lactosa) pueden
entrar en la célula bien en forma libre o fosforilada. En el primer caso, los disacáridos libres
son hidrolizados en monosacáridos por hidrolasas específicas. La translocación como azúcar
fosfato implica la presencia del sistema enzimático de carbohidrato fosfotransferasa, así
como las fosfohidrolasas específicas que se encargan de hidrolizar el disacárido fosfato en
dos monosacáridos, uno libre y otro con el grupo fosfato. Mientras que la glucosa es
fosforilada por una glucoquinasa y metabolizada a través de la vía glucolítica, la galactosa-6-

fosfato es metabolizada a través de la ruta tagatosa-6-fosfato (Figura 5.3). Asimismo, la
utilización de la sacarosa también está mediada por el sistema sacarosa fosfotransferasa.
Una vez internalizado, el catabolismo de este disacárido comienza con la hidrólisis por la
sacarosa-6-fosfato hidrolasa dando lugar a glucosa-6-fosfato y fructosa, las cuales entran en
la vía glucolítica (Figura 5.3). Tanto el sistema sacarosa fosfotransferasa como la sacrosa-6-
fosfato hidrolasa son inducidos por la presencia de sacarosa en el medio de cultivo. En el
caso de la glucosa, ésta es translocada al interior celular en su forma fosforilada a través del
sistema glucosa fosfotransferasa, siendo metabolizada posteriormente a través de la ruta
catabólica Embden-Meyerhoff.
La glucolisis (conocida también como ruta Embden-Meyerhoff) se caracteriza por la
formación de fructosa-1,6-difosfato que es hidrolizada por una aldolasa en dihidroxiacetona
fosfato y gliceraldehído-3-fosfato. Este último metabolito es transformado en piruvato
mediante la fosforilación a nivel de sustrato en dos sitios específicos distintos. De esta
forma, el piruvato es reducido a ácido láctico por una lactato deshidrogenasa dependiente
de NADH (el NADH formado en etapas previas se reoxida) bajo condiciones normales
(microaerófilas). Con la obtención de este balance rédox metabólico se obtiene ácido láctico
como producto final de la fermentación homoláctica. En el supuesto caso de que no exista un
aceptor adicional de electrones, el acetilfosfato es reducido a etanol vía acetil-CoA y
acetaldehído, dando lugar a la formación de etanol o ácido acético en la llamada
fermentación heteroláctica (Panesar et al., 2007; Salminen et al., 2004).
L. casei presenta un patrón de consumo secuencial que no ha sido descrito con
anterioridad para esta bacteria ácido láctica, siguiendo el siguiente orden: glucosa >
sacarosa > lactosa. Así, L. casei presenta un co-consumo de glucosa, sacarosa y lactosa
presentes en el suero de yogurt, siendo glucosa y la sacarosa los sustratos preferenciales.
Los patrones de consumo, bien sean secuenciales, simultáneos o mixtos, son modificados en
base a la presencia de diferentes carbohidratos. De hecho, se ha descrito cómo Pediococcus
pentosaceus ha experimentado diferente patrón de consumo en función de los diferentes
carbohidratos presentes en el medio de cultivo durante la producción de ácido láctico.
Mientras que la galactosa y arabinosa fueron co-metabolizados, el resto de azúcares
procedentes de residuos lignocelulósicos fueron metabolizados de forma secuencial (Adler
et al., 2012).
Figura 5.3. Rutas metabólicas involucradas en la producción de ácido láctico en L. casei a partir de glucosa,
lactosa y sacarosa.
Los altos niveles de lactosa residual, presentes en el medio una vez terminada la
fermentación, revelan la baja afinidad de L. casei por la lactosa en presencia de fuentes de
carbono extras como representan la glucosa y la sacarosa. Estos resultados sugieren que
este carbohidrato no comienza a ser consumido por L. casei hasta el agotamiento de las
fuentes de carbono preferenciales, como son la glucosa y sacarosa. La baja tasa de
asimilación de la lactosa se explica por la probable presencia de un sistema fosfotransferasa
inducido con mayor afinidad por la glucosa y la sacarosa. No obstante, L. casei fue capaz de
metabolizar la mayoría del contenido en glucosa y sacarosa en 48 horas consiguiendo
rendimientos superiores al 80% de producción de ácido láctico y con valores superiores a
42 g/L en el caso de una fermentación sin suplemento de extracto de levadura en el
preinóculo. El rendimiento (g/g consumido) de cada fermentación muestra cómo la mejora
de las condiciones operacionales repercutió favorablemente sobre la producción de ácido
láctico. Indudablemente, como anteriormente se ha resaltado, la mejora de las condiciones
operacionales unidas a la presencia de extracto de levadura en el preinóculo logró mejorar
el rendimiento global de la fermentación (42.54 versus 30.1 g/L de carbohidratos
residuales).
En relación al status fisiológico, como se ha mostrado en el subapartado correspondiente
(4.2), la fisiología de L. casei se ve fuertemente afectada por las diferentes condiciones
microambientales. Es indudable la importancia del conocimiento preciso de la fisiología
poblacional de la bacteria empleada como plataforma biocatalítica en un sistema. La

citometría de flujo multi-paramétrica permite la monitorización y el seguimiento de los
bioprocesos microbianos, conociendo asimismo con exactitud las distintas propiedades
celulares que influyen decisivamente sobre el status fisiológico poblacional (Díaz et al.,
2010; Müller y Nebe-von-Caron, 2010; Tracy et al., 2010). Así, el análisis multi-paramétrico
(enfocado al conocimiento de diferentes propiedades celulares) permite evaluar incluso el
impacto de diversos factores sobre la propia célula como tal, evaluando incluso la posible
toxicidad del metabolito producido o del propio sustrato (Figura 5.4). Esta caracterización a
nivel poblacional del bioproceso permite realizar modificaciones precisas sobre el sistema
con el fin de optimizar el proceso biocatalítico en términos de productividad y rendimiento
global. Son múltiples los estudios que analizan y caracterizan diferentes bioprocesos
mediante citometría de flujo, así esta técnica ha sido empleada con éxito en diferentes
bioprocesos involucrados en bioconversiones microbianas (Amanullah et al., 2002a, 2002b,
2003) e incluso en fermentaciones de cerveza (Kobayashi et al., 2007) o de sidra (Herrero et
al., 2006; Quirós et al., 2007).
Los conocimientos que aporta el empleo de esta técnica en estos sistemas microbianos
van desde la cuantificación de la viabilidad celular a la caracterización del status fisiológico
poblacional, dos parámetros que no pueden ser detectados ni cuantificados siguiendo
técnicas convencionales. Asimismo, esta técnica está siendo empleada en procesos
relacionados con la seguridad alimentaria con el fin de evaluar la eficacia de diferentes
técnicas antimicrobianas tales como el empleo de altas presiones o fluidos supercríticos. Por
tanto, a través del empleo de la citometría de flujo es posible monitorizar en tiempo real el
estado metabólico-fisiológico de toda población microbiana involucrada en cualquier
proceso de interés e incluso se presenta como una herramienta prometedora en la
evaluación de diferentes fenotipos microbianos surgidos a consecuencia de la ingeniería
metabólica.
Figura 5.4. Impacto directo de diferentes variables microambientales sobre el microorganismo y cómo éste
responde frente al stress con diferentes patrones de modulación fisiológica.
El protocolo de citometría de flujo empleado en el presente trabajo ha permitido evaluar

el estado fisiológico de L. casei en relación a dos importantes parámetros como son la
actividad metabólica y la integridad de membrana con el fin de elucidar y monitorizar la
heterogeneidad subpoblacional en las diferentes fermentaciones. Igualmente, la utilización
de un protocolo de citometría de flujo ha proporcionado una valiosa información acerca de
la influencia del microambiente sobre la fisiología celular de L. casei. Los ensayos de
viabilidad realizados en el presente trabajo se han llevado a cabo mediante tinción dual de
las muestras con el fin de discriminar células con membranas intactas y con actividad
enzimática (células metabólicamente activas) de las células dañadas y muertas (células con
la membrana dañada). La pérdida de la integridad de membrana (ligada al daño en la
membrana citoplásmica) implica la entrada en un estado fisiológico intermedio que puede
sobrellevar con la muerte celular debido a la depleción energética celular (se disipa la fuerza
protón motriz). Este estado fisiológico intermedio está constituido por subpoblaciones
celulares que presentan una doble dualidad funcional, presentando tanto actividad
enzimática como daños en la membrana. Por lo tanto, la evolución de un cultivo puede
implicar que las células vivas y con estrés fisiológico entren en un estado fisiológico
intermedio o en fase de muerte celular. Precisamente, un protocolo de citometría de flujo
con tinción dual puede distinguir los diferentes patrones en los cambios de viabilidad
celular dentro de un bioproceso como el que puede lleva a cabo L. casei en el lactosuero de
yogurt. De esta forma, ha sido posible asimismo elucidar el papel que representa el
suplemento con extracto de levadura sobre la viabilidad y la integridad de membrana.
La fermentación con control de pH se caracterizó por un progresivo cambio en el status
fisiológico de L. casei, sugiriendo una importante influencia del nivel de lactato producido
(~45 g/L) sobre la propia fisiología del microorganismo caracterizándose por un paulatino
aumento en las subpoblaciones dañada y muerta. Similar fenómeno ha sido descrito por
Rault et al. (2008) quienes han descrito la pérdida de la integridad de membrana de L.
delbrueckii ante concentraciones altas de lactato durante la fermentación. Por el contrario, la
falta de control de pH no influyo decisivamente sobre el status fisiológico de L. casei
manteniendo un robusto status fisiológico a lo largo de la fermentación a pesar de la acidez
(pH=3.5). Este patrón de comportamiento sugiere que L. casei exhibió una respuesta
tolerante frente a la acidez bajo estas condiciones a lo largo de la fase estacionaria, en la cual
no se detectó descenso alguno en la subpoblación metabólicamente activa. De hecho, se ha
descrito que la bajada en el pH intracelular no fue acompañada por una pérdida de la
viabilidad puesto que las bacterias ácido-lácticas desarrollan mecanismos de resistencia
para evitar la influencia del pH microambiental sobre la fisiología. Recientemente, Wu et al.
(2012) han descrito el mantenimiento de la funcionalidad de la membrana plasmática de L.

casei bajo condiciones de stress ácido. Así, L. casei desarrolla una respuesta tolerante al
stress mediante la transcripción de diferentes familias genes que regulan dicha respuesta
bajo estas condiciones microambientales (Broadbent et al., 2010). Asimismo, la presencia de
azúcares adicionales bajo condiciones de inanición puede suponer un incremento en la
viabilidad y/o mantenimiento de las funciones metabólicas de las bacterias ácido-láctica
durante prolongados periodos de almacenamiento (Chen et al., 2011).
En definitiva, el desarrollo de este sistema ha logrado la reducción del impacto final de
este residuo alimentario caracterizado por su alta concentración en carbohidratos
fermentables, constituyendo hasta la fecha la única aproximación para el aprovechamiento
biotecnológico de este residuo. De hecho, este residuo se ha mostrado como una matriz
adecuada para la producción de ácido láctico sin la utilización de pretratamientos asociados
a la sacarificación de residuos lignocelulósicos (John et al., 2009). A diferencia de otros
sistemas desarrollados con otros sustratos, las fermentaciones discontinuas aquí
desarrolladas fueron realizadas sin la adición de nutrientes extra durante el curso de las
mismas, enfatizando en el bajo coste del sistema aquí desarrollado. Asimismo, el empleo de
este residuo, una vez fermentado por L. casei y por lo tanto con altos niveles de ácido láctico,
puede ser de particular interés en la industria de alimentación animal ya que el empleo de
un suplemento alimentario con ácido láctico puede evitar las indeseables y recurrentes
diarreas por enterobacterias en lechones o terneros.
5.2. Producción biotecnológica de ácido lactobiónico
Como se ha venido resaltando a lo largo de la presente memoria, las múltiples y

novedosas aplicaciones del ácido lactobiónico en ámbitos tan diversos como el farmacéutico
o el alimentario han estimulado la búsqueda activa de nuevas metodologías relacionadas
con la producción del mismo. A diferencia de otros ácidos orgánicos como el succínico o el
láctico, la obtención microbiana de ácido lactobiónico no ha sido estudiada de forma profusa
hasta la fecha (en los subapartados 2.2 y 4.7 se pueden encontrar las comparaciones entre
los diferentes sistemas biotecnológicos desarrollados hasta la fecha con el objetivo de
producir este polihidroxi ácido). Indudablemente, uno de los objetivos fundamentales del
presente trabajo ha sido aportar información valiosa ante la falta de conocimientos relativa
a la producción biotecnológica de ácido lactobiónico. De igual forma, se ha intentado aportar
nuevos conocimientos que puedan ayudar en el avance tecnológico de dicho ácido orgánico.
Una síntesis de los principales hitos en la cronología histórica del ácido lactobiónico se
muestra en la Figura 5.5.
La producción biotecnológica de ácido lactobiónico es una metodología que en
comparación con la metodología tradicional (catálisis química) presenta múltiples ventajas,
entre las que destaca la escasa formación de subproductos durante la reacción (empleando
obviamente el microorganismo adecuado) a diferencia de muchos de los procesos de
catálisis química donde la selectividad es inferior al 100%, lo que significa la formación
paralela de subproductos durante el curso de la reacción.
A la hora de obtener ácido lactobiónico a escala de laboratorio, la catálisis enzimática
representa la estrategia adoptada más ampliamente. Sin embargo, esta estrategia necesita la
presencia de un adecuado aceptor de los protones liberados durante la oxidación (FAD).
Además, también precisa de laboriosas etapas previas necesarias para la solubilización y
purificación de los enzimas asociados a la membrana, caso de la oxidorreductosa de
Paraconiothyrium sp. KD-3 (Kiryu et al., 2008). Satory et al. (1997) han obtenido ácido
lactobiónico a través de una oxidación enzimática empleando como enzima a la glucosa-
deshidrogenasa purificada a partir de Zymomonas mobilis. Al igual que muchos trabajos, no
utilizan suero lácteo como sustrato en la bioconversión, otro inconveniente añadido es la
formación de otros metabolitos (caso del sorbitol) como consecuencia del empleo de un
enzima que presenta selectividad por diferentes sustratos. Estos investigadores han
conseguido un 44% de conversión del sustrato inicial, pero por el contrario también
obtienen una cantidad similar de sorbitol. Pedruzzi et al. (2011) también han utilizado una
estrategia similar empleando células de Z. mobilis previamente permeabilizadas. De hecho,
las enzimas contenidas en las células permeabilizadas de Z. mobilis biotransforman mezclas

de lactosa y fructosa, generando por lo tanto concentraciones equimolares de sorbitol como
subproducto de la reacción (Malvessi et al., 2012; Pedruzzi et al., 2011).
1889
Descubrimiento del
ácido lactobiónico
1947
(LBA) por Fischer y El Dr. Stodola
Meyer tras una descubre la capacidad
oxidación química. de Pseudomonas sp.
en la producción de
ácidos biónicos.
1960
El Dr. Bentley
confirma la oxidación 1988
de carbohidratos por
diferentes especies Desarrollo de la
de Pseudomonas. solución University of
Wisconsin por Belzer
y colaboradores.
1995
Los Drs. Van Scott y
Yu desarrollan la 1998
cosmética basada en
Solvay construye una
LBA en su compañía
planta industrial en
recientemente
Alemania con capacidad
creada, NeoStrata.
de producción de 1000 t
de LBA al año.
2002
Alianza entre Chr.
Hansen y Novozymes 2003
para crear y desarrollar Desarrollo de nuevos
LactoYIELD. Salida al biomateriales por parte
mercado en 2009. de Park et al. (2003).
Desarrollo de soportes
tisulares basados en LBA.
2007
Estudios de
funcionalización de
2011 nanopartículas y
Desarrollo de la sistemas de liberación
metodología de bio- controlada de fármacos
producción de LBA a con LBA.
partir de suero lácteo
con P. taetrolens.
Figura 5.5. Cronología temporal e histórica del desarrollo del ácido lactobiónico. Principales hitos en la
relativamente larga, intensa y a la vez desconocida historia del ácido lactobiónico.
Por lo tanto, la obtención biotecnológica del ácido lactobiónico surge como una
alternativa más que adecuada ante los innumerables inconvenientes del resto de estrategias
de producción. Sin embargo, hasta la fecha, en la bibliografía (véase subapartado 2.2) no se
emplea suero lácteo como sustrato de la bioconversión. Así, Miyamoto et al. (2000)
emplearon para la producción de ácido lactobiónico 50 g/L de lactosa inicial sobre la que
añadieron sales (KH2PO4 y MgSO4) junto con una fuente de nitrógeno (peptona). Estos
investigadores hallaron en sus investigaciones que la adición de 5-10 g/L de peptona como
fuente de nitrógeno al medio de cultivo promueve significativamente la producción de ácido
lactobiónico por parte de las cepas mutantes de Pseudomonas LS13-1, acortando además el
proceso de transformación a 60 h. Además, estos investigadores lograron rendimientos en
torno al 90% con una producción de hasta 290 g/L de ácido lactobiónico tras 155 h
empleando una metodología de alimentación en discontinuo (Miyamoto et al., 2000).
La obtención de ácidos aldobionicos por B. cepacia también ha sido abordada por
Meiberg et al. (1999). De hecho, han patentado un proceso biotecnológico con el fin de
obtener ácidos biónicos (maltobiónico y lactobiónico) a partir de disacáridos reductores
(lactosa y maltosa) utilizando como sustrato adicional sales minerales y diversas fuentes de
nitrógeno. Sin embargo, en este proceso presenta el inconveniente de utilizar un
microorganismo patógeno, así como el empleo de sustratos excesivamente complejos y
realmente costosos. Murakami y colaboradores también han llegado a conseguir la
bioconversion de 150 g/L de lactosa en ácido lactobiónico tras 24 horas de cultivo
empleando células en reposo de mutantes de B. cepacia (resistente a altas concentraciones
de lactosa) (Murakami et al., 2002, 2003, 2006).
P. taetrolens es una bacteria gram negativa aeróbica que se caracteriza por oxidar la
lactosa para dar ácido lactobiónico sin hidrólisis del enlace glucosídico. Se trata de una
especie originalmente aislada e identificada a partir de ovoproductos en proceso de
putrefacción (Levine y Anderson 1931; West, 2004). La capacidad oxidativa del
microorganismo fue descubierta a finales de los años 40 por Stodola y Lockwood (1947)
como miembros de la división de fermentación perteneciente a los Laboratorios Regionales
del Noreste (Illinois, USA). Estos investigadores llegaron en su momento a lograr un 77% de
bioconversión de la lactosa inicial utilizando una metodología de cultivo en matraz agitado.
De acuerdo con su origen, se puede aventurar que P. taetrolens es una bacteria que puede
sobre-expresar genes relacionados con el deterioro de diversas matrices o sustratos como
son los alimentos, y que asimismo posee la capacidad de oxidar determinados carbohidratos
como elemento evolutivo que le permite competir con otras bacterias presentes en el mismo
nicho. De esta forma se podría justificar la subida en los valores de pH en sustratos ricos en
nitrógeno que surgen como resultado de la liberación al medio de cultivo de determinados
compuestos básicos procedentes del metabolismo bacteriano.
La capacidad oxidativa de la lactosa llevada a cabo por P. taetrolens se ha puesto de
manifiesto con un screening oxidativo previo realizado sobre medio sólido. De hecho, esta
metodología ha puesto de manifiesto que la capacidad oxidativa no es inmediata, surge tras

el crecimiento de las colonias, tratándose, por lo tanto, de una capacidad asociada a la
entrada en fase estacionaria (véase subapartado 4.3). Como se observa en estas
bioconversiones, el ácido lactobiónico es un metabolito cuya producción por P. taetrolens no
está asociado al crecimiento y sí a la entrada en fase estacionaria, ya que durante la fase
activa de proliferación la bacteria no oxida la lactosa.
La influencia del tamaño de inóculo sobre el sistema de producción biotecnológica ha
sido uno de los parámetros estudiados. Así, se ha constatado el efecto positivo del tamaño
del inóculo sobre la producción de ácido lactobiónico para todas las bioconversiones
llevadas a cabo, tanto a pequeña escala en matraz Erlenmeyer como a nivel de biorreactor
(Figuras 5.6 y 5.7). Se ha constatado que un incremento en el tamaño de ínoculo indujo
mayores valores asociados al rendimiento y productividad debido al anticipo de la entrada
en fase de producción. Por el contrario, la tasa específica de crecimiento sí se ha visto
claramente reducida ante un incremento en el tamaño del inóculo. Estos resultados sugieren
que para tamaños de inóculo pequeños el margen existente para la proliferación celular es
muy amplio al existir mayor ratio sustrato/biomasa (evidenciado además por un
incremento significativo del valor máximo de salto del pH). Obviamente, el incremento en el
tamaño de inóculo provoca una reducción de la tasa específica de crecimiento puesto que los
requerimientos son mayores en términos de nutrientes y oxígeno disuelto. Así, para
tamaños de inóculos grandes, llega un momento en el que estos requerimientos son
mayores que la cantidad presente de nutrientes, reduciendo por tanto la tasa de
crecimiento. Igualmente, la cantidad de micronutrientes presentes en el suero lácteo
(vitaminas, sales minerales, aminoácidos, etc…) destinada al mantenimiento ha de ser
superior, con lo cual el sustrato destinado a la proliferación/crecimiento se reduce.
Durante un cultivo microbiano son varios los parámetros que influyen decisivamente
sobre la producción y rendimiento, se trata de parámetros que van desde la funcionalidad
celular (status fisiológico) del microorganismo, la cantidad de oxígeno disuelto en el medio
(influenciado por la agitación y la aireación), el pH o la riqueza de nutrientes del medio. El
aporte de oxígeno disuelto juega un papel decisivo en el sistema como se ha puesto de
manifiesto en el subapartado 4.4, ya que provee al microorganismo/sistema con la adecuada
demanda de oxígeno y más si se trata de un proceso oxidativo como el que se estudia en
parte de la presente memoria. Además, los estudios preliminares en matraz mostraron que
el sistema está limitado por la falta de aireación durante el proceso. De hecho, el aporte de
oxígeno es a menudo el factor limitante del crecimiento celular en matraz, ya que no se
incrementa la superficie de intercambio gaseoso y, en consecuencia, no se favorece la

transferencia de materia.
Figura 5.6. Relación entre el tamaño de inóculo y los diferentes parámetros del bioproceso a nivel de matraz.
Figura 5.7. Relación entre el tamaño de inóculo y los diferentes parámetros obtenidos del bioproceso tras el
cultivo en biorreactor.
Evidentemente, la concentración de oxigeno disuelto es un parámetro que cambia hasta

10 veces más rápido que la biomasa y la concentración de los sustratos. Este hecho sitúa al
aporte de oxígeno disuelto como la variable más importante a la hora de controlar y
optimizar bioproceso aeróbico como el aquí descrito. Por lo tanto, y dado que el pH también
es un parámetro que influye decisivamente en el proceso, se ha procedido a implementar la
bioconversión en biorreactor a diferentes niveles de agitación y aireación, además de
regular el pH del cultivo a 6,5 para evitar los posibles efectos inhibitorios por parte de la
acidez generada.
La constatación de que la disponibilidad de oxígeno disuelto constituye un parámetro

influyente en el bioproceso se ha puesto de manifiesto con la puesta en marcha de un cultivo
en iguales condiciones experimentales pero sin aireación. Así, como se ha reflejado en el
subapartado 4.4, en apenas 20 minutos se agota todo el oxígeno disuelto en el medio, una
circunstancia que impide la proliferación celular en los parámetros habituales a pesar de la
existencia de un leve crecimiento (véase subapartado 4.4). No se observa el habitual
incremento del pH durante el crecimiento en fase exponencial aún a pesar de la existencia
de un leve crecimiento. A la vista de la evolución del cultivo y de las condiciones (anoxia
completa), como cabría esperar los valores de producción de ácido lactobiónico y oxidación
de la lactosa no se modificaron a lo largo del mismo.
Tasas de agitación superiores a 350 rpm favorecen la formación de biomasa, mientras
que la producción de ácido lactobiónico se ve claramente mermada. Sorpresivamente esta
tendencia observada bajo altas condiciones de aporte de oxígeno a través de la agitación, no
se ha observado bajo tasas de aireación superiores a 1.5 Lpm. De hecho, un incremento en la
aireación no se tradujo en un incremento en la proliferación celular. De las distintas
variables microambientales estudiadas, se puede concluir que tanto el pH como la
disponibilidad de oxígeno disuelto juegan un papel decisivo sobre la buena evolución del
cultivo. Es importante tener en cuenta que la tasa de transferencia de oxígeno (OTR)
siempre se ha mantenido por encima de la demanda de oxígeno (OUR) como se observa en
la Figura 5.8.
Por lo tanto, se puede concluir que la adopción de estrategias operacionales con altos
niveles de agitación (500-1000 rpm) estimula el crecimiento y la proliferación celular, pero
por el contrario reduce la producción de ácido lactobiónico. Así, niveles medios de agitación
(350-500 rpm) y aeración (0.5-1.0 Lpm) se mostraron como los más idóneos desde el punto
de vista productivo (subapartado 4.4). Se ha comprobado igualmente que la presencia de un
periodo corto de disponibilidad de oxígeno limitado, favorece y estimula los parámetros
asociados a la producción de ácido lactobiónico (tanto el rendimiento, la entrada en fase de
producción como la productividad volumétrica). De hecho, la introducción de una fase de
microaerobiosis al final de la fase de crecimiento exponencial puede mejorar el
rendimiento, concentración y productividad dentro de determinados sistemas (Martínez et
al., 2010; Wieschalka et al., 2012). Es indudable que la ausencia total de aporte de oxígeno,
por el contrario, restringe tanto la proliferación celular como la producción de ácido
lactobiónico. La falta de oxígeno disuelto en el medio oxígeno inhibe por tanto la cadena
respiratoria, impidiendo tanto la generación de ATP como la oxidación del cofactor FADH2
involucrado en la deshidrogenación de la lactosa para formar ácido lactobiónico (véase

subapartado 2.2).
Figura 5.8. Comparación de los perfiles generados de biomasa y oxígeno disuelto (A), junto con los valores del
aporte de oxígeno disuelto y consumo netos (B), así como los coeficientes volumétricos de transferencia de
materia y el consumo específico (C) generados a lo largo de un bioproceso con una agitación de 350 rpm y una
tasa de aireación de 1 Lpm.
La industria biotecnológica ha reconocido de hecho la importancia de un detallado

conocimiento y caracterización de todo bioproceso asociado a la producción de metabolitos
de interés. Muchos de los bioprocesos están asociados a una heterogeneidad fisiológica que
puede constituir una verdadera limitación para el correcto desarrollo de un sistema
asociado a la producción de metabolitos de valor añadido (Lara et al., 2006). De hecho, las
aproximaciones reduccionistas representadas por las técnicas dependientes de cultivo
reflejan una visión reducida del status fisiológico de la biomasa celular (Hewitt y Nebe-von-
Caron, 2004) en contraposición con nuevas técnicas que permite conocer con exactitud el
estado saludable (healthy) de los microorganismos como la citometría de flujo multi-
paramétrica.
La evaluación a través de citometría de flujo del sistema biotecnológico aquí descrito, ha

revelado que tanto la heterogeneidad poblacional como la funcionalidad de las células de P.
taetrolens se ven claramente afectadas como respuesta a diferentes estímulos estresantes.
Es indudable que cambios microambientales, tales como la inanición o el aporte excesivo de
nutrientes externos pueden causar respuestas globales a nivel poblacional.
El estado fisiológico de una población bacteriana es clave a la hora de conseguir altas
eficiencias y rendimientos. Por lo tanto, es clave detectar la presencia de células doblemente
teñidas al inicio de un proceso de bioconversión, algo que se ha puesto de manifiesto con
imágenes obtenidas con microscopio láser confocal de células de P. taetrolens durante el
transcurso del proceso. Estos resultados demuestran la importancia del estado fisiológico
celular en el comportamiento general dentro del bioproceso. Así, la presencia de fases lag
sugiere un pobre status fisiológico intrínseco a los cultivos utilizados como inóculos de los
cultivos realizados en biorreactor. De hecho el predominio de células dañadas y muertas en
la fermentación inoculada con cultivos que presentan una edad fisiológica superior a 12-h
redujo drásticamente la productividad del sistema, obteniendo resultados pobres en
comparación con cultivos óptimos desde el punto de vista fisiológico (edad fisiológica de 12-
h). Macroscópicamente, la presencia de fases lag indica una mala adaptación del
microorganismo a las nuevas condiciones de cultivo. Descifrar la heterogeneidad fisiológica
subyacente asociada a inóculos con edad superior a 12-h ha permitido, en este caso,
optimizar y corregir los posibles errores operacionales que de otro modo serían inevitables
y para los cuales no habría una explicación coherente. La sobre-exposición de estos
precultivos a unas condiciones deletéreas (pH<3.4 a partir de 24 h) ha inducido una
respuesta fisiológica frente al stress ácido (Figura 5.9). Dichos datos fueron corroborados
igualmente a través de la visualización en microscopía de fluorescencia y a través de una
separación celular (fluorescence-activated cell sorting, FACS) y posterior cultivo en medio
sólido. De los resultados obtenidos, se puede inferir un 75% de las células dañadas
(correspondientes a un cultivo de 24-h) se corresponden con células viables no cultivables
(VBNC) (subapartado 4.5). A nivel celular, esta pérdida de viabilidad está asociada a
respuestas funcionales y estructurales con cambios irreversibles en la estructura de la
membrana o en la polarización de la misma (Figura 5.10). En definitiva, la heterogeneidad
fisiológica a nivel poblacional ha provocado una reducción en la funcionalidad celular, con la
consiguiente reducción en el rendimiento del bioproceso (Figura 5.11).
Figura 5.9. Diagrama que muestra grosso modo los cambios fisiológicos y estructurales en las células de P.
taetrolens bajo condiciones de pH libre. Como se observa, los cambios se asocian en primer lugar a una bajada
drástica en el pH tras la fase exponencial y a la posterior sobre-exposición bajo condiciones deletéreas (pH~3,4)
ya en fase estacionaria tardía (>24 h).
Figura 5.10. Daños funcionales y estructurales generados a nivel celular como respuesta frente a estímulos
estresantes.
El inadecuado pH microambiental ha provocado a su vez una respuesta inducida frente al

stress. De esta forma se ha podido conocer el origen de una reducción en los parámetros
asociados a la producción bajo diferentes condiciones operacionales como son el pH inicial
(Figura 5.12), el pH controlado en fase de producción (Figura 5.13) o durante todo el
proceso (Figura 5.14). En suma, el status funcional y la heterogeneidad fisiológica de P.
taetrolens son desde luego dos parámetros fundamentales y críticos en la producción de

ácido lactobiónico a partir de suero lácteo.
Figura 5.11. Relación directa entre la heterogeneidad fisiológica con el inadecuado pH operacional y la edad
fisiológica del inóculo.
Figura 5.12. Relación entre el pH inicial y los diferentes parámetros del bioproceso.
Figura 5.13. Relación entre el valor de pH controlado en fase de producción (pH-shift) y los diferentes
parámetros del bioproceso.
Tanto valores de pH inicial inferiores como superiores a 6,5 llevan consigo una reducción
en la productividad volumétrica, así como un incremento en la entrada en fase de
producción. Igualmente, la tasa específica de crecimiento ha encontrado en 6,5 el valor
óptimo de pH inicial (Figura 5.12).
En relación al valor más adecuado de pH controlado en fase de producción, se ha
constatado que 6,5 es el valor óptimo de operación en términos de productividad
volumétrica y rendimiento, a pesar de un incremento en la biomasa mantenida bajo valores
inferiores a 6,0 (Figura 5.13). Valores de pH controlado de 6,0-6,5 fueron encontrados como
los óptimos, así el empleo de valores fuera de este rango ha significado una reducción en la
tasa específica de crecimiento, la aparición de fases lag o un retraso en la entrada en fase de
producción (Figura 5.14).
Figura 5.14. Relación entre el valor de pH controlado (pH-stat) y los diferentes parámetros del bioproceso.
Tras la optimización de las diferentes variables, tanto el incremento de la productividad

volumétrica, así como la concentración producida de ácido lactobónico se han convertido en
los objetivos principales de la presente memoria. Por lo tanto, se ha procedido a
incrementar estos valores asociados a la producción a través de un cultivo discontínuo con
alimentación. Desde el desarrollo del cultivo en discontinuo con alimentación por Davey y
Johnson en 1953 para la producción de penicilina, esta estrategia ha sido ampliamente
adoptada por la industria como la solución para extender la fase de crecimiento/producción
con el fin de obtener mejoras en el rendimiento de sistemas centrados en la obtención vía
fermentativa de bio-productos tan diversos como ácidos orgánicos (Ding y Tan, 2006; Zhu
et al., 2011), antibióticos (Müller et al., 2012) o proteínas recombinantes tales como la
eritropoyetina (Soyaslan y Çalik, 2011). Se trata en definitiva de una metodología que
permite mejorar los parámetros de producción asociados a todo bioproceso. De hecho,

conocer la influencia de diferentes estrategias de alimentación sobre el sistema puede llevar
a la adopción de las estrategias más rentables desde el punto de vista productivo. Así, tasas
de alimentación superiores a 8.4 mL/h redujeron drásticamente el rendimiento y
productividad del sistema (subapartado 4.7). En este trabajo se ha desarrollado
adicionalmente una metodología eficiente y sencilla de co-alimentación que ha logrado
incrementar la concentración de lactosa disponible para la oxidación por parte de P.
taetrolens. Además, dicha estrategia ha logrado mantener la funcionalidad celular de P.
taetrolens a lo largo del proceso mediante el continuo aporte de nutrientes a tasas bajas. En
definitiva, se ha propuesto una eficiente alternativa de producción biotecnológica de ácido
lactobiónico que cumple los requisitos en cuanto a niveles de producción (80-100 g/L) se
refiere, que la industria considera como necesarios para proponer a cualquier sistema
biocatalítico de producción como alternativa factible a los sistemas basados en catálisis
química (Pollard y Woodley, 2007; Yang et al., 2007).
6. CONCLUSIONES
Conclusiones 217
6. CONCLUSIONES
El trabajo anteriormente expuesto permite inferir las siguientes conclusiones:
 El suero lácteo de yogurt se ha revelado como una adecuada fuente para la

producción de ácido láctico por L. casei a partir de una materia prima hasta ahora
considerada como un residuo y que se caracteriza por presentar altos niveles de
azúcares fermentables sin la necesidad de realizar pretratamiento alguno.
 L. casei metaboliza de forma jerárquica los azúcares presentes en el lactosuero de

yogurt. Los rendimientos y balances fermentativos han mostrado un consumo
simultáneo de la glucosa, lactosa y sacarosa. La evolución del contenido en azúcares
ha puesto de manifiesto que la glucosa y la sacarosa son metabolizadas de forma más
eficientemente, actuando como sustratos preferenciales en el catabolismo de L. casei
bajo las condiciones de cultivo ensayadas. Por el contrario, la mejora y control de las
condiciones operacionales no ha logrado incrementar la tasa de bioconversión de la
lactosa durante el proceso fermentativo. A pesar de la modificación de las
condiciones operacionales y de la presencia de extracto de levadura durante la
fermentación, no se han producido cambios en el patrón fermentativo homoláctico
de L. casei. A diferencia del resto de carbohidratos, el ratio de consumo de lactosa no
se ha visto afectado negativamente por la presencia de un ambientes más estresante,
menos controlado y pobre en nutrientes como es el caso de un proceso fermentativo
con pH libre y sin suplementación con extracto de levadura.
 Los parámetros operacionales y la metodología empleada durante la fermentación

acido láctica han influido decisivamente sobre el rendimiento productivo. De esta
forma, la producción de ácido láctico se ha visto afectada por la presencia de extracto
de levadura en el precultivo empleado como inóculo del proceso fermentativo final.
Asimismo, la regulación del pH y la presencia de nutrientes extra han favorecido un
incremento significativo en la cantidad de carbohidratos metabolizados por parte de
L. casei, quien presenta un consumo simultáneo y a la vez preferencial de los mismos
en el siguiente orden glucosa > sacarosa > lactosa.
 La monitorización del estado fisiológico de L. casei durante las fermentaciones ha

permitido establecer y determinar la presencia simultánea de hasta tres
218 Conclusiones
subpoblaciones distintas. Así, el incremento en la concentración de lactato

encontrado bajo condiciones de control de pH ha podido ser asociado con una
pérdida del status fisiológico de L. casei, mientras que un sistema sin control de pH
(pH~3.5) ha logrado mantener el nivel inicial de funcionalidad a lo largo de toda la
fermentación, sugiriendo una respuesta tolerante frente al stress por parte de L.
casei.
 El suero lácteo se ha revelado como una matriz adecuada para la producción

biotecnológica de ácido lactobiónico por parte de P. taetrolens, sugiriendo que este
microorganismo puede llevar a cabo una bioconversión oxidativa de la lactosa de
forma eficiente y competitiva. Asimismo, se ha descrito por primera vez cómo el
proceso de formación de ácido lactobiónico no está asociado al crecimiento. Mientras
que la fase de crecimiento exponencial de P. taetrolens se ha caracterizado por un
abrupto ascenso del pH del medio y por un agotamiento en el contenido del oxígeno
disuelto en el medio.
 Los datos obtenidos de la bioconversión a pequeña escala y en biorreactor han

mostrado que la eficiencia, el rendimiento y la productividad del bioproceso han
estado íntimamente ligados al tamaño de inóculo empleado. Así, una mayor cantidad
de biomasa al inicio del proceso logra mejores rendimientos en la bioconversión tras
promover la entrada en fase de producción bajo densidades celulares superiores,
consiguiendo asimismo una significativa reducción de la duración de la
fermentación.
 El papel del oxígeno disuelto se ha mostrado como determinante sobre el

rendimiento y la eficacia del bioproceso. Así, la disponibilidad plena de oxígeno
disuelto en el medio, bien a través de altas tasas de aireación (>1.5 Lpm) y/o
agitación (>500 rpm), ha implicado una reducción importante en los valores
producidos de ácido lactobiónico. Valores medios de aporte de oxígeno (350-500
rpm o 0,5-1 Lpm) se han mostrado como los óptimos en términos productivos.
 El sistema de producción de ácido lactobiónico por P. taetrolens lleva asociado una

heterogeneidad fisiológica subyacente que afecta decisivamente al rendimiento
productivo del sistema. Edades fisiológicas del inóculo superiores a 12-h reducen
Conclusiones 219
drásticamente la producción de ácido lactobiónico debido al predominio de células

dañadas y muertas durante el cultivo.
 La falta de control de pH (<4) unida al agotamiento en el contenido en oxígeno

disuelto (<1%) encontrado en matraz agitado ha llevado a la inducción de una
respuesta fisiológica frente al stress en P. taetrolens, caracterizada por la entrada
progresiva e irreversible en muerte celular. Esta heterogeneidad fisiológica,
caracterizada por la presencia de células dañadas y muertas, se ha mostrado como
clave en el correcto progreso del sistema productivo. Así, la presencia de células
dañadas y muertas ha implicado una reducción drástica en los valores asociados a la
producción de ácido lactobiónico.
 Se ha logrado determinar el valor de pH operacional óptimo con el fin de maximizar

el rendimiento del bioproceso. Así, estrategias con control de pH estricto (pH-stat) o
bien sólo en fase de producción (pH-shift) a 6.5, son las más adecuadas tanto desde
el punto de vista fisiológico como productivo. Igualmente, un valor de 6.5 como pH
inicial se ha mostrado como el más adecuado, puesto que valores inferiores a 5.5
inducen la entrada en fase de producción sin la necesaria etapa previa de
crecimiento.
 Se han alcanzado por primera vez niveles realmente atractivos para la industria
biotecnológica tanto en la concentración (180 g/L) como en el rendimiento (>90%),
empleando un residuo con escaso valor como sustrato en un sistema llevado a cabo
por un microorganismo no recombinante.
7. BIBLIOGRAFÍA
Bibliografía 223
7. BIBLIOGRAFÍA
Adler, P., Song, H.S., Kästner, K., Ramkrishna, D., Kunz, B., 2012. Prediction of dynamic
behavior of Pediococcus pentosaceus producing lactic acid from lignocellulosic sugars.
Biotechnol. Prog. 28, 623-635.
Affertsholt, T., 2007. Market developments and industry challenges for lactose and lactose
derivatives. Presentación en el IDF Symposium Lactose and its derivatives. 14-16 Mayo
2007.
Woodley, J.M., 2002a. Fed-batch bioconversion of indene to cis-indandiol. Enzyme
Microb. Tech. 31, 954-967.
Wooley, J.M., 2002b. Application of multi-parameter flow cytometry using fluorescent
probes to study substrate toxicity in the indene bioconversion. Biotechnol. Bioeng. 80,
239-249.
Woodley, J.M., 2003. Measurement of strain-dependent toxicity in the indene
bioconversion using multiparameter flow cytometry. Biotechnol. Bioeng. 81, 405-420.
Angermayr, S.A., Paszota, M., Hellingwerf, K.J., 2012. Engineering a cyanobacterial cell
factory for production of lactic acid. Appl. Environ. Microbiol. 78, 7098-7106.
Avery, S.V., 2006. Microbial cell individuality and the underlying sources of heterogeneity.
Nat. Rev. Microbiol. 4, 577-587.
Bailey, J.E., 1991. Toward a science of metabolic engineering. Science 252, 1668-1675.
Ben Amor, K., Breeuwer, P., Verbaarschot, P., Rombouts, F.M., Akkermans A.D.L., De Vos,
W.M., Abee, T., 2002. Multiparametric flow cytometry and cell sorting for the
assessment of viable, injured, and dead Bifidobacterium cells during bile salt stress.
Appl. Environ. Microbiol. 68, 5209-5216.
Bradford, M.M., 1976. A rapid and sensitive method for the quantification of microgram
quantities of protein utilizing the principle of dye-binding. Anal. Biochem. 72, 248-254.
Broadbent, J.R., Larsen, R.L., Deibel, V., Steele, J.L., 2010. Physiological and transcriptional
response of Lactobacillus casei ATCC 343 to acid stress. J. Bacteriol. 192, 2445-2458.
224 Bibliografía
Cao-Hoang, L., Marechal P.A., Lê-Thanh, M., Gervais, P., Waché, Y., 2008. Fluorescent probes
to evaluate the physiological state and activity of microbial biocatalysts: A guide for
prokaryotic and eukaryotic investigation. Biotechnol. J. 3, 890-903.
Cánovas, M., García, V., Bernal, V., Torroglosa, T., Iborra, J.L., 2007. Analysis of Escherichia
coli cell state by flow cytometry during whole cell catalyzed biotransformation for l-
carnitine production. Process Biochem. 42, 25-33.
Cavalheiro, J.M.B.T., de Almeida, M.C.M.D., Grandfils, C., da Fonseca, M.M.R., 2009. Poly(3-
hydroxybutyrate) production by Cupriavidus necator using waste glycerol. Process.
Biochem. 44, 509-515.
Chan, S., Kanchanatawee, S., Jantama, K., 2012. Production of succinic acid from sucrose and
sugarcane molasses by metabolically engineered Escherichia coli. Bioresour. Technol.
103, 329-336.
Chen, S., Ferguson, L.R., Shu, Q., Garg, S., 2011. The application of flow cytometry to the
characterization of a probiotic strain Lactobacillus reuteri DPC16 and the evaluation of
sugar preservatives for its lyophilization. LWT-Food Sci. Technol. 44, 1873-1879.
Choi, J.I., Lee, S.Y., 1999. High-level production of poly(3-hydroxybutirate-co-3-
hydroxyvalerate) by fed-batch culture of recombinant Escherichia coli. Appl. Environ.
Microbiol. 65, 4363-4368.
Choi, Y.J., Park, J.H., Kim, T.Y., Lee, S.Y. 2012. Metabolic engineering of Escherichia coli for the
production of 1-propanol. Metab. Eng. 14, 477-486.
Curran, K.A., Alper, H.S., 2012. Expanding the chemical palate of cells by combining systems
biology and metabolic engineering. Metab. Eng. 14, 289-297.
Demain, A.L., 2000. Small bugs, big business: the economic power of the microbe.
Biotechnol. Adv. 18, 499-514.
Demain, A.L., 2007. The business of biotechnology. Ind. Biotechnol. 3, 269-283.
Ding, S., Tan, T., 2006. L-lactic acid production by Lactobacillus casei fermentation using
different fed-batch feeding strategies. Process. Biochem. 41, 1451-1454.
Duan, S., Yuan, G., Zhao, Y., Li, H., Ni, W., Sang, M., Liu, L., Shi, Z., 2012. Enhanced
cephalosporin C production with a combinational ammonium sulfate and DO-stat
based soybean oil feeding strategy. Biochemical Eng. J. 61, 1-10.
Enfors, S.O., Jahic, M., Rozkov, A., Xu, B., Hecker, M., Jürgen, B., Krüger, E., Schweder, T.,
Hamer, G., O’Beirne, D., Noisommit-Rizzi, N., Reuss, M., Boone, L., Hewitt, C., McFarlane,
C., Nienow, A., Kovacs, T., Trägårdh, C., Fuchs, L., Revstedt, J., Friberg, P.C., Hjertager, B.,
Bibliografía 225
Blomstein, H., Hjort, S., Hoeks, F., Lin, H.Y., Neubauer, P., van der Lans, R., Luyben, K.,
Vrabel, P., Manelius, Å., 2001. Physiological responses to mixing in large scale
bioreactors. J. Biotechnol. 85, 175-185.
Erickson, B., Nelson, J.E., Winters, P., 2012. Perspective on opportunities in industrial
biotechnology in renewable chemicals. Biotechnol. J. 7, 176-185.
EuropaBio, ESAB, 2006. Industrial or white biotechnology. A driver of sustainable growth in
Europe. Working document from EuropaBio and ESAB for the Industrial
Biotechnology section of the European Technology Platform for Sustainable
Chemistry.
Gao, T., Wong, Y., Ng, C., Ho, K., 2012. L-lactic acid production by Bacillus subtilis MUR1.
Bioresour. Technol. 121, 105-110.
García-Ochoa, F., Gómez, F., 2009. Bioreactor scale-up and oxygen transfer rate in microbial
processes: an overview. Biotechnol. Adv. 27, 153-176.
García-Ochoa, F., Gómez, E., Santos, V.E., Merchuk, J.C., 2010. Oxygen uptake rate in microbial
processes: an overview. Biochemical Eng. J. 49, 289-307.
Hatti-Kaul, R., Törnvall, U., Gustafsson, L., Börjesson, P., 2007. Industrial biotechnology for
the production of bio-based chemicals-a cradle-to-grave perspective. Trends
Biotechnol. 25, 119-124.
Hermann, B.G., Blok, K., Patel, M.K., 2007. Producing bio-based bulk chemicals using
industrial biotechnology saves energy and combats climate change. Environ. Sci.
Technol. 41, 7915-7921.
Herrero, M., Quirós, C., García, L.A., Díaz, M., 2006. Use of flow cytometry to follow the
physiological states of microorganisms in cider fermentation processes. Appl Environ.
Microbiol. 72, 6725-6733.
Heo, J.H., Ananin, V., Kang, H.A., Rhee, S.K., Kim, C.H., 2008. Feeding strategies for the
enhanced production of recombinant human serum albumin in the fed-batch
cultivation of Hansenula polymorpha. Process. Biochem. 43, 918-924.
Hewitt, C.J., Nebe-von-Caron, G., 2001. An industrial application of multiparameter flow
cytometry: assessment of cell physiological state and its application to the study of
microbial fermentations. Cytometry. 44, 179-187.
monitor individual microbial cell physiological state. Adv. Biochem. Eng. Biot. 89, 197-
223.
Hewitt, C.J., Nebe-von-Caron, G., Nienow, A.W., McFarlane, C.M., 1999. The use of multi-
parameter flow cytometry to compare the physiological response of Escherichia coli
226 Bibliografía
W3110 to glucose limitation during batch, fed-batch and continuous culture

cultivations. J. Biotechnol. 75, 251-264.
Hewitt, C.J., Onyeaka, H., Lewis, G., Taylor, I.W., Nienow, A.W., 2007. A comparison of high
cell density fed-batch fermentations involving both induced and non-induced
recombinant Escherichia coli under well-mixed small-scale and simulated poorly
mixed large-scale conditions. Biotechnol. Bioeng. 96, 495-505.
IDF, International Dairy Federation, 2011. The World Dairy Situation 2010. Bulletin of
International Dairy Federation. 446.
IDFA, International Dairy Foods Association, 2008. Smart marketing 2008. Global beverage
trends… what is dairy´s role. New Orleans, Marzo 2008.
IEA, IEA Bioenergy-Task 42 Biorefinery Report, 2012. Bio-based chemicals: value added
products from biorefineries.
Jang, Y.S., Kim, B., Shin, J.H., Choi, Y.J., Choi, S., Song, C.W., Lee, J., Park, H.G., Lee, S.Y., 2012.
Bio-based production of C2-C6 platform chemicals. Biotechnol. Bioeng. 10, 2437-2459.
John, R.J., Anisha, G.S., Nampoothiri, M., Pandey, A., 2009. Direct lactic acid fermentation:
focus on simultaneous saccharification and lactic acid production. Bioresour. Technol.
27, 145-152.
Jung, Y.K., Lee, S.Y., 2011. Efficient production of polylactic acid and its copolymers by
metabolically engineered Escherichia coli. J. Biotechnol. 151, 94-101.
Kobayashi, M., Shimizu, H., Shioya, S., 2007. Physiological analysis of yeast cells by flow
cytometry during serial-repitching of low-malt beer fermentation. J. Biosci. Bioeng.
103, 451-456.
Koller, M., Bona, R., Chiellini, E., Grillo Fernandes, E., Horvat, P., Kutschera, C., Hesse, P.,
Braunegg, G., 2008. Polyhydroxyalkanoate production from whey by Pseudomonas
hydrogenovora. Bioresour. Technol. 99, 4854-4863.
Koln Paper, 2007. En route to the knowledge-based bio-economy. German Presidency of the
Council of the European Union.
Lara, A.R., Galindo, E., Ramírez, O.T., Palomares, L.A., 2006. Living with heterogeneities in
bioreactors: understanding the effects of environmental gradients on cells. Mol.
Levine ,M., Anderson, D.Q., 1931. Two new species of bacteria causing mustiness in eggs. J.
Bacteriol. 337-347.
Lidstrom, M.E., Konopka, M., 2010. The role of physiological heterogeneity in microbial
population behaviour. Nat. Chem. Biol. 6, 705-712.
Bibliografía 227
Looser, V., Hammes, F., Keller, M., Kovar, K., Egli, T., 2005. Flow-cytometric detection of
changes in the physiological state of E. coli expressing a heterologous membrane
protein during carbon-limited fedbatch cultivation. Biotechnol. Bioeng. 92, 69-78.
Martínez, I., Bennett, G.N., San, K.Y. 2010. Metabolic impact of the level of aeration during
cell growth on anaerobic succinate production by an engineered Escherichia coli
strain. Metab. Eng. 12, 499-509.
Meng, Y., Xue, Y., Yu, B., Gao, C., Ma, Y., 2012. Efficient production of L-lactic acid with high
optical purity by alcaliphilic Bacillus sp. WL-S20. Bioresour. Technol. 116. 334-339.
Müller, S., Harms, H., Bley, T., 2010. Origin and analysis of microbial population
heterogeneity in bioprocesses. Curr. Opin. Biotech. 21, 100-113.
Müller, S., Nebe-von-Caron, G., 2010. Functional single-cell analysis: flow cytometry and cell
sorting of microbial populations and communities. FEMS Microbiol. Rev. 34, 554-587.
Müller, J.M., Risse, J.M., Jussen, D., Flaschel, E., 2012. Development of fed-batch strategies for
the production of streptavidin by Streptomyces avidinii based on power input and
oxygen supply studies. J. Biotechnol. doi:10.1016/j.jbiotec.2012.10.021
Nampoothiri, K.M., Nair, N.R., John, R.P., 2010. An overview of the recent developments in
polylactide (PLA) research. Bioresour. Technol. 101, 8493-8501.
Navarro, J.M., Tormo, A., Martínez-García, E., 2010. Stationary phase in gram-negative
bacteria. FEMS Microbiol. Rev. 34, 476-495.
Nicolau, S.A., Gaida, S.M., Papoutsakis, E.T., 2010. A comparative view of metabolite and
substrate stress and tolerance in microbial bioprocessing: from biofuels and
chemicals, to biocatalysis and bioremediation. Metab. Eng. 12, 307-331.
Nyström, T., 2003. Conditional senescence in bacteria: death of the immortals. Mol.
Microbiol. 48, 17-23.
Nyström, T., 2005. Bacterial senescence, programmed death, and premeditated sterility.
ASM News. 71, 363-369.
OECD, 2007. OECD international futures project on “The bioeconomy to 2030: designing a
policy agenda”.
Oh, H., Wee, Y.J., Yun, J.S., Han, S.H., Jung, S., Ryu, H.W., 2005. Lactic acid production from
agricultural resources as cheap raw materials. Bioresour. Technol. 96, 1492-1498.
Ouyang, J., Ma, R., Zheng, Z., Cai, C., Zhang, M., Jiang, T., 2012. Open fermentative production
of l-lactic acid by Bacillus sp. strain NL01 using lignocellulosic hydrolyzates as low-cost
raw material. Bioresour. Technol. Doi: dx.doi.org/10.1016/j.biortech.2012.09.096
Papagianni, M., 2011. Organic acids. En: Moo-Young M, editor. Comprehensive
Biotechnology, vol. 1. Oxford: Elsevier Inc. 109-120.
228 Bibliografía
Panesar, P.S., Kennedy, J.F., Gandhi, D.N., Bunko, K., 2007. Bioutilisation of whey for lactic
acid production. Food Chem. 105, 1-14.
Park, I.K., Yang, J., Jeong, H.J., Bom, H.S., Harada, I., Akaike, T., Kim, S.I., Cho, C.S., 2003.
Galactosylated chitosan as a synthetic extracellular matrix for hepatocyte attachment.
Biomaterials 24, 2331-2337.
Parthuisot, N., Catala, P., Lemarchand, K., Baudart, J., Lebaron, P., 2000. Evaluation of
ChemChrome V6 for bacterial viability assessment in waters. J. Appl. Microbiol. 89,
370-380.
Patel, M.A., Ou, M.S., Harbrucker, R., Aldrich, H.C., Buszko, M.L., Ingram, L.O., Shanmugam,
K.T., 2006. Isolation and characterization of acid-tolerant, thermophilic bacteria for
effective fermentation of biomass-derived sugars to lactic acid. Appl. Environ.
Microbiol. 72, 3228-3235.
Pollard, D.J., Woodley, J.M., 2007. Biocatalysis for pharmaceutical intermediates: the future is
now. Trends Biotechnol 25, 66-73.
Quirós, C., Herrero, M., García, L.A., Díaz, M., 2007. Application of flow cytometry to
segregated kinetic modelling based on physiological states of microorganisms. Appl.
Environ. Microbiol. 73, 3993-4000.
Rault, A., Bouix, M., Béal, C., 2008. Dynamic analysis of Lactobacillus delbrueckii subsp.
bulgaricus CFL1 physiological characteristics during fermentation. Appl. Microbiol.
Reis, A., Lopes da Silva, T., Kent, C.A., Kosseva, M., Roseiro, J.C., Hewitt, C.J., 2005. Monitoring
population dynamics of the thermophilic Bacillus licheniformis CCMI 1034 in batch and
continuous cultures using multi-parameter flow cytometry. J. Biotechnol. 115, 199-
210.
Ryall, B., Eydallin, G., Ferenci, T., 2012. Culture history and population heterogeneity as
determinants of bacterial adaptation: the adaptomics of a single environmental
transition. Microbiol. Mol. Biol. R. 76, 597-625.
Salminen, S., von Wright, A., Ouwehand, A., 2004. Lactic acid bacteria. Microbiological and
functional aspects. 3ª ed. New York: Marcel Dekker.
Sauer, M., Porro, D., Mattanovich, D., Branduardi, P., 2008. Microbial production of organic
acids: expanding the markets. Trends Biotechnol. 26, 100-108.
Soetaert, W., Vandamme, E.J., 2010. The scope and impact of industrial biotechnology. En:
Industrial Biotechnology, sustainable growth and economic success. W. Soetaert y E.
Vandamme (Eds.). Wiley-VCH, Weinheim (Alemania).
Bibliografía 229
Soyaslan, E.S., Çalik, P., 2011. Enhanced recombinant human erythropoietin production by
Pichia pastoris in methanol fed-batch/sorbitol batch fermentation through pH
optimization. Biochemical Eng. J. 55, 59-65.
Tang, Y.J., Zhang, W., Zhong, J.J, 2009. Performance analyses of a pH-shift and DOT-shift
integrated fed-batch fermentation process for the production of ganoderic acid and
Ganoderma polysaccharides by medicinal mushroom Ganoderma lucidum. Bioresour.
Technol. 100, 1852-1859.
Tracy, B.P., Gaida, S.M., Papoutsakis, E.T., 2010. Flow cytometry for bacteria: enabling
metabolic engineering, synthetic biology and the elucidation of complex phenotypes.
Curr. Opin. Biotechnol. 21, 85-99.
Want, A., Thomas, O.R.T., Kara, B., Liddell, J., Hewitt, C.J., 2009. Studies related to antibody
fragment (Fab) production in Escherichia coli W3110 fed-batch fermentation
processes using multiparameter flow cytometry. Cytometry A. 75, 148-154.
West, T.P., 2004. Regulation of pyrimidine nucleotide formation in Pseudomonas taetrolens
ATCC 4683. Microbiol. Res. 159, 29-33.
Werpy, T., Petersen, G., eds., 2004. Top value added chemicals from biomass. US Department
of Energy.
Wieschalka, S., Blombach, B., Eikmanns, B.J., 2012. Engineering Corynebacterium glutamicum
for the production of pyruvate. Appl. Microbiol. Biotechnol. 94, 449-459.
Willke, T., Vorlop, K.D., 2004. Industrial bioconversion of renewable resources as an
alternative to conventional chemistry. Appl. Microbiol. Biotechnol. 66,131-142.
Wu, C., Zhang, J., Wang, M., Du, G., Chen, J., 2012. Lactobacillus casei combats acid stress by
maintaining cell membrane functionality. J. Ind. Microbiol. Biotechnol. 39, 1031-1039.
Yadav, V.G., De Mey, M., Lim, C.G., Ajikumar, P.K., Stephanopoulos, G., 2012. The future of
metabolic engineering and synthetic biology: Towards a systematic practice. Metab.
Eng. 14, 233-241.
Yang, S.T., Huang, H., Tay, A., Qin, W., De Guzman, L., San Nicolas, E.C., 2007. Extractive
fermentation for the fermentation of carboxylic acids. In: Yang ST, editor.
Bioprocessing for value-added products from renewable resources. Oxford: Elsevier
Inc. 421-446.
Yu, Z., Du, G., Zhou, J., Chen, J., 2012. Enhanced α-ketoglutaric acid production in Yarrowia
lipolitica WSH-Z06 by an improved integrated fed-batch strategy. Bioresour. Technol.
114, 597-602.
230 Bibliografía
Zhang, H., Zheng, Y., Liu, Q., Tao, X., Zheng, W., Ma, X., Wei, D., 2009. Development of a fed-
batch process for the production of anticancer drug TATm-survivin(T34A) in
Escherichia coli. Biochemical Eng. J. 43, 153-168.
Zhou, S., Causey, T.B., Hasona, A., Shanmugan, K.T., Ingram, L.O., 2003. Production of
optically pure d-lactic acid in mineral salts medium by metabolically engineered
Escherichia coli W3110. Appl. Environ. Microbiol. 69, 399-407.
Zhu, J., Thakker, C., San, K.Y., Bennet, G., 2011. Effect of culture operating conditions on
succinate production in a multiphase fed-batch bioreactor using an engineered
Escherichia coli strain. Appl. Microbiol. Biotechnol. 92, 499-508.
8. ANEXOS
Anexos 233
8. ANEXOS
8.1. Difusión de la tesis doctoral
8.1.1. Artículos científicos
 Alonso, S., Herrero, M., Rendueles, M., Díaz, M. 2010. Residual yoghurt whey for
lactic acid production. Biomass and Bioenergy 34, 931-938.
 Alonso, S., Rendueles, M., Díaz, M. 2011. Efficient lactobionic acid production from
whey by Pseudomonas taetrolens under pH-shift conditions. Bioresource Technology
102, 9730-9736.
 Alonso, S., Rendueles, M., Díaz, M. 2012. Role of dissolved oxygen availability on
lactobionic acid production from whey by Pseudomonas taetrolens. Bioresource
Technology 109, 140-147.
 Alonso, S., Rendueles, M., Díaz, M. 2012. Physiological heterogeneity of Pseudomonas

taetrolens during lactobionic acid production. Applied Microbiology and
Biotechnology 96, 1465-1477.
 Alonso, S., Rendueles, M., Díaz, M. 2012. Selection method of pH conditions to

establish Pseudomonas taetrolens physiological states and lactobionic acid
production. Applied Microbiology and Biotechnology. Aceptada, en prensa. doi:
10.1007/s00253-012-4607-x.
 Alonso, S., Rendueles, M., Díaz, M. Feeding strategies for enhanced lactobionic acid
production from whey by Pseudomonas taetrolens. Bioresource Technology. Aceptada.
 Alonso, S., Rendueles, M., Díaz, M. Bio-production of lactobionic acid: current status,
applications and future prospects. Enviada para su evaluación, bajo revisión.
234 Anexos
8.1.2. Comunicaciones a congresos
 Alonso, S., Herrero, M., Rendueles, M., Díaz, M. Whey fermentative processes with
industrial interest (póster).11th Mediterranean Congress of Chemical Engineering.
Barcelona (España). Octubre 21-24, 2008.
 Alonso, S., Herrero, M., Rendueles, M., Díaz, M. Monitorization of physiological

heterogeneity in a sugars-to-lactic bioprocess (comunicación oral, galardonada con
el 3er premio como mejor comunicación oral). 5th International Meeting on
Biotechnology BioSpain 2010. Pamplona (España). Septiembre 29-Octubre 1, 2010.
 Alonso, S., Rendueles, M., Díaz, M. Analysis of physiological states for bioreactor
scale-up (póster). The Bioprocessing Summit 2012. Boston (Estados Unidos). Agosto
20-23, 2012.
 Alonso, S., Rendueles, M., Díaz, M. Application of multi-parameter flow cytometry to

study the bioproduction of lactobionic acid (comunicación oral). 6th International
Meeting on Biotechnology BioSpain 2012. Bilbao (España). Septiembre 19-21, 2012.
 Alonso, S., Viña, J., Rendueles, M., Díaz, M. Effect of initial pH on shake-flask
cultivation of Pseudomonas taetrolens (póster). 6th International Meeting on
Biotechnology BioSpain 2012. Bilbao (España). Septiembre 19-21, 2012.
 Alonso, S., Rendueles, M., Díaz, M. Physiological states of Pseudomonas taetrolens

cultured in whey media under different oxygen and pH conditions (póster). 15th
European Congress on Biotechnology. Estambul (Turquía). Septiembre 23-26, 2012.
Anexos 235
8.2. Informe sobre el índice de impacto de los artículos de la tesis
Los artículos que conforman la presente memoria han sido publicados en revistas
incluidas en el Science Citation Index (Thomson Reuters), cuyos índices de impacto son los
siguientes:
Biomass and Bioenergy (2010)  3.84
Bioresource Technology (2011)  4.98
Applied Microbiology and Biotechnology (2011)  3.42

2013alonsobten PDF

Cargado por

Copyright:

Formatos disponibles

2013alonsobten PDF

Cargado por

Información del documento

Título original

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

2013alonsobten PDF

Cargado por

Copyright:

Formatos disponibles

Universidad de Oviedo

Departamento de Ingeniería Química y

Obtención de ácidos orgánicos por

Saúl Alonso Tuero

Al Prof. Mario Díaz le agradezco asimismo la libertad científica de la que he gozado en

Al Dr. Benjamín Paredes por su sapiencia, su curiosidad y su constante optimismo.

Quiero también particularizar el agradecimiento en varias personas. En primer lugar

En definitiva, quiero agradecer a todo el mundo que haya estado directamente o

1.1. Introducción ............................................................................................................................................................3

1.2. Objetivos ...................................................................................................................................................................7

1.3. Estructura de la memoria ..................................................................................................................................9

2. CONSIDERACIONES TEÓRICAS ......................................................................................................................... 13

2.1. Status actual de la producción biotecnológica de metabolitos microbianos ............................... 13

2.2. Producción biotecnológica de ácido láctico: status actual y aplicaciones .................................... 18

2.3. Producción biotecnológica de ácido lactobiónico: status actual y aplicaciones ......................... 23

3. MATERIALES Y MÉTODOS ................................................................................................................................... 81

3.1. Producción biotecnológica de ácido láctico ............................................................................................. 81

3.1.1. Microorganismo de cultivo ................................................................................................................... 81

3.1.2. Preparación del lactosuero de yogurt .............................................................................................. 81

3.1.3. Condiciones de cultivo ............................................................................................................................ 82

3.1.4. Métodos analíticos .................................................................................................................................... 82

3.1.5. Monitorización mediante citometría de flujo ................................................................................ 83

3.2. Producción biotecnológica de ácido lactobiónico .................................................................................. 86

3.2.1. Microorganismo de cultivo ................................................................................................................... 86

3.2.2. Preparación del suero lácteo ................................................................................................................ 86

3.2.3. Condiciones experimentales de cultivo ........................................................................................... 86

3.2.4. Métodos analíticos .................................................................................................................................... 87

3.2.5. Monitorización mediante citometría de flujo ................................................................................ 88

3.2.6. Separación celular .................................................................................................................................... 89

3.2.7. Microscopía confocal de fluorescencia ............................................................................................. 90

4.1. Producción de ácido láctico por Lactobacillus casei ............................................................................. 93

4.2. Monitorización fisiológica de Lactobacillus casei ............................................................................... 103

4.3. Producción de ácido lactobiónico por Pseudomonas taetrolens.................................................... 121

4.4. Influencia de la disponibilidad de oxígeno disuelto sobre la producción de ácido lactobiónico

4.6. Influencia del pH en la producción de ácido lactobiónico ................................................................ 155

5. DISCUSIÓN GENERAL .......................................................................................................................................... 195

5.1. Producción biotecnológica de ácido láctico .......................................................................................... 195

5.2. Producción biotecnológica de ácido lactobiónico ............................................................................... 203

6. CONCLUSIONES ...................................................................................................................................................... 217

7. BIBLIOGRAFÍA ........................................................................................................................................................ 223

8. ANEXOS ...................................................................................................................................................................... 233

8.1. Difusión de la tesis doctoral ......................................................................................................................... 233

8.1.1. Artículos científicos ............................................................................................................................... 233

8.1.2. Comunicaciones a congresos ............................................................................................................. 234

multiparamétrica, ha proporcionado el conocimiento preciso sobre el comportamiento

fermentation performance, productivity and efficiency, representing valuable information

El desarrollo de dos bioprocesos eficientes, efectivos, robustos y sostenibles para la

 Evaluar la capacidad de la fracción líquida residual procedente de productos lácteos

 Conocer el comportamiento fermentativo de L. casei en una matriz natural y

 Caracterizar y monitorizar los diferentes estados fisiológicos de L. casei durante la

 Desarrollar un sistema biotecnológico eficiente para la obtención de ácido

 Desarrollar protocolos de citometría de flujo multiparametrica como técnica que

 Proponer las condiciones óptimas de operación a escala de biorreactor con el fin de

1.3. Estructura de la memoria

La actual memoria se presenta como compendio de publicaciones que se enmarcan en la

como se podrá comprobar fácilmente todos comparten un mismo elemento central. Se ha

2.1. Status actual de la producción biotecnológica de metabolitos

En los últimos años, la producción biotecnológica de determinados componentes

de bio-productos obtenidos mediante fermentación (Figura 2.1). Es indudable que la