T41038 PDF
T41038 PDF
T41038 PDF
FACULTAD DE VETERINARIA
Sección Departamental de Tecnología de los Alimentos
TESIS DOCTORAL
PRESENTADA POR
Directores
Madrid
Ed. electrónica 2019
FACULTAD DE VETERINARIA
PRESENTADA POR
Madrid, 2018
Mª ISABEL CAMBERO RODRÍGUEZ, Profesora Titular de la Sección Departamental de
Tecnología de los Alimentos de la Facultad de Veterinaria de la Universidad
Complutense de Madrid, Mª ENCARNACIÓN FERNANDEZ VALLE y DAVID CASTEJÓN
FERRER, ambos Técnicos de Grado Superior de la Unidad de Resonancia Magnética
(CAI de Bioimagen Complutense) de la Universidad Complutense de Madrid
CERTIFICAN:
Que la Tesis Doctoral titulada “Aplicación de técnicas de Resonancia Magnética
Nuclear al estudio de miosistemas”, ha sido realizada bajo su dirección por Ana Belén
García García en la Sección Departamental de Tecnología de los Alimentos de la
Facultad de Veterinaria de la Universidad Complutense de Madrid. Asimismo,
autorizan su presentación para que sea juzgada por la comisión correspondiente para
optar al grado de Doctor.
A lo largo de estos años, muchas son las personas que han contribuido a la
realización de esta Tesis Doctoral.
Gracias a mis compañeros de laboratorio, Rosa, Lola y Carlos por los buenos
momentos y a mi amiga Karen que vivió conmigo esta experiencia desde el primer día
y con la que he compartido mucho más que trabajo.
Gracias a mis padres, Carmen y Rafa, por el apoyo y cariño que manifestáis día a día,
por haber dedicado todo vuestro esfuerzo a la educación de vuestros hijos, por haber
confiado en mí y haberme dado siempre libertad para elegir mi camino.
A Rubén, gracias por tu optimismo, por tu labor de coach, por recordarme que sí se
puede y por estar siempre a mi lado.
IV.2.7. Trabajo 7. Valoración de los exudados de cerdo como matriz apta para la
monitorización dela conservación de la carne tratada con electrones acelerados .............. 247
V. Discusión integradora ........................................................................................................... 269
V.1. Análisis de la estructura de derivados cárnicos ................................................................. 273
V.1.1. Aplicación de la IRM al estudio de la macroestructura de jamón curado .................. 273
V.1.2. Aplicación de la relaxometría T2 de RMN al estudio microestructural de distintas
matrices cárnicas ................................................................................................................... 276
V.1.2.1. Modificaciones derivadas del curado de piezas cárnicas enteras versus matrices
cárnicas picadas ................................................................................................................ 279
V.1.2.2. Efecto de la adición de proteasas ........................................................................ 283
V.1.3. RMN y características fisicoquímicas de derivados cárnicos ...................................... 285
V.1.4. RMN y comportamiento reológico de derivados cárnicos ......................................... 287
V.2. Análisis metabolómico de derivados cárnicos y exudados ................................................ 292
V.2.1. Derivados cárnicos ...................................................................................................... 292
V.2.2. Exudados cárnicos ....................................................................................................... 297
VI. Conclusiones / Conclusion ................................................................................................... 305
VII. Bibliografía .......................................................................................................................... 311
VIII. Índice de abreviaturas........................................................................................................ 323
Resumen / Abstract
1
Resumen
INTRODUCCIÓN
Las técnicas basadas en el principio de la Resonancia Magnética Nuclear (RMN) son
herramientas de análisis no destructivo que aportan valiosa información sobre las
matrices cárnicas, tanto a nivel estructural como a nivel bioquímico. Por un lado, se ha
estudiado el potencial la Imagen de Resonancia Magnética (IRM) para estudiar la
macroestructura de piezas musculares. Por otro lado, se ha evaluado la aptitud de la
relaxometría de RMN para el estudio y caracterización de la microestructura de
miosistemas cárnicos y, finalmente, se ha comprobado el potencial de la espectroscopía
de alta resolución de RMN para el estudio del perfil metabólico de muestras de músculo
íntegras y muestras de exudado. Estas metodologías permiten el análisis de matrices
complejas de forma rápida, sin apenas preparación de la muestra y su carácter no
invasivo y no destructivo aumenta su ámbito de aplicación.
PLANTEAMIENTO
Se ha empleado la RMN para el estudio de distintos miosistemas cárnicos. Por un
lado, se ha analizado un derivado cárnico elaborado a partir de una pieza entera, el
jamón curado. Por otro lado, se elaboraron distintos productos a partir de carne picada:
modelos cárnicos y embutidos crudos curados, tipo chorizo y tipo salchichón. Se planteó
un estudio longitudinal para monitorizar el proceso de maduración de estos productos,
realizando un seguimiento de los cambios estructurales (mediante IRM y relaxometría
de RMN) y bioquímicos (mediante espectroscopía de RMN) experimentados por la
matriz cárnica durante del tiempo de curado. Además, se empleó la espectroscopía de
RMN para el análisis de matrices líquidas procedentes de miosistemas cárnicos
(exudados) y evaluar su potencial como matriz para monitorizar el proceso de
conservación de la carne de cerdo.
OBJETIVOS
Se han abordado dos objetivos generales:
▪ Evaluar el potencial de las técnicas no destructivas de RMN (IRM y
relaxometría de RMN) para monitorizar el proceso de maduración de distintos
derivados cárnicos (jamón curado, modelos cárnicos y embutidos crudos
curados) a través de los cambios macro y microestructurales experimentados
por los citados miosistemas durante su elaboración.
3
Resumen
RESULTADOS
Se han desarrollado varias etapas en las que se han afrontado distintos objetivos
parciales, como seguidamente se indica. Los resultados obtenidos han dado lugar a siete
artículos incluidos en esta memoria.
4
Resumen
5
Resumen
6
Resumen
En este trabajo se presenta por primera vez el perfil metabólico de los exudados de
cerdo obtenido mediante la metodología 1H RMN HRMAS. Se propone emplear los
exudados de cerdo como matriz de análisis para identificar los procesos y las
trasformaciones habituales que tienen lugar durante el almacenamiento de la carne
sometida a un tratamiento de conservación basado en la aplicación de electrones
acelerados. Se observó que tanto la dosis del tratamiento como el tiempo de
almacenamiento de las muestras tratadas tienen un efecto notable sobre los
metabolitos cárnicos que pueden ser percibidos a partir de los datos espectrales
obtenidos. Además, la aplicación de herramientas quimiométricas al conjunto de datos
espectroscópicos puso de manifiesto cuáles fueron los metabolitos que experimentaron
variaciones significativas durante el almacenamiento de la carne. La clasificación de las
muestras de carne de acuerdo a su tiempo de almacenamiento se realizó a través del
7
Resumen
Análisis de Componentes Principales (PCA). Con este mismo procedimiento fue posible
la discriminación entre muestras tratadas y no tratadas con electrones acelerados.
CONCLUSIONES
Se concluye, de forma general, que las técnicas de RMN (IRM, relaxometría y
espectroscopía de alta resolución) presentan un gran potencial para monitorizar, desde
un punto de vista tanto estructural como bioquímico, los procesos de elaboración de
derivados cárnicos, como el jamón curado y los embutidos crudos curados, así como
procedimientos habituales en la industria alimentaria, como la conservación de
productos. Se confirma, por tanto, su aptitud para la trazabilidad y el control de calidad
de miosistemas cárnicos.
8
Abstract
INTRODUCTION
Nuclear Magnetic Resonance (NMR)-based techniques are non-destructive analytical
tools that provide valuable information about the meat matrices at both structural and
biochemical level. On the one hand, the potential of Magnetic Resonance Imaging (MRI)
to study the macrostructure of muscular pieces was studied. On the other hand, the
aptitude of T2 NMR relaxometry for the study and characterization of the microstructure
of meat myosistems was evaluated and, finally, the potential of the high-resolution NMR
spectroscopy for the study of the metabolic profile of intact samples of muscle and meat
exudates was verified. These techniques allow to analyze complex matrices quickly, with
minimum sample preparation and their non-invasive and non-destructive nature
increases their area of application.
EXPERIMENTAL DESIGN
NMR was used for the study of different meat myosistems. On the one hand, a meat
derivative elaborated from an entire piece, cured ham, was analyzed. On the other hand,
different products were elaborated from minced meat: meat models and raw cured
sausages, chorizo and salchichón type. A linear study to monitor the ripening process of
these products was outlined, in order to track the structural (by means of MRI and T2
NMR relaxometry) and biochemical changes (by means of NMR spectroscopy)
experienced by the meat matrices during processing. In addition, NMR spectroscopy was
used for the analysis of liquid samples coming from meat myosistems (exudates) and to
evaluate its potential as analytical matrix to monitor the conservation process of pork.
OBJECTIVES
Two general aims have been approached:
▪ To evaluate the potential of NMR non-destructive technologies (MRI and T2
NMR relaxometry) to monitor the ripening process of different meat
derivatives (cured ham, meat models and raw cured sausages) through the
macro and microstructural changes experienced by the mentioned
myosistems along their manufacturing process.
▪ To value high-resolution NMR spectroscopy aptitude for monitoring different
industrial processes using intact muscular samples, including ripening and
conservation, and samples of liquid nature (exudates).
9
Abstract
RESULTS
Different partial aims have been confronted along several stages, as indicated below.
The obtained results have led to seven articles included in this memory.
Paper 1. Use of MRI as a predictive tool for physicochemical and rheological features
during ham manufacturing.
In this work, MRI is described as predictive tool for the estimation of physicochemical
characteristics and textural parameters. This technique allowed the monitoring of the
manufacturing process of the product, revealing the principal macrostructural
modifications that take place in the myofibrilar matrix during ripening. For that purpose,
hams at different ripening times (raw, salted, post-salted, half-cured and cured) and
muscles with different anatomical location (biceps femoris, semimembranosus and
semitendinosus) were analyzed. T1, T2 and ADC parameters were obtained for each stage
and muscle and a progressive decrease was observed during ripening, related to the
dehydration kinetics and salt diffusion. Additionally, mathematical models were
established for the calculation of the water and salt content and rheological features
from the MRI parameters.
In order to delve into the structural modifications that occur throughout the
manufacturing process of dry cured ham, a T2 NMR relaxometry study was realized. This
technique allowed to separate the average values of T2 in several populations by
obtaining the constant distributions of the relaxation times. Thus, relevant information
was obtained at microstructural level. Three different T2 components were identified in
the meat samples, depending on the degree of association of the protons of the sample
with the protein matrix: T2b, T21, T22. T2b is the minority component with the shortest
relaxation time and it is related to the protons strongly associated with macromolecules;
T21 is the main component in the meat samples and it represents the protons of the
water retained by the myofibrillar proteins; finally, the T22 component is the population
representing the most mobile protons, with long relaxation times. In addition, a fourth
component (T2') was identified during the half-cured stage and it has been related to the
protons of the fat phase. The obtaining of scanning electron microscopy (SEM) images
10
Abstract
was very helpful to interpret the structural modifications. The SEM images analysis
allowed to observe the reduction of the intrinsic pore size of the meat matrix, due to
protein solubilization, absorption of salt and dehydration, typical transformations along
curing, as well as the visualization of the evolution of fat infiltration inside the muscle.
In addition, it was verified that T21 population is highly correlated to the breaking force
calculated by means of destructive uniaxial tensile test.
In this work, the proteolysis process that takes place during the ripening of meat
models elaborated from pork, particularly relevant to obtain the characteristics of the
final product, was studied. For that purpose, two batches of meat models were
elaborated, with (SMS) and without proteases (SMS+P), in order to study the differences
in the microstructure of both types of products by means of T2 NMR relaxometry. In
addition, electrophoretic analyses were conducted to compare the myofibrillar proteins
degradation pattern in both batches. The behavior of degradation was significantly
different (much more intense in the batch with proteases), as well as the distributions
of the T2 relaxation time obtained for each batch. In both cases, different water
populations (T2b, T21, T22) were identified. The mentioned populations experienced
changes during ripening, being reduced the relaxation times as the process progressed.
Structurally, models with proteases were characterized by a more fluid and much less
organized matrix, which was revealed by T2b and T22 populations.
T2 NMR relaxometry was used to study the microstructure of raw cured sausages,
recreating the typical conditions of a traditional process of production during a total of
14 days. Chorizo and salchichón type sausages with different levels of fat were analyzed
at several ripening times. Different water populations were obtained: T2b, T21, T22. The
progress of ripening was characterized by the microbial fermentation of the meat mass,
that generate changes in the myofibrillar matrix due to the decrease of pH and that have
11
Abstract
been related to the appearance of the T22 population, and the dehydration, which
marked the reduction of the relaxation times throughout the entire process. The texture
profile analysis (TPA) was also conducted to evaluate the structural evolution of the
meat matrix by a traditional method. The analysis of clusters revealed a clear
discrimination between samples according to their ripening time and Pearson's
correlation coefficient was used for establishing the degree of linear dependence
between physicochemical features-T2 parameters and T2- textural characteristics. The
main population T21 showed a stronger dependence, especially with water content
(r=0,93) and hardness (r=0,87), whereas other populations, T2b and T22, showed a
weaker association with the physicochemical and textural parameters.
1H RMN HRMAS (high resolution magic angle spinning) methodology was used to
monitor the changes in the metabolic profile of pork submitted to a traditional
manufacturing process. Principal components analysis (PCA) allowed to establish the
discrimination between samples and to group the obtained spectra depending on the
ripening stage. It was possible to identify a great quantity of metabolites. During cured
ham ripening an increase in the fatty acids signals, favored by the concentration of
macromolecules due to the dehydration, and to the enzymatic activity (lipolysis) was
reported. The degradation of nucleotides and the reduction of the carbohydrates
signals, due to sugars consumption was also observed. Additionally, an increase in the
signals corresponding to amino acids was detected, due to the proteolytic phenomena
involved in the development of the flavor, mainly remarkable in the aromatic amino
acids region, phenylalanine, tyrosine and tryptophan.
In order to monitor the biochemical changes happened during ripening and, thus, to
possess a global vision of the madurative changes, in this work the study of the
metabolic profile of salchichón type sausages by 1H RMN HRMAS spectroscopy was
performed. A great number of metabolites were identified, both meat metabolites and
12
Abstract
metabolites of addition, and it was possible to study their quantitative variation during
the stages of fermentation and ripening. Metabolomics allowed to separate the
processes of fermentation and ripening and the classification of the samples depending
on the ripening time by PCA.
Paper 7. Evaluation of E-beam irradiation and storage time in pork exudates using
NMR metabolomics
In this work, the metabolic profile of pork exudates was obtained by 1H RMN HRMAS
methodology for the first time. It proposes to use of exudates as a suitable matrix to
identify the processes and the habitual transformations that take place during the
storage of the meat treated with E-beam irradiation. It was observed that both the dose
of the treatment and the storage time of the treated samples have a notable effect on
the meat metabolites that can be perceived from the spectral information obtained. In
addition, the application of chemometric tools to the set of spectroscopic information
revealed which were the metabolites that experimented significant variations during the
meat storage. The classification of samples was performed according to the storage time
by PCA. The same procedure was used for the discrimination between samples treated
and not treated with E-beam irradiation.
CONCLUSIONS
NMR techniques (MRI, T2 relaxometry and high-resolution NMR spectroscopy) have
a great potential to monitor, from both a structural and biochemical point of view, the
manufacturing process of meat derivatives, cured ham and raw cured sausages, as well
as habitual procedures in the food processing industry, as the conservation of products.
NMR aptitude for the traceability and the quality control of meat myosistems is,
therefore, verified.
13
I. Introducción
15
La introducción de esta memoria se ha estructurado en cuatro apartados, en los que
se pretende abordar los distintos aspectos implicados en el desarrollo del trabajo
doctoral realizado. En un primer apartado se describen las particularidades de las
técnicas de Resonancia Magnética Nuclear que se proponen como herramientas de
análisis alternativas para el estudio de matrices alimentarias. En un segundo apartado
se ubica la industria cárnica española, a la que va orientado el estudio realizado, y se
trata la repercusión económica y características de este sector. La tercera parte se centra
en la carne de porcino y sus características. Finalmente, en la cuarta parte se abordan
las particularidades del proceso de elaboración de los productos cárnicos analizados:
embutidos y jamón curado. En conjunto, se pretende establecer el marco de trabajo y el
perfil de los objetivos propuestos.
17
I.Introducción
19
I.Introducción
20
I.Introducción
A comienzos de la década de los 80, el químico suizo Kurt Wütrich (ETH de Zúrich)
estudió la aplicación de la RMN a la elucidación estructural de proteínas, desarrollando
procedimientos generales para la asignación sistemática de las frecuencias de
resonancia de los diferentes núcleos magnéticos de estas biomoléculas y demostró que
a partir de esta información se podía deducir la estructura tridimensional de
determinadas proteínas (Wüthrich, 1974). Por ello, recibió el Premio Nobel de Química
en el año 2002.
En las últimas décadas, la RMN ha ido ganando protagonismo como técnica para la
elucidación estructural de moléculas aisladas y para el estudio del metabolismo, tanto
in vivo como in vitro. En la primera década del siglo XXI, se propone el término
Metabonómica (Fiehn, 2001) para hacer referencia a la medición cuantitativa de la
respuesta metabólica de naturaleza dinámica y multiparamétrica de los sistemas vivos
ante estímulos patofisiológicos o bien, ante la modificación genética. Desde entonces,
se ha trabajado para conseguir el aumento de la sensibilidad de la técnica gracias a la
utilización de campos magnéticos cada vez más intensos, a la combinación con técnicas
de polarización y al desarrollo de nuevas sondas de radiofrecuencia.
El gran potencial de análisis de las técnicas de RMN y sus ventajas frente a otras
técnicas analíticas han favorecido, en las últimas décadas, su incorporación a la Ciencia
y Tecnología de Alimentos. La RMN es una técnica rápida, no invasiva y no destructiva,
que no requiere preparación de la muestra (IRM) o es mínima (RMN). Es apta para el
análisis de matrices sólidas, semisólidas y líquidas, permitiendo obtener tanto el perfil
metabólico del alimento, como información molecular de forma cuantitativa. Además,
presenta un gran potencial para el estudio estructural de matrices alimentarias. En los
últimos años, la aplicación de esta técnica de alto rendimiento está permitiendo alcanzar
nuevos y ambiciosos objetivos en el análisis de alimentos (Laghi et al., 2014).
La RMN se basa en un fenómeno físico por el cual los núcleos atómicos con un
número impar de protones (Z) y/o un número impar de neutrones (N) (espín nuclear,
I≠0, no nulo) pueden absorber selectivamente energía generalmente en el rango de la
21
I.Introducción
Figura I.2. Secuencia de un proceso de RMN. A: orientación al azar de espines. B: aplicación de un campo magnético
B0 y orientación de espines. C: aplicación de un pulso de radiofrecuencia B1. D: absorción de energía. E: relajación
nuclear.
22
I.Introducción
23
I.Introducción
Después del pulso de radiofrecuencia todos los espines están en fase. Al retirar la
energía, con el tiempo, esta coherencia se va perdiendo, desfasándose progresivamente
debido a las diferencias en los campos magnéticos que perciben los núcleos (Figura I.6).
La relajación T2 es el resultado de interacciones entre los espines (relación +H -+H). Se
denomina también relajación spin-spin (Berendsen, 1992).
Figura I.6. Evolución de la proyección de los espines sobre el plano transversal después de un pulso de
radiofrecuencia.
24
I.Introducción
25
I.Introducción
que nos permitirá obtener una imagen de resonancia (Figura I.8). En este caso no se
adquiere la FID directamente ya que ésta desaparece muy rápidamente. Mediante el
uso de gradientes y pulsos de RF es posible reenfocar la señal de resonancia. La señal así
reenfocada se denomina eco. El conjunto de ecos se denomina espacio K y es la
representación del contenido en frecuencias espaciales del objeto.
Adquisición n ecos TF 2D
ECO
Espacio-k Imagen
Figura I.8. Formación de las imágenes de RM: adquisición de n ecos, construcción del espacio-k, obtención de la
imagen mediante una transformada de Fourier en 2 dimensiones.
Los alimentos son sistemas complejos tanto a nivel estructural como a nivel
bioquímico. Desde el punto de vista estructural, se pueden establecer distintos niveles
de organización. El modelo más general divide la estructura del alimento en
microscópica (en el rango de 0,1-100 m) y macroscópica (>100 m) (Ramírez et al.,
2009). Desde el punto de vista bioquímico, los alimentos están constituidos por un gran
y variado número de metabolitos. Ambas, la organización estructural y la composición
metabólica repercuten directamente sobre las características del producto final
(apariencia, textura, jugosidad, etc.) percibidas por el consumidor. El creciente interés
por ampliar conocimientos sobre la formación, estabilidad y composición de las
estructuras conlleva el empleo de nuevas técnicas analíticas. En este contexto, la imagen
de resonancia magnética (IRM) y la relaxometría son metodologías que pueden ser
empleadas para obtener información estructural, macro y microscópica,
26
I.Introducción
Como ya hemos visto, la IRM es una técnica a través de la cual las señales de RMN
son codificadas espacialmente utilizando gradientes de campo magnético y
reconstruidas en un formato de imagen (Liang & Lauterbur, 2000). La imagen de RMN
consiste en una matriz bidimensional de valores de intensidad (píxeles), que son las
señales de pequeños volúmenes dentro de la muestra (vóxeles). Cada píxel representa
la información tridimensional contenida en un vóxel (Figura I.9).
Figura I.9. Imagen de resonancia magnética de la sección interna de jamón. La imagen bidimensional consiste en
una matriz de píxeles; el valor numérico de cada píxel viene dado por la señal del elemento de volumen o vóxel. La
parte muscular y grasa se diferencian claramente (Fantazzini et al., 2009).
27
I.Introducción
28
I.Introducción
Para poder calcular los valores de los tiempos de relajación a partir de una serie de
imágenes, generalmente, se define un experimento en el que se fija uno de los
parámetros de adquisición, por ejemplo, TR o TE, y se hace variar otro, por ejemplo, TE
o TR. Así, se pueden obtener los mapas paramétricos. Los mapas son imágenes en los
que el valor de la intensidad del píxel es el valor del parámetro. Dependiendo del
parámetro que se haya calculado hablaremos de mapas del tiempo de relajación T1, T2,
mapas de DP, mapas de CDA, etc. El valor del parámetro se suele extraer como media
de todos los valores en una región de interés seleccionada [(region of interest (ROIs)]. A
diferencia de las imágenes, en los mapas, una intensidad mayor de la imagen
paramétrica siempre significa un valor mayor del parámetro (Tabla I.1).
Imagen Mapa
T1 1/T1 T1
T2 T2 T2
CDA 1/CDA CDA
DP [1H]
29
I.Introducción
I.1.3.2. Relaxometría
▪ Su frecuencia de resonancia.
30
I.Introducción
menos móviles, con mayor grado de inmovilización o ligazón, tendrán T2 cortos. A modo
de ejemplo, en la Figura I.10 se muestran las curvas de relaxometría que representan
los cambios en las componentes T2a (relajación rápida) y T2b (relajación lenta) en
muestras de queso manchego analizadas a distintos tiempos de maduración (Cruz et al.,
2014).
Figura I.10. Curvas de relaxometría registradas en un equipo de 200 MHz, a partir de muestras obtenidas a
diferentes tiempos de maduración durante el proceso de elaboración de queso manchego (Cruz et al., 2014).
Figura I.11. Representación de la secuencia CPMG. TE: tiempo de eco. : tiempo transcurrido entre el pulso de 90o y
el de 180o ( = TE/2).
31
I.Introducción
T1 y T2, características de un tejido trabajando a bajo campo, entre 0,5 y 1,5 Teslas, [RMN
de bajo campo o low-field NMR (LF-NMR)], o trabajando a campo alto [RMN de alto
campo o high-field NMR (HF-NMR)], con equipos de intensidad de campo magnético
superior a 1,5 Teslas.
32
I.Introducción
Figura I.12. Rangos de desplazamiento químico (ppm) para diferentes tipos de 1H, según su entorno químico.
33
I.Introducción
Figura I.13. Espectro 1H RMN de la quinina, alcaloide natural que se utiliza con frecuencia como muestra patrón en
RMN.
Figura I.14. Espectro 1H 1D de carne picada de cerdo en el que se representa la localización de los principales grupos
funcionales.
En la RMN 2D, los datos adquiridos dependen de dos variables de tiempo diferentes.
La primera continúa siendo el tiempo t2, que es el tiempo de detección habitual en los
experimentos de RMN 1D. La segunda es una nueva variable temporal, t1, tiempo de
evolución. Los experimentos 2D se basan, por tanto, en la adquisición de experimentos
34
I.Introducción
Figura I.15. Espectro bidimensional 1H-1H-COSY de exudado de carne de ternera tratada con 8kGy.
35
I.Introducción
Además del 1H, en las muestras es posible obtener información sobre otros núcleos
de interés (heteronúcleos) como el 13C, 15N o 31P que, en combinación con el 1H pueden
aportar nueva información estructural y contribuir a la elucidación estructural de la
muestra. Los experimentos 2D de correlación heteronuclear más habituales son:
36
I.Introducción
Preparación de la muestra
Asignación espectral
Datos cuantitativos
Análisis estadístico
37
I.Introducción
38
I.Introducción
39
I.Introducción
40
I.Introducción
41
I.Introducción
Por otro lado, es menos sensible que la MS, cuyo límite de detección es
sensiblemente menor.
42
I.Introducción
Figura I.20. Etapas del pre-procesado de datos de RMN y principales métodos quimiométricos de análisis.
43
I.Introducción
44
I.Introducción
La Figura I.21 muestra los gráficos de score plot y loading plot de un análisis de PCA
de los espectros de 1H RMN de muestras de exudado de merluza almacenada en hielo
entre 1 y 12 días. El análisis por PCA permite agrupar las muestras en función del tiempo
de almacenamiento e indicar los buckets más significativos en dicha agrupación
(Castejón et al., 2015).
45
I.Introducción
Figura I.21. Diagramas de scores y de loadings de un análisis de PCA (Castejón et al., 2015).
46
I.Introducción
Figura I.22. Análisis de PCA y PLS de mustras de suero procedentes de vacas sanas (rojo) y vacas enfermas (verde)
(Zhang et al., 2015)
47
I.Introducción
La industria cárnica es el cuarto sector productivo de nuestro país, sólo por detrás de
parcelas tan relevantes como la industria automovilística, la del petróleo y combustibles
y la de producción y distribución energética. El sector cárnico, formado por mataderos,
salas de despiece e industrias de elaborados, presenta en nuestro país un tejido
industrial constituido por unas 3.000 empresas, la mayoría en la categoría de pequeñas
y medianas, repartidas por toda la geografía nacional (ANICE, 2015). El 75% de las
industrias cárnicas españolas tienen menos de 20 empleados, según el Instituto Nacional
de Empleo (INE). Aunque su estructura está muy diseminada, la industria cárnica ocupa
con diferencia el primer lugar de toda la industria española de alimentos y bebidas,
representando una cifra de negocio de 22.000 millones de euros, más del 21% de todo
el sector alimentario, según datos publicados por el Ministerio de Agricultura, Pesca,
Alimentación y Medio Ambiente (MAPAMA, 2016a). Esta cifra de negocio supone
aproximadamente el 2 % del PIB total español, y en concreto el 14% del PIB de la rama
industrial. El empleo sectorial directo de las empresas cárnicas supera los 80.900
trabajadores, equivalente a más del 20% de la ocupación total de la industria alimentaria
española.
Por otra parte, la industria cárnica contribuye a paliar el déficit comercial de nuestro
país. Se trata del primer subsector exportador de la parcela agroalimentaria. En los
últimos años, se exportaron en torno a 1,7 millones de toneladas de productos frescos
y alrededor de 148.552 Tm de elaborados cárnicos, con un valor de más de 4.467
48
I.Introducción
millones de euros a distintos países del mundo, con una balanza comercial muy positiva
y en constante progresión, que ha pasado de una tasa de cobertura del 229% en 2010
al 375% en 2015. La industria cárnica es, por tanto, un sector estratégico que es
necesario proteger para conservar su posicionamiento internacional y su carácter
competitivo.
Como es sabido, España es uno de los países con mayor tradición en la elaboración y
consumo de una amplia gama de productos cárnicos, que engloba una extensa variedad
de embutidos y jamones, que forman parte del acervo cultural y gastronómico de cada
una de sus comunidades o regiones. Las extraordinarias características organolépticas
de estos productos cárnicos son la base de su gran aceptación tanto a nivel nacional
como internacional. El mercado actual, se caracteriza por la comercialización de
productos tradicionales de calidad, bien caracterizados y diferenciados, que conviven
con nuevos derivados cárnicos con los que se intenta cubrir las demandas del
consumidor actual y los requerimientos de su forma de vida. En el amplio elenco de
derivados cárnicos, sin duda, el producto más emblemático es el jamón curado, seguido
de otros productos entre los que cabe citar, lomo embuchado y embutidos curados
como chorizo, salchichón, fuet y salami además de fiambres y jamón cocido, galantinas,
bacon, morcillas, salchichas y elaborados para untar, como patés y sobrasadas. En los
últimos años, se ha modificado la forma de comercializar estos derivados cárnicos.
Mientras se ha reducido la venta de piezas enteras o troceadas en porciones, más o
menos grandes, se ha incrementado la oferta de productos loncheados listos para su
consumo o envasados en porciones pequeñas adaptadas al consumo individual o en
unidades familiares. Por otra parte, la industria cárnica está elaborando productos con
menor aporte calórico y con perfiles más saludables, adaptándose a los nuevos criterios
nutricionales.
49
I.Introducción
El sector porcino español tiene una importancia clave en la economía de nuestro país
ya que supone el 12,7% de la Producción Final Agraria. Dentro de las producciones
ganaderas, que representan en torno al 38% de la Producción Final Agraria, el sector
porcino ocupa el primer lugar en cuanto a su importancia económica alcanzando el
36,4% del total.
Tras varios años de evolución irregular del censo de porcino, como consecuencia de
las nuevas exigencias en materia de bienestar animal y de los vaivenes en los costes de
50
I.Introducción
Figura I.24. Evolución del número de animales sacrificados 2000-2015 (MAPAMA, 2016a).
51
I.Introducción
Por otro lado, España es uno de los países con mayor tradición en la elaboración y
consumo de derivados cárnicos. Lo diverso de la producción chacinera forma parte de
nuestro acervo cultural y gastronómico y es apreciada dentro y fuera de nuestras
fronteras.
Figura I.25. Importancia de los tipos de carne procesada sobre el total de carne procesada (MAPAMA, 2016b).
En España, el consumo per cápita de carne procesada durante el año 2015 fue de
8,25 kilos por persona y año. El 28,6% corresponde a fiambres siendo la ingesta media
aproximada de 2,36 kilos/persona/año. Le sigue el jamón curado y la paleta con un
consumo per cápita de 2,06 kilos/persona/año (MAPAMA, 2016b).
La importancia que tiene en España el consumo de derivados cárnicos ejerce una gran
influencia en todo el sector productivo y condiciona la mayor parte de las decisiones
técnicas y económicas. La producción de derivados cárnicos exige una materia prima
con atributos de alta calidad y características diferentes a las que requiere la
comercialización de la carne fresca, de ahí la necesidad de introducir mejoras en los
métodos de control y aseguramiento de la calidad, así como en los métodos y líneas de
producción.
52
I.Introducción
I.3. EL CERDO
La carne de cerdo proveniente de una buena crianza constituye un alimento nutritivo,
saludable, sabroso, muy equilibrado en su composición y por su relación calidad-precio
constituye uno de los productos cárnicos más consumidos por la población.
Estructuralmente, se considera un miosistema, por la importancia que tienen las
proteínas miofibrilares en la formación y consolidación de esta matriz cárnica. En este
apartado se abordan consideraciones generales de la carne y se detallan aspectos
relevantes relacionados con la estructura y composición de la misma.
Los esfuerzos encaminados a fijar las características deseadas de razas mejoradas han
conducido al reconocimiento de alrededor de 90 razas, con más de 200 variedades que
se han agrupado en tres tipos principales de cerdos: el céltico, el ibérico y el asiático.
Entre estos, la agrupación racial “tronco ibérico” con varias estirpes es la variedad
genuina de España (Suárez et al., 2002).
53
I.Introducción
En el animal vivo se distinguen, tanto por su estructura como por su función, tres
tipos de tejido muscular: el estriado esquelético, el músculo cardíaco y el músculo liso.
En la carne, la musculatura lisa sólo representa una pequeña proporción formando parte
fundamentalmente de los vasos sanguíneos, mientras que el músculo cardiaco se limita
al corazón. En consecuencia, esta memoria se referirá únicamente al músculo estriado
esquelético, que es el más importante cuantitativamente en la carne.
54
I.Introducción
lengua y los del diafragma y esófago, con o sin grasa, así como porciones de hueso, piel,
tendones, aponeurosis, nervios y vasos sanguíneos que normalmente acompañan al
tejido muscular y que no se separan de éste en las operaciones de preparación de la
carne.
55
I.Introducción
56
I.Introducción
La fibra muscular suele ser polinucleada. En la mayoría de las especies animales las
fibras rojas y blancas están más o menos mezcladas al azar en los haces musculares. Sin
embargo, en el cerdo, en la mayoría de los músculos, las fibras rojas se localizan en
forma de pequeños grupos cerca del centro de los fascículos y están rodeadas de fibras
blancas que forman la periferia de los haces (Aberle et al., 2012).
I.3.4.1. Agua
El contenido en agua de la carne de cerdo es, aproximadamente, del 75%, siendo muy
similar al de otras especies de abasto como pollo y vacuno (Primo, 1997; Fennema et al.,
2010). El agua de la canal se encuentra principalmente en el tejido muscular magro; el
tejido adiposo contiene poca agua. En conjunto se estima que en el tejido conectivo se
57
I.Introducción
58
I.Introducción
59
I.Introducción
total del músculo vivo, aunque en la carne, debido a los cambios postmortem,
puede ascender a más de un 15%. Se trata del agua que se encuentra fuera
de las células, en espacios macrocapilares y, por su movilidad, se considera
agua extramiofibrilar. La liberación del agua del espacio sarcoplásmico se
encuentra frenada por el comportamiento y la integridad del sarcolema. La
mayor cantidad de agua se encuentra en el interior de las miofibrillas,
atrapado en el retículo tridimensional constituido por los miofilamentos,
especialmente de actina y miosina. Éste es el agua embebida, agua de los
microcapilares o agua fijada a los miofilamentos (hasta 10 g de agua/g de
proteína). Según su movilidad, este tipo de agua se corresponde con el agua
intramiofibrilar.
2%
5%
13%
10%
agua de constitución
agua de interfase
agua extramiofibrilar
agua intramiofibrilar
agua libre
70%
Como se muestra en la Figura I.28, alrededor del 95% del contenido acuoso de la
carne se encuentra como agua libre o agua de movilidad reducida. El agua de estas
fracciones es congelable, puede disolver solutos e intervenir en diversas reacciones. Su
liberación de la masa cárnica exige una energía variable, que puede valorarse
sometiendo al sistema a distintas fuerzas; de esta forma, puede determinarse su
60
I.Introducción
61
I.Introducción
I.3.4.2. Proteínas
62
I.Introducción
I.3.4.3. Lípidos
La cantidad de grasa de un animal y, por tanto, de la canal, varía con la raza, el sexo,
la edad, la dieta, o el peso, entre otros factores (Marusic et al., 2013). El contenido en
lípidos de un determinado músculo es variable; su concentración está relacionada
inversamente con el contenido en agua. La mayoría de los lípidos se localizan en los
63
I.Introducción
depósitos de tejido adiposo y en el caso de los músculos (Bello, 2008), asociados a los
septos del tejido conectivo laxo que se encuentran entre los haces musculares (grasa
intramuscular), formando lo que se conoce como veteado o marmorización. Estos
lípidos dan jugosidad a la carne y, además, se comportan como un aislante, que permite
que la carne infiltrada pueda ser sometida a mayores tratamientos térmicos con mínima
pérdida de calidad.
La mayor parte de los ácidos grasos del material lipídico corresponde a los
triglicéridos, aunque hay pequeñas cantidades de mono y diglicéridos (Prändl et al.,
1994). La composición en ácidos grasos, además de ser importante para la consistencia,
influye en la calidad sensorial. Los porcentajes de grasa en carne de cerdo oscilan entre
un 17% y un 36%, correspondiendo el porcentaje de ácidos grasos más elevado a los
monoinsaturados (51%) seguido de los saturados (35%) y los poliinsaturados (13%). Por
ácidos grasos individuales los porcentajes más elevados corresponden al ácido oleico
(C18:1) (aproximadamente 48%), seguido del ácido palmítico (C16:11) (23%) y del ácido
esteárico (C18:0) (10.6%) (Piñeiro et al., 2001). Los fosfolípidos son componentes
esenciales de las membranas celulares, participan en la regulación del metabolismo
celular y su contenido es relativamente constante (0,8-1,0%) en los tejidos magros
(Bello, 2008). Estos lípidos estructurales contienen una proporción considerable de
ácidos grasos poliinsaturados, lo que resulta clave en la alteración química de la carne
por ser en estos ácidos grasos donde se inicia con mayor facilidad el enranciamiento
autooxidativo de los lípidos, originando cambios en el aroma, el color y el flavor de la
carne (Skiba et al., 2012).
I.3.4.4. Carbohidratos
64
I.Introducción
degradado, de la raza del animal o del tipo de músculo. En definitiva, la carne contiene
cierto porcentaje de ácido láctico, junto con cantidades vestigiales de algunos
metabolitos con estructura de azúcares, como glucosa, glucosa-6-fosfato y otros
azúcares-fosfatos (Bello, 2008).
65
I.Introducción
66
I.Introducción
Tabla I.3. Semáforo nutricional de distintos embutidos crudo-curados y de jamón curado según una ración de
consumo que va desde los 35 a los 60 gramos, en función del producto considerado.
Fuente: Tablas de composición de alimentos del CESNID (Farrán et al., 2003)
Indica la proporción aportada respecto de la cantidad Diaria Orientativa (CDO) que una persona adulta necesita
ingerir de cada nutriente: 2.000 kcal, 90 g de grasa, 70 g de saturadas, 20 g de azúcares, 6 g de sal y 24 g de fibra. El
semáforo nutricional se basa en un sistema de colores: verde-baja cantidad (la aportación es del 7,5% de la CDO),
amarillo-cantidad media (entre el 7,5% y el 20%) y naranja-cantidad alta (más del 20%). La fibra no tiene color
porque escasea en la dieta.
67
I.Introducción
cabra) ha sido clasificada como Grupo 2A, probablemente cancerígena para los seres
humanos. Dicha clasificación se basa en una evidencia limitada procedente de estudios
epidemiológicos que muestran una asociación positiva entre el consumo de carne roja
y el desarrollo de cáncer colorrectal. La carne procesada (carne que ha sido
transformada a través de la salazón, el curado, la fermentación, el ahumado, u otros
procesos para mejorar su sabor o su conservación) fue clasificada como Grupo 1,
cancerígeno para los seres humanos. Esta categoría se utiliza cuando hay suficiente
evidencia de carcinogenicidad en humanos. Los expertos afirman que el consumo de 50
gramos de productos cárnicos al día supone incrementar el riesgo de cáncer colorrectal
en un 18%, sin haber especificado qué tipo de productos ni el método de elaboración
utilizado. Esta evaluación del CIIC refuerza la recomendación de 2002 de la OMS acerca
de moderar el consumo de carne para reducir el riesgo de cáncer colorrectal. Algunas
otras directrices sobre la dieta también recomiendan limitar el consumo de carne roja o
carne procesada, pero éstas se centran principalmente en la reducción de la ingesta de
grasa y sodio, que son factores de riesgo para las enfermedades cardiovasculares y la
obesidad.
68
I.Introducción
I.4.1. GENERALIDADES
69
I.Introducción
70
I.Introducción
Según el citado R.D., los productos objeto de la presente Tesis Doctoral (jamón,
chorizo y salchichón curados) se incluyen en el grupo de derivados cárnicos no tratados
por calor, curado-madurados. Integran este grupo los productos sometidos a procesos
tecnológicos no térmicos, de salazón y de curado-maduración, suficiente para
71
I.Introducción
I.4.3.1. Definición
72
I.Introducción
No todos los jamones curados de cerdo blanco producidos en España son iguales, ya
que sus características dependen de diversos factores relativos al animal, tipo de
crianza, alimentación, edad de sacrificio y proceso de elaboración. Estos factores
proporcionan al producto final unas características físicas y sensoriales peculiares y
diferenciales.
73
I.Introducción
I.4.3.3.3. Salado
Esta etapa consiste en la incorporación de las denominadas sales del curado,
compuestas por cloruro sódico, sales nitrificantes (nitratos y nitritos) y coadyuvantes del
curado (azúcares y ascorbatos). La finalidad de esta operación es potenciar el sabor,
crear condiciones desfavorables para la multiplicación de la microbiota alterante y
74
I.Introducción
Los perniles están en contacto directo con las sales durante un tiempo que varía en
función de su tamaño. En la actualidad, la relación suele ser de 0,7 día/kg de peso. La
temperatura durante el salado puede considerarse como un factor regulador de la
duración del mismo. En las instalaciones actuales, se mantiene siempre por encima de
1 oC, para que no cristalice el agua, y por debajo de 3 oC, para impedir el crecimiento de
Clostridium botulinun. La humedad relativa de las cámaras de salado se mantiene en
torno a 90%, o incluso superior, para impedir en la medida de lo posible, la evaporación
del agua que, por ósmosis, afloraría desde el interior de los perniles.
I.4.3.3.5. Secado-maduración
Esta fase es la más larga del proceso, aunque su duración es variable, desde meses
en los cerdos de raza blanca hasta más de un año en los cerdos ibéricos, y es crucial para
el desarrollo del sabor y aroma merced a reacciones de degradación (proteólisis,
lipólisis, y oxidaciones) y síntesis (generación de moléculas sápidas de bajo peso
molecular y volátiles). Además, en esta etapa se estabiliza el color y se produce una
75
I.Introducción
I.4.4.1. Definición
Los embutidos son derivados cárnicos, generalmente constituidos por carne picada,
contenidos en una tripa o envoltura artificial (R.D. 474/2014).
Dentro de este grupo, los chorizos y salchichones son derivados cárnicos constituidos
por trozos de carne y grasa no identificables anatómicamente que, con carácter general
y no limitativo, se han sometido a un proceso de picado más o menos intenso, mezclados
con especias, ingredientes, condimentos y aditivos, embutidos en tripas naturales o
envolturas artificiales, y sometidos a un proceso de salazón seguido de curado-
maduración, acompañado de fermentación, suficiente para conferirles las
características organolépticas propias y su estabilidad a temperatura ambiente. Se les
añade pimentón, a los chorizos, y pimienta, a los salchichones, como ingrediente
caracterizante (R.D. 474/2014).
76
I.Introducción
festivas, culturales y hasta religiosas con más tradición en las sociedades rurales,
mediante la cual, las familias se abastecían de una reserva de carne para todo el año.
Por ello, son derivados cárnicos cuya producción está muy extendida por el territorio
nacional y casi todas las Comunidades Autónomas tienen sus embutidos característicos.
Como ocurre con el jamón curado, distintos chorizos y salchichones son productos
con marca de calidad garantizada, destacando las I.G.P.:
▪ Chorizo de Cantimpalo
▪ Chorizo Riojano
▪ Salchichón de Vic
77
I.Introducción
78
I.Introducción
Figura I.32. Efecto de la adición de sal sobre la CRA (Albarracín et al., 2011). PI: punto isoeléctrico.
79
I.Introducción
neutraliza por los aniones Cl-, cerrando los espacios interproteicos y, por tanto,
reduciendo la CRA (Albarracín et al., 2011) (Figura I.32).
I.4.2.4.3.2.1.2. Difusión
Este proceso es especialmente importante en derivados cárnicos elaborados a partir
de piezas musculares enteras, como el jamón curado, en cuya elaboración la etapa de
salado es fundamental para el desarrollo de las características del producto final. En
estos casos, la mayoría de estudios realizados muestran que el movimiento de la sal en
el interior de las piezas cárnicas se produce mediante fenómenos de difusión (Gou &
Comaposada, 2000), influenciados por la diferencia de concentraciones entre el exterior
y la superficie, y por el área de carne que permanece en contacto con la sal (Carrasco,
2001).
En las piezas cárnicas, el proceso de difusión será diferente según las características
de la zona de la carne por la que la sal penetra. El cloruro sódico difunde a través de la
fase acuosa líquida del músculo (Gou & Comaposada, 2000), pero en alimentos
sometidos a procesos de secado esta fase no permanece inmóvil, sino que se mueve
hacia el exterior y va arrastrando la sal en sentido opuesto al del agua. Otro factor
implicado en la penetración de la sal al interior del producto es el contenido en materia
seca del músculo y su composición, que van a determinar tanto la cantidad, como el
recorrido de la sal, e incluso va a actuar directamente sobre ella, fijándola a la estructura
de la proteína.
80
I.Introducción
I.4.4.3.2.3. Azúcares
Los azúcares fermentables (glucosa o sacarosa), se adicionan como sustrato
energético de los microorganismos que intervienen en el proceso de acidificación. El
descenso de pH puede ser más o menos pronunciado según el tipo y la cantidad de
azúcar adicionado. La glucosa es asimilada rápidamente por casi todos los
microorganismos, pero también se utilizan lactosa, sacarosa u otros azúcares menos
asimilables como el almidón o dextrinas para regular la velocidad de la acidificación
(Lücke, 1998). El descenso de pH tiene varios efectos beneficiosos, entre los que cabe
mencionar:
81
I.Introducción
I.3.4.3.2.4. Ascorbato
Es una sustancia antioxidante y coadyuvante del proceso de curado que ayuda a
disminuir la cantidad de nitrito residual puesto que favorece su transformación a óxido
nítrico con lo que, además, mejora la formación y estabilización del color. La legislación
española, R. D. 142/2002, permite la adición de ácido L-ascórbico o ascorbato sódico en
dosis quantum satis, habitualmente 0,5 g/Kg.
I.3.4.3.2.5. Especias
Son ingredientes vegetales con carácter aromático que se utilizan habitualmente en
pequeñas cantidades para conferir determinados sabores, aromas y colores a los
productos cárnicos. Además de sus propiedades aromáticas, muchas especias son
antioxidantes y antimicrobianas. Su empleo varía según el tipo de producto, pero, en
general, las más utilizadas son: pimentón, ajo, clavo, nuez moscada, cilantro, cebolla,
orégano, pimienta y tomillo.
82
I.Introducción
I.4.4.3.3. Embutido
Tras mezclar todos los ingredientes, se procede a embutir o introducir la masa en las
tripas para constituir las piezas del producto cárnico. Las tripas pueden ser naturales o
artificiales. Las tripas naturales se han utilizado tradicionalmente empleando tracto
intestinal de ovino, porcino y vacuno. Sin embargo, el proceso de obtención y
preparación de la tripa es muy laborioso por lo que son más costosas que las artificiales
y generalmente sólo se utilizan para productos de mayor calidad como el lomo
embutido. Las tripas artificiales de celulosa o colágeno son las más empleadas.
83
I.Introducción
Para llevar a cabo esta operación, la industria suele utilizar embutidoras continuas a
vacío que evitan la presencia de aire en la masa, reduciendo considerablemente la
posibilidad de que aparezcan defectos de aspecto y consistencia en el producto. Durante
la operación de embutido es recomendable que la temperatura de la masa cárnica no
sea superior a 1 o 2 oC para evitar fenómenos de embarrado, es decir, un
desprendimiento excesivo de grasa por rotura de adipocitos, lo cual confiere a la pasta
un aspecto pálido y grasiento.
I.4.4.3.4. Fermentación
Los embutidos se cuelgan a continuación en cámaras expuestos al aire con o sin
sistemas de control de la temperatura y humedad relativa (HR). En este proceso, la
temperatura puede variar entre 12 y 25 oC, y la HR se mantiene en el intervalo 90-95%
durante un periodo de tiempo que puede oscilar entre 24 y 72 horas.
La microbiota inicial, sin considerar la adición de cultivos iniciadores, suele ser muy
variada y similar a la presente en la carne fresca, incluyendo lactobacilos, micrococos,
enterobacterias, Pseudomonas spp, Achromobacter spp., Flavobacterium spp, Bacillus
spp., etc., hongos y levaduras (Ordoñez et al., 1999). Sin embargo, la conjunción de
diversos factores como la adición de las sales del curado, la reducción de la aw, los
efectos inhibitorios de nitratos y la baja tensión de oxígeno, favorece el establecimiento
de las condiciones óptimas para la inhibición del crecimiento de microorganismos
alterantes (psedomonas principalmente) y favorece el desarrollo de los deseables, de
manera que en los embutidos elaborados de forma tradicional, sin adición de cultivos
iniciadores, la microbiota típica está constituida por BAL y Micrococcaceae (Ordoñez et
al., 1999).
84
I.Introducción
I.4.4.3.5. Curado
Una vez finalizada la etapa de fermentación, los embutidos pasan por una etapa de
curado, donde empieza el proceso de maduración y, simultáneamente, el secado del
producto. Esta etapa implica el mantenimiento de los embutidos durante periodos de
tiempo variables en condiciones de Hr y temperatura controladas (Chang et al., 1996).
Los procedimientos más habituales suelen consistir en 5-10 días a 18-22 oC y HR entre
el 80-90% y, posteriormente, se mantienen a 12-15 oC y una HR del 65-80% (Ordóñez &
de la Hoz, 2001). La duración de este último periodo del curado es variable, en función
de la clase de producto y su diámetro, pero suele oscilar entre unos 15 (curado rápido)
y 90 (proceso tradicional) días (Flores, 1997).
85
I.Introducción
libres, contribuye, tanto al sabor y al aroma del producto final, como a su desecación,
pues acelera la pérdida de agua (DeMasi et al., 1990).
Por otra parte, la lipólisis también continúa en la fase de maduración. Los ácidos
grasos libres generados sufren diversas actividades oxidativas que conducen a la
aparición de sustancias volátiles y no volátiles que contribuyen al sabor y aroma del
embutido. Por último, cabe destacar que, durante esta fase, tiene lugar la completa
reducción del nitrito residual (Incze, 1992).
86
I.Introducción
Uno de los efectos más importantes que tiene el nitrito en los productos cárnicos
crudos curados es la formación del color típico. Este proceso tiene lugar a través de una
secuencia compleja de reacciones que conducen a la formación de nitrosomioglobina,
que imparte el característico color rojo violáceo a estos productos (Figura I.33).
87
I.Introducción
Figura I.33. Grupo hemo de la mioglobina (izquierda) y esquema de las reacciones del curado para la formación del
color (derecha). MbFe2+, mioglobina nativa; MbFe2+O2, oximioglobina; MMbFe3+, metamioglobina; MbFe2+NO,
nitrosomioglobina; AR, agentes reductores. Adaptado de: Skibsted, 1992.
88
I.Introducción
La evolución del contenido y composición del NNP va a estar condicionada tanto por
la variación de temperatura, como por el grado de secado del producto. El incremento
de la concentración se verá favorecido en las fases en las que la temperatura sea mayor.
Sin embargo, a la vez que el aumento de temperatura favorece la proteólisis, también
va a favorecer la desecación del producto, circunstancia en la que se verá disminuida la
actividad enzimática (Sánchez-Molinero & Arnau, 2010).
89
I.Introducción
I.4.5.2.2. Proteólisis
La proteólisis es un fenómeno de naturaleza enzimática, que consiste en la actuación
sucesiva de distintas enzimas proteolíticas sobre las proteínas cárnicas (Figura I.34).
Figura I.34. Esquema general de la proteólisis en productos cárnicos curados (Toldrá, 2006).
Las principales enzimas musculares que se asocian con la hidrólisis de las proteínas
son endoproteasas y exopeptidasas (Toldrá & Aristoy, 2010). A la acción hidrolítica de
las enzimas musculares hay que sumar la actividad proteolítica de los microorganismos
presentes en los productos curados, especialmente de las BAL. Aunque se desconoce el
papel que las enzimas proteolíticas juegan en el desarrollo del flavor de los productos
madurados, sí está claro que contribuyen a los cambios proteolíticos que suceden
90
I.Introducción
durante la maduración, sobre todo durante los primeros días, cuando dichas bacterias
hidrolizan proteínas a péptidos y aminoácidos para su utilización (Ordóñez et al., 1999).
Las calpaínas son cisteína endopeptidasas de origen muscular. Están formadas por la
calpaína I, la calpaína II y la calpastatina, proteína que inhibe de modo específico la
actividad de las anteriores (Koomaraie, 1994). Estas enzimas sólo actúan en la carne
cruda y en el primer estadio del periodo de maduración de productos cárnicos curados
(Ordóñez & De la Hoz, 2001). Estas enzimas son muy inestables, perdiendo su actividad
a los 10-14 días de procesado. Las calpaínas son capaces de hidrozilar proteínas como la
titina, la nebulina, las troponinas T e I, la tropomiosina y la desmina (Koohmaraie, 1994).
91
I.Introducción
92
I.Introducción
De forma general, se puede afirmar que los lípidos de los productos crudos-curados
son precursores de muchas sustancias aromáticas, que se forman como resultado de
fenómenos lipolíticos y oxidativos, durante el proceso de maduración, que van a tener
un importante papel en el desarrollo de las características sensoriales finales (Ordóñez
et al., 1999).
I.4.5.3.1. Lipólisis
La lipólisis consiste en la hidrólisis enzimática de los lípidos musculares o del tejido
adiposo y tiene como resultado la generación de ácidos grasos libres. Las enzimas
lipolíticas, tales como lipasas, esterasas, fosfolipasas y lisofosfolipasas, están
involucrados en los procesos de esta naturaleza que tienen lugar durante la maduración
de la carne y de los productos cárnicos (Ordóñez et al., 1999).
Pueden distinguirse dos tipos de lipólisis, según los lípidos afectados estén localizados
en el interior del músculo o en el tejido adiposo. La grasa muscular se compone
esencialmente de triglicéridos y fosfolípidos. En la lipólisis muscular, estos compuestos
son los sustratos de las lipasas (ácida lisosomal, neutra y básica) y las fosfolipasas (tipo
A1, A2, C y D), que reaccionan de forma hidrolítica (Figura I.35) liberando ácidos grasos
de estas moléculas (Motilva et al., 1993).
Figura I.35. Mecanismo de acción de las lipasas (izquierda) y las fosfolipasas (derecha) musculares (Mora, 2010).
93
I.Introducción
contenido de ácidos grasos libres desde un 2% del contenido lipídico total en el estado
inicial, a un 8%-20%, al final de su maduración (Gandemer, 2002). La actividad de las
lipasas neutra y básica es más importante durante los primeros 3-4 meses del proceso
de elaboración. Posteriormente, su actividad disminuye de forma progresiva, aunque la
lipasa ácida ejerce una baja actividad a lo largo de todo el procesado (Gandemer, 2002).
Los ácidos grasos que se liberan durante la fase de maduración de los productos
curados van a ser muy susceptibles de ser oxidados (Estévez et al., 2009), dando lugar a
diversos compuestos derivados de la oxidación (hidroperóxidos, epóxidos, furanos,
etc.), que actúan como precursores de un gran número de compuestos volátiles (Toldrá,
94
I.Introducción
95
I.Introducción
Tabla I.4. Productos específicos de la degradación de los ácidos oleico, linoleico y linolénico (Ordóñez y de la Hoz,
2001)
96
I.Introducción
Por todo ello, los resultados aquí obtenidos constituyen interesantes aportaciones a
la Tecnología Alimentaria, ahondan en el potencial de las técnicas basadas en la
resonancia magnética para el análisis de alimentos, y afianzan la importancia de estas
herramientas en el ámbito industrial.
97
II. Objetivos
99
II. Objetivos
España es uno de los países con una tradición más rica en la elaboración y consumo
de los más variados embutidos y jamones. Lo diverso de la producción, que se extiende
a todos los rincones de nuestro país, forma parte de nuestro acervo cultural y
gastronómico, y es apreciada dentro y fuera de nuestras fronteras.
En la presente Tesis Doctoral se abordan dos líneas de trabajo. Por una parte, se ha
estudiado el proceso de maduración de distintos productos derivados del cerdo blanco,
elaborados a partir de pieza entera (jamón curado) y elaborados a partir de carne picada
(modelos cárnicos y embutidos crudos curados) desde un punto de vista estructural.
Para ello se ha empleado la Imagen de Resonancia Magnética (IRM), con el fin de
estudiar los cambios en la macroestructura asociados al proceso, y la relaxometría T2 de
RMN, para estudiar la microestructura de los derivados cárnicos. Por otra parte, el
proceso de maduración de los mencionados derivados se ha estudiado también desde
un punto de vista bioquímico mediante la espectroscopía de 1H de alta resolución, que
permite la identificación y cuantificación de metabolitos representativos. Además, esta
técnica se ha empleado también para valorar la aptitud de los exudados liberados a
partir de la carne de cerdo como nueva matriz de análisis para estudiar los procesos y
los cambios metabólicos que tienen lugar durante la aplicación de un tratamiento de
electrones acelerados de la carne fresca de cerdo. Para afrontar las dos líneas de trabajo,
se plantean dos objetivos principales:
El objetivo principal del estudio estructural fue evaluar el potencial de las técnicas
no destructivas de RMN para monitorizar el proceso de maduración de distintos
productos cárnicos curados a través de los cambios macro y microestructurales
experimentados por los citados miosistemas durante su elaboración.
Para abordar este objetivo general de esta parte de la Tesis Doctoral se han
considerado los siguientes objetivos parciales:
101
II. Objetivos
Para abordar este objetivo general de esta parte de la Tesis Doctoral se han
considerado los siguientes objetivos parciales:
102
II. Objetivos
103
III. Material y métodos
105
III. Material y métodos
107
III. Material y métodos
109
III. Material y métodos
110
III. Material y métodos
Los perniles frescos (10-11 kg), así como las piezas de jamón en distintas etapas del
proceso de curado, analizados en este estudio fueron suministrados por la empresa
cárnica Incarlopsa (Tarancón, Cuenca). Todas las piezas procedían de hembras de
cerdos blancos grasos (Landrace × Large White) habitualmente utilizados para la
elaboración de jamón curado en la mencionada industria. El procedimiento seguido,
así como los momentos de toma de muestra (pernil fresco, salado, postsalado, secado
y bodega), y las determinaciones físicoquímicas realizadas se resumen en la Figura III.1.
Perfilado Sangrado
NaCl
NO2
Pernil salado Salado
NO3
después de la adición de sal Tº: 1-3 oC azúcares
Humedad: 90% ascorbatos
Tº: 4 oC
Pernil post-salado Post-salado
Humedad: 75-80%
tras la retirada de sal
Jamón curado
12 meses de bodega Tº: 12-17 oC
Maduración Humedad: 60-80%
Jamón curado
20 meses de bodega
Figura III.1. Principales etapas de la elaboración de jamón curado y análisis realizados. TPA: Análisis del Perfil de
Textura, ES: Extracto seco, MEB: Microscopía Electrónica de Barrido.
111
III. Material y métodos
Figura III.2. Localización anatómica de los músculos estudiados y procedimiento seguido para la toma de muestras.
SM:semimembranoso. BF: bíceps femoral; ST: semitendinoso.
Utilizando como base carne magra (Longissimus lumborum) molida en una picadora
dotada de una placa de control de granulometría con orificios de 5 mm, se
confeccionaron dos tipos de modelos cárnicos:
i. SMS (sistema modelo sin enzimas adicionadas) elaborado mezclando (%, p/p)
con la mencionada carne picada, grasa (19%), sales de curado [NaCl (2,5%),
KNO3 (0,02%), NaNO2 (0,01%)], dextrosa (0,8%), lactosa (1%) y dextrina (1,6%).
ii. SMS+P (sistema modelo adicionado con proteasas), elaborado como el modelo
anterior pero caracterizado por la adición de un combinado de proteasas. Se
utilizó pronasa E de Streptomyces griseus y aspartil proteinasa de Aspergillus
oryzae (300 y 100 U / Kg de carne, respectivamente). Las mencionadas
proteasas fueron suministradas por Sygma-Aldrich. Los detalles de la
elaboración de los dos modelos cárnicos se recogen en la Figura III.3.
Como embutidos crudos curados, se elaboró salchichón (S) y chorizo (CH) siguiendo
un procedimiento tradicional, sin utilizar cultivos iniciadores (Ordóñez et al. 1999;
112
III. Material y métodos
Toldrá et al., 2015). Para la fabricación de dichos productos se utilizaron dos niveles de
grasa: 20% (1) y 40% (2), de manera que se fabricaron cuatro lotes de productos: S1,
S2, CH1 y CH2. El esquema de elaboración de estos productos se recoge en la Figura
III.4.
113
III. Material y métodos
114
III. Material y métodos
115
III. Material y métodos
116
III. Material y métodos
117
III. Material y métodos
III.3. METODOLOGÍA
En la Figura II.6 se detalla gráficamente el planteamiento conjunto del trabajo
integrado en la presente Tesis Doctoral, así como las experiencias que constituyen
cada uno de los artículos o unidades publicables en los que se estructura.
ANÁLISIS DE MACRO Y
MICROESTRUCTURA
DE PRODUCTOS
ANÁLISIS DEL PERFIL
CÁRNICOS
METABÓLICO DE
DISTINTAS MATRICES
CÁRNICAS
118
III. Material y métodos
119
III. Material y métodos
120
III. Material y métodos
121
III. Material y métodos
122
III. Material y métodos
Figura III.6. Representación gráfica (fuerza-tiempo) de un análisis de perfil de textura (TPA). A: área; D: distancia.
123
III. Material y métodos
Para la realización de este ensayo, se emplearon unas pinzas A/MTG (Figura II.7)
fijadas, por una parte, a la base del texturómetro y, por otra, al brazo móvil del equipo,
conectado con la célula de carga. Se ensayaron probetas con una longitud de 7 cm y
una anchura entre 0,2 y 0,5 cm. Éstas se colocaron entre los dos extremos de las
pinzas, de forma que quedasen lo más estiradas posibles, y se inició la tracción a una
velocidad de 1 mm s-1 hasta la rotura. Se empleó una célula de carga de 5 Kg y las
medidas se realizaron a temperatura ambiente. Se realizaron unos 10 ensayos por tipo
de muestra.
124
III. Material y métodos
125
III. Material y métodos
El siguiente paso fue la evaporación (a vacío), que consiste en recubrir las muestras
con grafito. Para este procedimiento, se utilizó un evaporador Balzers SCD‐004 Sputter
Coater y el recubrimiento con grafito se realizó en un equipo Mini deposition System
MED 010 Balzers.
126
III. Material y métodos
Por último, se procedió a la metalización de las muestras con oro (Emitech K550x
Sputter Coater, 25 mA, 4 minutos), para su correcta visualización en el microscopio
electrónico de barrido (JEOL mod. JSM 6400, Japón). Las muestras utilizadas no son
buenas conductoras, por lo que es necesario hacer conductora la superficie, eliminar la
electricidad estática, minimizar el daño por radiación y aumentar la reflectividad
electrónica.
127
III. Material y métodos
128
III. Material y métodos
129
III. Material y métodos
Figura III.9. Espectrómetro Bruker BIOSPEC 47/40 utilizado en el análisis de IRM y de relaxometría T2 de alto campo.
En los experimentos de IRM, para las medidas del tiempo de relajación espín-espín
o transversal (T2), se obtuvieron 60 imágenes a diferentes tiempos de eco (TE = 5-300
ms). El resto de parámetros permanecieron constantes (NA = 1; TR = 3046,1 ms; MTX =
256 x 128, FOV = 5 x 2,5 cm; grosor del corte= 1 mm; número de corte = 1).
Los datos así obtenidos se ajustaron a una exponencial de acuerdo con la ecuación:
donde S(TE) es la señal de la imagen a cada tiempo de eco (TE) y S0 es la señal cuando
TE es igual a cero.
130
III. Material y métodos
Los datos así obtenidos se ajustaron a una exponencial de acuerdo con la ecuación:
Para las medidas del Coeficiente de Difusión Aparente (CDA) las imágenes se
adquirieron con seis valores de ponderación de la difusión (valor b = 48,4-54969,9 s
mm-2). El resto de parámetros de adquisición que permanecieron constantes entre los
diferentes experimentos fueron: (NA = 1; TR = 2000 ms; TE = 45 ms; Δ = 25ms; δ = 10
ms; MTX = 256 x 128; FOV = 5 x 2,5 cm; grosor del corte = 1 mm; número de cortes =
1).
Los datos así obtenidos se ajustaron a una exponencial de acuerdo con la ecuación:
131
III. Material y métodos
[90xo--180yo-) n-adquisiciones]
con un valor (tiempo entre el pulso de 90o y el pulso de 180o) de 150 s. Los datos
adquiridos resultaron de la suma de 16 scans. El tiempo de repetición entre dos
adquisiciones sucesivas fue 2 s. Las muestras cárnicas (4 cm de longitud, 1 cm de
diámetro y aproximadamente 5 g de peso) se depositaron en un tubo cilíndrico de
vidrio (5 cm de longitud, 14 mm de diámetro) que se introdujo directamente en el
equipo. Antes del ensayo, las muestras se atemperaron a 25 oC en un baño de agua
durante unos 15-20 minutos. Los datos de T2 obtenidos se analizaron con el programa
MatLab, versión 6,5, utilizando scripts internos, según el algoritmo de regularización
propuesto por Butler et al., (1981). Las distribuciones de ajuste exponencial se
muestran en una gráfica que refleja la intensidad de la relajación vs tiempo de
relajación. Se realizaron seis medidas de cada producto y tiempo de maduración.
132
III. Material y métodos
Se utilizaron dos equipos distintos para llevar a cabo los experimentos de RMN de
alta resolución, 1H RMN HRMAS (high resolution magic angle spinning).
Todos los espectros fueron procesados utilizando el software Topspin, versión 2.1.
A continuación, se ajustó la fase y la línea base de los espectros y se referenciaron al
TSP δ = 0 ppm.
Figura III.11. Espectrómetro Bruker AMX 500 MHz equipado con una sonda HRMAS (high resolution magic angle
spinnig) para el análisis de muestras semi-sólidas.
133
III. Material y métodos
600,13 MHz, equipado con una sonda de HRMAS (Figura III.12). Una solución de D2O
(15μL) junto con TSP (0,1 mM) se añadió a una cápsula HRMAS desechable Kel-F con
capacidad de 30 μL (Bruker Biospin, Rheinstetten, Germany) junto con la muestra
cárnica (10 ± 2 mg). La cápsula se introdujo en un rotor de zirconio de 4 mm. Los
espectros 1H RMN fueron adquiridos a 4 oC y una velocidad de giro de 5 MHz. Para ello,
se utilizó la secuencia espín-eco Carr-Purcell-Meiboom-Gill (CPMG) con un tiempo de
eco de 300 ms con el fin de eliminar las señales más intensas correspondientes a
lípidos y macromoléculas. Los espectros fueron adquiridos a 32k puntos de datos, con
un tiempo de repetición de 5 s. Se corrigió su fase, su línea base y finalmente se
referenciaron a TSP utilizando el software Topspin, versión 2.1.
Figura III.12. Espectrómetro Bruker Avance 600 MHz equipado con una sonda HRMAS (high resolution magic angle
spinnig) para el análisis de muestras semi-sólidas.
134
IV. Resultados
135
IV.1. ESTUDIO DE LA ESTRUCTURA (MACRO Y
MICRO) DE DISTINTOS PRODUCTOS CÁRNICOS CURADOS
137
IV. Resultados
Trabajo 1
Empleo de la IRM como herramienta predictiva para la estimación de parámetros
fisicoquímicos y reológicos durante el proceso de elaboración de jamón curado
141
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
*Corresponding author. Tel: +34 913943745; Fax: +34 913943743; E-mail: [email protected]
Abstract: Magnetic resonance imaging (MRI) was used to study the structural changes during
dry-cured ham manufacturing. T1, T2 and apparent diffusion coefficient (ADC) were determined.
Dry cured hams were analysed at different steps of the manufacturing process (raw, salted, post
salted, half-cured and cured). Structural changes have been linked with the rheological
behaviour, estimated by texture profile analysis (TPA) performed in three different muscles of
hams (semimembranosus, semitendinosus and biceps femoris). A decrease for T1, T2 and ADC
parameters was observed, connected to the dehydration kinetics and salt diffusion. Curing
process increased hardness and chewiness and reduced elasticity and cohesiveness.
Mathematical models were defined to obtain useful equations to monitor ripening. Multiple and
simple linear regression models were performed to predict water and salt content and rheological
features evolution through MRI parameters. Best settings were achieved with water and salt
content for the three studied muscles (R2 around 0.90). T1, T2 and ADC showed a negative
correlation with hardness and a positive relation with springiness and cohesiveness.
Keywords: cured ham, magnetic resonance imaging (MRI), texture profile analysis (TPA),
predictive models.
143
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
1.Introduction
Dry-curing has evolved from a mere meat preservation process to an improved procedure, in
order to obtain a flavourful and attractive meat product with particular and characteristic sensory
traits (Toldrá & Flores, 1998). Cured ham, in particular, is one of the most valuable meat
products, both inside and outside Spain, with a first-rate consumer acceptance. It is a non-
homogeneous product that undergoes a salting stage and then is subjected to a long
dehydration/ripening process (taking between 12 to 24 months) during which the dynamics of the
migration of the water and salts is balanced (Arnau et al., 1995) and flavour is developed. Given
the technological and economic importance of this product, the development of suitable analytical
tools is essential to know the quality of ham at its best.
A non-invasive and non-destructive way of measuring moisture profiles is through the
application of nuclear magnetic resonance (NMR) techniques. Proton NMR is performed by
applying a radio frequency pulse to a sample within a magnetic field and assessing the relaxation
properties of the hydrogen nuclei of the sample (Charlton, 2009). Magnetic resonance imaging
(MRI) permits the quantification of water content and distribution, ant its interaction with the food
matrix by magnetic resonance parameters such as transverse relaxation time (T2), longitudinal
relaxation time (T1) and the apparent diffusion coefficient (ADC). (Fantazzini et al., 2009; Marcone
et al., 2013). These magnetic resonance parameters are potentially sensitive to variations of
water mobility and enviroment (T1 and T2) and facility of diffusion of the water in all directions
(ADC) resulting from the modification of water macromolecule interactions and changes in tissue
structure (Boulby & Rugg-Gunn, 2003; Gowland & Stevenson, 2003; Wheeler- Kingshott et al.,
2003). Due to the extensive information gained through this technique, it has been used in meat
science to study carcass composition, adipose tissue distribution, connective tissue and muscle
fibre type. These parameters have also been correlated with meat properties, including pH,
water-holding capacity, moisture, texture and sensory attributes. (Tingle et al. 1995; Bonny et al.,
2000, Cernadas et al., 2005, Laurent, Bonny & Renou, 2000, Guiheneuf et al. 1997; Guiheneuf et
al 1996; Herrero et al., 2007, Herrero et al. 2009; Mitchell et al., 2001, Renou, Foucat & Bonny
2003, Ruiz-Cabrera et al., 2004 and Shaarani, Nott & Hall, 2006). However, few studies have
been performed on dry-cured ham (Manzocco, L., 2013; Fantazziniet al., 2009) and even this
technique has not been used so far to carry out a detailed study of the changes that take place
during Spanish dry cured ham manufacturing.
Moreover, texture is one of the key quality attributes used in the food industry to assess
product quality and acceptability. Meat products obtained by processing operations, such as
curing, show a non-isotropic behaviour, influenced by the internal structure, including different
tissues and different chemical composition. The evaluation of the rheological behaviour
contributes to understand the effect of processing on products, probes the system structure and
reveals critical aspects of food texture (Foegeding, 2003). Nowadays, the most commonly used
instrumental method is probably, the texture profile analysis (TPA) test, which is based on the
imitation of chewing process and is performed with double-compression cycles. TPA has been
144
IV. Resultados
García-García et al., 2018, IV.1.1., 1-23
employed for many authors as a usual way to evaluate sensorial properties of cured ham (Tabilo
et al., 1999; Morales et al., 2007a, 2007b; Aliño et al., 2010; Benedini et al., 2012).
The aim of this study was to apply MRI technique to pork in different stages of the curing
process to determine relaxation times T1 and T2, and ADC, relating these results to those
obtained from compression textural analysis (TPA) and physicochemical characteristics in order
to obtain useful equations to monitor the ripening process. Linear regression statistics could be
used to predict physicochemical and rheological behaviour using the imaging parameters as
predictor variables.
This study constitutes the first attempt to relate MRI parameters, which provide structural
information, with textural attributes of meat products. The establishment of suitable mathematical
models would allow using MRI to estimate the rheological behaviour of meat products without
realizing complex instrumental analyses, such as the TPA, which requires a laborious procedure.
Finally, this work is focused on increasing the existing knowledge on the potential application of
the MRI for the estimation of food quality and process control.
A total of 35 fresh hams (10-11 kg) obtained from white-breed pigs, (Landrace x Large White
female animals slaughtered at 100-120 kg live weight), were purchased from a local
manufacturing plant. The legs were randomly chosen. The process of salting and ripening was
the one usually followed by a standard production plant of Spanish dry-cured hams (Santos et al.,
2008). Sampling was carried out at the most relevant stages of the process (Figure 1):
▪ Raw hams (R). The pieces were taken at 3 days after slaughter.
▪ Salted hams (S). Raw hams were covered with a mixture of salts (NaCl amount
equivalent to 2% of the weight of the ham, 200 ppm of KNO 3 and 100 ppm of NaNO2).
Samples were taken one hour, approximately, after salt addition.
▪ Post-salted hams (PS). Hams covered with salts were staked for a period of 0.75 day
per kg of fresh ham. The samples were taken after removing salt left on the surface of
the hams.
▪ Hams in drying period for 6 months (Half-cured hams, 6M). Post-salted hams were
washed with cold water and hung for 30 days at 4 ± 1 oC. Then the pieces were
maintained at 13–15 oC and RH 75–85% for three months, and at 22–26 oC, RH 60–
75% for two months.
▪ Hams analysed at 8 months corresponding to the end of the drying process (Half-
cured hams, 8M). Pieces processed as previously mentioned but taken after three
months at 13–15 oC and RH 75–85%, and four months at 22–26 oC, RH 60–75%.
145
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
▪ Hams at ripening process with a total ageing of 12 months (Ageing, 12 M). After a
drying process of 8 months, the cured hams were left for four months at 12-17 oC and
RH 60-80%
Figure 1. Sample collection and sampling procedure, showing the Anatomical localization of muscles under analysis
(Bíceps femoris [BF], Semimembranosus [SM] and Semitendinosus [ST]) in the pork leg.
146
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1, 1-23
▪ Hams ripened up to 20 months (Ageing, 20 M): dry-cured hams with a total ripening
period of 20 months.
5 pieces were analyzed at each stage. All the hams used corresponded to the right side of the
animal. Once the bones, skin and fat had been removed, muscles corresponding to different
anatomical locations [biceps femoris (BF) and semitendinosus (ST), internal-type muscles, and
semimembranosus (SM), an external-type one] were taken from the leg, as displayed in Figure 1.
The visible fat and connective tissue were totally removed from the separated muscles pieces.
Water activity (aw), pH, dry matter (DM), salt content and fat content were determined in
triplicate samples of each of the considered muscles. aw was measured using a Decagon CX1
hygrometer (Decagon Devices Inc., Pullman,WA, USA) at 25 oC. The pH was determined in a
homogenate of the sample with distilled water (1:10) (w/v), using a Crison Digit-501 pH meter
(Crison Instruments LTD, Barcelona, Spain). DM was determined by drying the sample at 110 oC
to constant weight and the results were expressed as a percentage (AOAC, 2006). The salt
content was measured by AOAC method 935.47 (AOAC, 1995). Fat content was determined
using the method of Bligh and Dyer, described by Hanson and Olley (1963).
MRI experiments were performed using a Bruker BIOSPEC 47/40 spectrometer (Bruker
GmBH, Ettlingen, Germany) operating at 4.7 T (200 MHz). Magnetic field gradients for imaging in
the three orthogonal directions were generated by a 6 cm actively shielded gradient set capable
of reaching 450 mTm-1. Each of the samples for imaging, coming from the inner part of each of
the three muscles studied (BF, ST, SM), was cut approximately 4 cm long, 2 cm wide and 1.5 cm
thick. Individual specimens were placed in a radio frequency birdcage coil with an inner diameter
of 3.5 cm. The temperature inside the magnet was maintained at 18 °C using a water circulator.
For the MRI experiments a first global shimming was performed and then three scout spin-echo
experiments in axial, sagittal and coronal directions were acquired.
For measurements of T2, 60 separate images were acquired at different echo times (TE = 5-
300 ms). All other imaging parameters remained constant between images (NA = 1; TR = 3046.1
ms; MTX = 256 x 128, FOV = 5 x 2.5 cm; slice thickness = 1 mm; slice number = 1).
T1 relaxation time measurements were conducted on the same slice using a spin-echo
sequence. 13 separate images were acquired at different recovery times. TR varied from 52.5 to
6002.5 ms. All other imaging parameters remained constant between images (NA = 1; TE = 5 ms;
MTX = 256 x 128; FOV = 5 x 2.5 cm; slice thickness = 1 mm; slice number = 1).
To measure the apparent diffusion coefficient (ADC), separate images were acquired at six
diffusion weightings (b value varied from 48.4 to 54969.881 s mm-2). All other imaging parameters
147
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
remained constant between images (NA = 1; TR = 2000 ms; TE = 45 ms; Δ = 25ms; δ = 10 ms;
MTX = 256 x 128; FOV = 5 x 2.5 cm; slice thickness = 1 mm; slice number = 1).
From these sequences, T1, T2 and ADC maps were calculated using Image J 1.44p (Rasband,
W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA). The mean values at
ten different Region of Interest (ROIs) in each map were measured to obtain the T1, T2 and ADC
values.
Texture profile analysis (TPA) was performed using a TA.XT2i SMS Stable micro Systems
Texture Analyser (Stable Microsystems Ltd., Surrey, England) with the Texture Exponent 32,
6.0.7.0. version. Eight sample cores (diameter = 2.5 cm, height = 1.0 cm) were obtained from the
inner part of each of the studied muscles. A double compression cycle test was performed of up
to 50% compression of the original height with an aluminium cylinder probe P/25. Force–time
deformation curves were obtained with a 5 kg load cell applied at a cross-head speed of 1.0
mm/s. Hardness (N); springiness (cm); adhesiveness (N s); cohesiveness (dimensionless);
gumminess (N), (hardness × cohesiveness) and chewiness (J) (hardness × cohesiveness ×
springiness) were quantified according to Segura et al., 2015. The compression force was
performed in longitudinal direction to muscle fibres. Measurements of samples were carried out at
room temperature.
Statistical analyses were perfomed using Statgraphics Centurion XVI for Windows (Statistical
Graphics Corporation, Rockville, MD, USA). Data were presented as mean values and standard
deviation (SD). Chemical and TPA analyses were carried out in triplicate. To check the normal
distribution (90% confidence) of samples, the Shapiro–Wilks test was applied. When samples
fitted the normal distribution, the one-way ANOVA analysis was performed. When samples did
not fit the normal distribution, the Kruskal–Wallis test was used to test the null hypothesis that the
medians of variable within each of the levels in the samples were the same. Both Duncan´s test
for multiple mean comparisons and the Pearson product moment correlation (R), using the
Durbin–Watson statistic tests at 95%, of confidence level, were performed to determine the
relationships between data from TPA, physicochemical analysis and MRI parameters.
The multiple linear regressions were performed to determine the relationships between data
obtained by water and salt analysis, TPA and NMR parameters. To describe the relation between
textural and MRI parameters, surface models were obtained by regressions analysis to estimate
the response function as shown below:
148
IV. Resultados
García-García et al., 2018, IV.
where Y is the predicted response (estimated textural parameters values), o, i, ii, and ij, are
the estimated coefficients from regression. They represent the linear, quadratic and cross-product
relation among MRI parameters [T1 (X1), T2 (X2) and ADC (X3)] and textural features on the
response. ANOVA was applied to test the statistical significance of each factor introduced in the
model and non-significant factors, with a confidence level of 95% (p > 0.05), were gradually
eliminated. In order to check the goodness of fit of the calculated models, Root Mean Square
Error (RMSE) was also obtained according to the following equation:
RMSE
Xpred Xobs 2
n
where, Xpred is the estimated data, Xobs is the result obtained in the experimental analyses and
n the number of samples.
The mean values and their corresponding deviation of the physicochemical characteristics
obtained for each of the main stages of the curing process are shown in Table 1. aw dropped after
12 months of ripening in the three considered muscles. This decrease in a w is the main
mechanism to ensure the microbiological product safety since pH underwent little variation during
the process and remained at values ranging from 5.57 to 6.30. Dry matter (DM) content of the
muscles selected increased significantly throughout the curing process (p < 0.05). Lower moisture
content and aw are related to salting due to the osmotic effect of salt added to the surface of ham
and dehydration during resting and maturation. Significant differences (p < 0.05) were found
among muscles. As expected, the moisture content decreased quicker in the external muscle
(SM) not covered by skin and/or fat than in the inner ones (BF and ST) (Arnau, Casademont &
Gou, 1995) so SM presented the highest DM values and the lower a w during the hole process.
Although ST is located in a deeper layer of the pieces, BF is surrounded by a thick layer of fat,
which reduces its water loss (DM around 44.0 g/100 g product in BF and around 49.9 g/100 g
product in ST at the end of processing time). An increase on NaCl content was observed during
the studied stages, related to the penetration of the salt added to the raw pieces. At the end of the
ripening time, the highest NaCl content (p < 0.05) was observed in BF muscle (5.7±0.7) followed
by SM and ST, with values of 5.1±0.5 and 4.7±0.2, respectively. These results are in agreement
with the findings obtained by several authors (Arnau, Casademont & Gou, 1995; Fantazzini et al.,
2009). Raw ST presented the highest fat content (5.46±0.2 g/100 g product), followed by BF and
SM (2.40±0.1 and 2.13±0.1 g/100 g product, respectively). With the progressive loss of water
occurred the ensuing increase in the percentage of fat from pieces. According to previous studies
(Kim et al, 2008), at the end of the process, ST presented the higher value (13.94±1.3 g/100 g
product) followed by BF (12.91±1.2 g/100g product) and SM (11.11±1.2 g/100 g product). In
149
IV. Resultados
García-García et al., 2018, enviado a Meat Science, V.1.1., 1-23
Table 1. Physicochemical properties and chemical composition of muscles in different stages of the curing process of ham
a, b: values in the same column for a same muscle with different lyrics differ significantly (P<0.05) (differences in a same muscle throughout the process)
: values in the same column for the same curing stage of different muscles with different lyrics differ significantly (P<0.05) (differences between muscles throughout
the process)
150
IV. Resultados
García-García et al., 2018, IV.
general, the estimated parameters are within the range of values for different hams cured
Spanish type (Arnau, Casdemont & Gou, 1995; Fernández et al, 2007; Santos et al, 2008).
BF SM ST BF SM ST BF SM ST
Raw
Salted
1600.0 ms 100.0 ms 1.0 · 10-3 mm 2/s
Post-salted
Half-cured (6 months)
Half-cured (8 months)
Figure 2. T1, T2 and ADC maps obtained using MRI tool. A: T1 maps, B: T2 maps; C: ADC maps.
T1 maps were characterised by areas with different intensity level. Fat appeared dark than
lean tissue due to the ease in the relaxation of protons in the lipid molecules compared to those
present in muscle.
Table 2 shows the median value of MRI parameters calculated for different Regions of Interest
(ROIs) of BF, SM and ST. A progressive reduction of average T1 was observed throughout
ripening time in each of the concerned muscles. In BF, T1 value decreased from 1409.0±78.2 ms
in raw product to 707.5±13.4 ms in finished ham. The same trend was observed in the SM and
ST muscles in which this parameter went from 1388.3±73.9 to 483.6±29.8 ms and from
1403.4±47.5 to 698.0 ± 9.3 ms, respectively.
The observed T1 reduction is due to the salt diffusion and water evaporation that takes place
during the dry-curing progress. From the salting stage, sets a gradient of moisture from the inside
to the surface of the meat pieces and from the surface to the environment. These facts, respond
to the flow of water out by evaporation, osmotic dehydration and dissemination, all favoured by
environmental conditions in which develops the process and salt addition (Arnau & Gou, 2001).
151
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
Table 2. IRM parameters from Bíceps femoris (BF), Semimembranosus (SM) and Semitendinosus (ST) muscles in different stages of the process.
BF SM ST
T1 T2 ADC T1 T2 ADC T1 T2 ADC
(ms) (ms) (10-3 mm2/s) (ms) (ms) (10-3 mm2/s) (ms) (ms) (10-3 mm2/s)
Raw 1286.2±58.2 a, 45.6±1.7 a, 0.26±0.02 b, 1509.0±136.7 a, 47.1±1.1 a, 0.27±0.04 b, 1465.8±100.5 a, 46.9±1.2 a, 0.15±0.01 d,
Salted 1232.7±32.5 a, 42.5±1.1 b, 0.24±0.01 b, 1255.3±63.7 b, 41.2±2.2 b, 0.70±0.04 a, 1189.0±70.0 b, 46.2±1.0 a, 0.89±0.11 a,
Postsalted 1136.5±75.3 b, 40.9±2.4 b, 0.89±0.07 a, 1026.5±34.2 c, 35.7±2.6 c, 0.77±0.07 a, 1121.3±94.3 b, 39.8±1.5 b, 0.76±0.04 b,
Half-cured (6 months) 965.3±33.5 c, 28.3±6.0 c, 0.08±0.03 c, 778.8±30.6 d, 27.7±2.6 d, 0.16±0.03 c, 806.8±35.2 c, 30.9±2.0 c, 0.18±0.03 c,
Half-cured (8 months) 737.7±27.9 d, 26.9±2.2 c, 0.06±0.01 c, 694.1±48.0 e, 19.7±3.2 e, 0.02±0.00 e, 800.7±47.4 c, 26.1±1.7 d, 0.05±0.00 e,
Ageing (12 months) 707.6±29.9 d, 24.7±3.6 c, 0.08±0.03 c, 592.7±20.2 f, 19.2±2.2 e, 0.06±0.03 d, 693.0±33.5 d, 24.8±1.9 d, 0.20±0.03 c,
Ageing (20 months) 704.7±19.0 d, 15.4±1.1 d, 0.04±0.01 d, 478.5±28.7 g, 9.7±1.2 f, 0.01±0.00 e, 695.9±26.5 d, 12.8±0.9 e, 0.02±0.00 f,
a, b: values in the same column for a same muscle with different lyrics differ significantly (P<0.05) (differences in a same muscle throughout the process)
: values in the same column for the same curing stage of different muscles with different lyrics differ significantly (P<0.05) (differences between muscles throughout the
process)
10
152
IV. Resultados
García-García et al., 2018, IV.
T1 values are also associated with muscle location within the ham. BF and ST are inner muscles
of ham cushion; moisture decrease and salt uptake occur slowly, whereas SM is an outer muscle
of the thigh, undergoing a direct exposure to air dehydration from salting to maturing. It is
characterized by a marked moisture decrease that results in the lowest T1 among assayed
muscles at the same processing time. Moisture content in BF exceeds the values found in SM
and also in ST during the whole process, so that T1 gets the highest value for this muscle.
Moreover, fat protons have lower relaxation time T1 than muscle protons. This phenomenon is
related to lower average T1 of marbled muscles like ST.
11
153
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV1.1.,1-23
Throughout ripening a decrease in ADC was observed in almost every of the considered times,
except during the salting and post salting stages, in which an increase of ADC occurred due to
salt-induced osmotic effect and protein solubilization. The higher ADC value for SM (0.89*10-3 ±
0.05 mm2 s-1) was reached during the salting stage due to its direct contact with the added salt at
this time. The inner location of muscles BF and ST make the salt uptake slower and is not even
reached the resting phase when a higher diffusion of water is observed (0.77*10-3 ± 0.07 mm2s-1
in BF and 0.76*10-3 ± 0.04 mm2s-1 in ST).
As the process progresses, the restricted diffusive motion of water (low ADC) is observed. The
water that remains after dehydration is strongly linked by myofibrillar structures. This retention is
favoured by the presence of salt in the middle, since salt absorption is related to higher water
holding capacity and protein solubilisation. Both phenomena, dehydration and salt uptake, result
in a gradually descent of ADC that reaches the minimum value at the end of the manufacturing
process (0.03*10-3 ± 0.01 mm2s-1 in BF; 0.008*10-3 ± 0.002 mm2s-1 in SM and 0.02*10-3 ± 0.002
mm2s-1 in ST). As expected, ADC temporal changes were more pronounced in SM due to its
external location, whereas the values in BF and ST were very close, not finding significant
differences (p > 0.05) between both muscles for the same processing time. Although ST is a
muscle with higher fat content, fat had not led a barrier to water diffusion.
Drying and salting, however, did not affect significantly to the T2 relaxation time values
because the contribution of the myofibrillar water on this parameter is much larger than the input
of another more mobile water populations, that would be the more involved in the above-
mentioned ADC increase. In the post-salting stage, when dehydration was stronger, both
parameters, T2 and ADC, decreased, suggesting a remarkable loss of water in the whole product.
Values of rheological parameters obtained from the texture profile analysis (TPA) are shown in
Table 3. Significant differences (p < 0.05) were found between different stages of the process and
between the three studied muscles. In dry-cured ham, texture is a characteristic directly related to
the muscle structure, especially related to the degradation of myofibrillar protein and collagen as
well as to the intramuscular fat content and drying rate (Toldrá, 1998).
In accordance with previous studies, Monin et al., 1997, Serra et al. 2005, hardness increased
during ripening. When lower water activity and water content values were reached, hardness
increased significantly. Similarly, it also varies the chewiness since it is a secondary textural
characteristic (product of hardness x cohesiveness x elasticity). Salt could also increase the
stability and rigidity of the myofibrillar muscle structure since stiffness is increased by
mineralization. In the three muscles considered, a progressive trend was observed and the
maximum value of hardness and chewiness was reached at the end of the ripening. The muscle
that kept the highest hardness and chewiness values throughout the process was SM, followed
by BF and finally ST, the softest and the weakest of the three muscles studied. ST showed higher
fat content, lower mineralization level due to its intern location and medium moisture level that
12
154
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
Table 3. Texture parameters of Biceps femoris (BF), Semimembranosus (SM) and Semitendinosus (ST) muscles in different stages of the dry-cured ham manufacturing process
a, b: values in the same column for a same muscle with different lyrics differ significantly (P<0.05) (differences in a same muscle throughout the process)
: values in the same column for the same curing stage of different muscles with different lyrics differ significantly (P<0.05) (differences between muscles throughout the
process)
13
155
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
cause a lower mechanical resistance in that muscle compared with BF and, especially with SM,
characterized for an intense dehydration.
According to Serra et al., 2005, cohesiveness and springiness showed a positive linear
relationship with aw and moisture content. With increasing dehydration, cohesiveness (extent to
which a material can be deformed before it ruptures) and springiness (rate at which a deformed
material goes back to its initial form after the deforming force is removed) were reduced. SM
presented also the higher values for cohesiveness and springiness during the process. The
adhesiveness (degree to which the surface of the ham muscles adheres to the palate during
chewing process) not experienced regular changes during ripening. In each of the three muscles
studied, a decrease in adhesiveness since raw ham to cured ham was observed, although there
was not a progressive reduction. BF resulted to be the one presenting the lower value of
adhesiveness at the end of the process.
14
156
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
Table 4. Multiple linear regression analysis of physicochemical characteristics (water content and salt content) versus MRI parameters (T1, T2 and ADC) of ham
15
157
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
MRI parameters are related to water and salt content and distribution, so the results obtained
by regression analysis are in accordance with those obtained by other authors throughout the
curing process (Monin et al.1997, Serra et al. 2005).
16
158
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV1.1., 1-23.
Table 5. Estimated response surface model relating textural properties to magnetic resonance image (MRI) parameters of different muscle [biceps femoris (BF), semitendinosus (ST), and
semimembranosus (SM)] in different stages of the dry-cured ham manufacturing process.
R2adjusted
Muscle R2
Regression models for d.f. P RMSE
type (percent)
(percent)
Hardness = 38.58 + 0.13*T1 - 5.07*T2 + 254.63*ADC + 0.34*T22 - 0,01*T1*T2 + 2.05*T1*ADC – 63.23*T2 0.0000
97.44 96.06
*ADC
Cohesiveness = 1.02 – 0.09* T2 + 19.67*ADC + 0.000002* T12 + 0.006* T22 + 7.76*ADC2 -
91.01 83.65 0.0000
0.0002* T1* T2 + 0.02* T1*ADC – 1.43* T2*ADC +0.0003* T1* T2*ADC
BF
Springiness= -4.48+ 0.01* T1 – 0.34* T2+ 245.19*ADC + 0.03* T22+ 36.40*ADC2 – 0.0009* T1 * T2 – 12.05*
87.06 78.43 0.001
T2*ADC + 0.004* T1 * T2*ADC
Adhesiveness= 3.32 – 0.38* T2 + 31.26*ADC + 0.00001* T12+ 0.04* T22- 0.002* T1 * T2 + 0.21* T1 *ADC –
94.66 91.78 0.0000
6.49* T2*ADC
Hardness= 305.08 – 0.57*T1 + 0.0002*T12 + 0.08**T22 +0.02*T1*ADC 92.34 90.29 0.0000
Cohesiveness = 1.21 – 0.001* T1 - 2.93*ADC + 0.000004* T12
+ 0.004* T22
– 0.0002* T1 * T2 + 0.005* T1
83.31 74.32 0.001
SM *ADC -0.00006* T1* T2 *ADC
Springiness= -5.43+ 0.61*T2 – 14.31*ADC + 0.00007*T12+ 0.05* T22 + 13.27*ADC2 – 0.004* T1* T2 95.55 93.64 0.001
2 2 2
Adhesiveness= -0.36 – 0.007* T1 + 0.21* T2 +0.00003* T1 + 0.03* T2 + 9.82*ADC – 0.002* T1*T2 - 0.42*
86.94 78.23 0.005
T2*ADC + 0.0002* T1* T2*ADC
Hardness = -92.43 + 0.26*T1 – 1.47*T2 + 609.53*ADC +0.06*T22 – 90.69*ADC2 – 0.00579*T1*T2 – 99.70
99.44 0.0000
0.55*T1*ADC – 13.49*T2*ADC + 0.01*T1*T2*ADC
Cohesiveness = 0.04 +0.0008* T1 +3.12*ADC - 0.0003* T22 - 0.003* T1 *ADC - 0.05* T2 *ADC + 0.00006*T1*
92.20 88.59 0.005
T2*ADC
ST
Springiness= -28.62 + 0.05*T1 + 0.50*T2 + 89.03*ADC -18.40*ADC2 – 0.001* T1*T2 – 0.10*T1*ADC – 1.93*
93.27 88.78 0.0001
T2*ADC+ 0.002* T1 * T2*ADC
Adhesiveness=- 12.88 + 0.02* T1 + 53.66*ADC – 10.70*ADC2- 0.0003* T1 * T2 – 0.06* T1 *ADC – 0.81*
92.90 89.08 0.0001
T2*ADC +0.001* T1 * T2*ADC
R2= coefficient of determination. P-value for R2 was determined using F-test. RMSE: Root Mean Square Error
17
159
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
4. Conclusions
MRI has proven to be a suitable tool in order to study structural changes during ham
manufacturing process, since T1, T2 and ADC parameters allow getting complete information of
the studied tissues. The multivariate analysis confirms that MRI parameters could be used to
construct regression models to predict water and salt content and TPA parameters. Therefore,
with only non-destructive MRI analyses it could be possible to obtain structural, physicochemical
and textural information. This type of predictive models was accurate enough to consider MRI as
a useful tool for monitoring and optimizing dry cured ham manufacturing process.
Acknowledgements
The authors are grateful for the financial support of UCM-BSCH Group 920176GR35/10A, project
AGL2010‐19158 funded by the Spanish Secretary of State of Research, Development and
Innovation within the Ministry of Economy and Competitiveness. A.B.G.G. was awarded a grant
(BES-2011-047485) from the same institution.
18
160
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
References
Aliño, M., Grau, R., Toldrá, F., Barat, J.M. (2010). Physicochemical changes in dry-cured hams salted with
potassium, calcium and magnesium chloride as a partial replacement for sodium chloride. Meat Science
86, 331-336.
Arnau, J. & Gou, P. (2001). Effect of air relative humidity on ham rind and subcutaneous
salted fat during the resting period. Meat Science, 58, 65-68.
Arnau, J., Guerrero, L., Casademont, G. & Gou, P. (1995). Physical and chemical changes in different zones
of normal and PSE dry-cured ham during processing. Food Chemistry, 52, 63-69.
Benedini, R., Parolari, G., Toscani, T., Virgili, R. (2012). Sensory and texture properties of Italian typical dry-
cured hams as related to maturation time and salt content. Meat Science, 90, 431-437.
Bonny, J. M., Laurent, W., Labas, R., Taylor, R., Berge, P., Renou, J. P. (2000). Magnetic resonance
imaging of connective tissue: A non-destructive method for characterizing muscle structure. Journal of
the Science of Food and Agriculture, 81, 337–341.
Boulby, P. A. & Rugg-Gunn, F. (2003). T2: The transverse relaxation time. In: P. Toofts (Ed.), Quantitative
MRI of the brain. Measuring changes caused by disease (pp. 143-173). West Sussex, England: John
Wiley & Sons.
Cernadas, E., Carrion, P., Rodriguez, P. G., Muriel, E., Antequera, T. (2005). Analyzing magnetic resonance
images of Iberian pork loin to predict its sensorial characteristics. Computer Vision and Image
Understanding, 98, 344–360.
Charlton, J. A. (2009). High resolution NMR analysis of complex mixtures. In: M. Gudjónsdóttir, P. S. Belton,
& G. A. Webb (Eds.), Magnetic resonance in food science: challenges in a changing world (pp. 3–12).
Cambridge: Royal Society of Chemistry.
Fantazzini, P., Bortolotti, V., Garavaglia, C., Gombia, M., Riccardi, S., Schembri, P. (2005). Magnetic
resonance imaging and relaxation analysis to predict noninvasively and non-destructively salt-to-
moisture ratios in dry-cured ham. Magnetic Resonance Imaging, 23, 359-361.
Fantazzini, P., Gombia, M., Schembri, P., Simoncini, N., Virgili, R. (2009). Use of magnetic resonance
imaging for monitoring Parma dry-cured ham processing. Meat Science, 82, 219-27.
Fernández, M., Ordóñez, J.A., Cambero, I., Santos, C., Pin, C., de la Hoz, L. (2007). Fatty acid composition
of selected varieties of Spanish dry ham related to their nutritional implications. Food Chemistry, 101,
107-112.
Foegeding, E. A., Brown, J., Drake, M. A., Daubert, C. R. (2003). Review. Sensory and mechanical aspects
of cheese texture. International Dairy Journal, 13, 585-591.
Gowland, P.A.; Stevenson, V.L. (2003). T1: The longitudinal relaxation time. In: Quantitative MRI of the
brain. Measuring changes caused by disease. (pp. 111–143). Ed. John Wiley & Sons, West Sussex,
England.
Guiheneuf, T. M., Tessier, J. J., Herrod, N. J., Hall, L. D. (1996). Magnetic resonance imaging of meat
products: Automated quantitation of the NMR relaxation parameters of cured pork, by both bulk NMR
and MRI methods. Journal of the Science of Food and Agriculture, 71, 163–173.
Guiheneuf, T. M., Parker, A. D., Tessier, J. J., Hall, L. D. (1997). Authentication of the effect of
freezing/thawing of pork by quantitative magnetic resonance imaging. Magnetic Resonance in
Chemistry, 35, 112-118.
Hanson, S.W.F. & Olley J. (1963). Application of the Bligh and Dyer method of lipid extraction to tissue
homogenates. Biochemical Journal, 89, 101-102.
Herrero, A. M., Cambero, M. I., Ordonez, J. A., Castejon, D., Romero de Avila, M. D., de la Hoz, L. (2007).
Magnetic resonance imaging, rheological properties, and physicochemical characteristics of meat
systems with fibrinogen and thrombin. Journal of Agricultural and Food Chemistry, 55, 9357-9364.
19
161
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
Herrero, A.M., de la Hoz, L., Ordóñez, J.A., Castejón, D., Romero de Ávila, M.D., Cambero, M.I. (2009).
Magnetic resonance imaging studies of the cold-set gelation on plasma powder and meat systems. Food
Research International, 42, 1362-1372.
Kim, J. H., Seong, P. N., Cho, S. H., Park, B. Y., Hah, K. H., Yu, L. H., Lim, D. G., Hwang, I. H., Kim, D. H.,
Lee, J. L., Ahn, C. N. (2008). Characterization of nutritional value for twenty‐one pork muscles. Asian-
Australasian Journal Animal Science, 21, 138-144.
Laurent, W., Bonny, J. M. & Renou, J. P. (2000). Muscle characterization by NMR imaging and
spectroscopic techniques. Food Chemistry, 69, 419–426.
Marcone, M.F., Wuang, S., Albabish, W., Nie, S., Somnarain, D., Hill, A. (2013). Diverse food-based
applications of nuclear magnetic resonance (NMR) technology. Food Research International, 51, 729-
747.
Mitchell, A. D., Scholz, A. M., Wang, P. C., Song, H. (2001). Body composition analysis of the pig by
magnetic resonance imaging. Journal of Animal Science, 79, 1800-1813.
Monin, G., Marinova, P., Talmant, A., Martin, J.F., Cornet, M., Lanore, D., Grasso, F. (1997). Chemical and
structural changes in dry-cured hams (bayonne hams) during processing and effects of the dehairing
technique. Meat Science, 47, 29-47.
Manzocco, L., Anese, M., Marzona, S., Innocente, N., Lagazio, C., Nicoli, M.C. (2013). Monitoring dry-curing
of S. Daniele ham by magnetic resonance imaging. Food Chemistry, 141, 2246-2252.
Morales, R., Serra, X., Guerrero, L., Gou, P. (2007a). Instrumental evaluation of defective texture in dry-
cured hams. Meat Science, 76, 536-542.
Morales, R., Serra, X., Guerrero, L., Gou, P. (2007b). Softness in dry-cured porcine biceps femoris muscles
in relation to meat quality characteristics and processing conditions. Meat Science 77, 662-669.
Renou, J. P., Foucat, L., & Bonny, J. M. (2003). Magnetic resonance imaging studies of water interactions in
meat. Food Chemistry, 82, 35-39.
Romero de Ávila, M.D., Escudero, R., Ordóñez, J.A., Cambero, M.I. (2014). Weibull analysis characterizes
the breaking properties of dry-cured ham slices. Meat Science, 97, 451-458.
Ruiz-Cabrera, M.A., Gou, P., Foucat, L., Renou, J.P., Daudin, J.D. (2004). Water transfer analysis in pork
meat supported by NMR imaging. Meat Science, 67, 169-78.
Santos, C., Hoz, L., Cambero, I., Cabeza, C., Ordóñez, J.A. (2008). Enrichment of dry cured ham with α-
linolenic acid and α-tocopherol by the use of linseed oil and α-tocopheryl acetate in pig diets. Meat
Science, 80, 668-674.
Segura, J., Escudero, R., Romero de Ávila, M.D., Cambero, M.I., López-Bote, C.J. (2015). Effect of fatty
acid composition and positional distribution within the triglyceride on selected physical properties of dry-
cured ham subcutaneous fat. Meat Science, 103, 90-95.
Serra, X., Ruiz-Ramirez, J., Arnau, J., Gou, P. (2005). Texture parameters of dry-cured ham, muscle biceps
femoris samples, dried at different levels as a function of water activity and water content. Meat Science,
69, 249-54.
Shaarani, S. M., Nott, K. P., & Hall, L. D. (2006). Combination of NMR and MRI quantitation of structure and
structure changes for convection cooking of fresh chicken meat. Meat Science, 72, 398-403.
Simpson, N.E., Grant, S.C., Blackband, S.J., Constantinidis, I. (2003). NMR properties
of alginate microbeads. Biomaterials, 24, 4941-4948.
Tabilo, G., Flores, M., Fiszman, S.M., Toldrá, F. (1999). Postmortem meat quality and sex affect textural
properties and protein breakdown of dry-cured ham. Meat Science, 51, 255-260.
Tingle, J. M., Pope, J. M., Baumgartner, P. A., Sarafis, V. (1995). Magnetic resonance imaging of fat and
muscle distribution in meat. International Journal of Food Science and Technology, 30, 437–446.
Toldrá, F. & Flores, M. (1998). The role of muscle proteases and lipases in flavour development during the
processing of dry-cured ham. Critical Review of Food Science and Nutrition, 38, 331-52.
20
162
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1, 1-23
Toldrá, F. (1998). Desarrollo da las características de textura y flavor: contribución enzimática. In:
International Congress of Meat Science and Technology, 44,42-54. Barcelona. Simposio Especial.
Wheeler-Kingshott, A. M., Barker, G. J., Steens, S. C. A., Van Buchem, M. (2003). ADC: The diffusion of
water. In P. Toofts (Ed.), Quantitative MRI of the brain. Measuring changes caused by disease (pp. 203–
210). West Sussex, England: John Wile
21
163
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
SUPPLEMENTARY MATERIAL
(2 pages)
Table of contents
1). Figure S1. Relation among textural features and transverse relaxation time (T2, ms) and
longitudinal relaxation time (T1, ms) obtained by Magnetic resonance image (MRI) of different
muscle [biceps femoris (BF), semitendinosus (ST), and semimembranosus (SM)] in different
stages of the dry-cured ham manufacturing process (considering an apparent diffusion
coefficient value of 0.5* 10-3 mm2/s)……………………………………………………………………………………. p.28
22
165
IV. Resultados
García-García et al., 2018, enviado a Meat Science, IV.1.1., 1-23
1) Figure S1. Relation among textural features and transverse relaxation time (T2, ms) and longitudinal relaxation
time (T1, ms) obtained by Magnetic resonance image (MRI) of different muscle [biceps femoris (BF),
semitendinosus (ST), and semimembranosus (SM)] in different stages of the dry-cured ham manufacturing process
(considering an apparent diffusion coefficient value of 0.5* 10-3 mm2/s)
23
166
IV. Resultados
Trabajo 2
Estudio comparado de la microestructura de jamón curado mediante relaxometría T2 de
RMN, microscopía electrónica de barrido (MEB) y ensayos de tracción uniaxial.
167
IV. Resultados
169
IV. Resultados
García-García et al., 2018, enviado a Food Structure, V.1.2, 1-14
*Corresponding author. Tel: +34 913943745; Fax: +34 913943743; E-mail: [email protected]
Abstract: NMR relaxation analyses were performed on dry-cured hams at different processing
times to evaluate the ranges of variation of 1H transverse relaxation time (T2) in representative
ham muscle tissues, due to dehydration and salt uptake. Changes in the mobility and distribution
of water and fat marbling were detected throughout ripening. Three to four different 1H populations
were identified: T2b, T21, T22 and T2´. The shift of the main population, T21, toward faster relaxation
times indicates a higher degree of immobilization of water, and the change is related to the
increase in the consistency of the meat matrix. NMR results, supported by SEM images, suggest
that ripening favors the compaction of ham structure and leads to a high organized myofibrilar
matrix in which the remaining water is trapped within the protein tridimensional network. Fat
marbling became visible through the emergence of T2´ population during the maturing stage,
whose presence is linked to protons related to the fat phase. Breaking strength and elastic stress
limit values for BF and SM muscles changed according with the extent to which they are affected
by the curing process. The main water population, T 21 showed a positive linear relationship with
water content and it was negative related with the breaking strength. Consequently, the study
demonstrated that NMR T2 relaxometry is a promising technique to elucidate associations
between the biophysical state of intrinsic water and progression in processing of dry-cured ham.
171
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
1. Introduction
Foods are complex, dynamical and heterogeneous mixtures of macromolecules, solutes and
solvent (including water). This complex structure gives rise to three states of water; ‘structural’ or
‘bound’ water, water which is hydrogen-bonded inside the grooves and cavities of proteins and
polysaccharides; ‘surface’ water, water at the biopolymer surface having a dynamic state agitated
by the presence of the surface; and the third state is bulk water (Fullerton & Cameron, 1988; Hills,
Manning & Ridge, 1996). Thus, the distribution of water in the different states is highly dependent
on the microstructure of the food, which can be altered by changing formulation, processing and
storage conditions. However, the water availability for chemical and biochemical reactions
depends on the type of interaction it might have with the food constituents. This close relationship
between microstructure and water also influences the textural and other quality attributes; among
others, by the plasticising effect of water influencing the physical states of foods, such as
solubilisation, softening due to melting, glassy-to-rubbery transition and staling (Cornejo &
Chinachoti, 2003; Hermansson, 1983; Mathlouthi, 2001; Ruan & Chen, 1998). Because of this
intimate relationship between water, food quality and food safety, a more complete understanding
of water and its properties, behavior and influence, alone and in foods, is of prime importance.
In this regard, dry-cured ham is a whole muscle product whose processing involves
dehydration and salting of the raw pieces and the final attributes of the product depends on the
proper development of the techniques and procedures (Toldrá, 2006). A profound knowledge
about water mobility and distribution during ripening is essential to improve industrial ham
manufacturing.
Nuclear magnetic resonance (NMR) relaxometry is a widely used method in food science with
the ability to provide information about the mobility and compartmentalization of water and about
molecular interactions between water and myofibrilar components (Bertram, Dønstrup, Karlsson,
& Andersen, 2001; Bertram & Andersen, 2004). Thus, NMR relaxometry is able to provide detailed
information about the water in meat and thereby give a measure of the water availability, which is
decisive for food stability and safety (Bertram et al., 2001; Bertram, Purslow, & Andersen, 2002).
In the other hand, the rheological behavior of the product is highly related to textural properties
and, therefore, it determines technological usages and consumers´ acceptance. Among the
essential methods that can be used to study the rheological behavior, the tensile test is the best
suited for structural investigations (Purslow, 1985). The performance of this test allows obtaining
a load deformation curve to complete rupture, in which the fracture strength corresponds to the
maximum force supported by the material (Romero de Ávila, Escudero, Ordóñez & Cambero,
2014). The data obtained by tensile test offer complementary valuable information about muscles
behavior by applying a tensile stress force and how this external force affect meat structure.
The aim of this work was to investigate the potential use of NMR transverse relaxation time
(T2) as a technique capable of monitoring local changes occurring in dry-cured ham during
processing and affecting the quality of the final product, mainly related to salt and moisture content
and marbling fat amount and distribution. A detailed study of the microstructure of the product
172
IV. Resultados
García-García et al., 2018, IV.1.2, 1-14
was completed by the acquisition of SEM images, together with the characterization of the
breaking strength and elastic stress limit for each of the muscles and stages considered.
Samples of a total of 35 hams were taken at different steps of the manufacturing process: raw
(3 days after slaughter), during salting (after salt addition, in the first 2-3 days of processing), post-
salting (hams taken during the resting period after salt removal), maturing (half-cured hams taken
after 8 months of processing) and ageing (fully maturated hams after 20 months of processing).
Salting and post-salting were managed at low temperature (0-4 oC); two different processing
conditions were established during maturing: a cold period (13-15 oC, RH: 75-85%) and a warm
period (22-26 oC, RH: 60-75%) and ageing was performed at room temperature (12-17 oC, RH:
60-80), in conformity with Santos et al., 2008. Two different muscle types, corresponding to
different anatomical locations [bíceps femoris (BF) and semimembranosus (SM)] were
considered for analysis. The hams from white-breed pigs, (Landrace x Large White female
animals slaughtered at 100-120 kg live weight), were purchased from a local manufacturing plant.
Manufacturing process and sample collection was carried out according to García-García et al
(2018).
The water content was determined by drying the sample at 110 oC to constant weight and the
results were expressed as a percentage (AOAC, 2006).
173
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
calculation, thus, the effective echo time was 4 ms. The distributions were obtained using
UpenWin 1.01 (Università di Bologna, Bologna, Italy).
A preliminary T2 relaxometry study was performed in ham pieces, so one single analysis for
each ripening stage and muscle type was completed.
3. Results
174
IV. Resultados
García-García et al., 2018, IV.1.2, 1-14
salted meat. The post-salted hams were analyzed immediately after brushing the surface salt,
therefore these images show the effect of salt on the structure of the meat.
Figure 1. SEM images (300x) obtained for the two considered muscles [bíceps femoris (BF) and semimembranosus
(SM)] at different processing stages. Micrographs corresponding to raw, salted, post-salted, half-cured and cured hams
are shown)
Muscle packages experienced a slight swelling during this stage due to the property of salt to
increase the CRA, due to the shift of the isoelectric point of myofibrillar protein towards lower
values (Albarracín, Sánchez, Grau & Barat, 2011). The separation between bundles of muscle
fibers is not as clear and the fibers are not as sharp due to the degradation of cell membranes
and the solubilization of myofibrillar proteins by the effect of salt, especially in SM. Maturing and
ageing images mainly show the dehydration effect favored by the conditions of humidity and
temperature of the ripening rooms. The bundles of muscle fibers are much less distinguishable
and its diameter decreased due to water loss. Towards the end of ripening, the boundaries
between bundles of muscle fibers are hardly visible, favored by the lipolytic and proteolytic
phenomena that occur during these phases. The products generated are deposited in the inter-
muscular spaces (Larrea et al., 2007b) creating a sort of amorphous matrix between bundles of
muscle fibers. Especially notable is this effect in SM. BF is maintained to a greater extent fibrillar
structure, being possible to appreciate the packages of muscle fibers surrounded by connective
tissue (perimysium) dehydrated. The result is a characteristic structure of porous appearance.
3.2. T2 measurements
T2 distributed data (Figure 2) revealed the existence of four different T2 populations in dry-
cured samples, depending on the degree of association of the protons of the sample with the
protein matrix. T2b population, with a fast relaxation time, between 1-10 ms, is related to protons
strongly linked to macromolecules, with a high degree of association to the matrix. The main
population, T21, with a relaxation time around 50 ms in fresh meat, is related to protons located
within highly organized structures and corresponds to the water retained by the three-dimensional
network of myofibrillar proteins. T22 population, with slower relaxation time, more than 300 ms, is
related to the most mobile protons located in extramiofibrilar spaces. These three components
have been previously described in the fresh meat (Bertram, Dønstrup, Karlsson & Andersen,
175
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
2001; Bertram et al., 2001; Bertram & Andersen, 2004; Andersen, Andersen, & Bertram, 2006)
and also in dry-cured ham (Fantazzini et al., 2009).
Figure 2. NMR T2 distributed data obtained for bíceps femoris (BF) and semimembranosus (SM) muscles at different
processing times. A: raw ham. B: salted ham. C: post-salted ham. D: Half-cured ham. E: Cured ham.
176
IV. Resultados
García-García et al., 2018, IV.1.2, 1-14
Moreover, it should be noted the emergence of a new signal during maturing, T2´ population,
with a relaxation time of 80 ms approximately, not covered so far in the literature. To the authors´
knowledge, this is the first time that this signal has been associated with the protons of the fat
phase given that, at this point in the process, the infiltration degree of fat in the muscle matrix has
reach an important progress. The progress of ripening is reflected in the distributions obtained,
enabling the monitoring of the process. On the one hand, the dehydration of the protein matrix
manifested itself through changes in the protons distribution and mobility, so that a shift towards
T2 shorter relaxation times was observed, especially notable in the main component T21, and
related to the extinction of T22 population. This shift is related to a greater degree of binding of the
protons with the matrix, which suggests a higher degree of matrix compaction as the process
progresses. On the other hand, it was possible to observe changes in the distribution of fat, due
to the emergence of the T2´ signal, whose presence can be used as a marker for later stages of
processing. SEM images were of great help to interpret structural modifications. Their analysis
allowed to observe the reduction of the intrinsic size of pore of the protein matrix, due to the
phenomena of protein solubility, absorption of salt and dehydration, typical of curing, as well as
the display of the evolution of fat marbling in the muscle. In addition, differences in behavior
between muscles, were detected since the effect of ripening was more evident in SM muscle.
The T21 component distributions obtained for this muscle from the post-salting stage presented
faster relaxation times in comparison with the BF, suggesting that the ripening phenomena
affected faster to SM muscle, since it is an outer muscle, with direct contact with the air, which
determine the accelerated effects of the dynamics of diffusion of water and salt in this muscle.
Based on a previous work (García-García et al., 2018), the obtained values of moisture were used
in the present study to establish correlations between physicochemical parameters and T2 NMR
measurements. A linear correlation model between the main water population (T21) and the water
content for BF and SM muscles was calculated. A strong linear positive dependence was found
between both variables, suggesting that T21 population is a highly valuable parameter to monitor
the undergoing of the ripening process. The linear correlation models obtained were:
177
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
Table 1. Parameters obtained from tensile tests for bíceps femoris (BF) and semimembranosus (SM) in different stages
of the curing process
: values in the same column for the same curing stage of different muscles with different lyrics differ significantly (P<0.05)
(differences between muscles throughout the process)
The breaking strength of the two studied muscles (Ncm -2) increased significantly (P>0.05)
throughout the curing process. SM muscle was the most resistant, presenting the higher breaking
strength. This behavior suggests that rupture of the muscle would take place along several lines
of weakness in the bundles of muscle fibers, by successive microfractures in the muscle fibers
and in the destructuration of intramuscular connective tissue (Romero de Ávila, Escudero,
Ordóñez & Cambero, 2014). The enlargement in resistance in BF was much smaller than the one
showed by SM (the first tripled the value while in the second one was 20 times higher in cured
than in fresh meat). In both cases, this growing is associated with an increase in the elastic limit,
point above tissue deformation is irreversible. These results relate the increase of toughness with
the moisture loss and would be associated with data provided by other authors in different tissues
(Zeugolis, Paul, & Attenburrow, 2008; Jager & Fratzl, 2000; Landis, Librizzi, Dunn, & Silver, 1995).
Similarly, it has been reported that salt could increase the stability and rigidity of the myofibrillar
muscle structure (Andrés et al., 2004; Ruiz-Ramírez, Arnau, Serra, & Gou, 2005).
When a tensile test was applied, the resultant internal strains were absorbed causing a
characteristic deformation in each muscle tissue and strength-distance curves with different
shapes were obtained (Figure 3). Their analysis reveals a different mechanical behavior for BF
and SM. The form of the curves for each muscle can be related to several factors such as
morphology characteristics (Maltin, Balcerzak, Tilley, & Delday, 2003), sarcomere length (Serra,
Ruiz-Ramírez, Arnau, & Gou, 2005), intramuscular fat and connective contents (Serra, Ruiz-
Ramírez, Arnau, & Gou, 2005) and proteolytic activity (Parolari, Virgili, & Schivazappa, 1994).
178
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
Figure 3. Evolution of strength-distance curves obtained by means of uniaxial tensile test for bíceps femoris (BF) and
semimembranosus (SM) at different ripening stages.
Depending on the position of the muscle on the leg, the processing stage can also determine
the form of the obtained curves. BF experienced an increase of the breaking strength during
salting and drying and a slight decline during ageing. Samples reach the breaking point after an
elastic period, without a phase of plastic behavior. The shape of the curves changed during
ripening and a change in the slope was observed. SM curves are characterized by a period of
elastic deformation, which concludes with the break of the stressed sample. Modification of the
slope in this case is less remarkable than in BF, although the breaking strength increase is
progressive and most notably, reaching its maximum towards the end of the ripening. As it is
known, meat myofibrillar proteins have great influence on the structural and rheological properties
of food. BF myofibrillar degradation which takes place during ripening is slower than the
degradation experienced by SM due to its internal localization (slower loss of moisture and late
diffusion of salt), so properties as elasticity or the water holding capacity make stress behavior
different for the SM and BF muscles. At the end of the process, inside the product is still a three-
dimensional protein structure, while in the surface moisture loss has been more pronounced and
muscular structure is compact, the spaces between myofibrils are reduced and there is no
retained water. As a result, by applying tension forces, SM muscle, more compact structurally,
suffers less deformation than BF.
This finding agrees with a previous work, Romero de Ávila, Escudero, Ordóñez & Cambero,
2014, where stress-strain curves were obtained for different tissues of dry-cured ham in order to
provide a complete characterization of the mechanical behavior of this product.
SEM images resulted very helpful to understand what happens at the microstructural level by
applying tension forces. Figure 4 shows the behavior of BF muscle fibers with a representative
strength-distance curve obtained in a stress test. In the first part of the curve (area between A
and B) the applied force is proportional to the deformation suffered by the sample, stablishing the
region of linear deformation or elastic region. From point B, the deformation ceases to be linear
and the elastic limit is reached. From then on, the increase of tensile stress also increases the
179
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
deformation but not proportionally, reaching the peak of maximum force (C), from which the stress
begins to descend while the deformation continues the breaking point is reached.
Figure 4. Evolution of bíceps femoris (BF) muscular fibers behavior during an uniaxial tensile test.
SEM was also used to study the changes suffered by fresh and cured muscles when they
were subjected to the action of an external force of tensile stress. The stretching of SM muscle
fibers during the elastic period followed by their fully breaking is shown in Figure 5. The stretching
is most evident in fresh muscle, while in the cured one, due to the degradation of the myofibrillar
structure stretching effect is less noticeable. Especially striking is the structural difference which
can be seen when looking at the inner surface of a broken fiber bundle. Myofibrils can be
distinguished in fresh muscle, but not in the cured muscle, in which the fibers package is a
compact mass.
Figure 5. SEM images illustrate the evolution of semimembranosus (SM) microstructure by applying a tensile stress
force. Relaxed and stretched fibers are magnified 300x and broken fibers, 1000x.
10
180
IV. Resultados
García-García et al., 2018, IV.1.2, 1-14
As previously mentioned, microstructural changes in dry-cured ham during ripening are mainly
linked to salting and dehydration. In this regard, the main water population detected by NMR
relaxometry, T21, has demonstrated to be positive related to water content, therefore, it is possible
to follow the course of ripening by analyzing this NMR parameter. In this context, it is predictable
that T21 population have an inverse relationship with the breaking strength. Figure 6 shows the
evolution of both parameters, T21 population and breaking strength, during ripening for the two
studied muscles. The breaking strength is maximum when T21 values are minimum, which
suggests an antagonistic behavior of both variables. Short relaxation times involve matrix
compaction, therefore, more integrity and less plasticity of the tissue, which require the application
of higher external stress forces to get the breaking point.
The results obtained here suggest that dehydration, expressed in T21 terms, had a marked
effect on the mechanical properties of dry cured ham muscular matrix. It should be mentioned
that the breaking strength for the two studied muscles showed a linear relationship with T21,
especially for SM muscle (Figure 6). This finding is related to the dehydration and salting
dynamics in muscle tissues, since BF is located in a deeper layer of the ham and it is surrounded
in the outside by a thick layer of fat which reduces its water loss (Romero de Ávila, Escudero,
Ordóñez & Cambero, 2014), while SM occupies more external positions and would therefore be
subjected to the greatest degrees of dehydration and salt uptake. On the other hand, the intrinsic
properties of each muscle should be considered to explain their behavior: BF, in comparison to
SM, presents higher collagen content (Wheeler, Shakelford & Koomaraie, 2000) and it is a high
proteolyzed muscle which stays softer at the end of ripening (Parolari, Virgili, & Schivazappa,
1994).
Figure 6. T21 population and breaking strength evolution during ripening for bíceps femoris (BF) and semimembranosus
(SM) muscles.
4. Conclusions
Changes in water mobility and distribution during dry-cured ham processing can be analyzed
by means of NMR T2 relaxometry. Drying induced a notable shift toward shorter T21 relaxation
times related to the dehydration process, salt uptake and the consolidation of a firmer meat matrix.
The mentioned shift was especially notable from the post-salted stage in SM, suggesting that this
external muscle underwent a faster ripening than BF. The emergence of the T2´ population during
11
181
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
the maturing stage was linked to protons related to the fat phase and to the progress of fat
marbling.
Moreover, the main population, T21 showed a strong positive linear dependence with water
content and it was negative related to the evolution of the breaking strength during processing.
The study has demonstrated the potential of NMR T2 relaxometry to elucidate associations
between the biophysical state of intrinsic water and processing of dry-cured ham, being possible
to monitor the whole manufacturing process.
Acknowledgments
The present work received financial support from project AGL2010-19158 funded by the Spanish
Secretary of State of Research, Development and Innovation within the Ministry of Economy and
Competitiveness. A.B.G.G. was awarded a grant (BES-2011-047485) from the same institution.
12
182
IV. Resultados
García-García et al., 2018, IV.1.2, 1-14
References
Albarracin, W., Sánchez, I.C., Grau, R., & Barat, J.M. (2011). Salt in food processing; usage and
reduction: A review. International Journal of Food Science and Technology, 46, 1329-1336.
Andersen, R.H., Andersen, H.J., & Bertram, H.C. (2007). Curing-induced water mobility and distribution
within intra- and extra-miofibrilar spaces of three pork qualities. International Journal of Food Science and
Technology, 42, 1059-1066.
Andrés, A., Cava, R., Ventanas, J., Thovar, V., & Ruiz, J. (2004). Sensory characteristics of Iberian ham:
Influene of salt content and processing conditions. Meat Science, 68, 45-51.
AOAC (2006). Official methods of analysis (18th ed.). Washington D. C.: Association of Official Analytical
Chemists.
Bertram, H.C., Dønstrup, S., Karlsson, A.H., & Andersen, H.J. (2001). Continuous distribution analysis
of T2 relaxation in meat- an approach in the determination of water holding capacity. Meat Science, 60, 279-
285.
Bertram, H.C., Karlsson, A.H., Rasmussen, M., Pedersen, O.D., & Andersen, H.J. (2001). Origin of
multiexponential T2 relaxation in muscle myowater. Journal of Agricultural and Food Chemistry, 49, 3092-
3100.
Bertram, H.C., Purslow, P.P., & Andersen, H.J. (2002). Relationship between meat structure, water
mobility, and distribution: a low-field nuclear magnetic resonance study. Journal of Agricultural and Food
Chemistry, 50, 824-829.
Bertram, H.C., & Andersen, H.J. (2004). Applications of NMR in meat science. Annual Report on NMR
Spectroscopy, 53,158-202.
Cornejo, F., & Chinachoti, P. (2003). NMR in foods: oppotunity and challanges. in Magnetic Resonance
in Food Science: Latest Developments. P. S. Belton, A. M. Gil, G. A. Webb, and D. Rutledge, ed. Royal
Society of Chemistry, Cambridge.
Fantazzini, P., Gombia, M., Schembri, P., Simoncini, N., & Virgili, R. (2009). Use of magnetic resonance
imaging for monitoring Parma dry-cured ham processing. Meat Science, 82, 219-27.
Fullerton, G. D., & Cameron, I. L. (1988). Relaxation of Biological Tissues. Pages 115-155 in Biomedical
Magnetic Resonance Imaging. F. W. Werhli, D. Shaw, and J. B. Kneeland, ed. VCH Publishers Inc., New
York.
García-García, A.B., Fernández-Valle, M.E., Castejón, D., Escudero, R., Cambero, M.I. (2018). Use of
MRI as a predictive tool for physicochemical and rheological features during cured ham manufacturing. Meat
Science (under revision)
Hermansson, A. M. (1983). Protein functionality and its relation to food microstructure. Plant Foods for
Human Nutrition (Formerly Qualitas Plantarum), 32(3), 369-388.
Herrero, A.M., de la Hoz, L., Ordóñez, J.A., Herranz, B., Romero de Ávila, M.D., & Cambero, M.I. (2007).
Breaking strength of dry-fermented sausages and their correlation with texture profile analysis (TPA) and
physico-chemical characteristics. Meat Science, 77, 331-338.
Herrero, A.M., de la Hoz, L., Ordóñez, J.A., Herranz, B., Romero de Ávila, M.D., & Cambero, M.I. (2008).
Tensile properties of cooked meat sausages and their correlation with texture profile analysis (TPA)
parameters and physico-chemical characteristics. Meat Science, 80, 690-696.
Hills, B. P., Manning, C. E., & Ridge, Y. (1996). New theory of water activity in heterogeneous systems.
Journal of the Chemical Society, Faraday Transactions, 92(6), 979-983.
Honikel, K.O. (1998). Reference methods for the assessment of physical characteristics of meat. Meat
Science, 49, 447-457.
Hoz, L., Cambero, M.I., Cabeza, M.C., Herrero, A.M., & Ordóñez, J.A. (2008). Elimination of Listeria
monocytogenes from vacuum-packaged dry-cured ham by E-beam irradiation. Journal of Food Protection,
71, 2001-2006.
Huff-Lonergan, E., & Lonergan, S. (2005). Mechanisms of waterholding capacity of meat: the role of
postmortem biochemical and structural changes. Meat Science, 71, 194-204.
Jager, I., & Fratz, P. (2000). Mineralized collagen fibrils: A mechanical model with a staggered
arrangement of mineral particles. Biophysics Journal, 79,1737-1746.
13
183
IV. Resultados
García-García et al., 2018, enviado a Food Structure, IV.1.2, 1-14
Landis, W., J., Librizzi, J.J., Dunn, M.G., & Silver, F.H. (1995). A study of the relationship between mineral
content and mechanical properties of turkey gastrocnemius tendon. Journal of Bone Mineral Research, 10,
859-867.
Larrea, V., Perez-Munuera, I., Hernando, I., Quiles, A., & Lluch, M.A. (2007a). Chemical and structural
changes in lipids during the ripening of Teruel dry-cured ham. Food Chemistry, 102, 494-503.
Larrea, V., Perez-Munuera, I., Hernando, I., Quiles, A., Llorca, E., & Lluch, M.A. (2007b). Microstructural
changes in Teruel dry-cured ham during processing. Meat Science, 76, 574-82.
Maltin, C., Balcerzak, D., Tilley, R., & Delday, M. (2003). Determinants of meat quality: Tenderness.
Proceedings of the Nutrition Society, 62, 337-347.
Mathlouthi, M. (2001). Water content, water activity, water structure and the stability of foodstuffs. Food
Control, 12, 409-417.
Parolari, G., Virgili, R., & Schivazappa, C. (1994). Relationship between cathepsin B activity and
compositional parameters in dry-cured hams of normal and defective texture. Meat Science, 38, 117-122.
Purslow, P. (1985). The physical basis of meat structure: Observations on the fracture behavior of cooked
bovine M. semitendinosus. Meat Science, 12, 39-60.
Romero de Ávila, M.D, Escudero, R., Ordóñez, J.A., & Cambero, M.I. (2014). Weibull analysis
characterizes the breaking properties of dry-cured ham slices. Meat Science, 97, 451-458.
Ruan, R. R., & Chen, P. L. (1998). Water in Foods and Biological Materials: A Nuclear Magnetic
Resonance Approach. 1st edition ed. Technomic Publishing Company, Inc, Lancaster, Pennsylvania.
Ruiz-Ramírez, J., Arnau, J., Serra, X., & Gou, P. (2005). Relationship between water content, NaCl
content, pH and texture parameters in dry-cured muscles. Meat Science, 70, 579-587.
Santos, C., Hoz, L., Cambero, M.I., Cabeza, C., & Ordóñez, J.A. (2008). Enrichment of dry-cured ham
with -linolenic acid and -tocopherol by the use of linseed oil and -tocopheryl acetate in pig diets. Meat
Science, 80, 668-674.
Serra, X., Ruiz-Ramírez, J., Arnau, J., & Gou, P. (2005). Texture parameters of dry-cured ham m. biceps
femoris samples dried t different levels as a function of water activity and water content. Meat Science, 69,
249-254.
Toldrá, F. (2006). Dry-cured ham. En: Handbook of Food Science Technology and Engineering. Ed.
CRC Press, Boca Raton, E.E.U.U.: pp 164.1-164.11
Wheeleer, T.L., Shakeldord, S.D., & Koohmaraie, M. (2000). Variations in proteolysis, sarcomere length,
collagen content, and tenderness among major pork muscles. Journal of Animal Science, 78, 958-965.
Zeugolis, D.I., Paul, R.G., & Attenburrow, G. (2008). Post-self-assembly experimentation on extruded
collagen fibers for tissue engineering applications. Acta Biomaterials, 4, 1646-1656.
14
184
IV. Resultados
Trabajo 3
Estudio del proceso de proteólisis durante la maduración de modelos cárnicos elaborados
a base de carne picada
187
IV. Resultados
Article
pubs.acs.org/JAFC
ABSTRACT: Proton NMR relaxation analyses were performed in sausage model systems (SMS) at different manufacturing
times (0, 1, 3, 5, 7, and 9 days) to evaluate changes in water distribution and mobility. Three different water populations were
identified, T2b (5−10 ms), T21 (30−70 ms), and T22 (100−300 ms), and the progress of ripening could be followed as a shift
toward shorter relaxation times. In addition, the combined effect of adding commercial proteases (Pronase E and aspartyl
proteinase) on protein breakdown and structural integrity of sausage models (SMS+P) was investigated, resulting in the
formation of a more fluid and less organized meat matrix that led to changes in water populations T2b2 and T22 compared with
SMS. A very different protein degradation pattern between SMS and SMS+P was observed by means of SDS-PAGE and
fluorescamine assay, supporting that some degree of protein aggregation is needed for the presence of the T22 population in
fermented sausages.
KEYWORDS: sausage models, ripening, proteases, NMR relaxation, water distribution, meat fermentation
■ INTRODUCTION
In many foods the intrinsic distribution and mobility of water in
and water properties in sausage models by means of low-field
NMR. To obtain additional data that may contribute to a
meat have a profound influence on essential quality attributes further understanding of the relationship between proteolysis
such as juiciness, tenderness, firmness, and appearance.1 and the NMR T2 relaxation pattern, the study also included
Fermented meat sausage manufacture entails three main sodium dodecyl sulfate−polyacrylamide gel electrophoresis
steps: formulation, fermentation, and ripening/drying.2 During (SDS-PAGE) to study changes in myofibrillar proteins during
these stages, different technological processes such as chopping, ripening, because SDS-PAGE has proved to be a suitable tool
to study the proteolysis process in fermented sausages.15−17
■
salting, and pH decline result in changes in the structure of the
native muscle proteins that may affect water properties.3 It is MATERIALS AND METHODS
well-known that compounds resulting from protein breakdown
and those generated from amino acid transformation are Model Formulation and Manufacture. Two different batches of
sausage model systems were developed: SMS (model systems without
involved in flavor development in meat products.4 One of the enzyme) and SMS+P (model systems with proteases). SMS with pork
most established methods for accelerating the ripening of (Longissimus lumborum) were produced, consisting of (% w/w)
fermented products involves the addition of exogenous minced lean meat, fat (19%), and curing agents [NaCl (2.5%), KNO3
proteases to accelerate the reactions that modify texture and (0.02%), NaNO2 (0.01%), dextrose (0.8%), lactose (1%), and dextrin
generate flavor.5,6 Consequently, it is essential to elucidate how (1.6%)]. SMS meat mixture was treated with Pronase E from
these enzymes affect the microstructure and, therefore, the Streptomyces griseus and aspartyl proteinase from Aspergillus oryzae at
distribution and mobility of water in fermented sausages to concentrations of 300/100 enzyme units/kg of meat mixture,
understand how the process can be optimized. respectively, to obtain the SMS+P batch. The protease addition was
Low-field NMR relaxation techniques (LF-NMR) can performed according to previous studies conducted with the use of
selective exogenous proteases.4,5 The amount of enzyme added in the
provide information about the state of water in foods, and present study has proved to be suitable to obtain acceptable products
numerous applications in a variety of food matrices have been with viable final quality attributes. No starter culture was used in order
reported.7−10 In fresh meat, NMR relaxation data were found to develop a natural fermentation process. Lean meat and fat were
to indicate the existence of more than one type of water chopped, using an MR-9 mincer (Garhe, Amorebieta, Spain), then
population.11−13 It was also shown that the changes in water mixed with the rest of the ingredients at cool temperature (4 °C), and,
attributes as a function of drying as detected by LF-NMR were after a resting period of 24 h at 4 °C, the sausage mixtures were stuffed
useful to elucidate the microbial safety of fermented meat into 50 mm diameter collagen casings, in pieces of 400 g. Eighteen
products.14 Nevertheless, knowledge about water properties in independent units were manufactured for each batch and placed in an
fermented sausages is still limited, and no detailed inves-
tigations on proteolysis have been reported. Consequently, the Received: February 4, 2015
aim of this work was to study changes in water mobility and Revised: March 9, 2015
distribution during the manufacturing process of sausage model Accepted: March 9, 2015
systems and to elucidate the relationship between proteolysis Published: March 9, 2015
FED 53 ripening cabinet (Binder, Tuttlingen, Germany) for a period protein concentration was adjusted with water to give a final
of 9 days. During the first 24 h the units were kept at 22 °C and the concentration of 2 mg/mL and diluted 1:1 with SDS-PAGE sample
relative humidity of 95%. Within 48 h, storage conditions were buffer [2% SDS, 20% glycerol, 20 mM Tris-Cl, pH 6.8, 2 mM
gradually reduced to 89% humidity and 17 °C. Models were stored at ethylenediaminetetraacetic acid (EDTA), 160 mM dithiothreitol
12 °C the remaining days. Sampling was carried out at different (DTT), and 0.1 mg/mL bromphenol blue] to give a final
ripening days: 0 (24 h after batter meat preparation and just before concentration of 1 mg/mL. The samples were heated at 100 °C for
stuffing into casings), 1, 3, 5, 7, and 9 days. A total of three units of 5 min prior to electrophoresis. SDS-PAGE was carried out under
each batch were collected for each of the studied times. reducing conditions,24 using Criterion TGX (any kDa) precast ready-
Physicochemical analyses were performed at each sampling time made gels (Bio-Rad Laboratories, Hercules, CA, USA). The amount of
and, in any case, before freezing. Then sausages were vacuum packaged protein solution injected into the electrophoresis gels was 20 μL in
and stored at −20 °C until further analyses. each lane. After electrophoresis, the gels were covered with fixative
Physicochemical Analysis. Water activity (aw), pH, and dry solution (40% methanol, 10% acetic acid) and then stained using
matter (DM) were determined in triplicate samples of each of the Coomassie Brilliant Blue R-250. The molecular weights of the
considered products and ripening times. Water activity (aw) was products of proteolysis were estimated by reference to the relative
measured using a Decagon CX1 hygrometer (Decagon Devices Inc., motilities of standard proteins.
Pullman, WA, USA) at 25 °C. The pH was determined in a Statistical Analysis. Statistical analysis was carried out using
homogenate of the sample with distilled water (1:10 w/v). DM was Statgraphics, ver. 5.1. Plus. The analyses were conducted across the
determined by drying the sample at 110 °C to constant weight, and two manufactured batches (SMS and SMS+P) and six different
the results were expressed as a percentage.18 ripening times (0, 1, 3, 5, 7, and 9 days), so 18 individual units per
NMR Analyses. Sausage models were thawed prior to NMR treatment were analyzed. Data are presented as mean and standard
analysis. To carry out thawing, samples were kept at room temperature deviation (SD). One-way ANOVA was performed. Bartlett’s test was
for approximately 6 h. Sausage behavior was similar in both SMS and used to test the null hypothesis that the standard deviations of data
SMS+P, and syneresis or exudate loss was never observed. Samples (within each of the model systems) were the same. Duncan’s test for
(approximately 4 cm long and 1 cm in diameter, weight approximately multiple mean comparisons procedure was used to determine which
5 g) were placed in a cylindrical glass tube (5 cm long by 14 mm means are significantly different from others (at 95% of confidence
diameter). Before measurement, the sausage samples were thermo- level).
■
stated to 25 °C in a water bath for 15−20 min. Proton NMR T2
relaxation measurements were performed on a Maran Benchtop
Pulsed NMR Analyzer (Resonance Instruments, Witney, UK)
RESULTS AND DISCUSSION
operating at 23.2 MHz and equipped with an 18 mm variable- Physicochemical Parameters. DM, pH, and aw mean
temperature probe where the cylindrical glass tubes were placed. A values and standard deviations for SMS and SMS+P are shown
total of six replicates of each product and ripening time were analyzed in Figure 1. The products showed significant differences (P <
at 25 °C. Transverse relaxation time, T2, was measured using the 0.05) for the three physicochemical parameters studied
Carr−Purcell−Meiboom−Gill sequence (CPMG)19,20 with a τ value
throughout all stages of the manufacturing process. As
(time between 90 °C pulse and 180 °C pulse) of 150 μs. Data from
4096 echoes were acquired as 16 scan repetitions. The repetition time expected, results showed that the lowest DM content (P <
between two succeeding scans was 2 s. The obtained T2 data were 0.05) could be attributed to the samples collected on day 0
analyzed using algorithm21 and carried out in MatLab (The (28.11 ± 0.31% for SMS and 27.16 ± 0.30% for SMS+P). As
Mathworks Inc., Natick, MA, USA) using in-house scripts. Distributed ripening progressed, a steady and considerable increase in DM
exponential fittings are shown in a plot of relaxation amplitude versus was observed, and the highest value was reached at day 9 (56.12
relaxation time, over a predefined range of characteristic relaxation ± 0.98% for SMS and 51.66 ± 0.62% for SMS+P). There were
times. also significant differences (P < 0.05) in pH, so the highest
Measurement of Proteolysis. Degree of proteolysis was value was detected at day 0 (5.64 ± 0.04 for SMS and 5.84 ±
estimated by measurement of free amino terminal groups using the
0.03 for SMS+P). Lactic acid production due to the
fluorescamine assay described earlier.22 One hundred milligrams of
each sample was solubilized in 1 mL of phosphate-buffered saline (0.05 fermentation of sugars by lactic acid bacteria led to the lowest
M phosphate buffer, 0.14 M NaCl, pH 6.7). To 500 μL of 1.67 M pH values at day 3 (around 5.00). From day 5, a slight increase
(24%) trichloroacetic acid in water was added an equal volume of in pH was observed in both batches. This may be due to
solubilized sample, and the mixture was held for 20 min on ice to allow production of free amino acids and ammonia as a result of
precipitation of intact proteins. The precipitated mixture was enzymatic activity.2,25 With the increase in DM, previously
centrifuged (17000g for 20 min) and then filtered through a no. 41 mentioned, aw decreased constantly (from 0.960 ± 0.001 in
filter paper (Whatman PLC, Brentford, UK). A 12.5 μL aliquot of this SMS and 0.953 ± 0.001 in SMS+P to 0.911 ± 0.001 and 0.903
filtrate was mixed with 1 mL of 100 mM borate buffer and 500 μL of ± 0.001, respectively), which results in a microbial stabilization
719 μM fluorescamine in acetone, and 250 μL was then transferred to
of the transformed product. The measured parameters are
each well in a microtiter plate. Fluorescence of the samples was
measured, 18 min after the addition of the fluorescamine, at an within the normal range previously described for dry fermented
excitation wavelength of 390 nm and an emission wavelength of 480 sausages,26,27 taking into account that the products manufac-
nm using a model Synergy 2 Multi-Mode microplate reader (Holm & tured for this study had a shorter ripening time (9 days)
Halby, Brøndby, Denmark). Quantitation was achieved by calculating compared to commercial brands (around 20−28 days), which
leucine equivalents using an external leucine standard curve prepared explains that higher DM values have not been reached at the
as follows: a 0.1 M leucine stock solution was made (327.5 mg of end of the process.
leucine in 25 mL of 1 mM HCl), and from the stock solution six NMR Studies on Sausage Model Systems. Water
different concentrations were made (0.00050−0.0030 M leucine) and Mobility and Distribution throughout Sausage Model
mixed with 10 mL of 1 mM HCl and analyzed as the sausage samples. Manufacturing. Distributed exponential analysis of the NMR
SDS-PAGE. Myofibrillar proteins were extracted by homogeniza-
tion of 1 g of sausage in 5 mL of a solution containing 7 M urea, 2 M T2 relaxation data revealed the presence of three different water
thiourea, and 400 mM Tris-base. The homogenate was centrifuged for populations in SMS (Figure 2A): a minor component, with
20 min at 10000g at 4 °C, and the supernatant containing myofibrillar relaxation times around 1−10 ms (T2b), a major component
proteins was separated. The protein concentration of the myofibrillar around 30−70 ms (T21), and a slower relaxing component
protein extracts were determined by using the Bradford assay.23 The (T22) around 100−300 ms. Chopping results in disintegration
3040 DOI: 10.1021/acs.jafc.5b00660
191
J. Agric. Food Chem. 2015, 63, 3039−3045
IV. Resultados
Table 1. Mean Relaxation Time and Relative Area for the T2 Relaxation Populations T21 and T22 Obtained for Sausage Model
Systems (SMS) and Models with Proteases (SMS+P) on Days 0, 1, 3, 5, 7, and 9 after Productiona
T21 T22
sample/ripening time (days) relative area relaxation time relative area relaxation time
SMS/0 96.3 ± 0.3 a,α 56.7 ± 0.9 a,β
SMS/1 93.7 ± 0.3 b,β 52.4 ± 1.3 b,β 3.2 ± 0.4 c 228.9 ± 71.1 a
SMS/3 84.4 ± 0.5 c,β 44.4 ± 1.2 c,β 12.4 ± 0.5 a 260.5 ± 12.0 a
SMS/5 85.0 ± 0.9 c,β 38.3 ± 1.2 d,β 11.7 ± 0.7 b 215.8 ± 8.0 a
SMS/7 85.6 ± 0.6 c,β 37.1 ± 1.4 d,β 11.5 ± 0.3 b 202.7 ± 7.0 b
SMS/9 84.3 ± 0.7 c,β 30.0 ± 1.4 e,β 12.8 ± 0.2 a,α 149.1 ± 21.3 c,β
mentioned above, the temperature and the pH descent are key with large cavities where the T22 can be related to the mobile
factors in promoting the structural changes in the protein water in the cavities. Distributed T2 relaxation times in SMS+P
matrix. compared to that obtained for SMS showed how the structural
Effect of Proteases on Sausage Models Structure. integrity was changed by the introduction of proteases. In fact,
Figure 2B shows distributed T2 relaxation times obtained on the ripening-induced shift of T21 and T22 populations in SMS+P
SMS+P. T2b1, T2b2, and T21 populations were identified for all is not as noticeable as it is in SMS, suggesting that water
ripening times. Previous studies14,33 have related the addition of immobilization is not taking place until day 7, as affected by
salt to the appearance of a T2b2 population (5−10 ms), absent protein degradation.
in fresh meat, and also identified in this batch throughout the Time constants and relative areas (corresponding to the
entire manufacturing process. The presence of this new proportion of protons represented by the specific T 2
population could be associated with a different distribution of population) of the two main components (T21 and T22) for
salt in both batches. For the SMS batch, it was possible to SMS and SMS+P are shown in Table 1. Significant differences
distinguish the T2b2 component only on days 0, 1, and 3 of were detected by comparison between T21 relaxation times
production, although it was not a clear peak and appeared corresponding to SMS and SMS+P throughout ripening time,
almost merged with T2b1. In this batch, ripening took place on a so that SMS+P consistently maintained higher T21 values over
conventional basis, which means that the myofibrils were able time. These findings can be related to the fact that the water
to swell and retain water, and the added salt was easily absorbed content was also higher in SMS+P than SMS for the same
or incorporated into protein structure as ripening progressed. studied stage, ranging from 71.9 ± 0.3 to 43.9 ± 0.9 g of water/
The proteins dissolved as a result of the mincing and the 100 g of sample in SMS and from 72.8 ± 0.3 to 48.3 ± 0.6 g of
addition of salt produce filament-shaped aggregates that water/100 g sample in SMS+P. The drying process resulted in a
interact and contribute to the stability of the gel.2 However, decrease of the relative areas of both populations, although the
in SMS+P models, protein breakdown prevented an efficient decrease was more progressive in SMS and values were higher
absorption of salt by myofibrils in such a way that it together in SMS+P, especially from day 3 on.
with water constituted an independent compartment with faster Proteolysis in Model Samples As Measured by
relaxation time, different from the major myofibrillar protein Increase in Free Amino Groups. Proteolysis was quantified
component (T21) that could be clearly identified during the by assaying for protein breakdown products through
entire process. The presence of the T2b2 population suggests determination of free amino terminals (N-terminal + lysine
that the protein matrix did not present a strong and well- side chains, as well as other amines that may be present).
organized structure. The T22 component was not present in Although the concentration of free amino groups increased for
SMS+P samples until day 9, which constitutes an important both batches over time (Figure 3), higher values were found for
difference compared to T2 distributions in SMS. The absence of SMS+P (concentration varied from 657.83 ± 1.54 to 1487.09 ±
the T22 component suggests a nonaggregation behavior of the 5.06 mg Leu equiv/DM at days 0 and 9, respectively) than for
proteins. Partial denaturation of protein is required as an initial SMS (values from 290.35 ± 3.71 to 487.78 ± 0.93 mg Leu
step to achieve the ordered protein−protein interaction and equiv/DM, respectively), showing that the proteolysis was
aggregation;34 however, the hydrolysis induced by proteases more prominent in the protease-added batch, SMS+P. This
prevents the proper establishment of the protein network and finding is in agreement with previous works4,5 in which the
severely affects final texture resulting in a softening of the nitrogen fractions achieved higher values in protease-added
product, as previously reported.4,5 Consequently, T2b2 and T22 batches than those of the control. Significant differences were
populations are both related to the absence of a strong and firm found over time for both batches regarding free amino group
protein matrix. Proteolysis can also produce an extensive concentration, although the lower increase occurred toward the
unfolding of the protein inducing the formation of a network end of the process for SMS+P, during days 7 and 9 (5.5%),
3042 DOI: 10.1021/acs.jafc.5b00660
Figure 4. SDS-PAGE electrophoretograms of myofibrilar proteins throughout the ripening of sausages. Lanes: 1, protein standards; 2−7, sausage
model systems (SMS), days 0, 1, 3, 5, 7, and 9; 8−13, models with proteases (SMS+P), days 0, 1, 3, 5, 7, and 9. MHC, heavy myosin chain; α-act., α-
actinin; TN-T, troponin-T; TM, tropomyosin; MLC 1, 2, 3, myosin light chains 1, 2, and 3; TN-I, troponin I; TN-C, troponin C.
193
J. Agric. Food Chem. 2015, 63, 3039−3045
IV. Resultados
■
intensely stained at the end of ripening than at day 0. The loss
in myofibrillar protein solubility has been previously described REFERENCES
by other researchers.15,40 A dramatically different protein band
pattern was observed for SMS+P, where high-molecular-weight (1) Trout, G. R. Techniques for measuring water-binding capacity in
muscle foods − a review of methodology. Meat Sci. 1988, 23, 235−
compounds cannot be distinguished as a result of the 252.
proteolysis to which samples have been subjected. However, (2) Ordóñez, J. A.; Hierro, E. M.; Bruna, J. M.; Hoz, L. Changes in
protein insolubilization during drying was clearly evident. the components of dry-fermented sausages during ripening. Crit. Rev.
These results, together with proteolysis measurements (Figure Food Sci. 1999, 34, 329−367.
3), prove that protein denaturation on SMS+P was very intense (3) Tornberg, E. Effects of heat on meat proteins-Implications on
from day 0; consequently, the molecular weight of the protein structure and quality of meat products. Meat Sci. 2005, 70, 493−508.
chains is too low, and they solubilize completely before (4) Díaz, O.; Fernández, M.; García de Fernando, D. G.; de la Hoz,
aggregating. The solubility of myofibrillar proteins in SMS+P L.; Ordóñez, J. A. Proteolysis in dry fermented sausages: the effect of
decreased with ripening time due to the dehydration process selected exogenous proteases. Meat Sci. 1997, 46, 115−128.
and the increase in salt concentration, which leads to the (5) Díaz, O.; Fernández, M.; García de Fernando, G. D.; De la Hoz,
L.; Ordóñez, J. A. Effect of the addition of Pronase E on the
aggregation of protein chains and the formation of higher
proteolysis in dry fermented sausages. Meat Sci. 1993, 34, 205−216.
molecular weight compounds, so the bands on days 7 and 9 (6) Fernández, M.; Ordóñez, J. A.; Bruna, J. M.; Herrans, B.; de la
were more intensely stained than on previous days. With regard Hoz, L. Acelerated ripening of dry fermented sausages. Food Sci.
to T2 distributions, the stronger protein chain association at day Technol. 2000, 11, 201−209.
9 compared with earlier stages by reducing the pore size of the (7) Capitani, D.; Mannina, L.; Proietti, N.; Sobolev, A. P.; Tomassini,
matrix due to the water loss, together with the extensive A.; Miccheli, A.; DiCocco, M. E.; Capuani, G.; DeSalvador, F. R.;
unfolding suffered by proteins, could be associated with the Delfini, M. Metabolic profiling and outer pericarp water starte in
appearance of T22 population, suggesting that some degree of Zespri CI.GI and Hayward Kiwifruits. J. Agric. Food Chem. 2013, 61,
aggregation is needed for the presence of the water population 1727−1740.
T22 in dried sausages. The fact that the T2b2 population was also (8) Li, T.; Rui, S.; Li, W.; Chen, X. H.; Jiang, M.; Dong, M. S. Water
distribution in tofu and application of T2 relaxation measurements in
present in the final stages supports the hypothesis that a true determination of tofu’s water holding capacity. J. Agric. Food Chem.
coagulation is not taking place and the structural modification is 2014, 62, 8594−8601.
not complete and, further, related to the dehydration taking (9) Hager, A. S.; Bosmans, G. M.; Delcour, J. A. Physical and
place during ripening. molecular changes during the storage of gluten-free rice and oat bread.
In conclusion, the present study revealed that (i) changes in J. Agric. Food Chem. 2014, 62, 5682−5689.
water mobility and distribution during the manufacture of (10) Greiff, K.; Fuentes, A.; Aursand, I. G.; Erikson, U.; Masot, R.;
sausage model systems were detectable by means of LF-NMR Alcaniz, M.; Barat, J. M. Innovative non-destructive measurements of
relaxation, where ripening produced a shift toward faster water activity and the content of salts in low-salt hake minces. J. Agric.
relaxation times related to the reduction of the intrinsic pore Food Chem. 2014, 62, 2496−2505.
size of the meat matrix; and (ii) the distribution of water in (11) Bertram, H. C.; Dønstrup, S.; Karlsson, A. H.; Andersen, H. J.
Continuous distribution analysis of T2 relaxation in meat- An approach
sausage models was significantly affected by the addition of in the determination of water holding capacity. Meat Sci. 2001, 60,
proteases to the meat batter, supporting the role of myofibrillar 279−285.
proteins in meat coagulation and suggesting that some (12) Bertram, H. C.; Karlsson, A. H.; Rasmussen, M.; Pedersen, O.
aggregation degree is needed for the presence of a T22 D.; Andersen, H. J. Origin of multiexponential T2 relaxation in muscle
population in fermented sausages. Finally, the extinction of myowater. J. Agric. Food Chem. 2001, 49, 3092−3100.
the T22 population in meat systems could be associated with a (13) Bertram, H. C.; Andersen, H. J. Applications of NMR in meat
high enzymatic proteolysis degree and, consequently, the science. Annu. Rep. NMR Spectrosc. 2004, 53, 158−202.
presence of great amounts of free amino groups. (14) Møller, S.; Gunvig, A.; Bertram, H. C. Effect of starter culture
■
and fermentation temperature on water mobility and distribution in
fermented sausages and correlation to microbial safety studied by
AUTHOR INFORMATION nuclear magnetic resonance relaxometry. Meat Sci. 2010, 86, 462−467.
Corresponding Author (15) García de Fernando, D. G.; Fox, P. F. Study of proteolysis
during the ripening of a dry fermented pork sausage. Meat Sci. 1991,
*(H.C.B.) Phone: +45 87 158 353. E-mail: HanneC.Bertram@ 30, 367−383.
food.au.dk. (16) Hughes, M. C.; Kerry, J. P.; Arendt, E. K.; Kenneallu, P. M.;
Funding McSweeney, P. L.; O’Neill, H. Characterization of proteolysis during
The present work received financial support from Project the ripening of semi-dry fermented sausages. Meat Sci. 2002, 62, 205−
AGL2010-19158 funded by the Spanish Ministry of Economy 2016.
(17) Ohashi, K.; Negishi, H. Hydrolysis of pork myofibrillar proteins
and Competitivity − General Directorate of Scientific Research,
during fermentation using starter cultures of Lactobacillus bulgaricus
and A.B.G.G. was awarded a grant (BES-2011-047485) from and Streptococcus thermophilus. Food Sci. Technol. Res. 2014, 20, 679−
the same institution. H.C.B. acknowledges the Danish Research 685.
Council FTP for financial support through Project “Advances (18) Official Methods of Analysis of the Association of Official Analytical
in food and nutrition research through implementation of Chemists, 18th ed., 1st rev.; AOAC International: Gaithersburg, MD,
metabolomics technologies (274-09-0107)”. USA, 2006.
(19) Carr, H. Y.; Purcell, E. M. Effects of diffusion on free precession (38) Porcio, M. A.; Pearson, A. M. Improved resolution of
in nuclear magnetic resonance experiments. Am. J. Physiol. 1954, 94, myofibrillar proteins with sodium dodecyl sulfate-polyacrilamide gel
630−638. electrophoresis. Biochim. Biophys. Acta 1977, 490, 27−34.
(20) Meiboom, S.; Gill, D. Modified spin-echo method for measuring (39) Roseiro, L. C.; Santos, C.; Sol, M.; Borges, M. J.; Anjos, M.;
nuclear times. Rev. Sci. Instrum. 1958, 29, 688−691. Goncalves, H.; Carvalho, A. S. Proteolysis in Painho de Portalegre dry
(21) Butler, J. P.; Reeds, J. A.; Dawson, S. V. Estimating solutions of fermented sausage in relation to ripening time and salt content. Meat
1st kind integral equations with nonnegative constraints and optimal Sci. 2008, 79, 784−794.
smoothing. SIAM J. Numer. Anal. 1981, 18, 381−397. (40) Astiasarán, I.; Villanueva, R.; Bello, J. Analysis of proteolysis and
(22) Jansson, T.; Clausen, M. R.; Sundekilde, U.; Eggers, N.; protein insolubility during the manufacture of some varieties of dry
Nyegaard, S.; Larsen, L.; Ray, C.; Sundgren, A.; Andersen, H. J.; sausage. Meat Sci. 1990, 28, 111−117.
Bertram, H. C. Lactose-hydrolyzed milk is more prone to chemical (41) Wardlaw, F. B.; Skelley, G. C.; Johnson, M. G.; Acton, J. C.
changes during storage than conventional UHT milk. J. Agric. Food Changes in meat components during fermentation, heat processing
and drying of a summer sausage. J. Food Sci. 1973, 38, 1228−1231.
Chem. 2014, 62, 7886−7896.
(23) Bradford, M. M. A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 1976, 72, 248−254.
(24) Laemmli, U. K. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 1970, 227, 680−
685.
(25) Lücke, F. K. Fermented sausages. In Microbiology of Fermented
Foods; Wood, B. J. B., Ed.; Blackie Academic and Professional: New
York, 1998; pp 441−483.
(26) Ordóñez, J. A.; Hoz, L. Embutidos crudos curados. Tipos.
Fenomenos madurativos. Alteraciones. In Enciclopedia de la Carne y los
Productos Cárnicos; Martin and Macias: Cáceres, Spain, 2001; Vol. II,
pp 1063−1090.
(27) Herrero, A. M.; Ordóñez, J. A.; Romero de Á vila, M. D.;
Herranz, B.; de la Hoz, L.; Cambero, M. I. Breaking strength of dry
fermented sausages and their correlation with texture profile analysis
(TPA) and physico-chemical characteristics. Meat Sci. 2007, 77, 331−
338.
(28) Bertram, H. C.; Purslow, P. P.; Andersen, H. J. Relationship
between meat structure, water mobility, and distribution: a low-field
nuclear magnetic resonance study. J. Agric. Food Chem. 2002, 50, 824−
829.
(29) Venturi, L.; Rocculi, P.; Cavani, C.; Placucci, G.; Rosa, M. D.;
Cremonini, M. A. Water absorption of freeze-dried meat at different
water activities: a multianalytical approach using sorption isotherm,
differential scanning calorimetry, and nuclear magnetic resonance. J.
Agric. Food Chem. 2007, 55, 10572−10578.
(30) Albarracín, W.; Sánchez, I. C.; Grau, R.; Barat, J. M. Salt in food
processing; usage and reduction: a review. Int. J. Food Sci. Technol.
2009, 46, 1329−1336.
(31) Katsaras, K. Microstructure of fermented sausage. Meat Sci.
1992, 31, 121−134.
(32) Sun, X. D.; Holley, R. A. Factors influencing gel formation by
myofibrillar proteins in muscle foods. Compr. Rev. Food Sci. Food Saf.
2010, 10, 33−51.
(33) Andersen, R. H.; Andersen, H. J.; Bertram, H. C. Curing-
induced water mobility and distribution within intra- and extra-
miofibrilar spaces of three pork qualities. Int. J. Food Sci. Technol. 2007,
42, 1059−1066.
(34) Barbut, S. Importance of fat emulsification and protein matrix
characteristics in meat batter stability. J. Muscle Foods 1995, 6, 161−
177.
(35) Toldrá, F. Biochemical proteolysis basis for improved processing
of dry-cured meats. In Advanced Technologies for Meat Processing;
Nollet, L. M. L., Toldrá, F., Eds.; CRC Press: Boca Raton, FL, USA,
2006; pp 329−354.
(36) Toldrá, F. The role of muscle enzymes in dry-cured mat
products with different drying conditions. Trends Food Sci. Technol.
2004, 17, 164−168.
(37) Andersson, K.; Andersson, A.; Tornberg, E. Microstructure as
related to the water-holding, textural and sensory properties of
emulsion sausages. In Food Colloids − Proteins, Lipids and
Polysaccharides; Dickinson, D., Bergenstahl, B., Eds.; Royal Society of
Chemistry: Cambridge, UK, 1997; pp 29−42.
195
J. Agric. Food Chem. 2015, 63, 3039−3045
IV. Resultados
Trabajo 4
Distribución y movilidad de agua durante el proceso de elaboración de embutidos crudos
curados y su relación con características fisicoquímicas y reológicas
197
IV. Resultados
199
IV. Resultados
Eur Food Res Technol (2017) 243:455–466
DOI 10.1007/s00217-016-2759-0
ORIGINAL PAPER
Received: 18 April 2016 / Revised: 16 June 2016 / Accepted: 16 July 2016 / Published online: 6 August 2016
© Springer-Verlag Berlin Heidelberg 2016
Abstract Changes in the mobility and distribution of water Keywords Fermented sausages · Ripening · NMR
in dry-fermented sausages Spanish type (“chorizo” and relaxation · Physicochemical features · Textural properties
“salchichón”, 20 and 40 % fat) throughout ripening were
detected by NMR T2 relaxation. Three to four different
water populations (T2b1, T2b2, T21 and T22) were detected. Introduction
Multivariate cluster analysis revealed a clear discrimina-
tion between samples of different ripening times, and sig- A large variety of dry-cured meat products are produced
nificant differences were found in the physicochemical and consumed in Spain. Dry-fermented sausages, in par-
parameters, T2 relaxation data and textural properties of ticular, are considered traditional products with a high gas-
sausages according to the production day (0, 2, 4, 7, 11 and tronomic value for consumers worldwide. Dry-fermented
14). Pearson product-moment correlation coefficient (PCC) sausages consist of a mixture of meat and fat particles, salt,
was used to establish the degree of linear dependence curing agents, spices and condiments, which have been
between T2 relaxation–physicochemical characteristics and stuffed into casings, fermented and dried [1]. Fermenta-
T2 relaxation-textural properties. A strong correlation was tion refers to the breakdown of carbohydrates present in
found between the main component T21 and the water con- the meat batter, mainly to lactic acid, due to the action of
tent (r = 0.93), T21 and aw (r = 0.87), as well as between lactic acid bacteria (LAB), which results in a pH decrease
T21 and hardness (r = −0.78), although the other two at this stage [2]. The ongoing acidification favors color for-
water populations, mainly T2b, showed a lower degree of mation and muscle protein coagulation, increasing firmness
association with the physicochemical and textural param- and cohesiveness of the product [3]. In the final ripening/
eters. Consequently, the study demonstrated that LF-NMR drying phase, sausages are further dried, and the final tex-
relaxometry is a promising technique to elucidate associa- ture and flavor of the product are developed as a result of
tions between the biophysical state of intrinsic water and different biochemical reactions affecting macromolecules
progression in textural properties of dry-fermented meat (proteolysis, lipolysis and lipid oxidation) during this
products. period [4]. Given the nonhomogeneous nature of meat sau-
sages, the study of how water is distributed within the prod-
uct offers both theoretical and practical interest, since the
structural properties and the final quality of the sausages
are determined by the type of interaction water may have
* Hanne Christine Bertram with solutes present in the product, mainly water binding
[email protected] with proteins, since meat proteins represent the main func-
1 tional ingredient in meat products. Different studies have
Departamento de Nutrición, Bromatología y Tecnología
de los Alimentos, Facultad de Veterinaria, Universidad been previously conducted in order to study distribution
Complutense de Madrid, 28040 Madrid, Spain and mobility of water in meat systems by means of nuclear
2
Department of Food Science, Aarhus University, Research magnetic resonance (NMR) T2 relaxation. Proton relaxa-
Centre Aarslev, Kirstinebjergvej 10, 5792 Årslev, Denmark tion is governed by molecular movement and interaction
13
201
IV. Resultados
456 Eur Food Res Technol (2017) 243:455–466
between nuclei, and T2 relaxation provides valuable infor- get a product “salchichón type” (g or ml/Kg of product):
mation about the nuclei environment [5]. It has been previ- sodium chloride (24), lactose (30), dextrose (5), black
ously proved [6] that NMR relaxometry is a useful tech- pepper (2.5) and water (10). For “chorizo type,” sodium
nique to study the drying process and to elucidate microbial chloride (24), dextrose (7), paprika pepper (12), garlic (4),
safety in fermented meat products. A recent study [7] has oregano (2) and water (10) were added. Sausages were
been also conducted in order to establish the influence ripened for 14 days, and the process was conducted using
of lipid type in water and fat distribution in these meat the conditions previously described by García-García [8]
derivatives. Moreover, our group has recently conducted (Fig. 1). Sampling was carried out at different times: days 0
a T2 relaxometry study to elucidate structural changes in (24 h after batter meat preparation and just before stuffing
the protein matrix according to the ripening time and the into casings), 2, 4, 7, 11 and 14. Eighteen independent units
addition of proteases in fermented sausage model systems were manufactured for each batch (72 units were analyzed
[8]. Nevertheless, knowledge about water properties in in total, considering the four mentioned batches). Three dif-
fermented sausages is still only partially explored, despite ferent sausages were collected of each product and manu-
the fact that this type of meat product is one of the most facturing time. Physicochemical and texture analyses were
valuable and consumed meat-based product around the immediately performed, and then units were vacuum-pack-
world. In addition, given the importance of texture in the aged and stored at −20 °C until NMR analyses.
final product, and taking into account the effect of fermen-
tation and drying process on meat structure, it is interest- Physicochemical analysis
ing to evaluate the changes in textural parameters through-
out the manufacturing process, since it may contribute to The water activity (aw) measurements were carried
understand the effect of processing on products, probe the out at 25 °C using a Decagon CX1 hygrometer (Deca-
system structure and reveal critical aspects of food texture gon Devices Inc., Pullman, WA, USA). The dry matter
[9]. Texture profile analysis (TPA) is the most commonly (DM) of the samples was determined by drying the sam-
used instrumental method to study texture of foods nowa- ple at 110 °C to constant weight, and the results were
days, and it has been previously used by different authors expressed as a percentage [18]. The pH was determined
to study sausages [10–17]. However, TPA parameters have in a homogenate of the sample with distilled water (1:10)
not been related to T2 relaxation data, so far. Additionally, (w/v), using a Crison Digit-501 pH meter (Crison Instru-
since texture is a sensory property which derives from the ments LTD, Barcelona, Spain). Six physicochemical
structure of the product (molecular, microscopic and mac- measurements were performed per batch and ripening
roscopic) it can be related to the relaxometry data, since time. To estimate these parameters, two different portions
T2 provides valuable information about food structure of sample (2-cm thickness slices) were removed from the
[5]. Consequently, the aim of this study was to investigate inner central part of the sausage units, and analyses were
the potential of low-field NMR relaxometry for elucidat- performed in triplicate per slice.
ing changes in water mobility and distribution during the
manufacturing process of dry-fermented sausages “Spanish NMR analysis
type,” relating these results to those obtained from com-
pression texture analysis (TPA) and physicochemical fea- Proton NMR relaxometry data were recorded on a Maran
tures in order to obtain a better understanding of the rela- Benchtop Pulsed NMR Analyzer (Resonance Instruments,
tion between the biophysical state of intrinsic water and Witney, UK) operating at 23.2 MHz and equipped with
product texture characteristics. an 18-mm variable temperature probe. Transverse relaxa-
tion time, T2, was measured using the Carr–Purcell–Mei-
boom–Gill sequence [19, 20]. The T2 measurements were
Materials and methods performed with a τ-value of 150 µs (time between 90° and
180° pulse), and data were acquired as 16 scan repetitions.
Production of dry‑fermented sausages “Spanish type” The repetition time between two succeeding scans was
2 s. The obtained T2 data were analyzed using distributed
Two different dry-fermented sausages “Spanish type” were exponential fitting analysis according to the regularization
produced: “salchichón” (S) and “chorizo” (CH). The differ- algorithm by Butler [21] and carried out in MATLAB (The
ent products were manufactured using a basis of pork meat, Mathworks Inc., Natick, MA, USA) version 6.5 using in-
150 ppm NaNO2, 150 ppm NaNO3 and two different levels house scripts. Sausage units were thawed prior to NMR
of fat (20 and 40 % w/w). Four different batches were pro- analysis at room temperature for approximately 6 h. Six
duced: S1, CH1 (20 % fat) and S2, CH2 (40 % fat). The replicates (approx. 4 cm long, 1 cm in diameter) of each
following ingredients were added to this basis in order to product and ripening time were analyzed at 25 °C.
13
202
IV. Resultados
Eur Food Res Technol (2017) 243:455–466 457
Fig. 1 Dry-fermented sausages production and sampling procedure. S1: Salchichón (20 % fat), CH1: Chorizo (20 % fat); S2: Salchichón (40 %
fat); CH2: Chorizo (40 % fat); PhCh: physicochemical analysis; TPA: texture profile analysis; NMR: nuclear magnetic resonance analysis
13
203
IV. Resultados
458 Eur Food Res Technol (2017) 243:455–466
13
204
IV. Resultados
Eur Food Res Technol (2017) 243:455–466 459
13
205
IV. Resultados
460 Eur Food Res Technol (2017) 243:455–466
Textural properties
13
206
IV. Resultados
Eur Food Res Technol (2017) 243:455–466 461
Cluster analysis
362.1 ± 14.1
296.7 ± 34.0
219.1 ± 20.5
153.2 ± 5.1
355.8 ± 6.1
Relaxation
Table 1 Mean values and standard deviation of relaxation time and relative area for T21 and T22 populations obtained on batches CH2 and S2 on days 0, 2, 4, 7, 11 and 14 after production
b, α
d, α
a, α
a, α
c, α
T22
–
of individuals into homogeneous groups. Sausage samples
13.8 ± 1.1
13.8 ± 0.6
were associated in four different clusters, using the textural
9.4 ± 1.2
7.6 ± 0.7
9.4 ± 0.5
Relative
b, β
b, β
a, α
c, α
a, β
T22
–
ing time is observed along Factor 1, while Factor 2 seems
Relaxation
39.3 ± 2.8
56.7 ± 0.9
44.9 ± 0.6
44.4 ± 0.8
33.5 ± 1.7
23.5 ± 1.2
to group samples according to the sausage type (CH and
S). The centroid method was the procedure used for com-
time
b, α
b, α
d, α
c, α
a, α
f, α
T21
83.0 ± 1.1
82.7 ± 0.9
87.2 ± 1.2
88.7 ± 1.0
86.6 ± 0.7
the concept of an “average sausage” is used to represent
Relative
d, α
d, α
b, α
b, α
c, α
a, α
T21
S2
226.4 ± 17.1
177.2 ± 32.0
132.2 ± 13.3
prising the cluster. The cluster is then assumed to consist
277.1 ± 6.5
Relaxation
d, β
c, α
a, β
a, β
T22
15.9 ± 1.3
11.8 ± 1.0
10.7 ± 0.9
13.5 ± 0.7
a, b, c, d, e, f: Values with different letters within a column are statistically different (P < 0.05). Effect of ripening time is shown
b, α
d, α
a, α
e, α
c, α
T22
37.3 ± 1.7
54.8 ± 0.8
41.3 ± 0.7
40.6 ± 0.5
29.9 ± 2.2
22.9 ± 1.5
d, α
b, β
e, α
a, α
c, β
f, α
T21
α, β: Values with different letters within a row are statistically different (P < 0.05). Effect of product type is shown
78.0 ± 0.6
81.1 ± 1.5
86.1 ± 1.3
82.8 ± 1.0
into account how different from the overall mean value that
Relative
CH2
b, α
d, β
b, β
c, α
a, α
c, β
T21
189.5 ± 19.9
151.5 ± 15.2
359.9 ± 8.5
b, α
d, α
a, α
c, α
e, α
T22
14.1 ± 0.7
8.9 ± 1.6
6.7 ± 0.4
8.6 ± 0.9
Relative
b, β
b, β
a, β
c, β
a,α
–
31.2 ± 2.5
43.6 ± 0.9
39.9 ± 1.0
25.0 ± 2.8
56.0 ± 0.9
43.7 ± 0.6
e, α
c, α
a, α
f, α
T21
87.4 ± 1.9
139.8 ± 12.8 86.9 ± 1.1
95.7 ± 0.4
82.4 ± 1.1
d, α
c, α
c, α
a, α
T21
S1
263.3 ± 9.9
224.5 ± 6.2
Relaxation
b, β
d, β
c, α
a, β
a, β
T22
15.3 ± 2.4
11.0 ± 0.3
14.7 ± 1.4
Relative
b, α
b, α
c, α
c, α
a, α
T22
and water content, aw, T2b, T22, T21 mean values below the
–
33.0 ± 2.8
39.4 ± 1.1
37.6 ± 1.1
23.9 ± 1.3
54.8 ± 1.7
41.0 ± 0.5
d, α
b, β
e, α
a, α
c, β
f, α
T21
85.4 ± 1.5
81.6 ± 1.8
95.2 ± 0.4
77.3 ± 0.7
Relative
area
b, α
b, β
c, α
a, α
c, β
e, β
T21
14
4
7
0
13
207
IV. Resultados
462 Eur Food Res Technol (2017) 243:455–466
Fig. 4 Textural profile evolution obtained on fermented sausages, ness (adimensional) (×10); gumminess (N); chewiness (Nm) Dif-
batches CH1, CH2, S1 and S2 on days 4, 7, 11 and 14 of produc- ferent symbol within the same line indicates significant differences
tion. Mean values of the following parameters were plotted: hardness (P < 0.05) throughout ripening time
(N); adhesiveness (Ns) × (−10); springiness (m) (×10); cohesive-
Relationship between T2 relaxation time, association between pH and T22. According to T2 distri-
physicochemical characteristics and textural properties butions, a pH decline is required for the presence of T22,
and when the pH reached its lowest value, T22 achieved its
The degree of association between T2 relaxation data maximum.
and physicochemical characteristics as well as the asso- Regarding the textural properties, hardness showed
ciation between T2 relaxation data and textural properties a strong negative correlation with T21 (r = −0.79,
was investigated using bivariate analysis Pearson prod- P < 0.0005) and T22 (r = −0.63, P < 0.0005), and an inter-
uct-moment correlation coefficient (PCC). The correla- mediate correlation with T2b (r = −0.40, P < 0.0005).
tion coefficient for each of the considered pairs is shown Gumminess and chewiness, as secondary characteristics
in Table 3. A strong correlation was found between the dependent on hardness, showed a similar degree of depend-
main T2 population, T21, and physicochemical parameters. ence with the T2 populations. Springiness showed a high
A high association degree is shown between T21 and aw degree of association with T21 (r = −0.56, P < 0.0005),
(r = 0.88, P < 0.0005) and T21 and water content (r = 0.94, a low degree of association with T22 (r = −0.18, non-
P < 0.0005). This high degree of association confirms that significant) and moderate degree of association with T2b
T21 represents a high percentage of the total water content (r = −0.32, P < 0.005). Cohesiveness was highly corre-
within the sample, contrary to T22 and, especially T2b which lated with T21 (r = −0.68, P < 0.0005) and moderately cor-
showed a moderate and low degree of association, respec- related with T22 (r = −0.35, P < 0.005) and T2b (r = −0.40,
tively, with these parameters. Regarding pH, a high degree P < 0.0005). On the other hand, adhesiveness showed a
of association (r = 0.68, P < 0.0005) was found with T21, strong positive correlation with T21 (r = 0.68, P < 0.0005)
since pH has a great influence on the water-holding capac- and T22 (r = 0.60, P < 0.0005) and low degree of associa-
ity of myofibrillar proteins [40]. T2b and T22 populations tion with T2b (r = 0.28, P < 0.005). Intriguingly, signifi-
are less related to pH, but a striking finding is the negative cant associations were also found between water content
13
208
IV. Resultados
Eur Food Res Technol (2017) 243:455–466 463
Fig. 5 Plot of the clusters obtained on data acquired on fermented achieved according to the ripening time (0, 2, 4, 7, 11, 14 days). The
sausage samples using textural features as association parameters. direction of the arrow reflects the progress of ripening
The centroid of each cluster is represented. Sample separation was
and textural parameters. A negative correlation was found products revealed a similar development in T2 relaxation
between water content and hardness, springiness and cohe- behavior throughout ripening, and no significant differ-
siveness, while a positive correlation was found between ences between the two levels of fat included (20 and 40 %)
water content and adhesiveness. were detected.
A strong linear dependence was found between the main
component T21 and water content, T21 and aw, as well as
Conclusions between T21 and hardness, although the other two T2 water
populations, mainly T2b, showed a lower degree of associa-
Changes in water mobility and distribution during dry- tion with the physicochemical and textural parameters. No
fermented sausage production were detectable by means significant differences were found when comparing prod-
of LF-NMR. The drying process induced a shift toward uct types (CH and S), suggesting that the drying-induced
shorter T21 and T22 relaxation times related to the loss of changes was very similar throughout ripening for the two
water and the consolidation of a firmer meat matrix. Sig- types of sausages.
nificant changes were also detected in T2 relaxation popula- Overall, the study has demonstrated the potential of LF-
tions in the 1–10 ms region, since two different populations NMR relaxometry to elucidate associations between the
were found at the beginning of the drying process (T2b1 and biophysical state of intrinsic water and progression in tex-
T2b2), and only a single component (T2b) appeared from day tural properties of dry-fermented meat products. Implica-
4 on. The study on “chorizo” and “salchichón” (CH and S) tions of these findings involve that LF-NMR relaxometry
13
209
IV. Resultados
464 Eur Food Res Technol (2017) 243:455–466
Table 2 Cluster Characteristic variables Cluster mean Overall mean Cluster SD Overall SD Test value
characterization
CLUSTER 1 (33.33 %)
Water content*** 67.543 56.329 5.280 9.459 10.02
aw*** 0.957 0.928 0.004 0.027 9.24
Adhesiveness*** −0.020 −0.390 0.000 0.351 8.90
T21*** 49.027 39.132 6.731 9.969 8.41
pH*** 5.364 5.145 0.321 0.245 7.57
T22*** 305.279 245.047 38.316 74.362 4.42
T2b*** 3.392 2.708 1.813 1.472 3.93
Chewiness*** 0.000 0.016 0.000 0.018 −7.46
Gumminess*** 5.000 15.658 0.000 11.301 −7.97
Hardness*** 4.900 24.898 0.000 19.614 −8.62
Springiness*** 0.000 0.001 0.000 0.001 −9.28
Cohesiveness*** 0.450 0.553 0.000 0.083 −10.51
CLUSTER 2 (37.50 %)
Springiness*** 0.001 0.001 0.001 0.001 6.10
Cohesiveness*** 0.585 0.553 0.047 0.083 3.59
T22* 266.004 245.047 71.202 74.362 2.79
Water content*** 52.673 56.329 5.337 9.459 −3.58
pH*** 5.007 5.145 0.033 0.245 −5.20
CLUSTER 3 (20.83 %)
Gumminess*** 25.609 15.658 6.625 11.301 5.40
Cohesiveness*** 0.624 0.553 0.035 0.083 5.26
Hardness*** 40.747 24.898 9.876 19.614 4.96
Chewiness** 0.025 0.016 0.008 0.018 2.97
T22*** 197.310 245.047 41.860 74.362 −3.96
Water content*** 49.602 56.329 2.989 9.459 −4.36
T21*** 30.986 39.132 4.657 9.969 −4.91
aw*** 0.901 0.928 0.005 0.027 −6.12
Adhesiveness*** −0.900 −0.390 0.220 0.351 −8.92
CLUSTER 4 (8.33 %)
Chewiness*** 0.064 0.016 0.010 0.018 9.29
Hardness*** 65.886 24.898 15.853 19.614 7.53
Gumminess*** 38.358 15.658 10.795 11.301 7.24
Cohesiveness*** 0.643 0.553 0.039 0.083 3.92
Springiness*** 0.002 0.001 0.000 0.001 3.68
T2b* 1.742 2.708 0.325 1.472 −2.37
Water content*** 44.746 56.329 1.119 9.459 −4.42
aw*** 0.893 0.928 0.003 0.027 −4.66
T22*** 145.642 245.047 15.206 74.362 −4.86
T21*** 24.475 39.132 2.261 9.969 −5.30
Mean values and standard deviation for each cluster against overall values. The percentage of variability
explained by the cluster is also included
*** P < 0.0001; ** P < 0.001: * P < 0.01
13
210
IV. Resultados
Eur Food Res Technol (2017) 243:455–466 465
r near ± 1 reveal a stronger correlation between the considered parameters. T2b: one single measurement was considered in the 1–10 ms region. W. cont: water content; Hard: hardness; Adh:
−0.39***
−0.70***
−0.49***
−0.72***
−0.33***
−0.66***
0.96***
−0.64***
0.72***
0.86***
0.98***
Chew ing parameters affect the ripening process and final product
characteristics.
−0.66***
0.99***
−0.63***
0.67***
0.85***
0.98***
project AGL2010-19158 funded by the Spanish Secretary of State of
−0.33**
Research, Development and Innovation within the Ministry of Econ-
Gum
−0.77***
−0.50***
−0.75***
0.86***
−0.67***
0.89***
0.85***
0.86***
olomics technologies (#274-09-0169)’.
−0.35**
Coh
−0.63***
−0.56***
−0.68***
0.67***
−0.72***
0.90***
0.67***
0.72***
−0.32**
−0.18*
−0.72***
−0.66***
−0.63***
−0.64***
0.28**
References
Adh
−0.65***
0.67***
0.86***
0.97***
Technol 11:201–209
Table 3 Pearson correlation coefficients matrix for T2 relaxation times, physicochemical and textural properties
Hard
−0.76***
0.73***
−0.68***
−0.75***
−0.69***
−0.66***
0.47***
0.74***
−0.37***
0.44***
−0.56***
−0.50***
−0.33***
−0.33**
−0.21*
0.47***
0.90***
−0.81***
0.75***
−0.63***
−0.77***
−0.75***
−0.72***
0.83***
0.87***
−0.63***
0.60***
−0.55***
−0.49***
−0.35**
−0.21*
−0.18
63:3039–3045
T22
13:585–591
0.37***
0.94***
0.88***
0.68***
0.94***
−0.79***
0.68***
−0.56***
−0.68***
−0.73***
−0.69***
0.34***
0.36***
−0.40***
−0.39***
−0.39***
−0.39***
0.26**
0.28**
−0.32**
0.22*
Gum
Adh
Coh
Spr
T2b
T21
T22
13
211
IV. Resultados
466 Eur Food Res Technol (2017) 243:455–466
14. Houben JH, Van’t Hooft BJ (2005) Variations in product-related 28. Gücükoglu A, Küplülü O (2010) The effect of different starter
parameters during standardized manufacture of a semi-dry fer- cultures and ripening temperatures on formation of biogenic
mented sausage. Meat Sci 69:283–287 amine in Turkish fermented sausages. Eur Food Res Technol
15. González-Fernández C, Santos EM, Rovira J, Jaime I (2006) The 230:875–884
effect of sugar concentration and starter culture on instrumental 29. Mokhtar S, Mostafa G, Taha R, Eldeep GSS (2012) Effect of
and sensory textural properties of chorizo-Spanish dry-cured different starter cultures on the biogenic amines production as a
sausage. Meat Sci 74:467–475 critical control point in fresh fermented sausages. Eur Food Res
16. Herrero AM, Ordóñez JA, Romero de Avila MD, Herranz B, de la Technol 235:527–535
Hoz L, Cambero MI (2007) Breaking strength of dry fermented 30. Andersen RH, Andersen HJ, Bertram HC (2007) Curing-induced
sausages and their correlation with texture profile analysis (TPA) water mobility and distribution within intra- and extra-miofi-
and physico-chemical characteristics. Meat Sci 77:331–338 brilar spaces of three pork qualities. Int J Food Sci Technol
17. Lorenzo J, Franco D (2012) Fat effect on physico-chemical,
42:1059–1066
microbial and textural changes through the manufactured of dry- 31. Bertram HC, Dønstrup S, Karlsson AH, Andersen HJ (2001)
cured foal sausage lipolysis, proteolysis and sensory properties. Continuous distribution analysis of T2 relaxation in meat- an
Meat Sci 92:704–714 approach in the determination of water holding capacity. Meat
18. AOAC, Association of Official Agricultural Chemists
Sci 60:279–285
(2006) Official Methods of Analysis of the Association of 32. Bertram HC, Karlsson AH, Rasmussen M, Pedersen OD,
Official Agricultural Chemists, 18th edn, 1st rev. AOAC Int, Andersen HJ (2001) Origin of multiexponential T2 relaxation in
Gaithersburg muscle myowater. J Agric Food Chem 49:3092–3100
19. Carr HY, Purcell EM (1954) Effects of diffusion on free prece- 33. Bertram HC, Purslow PP, Andersen HJ (2002) Relationship
sion in nuclear magnetic resonance experimets. Am J Physiol between meat structure, water mobility, and distribution: a low-
94:630–638 field nuclear magnetic resonance study. J Agric Food Chem
20. Meiboom S, Gill D (1958) Modified spin-echo method for meas- 50:824–829
uring nuclear times. Rev Sci Instrum 29:688–691 34. Venturi L, Rocculi P, Cavani C, Placucci G, Rosa MD, Cre-
21. Butler JP, Reeds JA, Dawson SV (1981) Estimating solutions of monini MA (2007) Water absorption of freeze-dried meat at dif-
1st kind integral equations with nonnegative constraints and opti- ferent water activities: A multianalytical approach using sorption
mal smoothing. SIAM J Num Anal 18:381–397 isotherm, differential scanning calorimetry, and nuclear magnetic
22. Bourne MC (1978) Texture profile analysis. Food Technol
resonance. J Agric Food Chem 55:10572–10578
32:62–66 35. Li Y, Jia W, Zhang CH, Li X, Wang LC, Zhang DQ, Mu GF
23. Ordóñez JA, de la Hoz L (2001) Embutidos crudos curados. (2014) Fluctuated low temperature combined with high-humidity
Tipos. Fenómenos madurativos. Alteraciones. In: Martin & thawing to reduce physicochemical quality deterioration of beef.
Macías, Enciclopedia de la carne y los productos cárnicos, vol II, Food Bioprocess Tech 7:3370–3380
Caceres, pp 1063–1090 36. Stangierski J, Baranowska HM (2015) The influence of heating
24. Herrero AM, de la Hoz L, Ordóñez JA, Herranz B, Romero de and cooling process on the water binding in transglutaminase-
Ávila MD, Cambero MI (2008) Tensile properties of cooked modified chicken protein preparation, assessed using low-field
meat sausages and their correlation with texture profile analysis NMR. Food Bioprocess Technol 8:2359–2367
(TPA) parameters and physico-chemical characteristics. Meat 37. Gordon A, Barbut S (1992) Mechanisms of meat batter stabiliza-
Sci 80:690–696 tion: a review. Crit Rev Food Sci Nutr 32:299–332
25. Ledesma E, Laca A, Rendueles M, Díaz M (2016) Texture, color 38. Shao J, Deng YM, Song L, Datur A, Jia N, Liu DY (2016) Inves-
and optical characteristics of a meat product depending on smol- tigation of the effects of protein hydration states on the mobility
ing time and casing type. Food Sci Technol 65:164–172 water and fat in meat batters by LF-NMR technique. Food Sci
26. Casquete R, Benito MJ, Martín A, Ruíz-Moyano S, Hernández Technol 66:1–6
A, Córdoba MG (2011) Effect of autochthonous starter cultures 39. Roca M, Incze K (1990) Fermented sausages. Food Rev Int
in the production of “salchichón”, a traditional Iberian dry- 6:91–118
fermented sausage, with different ripening processes. Food Sci 40. Huff-Lonergan E, Lonergan S (2005) Mechanisms of water-
Technol 44:1562–1571 holding capacity of meat: the role of postmortem biochemical
27. Marco A, Navarro JL, Folres M (2008) The sensory quality of and structural changes. Meat Sci 71:194–204
dry fermented sausages as affected by fermentation stage and
curing agents. Eur Food Res Technol 226:449–458
13
212
IV. Resultados
213
IV. Resultados
Trabajo 5
Monitorización del proceso de elaboración de jamón curado mediante espectroscopía de
RMN de alta resolución
217
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
* Corresponding author. Tel: +34 913943745; Fax: +34 913943743; E-mail: [email protected]
219
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
1. Introduction
Dry-cured ham is a traditional meat product highly appreciated by consumers. Its production
is a time-consuming process which varies between different ham types. There are many factors
affecting the final characteristics of dry-cured ham. The quality of the raw material and the process
conditions mainly influence the rate and the extent of biochemical reactions which are in turn
responsible for the formation of specific flavor and texture (Ventanas, 2001). Given the
technological and economic importance of this product, the development of suitable analytical
tools is essential to know the quality of the meat at its best, both structurally and compositionally.
Understanding what changes and transformations happen in each stage of production would
allow to evaluate ham features and quality under different processing conditions. To realize this
work on so valuable samples it is necessary to use non-destructive technologies. High resolution
Nuclear Magnetic Resonance (NMR) spectroscopy is a well-established technique that, when
applied to complex mixtures, allows the simultaneous detection of a large number of their low
molecular weight components (Bales et al., 1984; Wishart, 2008). Hence, NMR spectroscopy is
gaining wide acceptance in food analysis due to its potential for giving very exhaustive
representations of the chemical composition of the food matrix without extensive manipulation
(Manina et al., 2011; Cheng et al., 2013; Mazzei et al., 2017). For this goal, the NMR spectroscopy
modality known as High Resolution Magic Angle Spinning (HRMAS) has proved to be very useful
for reliably assessing the metabolic profile (metabolome) of intact muscle samples (Ritota et al.,
2012; Mazzei et al., 2017; Sundekilde et al., 2017) Thus, this modality has been applied to
investigate factors that somehow affect meat properties, such as the geographical origin of cattle
(Jung et al., 2010), type of breeds (Ritota et al., 2015), the characterization of metabolic changes
during beef meat ageing (Castejón et al., 2015) or meat derivatives composition (García-García
et al., 2018). Nevertheless, in comparison with other foods, meat has hardly been investigated by
NMR and, still less, the examination of a process as complex as the production of dry cured ham.
The study of ham implies the evaluation of a very heterogeneous product constituted by
several muscles with very different features, such as composition, and therefore it is possible that
they are affected in different ways for the diverse processes that the ham suffers during the
manufacturing process (Arnau et al., 1995). Therefore, the aim of this work was: i) to classify
samples according to the ripening time and ii) to assess the potential of the innovative 1H NMR
HRMAS methodology to rapidly identify the presence of the main metabolites of three different
intact pork leg muscles and iii) to monitor their evolution throughout ripening, from a very small
sample amount with minimum preparation.
2.1. Samples
Raw hams obtained from white-breed pigs, Landrace x Landrace White female animals of 10-
11 kg live weight, were purchased from the manufacturing plant Incarlopsa (Tarancón, Cuenca)
at the following stages: raw (3 days after slaughter), during salting (after salt addition, in the first
220
IV. Resultados
García-García et al., 2018, IV.2.5, 1-13
2-3 days of processing), post-salting (hams taken during the resting period after salt removal),
half-cured (half-cured hams taken after 6 and 8 months of processing) and cured (fully maturated
hams after 12 and 20 months of processing). Salting and post-salting were managed at low
temperature (0-4 oC); two different processing conditions were established during maturing: a
cold period (13-15 oC, RH: 75-85%) and a warm period (22-26 oC, RH: 60-75%) and ageing was
performed at room temperature (12-17 oC, RH: 60-80), in conformity with Santos et al., 2008.
Once the bones, skin and fat had been removed, muscles corresponding to different anatomical
locations (biceps femoris, semimembranosus and semitendinosus) were taken from the lean
tissue. The visible fat and connective tissue were totally removed from the muscles pieces.
Manufacturing process and sample collection was carried out according to García-García et al.,
2018. The effect of ripening time on the metabolite profile was studied by analyzing one sample
of each muscle at each processing time, representing a total of 21 separate analyses.
About 8–10 mg of lean tissue of each of the muscles studied was examined using HR-MAS at
4 ◦C to minimize tissue deterioration and avoid degradation of thermolabile compounds. 1H NMR
HRMAS spectroscopy was carried out at 500.13 MHz using a Bruker AMX500 spectrometer
11.7T.
The samples were placed into a 50 μl zirconium oxide rotor with a cylindrical insert, together
with 20 μl of 0.1mM solution of TSP in D2O. D2O (99.9%) and sodium (3-trimethylsilyl)-2,2,3,3-
tetradeuteriopropionate (TSP) were purchased from SDS and Sigma–Aldrich, respectively.
Experimental runs were performed at the lowest possible spinning rates to minimize structural or
chemical changes during analysis.
Standard solvent suppressed spectra (NOESYPRESAT) were acquired into 16 k data points,
averaged over 256 acquisitions. The total acquisition time was ∼14 min, with a relaxation delay
of 2 s, a mixing time of 150 ms and a spectral width of 8333 Hz. All spectra were processed using
MestReNova 9.0.1 (Mestrelab Research SL, Santiago de Compostela, Spain) and TopSpin 2.1
(Bruker Biospin, Rheinstetten, Germany). Prior to Fourier transformation, the FIDs were multiplied
by an exponential weight function corresponding to a line broadening of 0.3 Hz. Spectra were
phased, baseline-corrected and referenced to the TSP at δ = 0 ppm. For some of the analyzed
samples, 1D CPMG experiments were also performed to attenuate contribution from high
molecular weight metabolites. Two-dimensional (2D) homonuclear 1H-1H (COSY and TOCSY)
NMR experiments were also measured on the pork samples, and the resulting 2D spectra were
used to assist in assigning signals from 1H HR-MAS NMR spectra. Between consecutive 2D
spectra, a control 1H NMR spectrum was always measured to ensure that the sample remains
uncharged.
221
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
Assignment of resonances in the 1H NMR spectra from beef exudates was based on both spin
connectivity information obtained from 2D experiments and the use, as guidelines, of both data
reported in the literature (Castejón et al., 2015), and data obtained from HMDB database (Human
Metabolite Data Base http://www.hmdb.ca/ ).
Multivariate statistical analysis was conducted on NMR data. A total of 21 ham spectra were
subjected to PCA using AMIX software (version 3.9.11, Bruker BioSpin). Prior to the multivariate
analyses, each individual spectrum was data reduced over the 9.0–0.5 ppm range by dividing it
into spectral regions (buckets) of 0.04 ppm each. Some regions (i.e. unsuppressed water region
and fat signals) were excluded from the bucketing. In order to account for the variable
concentration of each reconstituted sample, its bucket intensities were normalized to the total
spectral intensity over the whole spectrum. The number of principal components (PCs) employed
for PCA was established as the minimum required to explain 95% of the total variance.
Significance analysis of variables (buckets) was based on the procedure of Goodpaster et al.,
2010) using a confidence level of 95%.
Unsupervised multivariate analysis by PCA is a useful tool to classify a set of individuals into
homogeneous groups. The whole data set of bucket-reduced 1H NMR spectra for 21 ham
samples was analyzed. Fig. 1A shows the scatter plot of PC1 vs PC2. PC1 and PC2 explain
62.32% and 10.77% of the total variance, respectively. A clear discrimination between samples
of different ripening time was revealed, suggesting that the total spectra contain useful information
for sample classification and to monitor and predict the maturation degree of the samples
analyzed. Fig. 1B shows the main buckets responsible of the variation between groups (red color).
222
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
Figure 1. A.PC1 vs PC2. Each point represents one single sample. The direction of the arrow indicates the progress of
ripening. B. Significant buckets involved in the discrimination between groups.
The principal metabolites contained within these buckets are shown in Table 1. These are
mainly amino acids, followed by carbohydrates and nucleosides. Although it would be expected
that the fats also had an important role in this ordering, it is necessary to remark that to perform
this multivariate analysis the intense fat signals were disregarded.
Table 1. Spectral region (ppm) showing significant differences during processing (P < 0.05). The table includes the
assignment of the main metabolites present in each region and their variation during ripening.
Spectral Variation
region p-value Metabolite during
(ppm) ripening
0.94 5.14E-05 Isoleucine (Ile) ↑
0.98 4.87E-08 Leucine (Leu) ↑
1.02 2.36E-08 Isoleucine (Ile) ↑
1.06 4.90E-08 Valine (Val) ↑
1.50 1.29E-06 -Alanine (-Ala) ↑
1.70 2.02E-06 Arginine (Arg) ↑
1.74 5.84E-10 Leucine (Leu) ↑
1.90 5.92E-08 Lysine (Lys) ↑
1.94 1.20E-07 Acetic acid (AA) ↑
2.34 4.36E-07 γ-Aminobutiric Acid (GABA) ↑
2.38 2.62E-08 Glutamate (Glu) ↑
223
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
224
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
225
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
Figure 2. Assignation of the main signals in 1H NMR spectra obtained for SM muscle at different processing times. A: raw ham; B: post-salted, C: half-cured, D: cured.
226
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
Figure 3. Changes in aromatic signals of 1H NMR HRMAS-spectra of Semimbranosus muscle during ripening. Trp:
tryptophan; Phe: Phenilalanine; Tyr: Tyrosine; Car: Carnosine; Ans: Anserine
227
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Compositiosn and Analysis, IV.2.5, 1-13
dehydration of samples, since the lipolytic activity remain constant during processing (Virgili et
al., 1998).
Figure 4. Changes in fatty acids signals of 1H NMR HRMAS-spectra of Semimbranosus muscle during ripening. FA:
fatty acids.
6,861*Trp (R2=0,98)
where ripening period values < 1 correspond to raw ham; values within the range ≥ 1 and < 2
correspond to post salted ham; values within the interval ≥ 2 and < 3 correspond to half-cured
ham and values ≥ 3 correspond to cured ham].
Table 2. Regression model for ripening time using aromatic amino acids signal intensity. (Phe: phenilalanine; Trp:
tryptophan; Tyr: tyrosine)
228
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
Dependent Independent
R2 SE Coefficient F-ratio P-value
variable variable
4. Conclusions
As far as we know, this is the first 1H NMR study of dry-cured ham manufacturing, where the
complete metabolite profile of intact BF, SM and ST muscles and their evolution throughout
ripening was reported. The applied 1H NMR HRMAS methodology has been found to be a rapid
and effective technique to monitor the changes associated with the drying-curing processes. The
non-invasive analysis (just a needle of biopsy is required) proposed along the whole ripening
process allows to create predictive models that facilitate the monitoring and control of processing,
as well as to evaluate the possible effects that occur on the metabolic profile as a consequence
of modifications in the ripening conditions (temperature, humidity, time...). Consequently, 1H NMR
HRMAS spectral profiles of dry-cured products can be very useful in evaluating the metabolome
variations as a function of different muscle processing conditions.
Acknowledgements
The authors are grateful for the financial support of UCM-BSCH Group 920176GR35/10A,
project AGL2010‐19158 funded by the Spanish Secretary of State of Research, Development and
Innovation within the Ministry of Economy and Competitiveness and the cooperative collaboration
with the members of the CAI of NMR and RSE (UCM). A.B.G.G. was awarded a grant (BES-
2011-047485) from the same institution.
229
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analyisis, IV.2.5, 1-13
References
Virgili, R., Schivazappa, C., Parolari, G., Bordini, C. S., Degni, M. (1998). Proteases in fresh pork muscle
and their influence on bitter taste formation in dry-cured ham. Journal of Food Biochemistry, 22, 53-63.
Ruiz-Ramirez, J., Arnau, J., Serra, X., Gou, P. (2006). Effect of pH24, NaCl content and proteolysis index
on the relationship between water content and texture parameters in biceps femoris and
semimembranosus muscles in dry-cured ham. Meat Scenc,i 72, 185-194.
Zhao GM, Tian W, Liu YX, Zhou GH, Xu XL, Li MY (2008) Proteolysis in biceps femoris during Jinhua ham
processing. Meat Science, 79, 39–44.
Harkouss, R., Astruc, T., Lebert, A., Gatellier, P., Loison, O., Safa, H., Portanguen, S., Parafita, E., Mirade,
P. S. (2015). Quantitative study of the relationships among proteolysis, lipid oxidation, structure and
texture throughout the dry-cured ham process. Food Chemistry, 166, 522-530.
Ventanas, J. (2001). Factores del proceso que determinan la calidad del jamón. En: Jamón Ibérico y
Serrano. Fundamentos de la elaboración y de la calidad. Ed. Paraninfo, Madrid: pp. 63-103
Wishart, D.S. (2008). Quantitative metabolomics using NMR. TrAC Trends in Analytical Chemistry, 27, 228-
237.
Bales, J.R., Higham, D.P., Howe, I., Nicholson, J K., Sadler, P.J. (1984). Use of high-resolution proton
nuclear magnetic resonance spectroscopy for rapid multi-component analysis of urine. Clinical
chemistry, 30, 426-432.
Mannina, L., Sobolev, A.P., Viel, S. (2012). Liquid state 1H high field NMR in food analysis. Progress in
Nuclear Magnetic Resonance Spectroscopy, 66, 1-39.
Mazzei, P., Piccolo, A., Valentini, M., & Webb, G.A. (2017). Intact Food Analysis by Means of HRMAS-NMR
Spectroscopy. In: Web G. (eds.) Modern Magnetic Resonance, 1-16.
Arnau, J.; Guerrero, L.; Casademont, G.; Gou, P. (1995). Physical and chemicalchanges in different zones
of normal and pse dry-cured ham during processing. Food Chemistry, 52, 63-9.
Cheng, J.H., Dai, Q., Sun, D.W., Zeng, X.A., Liu, D., Pu, H.B. (2013). Applications of non-destructive
spectroscopic techniques for fish quality and safety evaluation and inspection. Trends in food science &
technology, 34, 18-31.
Ritota, M., Casciani, L., Failla, S., Valentini, M. (2012). HRMAS-NMR spectroscopy and multivariate analysis
meat characterisation. Meat science, 92, 754-761.
Sundekilde, U.K., Rasmussen, M.K., Young, J.F., Bertram, H.C. (2017). High resolution magic angle
spinning NMR spectroscopy reveals that pectoralis muscle dystrophy in chicken is associated with
reduced muscle content of anserine and carnosine. Food chemistry, 217, 151-154.
Jung, Y., Lee, J., Kwon, J., Lee, K.S., Ryu, D.H., Huang, G.S. (2010). Discrimination of the Geographical
Origin of Beef by 1H NMR-Based Metabolomics. Journal of Agricultural and Food Chemistry, 58, 10458–
10466.
Ritota, M., Casciani, L., Failla, S., Valentini, M. (2015). HRMAS-NMR spectroscopy and multivariate analysis
meat characterization. Meat Science, 92, 754–761.
Castejón, D., García-Segura, J.M., Escudero, R., Herrera, A., Cambero, M.I. (2015). Metabolomics of meat
exudate: Its potential to evaluate beef meat conservation and aging. Analytical Chemical Acta, 901,1 –
11.
García-García, A.B., Lamichhane, S., Castejón, D., Cambero, M.I., Bertram, H.C. (2018). 1H HR-MAS NMR-
based metabolomics analysis for dry-fermented sausage characterization Food Chemistry, 240, 514-
523.
García-García, A.B., Fernández-Valle, M.E., Castejón, D., Escudero, R., Cambero, M.I. (2018). Use of MRI
as a predictive tool for physicochemical and rheological features during cured ham manufacturing. Meat
Science (under revision).
Santos, C., Hoz, L., Cambero, M.I., Cabeza, C., & Ordóñez, J.A. (2008). Enrichment of dry-cured ham with
-linolenic acid and -tocopherol by the use of linseed oil and -tocopheryl acetate in pig diets. Meat
230
IV. Resultados
García-García et al., 2018, enviado a Journal of Food Composition and Analysis, IV.2.5, 1-13
Goodpaster, A. M., Romick-Rosendale, L. E., & Kennedy, M. A. (2010). Statistical significance analysis of
nuclear magnetic resonance-based metabonomics data. Analytical Biochemistry, 401, 134–143.
231
IV. Resultados
Trabajo 6
Identificación de los principales metabolitos presentes en embutidos crudos curados y
evolución de los mismos durante el proceso de maduración
233
IV. Resultados
235
IV. Resultados
Food Chemistry 240 (2018) 514–523
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
1
H HR-MAS NMR-based metabolomics analysis for dry-fermented sausage MARK
characterization
Ana Belén García-Garcíaa, Santosh Lamichhanec, David Castejónb, Mª Isabel Camberoa,
⁎
Hanne Christine Bertramc,
a
Departamento de Nutrición, Bromatología y Tecnología de los Alimentos, Facultad de Veterinaria, Universidad Complutense de Madrid, 28040 Madrid, Spain
b
CAI de Resonancia Magnética Nuclear y Espín Electrónico, Universidad Complutense de Madrid, 28040 Madrid, Spain
c
Department of Food Science, Aarhus University, Research Centre Aarslev, Kirstinebjergvej 10, DK-5792 Årslev, Denmark
A R T I C L E I N F O A B S T R A C T
Keywords: Proton high-resolution magic angle spinning (1H HR-MAS) nuclear magnetic resonance (NMR) spectroscopy in
Dry-fermented sausages combination with principal component analysis (PCA) was employed to characterize dry-fermented sausages
Fermentation salchichón type throughout the manufacturing process. 1H HR-MAS NMR metabolite profiling was achieved from
Ripening a small sample of intact sausage after 0, 2, 4, 7, 11 and 14 days of drying. Intriguingly, the obtained results
1
H HR-MAS NMR spectroscopy
enabled the identification of the three main stages in the traditional production of salchichón. Formulation,
Meat metabolites
Culinary meat
fermentation and drying-ripening periods showed distinct and characteristic metabolomic profiles.
Compositional changes related to microbial activity, as well as proteolytic and lipolytic phenomena, decisive
steps in such a ripening process, could be monitored through the NMR spectra. This study shows the potential of
1
H HR-MAS as a rapid method for probing metabolomic profiles and compositional changes during sausages
processing.
⁎
Corresponding author.
E-mail address: [email protected] (H.C. Bertram).
http://dx.doi.org/10.1016/j.foodchem.2017.07.150
Received 5 April 2017; Received in revised form 24 July 2017; Accepted 27 July 2017
Available online 29 July 2017
0308-8146/ © 2017 Elsevier Ltd. All rights reserved.
237
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
the beef metabolome (Zanardi et al., 2015), the wooden breast ab- sausage units for each ripening time were analyzed. Samples of lean
normality in chicken (Sundekilde, Rasmussen, Young, & Bertram, 2017) meat were obtained from the inner part of the salchichón units. A D2O
and the evaluation of beef exudates as suitable matrix to study beef solution (15 μL) containing trimethylsilyl 3-propionic acid sodium salt
ageing (Castejón, García-Segura, Escudero, Herrera, & Cambero, 2015). (TSP, 0.1 mM) was added to the 30 μL disposable Kel-F HR-MAS inserts
However, few NMR spectroscopic studies focused on the examination of (Bruker Biospin, Rheinstetten, Germany) together with the meat tissue
fermented sausages have been conducted. Siciliano et al., 2013 de- sample (10 ± 2 mg). TSP was used as internal chemical shift reference
termined the variations in the fatty acid composition based on analyses (0.00 ppm). The insert was placed into a 4 mm zirconium rotor. Proton
of lipid extracts and Mati, Staruch, & Soral, 2015 studied the quantita- NMR spectra were recorded on a Bruker Avance 600 spectrometer,
tive changes in carnosine and free amino acids in the aqueous fraction operating at a 1H frequency of 600.13 MHz, equipped with a HR-MAS
during ripening. Consequently, studies on the application of NMR- probe (Bruker BioSpin, Rheinstetten, Germany). A water-suppressed
based metabolomics for the detection of compounds in dry-fermented Carr-Purcell-Meiboom-Gill (CPMG) spin echo sequence [90°-(t-180°-
meat are still limited and no detailed investigations on the complete t)n–acquisition] (Mannina, Sobolev, & Viel, 2012) was used to eliminate
profiling of intact sausages have been reported. Therefore, the aim of signals from lipids and macromolecules with an effective echo time of
this study was to investigate the potential of a 1H HR-MAS methodology 300 ms. This effective echo time was previously optimized in our la-
for an explorative and untargeted elucidation of changes in the meta- boratory (Lamichhane et al., 2015). Proton spectra were acquired at
bolites of dry fermented sausages salchichón type during a traditional 4 °C with a spin rate of 5 kHz. Two hundred and fifty-six scans over a
manufacturing process and to explore the ability of this technique to spectral region of 10 kHz were collected into 32k data points with a
monitor and predict the early days of ripening. relaxation delay of 5 s. The free induction decay (FIDs) obtained were
multiplied by an exponential weight function corresponding to a line
2. Materials and methods broadening of 0.3 Hz before Fourier transformation. The spectra were
phased, baseline-corrected and referenced to TSP using the Bruker
2.1. Dry fermented sausages manufacture Topspin software (version 2.1).
238 515
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
3. Results and discussion the databases: HMDB (Human Metabolite Data Base http://www.hmdb.
ca/) and Chenomx (NMR Suite version 7.6, Edmonton, Canada). The
3.1. 1H HR-MAS NMR spectra assignment of the main signals present in the 1H HR-MAS spectra
(Fig. 1) is reported in Table 1. Several metabolites were identified:
A comparison of representative one-dimensional 1H HR-MAS NMR amino acids, peptides and analogues; carbohydrates; organic acids and
spectra of intact salchichón samples collected on days 0 (meat batter), 2 derivatives; nucleosides, nucleotides and analogues; fatty acids and
(fermentation stage) and 14 (final product) is shown in Fig. 1. These miscellaneous, including Acetone, Crn, Cho, PCho, GPCho or IMP (ab-
spectra represent the distinctive stages of the ripening process where breviations in Table 1).
the metabolic information shows the greatest differences throughout From the comparison of the three spectra it is evident that some
the process. The spectroscopic changes observed are correlated with the metabolites experienced a remarkable change during manufacturing.
three well-defined periods in the sausages manufacturing process, These major spectral differences will be discussed below in further
previously described by conventional analytical methods: formulation, detail.
fermentation and drying-ripening (Ordóñez et al., 1999). Table 1 includes the spectral information of each 1H NMR signal
Metabolite assignments were based on information previously (chemical shift and group, multiplicity and coupling constants) and also
published by our research group (Castejón et al., 2010, 2015) and on its intensity of variation during the fermentative and ripening/drying
516 239
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
Table 1
Chemical shift assignment of 1H NMR signals for the main components in salchichón samples with the peak notation shown in Fig. 1. The table includes the bucket numbers assigned in
Table S1, and the magnitude of their variation during fermentation and ripening/drying.
1
Peak Compound Group H(ppm) Mult.:J(Hz) Bucket Variation Variation
(fermentation) (ripening/drying)
240 517
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
Table 1 (continued)
1
Peak Compound Group H(ppm) Mult.:J(Hz) Bucket Variation Variation
(fermentation) (ripening/drying)
518 241
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
Table 1 (continued)
1
Peak Compound Group H(ppm) Mult.:J(Hz) Bucket Variation Variation
(fermentation) (ripening/drying)
processes. The magnitude of the variation experienced by metabolites from carbohydrates but this spectral region also contains resonances of
during fermentation and ripening (Table 1) is based on the integration other metabolites, such as amino acids, and quantitative changes in
of the buckets previously defined in Table S1. In some instances, one amino acids signals are subjected to the variation of sugars and their
single bucket region contained contribution from more than one me- real variation is not revealed. The same would apply to lipid signals, as
tabolite, and therefore quantifying the level of a singular metabolite in resonances from fatty acids may hide the real variation of adjacent
salchichón samples was a difficult task because of spectral crowding and metabolites. Spectral crowding together with the pronounced decrease
spectral overlap and great care is required with integration quantifi- in pH during the fermentation stage (Fig. S1) are responsible for the
cation approaches. Moreover, it must be noted that chemical shifts are mentioned inconsistencies. Conclusions concerning the increase or de-
highly sensitive to environmental changes. In particular, changes in pH crease of a singular metabolite have been made considering specific
between samples will alter the ionization of the functional groups and buckets in which the relevant metabolite is hardly overlapped or
thus affect the final chemical shift of some hydrogen nucleus (Dona slightly influenced by the presence of other metabolites.
et al., 2016). Indeed, AA and SA resonances (in Table 1, signal 24 and The main metabolites that define the spectral fingerprint of each
41, respectively) were visibly affected by the decrease in pH between manufacturing stage are described below.
day 0 and 2 (see Fig. S1), since an obvious shift from their usual spectral
positions was observed.
3.1.1. Formulation stage profile
Despite the CPMG pulse sequence employed to attenuate contribu-
The representative 1H NMR spectra of salchichón at 0 day (Fig.1A)
tion from high molecular weight metabolites, it was possible to detect
appeared to be very similar to that obtained recently for raw beef meat
signals corresponding to FAs, however, signals from triglycerides that
(Castejón et al., 2015), reflecting that at this stage the unripened matrix
have a very short relaxation time, were completely removed from the
1 made from lean meat, fat and non-meat ingredients to a great extent
H spectrum. The lipid profile obtained in this study was expected for
influences the overall metabolite composition. At first glance, the
fermented sausages (Siciliano et al., 2013). Table 1 contains data of the
spectra of salchichón were dominated by the intense resonances of lactic
characteristics signals of saturated fatty chains (signals 1, 11, 15, 36),
acid (LA) (signals 12, 127), creatine/phosphocreatine (Cr/PCr) (signals
monounsaturated fatty chains (signals 27, 149) and polyunsaturated
57, 118) and carnosine (Car) (signals 50, 60, 66, 69, 135, 159, 176).
chains (signal 52). Moreover, signals from phospholipids, PCho (signals
The signals of fatty acids are also noticeable in the spectra of day 0,
64, 88, 128) and GPCho (signals 93, 131), could be observed.
since fat is one of the main component of sausages, representing20% of
In table Table 1, it is possible to identify some lack of agreement
the total composition in the present study.
between the quantitative fluctuations experienced by the different
Some of the additives used in the elaboration of the salchichón are in
chemical shifts for the same metabolite. In this regard, it should be
concentration range to be observed in the 1H spectrum, especially in the
noted the spectral region 3.25 to 5.50 ppm contains several resonances
carbohydrate region. This spectral part showed remarkable differences
242 519
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
when comparing raw and processed meats (data not shown) due to the degradation of α-Glc and β-Glc as a result of microbial consumption is
fact that dextrose, dextrin and lactose were intentionally added to the revealed and significant differences between days 0 and 2 (P < 0.05)
minced meat in order to improve microorganism growth. Dextrose is were observed. On the contrary, the remaining metabolites, EtOH, LA,
the D-isomer for glucose (D-glucose) and glucose signals in the 1H NMR AA, FoA and 2,3-butanediol, (Fig. 2B–F) exhibited a different behavior
spectrum were considerably more prominent in sausage samples than in with a strongly and significant (P < 0.05) increase during the first
raw meat. On the other hand, dextrins which are polymers of D-glucose, 48 h, thus they are byproducts resulting from fermentative metabolism.
contain singular anomeric signals in the 1H NMR spectra (signals 144, There are also significant changes (P < 0.05) throughout the ripening
150) that are not present in raw meat (Castejón et al., 2015; Rodrigues process (2–14 days) in EtOH, LA and AA. These metabolites, clearly
et al., 2010). Finally, lactose was detectable through the resonances of linked to fermentation, decreased noticeably as ripening proceeded. On
their constituent monosaccharides, glucose and galactose. the other hand, Glc, FoA and 2,3-butanediol remained relatively stable
(non-significant differences) during the subsequent production days.
Regarding amino acids, Ile, Leu, Val, α-Ala Lys, Arg, Met and Glu
3.1.2. Fermentation stage profile were increased during the fermentation stage, as well as the aromatic
As shown in Fig.1B, the 1H NMR signals corresponding to amino amino acids signals intensities (Phe, Tyr, Trp and Tyrm). Moreover,
acids, carbohydrates and organic acids experienced notable changes notable changes in nucleotides during the first 48 h of the process were
compared with the previous stage and the most striking finding was the also detected, as well as changes in purine derivatives. While, AMP, IMP
pronounced increase in EtOH (signal 9), AA (signal 24) and 2,3-buta- and Ino underwent a significant decrease, Hx was notably increased at
nediol (signals 7, 8, 89). Variation in these metabolites can be explained this stage. These changes are consistent with the degradation process of
by the intense microbial activity during the first 48 h of the process as a ATP in muscle, according to which: ATP → ADP → AMP → IMP →
result of the LAB anaerobic fermentative metabolism and the microbial Ino → Hx → Xa → uric acid.
enzymatic activity related to the onset of proteolysis (Aymerich,
Martín, Garriga, & Hugas, 2003). Related to this microbial activity, a
pronounced decrease in the signals assigned to α-Glc, β-Glc, α-Gal and 3.1.3. Final product profile
β-Gal in day 2 was also observed, as carbohydrates serve as energy Fig.1C shows the representative 1H HR-MAS NMR spectra of a sal-
substrate for LAB. chichón type product after 14 days of ripening. The analysis of the me-
Fig. 2 shows the variation of the integral values (arbitrary units) of tabolite profile revealed a decrease or even an almost complete ex-
the mentioned metabolites during processing. In Fig. 2A, the tinction of these compounds linked to the fermentative metabolism in
Fig. 2. Changes in representative metabolites integral (mean values and standard deviation) during salchichón manufacturing (days 0, 2, 4, 7, 11, and 14). A: α-Glc; β-Glc. B:EtOH. C: LA,
D: AA, E: FoA, F: 2,3-butanediol.
520 243
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
244 521
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
Fig. 4. Plot of the buckets which experienced notable changes A: during fermentation and B: during ripening/drying. The variation of the integral of each bucket (A: 0–2 days and B:
0–14 days) compared to the final value of the bucket of each stage is represented.
522 245
IV. Resultados
A.B. García-García et al. Food Chemistry 240 (2018) 514–523
Castejón, D., Villa, P., Calvo, M. M., Santa-María, G., Herraiz, M., & Herrera, A. (2010). 9781782622741-00171 pp. 171–180.
1
H-HRMAS NMR study of smoked Atlantic salmon (Salmo salar). Magnetic Resonance Mannina, L., Sobolev, A. P., & Viel, S. (2012). Liquid state 1H high field NMR in food
in Chemistry, 48, 693–703. analysis. Progress in Nuclear Magnetic Resonance Spectroscopy, 66, 1–39.
Díaz, O., Fernández, M., García de Fernando, D. G., de la Hoz, L., & Ordóñez, J. A. (1997). Mati, M., Staruch, L., & Soral, M. (2015). Use of NMR spectroscopy in the analysis of
Proteolysis in dry fermented sausages: The effect of selected exogenous proteases. carnosine and free amino acids in fermented sausages during ripening. Chemical
Meat Science, 46, 115–128. Papers, 69, 1319–1324.
Dona, A. C., Kyriakides, M., Scott, F., Shephard, E. A., Varshavi, D., Veselkov, K., et al. Ordóñez, J. A., Hierro, E. M., Bruna, J. M., & de la Hoz, L. (1999). Changes in the
(2016). A guide to the identification of metabolites in NMR-based metabonomics/ components of dry-fermented sausages during ripening. Critical Reviews in Food
metabolomics experiments. Computational and Structural Biotechnology Journal, 14, Science and Nutrition, 39, 329–367.
135–153. Ritota, M., Casciani, L., Failla, S., & Valentini, M. (2015). HRMAS-NMR spectroscopy and
Fernández, M., Ordóñez, J. A., Bruna, J. M., Herranz, B., & de la Hoz, L. (2000). multivariate analysis meat characterization. Meat Science, 92, 754–761.
Accelerated ripening of dry fermented sausages. Food Science and Technology, 11, Rodrigues, J. E. A., Erny, G. L., Barros, A. S., Esteves, V. I., Brandao, T., Ferreira, A. A.,
201–209. et al. (2010). Quantification of organic acids in beer by nuclear magnetic resonance
Fernández-López, J., Sendra, E., Sayas-Barberá, E., Navarro, C., & Pérez-Álvarez, J. A. (NMR)-based methods. Analytica Chimica Acta, 674, 166–175.
(2008). Physico-chemical and microbiological profiles of “salchichón” (Spanish dry- Siciliano, C., Belsito, E., De Marco, R., Di Gioia, M. L., Leggio, A., & Liguori, A. (2013).
fermented sausage) enriched with orange fiber. Meat Science, 80, 410–417. Quantitative determination of fatty acid chain composition in pork meat products by
Goodpaster, A. M., Romick-Rosendale, L. E., & Kennedy, M. A. (2010). Statistical sig- high resolution 1H NMR spectroscopy. Food Chemistry, 136 564-554.
nificance analysis of nuclear magnetic resonance-based metabonomics data. Sundekilde, U. K., Rasmussen, M. K., Young, J. F., & Bertram, H. C. (2017). High re-
Analytical Biochemistry, 401, 134–143. solution magic angle spinning NMR spectroscopy reveals that pectoralis muscle
Jung, Y., Lee, J., Kwon, J., Lee, K. S., Ryu, D. H., & Huang, G. S. (2010). Discrimination of dystrophy in chicken is associated with reduced muscle content of anserine and
the Geographical Origin of Beef by 1H NMR-Based Metabolomics. Journal of carnosine. Food Chem. 217, 151–154. http://dx.doi.org/10.1016/j.foodchem.2016.
Agricultural and Food Chemistry, 58, 10458–10466. 08.104.
Laghi, L., Picone, G., & Capozzi, F. (2014). Nuclear magnetic resonance for foodomics Weckwerth, W. (2003). Metabolomics in systems biology. Annual Review of Plant Biology,
beyond Food analysis. Trends in Analytical Chemistry, 59, 93–102. 54, 669–689.
Lamichhane, S., Yde, C. C., Mielby, L. H., Kidmose, U., Møller, J. R., Hammershøj, M., Zanardi, E., Caligiani, A., Palla, L., Mariani, M., Ghidini, S., Di Ciccio, P. A., et al. (2015).
et al. (2015). High-resolution magic angle spinning studies of semi-hard Danbo (30+) Metabolic profiling by 1H NMR of ground beef irradiated at different irradiation
cheese-impact of processing condition and relation to sensory perception. Magnetic re- doses. Meat Science, 103, 83–89.
sonance in food science. defining food by magnetic resonancehttp://dx.doi.org/10.1039/
246 523
IV. Resultados
Trabajo 7
Valoración de los exudados de cerdo como matriz apta para la monitorización de la
conservación de la carne tratada con electrones acelerados
Enviado a Food Research International
247
IV. Resultados
249
IV. Resultados
García-García et al., 2018, enviado a Food Research International, V.2.7, 1-17
* Corresponding author. Tel: +34 913943745; Fax: +34 913943743; E-mail: [email protected]
Abstract: Seventy-two exudates from pork tenderloin samples, subjected to E-beam irradiation
treatments, have been employed to monitor, through 1H-NMR analysis, the effects of irradiation dose
(0, 1, 2 and 6 kGy) and storage time (1, 6 and 12 days). As far as we know, this is the first study where
meat exudate is employed to monitor the effects of irradiation dose and storage time. The 1H-NMR
spectra, obtained after ~ 2 minutes, allowed to determine the main components of the pork exudate.
Results show that 1H-NMR-based metabolomics provides valuable information about the metabolic
changes suffered during storage and how these transformations could be affected by E-beam irradiation
treatment. The ease to obtain exudates, the simple NMR sample preparation, the good correlation
between the selected metabolites, the irradiation treatment and the storage times point to that this study
could be the first step to develop a new method for analysis and control of meat conservation and to
evaluate its irradiation treatment.
Keywords: Exudate, 1H-NMR, Pork, Meat, Storage, Quality & Safety, Traceability, Irradiation
251
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
1. Introduction
Nowadays, quality and safety remains a major concern facing the food industry and it is a critical
component in ensuring food and nutrition security worldwide (FAO, 2012). Regarding meat products,
quality refers to intrinsic attributes critical for the technological, nutritional and sensory suitability of meat
for eating, processing, and storage (Listrat et al., 2016). Therefore, it is imperative to pay close attention
to meat quality and safety. These targets are becoming dynamic and challenging concerns which require
the generation of new information and continuous reevaluation of the existing knowledge for meeting
market’s demands (Sofos, 2008). One of the successful techniques to preserve food with minimum
interruption to the functional, nutritional, and sensory properties is irradiation (Chauhan et al., 2009).
This processing involves controlled application of energy from ionizing radiations such as gamma rays,
X-rays, and electron beam for food preservation. Irradiation preserves the food by disrupting the
biological processes that lead to decay of food quality (Farkas, 2006). Radiation interacts with water and
other biological molecules in a food system and produces various radiolytic products, which generally
act as oxidizing agents and can cause several changes in the molecular structure of organic matter.
Radiations also damage DNA molecules effectively, therefore living cells such as in microorganisms,
insects, and gametes are prevented from reproduction, resulting in a preservative effect. In contrast,
irradiation, like other processing techniques, results in physicochemical changes in food products. The
nature and extent of these changes depend on the kind of food subjected to irradiation and the irradiation
dose (Chauhan et al., 2009). Recently, a study developed by our group reveled the suitability of electron
beam (E-beam) irradiation to increase the shelf life of whole fresh pork loin stored at 4°C (García-
Márquez et al., 2012).
In general, the traditional techniques and methods for evaluation and detection of meat quality and
safety are tedious, laborious, expensive and time-consuming while spectroscopic techniques have
successfully overcome some of these disadvantages and can supplement or replace them. There are
growing interests in spectroscopic techniques, such as Nuclear Magnetic Resonance (NMR), due to
their high specificity, convenience, non-destructive, non-invasive, cost effective and quick response
(Cheng et al., 2013). NMR spectroscopy for structural and compositional characteristics of complex
biological samples (Martínez et al., 2005) has frequently been applied in food analysis to a broad range
of analytical challenges (Capitani et al., 2012; Rodrigues et al., 2010; Straadt et al., 2014). In studies on
meat, NMR has shown a high potential for compositional analysis of muscles (Ritota et al., 2012;
Sundekilde et al., 2017), storage (Castejón et al., 2015), geographical origin (Jung et al., 2010), quality
(Bertram, 2017) or fatty acids profiling (Sicialiano et al., 2013). Traditionally the high-resolution NMR
sampling of biological tissue starts with an extraction procedure followed by a conventional liquid NMR
spectroscopy acquisition (Mannina et al., 2012). However, magic angle spinning (MAS-)NMR allowed
measuring directly on a few milligrams of intact tissue making the extraction procedure unnecessary
and thus much faster and less resource demanding (Castejón et al., 2010).
252
IV. Resultados
García-García et al., 2018, IV.2.7, 1-17
In this paper, we present an alternative that consists in the direct study of pork exudate. The loss of
exudates from muscle is unavoidable due to some loss of moisture that occurs due to the presence of
water in a free form in muscle tissue. Therefore, the exudate is an easy to obtain substrate which
manipulation does not mean loss or alteration of the sample from which it comes from. They are
valueless, easy-to-obtain and, their presence in small amounts in commercial meat packages cannot
be associated with significant degradation or alteration of the meat from which they proceed. Rather
than this, they should be considered as the natural juice from raw meat that conveniently analyzed could
provide homogeneous information of the whole meat sample. Despite all these advantages, pork
exudate has been poorly studied. Recently, the influence of accumulated exudate on the quality
characteristics of fresh and freeze-thawed pork during cold storage was considered (Kim et al., 2013).
Moreover, in order to assure meat processing and final meat quality, identification of proteins (Pioselli
et al., 2011) and fatty acids (Benet et al., 2015) in cooked ham exudates was conducted. NMR spectra
of complex mixtures show hundreds of signals, coming from a great number of diverse metabolites. This
fact and the overlap between signals, sometimes, make it difficult to extract information, either “visually”
or by simple processing of the data. The most effective way to analyse these “holistic profiles” is by
using chemometric tools which enable the visualization of the data in a reduced dimension and the
classification of the samples into established classes based on inherent patterns in a set of spectral
measurements (Alonso-Salces et al., 2007).
Fresh meat is usually commercialized in trays, vacuum packed with or without modified atmosphere
packaging (MAP). Therefore, exudates naturally extracted from meat are accumulated and contained
inside these packaging. The objective of the present work is to determine if, under the commercial
conditions mentioned, meat exudates can be considered a suitable matrix to study the metabolic
changes that occur in the meat during the storage and conservation, as well as during the application
of an irradiation treatment. This statement is based in a previous work (Castejón et al., 2015) that
verifies that exudates extracted from beef provide important metabolic information of the meat piece
from which it proceeds due to the strong correlation found between meat NMR spectra and exudates
NMR spectra. Thus, 1H NMR would be a suitable technology for non-destructive analysis allowing the
monitoring of different procedures associated with meat processing.
Therefore, the aim of the present work was: i) to evaluate the feasibility of NMR techniques to analyze
pork exudates as a new suitable matrix to obtain fresh meat metabolome ii) to identify changes in the
molecular profile of pork exudate during storage and iii) to study whether irradiation has a positive effect
on the degradation of pork exudates during storage and also, to determine the suitable irradiation doses.
According to these objectives, samples under four different irradiation doses (0, 1, 2 and 4 kGy) stored
at three different times (1, 6 and 12 days) were analyzed, and the ability of NMR together with
chemometric tools to differentiate between irradiated and non-irradiated samples was evaluated.
253
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
Twenty-four hours postmortem, pork tenderloin pieces (3.5 to 5 kg in weight) were collected from
two local butcher markets. Meat samples did not differ in their postmortem period. Thermally insulated
containers were employed during transportation of the samples to the laboratory, which, in all cases,
took no longer than 1 hour. Afterwards, ~ 50 g pieces of each pork tenderloin sample was vacuum-
packed at 20 kPa in 10 x 10 cm laminated film bags (90 μm copolymer of polyamide/polyethylene) of
low gas permeability (transmission rates of 35 cm 3 24h-1 m-2 bar-1 and 150 cm3 24h-1 m-2 bar-1 for O2 and
CO2, respectively) and stored at 4°C until irradiation treatment, which was performed in the next 12-14
h. To evaluate pork exudates as a suitable matrix to analyze meat metabolites, a preliminary study was
conducted to compare both, intact muscle and exudates spectra. Results are shown in Figure S2
(Suplementary information)
Seventy-two samples (6 pork x 4 doses x 3 storage times) were transported under refrigeration (2-3
ºC) in insulated boxes to the irradiation plant (Ionmed S.A., Tarancón, Cuenca, Spain), and treated
under an electron-beam (E-beam) radiation source operating at 10 MeV. The radiation treatment was
adjusted to apply doses of 1, 2 and 6 kGy, dose lower than 10 kGy, which is considered the safety dose
by the World Health Organization (WHO, 1999). The dose absorbed by the samples was checked by
determining the absorbance of cellulose triacetate dosimeters (ASTM, 2000) simultaneously irradiated.
Non-treated samples were used as control (0 kGy). After irradiation treatment samples were kept
refrigerated at 4 °C.
At 1, 6 and 12 days of storage, the released exudate was collected from vacuum-packed samples
(twenty-four samples per storage point), frozen at -80 ºC, and freeze-dried at room temperature. Freeze-
dried exudates were stored at -80 ºC until NMR analysis.
For 1H-NMR analysis 23.6 ± 0.3 mg of the lyophilized pork tenderloin exudates were reconstituted in
an eppendorf by adding 650 μL phosphate buffer solution in D 2O (0.1M; pH=7.0) containing 1 mM
sodium trimethylsilyl-2,2,3,3-tetradeuteroproprionate (TSP), an internal chemical shift standard.
Samples were vortexed for 30 seconds and transferred to 5 mm NMR tube. The pH of the samples was
measured after the NMR experiment, pH= 6.57 ± 0.05.
254
IV. Resultados
García-García et al., 2018, IV.2.7, 1-17
1H-NMR spectra of the reconstituted exudates were recorded randomly at 298K on a Bruker Avance
500MHz spectrometer using a 5-mm multinuclear direct detection broadband probe (Bruker BioSpin
GmbH, Rheinstetten, Germany). 1D NOESY (NOESYPRESAT) standard solvent suppressed sequence
was acquired into 32 k data points, averaged over 32 scans. The total acquisition time was 2 min and
23 s, with a relaxation delay of 2 s. A mixing time of 150 ms and a spectral width of 9058 Hz were used.
The free induction decays (FIDs) obtained were processed using Bruker Biospin TOPSPIN software
(version 2.1). Prior to Fourier transformation, the FIDs were multiplied by an exponential weight function
corresponding to a line broadening of 0.3 Hz. Spectra were phased, baseline-corrected and referenced
to the TSP at δ = 0 ppm.
Assignments of NMR signals were confirmed through the standard two-dimensional experiments;
namely, 1H,1H-COSY, 1H,1H-TOCSY, 1H,13C-HSQC, and 1H,13C-HMBC. Standard homonuclear COSY
and TOCSY experiments with water suppression were acquired with 2k data points in t 2 domain and
384 increments in t1, each with 64 scans. A spectral width of 13 ppm for both dimensions was employed.
A mixing time of 70 ms was used for TOCSY experiment. The parameters selected for heteronuclear
HSQC and HMBC experiments were: 2k data points in t2 domain and 384 increments in t1, each with
256 scan; a spectral width of 13 ppm for f2 dimension and 260 ppm for f1 dimension. In addition, short-
range and long-range couplings were set to 145 Hz and 10 Hz, respectively. Two-dimensional data sets
were enhanced using the technique of forward linear prediction in first dimension on the real data points
to 512 and zero filling to 1024 and squared sine bell window functions in both dimensions prior to Fourier
transformation.
Assignment of resonances in the 1H-NMR spectra from pork tenderloin exudates was based on
previous studies (Castejón et al., 2015). These assignations were confirmed through the information of
connectivity between the spins obtained from 2D experiments.
A total of 72 spectra were subjected to multivariate analysis. Principal Component Analysis (PCA)
and Partial Least Squares (PLS) were performed using AMIX software (version 3.9.11, Bruker Biospin)
and MetaboAnalyst (Version 3.0, URL: http://www.metaboanalyst.ca) (Xia, Sinelnikov, Han, & Wishart,
2015; Xia, & Wishart, 2016). The 1D spectra of pork exudates were data reduced over the shift range
of δ= 8.94 - 0.50 ppm into spectral integral regions (buckets) of 0.04 ppm. These constant buckets
widths were chosen due to work with buffered exudate samples. Regions with spectral artefacts (i.e.,
255
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
unsuppressed water region) were excluded from bucketing. A total of 212 buckets were defined, and
the integral of each of them, calculated by summing up the intensities of all its points, was divided by
the total number of points in the bucket. PCA and PLS were performed using the data from the bucketed
1D spectra of pork exudates.
In the first part of the work, exudate profiles of the unirradiated samples were evaluated. Figure 1
shows the comparison between representative 1H-NMR spectra of exudate samples stored during 1 (A),
6 (B) and 12 (C) days. The chart gives a general overview of the main metabolites present in the
spectrum (carnosine, creatine and lactate) and the spectral regions where fundamental nutrients
appear.
Fig. 1. 1H-NMR spectrum, obtained after ~ 2 minutes of analysis time in a 500 MHz spectrometer, which enables the determination
of the main components of the non-irradiated pork exudates at 1(A), 6(B) and 12(C) days of storage.
As previously mentioned, signal assignments for 1H-NMR spectra from pork exudates were checked
out though the 2D NMR spectra. As a result, it allowed us to confirm that all the metabolites identified in
256
IV. Resultados
García-García et al., 2018, IV.2.7, 1-17
the pork exudates spectra had been described previously in studies carried out in our laboratory
(Castejón et al.,2015). Therefore, our research was focused on, as set forth below, changes in the main
metabolites between samples subjected to different storage times and conservation irradiation dose.
At this point, it was noticed that it was possible to extract additional information though the spectral
data that allowed to predict samples pH value. The pH-dependent 1H chemical shift (δ) of the aromatic
histidine signals from carnosine showed a statistically significant relationship (P < 0.05) between pH
value of the sample and its δ. This fact has been recently reported in salchichón sausages (García-
García et al., 2018). This finding constitutes an interesting approach to study changes in meat pH directly
from the spectral information, especially taking into account that pH of muscle tissue is extremely
important to meat science since the pH at specific times during the conversion of muscle to meat, as
well as the ultimate pH of meat, affects many quality factors (Li et al., 2014).
Exudate pork samples were analyzed according to their storage time under an unsupervised
dimensionality reduction approach, Principal Component Analysis (PCA). Eighteen samples of 212
buckets each, ranging from 1, 6 and 12 days after the purchase. As a result, 96.1% of the variance was
explained by the first six Principal Components (PCs).
A) B)
B - 6 days
A - 1 day
C - 12 days
Figure 2. Left A) Scores plot between PC1 and PC2. Each data point corresponds to one pork exudate sample. The explained
variances are shown in brackets. Right B) Expanded regions of 1H spectra of pork exudates at 1, 6 and 12 days of storage. The
graphic shows the evolution of nucleotides (IMP, Ino and Hx) and degradation of sugars (α, β-Glc and α-Man) involved in meat
storage processes.
257
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
The PCA score plot for the first two PCs (Fig. 2A) shows a clear cluster of the exudate spectra
according with their storage time explaining 81.3% of the total variance. The best direction that explains
the variance in the data set (PC1) resulted closely related to sample storage.
The important features identified by one-way ANOVA and post-hoc analysis (Table S1 in
supplementary data) correspond with buckets connected with the expected metabolic changes that take
place during beef storage (Castejón et al., 2015). Fig. 2B illustrates, in terms of spectral peak variations,
the main nucleosides (IMP, Ino and Hx) and sugar (α, β-Glc and α-Man) changes that can be detected
in pork exudate spectra due to the meat evolution during the storage times considered. Regarding
nucleotides, figure reflects the well-known storage degradation pathway
(ATP/ADP>AMP>IMP>Ino>Hx>Xn). It is important to note that the nucleotide profile in pork exudate
had the same behavior that previously was observed on samples of beef exudate. In fact, this profile
after 12 days of storage was practically the same in both species. On the other hand, figure shows the
signals from anomeric protons of carbohydrates. Signals from α, β-Glc and α-Man were observed at
one postmortem day but after 6 days of storage they were not longer detected. This evolution, related
to microbial degradation, is contrary to what was observed in beef exudates where after 12 days of
storage the characteristic doublets of the carbohydrates could be observed. Therefore, the analysis of
the evolution of sugars could serve as an indicator of the microbial load present and also, depending on
the final fermentation products detected in the 1H-NMR spectrum, could get information about the types
of microorganisms present (Müller, 2008). There were also significantly buckets related with the increase
levels of free amino acids, associated with the proteolysis process which starts after slaughtering, and
the transformation of Gln to Glu that happen during storage.
C - 12 days
A - 1 day
B - 6 days
Figure 3. Multivariate analysis using PLS-DA. PLS-DA 2D score plot of component 1 versus component 2 (66.8% and 8.8%
respectively) for comparison of the global metabolite profiles of pork exudates stored at 1, 6 and 12 days
258
IV. Resultados
García-García et al., 2018, IV.2.7, 1-17
To maximize the differences between groups and to determine the main variables that contribute to
discrimination, a supervised pattern recognition method was further employed for spectral data. As
shown in the PLS-DA score plots (Fig. 3), the exudate samples with different storage time were clearly
separated into three groups according with the three storage times analyzed. Therefore, both the
unsupervised (PCA) and the supervised (PLS-DA) analysis revealed a notable and consistent
separation between the three groups according with their storage time.
In order to isolate the effect of irradiation treatment, we focused on exudate samples at one day of
storage. Analysis of these exudate NMR spectra at 0, 1, 2 and 6 kGy did not show significant differences
or characteristic signals that could be associated with preservative treatments that use beta radiation as
previously was described in the literature (Castejón et al., 2015; Zanardi et al., 2015; Villa et al., 2013).
The PLS-DA supervised analysis of the four groups did not show a classification according to the dose
of irradiation applied although a trend was observed (see Figure S1 in supplementary information). This
tendency has been associated with the conservative effect of irradiation. The non-irradiated samples
already undergone an evolution during the first 24 hours of storage.
In order to evaluate jointly the effect of storage time and the dose of irradiation applied, an
unsupervised multivariate analysis of the 72 samples (6 porks x 4 doses x 3 storage times) was
performed, where a Pareto data scaling showed the best PCA classification (Figure 4).
A) B)
B - 6 days
0 kGy
1 kGy 0 kGy
2 kGy
6 kGy
1 kGy
C - 12 days
A - 1 day
2 kGy
6 kGy
Figure 4. A) Scores plot between PC1 and PC2. Each data point corresponds to one pork exudate sample. The explained
variances are shown in brackets. B) Expanded regions of 1H spectra from non-irradiated (0 kGy) and irradiated (1, 2 and 6 kGy)
samples of pork exudates stored during 1, 6 and 12 days. Abbreviations: Glc, Glucose; Man, Mannose.
259
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
As it is known, PCA is an unsupervised method aiming to find the directions that best explain the
variance in a data set (X) without referring to class labels (Y). The explained variance of each PC is
shown in the corresponding diagonal cell. In our analysis with only 5 PCs, 87.7% of the variance was
explained. The first principal component (PC1) explained 42.3% of the variance in the initial data set,
the second one (PC2) explained 25.9% and the third (PC3) 8.3%. Figure 4 shows the 2D PCA score
plot where the classification of the groups according to their storage time is clearly observed. In addition,
the longer the storage time the easier is the classification between groups of the same storage time as
a function of the dose of irradiation applied. As reported by the chart, PC1 mostly accounted for exudate
classification according to meat storage while PC2 seems to separate the samples in accordance with
irradiation dose. In Fig. 4B, the 1H-NMR expanded regions shows metabolite degradation during storage
and how these changes could be slowed down by the effect of E-beam irradiation treatment. As shown
in the figure, the intensity of Glc and Man signals and therefore, their concentration, seems to have a
clear correlation with applied irradiation dose. After 1 kGy a preservative effect is shown and this effect
increase with the applied dose. This behavior had a good correlation with previous works developed in
our laboratory on samples of smoked and fresh salmon (Villa et al., 2013; Castejón et al., 2016).
However, in this case, the conservative effect of the E-beam irradiation was noticeably lower. This fact
is not conflicting, since, as it has been demonstrated, and it was expected, the irradiation dose needed
to achieve the desired benefit in each product depends on its intrinsic characteristics.
3.2 Correlations among metabolite peak integrals, storage time and irradiation dose.
10
260
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
Figure 5. Response surface plot showing the relation of the percentage of metabolite integral versus storage time and irradiation
dose.
The figure clearly shows how metabolite integral is affected by storage time and irradiation dose
applied. The correlation study carried out between the values of the integrals of the selected metabolites,
storage times and irradiation dose applied allowed to define a predictive model according to the
regression equation (R2 = 0.96), meaning, T= Time (days) and ID= Irradiation dose (kGy):
According to the results, the effect and the interaction of the irradiation dose and the storage time
had a clear influence on the value of the integral of the selected metabolites from exudate of pork and
this effect could be predicted with the proposed model. This study reaffirms the potential of the exudate
as a new food matrix to monitor and to control meat systems and raises the possibility that the NMR-
based metabolomic analysis of meat exudate could lead to be an official method to myosistems
monitorization.
4. Conclusions
The analysis of the metabolic profile of pork exudate provides valuable information about the
changes that the different compounds suffer during storage and how these changes have been affected
by the conservative treatment with E-beam irradiation. The spectral information allowed to monitor the
metabolic changes associated with the storage process and to evaluate as much the conservative as
dose effects of the irradiation treatment. In addition, it was possible to establish relationships among the
value of the metabolites integral with the absorbed radiation dose and the storage time, which permitted
to obtain a predictive model. These findings could be the first step to develop a new method for non-
destructive analysis and for the control of meat conservation as well as to evaluate irradiation treatment.
11
261
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
5. Supporting Information
6. Acknowledgements
The authors acknowledge financial support from MICINN (grants AGL2010-19158 and CTQ2010-
14936) and the members of the CAI of NMR and RSE of the Complutense University for the cooperative
collaboration, with special mention to Elena Sáez.
12
262
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
References
Alonso-Salces, R. M., Moreno-Rojas, J. M., Holland, V. M., Reniero, F., Guillou, C., Serra, F., & Segebarth, N.
(2007). NMR and Isotopic Fingerprinting for Food Characterization. European Comission.
[ASTM] American Society for Testing and Materials. (2000). E1650-97e1 Standard Practice for Use of Cellulose
Acetate Dosimetry Systems. Standards on Disc, 12.
Benet, I., Ibañez, C., Guàrdia, M. D., Solà, J., Arnau, J., & Roura, E. (2015). Optimization of stir-bar sorptive
extraction (SBSE), targeting medium and long-chain free fatty acids in cooked ham exudates. Food Chemistry,
185, 75-83.
Bertram, H. C. (2017). NMR spectroscopy and NMR metabolomics in relation to meat quality. In Elsevier (1st Ed.)
New Aspects of Meat Quality: From Genes to Ethics, (Chapter 13).
Capitani, D., Sobolev, A. P., Tomassini, A., Sciubba, F., De Salvador, F. R., Mannina, L., & Delfini, M. (2012). Peach
fruit: metabolic comparative analysis of two varieties with different resistances to insect attacks by NMR
spectroscopy. Journal of Agricultural and Food Chemistry, 61(8), 1718-1726.
Castejón, D., Villa, P., Calvo, M. M., Santa‐María, G., Herraiz, M., & Herrera, A. (2010). 1H‐HRMAS NMR study of
smoked Atlantic salmon (Salmo salar). Magnetic Resonance in Chemistry, 48(9), 693-703.
Castejón, D., García-Segura, J. M., Escudero, R., Herrera, A., & Cambero, M. I. (2015). Metabolomics of meat
exudate: Its potential to evaluate beef meat conservation and aging. Analytica chimica acta, 901, 1-11.
Castejón, D., García-Segura, J. M., Herrera, A., & Cambero, M. I. (2016). NMR-detection of methylamine
compounds in Atlantic salmon (Salmo salar) subjected to E-beam irradiation. Food Control, 60, 455-460.
Chauhan, S. K., Kumar, R., Nadanasabapathy, S., & Bawa, A. S. (2009). Detection methods for irradiated foods.
Comprehensive Reviews in Food Science and Food Safety, 8(1), 4-16.
Cheng, J. H., Dai, Q., Sun, D. W., Zeng, X. A., Liu, D., & Pu, H. B. (2013). Applications of non-destructive
spectroscopic techniques for fish quality and safety evaluation and inspection. Trends in Food Science &
Technology, 34(1), 18-31.
Farkas, J. (2006). Irradiation for better foods. Trends in Food Science & Technology, 17(4), 148-152.
FAO, W. F. P. IFAD (2012). The state of food insecurity in the world 2012. Economic Growth is necessary but not
Sufficient to Accelerate Reduction of Hunger and Malnutrition. FAO, Rome, Italy, 1-61.
García-García, A. B., Lamichhane, S., Castejón, D., Cambero, M. I., & Bertram, H. C. (2018). 1H HR-MAS NMR-
based metabolomics analysis for dry-fermented sausage characterization. Food Chemistry, 240, 514-523.
García-Márquez, I., Cambero, M. I., Ordóñez, J. A., & Cabeza, M. C. (2012). Shelf-life extension and sanitation of
fresh pork loin by E-beam treatment. Journal of Food Protection, 75(12), 2179-2189.
Kim, G. D., Jung, E. Y., Lim, H. J., Yang, H. S., Joo, S. T., & Jeong, J. Y. (2013). Influence of meat exudates on the
quality characteristics of fresh and freeze-thawed pork. Meat Science, 95(2), 323-329.
Jung, Y., Lee, J., Kwon, J., Lee, K. S., Ryu, D. H., & Hwang, G. S. (2010). Discrimination of the geographical origin
of beef by 1H NMR-based metabolomics. Journal of Agricultural and Food Chemistry, 58(19), 10458-10466.
Li, P., Wang, T., Mao, Y., Zhang, Y., Niu, L., Liang, R., & Luo, X. (2014). Effect of ultimate pH on postmortem
myofibrillar protein degradation and meat quality characteristics of Chinese yellow crossbreed cattle. The
Scientific World Journal, 2014.
Listrat, A., Lebret, B., Louveau, I., Astruc, T., Bonnet, M., Lefaucheur, L., & Bugeon, J. (2016). How muscle structure
and composition influence meat and flesh quality. The Scientific World Journal.
http://dx.doi.org/10.1155/2016/3182746.
Mannina, L., Sobolev, A. P., & Viel, S. (2012). Liquid state 1H high field NMR in food analysis. Progress in Nuclear
Magnetic Resonance Spectroscopy, 66, 1-39.
13
263
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
Martinez, I., Bathen, T., Standal, I. B., Halvorsen, J., Aursand, M., Gribbestad, I. S., & Axelson, D. E. (2005).
Bioactive compounds in cod (Gadus morhua) products and suitability of 1H NMR metabolite profiling for
classification of the products using multivariate data analyses. Journal of Agricultural and Food Chemistry,
53(17), 6889-6895.
Müller, V. (2008). Bacterial Fermentation. In Encyclopedia of Life Sciences (ELS), John Wiley & Sons, Chichester,
1-8.
Pioselli, B., Paredi, G., & Mozzarelli, A. (2011). Proteomic analysis of pork meat in the production of cooked ham.
Molecular Biosystems, 7(7), 2252-2260.
Ritota, M., Casciani, L., Failla, S., & Valentini, M. (2012). HRMAS-NMR spectroscopy and multivariate analysis
meat characterization. Meat Science, 92(4), 754-761.
Rodrigues, J. E. A., Erny, G. L., Barros, A. S., Esteves, V. I., Brandão, T., Ferreira, A. A., & Gil, A. M. (2010).
Quantification of organic acids in beer by nuclear magnetic resonance (NMR)-based methods. Analytica
Chimica Acta, 674(2), 166-175.
Siciliano, C., Belsito, E., De Marco, R., Di Gioia, M. L., Leggio, A., & Liguori, A. (2013). Quantitative determination
of fatty acid chain composition in pork meat products by high resolution 1H NMR spectroscopy. Food chemistry,
136(2), 546-554.
Sofos, J. N. (2008). Challenges to meat safety in the 21st century. Meat science, 78(1), 3-13.
Straadt, I. K., Aaslyng, M. D., & Bertram, H. C. (2014). An NMR-based metabolomics study of pork from different
crossbreeds and relation to sensory perception. Meat Science, 96(2), 719-728.
Sundekilde, U. K., Rasmussen, M. K., Young, J. F., & Bertram, H. C. (2017). High resolution magic angle spinning
NMR spectroscopy reveals that pectoralis muscle dystrophy in chicken is associated with reduced muscle
content of anserine and carnosine. Food Chemistry, 217, 151-154.
Villa, P., Castejón, D., Herraiz, M., & Herrera, A. (2013). 1H‐HRMAS NMR study of cold smoked Atlantic salmon
(Salmo salar) treated with E‐beam. Magnetic Resonance in Chemistry, 51(6), 350-357.
World Health Organization. (1999). High-Dose Irradiation: Wholesomeness of Food Irradiatied with Doses above
10 kGy. World Health Organization, No. 890.
Xia, J., Sinelnikov, I. V., Han, B., & Wishart, D. S. (2015). MetaboAnalyst 3.0—making metabolomics more
meaningful. Nucleic Acids Research, 43(W1), W251-W257.
Xia, J., & Wishart, D. S. (2016). Using MetaboAnalyst 3.0 for Comprehensive Metabolomics Data Analysis. Current
Protocols in Bioinformatics, http://dx.doi.org/10.1002/cpbi.11.
Zanardi, E., Caligiani, A., Palla, L., Mariani, M., Ghidini, S., Di Ciccio, P. A., & Ianieri, A. (2015). Metabolic profiling
by 1H NMR of ground beef irradiated at different irradiation doses. Meat science, 103, 83-89.
14
264
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
SUPPLEMENTARY MATERIAL
(3 pages)
Table of contents
1). Table S1. Variable buckets used for the multivariate data analysis……………………….…p. 16
15
265
IV. Resultados
García-García et al., 2018, enviado a Food Reseach International, IV.2.7, 1-17
1). Table S1. Variable buckets used for the multivariate data analysis. The table includes the magnitude
of their variation during storage and the assignment of the main metabolites present in each bucket.
Name f.value p.value menosLOG10(p) FDR Post-hoc tests 0-6 days 6 -12 days Main Metabolite
2.54 438.11 4.96E-14 13.30 5.16E-12 1 - 12; 6 - 12 Gln
3.42 423.16 6.41E-14 13.19 5.16E-12 1 - 12; 1 - 6 a,b-Glc
8.22 412.49 7.74E-14 13.11 5.16E-12 12 - 1; 6 - 1 Hx
0.98 387.91 1.22E-13 12.92 5.16E-12 12 - 1; 6 - 1; 12 - 6 Ile, Val, Leu
3.62 387.87 1.22E-13 12.92 5.16E-12 12 - 1; 6 - 1 Thr, Val
3.26 376.94 1.50E-13 12.82 5.31E-12 12 - 1; 12 - 6 NI
4.66 329.28 4.05E-13 12.39 1.13E-11 1 - 12; 1 - 6 b-Glc
3.46 327.15 4.25E-13 12.37 1.13E-11 1 - 12; 1 - 6 b-Glc
5.26 237.17 4.45E-12 11.35 1.05E-10 1 - 12; 1 - 6 a-Glc
2.74 232.72 5.11E-12 11.29 1.08E-10 12 - 1; 12 - 6 NI
3.50 228.42 5.85E-12 11.23 1.13E-10 1 - 12; 1 - 6 a-Glc
2.50 206.83 1.20E-11 10.92 2.12E-10 1 - 12; 1 - 6; 6 - 12 Gln
3.74 182.64 2.95E-11 10.53 4.81E-10 1 - 12; 1 - 6 b-Glc
3.82 165.90 5.89E-11 10.23 8.92E-10 1 - 12; 1 - 6 a-Glc
7.14 153.96 1.01E-10 10.00 1.34E-09 1 - 12; 6 - 12 NI
8.26 153.57 1.02E-10 9.99 1.34E-09 1 - 12; 1 - 6 Ino, AMP
5.22 152.55 1.07E-10 9.97 1.34E-09 1 - 12; 1 - 6 a-Glc
6.10 140.83 1.90E-10 9.72 2.04E-09 1 - 12; 1 - 6 Ino
2.46 140.63 1.92E-10 9.72 2.04E-09 1 - 12; 1 - 6 Gln
3.86 139.75 2.01E-10 9.70 2.04E-09 1 - 12; 1 - 6 a-Glc, Ino
2.82 139.60 2.02E-10 9.69 2.04E-09 1 - 12; 6 - 12 NI
7.02 134.44 2.64E-10 9.58 2.55E-09 1 - 12; 6 - 12 NI
4.46 131.53 3.09E-10 9.51 2.85E-09 1 - 12; 1 - 6 Ino, IMP
Abbreviations: AMP, Adenosine 5´-phophate; Glc, Glucose; Gln, Glutamine; Hx, Hypoxanthine; Ile,
Isoleucine; IMP, Inosine 5´-phosphate; Ino, Inosine; Leu, Leucine; NI, Not Identified; Thr, Threonine;
Val, Valine.
16
266
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
2). Figure S1. PLS-DA scores plot between Component-1 and Component-2. Each data point
corresponds to one pork exudate spectrum at one day of store time at 0, 1, 2 and 6 kGy. The explained
variances are shown in brackets.
17
267
IV. Resultados
García-García et al., 2018, enviado a Food Research International, IV.2.7, 1-17
3). Figure S2. Representative exudates vs intact muscle samples spectra. Region between 1.00 and
4.5 ppm is shown. The same metabolites are present in each sample type.
18
268
V. Discusión integradora
269
Con el desarrollo de esta Tesis Doctoral se ha pretendido avanzar en la aplicación de
la RMN al análisis de alimentos, profundizando en el potencial de las distintas técnicas
de RMN para el estudio de miosistemas de naturaleza cárnica. Así, se ha empleado la
IRM para la monitorización del proceso de elaboración de jamón curado, se ha utilizado
la relaxometría T2 de RMN para el análisis de productos cárnicos elaborados a partir de
piezas cárnicas enteras (jamón curado) y a partir de carne picada (embutidos crudos
curados) con el fin de estudiar los cambios en la movilidad y distribución del agua y la
grasa y su relación con la microestructura del producto y se ha recurrido a la
espectroscopía de RMN para el análisis de distintas matrices cárnicas (jamón curado,
embutidos crudos curados y exudados cárnicos) tratando de evaluar y monitorizar
distintos procesos (curado, fermentación e irradiación). En definitiva, el objetivo global
planteado fue dar un paso más en la aplicación de la RMN a la monitorización y control
de procesos en la industria cárnica. Tras el análisis conjunto de los resultados obtenidos,
consideramos que se han conseguido importantes aportaciones de interés científico y
técnico.
271
V. Discusión integradora
El jamón curado es un derivado cárnico muy apreciado por los consumidores por su
gran calidad sensorial y nutritiva, condicionadas por las particularidades del animal del
que procede, de su alimentación y del proceso de elaboración. Se trata de un producto
de indiscutible valor añadido que requiere un control exhaustivo en todas las etapas de
su procesado.
273
V. Discusión integradora
En los mapas T1, las intensidades más claras y brillantes indican valores de T1 más
altos asociados a un mayor contenido de agua libre, con elevada movilidad y dificultad
para ceder su energía al medio. La asociación del agua a macromoléculas produce una
reducción del T1, lo que se manifiesta en tonalidades grisáceas, más oscuras cuanto
menor es el valor de T1. De forma similar, la pérdida de agua dará lugar a un
oscurecimiento de las zonas en proporción al descenso del valor de T1. Además, las
zonas de tejido adiposo en estos mapas aparecen oscuras, debido a que la grasa posee
un valor de T1 bajo.
En los mapas CDA las zonas donde hay mayor difusión de agua se muestran con
tonalidades claras y brillantes, mientras que las estructuras densas y organizadas, con
bajo CDA, aparecen oscuras.
274
V. Discusión integradora
En la presente Tesis Doctoral el empleo de la IRM resultó muy útil para la obtención
de imágenes de la sección interna de distintos músculos de jamón. Frente a otras
técnicas de imagen, la IRM ofrece la posibilidad de obtener parámetros cuantitativos,
como son los valores medios de T1, T2 y CDA, que permiten el procesado estadístico y el
establecimiento de relaciones entre atributos estructurales, características
fisicoquímicas y propiedades reológicas. El análisis de las imágenes, a través del cálculo
de los tiempos de relajación anteriormente mencionados y el CDA, permitió la
discriminación entre el tejido muscular y los depósitos de grasa. Esta técnica ha
permitido el seguimiento de los cambios estructurales de los componentes de la pieza
cárnica a lo largo de las distintas etapas de elaboración (véase apartado I.4.3.3.). En
estudios previos, se ha demostrado que las medidas volumétricas de grasa y músculo de
regiones bien definidas de la espalda y pata del cerdo proveen información precisa de la
composición corporal total del cerdo (Mitchell et al., 2001) pudiendo incluso utilizarse
esta técnica como método de clasificación de canales. En este trabajo se refuerza la idea
de emplear la IRM para monitorizar, de forma no destructiva, el proceso de producción
de jamón curado, principalmente, a través del seguimiento del curso de la transferencia
de agua y sal. Este procedimiento facilitaría en gran medida el control en las líneas de
producción. Adicionalmente, en esta Tesis Doctoral se ha propuesto establecer
relaciones matemáticas entre los parámetros de relajación de IRM (T2, T1 y CDA) y el
comportamiento reológico del derivado cárnico lo que permitiría la predicción de
parámetros de textura, como la dureza. Con esta nueva proyección de aplicación de los
parámetros de IRM, se abre la posibilidad de incorporar las técnicas no destructivas de
resonancia a la estimación de la calidad de las piezas de jamón y a la determinación de
los tiempos más adecuados de consumo. Este aspecto requiere mayores investigaciones
275
V. Discusión integradora
para relacionar los parámetros de IRM con la percepción sensorial del comportamiento
reológico y constituyen una de las líneas de trabajo futuro abiertas para proseguir a
partir de los resultados expuestos en la presente Tesis Doctoral.
276
V. Discusión integradora
En la Figura V.I se representa la distribución de los distintos tipos de agua que han
sido descritos en carne fresca mediante el uso de la relaxometría. En ella se pone de
manifiesto el comportamiento multicomponente de la relajación T2, puesto que cada
señal se corresponde con una población o grupo de protones con un grado de asociación
determinado a la matriz proteica. En general, cuanto menor es T2, mayor será el grado
de asociación o ligazón de los protones a las estructuras de la muestra.
Figura V.I. Distribución de los tiempos de relajación T2b, T21 y T22 en músculo de cerdo. Cuanto más fuertemente
ligadas se encuentran las moléculas de agua, más corto es el tiempo de relajación T 2. (Bertram y Andersen, 2008).
277
V. Discusión integradora
Figura V.2. Principales componentes musculares y localizaciones de los compartimentos del agua muscular
(Adaptado de Pearce et al., 2011).
278
V. Discusión integradora
V.1.2.1. Modificaciones derivadas del curado de piezas cárnicas enteras versus matrices
cárnicas picadas
El análisis de las distribuciones T2 obtenidas para jamón curado (Trabajo 2), revela la
existencia de distintos componentes T2 distintos en las muestras cárnicas, en función del
grado de asociación de los protones de la muestra con la matriz proteica: T2b, T21, T22.
Como se mencionó anteriormente en esta memoria, estos tres componentes han sido
descritos previamente en la carne fresca (Bertram et al., 2001; Bertram et al., 2002;
Bertram et al., 2004) y son comunes a los distintos derivados cárnicos [modelos cárnicos
(Trabajo 3) y embutidos crudos curados (Trabajo 4)] examinados, si bien su
presencia/ausencia se ve alterada dependiendo del producto y/o etapa de maduración.
En este sentido, el componente mayoritario, T21 está presente siempre, sin excepción, y
las diferencias entre productos y tiempos se hacen visibles a través de los componentes
T2b y T22.
279
V. Discusión integradora
Figura V.3. Transformación desde estado sol hasta estado coloidal gel y constitución de la red tridimensional
proteica.
En todos los productos estudiados, el avance del curado se hizo patente en las
distribuciones obtenidas. Por un lado, la deshidratación de la matriz proteica, propia del
curado/maduración, se manifestó a través de cambios en la distribución y movilidad de
los protones, de manera que en las distribuciones T2 se produjo un desplazamiento de
las señales hacia tiempos de relajación más cortos, especialmente visible a través del
componente mayoritario T21. Este comportamiento fue descrito previamente por MØller
280
V. Discusión integradora
En el caso concreto del jamón curado (Trabajo 2), cabe destacar la presencia de una
nueva población, T2´, con un tiempo de relajación entorno a los 80 ms, no contemplado
hasta el momento en la bibliografía. Este componente no está presente en todas las
distribuciones de jamón, sino que se hace visible a partir de un cierto grado de curado,
concretamente, durante la etapa de secado. El valor del tiempo de relajación de dicha
señal (más largo que el agua intramiofibrilar y más corto que el agua extramiofibrilar),
junto con el alto grado de infiltración de la grasa en la matriz muscular en este punto del
proceso (Fernández et al., 2007), han llevado a asociar el componente T2´ con los
protones de la fase grasa. Además, la obtención de imágenes de microscopía electrónica
de barrido (MEB) resultó de gran ayuda para interpretar las modificaciones
estructurales. Su análisis permitió observar la reducción del tamaño intrínseco de poro
de la matriz proteica, debido a los fenómenos de solubilización proteica, absorción de
sal y deshidratación, típicos del curado, así como la visualización de la evolución de la
infiltración grasa en el músculo.
Por tanto, en el caso del jamón curado (Trabajo 2), fue posible tanto monitorizar la
maduración como observar cambios en la distribución de la grasa, gracias a la aparición
de la señal T2´, cuya presencia puede utilizarse como marcador de etapas avanzadas del
curado. Además, se detectaron diferencias de comportamiento entre músculos, puesto
que el efecto de la maduración se hizo más notorio en el músculo semimembranoso. El
componente T21 en las distribuciones obtenidas para este músculo a partir de la etapa
de post-salado presentó tiempos de relajación más bajos en comparación con el
músculo bíceps femoral, sugiriendo que los fenómenos madurativos afectan más
rápidamente al músculo semimembranoso que, por tratarse de un músculo externo, de
contacto directo con el aire, experimenta antes los efectos de las dinámicas de difusión
de agua y sal (Arnau et al., 1995) y eso se refleja en los tiempos de relajación.
Cabe destacar que el componente T2´ presente en jamón curado, no se hizo visible ni
en las distribuciones T2 obtenidas en los modelos cárnicos ni en los embutidos curados
281
V. Discusión integradora
(Trabajos 3 y 4). Sin embargo, sí aparece una señal similar en otros productos
madurados, como el queso curado de oveja (Cruz-Díaz, 2016) en el que se describió un
comportamiento bimodal de la relajación transversal T2, de forma que el componente
que presentaba un tiempo de relajación más bajo (T2 < 20 ms) se relacionó con una
fracción de agua menos móvil asociada a la fracción proteica y el componente con un
tiempo de relajación más largo (T2 = 20-80 ms), se relacionó con la fracción de agua más
móvil, asociada a la fase grasa. En relación con los quesos, otros autores coinciden en
dicho comportamiento bimodal. Así, Kuo et al. (2001) y Luo et al. (2013) lo describieron
en queso Mozzarella y Mulas et al. (2013) en queso Pecorino y queso Fiore Sardo.
282
V. Discusión integradora
Músculo Picado y
+ fermentación
MP Gel
Grasa Tª, HR Tª, HR
+ masa cárnica
Ingredientes no
cárnicos
Extracción proteica
Desnaturalización
Solubilización proteica Interacción proteína-proteína
Red tridimensional
Exposición de grupos
Figura V.4. Proceso de gelificación proteica en músculo íntegro y masa cárnica. MP: proteínas miofibrilares. HR:
humedad relativa.
Respecto a los sistemas modelo con y sin adición de proteasa (Trabajo 3), cabe
destacar que las distribuciones T2 mostraron notables diferencias al comparar ambos
lotes de productos (Díaz et al., 1997; Fernández et al., 2000). Estas diferencias se hacen
visibles, sobre todo, en los componentes minoritarios, T2b y T2b2, y en el componente de
relajación más lenta, T22. Además, para ayudar a interpretar las distribuciones T2, se
realizaron pruebas clásicas, como electroforesis SDS-PAGE y ensayo de fluorescamina.
Con estas técnicas se pretendía observar desde distintos ámbitos qué ocurre con las
proteínas miofibrilares durante un proceso de maduración con y sin adición de enzima.
En este sentido, el patrón electroforético obtenido para las proteínas miofibrilares de
ambos lotes fue significativamente diferente. En los modelos sin adición de proteasas se
283
V. Discusión integradora
produjo una degradación proteica progresiva que se corresponde con una situación
normal de maduración. Sin embargo, en los modelos con proteasas, la degradación fue
patente desde el principio, donde ya no fue posible distinguir compuestos de alto peso
molecular debido a la proteólisis experimentada desde las etapas iniciales del proceso.
En los modelos sin proteasa, las poblaciones T2b y T2b2 constituyen dos señales
distintas hasta el día 3 de producción, aproximadamente. A partir de entonces, en la
región entre 1-10 ms solo se distingue una única señal. Este hallazgo, también visible en
los embutidos curados, se ha atribuido a la absorción de sal por parte de la matriz cárnica
que tiene lugar durante el proceso de maduración. Sin embargo, en los modelos cárnicos
a los que se les ha añadido proteasas exógenas, ambas poblaciones, T 2b y T2b2,
permanecen separadas casi durante todo el proceso y no es hasta el día 7,
aproximadamente, cuando comienza a observarse un solo componente en la región
comprendida entre 1-10 ms. Lo que ocurre en este lote de producto es que el efecto de
la proteólisis impide que se produzca una adecuada absorción de la sal por la matriz
proteica, de hecho, la aparición de una sola señal de T2 bajo coincide con la aparición
del componente T22 en este tipo de productos, lo cual sugiere la necesidad de cierto
grado de agregación proteica para la aparición de estos fenómenos. La electroforesis
apoya la teoría de que a partir del día 7 comienza a producirse un acercamiento entre
las cadenas peptídicas previamente degradadas, probablemente favorecido por la
deshidratación en este punto del proceso, cuya reorganización favorece la
compactación de la matriz y los cambios mencionados previamente (García de Fernando
et al., 1991, Ordóñez et al., 1999).
284
V. Discusión integradora
La posibilidad de utilizar la IRM para monitorizar los cambios del contino en agua y la
concentración en sal en jamón curado ha sido previamente documentada por otros
autores. Guiheneuf et al., 1996 obtuvo mapas paramétricos de IRM (T1, T2 y densidad
protónica) de muestras de carne de cerdo curado que mostraron una disminución de
estos parámetros en función del aumento de la concentración de NaCl. Además, se han
establecido relaciones lineales (p < 0,0001) entre los contenidos de agua y de sal y los
parámetros de IRM (T1 y T2) en el músculo Longissimus dorsi salado con distintos
porcentajes de NaCl (Fantazzini, 2005). También se emplearon modelos de regresión
múltiple para construir un modelo que permita determinar el contenido en sal en los
músculos semitendinoso y bíceps femoral en jamón de Parma. T1, T2, su proporción
(T1/T2) y la densidad protónica fueron los parámetros empleados para establecer este
modelo, existiendo una alta correlación entre el contenido en sal (R2=0,90, p < 0,001) y
los parámetros de IRM (Fantazzinni, 2009).
285
V. Discusión integradora
286
V. Discusión integradora
En cualquier caso, los ensayos de tracción y el TFA, al igual que cualquier otro análisis
mecánico del comportamiento reológico, son destructivos y requieren grandes
287
V. Discusión integradora
Conforme a estudios anteriores (Monin et al., 1997, Serra et al., 2005) la dureza
aumentó durante la maduración. El descenso de la actividad de agua y el contenido en
agua fue acompañado de un aumento notable de la dureza. Del mismo modo, la
masticabilidad también aumentó, ya que es una característica reológica secundaria
(producto de la dureza por la cohesividad por la elasticidad). La sal también puede
incrementar la estabilidad y la rigidez de la estructura miofibrilar mediante su
mineralización. En relación con estos hechos, en los tres músculos considerados, se
observó un aumento progresivo de la dureza y la masticabilidad, alcanzando su valor
máximo al final de la maduración. El músculo que mostró los valores más altos de dureza
durante todo el proceso de maduración fue el semimembranoso, seguido del bíceps
femoral y, finalmente, el semitendinoso, el más suave de los tres músculos estudiados.
Este último mostró el contenido en grasa más alto, el nivel de mineralización más bajo
debido a su posición más interna y un nivel de humedad medio. La conjunción de estos
factores explicaría su menor resistencia mecánica en relación con el bíceps y, sobre todo
con el semimembranoso, que por su posición más externa experimenta una rápida e
intensa deshidratación
288
V. Discusión integradora
semimembranoso presentó los valores más altos para cohesividad y elasticidad durante
el proceso de maduración. El valor de la adhesividad también se redujo con el avance
del curado. El bíceps fue el músculo que presentó el valor más bajo al final del proceso.
Este estudio conjunto de los parámetros de IRM y derivados del TPA, ha permitido
obtener modelos de predicción de los atributos de textura (dureza, elasticidad,
adhesividad y cohesividad) a partir de T1, T2 y CDA que vendrían a indicar la existencia
de una fuerte relación entre los parámetros de IRM y la estructura muscular. La dureza
mostró una correlación negativa con T1, T2 y CDA, mientras que la elasticidad y la
cohesividad mostraron una relación positiva con dichos parámetros.
Además del análisis de TPA, en el caso del jamón, el análisis del comportamiento
reológico de distintos músculos durante el proceso de curado se ha completado con la
estimación de su resistencia a la rotura y del límite elástico mediante análisis de tracción
destructivos (Trabajo 2). En términos generales, se puede decir que ambos parámetros
se vieron incrementados durante la maduración, siendo especialmente notable su
progresión en el músculo externo semimembranoso.
289
V. Discusión integradora
290
V. Discusión integradora
de elevada calidad y valor añadido. Una de las ventajas más interesantes de los equipos
de IRM es que son lo suficientemente grandes como para permitir el estudio de piezas
cárnicas enteras, por ejemplo, jamones curados sin deshuesar. Esto permitiría el trabajo
en continuo, siendo posible el acoplamiento de cintas transportadoras a los equipos
para un análisis detallado de las piezas, lo cual resultaría muy atractivo desde el punto
de vista productivo. Por otro lado, el empleo de relaxómetros, equipos específicos para
los estudios de relaxometría, permitiría disponer de herramientas de gran utilidad para
el análisis estructural con un ahorro de costes y espacio asociado, simplificando los
requerimientos técnicos.
291
V. Discusión integradora
Este estudio constituye una aportación novedosa al ámbito cárnico, puesto que es el
primer trabajo, hasta donde los autores han podido averiguar, en el que se monitoriza
292
V. Discusión integradora
la evolución del perfil metabólico del jamón curado durante su elaboración a partir de
muestras íntegras de tejido.
293
V. Discusión integradora
En este punto, cabe destacar que durante el desarrollo de esta Tesis también se ha
realizado el estudio bioquímico sobre muestras de chorizo. Los datos obtenidos a partir
de estos experimentos mostraron, como era de esperar, que el perfil metabólico del
producto, así como su evolución durante la maduración era bastantes similares a los
obtenidos con el salchichón. En la Figura V.5 se muestra la ampliación de las zonas del
espectro en las que se observó mayor evolución de señales para este producto. Estos
resultados forman parte de una comunicación “Caracterización metabólica de chorizo
artesanal mediante espectroscopía de RMN”, presentada en las Jornadas UCM de
Investigación en Ciencias de la Salud para Pregraduados (Madrid, 2016).
Figura V.5. Principales cambios observados en los espectros de chorizo durante los días 0, 2 y 14 de maduración. Val:
valina; Ile: isoleucina; Ala: alanina; AA: ácido acético; EtOH: etanol; Glu: glucosa; Phe: fenilalanina; Trp: triptófano;
Tyr: tirosina.
294
V. Discusión integradora
Figura V.6. Comparación de espectros salchichón vs chorizo en día 0 de producción. A: espectros completos. B:
ampliación de la región de carbohidratos. C: asignación de las señales características de la sacarosa.
295
V. Discusión integradora
Figura V.7. Principales vías de fermentación de azúcares, microorganismos implicados y productos finales (Müller,
2008).
Durante estos primeros días del proceso y, también relacionado con la actividad
microbiana, comenzó a desencadenarse una proteólisis temprana que se tradujo en el
aumento de las señales correspondientes a aminóacidos: Ile, Leu, Val, α-Ala Lys, Arg,
Met y Glu aumentaron durante la etapa fermentativa, así como las señales aromáticas
de Phe, Tyr, Trp y Tyrm.
296
V. Discusión integradora
Al igual que en el caso del jamón curado, el estudio en embutidos resulta muy
novedoso, puesto que, hasta donde los autores han podido averiguar, es la primera vez
que se aplica la espectroscopía 1H HRMAS al análisis de muestras íntegras de embutidos
crudos curados con el fin de obtener el perfil metabólico completo. El trabajo revela que
la RMN es una herramienta con gran potencial para el análisis de muestras intactas, con
el fin de obtener información cualitativa (identificación de metabolitos) y cuantitativa
(integración de señales y variación) del producto.
297
V. Discusión integradora
El análisis por RMN del exudado liberado por la carne de vacuno irradiado permitió
seguir de forma indirecta la evolución de la matriz miofibrilar frente al tratamiento con
electrones acelerados. En conjunto, se observaron diferencias significativas en las
señales correspondientes a carnosina, carnitina, glutamato y metionina, entre muestras
irradiadas y no irradiadas (Figura V.8).
298
V. Discusión integradora
Figura V.8. Ampliación de la región del espectro de 1H RMN de exudado de ternera que muestra la diferencia en la
intensidad de señales en muestras irradiadas y no irradiadas.
299
V. Discusión integradora
Este estudio podría constituir el primer paso para el desarrollo de un nuevo método
de análisis para el control de la conservación de la carne y para la evaluación del
tratamiento de irradiación. La combinación de la metodología 1H RMN HRMAS junto con
el empleo de exudados cárnicos, permite el establecimiento de un protocolo de análisis
completamente no invasivo, no destructivo, con grandes ventajas para la industria
cárnica, desde el punto de vista de la rapidez analítica y la preservación del producto.
Cabe destacar que los resultados obtenidos en relación con los exudados
procedentes de miositemas no se limitan solamente a muestras cárnicas, sino que
también se han realizado pruebas con resultados muy satisfactorios en matrices
miofibrilares de pescado.
300
V. Discusión integradora
Figura V.9. Regiones del espectro de 1H RMN que muestra la variación de metabolitos en los días 1 y 12 de
almacenamiento(A) y PCA (B). OTMA: óxido de trimetilamina, DMA: dimetilamina, IMP: Inosina monofosfato
301
V. Discusión integradora
cambios que se producen tras la muerte del animal es la degradación del ATP para
formar adenosina difosfato (ADP), adenosina monofosfato (AMP), inosina monofosfato
(IMP), inosina (Ino) e hipoxantina (Hx). La relación de estos metabolitos constituye el
valor o índice K, utilizado como indicador de frescura del pescado en diversos países. La
espectroscopía de 1H RMN HRMAS permitió seguir la evolución de IMP a Ino y de Ino a
Hx. Además, este espectro también permitió identificar las señales del óxido de
trimetilamina (OTMA) y trimetilamina (TMA), cuya relación también puede utilizarse
como indicador de frescura. Los datos obtenidos mostraron claramente una reducción
de OTMA y un incremento de TMA con el tiempo de almacenamiento. Estos resultados
concuerdan con los obtenidos en estudios preliminares realizados en muestras de
salmón (Castejón et al., 2010; Castejón et al., 2016).
Figura V.10. Ampliación de región del espectro de 1H RMN de exudados de salmón que muestra el aumento de TMA
con el tiempo de almacenamiento. TMA: trimetilamina.
302
V. Discusión integradora
303
VI. Conclusiones / Conclusion
305
VI. Conclusiones
Los principales resultados del trabajo presentado en esta Tesis Doctoral se pueden
resumir en las siguientes conclusiones:
7. Los espectros de RMN de protón (1H RMN) de los exudados liberados por la carne
de cerdo son un fiel reflejo de los obtenidos de los músculos de los que provienen
y permiten monitorizar los mismos procesos descritos en las muestras sólidas.
Por tanto, se propone este material para el análisis quimiométrico no invasivo,
307
VI. Conclusiones
Conclusión final
308
VI. Conclusion
The principal results of the work presented in this Doctoral Thesis can be summarized
in the following conclusions:
2. T2 distributed data are typical of each production process and T2 populations can
be used as indicators of the ripening stage in dry cured ham (T2') and of protein
gel formation in the raw cured sausages (T22), as well as of the effect of the
addition of proteases in meat models.
4. The relaxation times (T1 and T2) and the apparent diffusion coefficient (ADC)
calculated by Magnetic Resonance Imaging (MRI) allow to obtain predictive
models for the estimation of textural parameters and physicochemical
characteristics of dry cured ham.
5. Proton nuclear magnetic resonance (1H NMR) spectroscopy, together with the
high resolution magic angle spinning (HRMAS) probe allows the analysis of intact
samples of dry cured ham and raw cured sausages, making possible the
monitoring of the glycolysis, proteolysis and lipolysis processes typical of
ripening.
7. Proton NMR spectra (1H RMN) of exudates released from pork are a faithful
reflection of the obtained for the muscle which they come from and allow to
monitor the same processes described in the solid samples. Therefore, pork
exudates are proposed as suitable matrix for the non-invasive chemometric
analysis, with a high potential for the study and control of pork quality.
309
VI. Conclusion
8. NMR metabolomics of pork exudates allows to estimate the storage time and to
evaluate the conservative effect of the E-beam treatment.
Final conclusion
The different NMR-based techniques have a high potential for the analysis of the
structure and composition of myosistems. The obtained information can be in used in
the meat industry for the monitoring of the manufacturing process, estimation of the
storage time and prediction of physicochemical and textural features.
310
VII. Bibliografía
311
VII. Bibliografía
Aberle, E. D.; Forrest, J. C.; Gerrard, D. E.; Mills, E.W. (2012). Principles of Meat Science. 5th ed.
Kendall Hunt Publishing Company, Dubuque, IA.
Albarracín, W.; Sánchez, I. C.; Grau, R.; Barat, J. M. (2011). Salt in food processing; usage and
reduction: a review. Int. J. Food Sci. Technol., 46, 1329-1336.
Andersen, R.H.; Andersen, H.J.; Bertram, H.C. (2007). Curing-induced water mobility and
distribution within intra- and extra-miofibrilar spaces of three pork qualities. Int. J. Food
Sci. Technol., 42, 1059-1066.
Andrés, A. I.; Cava, R.; Martin, D.; Ventanas, J.; Ruíz, J. (2005). Lipolysis in dry-cured ham:
Influence of salt content and processing conditions. Food Chem., 90, 523-533.
Antequera, T.; Caro, A.; Rodríguez, P. G.; Pérez, T. (2007). Monitoring the ripening process of
Iberian ham by computer vision on magnetic resonance imaging. Meat Sci., 76, 561-567.
Arnau, J.; Guerrero, L.; Casademont, G.; Gou, P. (1995). Physical and chemical changes in differ
ent zones of normal and PSE dry-cured hams during processing. Food Chem., 52, 63-69.
Batlle, N.; Aristoy, M. C.; Toldrá, F. (2000). Early postmortem detection of exudative pork meat
based on nucleotide content. J. Food Sci., 65, 413-416.
Bello, J. (2008). Jamón curado. Aspectos científicos y tecnológicos. Ed. Díaz de Santos, Madrid.
ISBN: 9788479788841.
Bligh, E. G., Dyer, W. J. (1959). Canadian Journal of Biochemistry and Physiology. Can J Biochem
Phys, 37, 911-917.
B.O.E. (2002). Real Decreto 142/2002, de 1 de febrero, por el que se aprueba la lista positiva
de aditivos distintos de colorantes y edulcorantes para su uso en la elaboración de
productos alimenticios, así como sus condiciones de utilización. Boletín Oficial del
Estado, 44, 20/2/2002, 6756-6799.
B.O.E. (2014). Real Decreto 474/2014, de 13 de junio, por el que se aprueba la norma de
calidad de derivados cárnicos. Boletín Oficial del Estado, 147, 18/06/2014, 46058-46078.
Berendsen, H. J. C. (1992). Rationale for using NMR to study water relations in foods and
biological tissues. Trends Food Sci. Technol., 3, 202-205.
313
VII. Bibliografía
Bertram, H. C.; Andersen, H. J. & Karlsson, A. H. (2001a). Comparative study of low-field NMR
relaxation measurements and two traditional methods in the determination of water
holding capacity of pork. Meat Sci, 57, 125-132.
Bertram, H. C.; Karlsson, A. H.; Rasmussen, M.; Pedersen, O. D.; Dønstrup, S.; Andersen, H. J.
(2001b). Origin of multiexponential T2 relaxation in muscle myowater. J Agric Food
Chem, 49, 3092-100.
Bertram, H. C.; Dønstrup, S.; Karlsson, A. H., Andersen, H. J. (2002). Continuous distribution
analysis of T2 relaxation in meat: An approach in the determination of water-holding
capacity. Meat Sci., 60, 279-285.
Bertram, H. C. & Andersen, H. J. (2004). Applications of NMR in meat science. Annu. Rep. NMR
Spectro., 53,158-202.
Bertram, H.C. & Andersen, H. J. (2006). Proton NMR Relaxometry in Meat Science. Modern
Magnetic Resonance, Springer, Netherlands, 1729-1733.
Bertram, H. C.; Andersen, H. J. (2007). NMR and the water-holding issue of pork. J. Anim.
Breed. Genet., 124, 35-42.
Bertram, H. C.; Meyer, R. L.; Zhiyun, W.; Zhou, X.; Andersen, H. J. (2008). Water distribution
and microstructure in enhanced pork. J. Agri. Food Chem., 56, 7201-7207.
Budiman, M.; Stroshine, R.L.; Cornillon, P. (2002). Moisture measurement in cheese analogue
using stretched and multi-exponential models of the magnetic resonance T2 relaxation
curve. Journal of Dairy Research, 69, 619-632.
Caplice, E. & Fitzgerald, G.F. (1999). Food fermentations: role of microorganisms in food
production and preservation. Int. J. Food Microbiol., 50, 131-149.
Caro, A.; Durán, M.; Rodríguez, P. G.; Antequera, T.; Palacios, R. (2003). Mathematical
morphology on MRI for the determination of Iberian ham fat content. En: Sanfeliu A.,
Ruiz-Shulcloper J. (eds). Progress in Pattern Recognition, Speech and Image Analysis.
CIARP 2003. Lecture Notes in Computer Science, vol. 2905. Springer, Berlin,
Heidelberg.
Carr, H. Y. & Purcell, E. M. (1954). Effects of Diffusion on Free Precession in Nuclear Magnetic
Resonance Experiments. Phys. Rev., 94, 630-638.
Castejón, D.; Villa, P.; Calvo, M. M.; Santa‐María, G.; Herraiz, M.; Herrera, A. (2010). 1H‐HRMAS
NMR study of smoked Atlantic salmon (Salmo salar). Magn. Reson. Chem., 48, 693-703.
Castejón, D.; García-Segura, J. M.; Escudero, R.; Herrera, A.; Cambero, M. I. (2015).
Metabolomics of meat exudate: Its potential to evaluate beef meat conservation and
aging. Anal. Chem. Acta, 901, 1-11.
314
VII. Bibliografía
Castejón, D.; García-Segura, J.M.; Herrera, A.; Cambero, M.I. (2016). Effect of e-beam
treatment on the methylamine compounds in Atlantic salmon. Food Control, 60, 455-
460.
Capitani, D. ;Mannina, L.; Proietti, N.; Sobolev, A.; Tomassini, A.; Miccheli, A.; Di Cocco, M.;
Capuani, G.; De Salvador, R.; Delfini, M. (2010). Monitoring of metabolic profiling and
water status of Hayward kiwifruits by nuclear magnetic resonance. Talanta, 82, 1826-
1838.
Cernadas, E.; Carrio, P.; Rodríguez, P. G.; Muriel, E.; Antequera, T. (2005). Analyzing magnetic
resonance images of Iberian pork loin to predict its sensorial characteristics. Comput.
Vis. Image Underst., 98, 345-361.
Charlton, J. A. (2009). High resolution NMR analysis of complex mixtures. En: M. Gudjónsdóttir,
P. S. Belton, & G. A. Webb (Eds.), Magnetic resonance in food science: challenges in a
changing world (pp. 3-12). Cambridge: Royal Society of Chemistry.
Chang, S. F., Huang, T. C., Pearson, A. M. (1996). Control of the dehydration process in
production of intermediate-moisture meat products: a review. En: Advances in food and
mutrition research, Taylor, S.L. (Ed), vol. 39, pp 71-161. Academic Press, Inc., San Diego,
California, EE.UU.
Cifuentes, A. (2013). Foodomics: Advanced mass spectrometry in modern food science and
nutrition, Vol. 52, John Wiley and Sons (Ed), Nueva York-USA.
Codex Alimentario (2005). CAC/RCP 58/2005: Código de Prácticas de Higiene para la Carne.
Disponible en: www.codexalimentarius.org/normas-oficiales/lista-de-las-normas.
Documento consultado: Marzo 2017.
Córdoba, J. J.; Antequera, T.; García, C.; Ventanas, J.; López-Bote, C.; Asensio, M. A. (1994).
Evolution of free amino acids and amines during ripening of Iberian cured ham. J. Agri.
Food Chem., 42, 2296-2301.
Cruz-Díaz, K., Fernández-Valle, M.E., García-García, A.B., Castejón, D.; Escudero, R.; Romero de
Ávila, M.D., Cambero, M.I. (2014). Relaxometry and MRI study of Manchego cheese
during ripening and its relationship with rheological behavior. XII International
Conference on the Applications of Magnetic Resonance in Food Science: Defining Food by
Magnetic Resonance, Cesena, Italia.
Cruz, K.P. (2017). Aplicación de técnicas de RMN al análisis de quesos de oveja y películas
comestibles. Evaluación de su potencial para la supervisión y predicción de
características reológicas y estructurales. TESIS DOCTORAL
Damez, J. L. & Clerjon, S. (2008). Meat quality assessment using biophysical methods related to
meat structure. Meat Sci, 80, 132-149.
DeMasi, T. W.; Wardlaw, F. B.; Dick, R. L; Acton, J. C. (1990). Non-protein nitrogen (NPN) and
free aminoacids contents of dry, fermented and non-fermented sausages. Meat Sci., 27,
1-12.
Díaz, O.; Fernández, M.; García de Fernando, G. D.; De la Hoz, L.; Ordóñez, J. A. (1993). Effect of
the addition of Pronase E on the proteolysis in dry fermented sausages. Meat Sci., 34,
205-216.
315
VII. Bibliografía
Díaz, O.; Fernández, M.; García de Fernando, D. G.; de la Hoz, L.; Ordóñez, J. A. (1997).
Proteolysis in dry fermented sausages: the effect of selected exogenous proteases. Meat
Sci., 46, 115-128.
Dolata, W.; Piotrowska, E.; Waidzik, J.; Tritt-Goc, J. (2004). The use of the MRI technique in the
evaluation of water distribution in tumbled porcine muscle. Meat Sci., 67, 25-31.
Emsley, J. & Feeney, J. (2007). Forty years of progress in nuclear magnetic resonance
spectroscopy. Prog. Nucl. Magn. Reso. Spectrosc., 50, 179-198.
Estévez, M.; Morcuende, D.; Ventanas, S. (2009). Determination of oxidation. En: Handbook of
processed meats and poultry analysis, Press, Taylor y Francis Group (Ed), pp. 141-162.
Boca Raton, Florida.
Fantazzini, P.; Bortolotti, V.; Garavaglia, C.; Gombia, M.; Riccardi, S.; Schembri, P. (2005).
Magnetic resonance imaging and relaxation analysis to predict noninvasively and non-
destructively salt-to-moisture ratios in dry-cured ham. Magn. Reson. Imaging, 23, 359-
361.
Farouk, M.M.; Zhang, S.X. Waller, J. (2005). Meat spaghetti tensile strength and extensibility as
indicators of the manufacturing quality of thawed beef. J. Food Quality, 28, 452-466.
Farrán, A.; Zamora, R.; Cervera, P. (2003). Tablas de composición de alimentos del CESNID. Ed.
Mc- Graw-Hill / Interamericana y Edicions Universitat de Barcelona.
Fernández, M.; Ordóñez, J. A.; Bruna, J. M.; Herrans, B.; de la Hoz, L. (2000). Accelerated
ripening of dry fermented sausages. Food Sci. Technol., 11, 201-209.
Fernández, M.; Ordóñez, J.A.; Cambero, I.; Santos, C.; Pin, C.; Hoz, L. (2007). Fatty acid
composition of selected varieties of Spanish dry ham related to their nutritional
implications. Food Chem., 101, 107-112.
Fiehn, O. (2002). Metabolomics the link between genotypes and phenotypes. Plant Molec.
Biol., 48, 155-171.
Flores, M.; Aristoy, M. C.; Toldrá, F. (1996). HPLC purification and characterization of soluble
alanyl aminopeptidase from porcine skeletal muscle. J. Agric. Food Chem., 44, 2578-
2583.
Flores, M.; Aristoy, M. C.; Antequera, T.; Barat, J. M.; Toldrá, F. (2009). Effect of prefreezing
ham on endogenous enzyme activity during the processing of Iberian dry-cured hams.
Meat Sci., 82, 241-246.
Froehlich, D. A.; Gullett, E. A.; Usborne, W. R. (1983). Effect of nitrite and salt on the color,
flavor and overall acceptability of ham. J. Food Sci., 48, 152-154.
316
VII. Bibliografía
Gálvez, M. M.; Farías A. M.; Asahi, K. T.; & Bravo C. E. (2005). Cálculo de tiempos T1 y T2 in
vitro. Revista chilena de radiología, 11, 109-115
Gandemer, G. (2002). Lipids in muscle and adipose tissues, changes during processing and
sensory properties of meat products. Meat Sci., 62, 309-321.
García-Cañas, V.; Simó, C.; Herrero, M.; Ibáñez, E.; Cifuentes, A. (2012). Present and future
challenges in food analysis: Foodomics. Anal. Chem., 84, 10150-10159.
García de Fernando, D. G.; Fox, P. F. (1991). Study of proteolysis during the ripening of a dry
fermented pork sausage. Meat Sci., 30, 367-383.
García-García, A. B.; Larsen, L. B.; Cambero, M. I.; Cruz-Díaz, K.; Bertram, H. C. (2015).
Proteolysis process in fermented sausage model systems as studied by NMR
relaxometry. J. Agric. Food Chem., 63, 3039-3045.
García- Márquez, I., Cambero, M. I., Ordóñez, J. A., Cabeza, M. C. (2012). Shelf life extension
and sanitation of fresh pork loin by E-beam treatment. J Food Protec.,75, 2179-2189.
Godber, J. S. (1994). Nutritional Value of Muscle Foods. En: Muscle Foods. Meat Poultry and
Seafood Technology. D.M. Kinsman, A.W. Kotula y B.C. Breidenstein (Eds). Chapman y
Hall, New York.
Gordon, A.; Barbut, S. (1992). Mechanisms of meat batter stabilization: a review. Crit. Rev.
Food Sci. Nutr., 32, 299-332.
Gou, P. & Comaposada, J. (2000). La transferencia de sal en el interior del jamón curado.
Eurocarne, 86, 35-64.
Guiheneuf, T. M.; Tessier, J. J.; Herrod, N. J.; Hall, L. D. (1996). Magnetic resonance imaging of
meat products: Automated quantitation of the NMR relaxation parameters of cured
pork, by both bulk NMR and MRI methods. J. Sci. Food Agri., 71, 163-173.
Hanson, S. W. F.; Olley J. (1963). Application of the Bligh and Dyer method of lipid extraction to
tissue homogenates. Biochem J., 89,101-102.
Herranz, B.; de la Hoz, L.; Hierro, E.; Fernández, M.; Ordóñez, J. A. (2005). Improvement of the
sensory properties of dry‐fermented sausages by the addition of free aminoacids. Food
Chem., 91, 673-682.
Herrero, A. M.; Cambero, M. I.; Ordonez, J. A.; Castejon, D.; Romero de Avila, M. D.; de la Hoz,
L. (2007). Magnetic resonance imaging, rheological properties, and physicochemical
characteristics of meat systems with fibrinogen and thrombin. J Agri Food Chem, 55:
9357–9364.
Herrero, A. M.; de la Hoz, L.; Ordóñez, J. A.; Herranz, B.; Romero de Ávila, M. D.; Cambero, M. I.
(2008). Tensile properties of cooked meat sausages and their correlation with texture
profile analysis (TPA) parameters and physico-chemical characteristics. Meat Sci., 80,
690-696.
Herrero, A.M.; de la Hoz, L.; Ordóñez, J.A.; Castejón, D.; Romero de Ávila, M.D.; Cambero, M.I.
(2009). Magnetic resonance imaging studies of the cold-set gelation on plasma powder
and meat systems. Food Res. Int., 42, 1362-1372
317
VII. Bibliografía
Honikel, K. (2008). The use and control of nitrate and nitrite for the processing of meat
products. Meat Sci., 78, 68-76.
Incze, K. (1992). Raw fermented and dried meat products. Fleischwirtschaft, 72, 58-72.
Joo, S. T. & Kim, G. D. (2011). Meat quality traits and control technologies. S.T. Joo (Ed.),
Control of meat quality, Research Signpost, 6-10, Kerala, India.
Jurado, A.; García, C.; Timón, M.; Carrapiso, A. (2007). Effect of ripening time and rearing
system on amino-acid related flavor compounds of Iberian ham. Meat Sci., 75, 585-594.
Kim, G. P.; Jung, E. Y.; Lim, H. J.; Yang, H. S.; Joo, S. T.; Jeong, J. Y. (2013). Influence of meat
exudates on the quality characteristics of fresh and freeze-thawed pork. Meat Sci., 95,
323-329.
Koohmaraie, M. (1994). Muscle proteinases and meat aging. Meat Sci., 36, 93-104.
Kuo, M. I.; Gunasekaran, S.; Johnson, M.; Chen, C. (2001). Nuclear magnetic resonance study of
water mobility in pasta filata and non-pastafilata Mozzarella. J. Dairy Sci., 84, 1950-1958.
Laghi, L.; Picone, G.; Capozzi, F. (2014). Nuclear magnetic resonance for foodomics beyond.
Food analysis. Trends Anal. Chem., 59, 93-102.
Larson, G.; Dobney, K.; Albarella, U.; Fang, M.; Matisoo-Smith, E.; Robins, J.; Lowden, S.;
Finlayson, H.; Brand, T.; Willerslev, E.; Rowley-Conwy, P.; Andersson, L. y Cooper, A.
(2005). Worldwide phylogeography of wild boar reveals multiple centers of pig
domestication. Science. 307, 1618-1621.
Li, Y.; Jia, W.; Zhang, C. H.; Li, X.; Wang, L. C.; Zhang, D. Q.; Mu, G. F. (2014). Fluctuated low
temperature combined with high-humidity thawing to reduce physicochemical quality
deterioration of beef. Food Bioprocess Tech., 7, 3370-3380.
Liang, Z. & Lauterbur, P. C. (2000). Principles of magnetic resonance imaging. SPIE Optical
Engineering, New York.
Luo, J.; Pan, T.; Gou, H. Y.; Ren, F. Z. (2013). Effect of calcium in brine on salt diffusion and
water distribution of Mozzarella cheese during brining. J. Dairy Sci., 96, 824-831.
Mannina, L.; D'Imperio, M.; Capitani, D.; Rezzi, S.; Guillou, C.; Mavromoustakos, T.; Vilchez, M.
D.; Fernández, A. H.; Thomas, F.; Aparicio, R. (2009). 1H NMR based protocol for the
318
VII. Bibliografía
detection of adulterations of refined olive oil with refined hazelnut oil. J. Agri. Food
Chem., 57, 11550-11556.
Mansfield, P. & Grannell, P. K. (1973). NMR diffraction in solids? Journal of Physics C: solid
state physics, 6, L422.
Marusic, N.; Aristoy, M. C.; Toldrá, F. (2013). Nutritional meat compounds as affected by ham
dry-curing. Meat Sci., 93, 53-60.
Martín, L.; Córdoba, J. J.; Ventanas, J.; Antequera, T. (1999). Changes in intramuscular lipids
during ripening of Iberian dry-cured ham. Meat Sci., 51, 129-134.
McDonnell, C. K.; Allen, P.; Duggan, E., Arimi, J. M.; Casey, E.; Duane, G.; Lyng, J. (2013). The
effect of salt and fibre direction on water dynamics, distribution and mobility in pork
muscle: A low field NMR study. Meat Sci., 95, 51-58.
Meiboom, S. & Gill, D. (1958). Modified Spin-Echo Method for Measuring Nuclear Relaxation
Times. Rev. Sci. Instrum., 29, 688-691.
Melody, J. L.; Lonergan, S. M.; Rowe, L. J.; Huiatt, T. W.; Mayes, M. S.; Huff-Lonergan, E. (2004).
Early postmortem biochemical factors influence tenderness and water-holding capacity
of three porcine muscles. J. Anim. Sci., 82, 1195-1205.
Mitchell, A. D.; Scholz, A. M.; Wang, P. C.; Song, H. (2001). Body composition analysis of the pig
Magnetic Resonance Imaging. J. Anim. Sci., 79, 1800-1813.
Møller, S.; Gunvig, A.; Bertram, H. C. (2010). Effect of starter culture and fermentation
temperature on water mobility and distribution in fermented sausages and correlation
to microbial safety studied by nuclear magnetic resonance relaxometry. Meat Sci., 86,
462-467.
Molly, K.; Demeyer, D.; Johansson, G.; Raemaekers, M.; Ghistelinck, M.; Geenen, I. (1997). The
importance of meat enzymes in ripening and flavor generation in dry fermented
sausages. First results of a European project. Food Chem., 59, 539-545.
Moreiras, O.; Carbajal, A.; Cabrera, L.; Cuadrado, C. (2013). Tablas de composición de
alimentos. 16 ed. Ed. Pirámide. ISBN: 9788436829037.
319
VII. Bibliografía
Morrissey, P. A. Sheehy, P. J. A.; Galvin, K.; Kerry, J. P.; Buckley, D. J. (1998). Lipid stability in
meat and meat products. Meat Sci., 49, 73-86.
Muguerza, E.; Ansorena, D.; Astiasaran, I. (2003). Improvement of nutritional properties of cho
rizo de Pamplona by replacement of pork backfat with soy oil. Meat Sci, 65, 1361-1367.
Mulas, G.; Roggio, T., Uzzau, S.; Anedda, R. (2013). A new magnetic resonance imaging
approach for discriminating Sardinian sheep milk cheese made from heat-treated or raw
milk. J. Dairy Sci., 96, 7393-7403.
Müller, V. In Bacterial Fermentation; John Wiley&Sons: Chichester. UK, 2008. DOI: 10.1002/
/9780470015902.a0001415.pub2.
Motilva, M. J.; Toldrá, F.; Nieto, P.; Flores, J. (1993). Muscle lipolysis phenomena in the
processing of dry-cured ham. Food Chem., 48, 121-125.
Monin, G.; Marinova, P.; Talmant, A.; Martin, J. F.; Cornet, M.; Lanore, D.; Grasso, F. (1997).
Chemical and structural changes in dry-cured hams (bayonne hams) during processing
and effects of the dehairing technique. Meat Sci., 47, 29-47.
Nawar, W.W. (1996). Lipids. En: Food Chemistry, O.R. Fennema (Ed), pp. 225-319. New York:
Marcel Dekker, Inc.
Olmos, J. V. (2006). Atlas Monte Nevado de anatomía del jamón serrano. Edita Jamones
Segovia, S.A.
Ordóñez, J. A.; Cambero M. I.; Fernández, L.; García M. L.; García de Fernando, G.; de la Hoz, L.;
Selgas, M. D. (1998). Cambios post mortem del músculo. En: Tecnología de los
Alimentos. Alimentos de origen animal. 170-184. Síntesis (Ed), Madrid.
Ordóñez J. A.; Hierro, E. M.; Bruna, J. M.; de la Hoz, L. (1999). Changes in the components of
dry-fermented sausages during ripening. Crit. Rev. Food Sci. Nutr., 39, 329-367.
Ordóñez J.A. & de la Hoz, L. (2001). Embutidos crudos curados. Tipos. Fenómenos
madurativos. Alteraciones. En: Enciclopedia de la carne y los productos cárnicos, vol. II,
Martin & Macías (Ed), pp. 1063-1090, Cáceres.
Pearce, K. L.; Rosenvold, K.; Andersen, K. J.; Andersen, H. J; Hopkins, D. L. (2011). Water
distribution and mobility in meat during the conversion of muscle to meat and ageing
and the impacts on fresh meat quality attributes: A review. Meat Sci., 89, 111-124.
Pichner, R.; Hechelmann, H.; Gareis, M.; Steinruck, H. (2006). Shigatoxin producing Escherichia
coli (STEC) in conventionally and organically produced salami products. Fleischwirtschaft,
86, 112-114.
Purcell, E. M.; Torrey, H.; Pound, R. V. (1946). Resonance absorption by nuclear magnetic
moments in a solid. Phys. Rev., 69, 37-38.
Prändl, O.; Fischer, A.; Schmidhofer, T.; Sinell, H. J. (1994). Tecnología e higiene de la carne.
Acribia (Ed), Zaragoza.
320
VII. Bibliografía
Piñeiro, M. E.; Punín, M. O.; Lage, M. A. (2001). Determinación de ácidos grasos en carne de
cerdo. En: Alimentaria: Revista de tecnología e higiene de los alimentos. ISSN: 0300-
5755, Vol. 327, 41-45.
Pospiech, E.; Grzes, B.; Lyczynski, A.; Borzuta, K.; Szalata, M.; Mikolajczak, B. (2003). Muscle
proteins and their changes in the process of meat tenderization. Anim. Sci. Pap. Rep., 21,
133-155.
Prieto, B. & Carballo, J. (1997). El control analítico de la calidad en los productos cárnicos
crudos-curados. Ciencia y Tecnología de Alimentos, 1, 112-120.
Price, J. F. & Schweigert, B. S. (1994). Ciencia de la carne y de los productos cárnicos. Acribia
(Ed), Zaragoza.
Ramírez, C.; Germain, J. C.; Aguilera, J. M. (2009). Image analysis of representative food
structures: application of the bootstrap method. J. Food Sci., 74, 65-72.
Romero de Ávila, M. D; Escudero, R.; Ordóñez, J. A.; Cambero, M. I. (2014). Weibull analysis
characterizes the breaking properties of dry-cured ham slices. Meat Sci., 97, 451-458.
Ruan, R. R. y Chen, P. L. (1998). Water in Foods and Biological Materials: A Nuclear Magnetic
Resonance Approach. 1st edition ed. Technomic Publishing Company, Inc, Lancaster,
Pennsylvania.
Savage, A. W.; Warriss, P. D.; Jelley, P. D. (1990). The amount and composition of the proteins
in drip from stored pig meat. Meat Sci., 27, 289-303.
Scheffler, T. L. & Gerrard, D. E. (2007). Mechanisms controlling pork quality development: The
biochemistry controlling postmortem energy metabolism. Meat Sci., 77, 7-16.
Schievano, E.; Pasini, G.; Cozzi, G.; Mammi, S. (2008). Identification of the production chain of
Asiago d’Allevo cheese by nuclear magnetic resonance spectroscopy and principal
component analysis. J. Agric. Food Chem., 56, 7208-7214.
Serra, X.; Ruiz-Ramirez, J., Arnau, J.; Gou, P. (2005). Texture parameters of dry-cured ham,
muscle biceps femoris samples, dried at different levels as a function of water activity
and water content. Meat Sci., 69, 249-54.
Skiba, G.; Raj, S.; Polawska, E.; Pastuszewska, B.; Elminowska-Wenda, G.; Bogucka, J.; Knecht,
D. (2012). Profile of fatty acids, muscle structure and shear force of
musculus longissimus dorsi (MLD) in growing pigs as affected by energy and protein or
protein restriction followed by realimentation. Meat Sci.e, 91, 339-346.
Skibsted, L. H. (1992). Cured meat products and their oxidative stability. En: The chemistry of
muscle based foods, D. A. Ledward, D. E. Johnston y M. K. Knight (Eds), pp. 266-286.
Londres: Royal Society of Chemistry.
Skibsted, L. H. (2011). Nitric oxide and quality and safety of muscle based foods. Nitric Oxide –
Biol. Chem., 24, 176-183.
321
VII. Bibliografía
Suárez, M.V.; Barba, C.; Forero, J.; Sereno, J. R. B.; Diéguez, E.; Delgado, J. V. (2002).
Reproductive characterization of various pig breed from Iberian origin. I. Descriptive
analysis. Archivos de Zootecnia, 51, 245-248.
Tautenhahn, R.; Cho, K.; Uritboonthai, W.; Zhu, Z.; Patti, G. J.; Siuzdak, G. (2012). An
accelerated workflow for untargeted metabolomics using the METLIN database. Nat.
Biotechnol., 30, 826-828.
Toldrá, F.; Cerveró, C.; Part, C. (1993). Porcine aminopeptidase activity as affected by curing
agents. J. Food Sci., 58, 724-726.
Toldrá, F.; Flores, M.; Aristoy, M. C.; Virgili, R.; Parolari, G. (1996). Pattern of muscle proteolytic
and lipolytic enzymes from light and heavy pigs. J. Sci. Food Agri., 71, 124-128.
Toldrá, F. (1998). Proteolysis and lipolysis in flavour development of drycured meat products.
Meat Sci., 49, 101-110.
Toldrá, F. (2006). Dry-cured ham. En: Handbook of Food Science Technology and Engineering.
CRC Press (Ed), pp 164.1-164.11, Boca Raton, E.E.U.U.
Toldrá, F. & Aristoy, M.C. (2010). Dry-cured ham. En: Handbook of Meat Processing, F. Toldrá
(Ed), pp.351-362. Ames, Iowa.
Totosaus, A., Montejano, G. J., Salazar, J. A., Guerrero, I. (2002). A review of physical and
chemical protein-gel induction. Int. J. Food Sci. Technol., 37, 589-601.
Ventanas, J. (2001). Factores del proceso que determinan la calidad del jamón. En: Jamón
Ibérico y Serrano. Fundamentos de la elaboración y de la calidad. pp. 63-103, Paraninfo
(Ed), Madrid.
Wakamatsu, J., Hayashi, N., Nishimura, T., y Hattori, A. (2010). Nitric oxide inhibits the
formation of zinc protoporphyrin IX and protoporphyrin IX. Meat Science, 84, 125– 128.
Windham, C. T.; Wyse, B. W.; Hansen, R.G. (1990). Thiamin, riboflavin, niacin and pantothenic
acid. En: Meat and Health, Advanced Materials Research. Volume: 6. M. Pearson y T.R.
Dutson (Ed), págs. 401-459. Elsevier Applied Science, London.
Wüthrich, K. (1974). Nuclear magnetic resonance in protein research. Experientia, 30, 577-585.
Xia, X.; Kong, B.; Liu, Q.; Liu, J. (2009). Physicochemical change and protein oxidation in
porcine longissimus dorsi as influenced by different freeze–thaw cycles. Meat Sci., 83,
239-245.
Zhang, G.; Hailemariam, D.; Dervishi, E.; Deng, Q.; Goldansaz, S. A.; Dunn, S. M.; Ametaj, B. N.
(2015). Alterations of innate immunity reactants in transition dairy cows before clinical
signs of lameness. Animals, 5, 717-747.
322
VIII. Índice de abreviaturas
323
VIII. Índice de abreviaturas
ABREVIATURAS Y ACRÓNIMOS
325
VIII. Índice de abreviaturas
326
VIII. Índice de abreviaturas
327
II. Objetivos
328