2019 Tesis Garlito Molina Borja

Programa de Doctorado en Ciencias
Escuela de Doctorado de la Universitat Jaume I
ANÁLISIS DE COMPUESTOS ENDÓGENOS Y EXÓGENOS EN

PRODUCTOS DE ACUICULTURA MEDIANTE TÉCNICAS
ANALÍTICAS AVANZADAS
Memoria presentada por Borja Garlito Molina para optar al grado de

doctor por la Universitat Jaume I
Borja Garlito Molina Roque Serrano Gallego Tania Portolés Nicolau
Castelló de la Plana, octubre de 2019

Financiación recibida
Agencias financiadoras del doctorando
 Universitat Jaume I (I-4663)

 SCORE-COST Action ES1307. Short Term Scientific Mission
(STSM) grant (COST-STSM-ES1307-38094)
Agencias financiadoras del proyecto de investigación o de los recursos

materiales específicos del grupo de investigación.
 EU Seventh Framework Programme

 Norwegian Research Council
 Generalitat Valenciana
 Ministerio de economia i competitividad
Tesis por compendio de las siguientes publicaciones:
I) T. Portolés, M. Ibáñez, B. Garlito, J. Nácher-Mestre, V. Karalazos, J. Silva,

Alm, R. Serrano, J. Pérez-Sánchez, F. Hernández, M.H.G. Berntssen.
Comprehensive strategy for pesticide residue analysis through the production
cycle of gilthead sea bream and Atlantic salmon. Chemosphere 179 (2017)
242-253. http://dx.doi.org/10.1016/j.chemosphere.2017.03.099. Factor de
impacto: 4.427
II) T. Portolés, B. Garlito, Jaime Nácher-Mestre, M.H.G. Berntssen, J. Pérez-

Sánchez. Multi-class determination of undesirables in aquaculture samples by
gas chromatography/tandem mass spectrometry with atmospheric pressure
chemical ionization: A novel approach for polycyclic aromatic hydrocarbons.
Talanta 172 (2017) 109–119. http://dx.doi.org/10.1016/j.talanta.2017.05.025.
Factor de impacto: 4.244
III) B. Garlito, M. Ibáñez, T. Portolés, R. Serrano, H. Amlund, A-K.

Lundebye, M. Sanden, M.H.G. Berntssen, F. Hernández. LC-MS/MS method
for the determination of organophosphorus pesticides and their metabolites in
salmon and zebrafish fed with plant-based feed ingredients. Analytical and
Bioanalytical Chemistry (2019). Factor de impacto: 3.286
IV) B. Garlito, T. Portolés, W.M.A. Niessen, J.C. Navarro, F. Hontoria, Ó.

Monroig, I. Varó, R. Serrano. Identification of very long-chain (>C24) fatty
acid methyl esters using gas chromatography coupled to quadrupole/time-of-
flight mass spectrometry with atmospheric pressure chemical ionization
source. Analytica Chimica Acta 1051 (2019) 103-109.
https://doi.org/10.1016/j.aca.2018.11.001. Factor de impacto: 5.256
“Esta tesis dispone de la aceptación de los coautores de las publicaciones
que el doctorando presenta como tesis y su renuncia expresa a presentarlas
como parte de otra tesis doctoral”
AGRAÏMENTS
Després de quatre anys, ha arribat el moment de posar punt i final a una etapa.
Quatre anys des que iniciàrem, inexperts, un camí ple d’alts i baixos, el de
l’assaig-error i la perseverança. A l’escriure aquestes línies, m’ha sobtat la
facilitat que tenim per a sintetitzar el treball d’un grapat d’anys - segurament
per l’experiència, coneixement i habilitats adquirides - però la dificultat que
suposa condensar en pocs paràgrafs la gratitud cap a totes aquelles persones
que han fet el camí més planer. A totes i cadascuna de vosaltres, vaja per
davant, moltíssimes gràcies!
Aquesta Tesi Doctoral ha estat realitzada en l’Institut Universitari de

Plaguicides i Aigües de la Universitat Jaume I, sota la direcció de la Dra. Tania
Portolés Nicolau i el Dr. Roque Serrano Gallego. Als meus directors, a la Dra.
Elena Pitarch i al Dr. Félix Hernández, m’agradaria agrair-los, en primer lloc,
l’oportunitat de formar part del IUPA i d’introduir-me en el món de la
investigació. A Roque, també, pel seu optimisme, serenor i consell quan se’ns
complicaven els articles. I a Tania, a qui estaré eternament agraït, referent i
punt de suport des del primer dia, per tot el que m’ha ensenyat, ajudat i per la
seua paciència infinita.
També voldria agrair al Dr. Joaquín Beltrán, qui fa sis anys va supervisar el
meu treball de final de grau, per posar-me, per primera vegada, en contacte
amb un espectròmetre de masses. Allà va començar tot. I al Dr. Juan V.
Sancho, a qui sempre és un goig escoltar, per l’ajuda rebuda quan els
instruments semblaven estar en contra de la font APCI.
I would like to thank Dr. Hans Mol and Marc Tienstra for allowing me to do
my research stay at RIKILT-Wageningen University and Research and for our
stimulating discussions in front of the QOrbitrap. I would also like to thank
Dr. Marc H.G. Berntssen for his advice in the field of toxicology that,
undoubtedly, has been fundamental in this work.
Amb molta estima, voldria agrair a la resta de professors i a tots els companys
i companyes, que estan o han estat al IUPA, la seua ajuda desinteressada
durant estos últims quatre anys. Feu fàcil allò que pot semblar difícil. Moltes
gràcies a Edu, Rubén i Clara per les partidetes de pàdel i les cervesetes de
després; a David pels debats enriquidors; a Nácher per ensenyar-me que
l’anàlisi de productes d’aqüicultura és cosa d’acostumar la pituïtària; a Carlos
per compartir la desesperació de l’APCI; a María pel comboi d’eixos dinars
tan necessaris a Vilavella; a Mercedes pel seu carisma i la seua energia
infinita; a Jose, perquè si no parlàrem del Castelló al passadís potser viuríem
uns anys menys; i a Montse, Alberto, Leticia, Jorge, Eddie, Robert...pels cafés,
cerveses i tertúlies que hem compartit.
Tampoc m’oblide dels amics i amigues amb qui he crescut i viscut, que sempre
han estat al meu costat recolzant-me cegament tot i no entendre ben bé allò
que feia. Especialmente no me olvido de ti, Alejandro. Tu ausencia es
irremplazable. Com no, a tota la gent que conforma la Conlloga, per ensenyar-
me que el treball cooperatiu i la companyonia inherent a les muixerangues ens
pot fer anar més lluny en qualsevol àmbit de la nostra vida. Milions de gràcies!
Y, por último, a mi familia. A mis padres, a Marcos y a mis abuelos por ser
mis principales referentes, por estar al pié del cañón, confiando y facilitando
mucho las cosas, siempre tendiéndome la mano. I a tu, companya i amiga
Isabel, per recolzar-me, escoltar-me, aconsellar-me i acompanyar-me en els
camins que se’ns presenten. Gracias a todos vosotros!
Resumen
RESUMEN
Ante el estancamiento de la producción de la pesca de captura desde finales

de la década de 1980, y la creciente demanda de pescado y productos
pesqueros, la industria acuícola se enfrenta a un importante reto de
sostenibilidad dada su dependencia económica y ecológica con respecto a
harinas y aceites de origen marino. Por ello, en la última década se han
introducido ingredientes vegetales en la composición de los piensos,
planteando un nuevo escenario en el ámbito de la seguridad alimentaria y la
salud animal.
En la presente Tesis Doctoral se ha estudiado el potencial analítico y la

aplicabilidad de diferentes acoplamientos cromatografía-espectrometría de
masas en el campo del análisis de contaminantes orgánicos y compuestos
endógenos en muestras de acuicultura. La variedad y el potencial de los
acoplamientos instrumentales han permitido optimizar el alcance de los
análisis y desarrollar metodología analítica sensible, selectiva y fiable, desde
un enfoque cualitativo y cuantitativo.
El trabajo presentado se ha estructurado en dos bloques: un primer bloque

donde se aborda el análisis de contaminantes orgánicos, y un segundo
dedicado al estudio de compuestos endógenos.
El primer bloque está formado por tres trabajos consecutivos y relacionados

entre ellos que abordan, desde diferentes perspectivas, las problemáticas en el
ámbito de la seguridad alimentaria y la salud animal surgidas en campo de la
acuicultura. En este bloque se desarrollan métodos analíticos para la
determinación de pesticidas, hidrocarburos policíclicos aromáticos (PAHs) y
bifenilos policlorados (PCBs) en ingredientes, piensos y pescado de
I
Resumen
acuicultura. En el primero de los trabajos presentados se estudia el potencial

de la cromatografía de líquidos acoplada a espectrometría de masas híbrida
cuadrupolo-tiempo de vuelo (LC-QTOF MS) y la cromatografía de gases
acoplada a QTOF MS con fuente de ionización APCI (GC-APCI-QTOF MS)
para el desarrollo de una estrategia de screening integral de pesticidas. A una
primera etapa de validación cualitativa para 253 pesticidas a 10 y 50 ng·g-1, le
sigue la aplicación del método para el análisis de muestras mediante las
aproximaciones target y suspect screening.
En el segundo trabajo, surgido a partir de los resultados obtenidos en el trabajo

anterior, se desarrolla un método basado en la cromatografía de gases acoplada
a espectrometría de masas en tándem con fuente de ionización APCI (GC-
APCI-MS/MS) para la determinación de PAHs, PCBs y los pesticidas
detectados en el screening anterior. El objetivo principal de este trabajo es
estudiar el potencial de la fuente APCI junto con el analizador de triple
cuadrupolo (QqQ) con el fin de explorar la posibilidad de reducir el efecto
matriz mediante dilución simple del extracto y, en consecuencia, cuantificar
usando curvas de calibración en solvente relativas a un patrón interno. El
método final se valida satisfactoriamente a 2, 5 y 50 ng·g-1 en diferentes
matrices para la mayoría de compuestos mediante la cuantificación por
calibrado en solvente.
Para finalizar, en el tercer trabajo de este bloque se desarrolla un método

analítico basado en la cromatografía de líquidos acoplada a espectrometría de
masas en tándem (LC-MS/MS) para la determinación de pirimiphos-methyl,
chlopyrifos-methyl y sus principales metabolitos en pez cebra y salmón
sometidos a la exposición dietaria de estos pesticidas. Este trabajo surge de los
resultados obtenidos en los dos trabajos anteriores y pretende dar respuesta,
II
Resumen
desde una aplicación analítica en el ámbito de la toxicología, a la ausencia de

límites máximos de residuo (MRLs) de los analitos mencionados en pescado.
Los resultados obtenidos permitirán generar un modelo cinético de
acumulación y eliminación de pirimiphos-methyl en el salmón que contribuirá
a establecer MRLs seguros y objetivos en pescado.
El segundo bloque aborda el estudio de ácidos grasos poliinsaturados de

cadena muy larga (VLC-PUFA) en muestras de retina de pescado. En el cuarto
trabajo presentado en la Tesis Doctoral se estudia el potencial de la fuente
APCI en el acoplamiento GC-APCI-QTOF MS, como alternativa a la
ionización electrónica, para la identificación de las series ω3 y ω6 de estos
compuestos. Para este fin, se genera un modelo estadístico basado en el
análisis ortogonal de mínimos cuadrados parciales discriminantes (OPLS-DA)
que destaca la abundancia relativa entre dos iones como la variable
diferenciadora de ambos grupos. La aplicación del modelo al análisis de
muestras reales, junto con la presencia de la molécula protonada en el espectro
de masas, permite diferenciar e identificar satisfactoriamente los isómeros de
las series ω3 y ω6.
III
Summary
SUMMARY
Due to the stagnation of capture fisheries since the late 1980s, and the growing
demand for fish and fishery products, the aquaculture industry faces a major
sustainability challenge given its economic and ecological dependence on
fishmeals and fish oils. Therefore, in the last decade plant ingredients have
been introduced in the composition of feed, posing a new scenario in the field
of food safety and animal health.
In this Doctoral Thesis the analytical potential and applicability of different

chromatography-mass spectrometry couplings in the field of organic
pollutants and endogenous compounds analysis in aquaculture samples have
been studied. The variety and potential of the instrumental couplings has
allowed optimizing the scope of the analysis and developing sensitive,
selective and reliable analytical methodology, in both qualitative and
quantitative approaches.
The work presented has been structured in two blocks: a first block focused
on the analysis of organic pollutants, and a second one dedicated to the study
of endogenous compounds.
The first block is formed by three consecutive and related works that address,
from different perspectives, the problems in the field of food safety and animal
health arising in the field of aquaculture. In this block, analytical methods for
the determination of pesticides, polycyclic aromatic hydrocarbons (PAHs) and
polychlorinated biphenyls (PCBs) in aquaculture ingredients, feed and fish are
developed. In the first work, the potential of liquid chromatography coupled
to hybrid quadrupole-time of flight mass spectrometry (LC-QTOF MS) and
gas chromatography coupled to QTOF MS with APCI ionization source (GC-
V
Summary
APCI-QTOF MS) are studied for the development of a comprehensive

pesticide screening strategy. In a first step, the method is qualitatively
validated for 253 pesticides at 10 and 50 ng·g-1, and then it is applied to the
analysis of samples using target and suspect screening approaches.
In the second work, a method based on gas chromatography coupled to tandem

mass spectrometry with APCI ionization source (GC-APCI-MS/MS) is
developed for the determination of PAHs, PCBs and the pesticides detected in
the previous screening. The main objective of this work is to study the
potential of the APCI source coupled to triple quadrupole mass analyzer
(QqQ) in order to explore the possibility of reducing the matrix effect by
simple dilution of the final extract and, consequently, quantify using
calibration curves with standards in solvent. The final method is successfully
validated at 2, 5 and 50 ng·g-1 in different matrices for most compounds using
calibration with standards in solvent for quantification.
Finally, in the third work of this block, an analytical method based on liquid
chromatography coupled to tandem mass spectrometry (LC-MS/MS) is
developed for the determination of pirimiphos-methyl, chlorpyrifos-methyl
and their main metabolites in dietary exposure experiments in zebrafish and
salmon. This work arises from the results obtained in the two previous works
and aims to respond, from an analytical application in the field of toxicology,
to the absence of maximum residue limits (MRLs) for the mentioned
pesticides in fish. The results obtained will be essential for developing a
kinetic model of accumulation and elimination of pirimiphos-methyl in
salmon, and will contribute to establish relevant MRLs for fish.
The second block addresses the study of very long chain polyunsaturated fatty
VI
Summary
acids (VLC-PUFA) in fish retina. In the fourth paper presented in the Doctoral
Thesis, the potential of the GC-APCI-QTOF MS coupling is studied, as an
alternative to electron ionization, for the reliable identification of the ω3 and
ω6 series of these compounds. To this aim, a statistical model is generated
based on the orthogonal partial least squares discriminant analysis (OPLS-
DA) that highlights the relative abundance between two ions as the
differentiating variable of both groups. The application of the model to the
analysis of real samples, together with the presence of the protonated molecule
in the mass spectrum, make it possible to differentiate and identify the isomers
of the ω3 and ω6 series.
VII
ÍNDICE GENERAL
OBJETIVOS.................................................................................................... 1
OBJECTIVES ................................................................................................. 5
CAPÍTULO 1. INTRODUCCIÓN GENERAL ............................................. 9
1.1. Retos actuales de la acuicultura .......................................................... 13
1.2. Seguridad alimentaria y salud animal ................................................. 16
1.3. Técnicas analíticas para el control alimentario ................................... 18
1.3.1. Técnicas cromatográficas............................................................. 19
1.3.2. Espectrometría de masas .............................................................. 23
1.4. Estrategias analíticas para el control alimentario ............................... 33
1.5. Tratamiento de muestra en matrices de acuicultura. .......................... 36
1.6. Estrategias de corrección del efecto matriz. ....................................... 38
1.7. Referencias ......................................................................................... 43
CAPÍTULO 2. DETERMINACIÓN DE COMPUESTOS EXÓGENOS

MEDIANTE EL USO DE ESPECTROMETRÍA DE MASAS CON
ANALIZADOR DE CUADRUPOLO-TIEMPO DE VUELO Y TRIPLE
CUADRUPOLO............................................................................................ 51
2.1. Introducción ........................................................................................ 55
2.2. Artículo científico I ............................................................................ 61
IX
Comprehensive strategy for pesticide residue analysis through the
production cycle of gilthead sea bream and Atlantic salmon
Chemosphere, 179, 242-253, 2017
2.3. Artículo científico II ......................................................................... 105
Multi-class determination of undesirables in aquaculture samples by gas

chromatography/tandem mass spectrometry with atmospheric pressure
chemical ionization: A novel approach for polycyclic aromatic
hydrocarbons
Talanta, 172, 109–119, 2017
2.4. Discusión de los resultados obtenidos .............................................. 153
2.5. Referencias ....................................................................................... 163
CAPÍTULO 3. ESTUDIO DE LA TRANSFERENCIA DE

CONTAMINANTES ORGÁNICOS A TRAVÉS DE LA DIETA EN
SALMÓN ATLÁNTICO MEDIANTE CROMATOGRAFÍA DE LÍQUIDOS
ACOPLADA A ESPECTROMETRÍA DE MASAS EN TÁNDEM ......... 167
3.1. Introducción...................................................................................... 171
3.2. Artículo científico III........................................................................ 173
LC-MS/MS method for the determination of organophosphorus pesticides

and their metabolites in salmon and zebrafish fed with plant-based feed
ingredients
Analytical and Bioanalytical Chemistry, 2017
X
3.4. Referencias ....................................................................................... 207
CAPÍTULO 4. DETERMINACIÓN DE COMPUESTOS ENDÓGENOS

ANALIZADOR DE CUADRUPOLO-TIEMPO DE VUELO Y FUENTE DE
IONIZACIÓN APCI ................................................................................... 209
4.1. Introducción ...................................................................................... 213
4.2. Artículo científico IV........................................................................ 217
Identification of very long-chain (>C24) fatty acid methyl esters using gas
chromatography coupled to quadrupole/time-of-flight mass spectrometry
with atmospheric pressure chemical ionization source
Analytica Chimica Acta, 1051, 103-109, 2019
4.4. Referencias ....................................................................................... 251
CAPÍTULO 5. CONCLUSIONES/CONCLUSIONS ............................... 255
SUGERENCIAS PARA FUTUROS TRABAJOS ..................................... 261
SUGGESTIONS FOR FUTURE WORKS ................................................. 263
ARTÍCULOS RELACIONADOS/RELATED ARTICLES ....................... 265
XI
ÍNDICE DE ACRÓNIMOS
2-DAMP 2-(diethylamino)-6-methyl-4-pyrimidinol
AOAC Asociación de Comunidades Analíticas
APCI Ionización química a presión atmosférica
API Ionización a presión atmosférica
ARA Ácido araquidónico
CEN Comité Europeo de Normalización
CI Ionización química
CID Disociación inducida por colisión
DC Corriente continua
DHA Ácido docosohexaenoico
d-SPE Extracción en fase sólida dispersiva
EFSA Autoridad Europea de Seguridad Alimentaria
EI Ionización electrónica
ELOVL4 Elongasa de ácidos grasos de cadena muy larga 4
EPA Ácido eicosapentaenoico
ESI Interfase electrospray
FAME Éster metílico de ácido graso
GAP Buenas prácticas agrícolas
GC Cromatografía de gases
GCB Carbón negro grafitizado
GC-APCI- Cromatografía de gases acoplada a espectrometría de
MS/MS masas en tándem con fuente de ionización química a
presión atmosférica
XIII
GC-APCI- Cromatografía de gases acoplada a espectrometría de
QTOF MS masas híbrida cuadrupolo-tiempo de vuelo con fuente
de ionización química a presión atmosférica
GC-MS Cromatografía de gases acoplada a espectrometría de
masas
GC-MS/MS Cromatografía de gases acoplada a espectrometría de
masas en tándem
GPC Cromatografía por permeación en gel
HE Función de alta energía
HPLC Cromatografía de líquidos de alta resolución
HRMS Espectrometría de masas de alta resolución
IP Punto de identificación
IS Patrón interno
LC Cromatografía de líquidos
LC-MS Cromatografía de líquidos acoplada a espectrometría
de masas
LC-MS/MS Cromatografía de líquidos acoplada a espectrometría
de masas en tándem
LC-QTOF MS Cromatografía de líquidos acoplada a espectrometría
de masas híbrida cuadrupolo-tiempo de vuelo
LC-PUFA Ácido graso poliinsaturado de cadena larga
LC-PUFA-ME Éster metílico de ácido graso poliinsaturado de cadena
larga
LE Función de baja energía
m/z Relación masa/carga
M+· Ion molecular
[M+H]+ Molécula protonada
XIV
MAE Extracción asistida por microondas
MRL Límites máximos de residuo
MS Espectrometría de masas
MS/MS Espectrometría de masas en tándem
NCI Ionización química negativa
N-Des-PM O-[2-(ethylamino)-6-methylpyrimidin-4-yl]O,O-
dimethylphosphorothioate
OPLS-DA Análisis ortogonal de mínimos cuadrados parciales
discriminante
PAHs Hidrocarburos policíclicos aromáticos
PBDEs Difeniléteres polibromados
PC Fosfatidilcolina
PCA Análisis de componentes principales
PCBs Bifenilos policlorados
PLE Extracción líquida presurizada
POPs Contaminantes orgánicos persistentes
PRM Monitorización de la reacción paralela
PSA Amina primaria-secundaria
PTV Vaporización con temperatura programada
Q Filtro de masas cuadrupolar
q/Q Ratio iónica
QqQ Analizador de masas de triple cuadrupolo
QTOF Analizador de masas híbrido cuadrupolo-tiempo de
vuelo
QuEChERS Rápido, fácil, barato, efectivo, robusto, seguro
RF Radiofrecuencia
S/N Relación señal/ruido
XV
SDL Límite de detección de screening
SFE Extracción con fluidos supercríticos
SIL-IS Patrón interno marcado con isótopos estables
SIM Monitorización del ion seleccionado
SOFIA Estado Mundial de la Pesca y la Acuicultura
SPE Extracción en fase sólida
SRM Monitorización de la transición seleccionada
TCP 3,5,6-trichloropyridinol
TOF Analizador de masas de tiempo de vuelo
UE Unión Europea
UHPLC Cromatografía de líquidos de ultra alta resolución
USE Extracción mediante ultrasonidos
VLC-PUFA Ácido graso poliinsaturado de cadena muy larga
VLC-PUFA-ME Éster metílico de ácidos graso poliinsaturado de cadena
muy larga
XVI
OBJETIVOS
Objetivos
OBJETIVOS
El objetivo principal de la presente Tesis Doctoral es explorar el potencial

analítico de la cromatografía de gases (GC) y de líquidos de ultra alta
resolución (UHPLC) acopladas a espectrometría de masas (MS), tanto en
modo simple como tándem (MS/MS), usando analizadores de triple
cuadrupolo (QqQ) e híbrido cuadrupolo-tiempo de vuelo (QTOF), para la
determinación de compuestos exógenos y endógenos en el campo de la
acuicultura. Con ello, se pretende avanzar en el conocimiento de los nuevos
retos que este campo está abordando en aras de la rentabilidad y sostenibilidad.
En primer lugar, se diseña una estrategia integral, mediante el uso de GC-

QTOF MS con fuente de ionización química a presión atmosférica (APCI),
UHPLC-QTOF MS con fuente electrospray (ESI) y GC-(APCI)-MS/MS con
QqQ para el análisis de contaminantes orgánicos en muestras procedentes de
la acuicultura. En segundo lugar, se desarrolla un método UHPLC-MS/MS
con QqQ para llevar a cabo un estudio de migración de contaminantes de
piensos a tejidos de pescado. Por último, se aplica el potencial de la técnica
GC-(APCI)-QTOF MS para la elucidación y caracterización de ácidos grasos
de cadena muy larga (VLC-PUFA) cuyos estándares y espectros de ionización
electrónica no están disponibles.
Con el fin de alcanzar el objetivo principal, se establecen los siguientes

objetivos específicos:
1. Desarrollar, validar y aplicar un método analítico basado en GC-(APCI)-

QTOF MS y UHPLC-QTOF MS para el screening de contaminantes
orgánicos, tanto en modo target como suspect, en diversos tipos de muestras
de acuicultura.
3
Objetivos
2. Desarrollar, validar y aplicar un método analítico basado en GC-(APCI)-

MS/MS con QqQ para la cuantificación y confirmación de los contaminantes
orgánicos detectados en el screening de las muestras analizadas anteriormente.
3. Explorar el potencial de la fuente APCI en combinación con GC-MS/MS

con QqQ, como técnica para la determinación de contaminantes orgánicos en
un amplio número de matrices complejas. Estudiar el efecto matriz y el
potencial del método con el fin de utilizar calibrado en solvente. Este objetivo
específico pone especial énfasis en los hidrocarburos policíclicos aromáticos.
4. Desarrollar y validar un método analítico basado en UHPLC-MS/MS con

QqQ para la cuantificación y confirmación de pirimiphos-methyl,
chlorpyriphos-methyl y sus principales metabolitos en piensos y tejidos de pez
cebra y salmón; y generar datos para un futuro estudio de migración de
contaminantes de piensos a tejidos de pescado.
5. Estudiar el potencial del acoplamiento GC-(APCI)-QTOF MS para la

elucidación y caracterización de VLC-PUFA, cuyos espectros de ionización
electrónica no se encuentran en librerías comerciales y cuyos estándares
analíticos no están disponibles comercialmente. Utilización de herramientas
estadísticas para el estudio de espectros de masas de compuestos isoméricos.
4
OBJECTIVES
Objectives
OBJECTIVES
The main objective of this Doctoral Thesis is to explore the analytical

capabilities of GC and UHPLC coupled to MS using triple quadrupole (QqQ)
and hybrid quadrupole-time of flight (QTOF) for the determination of
exogenous and endogenous compounds in the aquaculture field. With this, we
intend to advance in the knowledge of the new challenges that this field is
addressing in the interests of profitability and sustainability.
First, a comprehensive strategy is generated using GC-(APCI)-QTOF MS,

UHPLC-QTOF MS and GC-(APCI)-MS/MS with QqQ for the analysis of
organic contaminants in aquaculture samples. Secondly, a UHPLC-MS/MS
with QqQ method is developed to carry out a migration study of feed
contaminants to fish tissues. Finally, the potential of the GC-(APCI)-QTOF
MS technique is explored for the elucidation and characterization of
compounds whose standards and electron ionization spectra are not available.
Specific objectives of the Thesis are the following:
1. Development, validation and application of an analytical method based on

GC-(APCI)-QTOF MS and UHPLC-QTOF MS for the screening of organic
pollutants, both in target and suspect mode, in aquaculture samples.
2. Development, validation and application of an analytical method based on

GC-(APCI)-MS/MS with QqQ for the quantification and confirmation of
organic contaminants previously detected in the screening analysis.
3. Study the potential of APCI, in combination with GC-MS/MS with QqQ,

as a technique for the determination of organic pollutants in complex matrices
7
Objectives
using calibration in solvent. Emphasis on polycyclic aromatic hydrocarbons.
4. Development and validation of an analytical method based on UHPLC-

MS/MS with QqQ for the quantification and confirmation of pirimiphos-
methyl, chlorpyrifos-methyl and their main metabolites in feed and tissues of
zebrafish and salmon, to generate analytical data for a future feed-to-fillet
transfer study.
5. Investigation of the capabilities of GC-(APCI)-QTOF MS coupling for the

elucidation and characterization of compounds whose EI spectra and
analytical standards are not available. Use of statistical tools for the study of
mass spectra of isomeric compounds.
8
CAPÍTULO 1
INTRODUCCIÓN GENERAL
CAPÍTULO 1: INTRODUCCIÓN GENERAL
1.1. Retos actuales de la acuicultura
1.2. Seguridad alimentaria y salud animal
1.3. Técnicas analíticas para el control alimentario
1.3.1. Técnicas cromatográficas
1.3.2. Espectrometría de masas
1.3.2.1. Fuentes de ionización
1.3.2.2. Analizadores de masas
1.4. Estrategias analíticas para el control alimentario
1.5. Tratamiento de muestra en matrices de acuicultura.
1.6. Estrategias de corrección del efecto matriz
1.7. Referencias
11
Capítulo 1 Introducción general
1.1. Retos actuales de la acuicultura
Ante la previsión de crecimiento de la población mundial por encima de los

9.700 millones de personas en el año 2050 [1], la comunidad mundial se
enfrenta al gran reto de asegurar un suministro alimentario y nutricional
adecuado a una población cada vez más numerosa, garantizando un desarrollo
sostenible en términos económicos, sociales y ambientales.
Según el último informe del “Estado Mundial de la Pesca y la Acuicultura”

(SOFIA) [2], el pescado - que representa aproximadamente el 17% del
consumo de proteínas de origen animal a escala mundial - registró un aumento
en su consumo per cápita de 9,9 kg en la década de 1960 hasta superar los 20
kg en la actualidad. Asimismo, el comercio mundial de pescado y productos
pesqueros destinados al consumo humano se ha quintuplicado en los últimos
cuarenta años pese a que el estado de los recursos marinos mundiales no ha
mejorado. Actualmente, casi un tercio de las poblaciones de peces con valor
comercial se capturan a un nivel biológicamente insostenible. Por ello, la
producción de pescado y productos pesqueros ha dejado de basarse
mayoritariamente en la captura de peces salvajes [2].
Ante el estancamiento de la producción de la pesca de captura desde finales

de la década de 1980, y la creciente demanda de pescado y productos
pesqueros, la acuicultura ha sido la desencadenante del importante
crecimiento del suministro mundial (figura 1). Si bien la acuicultura
proporcionaba en torno al 6% del pescado consumido en 1970, este porcentaje
aumentó hasta el 30% en el año 2000, hasta superar el 50% en la actualidad;
y las previsiones indican que la producción acuícola crecerá un 39% respecto
al periodo comprendido entre el 2013 y 2015 [2].
13
PRODUCCIÓN MUNDIAL DE LA PESCA DE CAPTURA Y ACUICULTURA
210
180
Millones de toneladas
150
120
90
60
30
0
1950 1956 1962 1968 1974 1980 1986 1992 1998 2004 2010 2016
TOTAL ACUICULTURA CAPTURA
Figura 1: Producción mundial de pesca de captura y acuicultura. Datos

extraídos de la web http://www.fao.org/fishery/topic/16140/es (visitada el
20/05/2019).
Así pues, en el contexto actual de rápido crecimiento y aumento de la demanda

de productos de pesca, la industria acuícola se enfrenta a un importante reto
de sostenibilidad dada su dependencia económica y ecológica con respecto a
harinas y aceites de origen marino. Estas materias primas proceden de la pesca
extractiva y se utilizan para la producción de piensos. Sin embargo, el
suministro de estas se encuentra en su límite sostenible, lo que comporta que
la expansión de esta industria deba realizarse usando fuentes alternativas de
ingredientes. Por ello, en la última década se han introducido ingredientes
vegetales en la composición de los piensos. A modo de ejemplo, en el caso del
salmón atlántico, los piensos han pasado de estar formados en su totalidad por
ingredientes de origen marino a contener un 70% de ingredientes vegetales,
principalmente aceites [3].
14
Estos cambios en la composición de las dietas plantean varias problemáticas.

En primer lugar, la acuicultura moderna depende de la adecuada provisión de
ácidos grasos poliinsaturados de cadena larga (LC-PUFA) de la serie omega-
3, esenciales para satisfacer los requerimientos fisiológicos [4]. No obstante,
el uso de ingredientes vegetales, ricos en LC-PUFA omega-6, compromete la
correcta provisión de LC-PUFA omega-3. Por un lado, la importancia de los
LC-PUFA omega-3 ha sido puesta de manifiesto en numerosos estudios que
resaltan el papel fundamental en diversos procesos fisiológicos y metabólicos
que aseguran el correcto funcionamiento celular [5]. Por otra parte, el pescado
representa la principal fuente de LC-PUFA omega-3 para el consumo humano.
Así pues, el cambio en la composición de las dietas de los peces podría tener
efectos negativos tanto en la composición nutricional del producto como en la
salud de los individuos cultivados.
En segundo lugar, se ha demostrado que el uso de ingredientes vegetales, junto

con técnicas comerciales de descontaminación [6–8], se ha traducido en la
disminución de contaminantes orgánicos persistentes (POPs) en peces de
cultivo [9, 10]. Sin embargo, con la sustitución de los ingredientes
tradicionales se han incrementado los niveles de otros contaminantes como
hidrocarburos policíclicos aromáticos (PAHs), relacionados con el tratamiento
térmico de los ingredientes vegetales, y se han introducido nuevos
contaminantes no asociados a las actividades de cría, como plaguicidas [11–
13]. Para garantizar la seguridad alimentaria, la mayoría de alimentos tienen
establecidos límites máximos de residuo (MRLs) por la Unión Europea (UE).
Sin embargo, en el caso del pescado, piensos e ingredientes de acuicultura
todavía no los tienen. Por ello, el conocimiento de los factores de migración y
el efecto de estos contaminantes sobre los peces son necesarios para establecer
límites máximos apropiados.
15
Por consiguiente, el gran reto actual de la acuicultura es la producción de peces

alimentados con dietas sostenibles y seguras que posibiliten el normal
desarrollo y crecimiento del pez en cautividad y que garanticen la seguridad
alimentaria al consumidor.
1.2. Seguridad alimentaria y salud animal
El Reglamento (CE) Nº 178/2002 [14] define los principios de la política de

seguridad alimentaria de la UE con la finalidad de establecer una base común
para las medidas aplicables a los productos en el ámbito nacional y
comunitario, facilitar la libre circulación de productos entre los países de la
UE, garantizar un nivel elevado de protección de la salud de las personas y de
los animales, y proteger los intereses de los consumidores mediante acciones
a lo largo de la cadena de producción alimentaria. Para este fin, la legislación
se basa en datos científicos y en una evaluación minuciosa del riesgo
proporcionado por agencias independientes, como la Autoridad Europea de
Seguridad Alimentaria (EFSA), que tiene el cometido de prestar
asesoramiento científico-técnico, emitir dictámenes y actuar ante las crisis de
seguridad alimentaria.
Una de las principales competencias de la UE en materia de seguridad

alimentaria es garantizar el bienestar animal de las especies cultivadas para la
producción de alimentos. Este ámbito, que comprende factores como las
condiciones de espacio y temperatura de cría, las condiciones de transporte o
los métodos de sacrificio, está estrechamente vinculado con la salud animal y
las enfermedades transmitidas por éstos, comprometiendo, por tanto, la
seguridad de la cadena alimentaria. Por ello, todos los factores relacionados
16
con el bienestar animal fueron abordados en 1976 en la European Convention

for the Protection of Animals kept for Farming Purposes [15], donde se
establecieron los principios comunes para la protección de animales
cultivados.
De entre todos los factores, uno de los que más interés científico despierta y
que mayor efecto tiene sobre el bienestar animal y la seguridad alimentaria, es
la alimentación de los individuos de cría. Por una parte, según los principios
acordados en la European Convention for the Protection of Animals kept for
Farming Purposes, los animales deben ser provistos de alimentos que sean
adecuados para sus necesidades fisiológicas y etológicas. Con este fin, y en
relación con la problemática en la sostenibilidad de los piensos expuesta en el
primer punto de la introducción, la acuicultura moderna está centrando sus
esfuerzos en ampliar el conocimiento científico. De hecho, la UE ha
financiado proyectos como “AQUAMAX”, “ARRAINA” o “PUFAFEEDS”,
entre otros, dirigidos a esta materia. Por otra parte, dichos alimentos deben
estar exentos de sustancias que puedan causar daños o lesiones innecesarias.
Por ello, la UE ha establecido MRLs para determinadas sustancias tóxicas en
ingredientes y piensos para peces - como PAHs, bifenilos policlorados (PCBs)
o dioxinas [11] - con el objetivo de controlar y reducir los niveles de estas
sustancias así como su seguridad. A pesar de todo, los avances en el campo de
la acuicultura, con el desarrollo de nuevos piensos de composición
principalmente vegetal, han planteado un escenario escasamente contemplado
por la legislación vigente. Pese a la existencia de MRLs de plaguicidas para
gran parte de alimentos de carácter vegetal, los cuales han sido establecidos
en base a buenas prácticas agrícolas (GAP), los cultivos utilizados en la
alimentación animal, piensos y productos de acuicultura no han sido
contemplados. Es por ello que urge una evaluación del efecto de los residuos
17
de los plaguicidas y de los posibles riesgos para la salud del consumidor y del
animal, siendo estos estudios llevados a cabo normalmente por agencias como
la EFSA. A partir de estos se establecerían los MRLs correspondientes.
Para controlar el cumplimiento de dichos MRLs es necesario realizar análisis

periódicos, utilizando métodos analíticos fiables, robustos y estrictamente
validados siguiendo las últimas versiones de guías de calidad analítica como
la SANTE/11813/2017 [16]. Es aquí, pues, donde los últimos avances en la
investigación analítica se han convertido en un elemento imprescindible en el
área de la seguridad alimentaria. El desarrollo de metodología analítica
sofisticada, que incorpore los últimos avances en el ámbito instrumental,
resulta crucial a la hora de determinar residuos de plaguicidas u otras
sustancias contaminantes con un alto grado de fiabilidad, y capaz de cumplir
con las exigencias de la legislación vigente.
1.3. Técnicas analíticas para el control alimentario
En las últimas décadas, ha habido una creciente preocupación en el ámbito de

la seguridad alimentaria. Ante esta preocupación, el desarrollo de
metodologías analíticas para la determinación de contaminantes orgánicos y
compuestos endógenos ha supuesto un avance evidente en el campo del
control alimentario y ambiental. Sin embargo, su determinación continúa
siendo un desafío. El elevado número de compuestos de interés con
propiedades físico-químicas muy diferentes, la complejidad de las muestras a
analizar y los bajos niveles de concentración a los que habitualmente se
encuentran, requieren de instrumentación analítica poderosa. A este respecto,
los acoplamientos GC-MS y LC-MS se han consolidado como las técnicas
18
más ampliamente utilizadas en este ámbito [17]. Asimismo, la espectrometría

de masas presenta mucha versatilidad, debido sobre todo a los diferentes
analizadores que existen, así como las fuentes de ionización disponibles.
A lo largo de los trabajos presentados en esta tesis, y con el objetivo de abarcar

un amplio abanico de contaminantes orgánicos de diferente naturaleza, se han
utilizado tanto la GC como la LC acopladas a espectrometría de masas con
diferentes tipos de analizadores. A continuación, se describirán brevemente
los fundamentos de las técnicas utilizadas en esta Tesis.
1.3.1. Técnicas cromatográficas
La cromatografía se define como un método físico de separación en el que los

componentes de una mezcla se distribuyen entre dos fases, una estacionaria y
otra móvil que se mueve en una dirección determinada. El proceso
cromatográfico ocurre como resultado de la interacción de los componentes
con la fase estacionaria y la separación de éstos dependerá de las diferencias
en sus coeficientes de distribución [18].
La clasificación de las técnicas cromatográficas se basa en la naturaleza física

de la fase móvil, siendo un gas inerte en la cromatografía de gases (GC) y un
líquido en la cromatografía de líquidos (LC).
Cromatografía de gases
La cromatografía de gases es una técnica de separación basada en la presión

de vapor relativa de los analitos y la interacción de éstos con una fase
estacionaria contenida en el interior de la columna cromatográfica [19]. Se
19
trata de una técnica muy útil para la separación de compuestos volátiles o

semivolátiles y necesariamente termoestables como, por ejemplo, los PAHs,
PCBs o algunos plaguicidas.
El pilar central de la cromatografía de gases reside en la columna

cromatográfica. Actualmente, la mayoría de las aplicaciones utilizan
columnas capilares debido a su alta eficacia de separación y a la gran variedad
de características químicas, dimensiones y espesor de la fase estacionaria
disponibles comercialmente. Esto ha permitido optimizar sus propiedades a la
naturaleza de los analitos y matrices, mejorando así la sensibilidad de su
detección respecto a otras columnas. No obstante, una buena separación
cromatográfica está estrechamente ligada a la primera etapa de todo proceso
cromatográfico: la inyección de la muestra al sistema.
La inyección de la muestra en GC es una etapa crítica que debe realizarse

adecuadamente para asegurar una óptima transferencia de los analitos a la
columna. En función de las características de las muestras, la naturaleza de los
analitos, la volatilidad del disolvente y el sistema cromatográfico, existen
diferentes técnicas de inyección para columnas capilares. Entre las más
utilizadas se encuentran:
 Inyección con división de muestra o split: permite la introducción en

la columna capilar de una fracción de la muestra inyectada a alta temperatura
en la cámara de vaporización, mediante el ajuste de los flujos de gas portador
y gas de purga. Se usa principalmente para evitar la saturación de la fase
estacionaria al inyectar muestras muy concentradas o sucias [20].
 Inyección sin división de muestra o splitless: utilizando el mismo

sistema instrumental que la inyección en modo split, permite introducir
20
prácticamente la totalidad de la muestra inyectada en la columna capilar ya

que, a diferencia del anterior, la válvula de purga se encuentra cerrada en el
momento de la inyección. Es la técnica ampliamente utilizada en el análisis de
trazas pese a requerir un tratamiento de muestra severo [20]. En los artículos
científicos I, II y IV se ha empleado la inyección mediante el sistema pulsed
splitless. Se trata de un sistema splitless en el cual, cuando la muestra es
inyectada, aumenta notablemente el flujo de gas portador, favoreciendo la
transferencia de los analitos a la columna. Como resultado, se reducen posibles
fenómenos de discriminación, adsorción y/o degradación de los analitos en la
cámara de vaporización [21].
 Inyección en columna: permite la inyección de la muestra directamente

en la columna capilar, sin necesidad de una cámara de vaporización. Los
analitos se separan del disolvente mediante efectos térmico y de solvente. Esta
técnica evita problemas de discriminación y posible degradación térmica. No
obstante, requiere de mayor tratamiento de muestra para generar extractos
limpios y así evitar el deterioro de la cabeza de columna [20].
 Inyección de vaporización con temperatura programada (PTV): se trata

de un híbrido de las técnicas descritas anteriormente. La muestra se introduce
a baja temperatura en un inyector split/splitless y, posteriormente, la cámara
de vaporización se calienta rápidamente para desorber la muestra en la
columna capilar. Esta técnica permite la inyección de grandes volúmenes de
muestra, aportándole mayor versatilidad [20].
21
Cromatografía de líquidos
La cromatografía de líquidos es una técnica de separación basada en la

diferencia de afinidad de los compuestos de una muestra entre una fase móvil,
líquida, compuesta por una mezcla de disolventes con diferente fuerza
eluotrópica; y una fase estacionaria, sólida, empaquetada en una columna. Se
trata de una técnica muy versátil y especialmente útil para la separación de
compuestos con carácter polar, poco volátiles y termolábiles.
Al igual que en GC, la columna cromatográfica es determinante en el proceso

de separación. Así, en función del tipo de las partículas de relleno se producirá
un tipo u otro. El tipo de separación cromatográfica más ampliamente
utilizado en LC es la denominada fase reversa, es decir, la que utiliza una fase
estacionaria apolar formada normalmente por sílice modificada con grupos
octadecilo (C18). Como fase móvil, se utilizan mezclas de agua con disolventes
orgánicos, como metanol o acetonitrilo, que pueden contener modificadores
para mejorar la retención, forma de pico y/o sensibilidad de ciertos
compuestos. En este tipo de cromatografía, en la que la retención disminuye
generalmente al aumentar la polaridad del analito, las interacciones entre
analito y solvente son las que determinan su especificidad, ya que las
interacciones de los analitos con la fase estacionaria son relativamente débiles
e inespecíficas debido a interacciones de tipo Van der Waals.
Además del tipo de química de las fases estacionarias hay otros factores que
afectan a la separación, como es el tamaño de partícula que, junto con el flujo
de fase móvil, tiene un efecto directo sobre la eficacia cromatográfica según
la ecuación de Van Deemter. Esta ecuación relaciona la velocidad lineal de la
fase móvil con la eficacia de la columna para un determinado tamaño de
partícula. Con el objetivo de mejorar la eficacia, en los últimos años se han
22
desarrollado columnas cromatográficas con tamaño de partícula inferior a 2

µm, frente a las tradicionales columnas HPLC de 3.5 y 5 µm. Estas columnas,
denominadas UHPLC o de ultra alta resolución, aportan una mayor eficacia
en un amplio rango de flujos elevados, a diferencia de las columnas HPLC
cuyo rango de flujos óptimos es muy limitado. El hecho de permitir trabajar a
flujos mayores sin afectar la eficacia, favorece la reducción de los tiempos de
análisis. Además, al generar picos cromatográficos más estrechos también se
incrementa la capacidad de picos de la columna.
La cromatografía UHPLC en fase reversa ha sido utilizada en los artículos

científicos I y III presentados en los capítulos 2 y 3.
1.3.2. Espectrometría de masas
La espectrometría de masas es una técnica analítica que permite la separación

de iones en base a una propiedad intrínseca de la materia ionizada: la relación
masa/carga (m/z). Se trata de una técnica muy versátil y potente, habitualmente
acoplada a cromatografía, que permite obtener información estructural,
cualitativa y cuantitativa de las moléculas presentes en las muestras. Este
acoplamiento ha supuesto un gran avance en las técnicas de análisis, y su
potencial reside en el carácter bidimensional de separación que proporciona la
cromatografía junto con la detección sensible y selectiva de la espectrometría
de masas.
El espectrómetro de masas presenta tres componentes fundamentales: la

fuente de ionización, el analizador de masas y el detector. A continuación, se
describirán las dos primeras.
23
1.3.2.1. Fuentes de ionización
El acoplamiento del sistema cromatográfico al espectrómetro de masas se

realiza mediante una interfase. En el caso de GC, dada la naturaleza gaseosa
de la técnica, esta interfase consiste en una línea de transferencia caliente
(entre 200 – 350ºC) que contiene la parte final de la columna cromatográfica
en su interior y que la conecta con la fuente de ionización. No obstante, en el
caso de LC, son las propias interfases las que actúan como fuente de
ionización.
Ionización en GC
Ionización electrónica (EI)
La ionización electrónica es la fuente más comúnmente empleada en el

acoplamiento GC-MS. Se trata de una fuente robusta, reproducible y universal
que opera en condiciones de vacío y cuyo mecanismo de ionización se basa
en el bombardeo de las moléculas con un haz de electrones de alta energía
(habitualmente 70 eV) producidos desde un filamento caliente. Estos
electrones son acelerados mediante un campo eléctrico, interaccionan con las
moléculas y les extraen un electrón. En este proceso, las moléculas adquieren
un exceso energético que eliminan mediante movimientos vibracionales,
rotacionales y de reordenación molecular, lo que provoca la fragmentación de
las mismas.
La principal ventaja de la EI reside en la disponibilidad comercial de librerías

de espectros de masas, lo que facilita la identificación de compuestos. No
obstante, presenta una serie de inconvenientes. Debido a la elevada energía
involucrada en el proceso de ionización, ciertos tipos de moléculas son
24
altamente fragmentadas provocando, incluso, la pérdida total o parcial del ion

molecular en el espectro de masas. Ante estas circunstancias, la identificación
de compuestos especialmente los que no están incluidos en las librerías
espectrales supone un reto por la complejidad en la interpretación del
espectro y por la falta de información de la molécula intacta.
Ionización química (CI)
La ionización química es una técnica de ionización basada en la interacción

de iones reactivos de baja energía provenientes de un gas ionizado mediante
el bombardeo con un haz de electrones de alta energía con las moléculas de
los analitos. El proceso de ionización en CI puede implicar la transferencia de
un electrón, un protón u otra especie cargada. A diferencia de la EI, la
presencia de una atmósfera rica en moléculas neutras de gas reactivo reduce
la energía de los electrones emitidos y de las moléculas ionizadas, aportando
a esta técnica un carácter de ionización suave.
La ionización química puede dar lugar a una especie positiva y/o negativa, y
ambos procesos pueden ocurrir simultáneamente. En modo positivo, la
ionización suele producirse mediante la transferencia de un protón, aunque
también se producen otras reacciones ion-molécula como la formación de
aductos con los iones reactivos, la extracción de hidruro o la transferencia de
carga. En modo negativo, puede producirse la reacción ion-molécula de
desprotonación, aunque los principales mecanismos son la captura de electrón
por resonancia, la captura de electrón disociativa o la formación de pares
iónicos. Al igual que en EI, la fragmentación de las moléculas dependerá de
la energía involucrada en el proceso de ionización.
La principal ventaja de la CI reside en la selectividad y en la alta sensibilidad
25
que presenta en modo negativo para la determinación de compuestos

electrofílicos. Además, el carácter suave de ionización permite obtener, en
algunos casos, información sobre la molécula prácticamente intacta,
facilitando la identificación de compuestos desconocidos. Sin embargo, es una
técnica poco popular ya que, a diferencia de EI, no es universal, no dispone de
librerías espectrales y es poco robusta y reproducible.
Ionización química a presión atmosférica (APCI)
La ionización química a presión atmosférica combina la universalidad de la

EI con el carácter de ionización suave de la CI. Se trata de una fuente de
ionización a presión atmosférica (API) cuyo mecanismo se inicia con la
ionización de un gas reactivo (generalmente N2) mediante una descarga
eléctrica generada en un pin, originando un plasma de nitrógeno por efecto
corona. Como puede observarse en la figura 2, este plasma ioniza los analitos
mediante dos mecanismos simultáneos y competitivos entre ellos: la
transferencia de carga y/o la protonación (en presencia de humedad). La
ionización mediante transferencia de carga origina normalmente el ion
molecular (M+·) como pico base del espectro, mientras que la molécula
protonada ([M+H]+) suele ser la más abundante en la ionización por
protonación. Dada la dificultad de controlar las condiciones de humedad en la
fuente, en moléculas que presentan algún grupo protonable, es común
observar en el espectro de masas ambos iones. Para favorecer las condiciones
de protonación se pueden añadir en la fuente de APCI modificadores próticos
como agua, metanol o isopropanol, de modo que la abundancia de [M+H]+
aumenta para aquellos compuestos cuya afinidad prótica es superior a la del
modificador.
26
Figura 2: Esquema de la fuente APCI. Las flechas naranjas, azules y moradas

indican los flujos de N2 provenientes del gas de cono, gas auxiliar y gas de
make up, respectivamente. La flecha roja corresponde al eluato de la columna.
El potencial de la fuente APCI reside en la escasa o nula fragmentación

producida durante el proceso de ionización. Pese a la ausencia, de momento,
de amplias librerías espectrales, el carácter de ionización suave facilita la
identificación de compuestos desconocidos ya que mantiene la información
de la estructura molecular prácticamente intacta. Además, simplifica la
interpretación de los espectros de masas, ya que la información estructural
obtenida tras experimentos de masas en tándem (MS/MS) proviene, en la
mayoría de casos, de M+· y/o [M+H]+ como iones precursores. Este hecho
supone una gran ventaja en el desarrollo de métodos target o dirigidos
mediante el uso de MS/MS, ya que al ser más abundantes y selectivos los iones
precursores, las transiciones adquiridas serán también más sensibles y
específicas.
27
Ionización en LC
Ionización mediante electrospray (ESI)
La interfase de ionización electrospray es una técnica API ampliamente

utilizada en el acoplamiento LC-MS, que se basa en la producción de un
aerosol formado por pequeñas gotas cargadas bajo la influencia de un campo
eléctrico. Para ello, los analitos disueltos en la fase móvil pasan a través de un
capilar cuya punta se encuentra a una diferencia de potencial de unos 3 kV
respecto al contraelectrodo. Esta diferencia de potencial produce un intenso
campo eléctrico capaz de dispersar el eluato y generar un aerosol de pequeñas
gotas cargadas ayudado por un gas de nebulización. Al atravesar la región a
presión atmosférica, la mayor parte de las moléculas de disolvente se
evaporan. Debido a los flujos utilizados en LC, el proceso de desolvatación se
acelera mediante una corriente de nitrógeno caliente. A medida que se evapora
el disolvente, las gotas van reduciendo su tamaño hasta que las fuerzas
repulsivas entre los iones generados en su interior son capaces de superar las
fuerzas cohesivas de tensión superficial, produciendo la ruptura de la gota y
originando iones en fase gaseosa. Dado el carácter de ionización suave, los
espectros obtenidos suelen estar dominados por las moléculas protonadas o
desprotonadas, según la ionización se efectúe en modo positivo o negativo,
respectivamente.
La interfase ESI utilizada en los artículos científicos I y III presenta la

configuración Z-Spray. A diferencia de las fuentes ESI convencionales, en las
que la trayectoria de los iones desde el capilar hasta el espectrómetro de masas
es una línea recta, la configuración Z-Spray presenta una trayectoria en forma
de Z, como bien indica su nombre, permitiendo que únicamente las moléculas
28
ionizadas fluyan hacia el espectrómetro de masas mediante la aplicación de

potenciales. De este modo, se reduce la señal de fondo y se mejora la relación
señal/ruido (S/N).
Uno de los puntos fuertes de la interfase ESI es la posibilidad de generar iones

policargados. Esta característica ha facilitado el análisis de macromoléculas
biológicas mediante MS, pudiendo extender el rango de masas del instrumento
en un factor equivalente al número de cargas de la molécula del analito.
Además, es una técnica muy versátil ya que permite el análisis de compuestos
termolábiles, con un amplio rango de pesos moleculares y también de
polaridades: desde compuestos iónicos a otros con cierto carácter lipofílico.
Sin embargo, el principal inconveniente que presenta es el elevado efecto que
tiene la presencia de compuestos de la matriz sobre la eficacia en la ionización.
1.3.2.2. Analizadores de masas
Cuadrupolo simple (Q)
El filtro de masas cuadrupolar o cuadrupolo simple consta de cuatro

superficies hiperbólicas paralelas entre ellas a las que se les aplica una
corriente continua (DC) y un voltaje de radiofrecuencia (RF). La combinación
de ambos potenciales a cada par de superficies opuestas genera un campo
eléctrico oscilante de alta frecuencia que permite establecer trayectorias
estables, a través de la geometría del cuadrupolo, para iones con una
determinada relación masa/carga (m/z). De este modo, el resto de iones con
relación m/z distinta a la estabilizada por los potenciales se pierden a lo largo
del filtro de masas cuadrupolar.
29
Este tipo de analizador de masas, por su modo de funcionamiento, permite

adquirir en Full Scan, es decir, realizar barridos completos de iones en un
determinado rango de m/z, obteniendo así el espectro de masas completo; o,
por otra parte, trabajar en modo Selected Ion Monitoring (SIM), en el que se
monitorizan únicamente unos iones previamente seleccionados, aumentando
la sensibilidad y selectividad de la adquisición.
Triple cuadrupolo (QqQ)
El analizador de masas de triple cuadrupolo consiste en la combinación de dos

cuadrupolos (Q1, Q2) acoplados en serie a otro (q, habitualmente hexapolo u
octapolo) situado entre ellos, que normalmente actúa como celda de colisión.
Se trata de un sistema muy versátil ya que permite la adquisición en diferentes
modos, tanto en espectrometría de masas simple (MS) como en tándem
(MS/MS). Sin embargo, es la configuración MS/MS la que ofrece todo su
potencial, aportando gran sensibilidad y selectividad; y proporcionando
información estructural adicional de los compuestos por medio de la celda de
colisión, que produce la fragmentación de los iones mediante la disociación
por colisión inducida (CID) con un gas no reactivo (Ar, He o N2). Existen
diferentes modos de adquisición en un QqQ (ver Tabla 1), aunque el más
interesante en análisis cuantitativo es el modo Selected Reaction Monitoring
(SRM). En este modo, el Q1 aísla un ion precursor previamente seleccionado,
se fragmenta en la celda de colisión y el Q2 aísla un ion producto previamente
establecido, mejorando así la selectividad y la relación S/N del método.
30
Tabla 1: Modos de adquisición en analizadores de masas de triple cuadrupolo.
MODO Q1 q Q2 APLICACIÓN
Full Scan - - Scan Conocer la ionización de la molécula
Adquirir un ion específico de la

SIM - - SIM
molécula
Product Obtener información estructural de la

SIM CID Scan
Ion Scan molécula
Adquirir una reacción CID (métodos

SRM SIM CID SIM
cuantitativos)
Precursor Obtener información de compuestos

Scan CID SIM
Ion Search que presentan fragmentos idénticos
Neutral Obtener información de compuestos

Loss Scan CID Scan que presentan pérdidas neutras por
Search fragmentación idénticas
El sistema instrumental Xevo TQ-S utilizado en los artículos científicos I, II

y III, consiste en un triple cuadrupolo que dispone del sistema de transferencia
de iones StepWave. Este sistema consiste en una serie de electrodos anulares
apilados que se combinan para formar una lente de transmisión de iones fuera
del eje (en forma de escalón). Este diseño permite eliminar los compuestos no
ionizados que pudiesen entrar al sistema mediante el vacío, mientras que
aquellos ionizados entran al sistema conducidos por una serie de potenciales
aplicados a los electrodos. La principal ventaja que presenta es la mejora de la
relación S/N al eliminar parte de la señal de fondo, especialmente en matrices
complejas.
31
Híbrido cuadrupolo – tiempo de vuelo (QTOF)
El analizador de masas QTOF combina el filtro de masas cuadrupolar con la

alta resolución del analizador TOF. El funcionamiento del TOF se basa en la
separación de los iones en función del tiempo que tardan en atravesar un tubo
de vuelo. Al entrar al analizador de masas, los iones son sometidos a un pulso
de voltaje que los acelera a través del tubo de vuelo. Partiendo del principio
de conservación de la energía cinética de los iones - adquirida al ser
previamente acelerados a través de la fuente de ionización -, estos se separan
a diferentes velocidades, que son inversamente proporcionales a la raíz
cuadrada de su relación m/z. De este modo, para cada pulso de voltaje se
obtiene un espectro de masas completo. El instrumento TOF requiere estos
pulsos de voltaje para transferir de manera simultánea e instantánea los
paquetes de iones, así como evitar el solapamiento entre éstos, ya que la
frecuencia en la que son emitidos está determinada por el tiempo que tarda el
último ion en llegar al detector.
Para optimizar la transferencia de iones, minimizar la dispersión espacial de

los mismos y, por ende, aumentar la sensibilidad y determinar con mayor
precisión los tiempos de vuelo, los instrumentos TOF presentan,
habitualmente, la configuración de aceleración ortogonal; como es el caso del
sistema instrumental Xevo G2 QTOF utilizado en los artículos científicos I,
III y IV. En esta configuración, el haz de iones proveniente de la fuente de
ionización accede al analizador desde una dirección perpendicular al tubo de
vuelo. Además, los instrumentos TOF de alta resolución suelen disponer de
reflectrones. El reflectrón consiste en una serie de electrodos colocados al final
del tubo de vuelo que corrige los efectos de la dispersión espacial y energética
iniciales. Su funcionamiento se basa en la reducción de la velocidad de los
32
iones que entran al dispositivo mediante un campo eléctrico repelente, y su

posterior aceleración hacia el detector. De esta manera, los iones con la misma
relación m/z pero con velocidades iniciales diferentes, se reenfocan para ser
detectados simultáneamente. El sistema instrumental Xevo G2 QTOF
utilizado en esta tesis, presenta la configuración en V (con un reflectrón), lo
que permite duplicar la distancia de vuelo y, por tanto, el poder de resolución.
El hecho de combinar el cuadrupolo con el TOF permite trabajar en modos de

adquisición como Full Spectrum, Product Ion Scan y Data Independent
Acquisition (MSE, en el caso del instrumento utilizado en esta tesis). Para los
dos últimos modos de adquisición, el QTOF dispone de la celda de colisión
CID. En el modo Product Ion Scan, el ion precursor se aísla en el cuadrupolo,
se fragmenta en la celda de colisión a unas energías determinadas y se detectan
todos los iones producto en el TOF. En el modo MSE, por su parte, los iones
precursores no se aíslan, pero la celda de colisión actúa secuencialmente como
elemento de transmisión (función de baja energía, LE) y como celda de
colisión en la cual se aplica una rampa de potencial (función de alta energía,
HE). En este último modo, se adquieren en una única inyección ambas
funciones, obteniendo información conjunta de las moléculas “intactas” y
fragmentadas, permitiendo el análisis retrospectivo.
1.4. Estrategias analíticas para el control alimentario
En el apartado anterior se han descrito las diferentes opciones instrumentales

utilizadas a lo largo de esta tesis ya que, desde el punto de vista analítico,
resulta crucial conocer el potencial de cada instrumento y aprovecharlo para
desarrollar estrategias que permitan mejorar el alcance de los análisis. A
33
continuación se describen brevemente aquellas utilizadas a lo largo de esta

tesis.
Análisis pre-target
Se trata de un modo de análisis en el que se definen previamente los

compuestos de interés y se optimizan sus condiciones de medida con la
finalidad de detectarlos, identificarlos y, normalmente, cuantificarlos en las
matrices de estudio. Para esta estrategia es habitual la utilización de
analizadores de masas QqQ en modo MS/MS.
Las principales ventajas de este tipo de análisis residen en la posibilidad de

optimización de todo el proceso analítico mediante el uso de patrones de
referencia, incrementando así la sensibilidad del método. Además, este facilita
el procesamiento de datos dado que se adquiere un número limitado de
transiciones.
Por otra parte, el principal inconveniente de este modo de análisis se encuentra

precisamente en la limitación a los compuestos incluidos en el método. El
análisis pre-target, al centrar su alcance en una serie definida de compuestos,
no permite el análisis retrospectivo, es decir, el que se realiza a partir de datos
adquiridos en un análisis anterior, con el fin de investigar otros compuestos de
interés no considerados inicialmente.
Screening de amplio alcance
La alternativa al análisis pre-target pasa por la adquisición en modo full-scan

o full spectrum. Este tipo de adquisición permite llevar a cabo estrategias de
screening de amplio alcance ya que, al no requerir de la preselección de m/z
determinadas se evita la limitación de compuestos a analizar y permite el
34
análisis retrospectivo. Habitualmente, este tipo de análisis se lleva a cabo con

espectrómetros de masas de alta resolución, como QTOF o QOrbitrap, que
aportan espectros completos medidos con elevada exactitud de masa.
Sin embargo, este modo de análisis presenta el principal inconveniente de

generar gran cantidad de datos, muchos de ellos irrelevantes desde el punto de
vista analítico, que dificultan el procesamiento de los mismos. Por ello, es
necesario establecer aproximaciones que filtren los datos correspondientes a
los compuestos de interés. Basándonos en la nomenclatura utilizada por
Hernández et al. [22], podemos distinguir tres tipos de aproximaciones para
abordar el tratamiento de los datos:
- Target screening: se refiere al tipo de screening en el que, posteriormente a

la adquisición de los datos, se seleccionan unos compuestos a investigar para
los cuales se posee su información espectral y cromatográfica obtenida a partir
de los respectivos patrones de referencia.
- Suspect screening: a diferencia del target screening, no se posee

información espectral ni cromatográfica previa, dada la ausencia de patrones
de referencia, y únicamente se dispone de información molecular.
- Non-target screening: es el tipo de screening en el que, mediante algoritmos

de deconvolución espectral, se encuentran picos cromatográficos presentes en
la muestra para la posterior elucidación e identificación de aquellos que
puedan resultar relevantes analíticamente. En otras palabras, se investiga la
presencia de compuestos para los cuales no se dispone de información
espectral, ni cromatográfica, ni molecular previa. Esta aproximación no se ha
utilizado en los trabajos presentados en la presente tesis.
35
1.5. Tratamiento de muestra en matrices de acuicultura.
La aplicación de la química analítica en el campo de la acuicultura implica el

análisis de una gran variedad de matrices de diferente naturaleza y
complejidad, desde ingredientes de los piensos a pescado. Además, las
propiedades físico-químicas de los analitos también son muy amplias, lo que
dificulta encontrar métodos que funcionen correctamente para todas las
combinaciones matriz/analito. Estos factores, junto a la baja concentración en
la que suelen encontrarse los analitos en las muestras de análisis, hacen
necesaria la aplicación de tratamientos de muestra eficaces que permitan la
obtención de extractos limpios, compatibles con los sistemas instrumentales,
para proporcionar resultados fiables.
El análisis de muestras sólidas requiere, en primer lugar, la extracción de los

compuestos de interés de la matriz mediante el uso de disolventes orgánicos
y, en segundo lugar, la purificación o clean up del extracto para reducir o
eliminar componentes de la matriz que puedan interferir en el análisis.
Algunos de los sistemas de tratamiento de muestra tradicionalmente utilizados
en el análisis de muestras de acuicultura son el Soxhlet, la extracción con
fluidos supercríticos (SFE), la extracción asistida por microondas (MAE), la
extracción mediante ultrasonidos (USE), la extracción líquida presurizada
(PLE) y la cromatografía por permeación en gel (GPC) [23]. Sin embargo,
estos sistemas son costosos en cuanto a tiempo y volumen de disolventes,
como el Soxhlet; o en cuanto a equipamiento, en el caso de SFE, MAE, PLE
y GPC. Debido al creciente interés en desarrollar métodos más rápidos,
económicos, simples y verdes, la aplicación de QuEChERS goza de gran
popularidad en la actualidad.
36
Desde que fue diseñado en 2003 para la extracción de pesticidas en frutas y

vegetales [24], basado en la extracción con acetonitrilo, el método QuEChERS
original se ha sometido a modificaciones que lo convierten en un método
versátil, adaptable a matrices y analitos de diversa naturaleza, así como a las
características instrumentales. Entre las modificaciones desarrolladas, cabe
destacar, en el análisis de residuos de plaguicidas, dos de ellas que mejoran la
extracción del método original sobre pesticidas pH-dependientes: el método
15662 del Comité Europeo de Normalización (CEN), que emplea tampón de
citrato [25]; y el método oficial 2007.01 de la Asociación de Comunidades
Analíticas (AOAC), que usa acetato sódico como tampón [26]. Pese a ser
métodos de referencia en el análisis de plaguicidas en frutas y verduras, el uso
de métodos basados en QuEChERS se ha extendido en la determinación de
otros contaminantes como PAHs, PCBs o difeniléteres polibromados
(PBDEs) en matrices de acuicultura [27, 28], cambiando, en algunos casos, el
extractante a acetato de etilo para favorecer la extracción de los analitos [29,
30].
La versatilidad de QuEChERS no sólo reside en la etapa de extracción.

Actualmente, existen diferentes adsorbentes que facilitan la optimización del
proceso de clean up en función de las características de la matriz y de los
analitos. De hecho, en el método original, la etapa de purificación se realiza
mediante el uso de MgSO4 anhidro y amina primaria-secundaria (PSA). Sin
embargo, ahora es habitual encontrar aplicaciones en las que también se utiliza
carbón negro grafitizado (GCB), C18 o materiales de última generación como
Z-Sep, que permiten obtener extractos más limpios [27, 31–33].
A pesar de la efectividad del tratamiento de muestra para reducir el efecto

matriz, es decir, el efecto que tienen los compuestos coextraídos de la matriz
37
sobre la respuesta de los analitos, éste puede ser notorio en algunos tipos de
muestras. El efecto matriz puede originar errores de cuantificación y
detección, así como afectar la sensibilidad, reproducibilidad y linealidad del
método. Por estos motivos, es necesario evaluar cómo afecta la presencia de
la matriz en los compuestos a determinar para así definir las estrategias que lo
corrijan.
1.6. Estrategias de corrección del efecto matriz.
El mecanismo de supresión/exaltación iónica en cromatografía acoplada a

espectrometría de masas puede variar en función del tipo de cromatografía e
ionización.
Por un lado, en LC-MS con fuente de ionización ESI, se piensa que el efecto
matriz se produce en la fuente de ionización y se debe probablemente a un
proceso de competición entre los analitos y los componentes de la matriz
coeluídos [34], bien por acceder a la superficie de las pequeñas gotas donde
se produce el paso a fase gas - a mayor fuerzas atractivas para mantener las
gotas, menor eficiencia de ionización [35] -, o bien porque la concentración
de compuestos orgánicos es superior a 10-5 M, fijando esta concentración
como máxima para obtener una ionización eficiente [34, 36]. En este caso es
común observar supresión de señal.
Por otro lado, en GC, y especialmente en combinación con la fuente EI, el

efecto matriz se produce principalmente en el sistema de inyección y/o
cromatográfico debido a la interacción de la matriz con los puntos activos del
liner y de la columna cromatográfica. Los componentes de la matriz tienden a
38
bloquear estos puntos activos, reduciendo así las pérdidas de analitos causadas
por la adsorción i/o degradación en éstos y, por tanto, obteniendo una
respuesta mayor (exaltación de señal) en los extractos en comparación con los
estándares de solventes [37–39].
Independientemente del mecanismo de supresión/exaltación de la señal,

existen diferentes estrategias comunes para eliminar o compensar el efecto
matriz.
Estrategias en el proceso de cuantificación
Tienen como objetivo compensar el efecto que pueden producir los

interferentes de matriz en el proceso de análisis. Entre las estrategias más
utilizadas se encuentran: el uso de respuesta relativa, el calibrado en matriz y
el método de adiciones estándar.
El uso de la respuesta relativa consiste en la utilización de un patrón interno,

el cual debe estar ausente en las muestras y debe presentar un comportamiento
similar al de los analitos estudiados. Su adición puede realizarse sobre la
muestra cruda con el objetivo de corregir posibles pérdidas en el proceso de
tratamiento de muestra y el efecto matriz (surrogate), o adicionarse al extracto
final con el fin de corregir el efecto matriz y posibles variaciones en la
inyección (internal standard, IS). La dificultad de encontrar un patrón interno
que esté ausente en las muestras y presente un efecto matriz similar (además
del comportamiento durante el tratamiento de muestra, en el caso de ser
adicionado como surrogate) al de los compuestos de interés dificulta esta
aproximación. No obstante, existe la posibilidad de utilizar patrones internos
marcados isotópicamente (Stable Isotopically Labelled Internal Standards,
SIL-IS), es decir, compuestos estructuralmente iguales a los analitos de interés
39
– por lo que su comportamiento cromatográfico y espectrométrico es

prácticamente idéntico al compuesto nativo - a los que se les sustituyen
algunos átomos por sus respectivos isótopos. El principal requisito que deben
cumplir los SIL-IS es que contengan suficientes átomos sustituidos por sus
correspondientes isótopos para evitar el solapamiento entre la distribución
isotópica del compuesto marcado con el nativo. En esta aproximación, la
respuesta relativa entre la señal del compuesto de interés y la del SIL-IS - este
último adicionado siempre a una concentración fija en todas las muestras - se
utiliza para cuantificar la concentración del primero corrigiendo las
variaciones debidas a la matriz. Aunque el uso de SIL-IS es la opción más
adecuada, presenta como principales inconvenientes la limitada disponibilidad
comercial y su elevado precio.
El uso del calibrado en matriz es otra aproximación ampliamente utilizada

para la corrección del efecto matriz. Esta opción consiste en la preparación de
curvas de calibrado sobre extracto blanco de la matriz, sometido al mismo
tratamiento que las muestras a analizar, para cuantificar los analitos. De este
modo, se compensa el efecto matriz de las muestras con el de los patrones de
cuantificación. El principal requisito de esta alternativa es disponer de
muestras blanco representativas ya que, de no serlo, se puede incurrir en
errores de cuantificación debido a diferencias en la supresión/exaltación de la
señal.
Por último, aunque menos utilizadas, las adiciones estándar son una
posibilidad para corregir el efecto matriz. Esta alternativa consiste en añadir
cantidades crecientes de analitos de interés a una cantidad fija de muestra. Esta
aproximación presenta dos limitaciones principales: un aumento considerable
del tiempo de análisis, ya que se necesitarán varias inyecciones por muestra;
40
y la necesidad de prever el nivel de concentración esperado para poder realizar

las adiciones correctas.
Estrategias de reducción de la cantidad de matriz
Tienen como objetivo evitar el efecto matriz mediante la reducción de

interferentes inyectados en el sistema cromatográfico.
La opción más utilizada es la aplicación de etapas de purificación.

Actualmente, estas etapas suelen consistir en una extracción en fase sólida
(SPE) o extracción en fase sólida dispersiva (d-SPE), habitual en el método
QuEChERS, debido a la gran variedad de adsorbentes disponibles para
obtener una purificación adecuada. Aunque la SPE permite eliminar
interferentes y realizar una preconcentración de los analitos, ésta requiere de
mayores volúmenes de extracto a tratar, así como un aumento del tiempo de
análisis, a diferencia de los métodos QuEChERS. Además, el uso de técnicas
de purificación en métodos multianalito presenta limitaciones debido a las
posibles pérdidas de compuestos dada la variedad de propiedades físico-
químicas de los mismos.
Por ello, una de las soluciones más sencillas que se adoptan en métodos
multianalito es la dilución de los extractos. Con esta aproximación se
consigue, sin producirse la pérdida de analitos, reducir la carga de matriz
inyectada al sistema cromatográfico. Sin embargo, esta alternativa está muy
condicionada por la sensibilidad del método.
41
En la presente Tesis Doctoral se han utilizado técnicas basadas en

cromatografía acoplada a espectrometría de masas para la determinación de
compuestos endógenos y exógenos en matrices de acuicultura.
En el capítulo 2 se ha desarrollado metodología cuantitativa y cualitativa para

el análisis de contaminantes orgánicos. El artículo científico I se ha centrado
en el screening, utilizando UHPLC y GC con fuentes API acopladas al
analizador (Q)TOF. A partir de los resultados obtenidos, se desarrolló un
método multiresiduo basado en GC-APCI-QqQ MS/MS, focalizado en la
determinación de PAHs, que se detalla en el artículo científico II.
En el capítulo 3, se detalla el método desarrollado para la determinación de

chrlopyrifos-methyl, pirimiphos-methyl y sus principales metabolitos
mediante UHPLC-QqQ MS/MS, en muestras de pez cebra y de tejidos de
salmón. El objetivo final del artículo científico III es contar con un método
avanzado para determinar la cinética de acumulación y eliminación de los
plaguicidas de estudio y sus metabolitos principales en el tejido de los peces.
Por último, en el capítulo 4, compuesto por el artículo IV, se indaga en las

características de la fuente APCI como alternativa para la caracterización de
VLC-PUFA.
Los métodos analíticos presentados se han desarrollado con el fin de ampliar

el conocimiento sobre un campo de creciente interés, como es la acuicultura.
Además, a lo largo de este trabajo, se corrobora el potencial de la fuente APCI
en GC para dar solución a problemas concretos desde un punto de vista
espectrométrico.
42
1.7. Referencias
1. United Nations, Department of Economic and Social Affairs,

Population Division (2017) World Population Prospects: The 2017
revision
2. FAO (2018) El estado mundial de la pesca y la acuicultura 2018.

Cumplir los objetivos de desarrollo sostenible. Roma. Licencia: CC
BY-NC-SA 3.0 IGO
3. Shepherd CJ, Jackson AJ (2013) Global fishmeal and fish-oil supply:

Inputs, outputs and markets. J Fish Biol 83:1046–1066. doi:
10.1111/jfb.12224
4. Monroig Ó, Navarro JC, Tocher DR (2011) Long-Chain

Polyunsaturated Fatty Acids in fish: Recent advances on desaturases
and elongases involved in their biosynthesis. Av en Nutr Acuícola XI -
Memorias del Décimo Prim Simp Int Nutr Acuícola, 23-25 Noviembre,
San Nicolás los Garza, N L, México 257–283
5. Janssen CIF, Kiliaan AJ (2014) Long-chain polyunsaturated fatty acids

(LCPUFA) from genesis to senescence: The influence of LCPUFA on
neural development, aging, and neurodegeneration. Prog Lipid Res
53:1–17. doi: 10.1016/j.plipres.2013.10.002
6. Kawashima A, Watanabe S, Iwakiri R, Honda K (2009) Removal of

dioxins and dioxin-like PCBs from fish oil by countercurrent
supercritical CO2extraction and activated carbon treatment.
Chemosphere 75:788–794. doi: 10.1016/j.chemosphere.2008.12.057
43
7. Maes J, De Meulenaer B, Van Heerswynghels P, De Greyt W, Eppe G,

De Pauw E, Huyghebaert A (2005) Removal of dioxins and PCB from
fish oil by activated carbon and its influence on the nutritional quality
of the oil. JAOCS, J Am Oil Chem Soc 82:593–597. doi:
10.1007/s11746-005-1114-1
8. Oterhals Å, Solvang M, Nortvedt R, Berntssen MHG (2007)

Optimization of activated carbon-based decontamination of fish oil by
response surface methodology. Eur J Lipid Sci Technol 109:691–705.
doi: 10.1002/ejlt.200700083
9. Berntssen MHG, Olsvik PA, Torstensen BE, Julshamn K, Midtun T,

Goksøyr A, Johansen J, Sigholt T, Joerum N, Jakobsen J V., Lundebye
AK, Lock EJ (2010) Reducing persistent organic pollutants while
maintaining long chain omega-3 fatty acid in farmed Atlantic salmon
using decontaminated fish oils for an entire production cycle.
10. Nácher-Mestre J, Serrano R, Benedito-Palos L, Navarro JC, López FJ,

Pérez-Sánchez J (2009) Effects of fish oil replacement and re-feeding
on the bioaccumulation of organochlorine compounds in gilthead sea
bream (Sparus aurata L.) of market size. Chemosphere 76:811–817.
doi: 10.1016/j.chemosphere.2009.04.046
11. Berntssen MHG, Julshamn K, Lundebye A-K (2010) Chemical

contaminants in aquafeeds and Atlantic salmon (Salmo salar) following
the use of traditional- versus alternative feed ingredients. Chemosphere
78:637–646. doi: 10.1016/j.chemosphere.2009.12.021
44
12. Nácher-Mestre J, Serrano R, Portolés T, Berntssen MHG, Pérez-

Sánchez J, Hernández F (2014) Screening of pesticides and polycyclic
aromatic hydrocarbons in feeds and fish tissues by gas chromatography
coupled to high-resolution mass spectrometry using atmospheric
pressure chemical ionization. J Agric Food Chem 62:2165–2174. doi:
10.1021/jf405366n

Kaushik S, Pérez-Sánchez J (2010) Bioaccumulation of polycyclic
aromatic hydrocarbons in gilthead sea bream (Sparus aurata L.)
exposed to long term feeding trials with different experimental diets.
Arch Environ Contam Toxicol 59:137–46. doi: 10.1007/s00244-009-
9445-1
14. (2002) Reglamento (CE) No 178/2002 del parlamento y del consejo de

28 de enero de 2002 por el que se establecen los principios y los
requisitos generales de la legislación alimentaria.
15. Council of Europe (1976) European Convention for the Protection of

Animals kept for Farming Purposes. Strasbourg
16. European Commission (2017) Analytical quality control and method

validation procedures for pesticide residues analysis in food and feed;
document no. SANTE/11813/2017
17. Hernández F, Ibáñez M, Portolés T, Cervera MI, Sancho J V., López

FJ (2014) Advancing Towards Universal Screening for Organic
Pollutants in Waters. J Hazard Mater 282:86–95. doi: 10.1016/j.jha
zmat.2014.08.006
45
18. Poole CF, Poole SK (1991) Chromatography today. Elsevier
19. McNair HM, Miller JM (1998) Basic gas chromatography. Wiley-

Interscience
20. Grob RL, Barry EF (2004) Modern practice of gas chromatography, 4th
ed. Wiley-Interscience
21. Godula M, Hajšlová J, Alterová K (1999) Pulsed splitless injection and

the extent of matrix effects in the analysis of pesticides. HRC J High
Resolut Chromatogr 22:395–402. doi: 10.1002/(SICI)1521-4168(1999
0701)22:7<395::AID-JHRC395>3.0.CO;2-O
22. Hernández F, Sancho J V, Ibáñez M, Abad E, Portolés T, Mattioli L

(2012) Current use of high-resolution mass spectrometry in the
environmental sciences. Anal Bioanal Chem 403:1251–64. doi:
10.1007/s00216-012-5844-7
23. Plaza-Bolaños P, Frenich AG, Vidal JLM (2010) Polycyclic aromatic

hydrocarbons in food and beverages. Analytical methods and trends. J
Chromatogr A 1217:6303–6326. doi: 10.1016/j.chroma.2010.07.079
24. Anastassiades M, Lehotay SJ, Štajnbaher D, Schenck FJ (2003) Fast

and easy multiresidue method employing acetonitrile
extraction/partitioning and “dispersive solid-phase extraction” for the
determination of pesticide residues in produce. J AOAC Int 86:412–
431
25. Payá P, Anastassiades M, Mack D, Sigalova I, Tasdelen B, Oliva J,

Barba A (2007) Analysis of pesticide residues using the Quick Easy
46
Cheap Effective Rugged and Safe (QuEChERS) pesticide multiresidue

method in combination with gas and liquid chromatography and
tandem mass spectrometric detection. Anal Bioanal Chem 389:1697–
1714. doi: 10.1007/s00216-007-1610-7
26. Lehotay SJ, Maštovská K, Lightfield AR (2005) Use of buffering and

other means to improve results of problematic pesticides in a fast and
easy method for residue analysis of fruits and vegetables. J AOAC Int
88:615–629
27. Sapozhnikova Y, Lehotay SJ (2013) Multi-class, multi-residue analysis

of pesticides, polychlorinated biphenyls, polycyclic aromatic
hydrocarbons, polybrominated diphenyl ethers and novel flame
retardants in fish using fast, low-pressure gas chromatography-tandem
mass spectrometry. Anal Chim Acta 758:80–92. doi: 10.1016/j.aca
.2012.10.034
28. Johnson YS (2012) Determination of Polycyclic Aromatic

Hydrocarbons in Edible Seafood by QuEChERS-Based Extraction and
Gas Chromatography-Tandem Mass Spectrometry. J Food Sci 77. doi:
10.1111/j.1750-3841.2012.02758.x
29. Kalachova K, Pulkrabova J, Cajka T, Drabova L, Hajslova J (2012)

Implementation of comprehensive two-dimensional gas
chromatography-time-of-flight mass spectrometry for the simultaneous
determination of halogenated contaminants and polycyclic aromatic
hydrocarbons in fish. Anal Bioanal Chem 403:2813–2824. doi:
10.1007/s00216-012-6095-3
47
30. Kalachova K, Pulkrabova J, Drabova L, Cajka T, Kocourek V,

Hajslova J (2011) Simplified and rapid determination of
polychlorinated biphenyls, polybrominated diphenyl ethers, and
polycyclic aromatic hydrocarbons in fish and shrimps integrated into a
single method. Anal Chim Acta 707:84–91. doi: 10.1016/j.aca.2011
.09.016
31. Łozowicka B, Rutkowska E, Jankowska M (2017) Influence of

QuEChERS modifications on recovery and matrix effect during the
multi-residue pesticide analysis in soil by GC/MS/MS and
GC/ECD/NPD. Environ Sci Pollut Res 24:7124–7138. doi:
10.1007/s11356-016-8334-1
32. Rodríguez-González N, González-Castro M-J, Beceiro-González E,

Muniategui-Lorenzo S (2017) Development of a matrix solid phase
dispersion methodology for the determination of triazine herbicides in
marine sediments. Microchem J 133:137–143. doi: 10.1016/j.
microc.2017.03.022
33. Ramos S, Homem V, Santos L (2019) Development and optimization

of a QuEChERS-GC–MS/MS methodology to analyse ultraviolet-
filters and synthetic musks in sewage sludge. Sci Total Environ
651:2606–2614. doi: 10.1016/j.scitotenv.2018.10.143
34. Tang L, Kebarle P (1993) Dependence of Ion Intensity in Electrospray

Mass Spectrometry on the Concentration of the Analytes in the
Electrosprayed Solution. Anal Chem 65:3654–3668. doi:
10.1021/ac00072a020
48
35. King R, Bonfiglio R, Fernandez-metzler C, Miller-stein C, Olah T

(2000) Mechanistic Investigation of Ionization Suppression in
Electrospray Ionization. 305:
36. Kebarle P, Verkerk UH (2009) Electrospray: from ions in solution to

ions in the gas phase, what we know now. Mass Spectrom Rev 28:898–
917. doi: 10.1002/mas.20247
37. Fujiyoshi T, Ikami T, Sato T, Kikukawa K, Kobayashi M, Ito H,

Yamamoto A (2016) Evaluation of the matrix effect on gas
chromatography - mass spectrometry with carrier gas containing
ethylene glycol as an analyte protectant. J Chromatogr A 1434:136–
141. doi: 10.1016/j.chroma.2015.12.085
38. Maštovská K, Lehotay SJ, Anastassiades M (2005) Combination of

analyte protectants to overcome matrix effects in routine GC analysis
of pesticide residues in food matrixes. Anal Chem 77:8129–8137. doi:
10.1021/ac0515576
39. Cherta L, Portolés T, Beltran J, Pitarch E, Mol JGJ, Hernández F (2013)

Application of gas chromatography-(triple quadrupole) mass
spectrometry with atmospheric pressure chemical ionization for the
determination of multiclass pesticides in fruits and vegetables. J
Chromatogr A 1314:224–40. doi: 10.1016/j.chroma.2013.09.029
49
CAPÍTULO 2
DETERMINACIÓN DE COMPUESTOS
EXÓGENOS MEDIANTE EL USO DE
ESPECTROMETRÍA DE MASAS CON
ANALIZADOR DE CUADRUPOLO-
TIEMPO DE VUELO Y TRIPLE
CUADRUPOLO
CAPÍTULO 2: DETERMINACIÓN DE COMPUESTOS EXÓGENOS
ANALIZADOR DE CUADRUPOLO-TIEMPO DE VUELO Y TRIPLE
CUADRUPOLO
2.1. Introducción
2.2. Artículo científico I
“Comprehensive strategy for pesticide residue analysis through the production

cycle of gilthead sea bream and Atlantic salmon”
Chemosphere, 179, 242-253, 2017.
2.3. Artículo científico II
“Multi-class determination of undesirables in aquaculture samples by gas

chemical ionization: A novel approach for polycyclic aromatic hydrocarbons”
Talanta, 172, 109–119, 2017.
2.4. Discusión de los resultados obtenidos
2.5. Referencias
53
Capítulo 2 Determinación de compuestos exógenos
2.1. Introducción
Como se ha expuesto en la introducción general de la tesis, la industria

acuícola se enfrenta a un gran reto en términos de rentabilidad y sostenibilidad,
planteando un nuevo escenario en el ámbito de la seguridad alimentaria.
Estudios previos en proyectos de la UE como “AQUAMAX” han demostrado

que la sustitución de aceites y harinas de pescado por ingredientes vegetales
en piensos de acuicultura, junto con técnicas comerciales de
descontaminación, han supuesto un cambio en el perfil de los contaminantes
presentes en los mismos. Por una parte, la sustitución de los ingredientes
tradicionales se ha traducido en la disminución de contaminantes orgánicos
persistentes (POPs), como dioxinas y PCBs, en los peces de cultivo [1, 2]. Sin
embargo, también ha derivado en el incremento de los niveles de
contaminantes relacionados con el tratamiento térmico de los ingredientes
vegetales, como los hidrocarburos policíclicos aromáticos (PAHs), y la
introducción de otros asociados a la agricultura, como plaguicidas [3–5].
Ante la relevancia del cambio composicional de los piensos de acuicultura en

el ámbito socioeconómico y medioambiental [6], el desarrollo de métodos
analíticos para la determinación de contaminantes orgánicos se ha convertido
en una herramienta esencial para obtener datos fiables que contribuyan a
mejorar el proceso de producción y a asegurar la calidad del producto final al
consumidor. Así pues, este nuevo escenario en el campo de la acuicultura
plantea un desafío analítico dado el elevado número de potenciales
contaminantes, la diversidad de propiedades físico-químicas de los mismos,
la complejidad y variedad de matrices a analizar y los bajos niveles de
concentración a los que se pueden encontrar. Por ello, como se ha comentado
55
en el capítulo 1, resulta imprescindible establecer estrategias analíticas

basadas en el acoplamiento cromatografía-espectrometría de masas que
permitan mejorar el alcance de los análisis.
Tradicionalmente, la determinación de contaminantes orgánicos se ha llevado

a cabo mediante métodos pre-target basados en GC-MS/MS y LC-MS/MS
[7–10]. Sin embargo, estos métodos únicamente permiten el análisis de un
número preestablecido de compuestos, lo que supone un inconveniente a la
hora de ampliar el alcance del análisis. Por ello, en el ámbito de la
investigación de contaminantes orgánicos en el medio ambiente, ha crecido el
interés por los métodos de screening basados en la espectrometría de masas
de alta resolución (HRMS), dada la posibilidad de obtener rápidamente una
respuesta cualitativa sobre la presencia o ausencia de un mayor número de
compuestos de interés para su posterior confirmación y/o cuantificación. Estos
últimos métodos deben ser rápidos, fácilmente actualizables, con un
tratamiento de muestra lo más genérico posible y lo suficientemente sensibles
para poder controlar el cumplimiento de los MRLs. Y, además, deben ser
fiables, por lo que, al igual que los métodos pre-target cuantitativos, también
han de someterse a un proceso de validación.
El proceso de validación de métodos de screening basados en HRMS pretende

establecer el nivel de concentración mínimo (o límite de detección de
screening, SDL), mediante el análisis de muestras fortificadas a diferentes
niveles de concentración, para el cual el método detecta los compuestos de
interés con un grado determinado de confianza. Sin embargo, los criterios de
identificación que establecen dichos límites son objeto de debate entre la
comunidad científica. Por una parte, la directiva 2002/657/CE de la Comisión
Europea fundamenta la identificación en base a puntos de identificación (IPs),
56
es decir, cuantifica el grado de confianza de la identificación teniendo en

cuenta el número de iones diagnóstico que cumplen unas tolerancias
establecidas para la relación q/Q, así como el tipo de analizador de masas
utilizado [11]. Por otra parte, la guía SANTE/11813/2017 fundamenta la
identificación en base a la presencia y la exactitud de masa del ion precursor
y de, al menos, un ion fragmento [12]. Pese a que ambas guías tienen en cuenta
la resolución de los analizadores utilizados, únicamente la segunda de ellas se
centra en la exactitud de masa, valorando objetivamente la selectividad
instrumental. Por este motivo, los criterios de identificación en métodos de
screening seguidos en esta tesis han adoptado aquellos de la guía
SANTE/11813/2017, si bien se ha adaptado la nomenclatura a la utilizada por
Hernández et al [13] en función del uso o no de patrones de referencia para
confirmar la identidad de compuestos mediante el tiempo de retención
cromatográfico (ver figura 1).
Por definición, la identificación implica asegurar que la respuesta obtenida se

corresponde inequívocamente a un analito concreto. En caso contrario, es
habitual referirse a la detección o identificación tentativa de un compuesto,
requiriendo un segundo análisis para su identificación. Sin embargo, la guía
SANTE/11813/2017 también la define como el “resultado cualitativo de un
método capaz de proporcionar información estructural que cumple con
criterios aceptables para el propósito del análisis” [12]; y aclara en el apartado
D8 que aunque “los diferentes tipos de analizadores de masas y modos de
adquisición proporcionan diferentes grados de selectividad - lo que se
relaciona con la confianza en la identificación - los requisitos para la
identificación deben considerarse como criterios de orientación, no como
criterios absolutos para demostrar la presencia o ausencia de un analito” [12].
Tal es así que esta guía únicamente requiere un porcentaje menor del 5% de
57
falsos negativos en la validación de métodos de screening pero no contempla

un porcentaje mínimo de falsos positivos, recayendo la confirmación de los
mismos en un método de identificación, es decir, en el reanálisis de la muestra
mediante una metodología independiente a la utilizada [14]. Por esta razón,
posterior al screening, es habitual y recomendable desarrollar, validar y
aplicar métodos basados en analizador de QqQ, para la confirmación y
cuantificación de los compuestos identificados reportados.
Figura 1: Esquema general del procesamiento de datos en métodos de

screening.
58
En el presente capítulo se discute el desarrollo, validación y aplicación de una

estrategia analítica integral para la identificación, confirmación y
cuantificación de PAHs y pesticidas en muestras de acuicultura. La estrategia
propuesta en este capítulo se presenta como una herramienta de evaluación de
riesgos disponible para la industria acuícola con el objetivo de ampliar el
conocimiento sobre los nuevos ingredientes introducidos en los piensos, los
mismos piensos y las partes comestibles de las especies cultivadas.
En primer lugar, y como objetivo principal de artículo científico I, se valida

un método de screening de pesticidas para el análisis de más de veinte matrices
de acuicultura diferentes mediante el uso combinado de los acoplamientos LC-
QTOF MS y GC-APCI-QTOF MS. Tras la validación, se aplica a las muestras
siguiendo las aproximaciones target screening y suspect screening reportando
la identificación de dieciséis pesticidas diferentes. Con el propósito de
confirmar los compuestos reportados, cuantificarlos y ampliar el alcance del
análisis con la determinación de PAHs i PCBs, en el artículo científico II se
desarrolla, valida y aplica un método basado únicamente en GC-APCI-
MS/MS con analizador QqQ. Además, en este artículo se evalúa el potencial
de la técnica GC-APCI-MS/MS para realizar la cuantificación de matrices
complejas mediante el uso de calibrado en solvente.
Todos los trabajos realizados en este capítulo se han desarrollado en el marco

del proyecto europeo ARRAINA (Advanced Research Initiatives for Nutrition
and Aquaculture).
59
2.2. Artículo científico I
T. Portolés et al., Chemosphere 179 (2017) 242-253 61

Chemosphere 179 (2017) 242-253
Comprehensive strategy for pesticide residue analysis through the

production cycle of gilthead sea bream and Atlantic salmon
T. Portolés a, M. Ibáñez a, B. Garlito a, J. Nácher-Mestre a, b, 2, V. Karalazos c,

J. Silva d, Alm e, R. Serrano a, J. Pérez-Sánchez b, 1, *, F. Hernández a, 1, *,
M.H.G. Berntssen f, 1, *
a
Research Institute for Pesticides and Water (IUPA), Avda. Sos Baynat, s/n,
University Jaume I, 12071 Castellón, Spain
b
Institute of Aquaculture of Torre la Sal (IATS, CSIC), 12595 Ribera de
Cabanes, Castellón, Spain
c
BioMar R&D, Grangemouth, FK3 8UL, UK
d
BioMar AS, N-7010 Trondheim, Norway
e
European Fat Processors and Renderers Association (EFPRA), Boulevard
Baudouin, 1518, 4th Floor, BE - 1000, Brussels, Belgium
f
National Institute of Nutrition and Seafood Research, PO Box 2029
Nordnes, N-5817 Bergen, Norway
Abstract
Plant ingredients and processed animal proteins are alternative feedstuffs for
fish feeds in aquaculture. However, their use can introduce contaminants like
pesticides that are not previously associated with marine Atlantic salmon and
gilthead sea bream farming. This study covers the screening of around 800
pesticides by gas chromatography (GC) and liquid chromatography (LC)
coupled to high resolution time-of-flight mass spectrometry in matrices
throughout the entire marine food production chain. Prior to analysis of real-
62 T. Portolés et al., Chemosphere 179 (2017) 242-253

world samples, the screening methodology was validated for 252 pesticides to
establish the screening detection limit. This was 0.01 mg·kg-1 for 113
pesticides (45%), 0.05 mg·kg-1 for 73 pesticides (29%) and >0.05 mg·kg-1 for
66 pesticides (26%). After that, a quantitative methodology based on GC
coupled to tandem mass spectrometry with atmospheric pressure chemical
ionization source (GC-APCI-MS/MS) was optimized for the pesticides found
in the screening. Although several polar pesticides, of which pirimiphos
methyl and chlorpyriphos-methyl were most dominant, were found in plant
material and feeds based on these ingredients, none of them were observed in
fillets of Atlantic salmon and gilthead sea bream fed on these feeds.
Highlights
► Complementary use of GC and UHPLC coupled to single QTOF MS

platform. ► 800 pesticides were investigated in matrices from the marine food
production chain. ► SDLs and LOQs at 0.01 and 0.005 mg·kg-1 achieved by
GC&LC-QTOF and GC-APCI-MS/MS. ► Plant material showed highest
pesticide indices, with no feed-to-fillet transfer. ► Processed animal proteins
had low pesticide indices.
Keywords
Fish, feed, processed animal products, screening, pesticides, quadrupole time-

of-flight.

Graphical abstract

1. Introduction
The availability of wild fishery-derived raw materials is finite and the rapid
and sustained growth rate of global aquaculture have forced the industry to
explore alternative and more sustainable feed ingredients (Tacon and Metian,
2013). Much attention has been paid to plant ingredients and experimental
evidence supports a successful replacement of marine feedstuffs at relatively
high levels in most carnivorous farmed fish, such as Atlantic salmon (Salmo
salar) and gilthead seabream (Sparus aurata) (Benedito-Palos et al., 2016;
Ytrestøyl et al., 2015). Processed animal proteins (PAPs) from the rendering
industry, re-authorized for use in aquafeeds in the European Union (EU) in
2013 (EC, 2013a), are another valuable source of feed ingredients for farmed
marine fish (Hatlen et al., 2015).
The use of these alternative feed ingredients can introduce contaminants that
were previously not associated with marine salmon and gilthead sea bream
farming. One example are pesticides that are world-wide pre and post-harvest
used on crops or as anti-parasite agent in farming of terrestrial animals. Well
known organochlorine pesticides (OCP) such as DDT and HCB have been
mostly banned for agricultural use and are associated with fish oil (Berntssen
et al., 2010; Friesen et al., 2008; Nácher-Mestre et al., 2009). These OCP
pesticides have been replaced by less persistent and more water soluble
pesticides (Seiber, 2002). EU Maximum Residue Level (MRL) legislation for
non-OCP pesticides comprises most food commodities (EC, 2005), but for
feed ingredients and fish, specific harmonized EU MRLs are not yet
established (EC, 2013b). This emphasizes the need for data on the occurrence
of pesticides in feed ingredients and the edible part of fish farmed on plant-
based feeds. Extensive EU surveillance programmes exist on pesticide

residues in food (EFSA, 2013). Several surveillance studies report on

pesticides in terrestrial animals feed (i.e (Gómez-Pérez et al., 2015)) as well
as potential transfer of the pesticides to edible part of animals such as meat
milk and eggs (Kan and Meijer, 2007; Leeman et al., 2007). For farmed fish,
occurrence and feed-to-fillet transfer data on most (non POPs) pesticides, is
limited (Lovell et al., 1996; Nácher-Mestre et al., 2014).
In addition to the above findings, the different physicochemical characteristics

of pesticides, together with the low concentration levels established by current
legislation and the complexity of the matrices make necessary the use of last
generation analytical techniques. Multi-residue methods (MRM) are applied
with a clear tendency to liquid chromatography-mass spectrometry (LC-MS),
although gas chromatography-MS (GC-MS) is still required to widen the
number of compounds investigated. Thus, the combined use of GC and LC
with tandem MS with last generation triple quadrupole (QqQ) instruments is
one of the best options to get the sensitivity and selectivity required (Golge
and Kabak, 2015; Hernández et al., 2012, 2013). Additionally, accurate-mass
full-spectrum data obtained by high resolution MS opens the possibility to
provide a complete overview of pesticide pollution, and not only those
compounds initially targeted can be investigated. LC coupled to quadrupole-
time of flight (QTOF) MS with electrospray (ESI) source has been widely
explored for the screening of polar contaminants, their metabolites and
transformation products (TPs) (Hernández et al., 2014, 2015a; Nácher-Mestre
et al., 2013). In relation to GC-MS, the soft ionization obtained from
atmospheric pressure chemical ionization (APCI) source in GC-MS
instruments has offered attractive features for screening purposes (Portolés et
al., 2010, 2014). All this, thereby, opens fascinating perspectives in the
analytical field (Hernández et al., 2015b; Pitarch et al., 2016) towards the

screening of thousands of contaminants without standards (Castillo et al.,

2016; Hernández et al., 2015b; Krauss et al., 2010).
The present work is based on our previous research on screening of pesticide

residues (Nácher-Mestre et al., 2013, 2014) in farmed fish. A comprehensive
strategy is presented for screening, identification and quantification of around
800 pesticides in commercially available plant and novel PAP feed ingredients
and their transfer to the edible part of farmed Atlantic salmon and gilthead sea
bream (two main species of the European aquaculture). The screening
considers an initial qualitative validation of 252 pesticides using GC-
(APCI)QTOF MS and UHPLC-(ESI)QTOF MS, followed by a target
quantitative assessment by GC-(APCI)MS/MS QqQ for those pesticides
identified in the qualitative validation.
2. Materials and methods
2.1. Chemicals and reagents
All pesticides and isotopically labelled reference standards were purchased

from Dr. Ehrenstorfer (Augsburg, Germany) and Sigma Aldrich (St Louis,
MO, USA). Isotopically labelled internal standards (ILIS)
Hexachlorobenzene-13C6, Tebuconazole-D6 and 4,4’-DDE-D8 were also
purchased from Dr. Ehrenstorfer. All standards had purities higher than 95%.
Stock standard solutions (around 500 mg·L-1) were prepared in acetone and
were stored at -20ºC. Twenty-two mixtures of pesticide standards (individual
concentration of each pesticide around 50 mg·L-1) were prepared by dilution
of stock individual solutions in acetone. Working standard solutions

containing all pesticides were prepared by dilution of mixtures with acetone

(for sample fortification in GC), hexane (GC injection), methanol (for sample
fortification in LC) and water (instrument injection in LC). Stock standard
solutions were stored at -20ºC, whereas working solutions were stored at 4ºC.
HPLC-grade water was obtained from a MilliQ water purification system

(Millipore Ltd., Bedford, MA, USA). HPLC-grade methanol, HPLC-
supragradient acetonitrile, acetone (pesticide residue analysis quality) and n-
hexane (all ultra-trace quality) were purchased from Scharlab (Barcelona,
Spain). Formic acid (HCOOH, content > 98%), sodium hydroxide (NaOH,
reagent grade) and ammonium acetate (NH4Ac, reagent grade) were supplied
by Scharlab. Anhydrous magnesium sulfate (extra pure) and anhydrous
sodium acetate (reagent grade) were purchased from Scharlab. Leucine
enkephalin (used as lock mass in LC) and heptacosa (for GC calibration) were
purchased from Sigma Aldrich.
QuEChERS commercial clean-up kits were purchased from Teknokroma

(Barcelona, Spain). Each kit contains 50 mg of primary-secondary amine
(PSA), 150 mg of anhydrous magnesium sulfate, and 50 mg of C18, in 2 mL
microcentrifuge tubes for d-SPE.
2.2. Samples
A total of 76 samples were studied in this work as detailed in Table S1. The
list contains ingredients from different origin (plant, terrestrial animals and
marine), and also feeds based on these feed ingredients, as well as fillets of
Atlantic salmon and gilthead seabream reared on these feeds. Atlantic salmon

and gilthead seabream were fed by the produced feeds for 7 and 18 months,
respectively, and fillet samples were taken for analysis at the end of the
exposure trial. The same feeds were provided throughout the feeding trial. The
screening and quantification was performed on feed ingredients, feeds
produced from the same feed ingredients, and fish fillets of fish fed on these
feeds. The feed samples were analyzed at the beginning of the trial and no
stability assessment was made by analyzing the feed during storage.
Commercially available plant and marine feed ingredients were provided by

BioMar (Tech Center, Brande, Denmark) feed producer and PAPs from non-
ruminants were provided by the European Fat Processors and Renderers
Association (EFPRA). All PAPs were produced according the EU regulation
for PAP intended for use as feed-ingredients in animal feed (EC, 2001, EC,
2009). The ingredients selected represent the novelties in fish feed
compositions to reduce the inclusion of fish derivatives. Fish feeds for feeding
trials were based on plant feed ingredients, and not PAPs, as higher levels of
pesticide residues were found in plant feed ingredients (see Section 3). The
feeds were produced by BioMar under commercial aquafeed production
techniques based on high-temperature extrusion processes, which potentially
could affect pesticide residue levels.

70
Capítulo 2
Table S1. Samples analyzed

Ingredients Feed Fish
Terrestrial animal Marine animal
Plant Plant oil origin (PAP) origin Sea bream Salmon Sea bream Salmon
Pea protein (2) Linseed (1) Poultry blood meal (4) Krill meal (1) Experimental (8) Experimental (4) Experimental fillet after 8 months (4) Experimental fillet (4)
Pea (1) Palm (2) Poultry meal (4) Fish meal (3) Experimental fillet after 18 months (4)
Wheat (3) Rapeseed (5) Feather meal (3) Fish oil (2)
Wheat gluten (4) Pork blood meal (3)
Corn gluten (3) Pork meal (3)
Soya protein (4) Pork greaves (2)
Sunflower meal (1)
Rapeseed cake (1)
Number of samples are specified in brackets
T. Portolés et al., Chemosphere 179 (2017) 242-253

Determinación de compuestos exógenos
2.3. Wide scope screening work-flow
The sample procedure applied for pesticides screening and quantification is

illustrated in Fig. 1. Briefly, samples were thawed at room temperature and 5
g were accurately weighed and transferred to centrifuge tubes (50 mL). For
GC-analysis, samples were extracted with acetonitrile (10 mL) and the extract
was subsequently left in a freezer (at least for 2 h to precipitate proteins and
fix lipids to the tube walls). Then, a QuEChERS clean-up step was carried out
prior injection in the GC-system (Nácher-Mestre et al., 2014). In the LC-
screening, similar procedure was followed without any purification or
preconcentration step (Fig. 1). In this case, extraction of the samples was
carried out with acetonitrile/water 80:20 (0.1% formic acid (Nácher-Mestre et
al., 2013).
Pesticides found by the GC&LC-QTOF MS screening were subsequently

confirmed and quantified by GC-(APCI)MS/MS QqQ. Sample treatment was
similar to that applied for GC-screening with two slight variations: i) 1 g
(instead of 5 g) of sample was spiked with isotopically labelled internal
standards and extracted with 2 mL of acetonitrile (instead of 10 mL); ii) just
before injection, 50 µL of the final acetonitrile extract was diluted with 300
µL of acetone and 650 µL of hexane in order to make the solution miscible.

Fig. 1. Scheme of the sample procedure for screening and quantification of

pesticides in aquaculture field.

2.4. Screening validation
For qualitative analysis of GC-amenable compounds, an Agilent 7890A GC

system (Palo Alto, CA, USA) equipped with an Agilent 7693 autosampler was
coupled to the Xevo G2 QTOF (Waters, Manchester, UK), operating in APCI
mode. For qualitative analysis of LC-amenable compounds, a Waters Acquity
UHPLC system (Waters, Milford, MA, USA) was coupled to a XEVO G2
QTOF (Waters, Manchester, UK), with an orthogonal Z-spray-ESI interface
operating in both positive and negative ionization modes.
GC-(APCI)QTOF MS
The GC separation was performed using a fused silica DB-5MS capillary

column with a length of 30 m x 0.25 mm i.d. and a film thickness of 0.25 µm
(J&W Scientific, Folson, CA, USA). The oven temperature was programmed
as follows: 90 ºC (1 min); 5 ºC/min to 315 ºC (2 min). Pulsed splitless (50 psi)
injections of 1 µL of sample extracts were carried out with an injector
temperature of 280 ºC and with a splitless time of 1 min. Helium 99.999%
(Praxair, Valencia, Spain) was used as carrier gas at a constant flow of 2
mL/min. The interface and source temperatures were set to 310 ºC and 150
ºC, respectively. The make-up gas (N2) was set at 320 mL/min flow, the
auxiliary gas at 150 L/h and the cone gas at 16 L/h. The voltage of the
sampling cone was set at 20 V, the voltage of the extraction cone was 4 V, and
the APCI corona pin was fixed at a current 1.7 µA. The ionization process
occurred within an enclosed ion volume, which enabled control over the
protonation/charge transfer processes. A scan time of 0.4 s was selected and
MS data were acquired over an m/z range of 50-650.
For MSE experiments, two acquisition functions with different collision

energies were created and applied simultaneously in each sample injection:

the low energy function (LE), selecting a collision energy of 4 eV, and the
second one, the high energy function (HE), with a collision energy ramp
ranging from 10 to 40 eV. TOF MS resolution was approximately 20,000
(FWHM) at m/z 614.
Two injections were performed for sample: the first one promoting the
formation of the molecular ion, and the second one, promoting the formation
of the protonated molecule. Full-spectrum acquisition data generated at low
and high collision energy (MSE) were processed, using the specialized
application manager ChromaLynxXS (within MassLynx).
Heptacosa was used for the daily mass calibration. Continuous internal
calibration was performed using a background ion coming from the GC-
column bleeding as lock mass (m/z 223.0640).
UHPLC-(ESI)QTOF MS
An Acquity UHPLC BEH C18 1.7 µm particle size analytical column 100×2.1
mm (Waters) at a flow rate of 300 µL/min was employed for chromatographic
separation. Mobile phase consisted of water/methanol gradient both with
0.01% HCOOH. The percentage of organic modifier (B) was changed linearly
as follows: 0 min, 10 %; 14 min, 90 %; 16 min, 90 %; 16.01 min, 10 %; 18
min, 10 %. The column temperature was set at 40 ºC. Hybrid TOF MS
resolution was approximately 20,000 at full width half maximum (FWHM),
at m/z 556.2771. MS data were acquired on the m/z range of 50-1000. A
capillary voltage of 0.7 kV in positive mode and 2.5 kV in negative mode were
used with a cone voltage of 25 V. Collision gas was argon 99.995% (Praxair,
Valencia, Spain). The interface temperature was set to 650 ºC and the source

temperature at 130 ºC.
Similarly to GC-QTOF MS, for MSE experiments two acquisition functions

were created: the low energy function (LE), selecting a collision energy of 4
eV, and the second one, the high energy function (HE), with a collision energy
ramp ranging from 15 to 40 eV.
MS data were acquired in centroid mode and were processed by the

ChromaLynx XS application manager (within MassLynx v 4.1; Waters
Corporation).
Calibrations were conducted daily from m/z 50 to 1000 with a 1:1 mixture of
0.05M NaOH:5% HCOOH diluted (1:25) with acetonitrile:water (80:20), at a
flow rate of 20 µL/min. For automated accurate mass measurement, the lock-
spray probe was used, using as lock mass a solution of leucine enkephalin (10
mg/L) in acetonitrile:water (50:50) at 0.1% HCOOH pumped at 30 µL/min
through the lock-spray needle. The (de)protonated molecule of leucine
enkephalin at m/z 556.2771 in positive mode and m/z 554.2615 in negative
mode was used for recalibrating the mass axis and ensuring a robust accurate
mass measurement along time.
Validation of the two screening methods applied (GC-QTOF MS for 170

pesticides and UHPLC-QTOF MS for 162 pesticides) was performed for
qualitative purposes on the basis of European analytical guidelines (Sanco,
2013). To this aim, at least twenty sample matrices (including different feed
ingredients, feed and fish) were spiked with pesticides at two concentrations,
0.01 and 0.05 mg·kg-1 and, analyzed together with their respective non-spiked
samples (“blanks”) and method blanks to assure absence of contamination
along the procedure. The main parameter evaluated was screening detection

limit (SDL), which was the lowest concentration for which each pesticide was
detected in 95% of the spiked samples tested (e.g. 19 out of 20 samples)
independently of its recovery and precision. The detection of the compound
was made by using the most abundant ion measured at its accurate mass
(typically the molecular ion or (de)protonated molecule) in the LE function.
This implied that at least one m/z ion was observed at the expected retention
time (Rt) (deviation accepted ± 0.2 min, in comparison with the reference
standard) with mass error below 5 ppm.
The GC-QTOF MS qualitative screening had been previously validated for

131 pesticides in twenty aquaculture samples in a previous work (Nácher-
Mestre et al., 2014). In this work, validation was widened with 39 pesticides
in relation to our previous work and was tested for new sample matrices from
the fish growing trials. To this aim, 4 additional matrices (gilthead sea bream
fillet, salmon fillet and two additional fish feed) were spiked with the already
validated 131 pesticides together with the new 39 pesticides. For those 4 new
matrices, a criteria of 4 positives out of 4 analyzed was required to accept the
SDL. This allowed us to check and confirm the SDLs previously established
for the 131 pesticides, as well as establishing provisional SDLs for the
additional 39 pesticides studied. Furthermore, 6 PAP matrices were also
subjected to the same methodology, and spiked at the concentration levels
indicated above. With a requirement of 6 positives out of the 6 samples
analyzed, provisional SDLs were also established.
Regarding LC-QTOF MS screening, the validation has been performed in two

steps as a function of the availability of the samples. Similarly to GC-QTOF
MS, in a first step, the LC screening was qualitatively validated in twenty
different sample matrices (feed ingredients, feed and fish) spiked with 125

pesticides at two concentrations, 0.01 and 0.05 mg·kg-1, and the SDL was
established. The detection was made by using the (de)protonated molecule, so
at least one m/z ion was found at the expected Rt (deviation accepted ± 0.2
min) with mass error below 5 ppm. The LC qualitative screening was widened,
in a second step, with 36 pesticides. Also, 4 additional samples (the same as
in GC-QTOF MS) were spiked with the already validated 125 compounds and
36 more pesticides. This allowed us to confirm the SDLs already established
in the first step and also establish provisional SDLs for the 36 new pesticides.
Additionally, 6 PAPs were also spiked at the concentration levels cited above
and for those samples a criteria of 6 out of 6 was required to establish a
provisional SDL.
2.5. Qualitative screening of aquaculture samples
The overall strategy proposed was applied to the screening of aquaculture

samples from a multidisciplinary European funded project (EU Seventh
Framework Programme ARRAINA Project 288925). Samples analyzed
covered the whole production chain of Atlantic salmon and gilthead sea
bream. For these purposes, plant and marine ingredients as raw materials for
aquafeed compositions from feed producers were all studied (Table S1). Fish
tissues from gilthead sea bream and salmon feeding trials were analyzed in
parallel. After injection of the sample extracts, full-spectrum acquisition data
generated at low and high collision energy (MSE) were processed, using the
specialized application manager ChromaLynxXS (within MassLynx) in
combination with a home-made database containing 465 pesticides for GC and
527 for LC. Around 200 compounds were included in both databases;

therefore, the total number of pesticides searched in the comprehensive

screening was near 800. The screening was applied for those compounds that
were qualitatively validated (thus, reference standards were available) and
also for those other pesticides included in the database, for which reference
standards were not available (suspect screening). The detection of a potential
positive was based on the presence of the (de)protonated molecule/molecular
ion (occasionally adducts), measured at its accurate mass, in the LE function
(for both GC and LC-QTOF). For this purpose, nw-XICs at the m/z of all
compounds included in the database were automatically performed in the LE
function (150 ppm mass window) (Hernández et al., 2015b). Data from HE
function was used to confirm the identity based to the presence of fragment
ions.
When a sample was analyzed, the presence of chromatographic peak at the

expected Rt, together with the evaluation of the accurate-mass fragment ions
and characteristic isotopic ions, allowed the unequivocal confirmation of the
identity of the compound detected when the reference standard was available.
In the case of suspect analysis, the tentative identification was supported by
MS/MS product ions reported in the literature for the suspect compound
(either in exact or nominal mass) and by the compatibility of the fragment ion
with the chemical structure of the candidate. Tentative identification was
finally confirmed by subsequent acquisition of the reference standard, which
was made at a later step. MassFragment software (Waters) was used to
propose compatible structures from accurate mass measurements of the
observed fragment ions.
All compounds detected (only one ion with accurate mass and Rt agreement)
and/or identified (minimum two accurate-mass ions, with Rt and ion ratio

agreement) were included in the GC-(APCI)MS/MS quantitative method

developed. Those compounds from the suspect list that were just tentatively
identified were also included in the target quantitative method.
2.6. Quantitative analyses of aquacultural samples
Quantitative analysis of selected pesticides was performed in a GC system

(Agilent 7890B, Palo Alto, CA, USA) equipped with an autosampler (Agilent
7693) and coupled to QqQ mass spectrometer (Xevo TQ-S, Waters
Corporation, Manchester, UK), operating in APCI mode.
GC-(APCI)MS/MS
The GC separation was performed using a fused silica DB-17MS (50%

phenyl-methylpolysiloxane) capillary column with a length of 30 m x 0.25
mm i.d. and a film thickness of 0.25 µm with a 1 m x 0.25 mm i.d deactivated
post-column (J&W Scientific, Folson, CA, USA). The oven temperature was
programmed as follows: 70 ºC (1 min); 20 ºC/min to 230 ºC; 5 ºC/min to 320
ºC (6 min). The injector was operated in pulsed splitless mode (50 psi),
injecting 1 µL at 280 ºC. Helium was used as carrier gas at a ramped flow
mode programmed as follows: 2 mL·min−1 (23 min); a ramp of 4 mL·min−1 to
6 mL·min−1 (9 min). In the SRM method, automatic dwell time (values
ranging from 3 to 63 ms) was applied in order to obtain 15 points per peak.
The interface and source temperatures were set to 320 ºC and 150 ºC
respectively using N2 as auxiliary gas at 275 L·h−1, as make-up gas at 300
mL·min−1 and as cone gas at 200 L·h−1. The APCI corona discharge pin was
operated at 1.6 µA.

The ionization process occurred within an enclosed ion volume, which

enabled control over the protonation/charge transfer processes. In order to
work under proton transfer conditions, an uncapped vial containing water was
placed in a designed holder into the APCI source door to enhance protonation.
Targetlynx (a module of MassLynx) was used to process the data acquired.
A quantitative method was optimized for those pesticides found in the

screening of samples. Validation of the GC-(APCI)MS/MS method was
performed for 12 out of 16 pesticides detected and/or identified in the samples.
The remaining four compounds -flufenoxuron, tebufenozide, teflubenzuron
and carbofuran-3OH- were not included in the quantification step as they are
not GC-amenable compounds. Accuracy (estimated by means of recovery
experiments) was evaluated by analyzing quality control (QC) samples spiked
at 0.005 and 0.05 mg·kg-1 in 26 “blank” samples (corresponding to 19
different matrices). The limit of quantification (LOQ) was established as the
lowest concentration for which the method showed satisfactory recovery
(between 60 and 140%). Isotopically labelled internal standards were used to
correct matrix effects and potential errors associated to sample manipulation
(Portolés et al., 2017).

3. Results and discussion
3.1. Target and suspect screening of feed ingredients, feed and transfer to
farmed fish
Regarding GC-QTOF screening validation, among the 131 pesticides already

studied, 121 maintained the already established SDL: 0.01 mg·kg-1 (69
pesticides), 0.05 mg·kg-1 (34 pesticides) and >0.05 mg·kg-1 (18 pesticides),
and 4 improved/lowered this value (carbophenothion, chlorfenson,
pendimethalin and tau-fluvalinate) (see Table S2). Only six pesticides did not
pass the new criteria of 4 out of 4 in the new samples and sacrificed the SDL
from 0.01 to >0.05 mg·kg-1 (diphenylamine and leptophos), or from 0.05 to
>0.05 mg·kg-1 (chlorothalonil, heptachlor epoxide A, heptachlor epoxide B
and propoxur). For the 39 additional pesticides studied in four new samples
(including two fish feed and two fish fillets), a provisional SDL was
established as 0.01 mg·kg-1 for 13 pesticides, 0.05 mg·kg-1 for 17 pesticides
and >0.05 mg·kg-1 for 9 pesticides based on the 4 out of 4 criteria. SDLs
obtained for PAPs coincided with those of feed ingredients, feed and fish for
the great majority of analytes (82% of cases, corresponding to 141
compounds).
Regarding LC-QTOF screening validation, for the first 125 pesticides studied,
SDLs were established as 0.01 mg·kg-1 (49 pesticides), 0.05 mg·kg-1 (31
pesticides) and >0.05 mg·kg-1 (25 pesticides) and 18 improved/lowered this
value (see Table S3). Only two pesticides (chlorpropham and parathion-ethyl)
did not pass the new criteria of 4 out of 4 in the new samples and sacrificed
the SDL from 0.05 to >0.05 mg·kg-1. For the new 36 pesticides studied in four
matrices, a tentative/provisional SDL was established as 0.01 mg·kg-1 for 6

pesticides, 0.05 mg·kg-1 for 24 pesticides and >0.05 mg·kg-1 for 6 pesticides.
SDL obtained for PAPs coincided with those for feed ingredients, feed and
fish in 136 cases (84%) and for the rest showed worst results except for
hexaflumuron, butachlor and omethoate.
In general, the evaluation of the SDL for ethoxyquin (ETQ) was troublesome
due to the presence of the analyte at high concentrations in the samples used
for validation.
Table S2. SDLs estimated by GC-APCI-QTOF MS for the pesticides based

on a) 20 different sample matrices (feed ingredients, feed and fish), b) widen
with 4 additional samples (gilthead sea bream fillet, salmon fillet and two
additional fish feed) and c) 6 PAPs samples.
n=20a n=4b n=6c
2-Phenylphenol Diazinon Parathion ethyl
4-4'-Dichlorobenzophenone Dichlofenthion Parathion methyl
Alachlor* Dichloran Pirimicarb
Atrazine Dichlorvos Pirimiphos methyl
Atrazine desethyl Diflufenican Procymidone
Atrazine desisopropyl Endosulfan ether Propham*
Azoxystrobin EPN Propiconazole
Bifenthrin* Ethalfluralin Propyzamide
Bromophos Etofenprox* Pyriproxyfen
Bromophos ethyl Famphur Quinalphos
Bromopropilate Fenarimol Resmethrin
0.01 0.01 0.01
Buprofezin Fenitrothion Simazine
Carfentrazone ethyl Fenoxycarb Terbacil*
Chinomethionat Fenthion Terbumeton
Chlorfenapyr Fipronil Terbumeton desethyl
Chlorfenvinphos Folpet Terbuthylazine desethyl
Chlorpropham* Metalaxyl Terbutryn
Chlorpyrifos ethyl Methoxychlor* Tetradifon
Chlorpyrifos methyl Metolachlor Tolclofos methyl
Coumaphos Metribuzin Triadimefon
Cyanophos Molinate Trifluralin
Cyprodinil Oxyfluorfen Vinclozolin
Iprodione Malathion Methiocarb* 0.01 0.01 0.05
Diphenylamine Leptophos 0.01 - -
alpha-endosulphan Endrin p,p'-DDE
alpha-HCH* Flucythrinate* p,p'-DDT* 0.05 0.05 0.05
beta-endosulfan Fludioxonil Pentachlorobenzene

Table S2. (continuation)

n=20a n=4b n=6c
beta-HCH* gamma-HCH * Phosmet
Cadusafos HCB Tefluthrin
Carbaryl* Heptachlor Thiabendazole
0.05 0.05 0.05
delta-HCH* lamba-Cyhalothrin Permethrin*
Dieldrin Methamidophos
Dimethoate p,p'-DDD*
Carbofuran* Phorate Ethion
Dioxathion* Terbuthylazine Fenamiphos 0.05 0.05 0.01
Oxadixyl Cyanazine Triflumizole
Chlorothalonil Heptachlor epoxide B Propoxur
0.05 - -
Heptachlor epoxide A
Pendimethalin Carbophenothion - 0.05 0.05
Chlorfenson tau-Fluvalinate - 0.05 0.01
Cyfluthrin Deltamethrin Endosulfan sulfate - - 0.05
Ethoxyquin * * *
Aldrin Fenvalerate Mirex* - - -
Azinphos methyl Hexachlorobutadiene Propetamphos
Captafol Imazalil Tolyfluanid*
Captan* Isodrin trans-Chlordane
Cypermethrin Methidathion
Azaconazole Epoxyconazol Propazine 0.01 0.01
Bromuconazole Fluquinconazole Prosulfocarb
Clodinafop-propargyl Isopyrazam Tebuconazole
Cyproconazole Mephosfolan
Dimethachlor Metconazole
Bixafen Iprovalicarb Pyraclostrobin 0.05 0.01
Difenoconazole Mepanipyrim Tepraloxydim
Indoxacarb Metrafenon
Ioxynil-Octanoate Profenofos
Carbetamide Methabenzthiazuron Quintocene 0.05 0.05
Fenpropimorph Procloraz
Isoxaben Pymetrozine
Acequinocyl* Chlordecone Oxydemeton-methyl* - -
Bromoxynyl Flumetrine* Prothioconazole
Carbosulfan Ioxonyl Spiromesifen
* monitored as a fragment ion

bold: charge transfer conditions

Table S3. SDLs estimated by UHPLC-QTOF MS for the pesticides based on

a)
20 different sample matrices (feed ingredients, feed and fish), b) widen with
4 additional samples (gilthead sea bream fillet, salmon fillet and two
additional fish feed) and c) 6 PAPs samples.
n=20a n=4b n=6c
Atrazine Fipronil (-) Promecarb
Bentazone (-) Fludioxonil (-) Pyrifenox
Buprofezin Fluazifop-P-butyl Propanil (-)
Cadusafos Flutriafol Pyridaphenthion
Carbaryl Haloxyfop-2-ethoxyethyl Pyriproxyfen
Carbofuran Imazalil Quinalphos
Clomazone Linuron Simazine
0.01 0.01 0.01
Coumaphos Malaoxon Tebuconazole
Cyprodinil Malathion Terbumeton
Dichlorvos Metolachlor Terbuthylazine
Terbuthylazine desethyl Paclobutrazol Terbutryn
Diazinon Pirimicarb Thiobencarb
Dimethomorph Pirimiphos-methyl Triflumizole
Fenamiphos
Dicrotophos Isoproturon Teflubenzuron (-)
Fluazinam (-) Metalaxyl Tetraconazole 0.01 0.01 0.05
Haloxyfop-methyl Propiconazole
Azinphos-methyl 0.01 0.01 -
Azoxystrobin Quizalofop-ethyl Triadimenol
0.01 0.05 0.05
Chlorfenvinphos
Bensulide Spiroxamine Thiodicarb
0.01 0.05 -
Fenoxycarb
Atrazine desisopropyl Ethoxyquin dimer Pendimethalin
Boscalid Fenhexamid Simazine
Carfentrazone ethyl Fenoxaprop Tebufenpyrad
Chlorsulfuron Fenthion Terbacil (-)
Diflubenzuron Flufenoxuron Terbumeton desethyl
0.05 0.05 0.05
Diflufenican Methidathion Thiabendazole
Diuron Methiocarb Thiacloprid
Ethiofencarb Mevinphos Thiophanate-methyl
Ethion Monocrotophos Tolclofos-methyl
Ethofumesate Oxadixyl Tridemorph
Hexaflumuron (-) 0.05 0.05 0.01
Chlorpropham Parathion-ethyl 0.05 - -
Butachlor Omethoate - 0.05 0.05
Acetamiprid Clothianidin Methiocarb sulfoxide
Aldicarb sulfone Ethiofencarb sulfoxide Propamocarb - 0.05 -
Atrazine desethyl Hexythiazox
Ethoxyquin * * *
Simazine 2-hydroxy- Chloridazon Oxamyl
- - -
Terbuthylazine 2-hydroxy- Ethiofencarb sulfone Terbufos


n=20a n=4b n=6c
Aldicarb sulfoxide Fluroxypyr Thiamethoxam
Azinphos-ethyl Imidacloprid Thiram
Benomyl Lufenuron Tolyfluanid - - -
Bifenazate MCPA (-) Trichlorfon
Butocarboxym Methamidophos Triforine
Carbendazim Methiocarb sulfone
- - -
Carbofuran-3-OH Methomyl
Azaconazole Fenpropimorph Mephosfolane
0.01 0.05
Cyproconazole Mepanipyrim Propazine
Aldicarb Ioxynil (-) Prosulfocarb
Bixafen Isoxaben Pyraclostrobin
Bromuconazole Metconazole Tebufenozide
Clodinafop-Propargyl Methabenzthiazuron Tepraloxydim
0.05 0.05
Difenoconazole Metrafenone
Dimetachlor Oxydemeton-methyl
Fluquinconazole Prochloraz
Indoxacarb Profenofos
Carbetamide Isopyrazam Prothioconazole
0.05 -
Iprovalicarb
Benoxacor Carbosulfan Pymethrozine
- -
Bromoxynil (-) Dalapon (-) Trinexapac acid (-)
* monitored as a fragment ion

bold: ESI in negative mode
It is worth to mention that in some cases the same pesticide was included in
both screening methodologies, LC and GC. In those cases, the most favorable
SDL was selected. In this way, Table 1 summarizes the final SDLs established
for feed ingredient, feed and fish for the 252 pesticides studied (removing
duplicities resulting from LC and GC analysis of the same compound).
Overall, SDL values were 0.01 mg·kg-1 for 113 pesticides (45%), 0.05 mg·kg-
1
for 73 pesticides (29%) and a total of 66 pesticides could not be qualitatively
validated (26%) at these levels. For most of them, surely the method was not
sensitive enough for the analyte/matrix tested, and higher analyte
concentrations (>0.05 mg·kg-1) should be tested. In addition, some pesticides
and sample matrices might require specific sample treatments and/or

measurement conditions in order to reach the low concentration levels tested

in this work.
Table 1. SDLs obtained for each pesticide studied by GC-APCI-QTOF MS

and UHPLC-ESI-QTOF MS.
0.01 mg·kg-1 0.05 mg·kg-1 > 0.05 mg·kg-1
2-Phenylphenol Folpet alpha-endosulphan Acetamiprid
4-4'-Dichlorobenzophenone Haloxyfop-2-ethoxyethyl alpha-HCH* Aldicarb sulfone
Alachlor* Haloxyfop-methyl Bensulide Aldicarb sulfoxide
Atrazine desethyl Imazalil beta-endosulfan Aldrin
Atrazine desisopropyl Iprodione beta-HCH* Azinphos-ethyl
Atrazine/Atrazine Isoproturon Boscalid Benomyl
Azinphos-methyl Linuron Chlorsulfuron Bifenazate
Azoxystrobin Malaoxon Cyanazine Butachlor
Bentazone (-) Malathion/Malathion delta-HCH* Butocarboxym
Bifenthrin* Metalaxyl/Metalaxyl Dieldrin Captafol
Bromophos Methiocarb* Diflubenzuron Captan*
Bromophos ethyl Methoxychlor* Dimethoate Carbendazim
Bromopropylate Metolachlor/Metolachlor Dioxathion* Carbofuran-3-OH
Buprofezin/Buprofezin Metribuzin Diuron Carbophenothion
Cadusafos Molinate Endrin Chlorfenson
Carbaryl Oxyfluorfen Ethiofencarb Chloridazon
Carbofuran Paclobutrazol Ethion/Ethion Chlorothalonil
Carfentrazone ethyl Parathion ethyl Ethofumesate Clothianidin
Chinomethionat Parathion methyl Ethoxyquin dimer Cyfluthrin
Chlorfenapyr Pirimicarb/Pirimicarb Fenhexamid Cypermethrin
Chlorfenvinphos Pirimiphos/Pirimiphos Fenoxaprop Deltamethrin
Chlorpropham* Procymidone Flucythrinate* Diphenylamine
Chlorpyrifos ethyl Promecarb Flufenoxuron Endosulfan sulfate
Chlorpyrifos methyl Propanil (-) gamma-HCH * Ethiofencarb sulfone
Clomazone Propham* HCB Ethiofencarb sulfoxide
Coumaphos/Coumaphos Propiconazole/Propiconazole Heptachlor Ethoxyquin/Ethoxyquin
Cyanophos Propyzamide Hexaflumuron (-) Fenvalerate
Cyprodinil/Cyprodinil Pyridaphenthion lambda-Cyhalothrin Fluroxypyr
Diazinon/Diazinon Pyrifenox Methamidophos Heptachlor epoxide A
Dichlofenthion Pyriproxyfen/Pyriproxyfen Methidathion Heptachlor epoxide B
Dichloran Quinalphos/Quinalphos Mevinphos Hexachlorobutadiene
Dichlorvos/Dichlorvos Resmethrin Monocrotophos Hexythiazox
Dicrotophos Simazine/Simazine Oxadixyl/Oxadixyl Imidacloprid
Diflufenican Tebuconazole p,p'-DDD* Isodrin
Dimethomorph Teflubenzuron (-) p,p'-DDE Leptophos
Endosulfan ether Terbacil* p,p'-DDT* Lufenuron
EPN Terbumeton/Terbumeton Pendimethalin MCPA (-)
Ethalfluralin Terbumeton desethyl Pentachlorobenzene Methiocarb sulfone
Etofenprox* Terbuthylazine Permethrin* Methiocarb sulfoxide
Terbuthylazine-desethyl
Famphur Phorate Methomyl
/Terbuthylazine-desethyl
Fenamiphos Terbutryn/Terbutryn Phosmet Mirex*
Fenarimol Tetraconazole Quizalofop-ethyl Omethoate
Fenitrothion Tetradifon Spiroxamine Oxamyl
Fenoxycarb Thiobencarb Tebufenpyrad Propamocarb
Fenthion Tolclofos methyl Tefluthrin Propetamphos
Fipronil/Fipronil (-) Triadimefon Terbacil (-) Propoxur
Thiabendazole
Fluazifop-P-butyl Triflumizole Simazine 2-hydroxy
/Thiabendazole
Fluazinam (-) Trifluralin Thiacloprid tau-Fluvalinate
Fludioxonil* (-) Vinclozolin Thiodicarb Terbufos

Table 1. (continuation)
0.01 mg·kg-1 0.05 mg·kg-1 > 0.05 mg·kg-1
Flutriafol Thiophanate-methyl Terbuthylazine 2-OH
Triadimenol Thiamethoxam
Tridemorph Thiram
Tolyfluanid*/Tolyfluanid
trans-Chlordane
Trichlorfon
Triforine
0.01 mg·kg-1 0.05 mg·kg-1 > 0.05 mg·kg-1
Azaconazole/Azaconazole Aldicarb Acequinocyl
Bromuconazole Bixafen/Bixafen Benoxacor
Clodinafop-propargyl Carbetamide/Carbetamide Bromoxynyl/Bromoxynil (-)
Cyproconazole/Cyproconazole Carbosulfan/Carbosulfan Iprovalicarb/Iprovalicarb
Dimethachlor Indoxacarb/Indoxacarb Chlordecone
Epoxyconazol Ioxynil (-) Dalapon (-)
Fenpropimorph Ioxynil-Octanoate Flumetrine
Fluquinconazole Difenoconazole Oxydemeton-methyl
Isopyrazam Isoxaben/Isoxaben Spiromesifen
Mepanipyrim Methabenzthiazuron/Methabenzthiazuron Trinexapac acid (-)
Mephosfolan/Mephosfolan Metrafenon/Metrafenone
Metconazole Oxydemeton-methyl
Propazine/Propazine Procloraz/Procloraz
Prosulfocarb Profenofos/Profenofos
Tebuconazole Prothioconazole
Pymetrozine
Pyraclostrobin/Pyraclostrobin
Quintocene
Tebufenozide
Tepraloxydim/Tepraloxydim
italic: GC under charge transfer conditions

bold: LC
*: ion fragment
(-): ESI neg
All the samples described in the experimental section, which contain

ingredients from different origin (plant, terrestrial animals and marine), and
also different feed compositions and fish tissues, were analyzed following the
recommended procedure by both GC-(APCI)QTOF MS and UHPLC-
(ESI)QTOF MS. Fig. 2 illustrates the pesticides detected, identified
(confirmed with reference standard) and tentatively identified in the screening
of these aquaculture samples.

Fig. 2. Accumulated % of samples positives to pesticides detected or identified

in the screening of aquaculture samples by GC&LC-QTOF MS.
Pirimiphos methyl, was the compound more frequently identified by both

techniques in most ingredients (68% of plant protein, 75% plant oil ingredients
and 17% of marine ingredients) and in all the feed samples (salmon and sea
bream). However, no residues were found in the fish samples suggesting none
feed to fish fillet transfer. Similarly, chlorpyriphos methyl was detected,
mainly by GC-(APCI)QTOF MS, in plant protein (11%), plant oil ingredients
(25%), gilthead sea bream feed (50%) and salmon feed (25%) but not fish
fillets. Foodborne chlorpyriphos-methyl is readily metabolized and eliminated
by fish, and its relative low biomagnification potential compared to POPs
(Varó et al., 2002) could explain the non-detectable fillet levels in fish that

were fed on low background levels in the present study. In addition, long term
storage at high temperatures could potentially affect the level of chlorpyrifos-
methyl, but not pirimiphos-methyl residues in corn products (White et al.,
1997). In the present trial, no assessment of the pesticide level during storage
was made, which could have contributed to possible absence of detectable
pesticides in the fish fillets of fish fed on the feeds. Earlier surveillance studies
also identified chlorpyriphos methyl, and to a lesser extend pirimiphos methyl,
as some of the most frequent pesticide residues in terrestrial animal feeds
(Gómez-Pérez et al., 2015; Lovell et al., 1996). In contrast to the present study,
chlorpyriphos-methyl was also detected in fish from Taiwan markets and fish
feeds were suggested to be the main source of this compound in farmed fish
(Sun and Chen, 2008).
Other pesticides like the organochlorine pesticide HCB was found by GC-
(APCI)QTOF MS in one marine origin ingredient, which is a well-known
OCP pesticide that behaves as a POP with elevated levels in fish oil obtained
from pelagic fish species (Berntssen et al., 2010). The none-OCP pesticides,
tebuconazole, azoxystrobin, malathion and boscalid were found by UHPLC-
(ESI) QTOF MS in plant-based ingredients (specially in plant oil ingredients).
In contrast to chlorpyriphos-methyl and pirimiphos-methyl, these pesticides
were not found in feed samples. The absence of these pesticides in feed while
present in the plant-based feed ingredients is likely due to the dilution effect
occurred when plant ingredients are mixed with other ingredients such as fish
oil and meal to produce fish feeds, causing levels below SDL. Then,
flufenoxuron, tebufenozide and teflubenzuron were identified (tebufenozide
only detected) by UHPLC-(ESI)QTOF MS in gilthead sea bream feed samples
(among 13 and 38% of the analyzed samples), but not in the feed ingredients
used in these feeds or fillets of seabream fed on these feeds. Ethoxyquin, which

use is currently authorized as a feed ingredient antioxidant supplement, was

identified in all samples by both techniques except plant oil ingredients (75%)
and animal origin ingredients (only in 5%).
All cited compounds had been included in the target screening list, as
reference standards were available for them and had been previously included
in the qualitative screening validation protocol. Oppositely, the fungicide
fluazinam, included in the suspect list, was tentatively identified by UHPLC-
(ESI)QTOF MS in two gilthead sea bream feed samples. Fig. 3 illustrates the
detection and tentative identification of this compound in a gilthead sea bream
feed sample by UHPLC-QTOF MS. The deprotonated molecule of fluazinam
was detected in the LE function in ESI negative mode, with a mass error of
1.9 ppm. As the reference standard was not available, chemical structures for
the most abundant fragment ions were suggested based on their accurate
masses, using the Mass-Fragment software (Waters). In the HE function, 2
fragments (m/z 415.9433 and 397.9768) were observed with chromatographic
peaks at the same Rt, and mass errors lower than 1.2 ppm in relation to the
theoretical predicted exact masses. All structures proposed for the fragments
were compatible with the chemical structure of fluazinam and were in
accordance with the isotopic pattern observed for the chlorine atoms present
in the structure, making the identification even more reliable. Moreover, the
tentative identification was supported by the MS/MS product ions reported in
the literature (Pizzutti et al., 2009). After this careful evaluation process, the
reference standard was finally acquired and injected, allowing the ultimate
confirmation of this compound in the sample.

Fig. 3. Detection and identification of fungicide fluazinam by UHPLC-QTOF

MS in a gilthead sea bream feed sample (the reference standard was not
available at our laboratory in the time of the detection): (a) LE (bottom) and
HE (top) spectra of the compound eluting at 13.96 min. Proposed elemental
compositions for fragment ions; (b) extracted-ion chromatograms (150 ppm
mass width) for protonated molecule in LE function and different fragment
ions in HE function
3.2. Quantitative analysis of identified pesticides and feed-to fillet transfer
QC recoveries were obtained at 0.005 and 0.05 mg·kg-1 in 19 different

matrices, ranging between 60% and 130% for most matrix/analyte
combinations. A LOQ of 0.005 mg·kg-1 was obtained for azoxystrobin,
boscalid, malathion, pirimiphos-methyl, chlorpyriphos-methyl and
ethoxyquin-dimer (ETQ-D) while it was 0.05 mg·kg-1 for diphenylamine,
tebuconazole and imazalil, this being the lower MRL established in the current
legislation for crops that can be used as feed ingredients. Regarding
ethoxyquin, the evaluation of its recovery was not feasible due to the presence
of the analyte at high concentrations in the samples used for validation. For
the remaining two compounds, fluazinam and hexachlorobenzene, the method

was not suitable as they did not present satisfactory results in most of the
matrices.
All compounds reported as identified by QTOF screening were confirmed and

quantified by GC-(APCI)MS/MS. The only exceptions were fluazinam (2
samples), flufenoxuron (1 sample), teflubenzuron (1 sample) and carbofuran-
3OH (1 sample) which were identified by UHPLC-QTOF MS but could not
be included in GC-(APCI)MS/MS quantitative method due to their physico-
chemical characteristics. Additionally, there were another 37 detections in the
screening, for which only one ion measured at accurate mass at expected Rt
was found and therefore their identity could not be confirmed. 22 out of these
37 suspect positives could be confirmed and quantified by QqQ while for rest
seemed to be false detections. The greater sensitivity of GC-MS/MS with QqQ
in comparison to QTOF made it possible to report 47 new positive findings
that had not been detected previously or identified by QTOF (ethoxyquin,
ethoxyquin-dimer, boscalid, azoxystrobin, tebuconazole and imazalil). All of
them were quantified by QqQ at levels below 0.05 mg·kg-1 except for
ethoxyquin and ethoxyquin-dimer whose concentrations exceeded 0.05
mg·kg-1 in most of the salmon fillets analyzed. As regards identification, all
quantified pesticides were identified by the use of three transitions and the
compliance of at least one q/Q ratio.
Table 2 summarizes the pesticide concentrations determined in the analyzed

samples by GC-(APCI)MS/MS. Ethoxyquin and ethoxyquin-dimer were
found in all feed and fish samples. Concentrations were above 0.5 mg·kg-1 in
all feeds, in the range of 0.005-0.5 mg·kg-1 in salmon fillet and above 0.05
mg·kg-1 in gilthead sea bream. ETQ was found at concentration levels above
0.05 mg·kg-1 in all ingredients with the exception of one plant ingredient in

the range of 0.005-0.05 mg·kg-1. ETQ-D was found below 0.005 mg·kg-1
except four plant ingredients in the range of 0.005-0.05 mg·kg-1 and, above
0.005 mg·kg-1 in all marine origin ingredients. Earlier studies also reported the
overall presence of synthetic antioxidants such as ETQ in commercial feed
and ETQ and ETQ-D in farmed fish including Atlantic salmon, halibut, cod,
and rainbow trout, with mean (min.-max.) ETQ feed levels of 10 (1.4-32)
mg·kg-1 and mean (min.-max.) ETQ and ETQ-D levels of 0.06 (0.013-0.17)
and 0.7 (0.29-1.5) mg·kg-1, respectively, analyzed by means of HPLC coupled
to fluorescence detection (Lundebye et al., 2010).
A concentration level around 0.01 mg·kg-1 of fungicides boscalid and

azoxystrobin were found in one feed sample (0.009 mg·kg-1 for both analytes),
one PAP (0.007 and 0.008 mg·kg-1 respectively) and one plant oil (only
boscalid at 0.007 mg·kg-1) although not exceeding its MRL. The
organophosphorous insecticides pirimiphos-methyl and chlorpyriphos-methyl
were found in 66% and 25% of the feed samples in a range of 0.006-0.030
mg·kg-1 and 0.005-0.009 mg·kg-1, respectively. The highest concentration
level of these two OP insecticides was found in a wheat gluten sample at 0.037
mg·kg-1 for chlorpyriphos-methyl and 0.191 mg·kg-1 for pirimiphos-methyl.
Additionally, pirimiphos-methyl was also found in five plant oil and three
plant ingredient samples at concentration levels among 0.005-0.5 mg·kg-1. No
MRL exists for crop partly or exclusively used for feed ingredients (EC,
2013a). Until specific feed ingredient MRLs have been established, existing
EU MRLs for food crop would apply, taking into account an appropriate
processing (EFSA, 2015). As no standard factors are known for the processing
of whole wheat into animal feed graded wheat gluten, no clear MRL can be
set. However the wheat MRLs for chlorpyriphos-methyl and pirimiphos-
methyl are respectively 3.0 and 5.0 mg·kg-1 which is respectively 25 and 135-

fold higher than levels found in present study. Regarding tebuconazole, it was
found in two rapeseed oil samples at concentration around 0.01 mg·kg-1. As
an illustrative example, Fig. 4 shows the GC-(APCI)MS/MS chromatograms
obtained for the quantification and confirmation of boscalid in one salmon feed
(0.009 mg·kg-1), azoxystrobin in a poultry blood meal (0.008 mg·kg-1) and
chlorpyriphos-methyl in wheat gluten (0.037 mg·kg-1). Also, Fig. S1 (a)
shows the GC-(APCI)MS/MS chromatograms obtained for the quantification
and confirmation of pirimiphos-methyl in wheat gluten at concentration level
of 0.191 mg·kg-1. This ingredient is used to prepare a gilthead sea bream feed
shown at Fig. S1 (b) that contains the pirimiphos-methyl at concentration level
of 0.007 mg·kg-1. This feed represents a total replacement of 80% of fish meal
by plant meal and 84% of the fish oil by alternative plant oils. The gilthead
sea bream fish fillet reared on this feed does not shown any trace of pirimiphos
methyl (Fig. S1(c)).
Figure S1. GC-APCI-MS/MS chromatograms obtained for the quantification

and confirmation of pirimiphos-methyl in (a) plant ingredient, (b) feed and (c)
gilthead sea bream fish fillet. Q: Quantification transition; qi: qualification
transition.  q/Q within accepted tolerances.

Fig. 4. GC-(APCI)MS/MS chromatograms obtained for the quantification and

confirmation of boscalid in a) feed (0.009 mg·kg-1); b) azoxystrobin in animal
origin ingredient (0.008 mg·kg-1) and; c) chlorpyriphos-methyl in plant oil
(0.037 mg·kg-1). Q: Quantification transition; qi: qualification transitions. ✓
q/Q within accepted tolerances.

Table 2. Number of samples with quantified values in the mentioned range
96
Feed: sea bream (n=8) Feed: salmon (n=4) Fish: salmon (n=4) Fish: sea bream (n=8)
Capítulo 2
< 0.005 0.005-0.05 0.05-0.5 >0.5 < 0.005 0.005-0.05 0.05-0.5 >0.5 < 0.005 0.005-0.05 0.05-0.5 >0.5 < 0.005 0.005-0.05 0.05-0.5 >0.5
m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg
Diphenylamine 8 4 4 8
HCB 8 4 4 8
Ethoxyquin 8 4 4 7 1
Chlorpyriphos-methyl 6 2 3 1 4 8
Pirimiphos-methyl 4 4 4 4 8
Malathion 8 4 4 8
Imazalil 8 4 4 8
T ebuconazole 8 4 4 8
Ethoxyquin dimer 8 4 2 2 4 4
Boscalid 8 3 1 4 8
Azoxystrobin 8 3 1 4 8
Ingredients: animal origin (n=19) Ingredients: marine origin (n=6) Ingredients: plant oil (n=8) Ingredients: plant (n=19)
< 0.005 0.005-0.05 0.05-0.5 >0.5 < 0.005 0.005-0.05 0.05-0.5 >0.5 < 0.005 0.005-0.05 0.05-0.5 >0.5 < 0.005 0.005-0.05 0.05-0.5 >0.5
m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg m g/kg
Diphenylamine 19 6 7 1 17 2
HCB 19 5 1 8 19
Ethoxyquin 18 1 2 4 5 3 1 16 2
Chlorpyriphos-methyl 19 6 8 18 1
Pirimiphos-methyl 19 6 3 5 15 1 3 3
Malathion 19 6 8 19
Imazalil 19 6 8 19
T ebuconazole 19 6 6 2 19
Ethoxyquin dimer 19 2 2 2 6 2 17 2
Boscalid 18 1 6 7 1 19
T. Portolés et al., Chemosphere 179 (2017) 242-253

Azoxystrobin 18 1 6 8 19
4. Conclusions
The developed strategy faces the “universal” pesticide analysis in aquaculture

field by means of combined use of LC-QTOF MS and GC-QTOF MS for
screening, followed by confirmation and quantification by GC-(APCI)MS/MS
with QqQ, as most pesticide detected in the screening were GC-amenable. The
strategy proposed is among the most comprehensive and informative in the
pesticide analysis context as it covers a large list of pesticides from different
families. The overall strategy is presented as a risk assessment tool available
for the feed industry in order to widen the knowledge of novel and traditional
ingredients, feed and edible parts of consumed animals. The most dominant
polar pesticides found in plant feed ingredients and feed based on these
ingredients were pirimiphos-methyl and chlorpyriphos-metyl. These
pesticides were not found in the fillets of fish fed on these feeds.
Acknowledgments
This work has been funded by the EU Seventh Framework Programme

ARRAINA Project 288925 (Advanced Research Initiatives for Nutrition and
Aquaculture), and the Norwegian Research Council SAFETY-PAP project
(227387) from the Sustainable Innovation in Food and Bio-based Industries
(Bionær) program. Additional funding was obtained from Generalitat
Valenciana (research group of excellence PROMETEOII/2014/085;
PROMETEOII/2014/ 023; ISIC 2012/016). This work has been developed
within the framework of the Research Unit of Marine Ecotoxicology (IATS
(CSIC)-IUPA (UJI)).

References
Benedito-Palos, L., Ballester-Lozano, G.F., Simó, P., Karalazos, V., Ortiz, Á.,
Calduch-Giner, J., Pérez-Sánchez, J., 2016. Lasting effects of butyrate and
low FM/FO diets on growth performance, blood haematology/biochemistry
and molecular growth-related markers in gilthead sea bream (Sparus aurata).
Aquaculture 454, 8-18. http://dx.doi.org/10.1016/j.aquaculture.2015.12.008.
Berntssen, M.H.G., Julshamn, K., Lundebye, A.-K., 2010. Chemical

contaminants in aquafeeds and Atlantic salmon (Salmo salar) following the
use of traditional-versus alternative feed ingredients. Chemosphere 78, 637-
646. http://dx.doi.org/10.1016/j.chemosphere.2009.12.021.
Castillo, N.I., Ibáñez, M., Beltrán, E., Rivera-Monroy, J., Ochoa, J.C., Páez-
Castillo, M., Posada-Buitrago, M.L., Sulyok, M., Hernández, F., 2016.
Identification of mycotoxins by UHPLC-QTOF MS in airborne fungi and
fungi isolated from industrial paper and antique documents from the Archive
of Bogotá. Environ. Res. 144, 130-138. http://dx.doi.org/10.1016/j.envre
s.2015.10.031.
EC, 2001. Regulation (EC) no 999/2001 of the European Parliament and of

the council of 22 May 2001 laying down rules for the prevention, control and
eradication of certain transmissible spongiform encephalopathies. Official J.
Eur. Union L147, 1-38.
EC, 2005. Regulation (EC) NO 396/2005 of the European Parliament and of

the council of 23 February 2005 on maximum residue levels of pesticides in
or on food and feed of plant and animal origin and amending Council Directive
91/414/EEC. Official J. Eur. Union.

EC, 2009. Regulation (EC) No 1069/2009 of the European Parliament and of

the council of 21 October 2009 laying down health rules as regards animal by-
products and derived products not intended for human consumption and
repealing Regulation (EC) No 1774 (Animal by-products Regulation). Official
J. Eur. Union L300, 1-33.
EC, 2013a. Commission regulation (EU) No 56/2013 of 16 January 2013

amending Annexes I and IV to Regulation (EC) No 999/2001 of the European
Parliament and of the Council laying down rules for the prevention, control
and eradication of certain transmissible spongiform encephalopathies. Official
J. Eur. Union 21, 3-16.
EC, 2013b. Commission regulation (EU) No 212/2013 of 11 March 2013

replacing Annex I to regulation (EC) No 396/2005 of the European Parliament
and of the council as regards additions and modifications with respect to the
products covered by that Annex. Official J. Eur. Union.
EFSA, 2013. Scientific report of EFSA. The 2010 European Union report on
pesticide residues in food. EFSA J. 11 (3130).
EFSA, 2015. Reporting Data on Pesticide Residues in Food and Feed

According to Regulation (EC) No 396/2005 (2014 Data Collection) European
Food Safety Authority http://www.efsa.europa.eu/ sites/default/files/scientific
_output/fil (Accessed February 2016).
Friesen, E.N., Ikonomou, M.G., Higgs, D.A., Keng, P.A., Dubetz, C., 2008.
Use of terrestrial based lipids in aquaculture feeds and the effects on flesh
organohalogen and fatty acid concentrations in farmed Atlantic salmon.
Environ. Sci.Technol. 42, 3519-3523. http://dx.doi.org/10.1021/es0714843.

Golge, O., Kabak, B., 2015. Evaluation of QuEChERS sample preparation and
liquid chromatography-triple-quadrupole mass spectrometry method for the
determination of 109 pesticide residues in tomatoes. Food Chem. 176, 319-
322. http://dx.doi.org/10.1016/j.foodchem.2014.12.083.
Gómez-Pérez, M.L., Romero-González, R., Martínez Vidal, J.L., Garrido

Frenich, A., 2015. Analysis of veterinary drug and pesticide residues in animal
feed by high-resolution mass spectrometry: comparison between time-of-
flight and Orbitrap. Food Addit. Contam. - Part A Chem. Anal. control. Expo.
Risk Assess. 32 http://dx.doi.org/10.1080/19440049. 2015.1023742.
Hatlen, B., Jakobsen, J.-V., Crampton, V., Alm, M., Langmyhr, E., Espe, M.,
Hevrøy, E.M., Torstensen, B.E., Liland, N., Waagbø, R., 2015. Growth, feed
utilization and endocrine responses in Atlantic salmon (Salmo salar) fed diets
added poultry by-product meal and blood meal in combination with poultry
oil. Aquac. Nutr. 21, 214-725. http://dx.doi.org/10.1111/anu.12194.
Hernández, F., Ibáñez, M., Portolés, T., 2012. Comprehensive Environmental

Mass Spectrometry. ILM Publications, Glendale (USA).
Hernández, F., Cervera, M.I., Portolés, T., Beltrán, J., Pitarch, E., 2013. The
role of GC-MS/MS with triple quadrupole in pesticide residue analysis in food
and the environment. Anal. Methods 5, 5875-5894. http://dx.doi.org/10.1039/
c3ay41104d.
Hernández, F., Ibáñez, M., Bade, R., Bijlsma, L., Sancho, J.V., 2014.
Investigation of pharmaceuticals and illicit drugs in waters by liquid
chromatography-high-resolution mass spectrometry. TrAC - Trends Anal.
Chem. 63, 140-157. http:// dx.doi.org/10.1016/j.trac.2014.08.003.

Hernández, F., Ibáñez, M., Botero-Coy, A.-M., Bade, R., Bustos-López, M.C.,
Rincón, J., Moncayo, A., Bijlsma, L., 2015a. LC-QTOF MS screening of more
than 1,000 licit and illicit drugs and their metabolites in wastewater and
surface waters from the area of Bogotá, Colombia. Anal. Bioanal. Chem. 407,
6405-6416. http:// dx.doi.org/10.1007/s00216-015-8796-x.
Hernández, F., Ibáñez, M., Portolés, T., Cervera, M.I., Sancho, J.V., López,
F.J., 2015b. Advancing towards universal screening for organic pollutants in
waters. J. Hazard. Mater. 282, 86-95. http://dx.doi.org/10.1016/j.jhazmat
.2014.08.006.
Kan, C.A., Meijer, G.A.L., 2007. The risk of contamination of food with toxic
substances present in animal feed. Anim. Feed Sci. Technol 133, 81-108.
http:// dx.doi.org/10.1016/j.anifeedsci.2006.08.005.
Krauss, M., Singer, H., Hollender, J., 2010. LC-high resolution MS in

environmental analysis: from target screening to the identification of
unknowns. Anal. Bioanal. Chem. 397, 943-951. http://dx.doi.org/10.1007
/s00216-010-3608-9.
Leeman, W.R., Van Den Berg, K.J., Houben, G.F., 2007. Transfer of
chemicals from feed to animal products: the use of transfer factors in risk
assessment. Food Addit. Contam. 24, 1-13. http://dx.doi.org/10.1080/
02652030600815512.
Lovell, R.A., Mcchesney, D.G., Price, W.D., 1996. Organohalogen and

organophosphorus pesticides in mixed feed rations: findings from FDA's
domestic surveillance during fiscal years 1989-1994. J. AOAC Int. 79, 544-
548.

Lundebye, A.-K., Hove, H., Mage, A., Bohne, V.J., Hamre, K., 2010. Levels
of synthetic antioxidants (ethoxyquin, butylated hydroxytoluene and butylated
hydroxyanisole) in fish feed and commercially farmed fish. Food Addit.
Contam. 27, 1652-1657.
Nácher-Mestre, J., Serrano, R., Benedito-Palos, L., Navarro, J.C., López, F.J.,
Pérez-Sánchez, J., 2009. Effects of fish oil replacement and re-feeding on the
bioaccumulation of organochlorine compounds in gilthead sea bream (Sparus
aurata L.) of market size. Chemosphere 76, 811-817.
http://dx.doi.org/10.1016/j.chemosphere.2009.04.046.
Nácher-Mestre, J., Ibáñez, M., Serrano, R., Pérez-Sánchez, J., Hernández, F.,
2013. Qualitative screening of undesirable compounds from feeds to fish by
liquid chromatography coupled to mass spectrometry. J. Agric. Food Chem.
61, 2077-2087. http://dx.doi.org/10.1021/jf304478n.
Nácher-Mestre, J., Serrano, R., Portolés, T., Berntssen, M.H.G., Pérez-

Sánchez, J., Hernández, F., 2014. Screening of pesticides and polycyclic
aromatic hydrocarbons in feeds and fish tissues by gas chromatography
coupled to high-resolution mass spectrometry using atmospheric pressure
chemical ionization. J. Agric. Food Chem. 62, 2165-2174.
http://dx.doi.org/10.1021/jf405366n.
Pitarch, E., Cervera, M.I., Portolés, T., Ibáñez, M., Barreda, M., Renau-
Pruñonosa, A., Morell, I., López, F., Albarrán, F., Hernández, F., 2016.
Comprehensive monitoring of organic micro-pollutants in surface and
groundwater in the surrounding of a solid-waste treatment plant of Castellón,
Spain. Sci. Total Environ. 548-549, 211-220. http://dx.doi.org/10.1016/j.
scitotenv.2015.12.166.

Pizzutti, I.R., de Kok, A., Hiemstra, M., Wickert, C., Prestes, O.D., 2009.
Method validation and comparison of acetonitrile and acetone extraction for
the analysis of 169 pesticides in soya grain by liquid chromatography-tandem
mass spectrometry. J. Chromatogr. A 1216, 4539-4552.
http://dx.doi.org/10.1016/ j.chroma.2009.03.064.
Portolés, T., Sancho, J.V., Hernández, F., Newton, A., Hancock, P., 2010.
Potential of atmospheric pressure chemical ionization source in GC-QTOF
MS for pesticide residue analysis. J. Mass Spectrom. 45, 926-936.
http://dx.doi.org/10.1002/ jms.1784.
Portolés, T., Mol, J.G.J., Sancho, J.V., Hernández, F., 2014. Use of electron
ionization and atmospheric pressure chemical ionization in gas
chromatography coupled to time-of-flight mass spectrometry for screening
and identification of organic pollutants in waters. J. Chromatogr. A 1339, 145-
153. http://dx.doi.org/10.1016/j.chroma.2014.03.001.
Portolés, T., Garlito, B., Nácher-Mestre, J., Berntssen, M.H.G., Pérez-

Sánchez, J., 2017. Multi-class Determination of Undesirables in Aquaculture
Samples by Gas Chromatography/tandem Mass Spectrometry with
Atmospheric Pressure Chemical Ionization: a Novel Approach for Polycyclic
Aromatic Hydrocarbons (Submitted).
Sanco, 2013. SANCO/12571/2013. Guidance Document on Analytical

Quality Control and Validation Procedures for Pesticide Residues Analysis in
Food and Feed.
Seiber, J., 2002. Environmental fate of pesticides. In: Wheeler, W.B. (Ed.),
Pesticides in Agriculture and the Environment. New York.

Sun, F., Chen, H.-S., 2008. Monitoring of pesticide chlorpyrifos residue in

farmed fish: investigation of possible sources. Chemosphere 71, 1866-1869.
http:// dx.doi.org/10.1016/j.chemosphere.2008.01.034.
Tacon, A.G.J., Metian, M., 2013. Fish Matters: importance of Aquatic foods
in human nutrition and global food supply. Rev. Fish. Sci. 21, 22-38.
http://dx.doi.org/ 10.1080/10641262.2012.753405.
Varó, I., Serrano, R., Pitarch, E., Amat, F., López, F.J., Navarro, J.C., 2002.
Bioaccumulation of chlorpyrifos through an experimental food chain: study
of protein HSP70 as biomarker of sublethal stress in fish. Arch. Environ.
Contam. Toxicol. 42, 229-235. http://dx.doi.org/10.1007/s00244-001-0013-6.
White, N.D.G., Jayas, D.S., Demianyk, C.J., 1997. Degradation and biological
impact of chlorpyrifos-methyl on stored wheat and pirimiphos-methyl on
stored maize in Western Canada. J. Stored Prod. Res. 33, 125-135.
http://dx.doi.org/10.1016/ S0022-474X(96)00049-5.
Ytrestøyl, T., Aas, T.S., Åsgård, T., 2015. Utilisation of feed resources in
production of Atlantic salmon (Salmo salar) in Norway. Aquaculture 448,
365-374.http://dx.doi.org/10.1016/j.aquaculture.2015.06.023.

2.3. Artículo científico II
T. Portolés et al., Talanta 172 (2017) 109-119 105

Talanta 172 (2017) 109–119
Multi-class determination of undesirables in aquaculture samples by gas

chemical ionization: A novel approach for polycyclic aromatic
hydrocarbons
T. Portolésa,⁎, B. Garlitoa, Jaime Nácher-Mestreb, M.H.G. Berntssenc,

J. Pérez-Sánchezb
a
Research Institute for Pesticides and Water, University Jaume I, 12071
Castellón, Spain
b
Institute of Aquaculture of Torre la Sal (IATS, CSIC), 12595 Ribera de
Cabanes, Castellón, Spain
c
Abstract
In this work, a method for the analysis of 24 PAHs in 19 different matrices,

including fish tissues, feeds and feed ingredients, has been developed using
gas chromatography coupled to triple quadrupole tandem mass spectrometry
with atmospheric pressure chemical ionization source (GC-APCI-MS/MS).
The method is based on a modification of the unbuffered QuEChERS method,
using freezing as an additional clean-up step and applying a 20-fold dilution
factor to the final extract. The procedure was also tested for 15 pesticides and
7 polychlorinated biphenyl (PCB) congeners in order to widen the scope of
the method.
106 T. Portolés et al., Talanta 172 (2017) 109-119

The excellent sensitivity and selectivity provided by GC-APCI-MS/MS

allowed the dilution of the sample extracts and quantification using calibration
with standards in solvent for all the 19 matrices tested. The developed method
was evaluated at 2, 5 and 50 ng·g−1 spiking levels. LOQs were 2 ng·g−1 for
most compounds, and LODs ranged from 0.5 to 2 ng·g−1.
Analysis of real-world samples revealed the presence of naphthalene,

fluorene, phenanthrene, fluoranthene and pyrene at concentration levels
ranging from 4.8 to 187 ng·g−1. No PCBs, DDTs and pesticides were found in
fillets from salmon and sea bream.
Highlights
► GC-APCI-MS/MS to determine 24 PAHs in 19 different matrices from

aquaculture field. ► Elimination of matrix effect by a 20-fold dilution factor
of the final extract. ► Use of solvent calibration due to the excellent
sensitivity and selectivity gained. ► Use of both M+• and [M+H]+ as an
additional value to the selectivity. ► LOQs of 2 ng·g−1 for most compounds,
and LODs from 0.5 to 2 ng·g−1.
Keywords
Polycyclic aromatic hydrocarbons, atmospheric pressure chemical ionization,

gas chromatography, triple quadrupole, aquaculture, fish feed

Graphical abstract

1. Introduction
Farming of carnivorous marine fish species such as Atlantic salmon (Salmo

salar) and gilthead sea bream (Sparus aurata) has traditionally relied on the
use of marine feed ingredients such as fish oil and meal in their formulated
feeds [1–3]. Limited access to fish meal and fish oil for the rapidly growing
aquaculture industry has led to a need to develop more sustainable aquafeeds
that rely less on marine ingredients from feral fish stocks [4,5]. Vegetable oils
and proteins are the main alternative to marine feed ingredients in aquafeeds
[2,5].
The development of new sustainable plant based feeds for marine fish farming
introduces new challenges concerning contaminants that were previously not
relevant when using traditional marine feed ingredients. Unrefined plant oils
obtained from oilseeds such as soybeans, rapeseeds, olive seeds, and
sunflower seeds are known to contain elevated levels of polyaromatic
hydrocarbons (PAHs) [6–11]. Although fish oil also contains PAHs from
environmental pollution [12], the use of plant oils causes increased PAH levels
in plant-based salmon feeds compared to marine-based feeds [13]. In contrast,
in a study on fish oil replacement in sea bream (Sparus aurata) feeds, plant
oils did not have elevated PAH levels compared to fish oil, and it was
concluded that feed PAH levels did not correlate with the amount of fish or
plant oil used [14]. The genotoxic and carcinogenic “heavy” PAHs (>4–6
rings), such as benzo[a]pyrene (B[a]P), benzo[a]anthracene (B[a]A), chrysene
(Chr), benzo[b]fluoranthene (B[b]F), have received special attention with
regards to food safety [15]. Studies related to plant oil PAH contamination,
however, mainly focus on light (2–4 rings) PAHs such as fluoranthene,
naphthalene, anthracene, phenanthrene, as they are most dominantly present

in unrefined plant oils. These light PAHs are also on the US EPA
(Environmental Protection Agency) list for environmental relevant PAHs but
are mostly not carcinogenic and genotoxic [10,16].
The European union has set an upper limit for 4 PAHs (B[a]P + B[a]A + Chr
+ B[b]F) in food products to protect the consumer safety [15]. No upper limits
for PAH in feed ingredients or animal feeds currently exist. The PAH levels
of raw ingredients for feeds for farmed terrestrial animals such as cow,
chicken, and pig have been reported in only few studies [11] and little is
known on PAH levels in feeds for farmed fin fish species [17,18], especially
for feed ingredients used in novel plant based fish feed [13,14,19].
Traditionally, the determination of PAHs in solid fatty matrices have involved

time-consuming extractions prior to GC-MS with electron ionization (EI)
analysis [20]. However, the sensitivity restrictions of the GC-EI-MS(/MS)
systems [21,22] in addition to the inherent matrix effects make pre-
concentration and time consuming clean-up steps, as gel permeation
chromatography [23] or solid phase extraction [19,24], necessary. Recently,
some methodologies have applied the QuEChERS procedure in aquaculture
samples [25,26], reducing notably the time and cost of sample treatment.
However, despite all efforts to minimize the presence of interferences, the still
remaining matrix effects make necessary to use matrix-matched calibration for
quantification purposes [22]. In the last years, the application of atmospheric
pressure chemical ionization (APCI) source in combination with last
generation triple quadrupole instruments has made possible to notably
increase sensitivity and selectivity of the determination of organic
contaminants in food safety, environment and doping control fields [27–30] in
selected reaction monitoring (SRM) based methods. The improved

performance of GC-APCI-MS/MS is more than welcome in the aquaculture

field, where the analysis of a wide diversity of “difficult” matrices (oils, plant
ingredients, marine ingredients, terrestrial animal ingredients, fish, etc.) with
this technique can be simplified by reducing the matrix content in the final
extract by simple dilution which minimizes the matrix effect and avoids the
need of using matrix-matched calibrations.
In this work, we explore the potential of the new GC-APCI-MS/MS technique

for quantification and confirmation of PAHs, together with some selected
pesticides and PCBs, in very different sample matrices from aquaculture field.
Sensitivity of this new technique has been exploited in order to avoid the time-
consuming and expensive purification steps commonly used in these complex
matrices and with the aim of obtaining a cheap and efficient sample treatment.
The developed methodology has been applied to the quantification of
compounds that were previously found in wide-scope screening of
aquaculture samples by combined use of GC & LC-QTOF MS. Up to our
knowledge this is the first application of GC-APCI-MS/MS with triple
quadrupole to the analysis of PAHs in samples from aquaculture field.
2. Experimental
2.1. Reagents
Benzo[j]fluoranthene (B[j]F), 5-methylchrysene (5-MC), benzo[c]fluorene

(B[c]F), dibenzo[a,e]pyrene (D[a,e]P), dibenzo[a,h]pyrene (D[a,h]P),
dibenzo[a,i]pyrene (D[a,i]P), dibenzo[a,l]pyrene (D[a,l]P), and
cyclopenta[c,d]pyrene (C[c,d]P) individual standard solutions at 10 mg·L−1

(100 mg·L−1 for C[c,d]P), PAH MIX 9 containing 16 PAHs (10 mg·L−1 in
cyclohexane) and PCB mix 3 (100 ng·µL−1 in isooctane) were purchased from
Dr. Ehrenstorfer (Augsburg, Germany). Reference standards diphenylamine,
ethoxyquin, p,p′-DDD, p,p′- DDT, tebuconazole, azoxystrobin, fluazinam and
imazalil supplied by Dr. Ehrenstorfer GmbH (Augsburg, Germany); boscalid
and malathion by Riedel-de Haën (Seelze, Germany); ethoxyquin dimer by
Synthetica AS (Oslo, Norway) and hexachlorobenzene (HCB), pirimiphos-
methyl and chlopyrifos-methyl by Sigma-Aldrich (St Louis, MO, USA) with
a purity 97–99.9% were used for standard preparation (see structures for all
the compounds in Table S1). Stock standard solutions (around 500 mg·L−1)
were prepared in acetone. Two mixtures of pesticide standards (individual
concentration of each pesticide around 50 mg·L−1) were prepared by dilution
of stock individual solutions in acetone. Working standard solutions
containing all compounds were prepared by dilution of mixtures with acetone
(for sample fortification in GC) and hexane (GC injection). Stock standard
solutions and working solutions were stored in a freezer at −20 °C.
Stable isotopic labeled internal standards (SIL-IS) PCB 153-13C12 and PAH
surrogate cocktail containing acenaphthylene-D8, B[a]P-D12, B(g,h,i)Pe-D12,
fluoranthene-D10, naphthalene-D8, phenanthrene-D10 and pyrene-D10 were
purchased from Cambridge Isotope Laboratories (Andover, MA, USA).
Hexachlorobenzene-13C6, tebuconazole-D6 and p,p′-DDE-D8 were also
purchased from Dr. Ehrenstorfer.
HPLC-supragradient acetonitrile, acetone (pesticide residue analysis quality)

and n-hexane (ultra-trace quality) were purchased from Scharlab (Barcelona,
Spain). Anhydrous magnesium sulfate (extra pure) and anhydrous sodium
acetate (reagent grade) were purchased from Scharlab. The QuEChERS

commercial clean-up kits were purchased from Teknokroma (Barcelona,

Spain). Each kit contains 50 mg of primary-secondary amine (PSA), 150 mg
of anhydrous magnesium sulfate, and 50 mg of C18, in 2 mL microcentrifuge
tubes for d-SPE.
Table S1. Structures for all the compounds
Structure Compound Structure Compound
Naphthalene 5-Methylchrysene
Acenaphthylene Benzo[b]fluoranthene
Acenaphthene Benzo[k]fluoranthene
Fluorene Benzo[j]fluoranthene
Phenanthrene Benzo[a]pyrene
Anthracene Indeno[1,2,3-cd]pyrene
Fluoranthene Dibenzo[a,h]anthracene
Pyrene Benzo[g,h,i]perylene
Benzo[c]fluorene Dibenzo[a,l]pyrene
Benzo[a]anthracene Dibenzo[a,e]pyrene
Cyclopenta[c,d]pyrene Dibenzo[a,i]pyrene


Structure Compound Structure Compound
Chrysene Dibenzo[a,h]pyrene
PCB-28 Chlorpyrifos-methyl
PCB-52 Pirimiphos-methyl
PCB-101 Malathion
PCB-118 4,4’-DDE
PCB-153 Imazalil
PCB-138 4,4’-DDD
PCB-180 4,4’-DDT
Diphenylamine Tebuconazole
HCB Ethoxyquin dimer
Ethoxyquin Boscalid
Fluazinam Azoxystrobin

2.2. Samples
The sample material has been described before in detail [31]. A total of 76
samples (from 19 different matrices) were studied in this work. The list
contains ingredients from different origin (plant, terrestrial animals and
marine), and also feeds based on these ingredients (PAPs not included), as
well as fillets of Atlantic salmon and gilthead sea bream reared on these feeds.
Fish individuals were fed for, respectively, 7 and 18 months, and fillet samples
were taken for analysis at the end of the exposure trial and, additionally, for
sea bream at 8 months (commercial size). The same feed compositions were
provided throughout the feeding trial. The quantification was performed on
feed ingredients, feeds produced from the same feed ingredients, and fish
fillets of fish fed on these feeds. The feed samples were analyzed at the
beginning of the trial (additionally after 8 months for sea bream) and no
stability assessment was made by analyzing the feed during storage (below
7°C for salmon and sea bream feeds).
Commercially available plant and marine feed ingredients were provided by

Biomar (Tech Center, Brande, Denmark) feed producer and processed animal
proteins (PAPs) from non-ruminants were provided by the European Fat
Processors and Renderers Association (EFPRA). The ingredients selected
represent the novelties in fish feed compositions to reduce the inclusion of fish
derivatives. Fish feeds for feeding trials were based on plant feed ingredients,
and not PAPs, as higher levels of PAHs were found in plant feed ingredients.
The feeds were produced by Biomar under commercial aquafeed production
techniques based on high-temperature extrusion processes, which potentially
could affect pesticide residue levels.

2.3. Sample treatment
Samples were thawed at room temperature and 1 g was accurately weighed

and transferred to 15 mL falcon tubes and spiked with 0.2 mL of SIL-IS
solution of 1 µg·mL−1. After 30 min, 2 mL of acetonitrile were added and the
tube was vigorously shaken by vortex for 30 s. Then, 0.8 g of MgSO4 were
added and the tube was immediately shaken for 30 s. Subsequently, the tube
was centrifuged at 1893 rcf·g for 5 min and the upper layer of the extract was
transferred to a 2 mL eppendorf tube and stored for at least two hours in a
freezer to precipitate proteins and fix lipids to the tube walls (freezing
cleanup). Then, a QuEChERS clean-up step was carried out prior injection in
the GC-system [25]. Briefly, 1 mL of the extract was carefully transferred to
the cleanup QuEChERS vial (50 mg of PSA + 150 mg of MgSO4 + 50 mg of
C18), and it was vigorously shaken for 30 s and centrifuged at 12,557 rcf·g for
5 min. Then, 50 µL of the final acetonitrile extract was diluted with 300 µL of
acetone and 650 µL of hexane in order to make the solution miscible (20-fold
dilution factor).
2.4. Instrumentation
2.4.1. GC-APCI-MS/MS
Data were acquired using a GC system (Agilent 7890A, Palo Alto, CA, USA)
equipped with an autosampler (Agilent 7693) and coupled to a triple
quadrupole (QqQ) mass spectrometer (Xevo TQ-S, Waters Corporation,
Manchester, UK), operating in positive APCI mode. The GC separation was
performed using a fused silica DB-17MS (50% phenyl-methylpolysiloxane)

capillary column with a length of 30 m × 0.25 mm i.d. and a film thickness

of 0.25 µm with a 1 m × 0.25 mm i.d deactivated post-column (J & W
Scientific, Folson, CA, USA). The oven temperature was programmed as
follows: 70°C (1 min); 20°C·min−1 to 230°C; 5 °C·min−1 to 320°C (6 min).
The injector was operated in pulsed splitless mode (50 psi), injecting 1 µL
at 280°C. Helium was used as carrier gas at a ramped flow mode
programmed as follows: 2 mL·min−1 (23 min); a ramp of 4 mL·min−1 to 6
mL·min−1 (9 min). In the SRM method, automatic dwell time (values
ranging from 3 to 63 ms) was applied in order to obtain 15 points per peak.
The interface and source temperatures were set to 320°C and 150°C
respectively using N2 as auxiliary gas at 275 L·h−1, as make-up gas at 300
mL·min−1 and as cone gas at 200 L·h−1. The APCI corona discharge pin
was operated at 1.6 µA. The ionization process occurred within an enclosed
ion volume, which enabled control over the protonation/charge transfer
processes. In order to work under proton transfer conditions, an uncapped
vial containing 1.25 mL of water was placed in a designed holder into the
APCI source door to enhance protonation. Targetlynx (a module of
MassLynx) was used to handle and process the acquired data.
2.5. Study of matrix effects
Nine out of nineteen different matrices (fish fillets, feed and PAPs) were
selected for the study of matrix effects and different dilution factors were
tested (crude extract, 5-fold, 10-fold and 20-fold dilution). To this aim, matrix-
matched calibrations were prepared for each sample matrix by taking between
25 and 500 µL of the “blank” sample extract (for crude extract, 500 µL were

evaporated to dryness) and adding 50 µL (125 µL for the 25 ng·mL−1 level)

of the corresponding standard in hexane (between 10 and 100 ng·mL−1),
resulting in final analyte concentrations between 1 and 25 ng·mL−1. The final
volume of each calibration point was 500 µL in all cases (30% acetone, 5%
acetonitrile and 65% hexane). Matrix effect was evaluated by comparing the
calibration graphs obtained with standards in solvent and in matrix (at the
different dilution factors) [32]. The accepted relative error between the slopes
of calibration in solvent and matrix-matched calibration was ± 25%. Once the
optimum dilution factor was selected, the absence of matrix effect was also
tested for the remaining 10 matrices.
2.6. Validation and recovery experiments
Eleven SIL-IS were added at the initial stage of the procedure as quality
control (surrogates) in order to correct for possible losses during the overall
procedure and instrumental deviations. Linearity of relative response of
analytes was established by analyzing standard solutions injected in triplicate
in the range of 0.01–25 ng·mL−1 for all compounds except for C[c,d]P (0.005–
12.5 ng·mL−1). Linearity was assumed when the correlation coefficient (R2)
was higher than 0.99 with residuals lower than 30%.
A full validation of the method was carried out for two sample matrices,
salmon and sea bream fillets. Accuracy (estimated by means of recovery
experiments) was evaluated by analyzing six replicates spiked at 2 ng·g−1 (1
ng·g−1 for C[c,d]P) and 50 ng·g−1 (25 ng·g−1 for C[c,d]P) for the two matrices
tested. Precision was expressed as repeatability in terms of relative standard
deviation (RSD, %) (n=6) calculated for each fortification level. The limit of

quantification (LOQ) objective, for salmon and sea bream fillets, was defined
as the lowest concentration level validated following SANTE/11945/2015
guide criteria [33] (recovery 70–120% and RSDs below 20%). For those
compounds that could not be validated due to the high concentration in the
“blank” sample a statistically calculated LOQ was estimated as the analyte
concentration that produced a peak signal of ten times the background noise
in the chromatogram at the lowest fortification level tested for each compound.
The limit of detection (LOD) was estimated for all compounds considering a
signal-to-noise (S/N) ratio of three.
The methodology was validated for the remaining 17 sample matrices by

analysis of quality control (QC) samples, i.e. “blank” samples spiked at 5 and
50 ng·g−1 injected in each batch. A total of 48 QCs prepared from 24 “blank”
samples (corresponding to 17 different matrices) were analyzed. Recoveries
between 60% and 140% for each individual sample were considered
satisfactory following the acceptance criteria for routine recoveries according
to SANTE/11945/ 2015 guide criteria [33].
The selectivity of the method was assured by choosing selective SRM

transitions using M+• or [M+H]+ as precursor ion, when possible. Specificity
was evaluated by verification of the absence of interfering peaks at the
retention time of each compound in blank samples for the acquired SRM
transitions.
The qi/Q ratio, defined as the ratio between the intensity of the confirmation
ion (qi) and the quantification ion (Q), was used to confirm peak identity in
real and spiked samples. The experimental qi/ Q value for each compound was
calculated as the mean value obtained from three standard solutions injected

in triplicate (RSD below 15% in all cases) Confirmation of analytes detected

in samples was considered positive when the qi/Q ratio was within acceptable
tolerances (30% of the experimental qi/Q value calculated from standards)
according to SANTE/11945/2015 guide criteria [33].
3.1. Ionization and fragmentation behavior of PAHs in GC-APCI
In this work, the ionization behavior of the APCI interface was tested using
PAH standards in solvent. Two mechanisms of ionization were
simultaneously observed: i) charge transfer in which the nitrogen plasma
created by the corona discharge needle promotes the formation of M+•, and ii)
proton transfer, where the presence of water vapor traces in the source favors
the formation of the [M+H]+ ion.
The PAHs studied showed a mixture of two ions, M+• and [M+H]+, as base
peak of the spectrum. In the case of acenaphthene, fluorene, phenanthrene,
anthracene, fluoranthene, pyrene, B[c]F and C[c,d]P, the [M+H]+ intensity
was slightly higher than M+• (among 1.2 and 1.5 times higher). In contrast, the
intensity of M+• was around 1.1 and 1.2 times higher than [M+H]+ for
naphthalene, I[1,2,3-cd]P, B[g,h,i]Pe, D[a,h]A, D[a,h]P, D[a,e]P, D[a,i]P and
D[a,l]P. For the rest of PAHs, the abundance of M+• and [M+H]+ was similar.
Fig. 1 (up) shows the APCI spectrum of D[a,l]P where M+• and [M+H]+ can
be observed as base peak of the spectrum. After that, the fragmentation of the
PAHs in the collision cell was studied. M+• and [M+H]+ were selected as
precursor ions for all PAHs. Fragmentation was performed at collision

energies between 10–60 eV. The losses of H• and H2 from both M+• and
[M+H]+ were the most abundant and common to all PAHs studied. Losses as
CH3• or C2H2 among others were also commonly observed. The use of water
as modifier favored the formation of the [M+H]+ and the M+• decreased in all
cases (see Fig. 1, down). However, M+• still appeared in the spectra in a
percentage among 35–80%, depending on the PAH.
Fig. 1. APCI mass spectra of D[a, l]P under charge transfer conditions (up)
and under proton transfer conditions with water as modifier (down).
3.2. Sensitivity and repeatability. Effect of different modifiers.
Once the SRM transitions were optimized, the sensitivity and repeatability of
the GC-MS/MS signals were evaluated for all the transitions selected (those
coming from M+• as well as from [M+H]+). As explained above, both [M+H]+

and M+• were observed in the APCI spectra of all PAHs even under “dry”
conditions. Thus, we investigated the effect of different modifiers in order to
promote the formation of the protonated molecules and to increase the
sensitivity of SRM transitions coming from [M+H]+. Water, 0.5% formic acid,
methanol and isopropanol were added separately in an uncapped vial, which
was located within a specially designed holder placed in the source door. The
sensitivity and repeatability of the GC-MS/MS signals for PAHs were
evaluated under the different conditions (with and without the use of
modifiers).
As can be seen in Fig. 2 (up), under charge transfer conditions (without adding
a modifier), SRM transitions coming from M+• were more sensitive than those
coming from [M+H]+ for acenaphthene, fluorene, B(c)F, C[c,d]P, chrysene,
5-MC, B(b)F and B(j)F (among 2 and 4 times higher). In contrast, the
sensitivity of SRM transitions coming from [M+H]+ was higher than M+•
(among 3 and 14 times) for naphthalene, acenaphthylene, phenanthrene,
anthracene, fluoranthene, pyrene and B[g,h,i]Pe. For the rest of PAHs, they
were mostly similar. When adding water as modifier, the sensitivity of
transitions coming from the M+• decreased between 1.5 and 3 times for all
compounds, except for D[a,h]A which was rather similar; and consequently,
the intensity for [M+H]+ transitions improved 1.5–2 times for high molecular
weight PAHs (HMW-PAHs, with four or more aromatic rings).
When adding HCOOH 0.5%, the sensitivity of [M+H]+ transitions remained

almost the same as with only water (Fig. 2 (down)). Similar results were
obtained using MeOH as modifier, except for some compounds, such as
naphthalene, fluorene, phenanthrene, fluoranthene, chrysene and B(b)F,
among others, whose responses decreased considerably. Regarding

isopropanol, its use was not beneficial because sensitivity extremely decreased
for most of the compounds. Thus, the modifiers MeOH and isopropanol were
discarded for further optimization.
Fig. 2. Absolute sensitivity of SRM transitions coming from M+· and [M+H]+
for PAHs under charge transfer and proton transfer (with water) ionization
conditions (up); absolute sensitivity of SRM transitions coming from [M+H]+
under proton transfer conditions using different modifiers (bottom).
Once these modifiers were discarded, the repeatability of the response (n=10
at 10 ng·mL−1) was also evaluated under the remaining working conditions
(without water, with water and with HCOOH 0.5%) (Fig. S1). Our data
showed a slightly poorer repeatability for M+• and [M+H]+ transitions working
under dry conditions (relative standard deviations, RSD, between 20–30% for
most compounds) being a bit better (RSD 10–17%) for fluoranthene, pyrene

and low molecular weight PAHs (LMW-PAHs, 2–3 aromatic rings). This
parameter was improved under wet conditions using water as modifier for M+•
(RSD 7–20%) and specially for [M+H]+ transitions (RSD 2–8%), but
dramatically get worst when adding HCOOH 0.5%, especially for M+·
transitions (RSD 14–39%).
Overall, the use of water as modifier was selected for further experiments,
selecting the transitions from [M+H]+ for most of PAHs. In some particular
cases, the most sensitive transitions using M+• as precursor ion were used for
confirmation purposes (except for b[j]F which was used for quantification)
(Table 1).
In a later stage, the optimum conditions selected for PAHs were tested for the
PCBs and pesticides in order to widen the applicability of the methodology to
other relevant pollutants included in the method. In previous works, it was
reported that sensitivity for halogenated hydrocarbons without any other
heteroatoms, e.g. PCBs and many organochlorine pesticides (such as
hexachlorobutadiene, pentachlorobenzene, hexachlorobenzene, HCHs,
DDTs, trans-chlordane, mirex) was negatively affected by the introduction of
water in the source [34,35]. Despite this fact, the sensitivity and repeatability
of PCBs, p,p′-DDT, p,p′-DDD, p,p′-DDE and HCB was still enough to reach
the maximum residue level regulated. Regarding the other pesticides included
in the method, ionization efficiency was favored under proton-transfer
conditions, i.e. selecting [M+H]+ as precursor ion, except for ethoxyquin and
ethoxyquin dimer that were scarcely affected.

Figure S1. Repeatability of the response (n=10 at 10 ng/mL) evaluated under

the following working conditions (without water, with water and with
HCOOH 0.5%).

Table 1. Experimental conditions of the optimized GC-APCI-MS/MS method using water as

modifier. Quantifier (Q) and qualifier (q) transitions.
Rt (min) Compound Internal Standard Precursor ion Product ion CE (EV) q/Q ratio
5.53 Naphthalene-D8 136 134.1 30 Q
5.55 Naphthalene Naphthalene-D8 129 128.1 20 Q
126.8 20 0.273
103 20 0.011
79 20 0.002
128 102.1 20 0.063
7.65 Acenaphthylene-D8 160 158.3 30 Q
Acenaphthylene-
7.66 Acenaphthylene 153 152 20 Q
D8
150.1 30 0.010
125.9 30 0.018
152 126.1 30 0.018
Acenaphthylene-
7.82 Acenaphthene 155 154 20 0.539
D8
153 20 Q
129.2 20 0.367
128 30 0.317
154 153 20 0.852
Acenaphthylene-
8.42 Fluorene 167 166 20 0.349
D8
165.1 20 Q
152 20 0.956
115 30 0.064
166 165.1 20 0.607
8.72 Diphenylamine Tebuconazole-D6 170 93 30 Q
168 30 0.035
153 20 0.086
152 20 0.070
169 168 20 0.506
8.87 HCB HCB-13C6 282 247 30 Q
212.1 30 0.718
177 40 0.557
142 50 0.435
8.86 HCB-13C6 292 222 40 Q
9.01 Ethoxyquin Tebuconazole-D6 218 202 20 0.576
174 20 Q
148 20 0.671
160 30 0.510
217 202 10 0.273
9.81 Phenanthrene-D10 188 186.4 20 Q
9.85 Phenanthrene Phenanthrene-D10 179 178.1 20 Q
176.7 30 0.025
176.1 40 0.044
152.1 30 0.083
178 152 30 0.083
9.88 Anthracene Phenanthrene-D10 179 178 20 Q
176.7 30 0.025
176.1 50 0.026
152.1 30 0.069
178 152 30 0.112
9.98 PCB 28 4-4’-DDE D8 256 221 20 0.060
186 30 Q
151 40 0.177
10.07 Fluazinam Tebuconazole-D6 465 372.9 20 Q
337.7 40 0.087
303.9 40 0.077
268.9 50 0.045

Table 1 (continuation)
Rt (min) Compound Internal Standard Precursor ion Product ion CE (EV) q/Q ratio
10.25 Chlorpyrifos-methyl Tebuconazole-D6 322 290 10 0.161
212 30 0.065
177 40 0.021
125 20 Q
10.39 PCB 52 4-4’-DDE D8 290 220 30 Q
292 257 20 0.605
220 30 0.689
150 50 0.334
10.41 Pirimiphos-methyl Tebuconazole-D6 306 164 20 Q
108 40 0.505
125 30 0.269
95 20 0.145
10.71 Malathion Tebuconazole-D6 331 211 10 0.026
143 30 0.017
125 30 Q
117 20 0.047
99 10 0.732
11.69 PCB 101 4-4’-DDE D8 324 254 40 Q
326 291 20 0.938
256 40 0.936
184 50 0.081
12.17 Fluoranthene-D10 212.2 210.3 30 Q
12.22 Fluoranthene Fluoranthene-D10 203 202.1 30 Q
200 40 0.069
175.9 50 0.007
152 50 0.014
202 200.8 30 0.053
12.30 4,4'-DDE-D8 324 254 40 Q
12.34 4,4'-DDE 4-4’-DDE D8 316 246.1 30 Q
210 40 0.083
281 20 0.162
176 50 0.483
12.51 Imazalil Tebuconazole-D6 297 254.8 10 0.031
201.1 10 0.050
176.3 20 0.334
159 20 Q
109 20 0.259
12.93 Pyrene-D10 212.2 210.3 30 Q
12.97 Pyrene Pyrene-D10 203 202.2 30 Q
200.8 40 0.067
200 50 0.041
176.1 50 0.003
202 200.8 30 0.039
13.03 PCB 118 4-4’-DDE D8 326 291 20 0.107
256 40 Q
184 50 0.470
324 254 40 0.968
13
13.28 PCB 153 C12-PCB 153 360 325 20 0.478
290 40 Q
218 60 0.551
358 253 50 0.151
13.27 PCB 153-13C12 376 306 40 Q
13.65 4,4'-DDD Tebuconazole-D6 235 165 20 Q
115 50 0.028
99 40 0.029

Rt Compound Internal Precursor Product CE q/Q
(min) Standard Ion Ion (eV) ratio
13.98 B[c]F Pyrene-D10 217 215.9 20 0.129
215 30 0.370
202 30 Q
200.1 40 0.048
216 215 30 0.393
13
14.27 PCB 138 C12-PCB 153 360 325 20 0.683
290 40 Q
218 60 0.559
358 253 50 0.150
14.31 4,4'-DDT 4-4’-DDE D8 235 165 20 Q
115 50 0.026
99 40 0.023
14.64 Tebuconazole-D6 314 296 10 Q
14.70 Tebuconazole Tebuconazole-D6 308 290 10 Q
165 20 0.430
151 20 0.246
125 40 0.865
13
15.67 PCB 180 C12-PCB 153 392 322 40 0.557
394 324 30 Q
396 361 20 0.454
254 60 0.448
16.80 B[a]A Pyrene-D10 229 228 40 Q
226.9 30 0.262
226 40 0.629
202.2 30 0.148
228 226 40 0.568
17.07 C[c,d]P Pyrene-D10 227 226 30 Q
225 40 0.076
200.8 40 0.002
226 224.9 40 0.039
17.11 Chrysene Pyrene-D10 229 227.9 40 Q
227 30 0.440
226 40 0.842
202.1 30 0.219
228 226 40 0.579
18.75 5-MC Pyrene-D10 243 242 30 0.043
228 30 Q
226.1 40 0.261
202.1 40 0.066
242 241.1 20 0.227
19.96 Ethoxyquin dimer Tebuconazole-D6 433 418 10 0.076
216 20 Q
188 30 0.744
432 417 20 0.307
173 50 0.354
21.06 B[b]F B[a]P-D12 253 251.9 30 Q
250.8 40 0.064
250.1 50 0.349
226 50 0.051
252 250 40 0.188
21.16 B[k]F B[a]P-D12 253 251.9 30 Q
250.8 40 0.046
250.1 50 0.174
226 50 0.011
252 250 40 0.165

Rt Compound Internal Precursor Product CE q/Q
(min) Standard Ion Ion (eV) ratio
22.28 B[j]F B[a]P-D12 253 251.9 30 0.899
248.6 60 0.020
226 50 0.269
224.1 60 0.120
252 250 40 Q
21.97 Boscalid Tebuconazole-D6 343 307 20 0.336
140 20 Q
112 40 0.413
342 140 20 0.039
22.75 B[a]P-D12 264.3 262.3 30 Q
22.66 B[a]P B[a]P-D12 253 252.1 30 Q
250.7 40 0.023
250.2 50 0.201
226 50 0.037
252 250 40 0.146
24.97 Azoxystrobin Tebuconazole-D6 404 372 10
344 20 0.243
329 30 0.414
172 50 0.212
156 50 0.211
25.19 I[1,2,3-cd]P B[g,h,i]Pe -D12 277 276.1 30 Q
275.4 50 0.123
274.3 50 0.192
25.22 D[a,h]A B[a]P-D12 279 278.1 30 Q
276.8 40 0.161
275.9 40 0.333
263.3 30 0.267
25.87 B[g,h,i]Pe -D12 288.2 286.3 30 Q
25.93 B[g,h,i]Pe B[g,h,i]Pe -D12 277 276.2 40 Q
275.1 50 0.153
274 50 0.154
272.7 60 0.005
276 275.1 50 0.053
28.86 D[a,l]P B[g,h,i]Pe -D12 303 302.1 30 Q
300.8 30 0.067
300 50 0.455
276 40 0.057
302 300 50 0.288
29.92 D[a,e]P B[g,h,i]Pe -D12 303 302.1 30 Q
300.1 50 0.341
298.6 60 0.008
297.9 60 0.022
302 300 50 0.183
30.59 D[a,i]P B[g,h,i]Pe -D12 303 302.1 30 Q
300.1 50 0.245
297.9 60 0.019
276 50 0.033
302 300 50 0.178
31.00 D[a,h]P B[g,h,i]Pe -D12 303 302.1 30 Q
300.1 50 0.274
297.9 60 0.019
276 50 0.033
302 300 50 0.219

3.3. Selectivity of the SRM transitions for PAHs
Selectivity of the selected SRM transitions, especially in cases where

coelution occurred, was also studied during the optimization process. In such
cases, the possibility of selecting different precursor ions for the same
compound (M+• or [M+H]+) was of great advance in terms of specificity. This
was the case of the determination of cyclopenta(c,d)pyrene (C18H10,
MW=226), which nearly coelutes with chrysene (C18H12, MW=228) (see Fig.
S2). Chrysene shows 226 as an in-source fragment (coming from M+• 228).
Thus, when monitoring C(c,d)P with the transition coming from its M+• 226,
an interference from chrysene (coming from an in-source fragment) is
observed in the chromatogram, which makes the peak area integration
troublesome, especially when the concentration of chrysene is higher than
cyclopenta(c,d)pyrene. However, quantification and confirmation transitions
for C(c,d)P coming from [M+H]+ 227 did not suffer from this interference and
showed higher specificity, which is commonly observed on EI-based methods
[36]. A similar case occurred for the determination of indeno(1,2,3-cd)pyrene
(C22H12, MW=276) due to the nearly coelution of dibenzo(a,h)anthracene
(C22H14, MW=278). D(a,h)A shows an in-source fragment at m/z 276, so the
transition coming from M+• 276 for I(1,2,3,cd)P was interfered by D(a,h)A,
while those coming from [M+H]+ 277 were much cleaner.

Figure S2. GC-APCI-MS/MS chromatogram of the triplet B[a]A, C[c,d]P and

chrysene for (up) SRM transition coming from [M+H]+ and (down) SRM
transition coming from M+·.

3.4. Study of matrix effects
Matrix effects observed under GC-APCI-MS come from the GC liner

(normally signal enhancement) and from the APCI source (normally signal
suppression) [27]. As a summary of the matrix effect study for 9 out of 19
matrices (where different dilution factors were tested: crude extract, 5-fold,
10-fold and 20-fold dilution), signal enhancement/suppression was observed
for: i) 36% of the combinations matrix/PAHs, 58% for matrix/pesticides and
52% for matrix/PCBs (crude extract); ii) 35% for matrix/PAHs, 46% for
matrix/pesticides and 17% for matrix/PCBs (5-fold dilution factor); iii) 26%
of matrix/PAHs, 33% for matrix/pesticides and 14% for matrix/PCBs (10-fold
dilution factor); and iv) 11% for matrix/pesticides (20-fold dilution factor). As
an illustrative example, Fig. 3 shows how the matrix effect for PAHs in sea
bream fillet decreases when the different dilution factors studied are applied.
Thus, 20-fold dilution of the final extract was finally selected for further
experiments.
As a summary, for a 20-fold dilution, signal enhancement was observed only

for ethoxyquin and ethoxyquin-dimer, whereas signal suppression was
observed for boscalid and azoxystrobin (mainly in plant and marine
ingredients). No significant matrix effect was observed for PAHs, PCBs and
the rest of pesticides (Fig. S3). According to our data, 20-fold dilution of the
final sample extract eliminated matrix effect for most analyte/matrix
combinations, and this allowed us to perform the quantification in the analysis
of the 19 aquaculture matrices using calibration curve in solvent. The
remaining matrix effects for boscalid and azoxystrobin could be corrected with
the use of SIL-IS. Matrix effect for ethoxyquin and ethoxyquin-dimer could
not be properly evaluated in feed, krill meal and fishmeal matrices because the

analyte concentration in these samples was typically higher than the spiking
levels assayed.
Fig. 3. Matrix effect study for PAHs in sea bream fillet. Relative error between the slopes of calibration in
solvent and matrix-matched calibration at different dilution factors.

Figure S3. Matrix effect observed for a 20-fold dilution factor.
3.5. Analytical parameters
Validation of the analytical procedure was carried out in salmon and sea bream
fillets considering the parameters of linearity, accuracy, precision, LODs and
LOQs. Relative responses to the selected SIL-IS were used (see Table 1).

The study of the linearity revealed that correlation coefficients (R2) were
higher than 0.99, with residuals lower than 30% for most compounds in the
range 0.01–25 ng·mL−1. Some exceptions were: 0.0125–12.5 ng·mL−1
(C[c,d]P); 0.025–25 ng·mL−1 (naphthalene, acenaphthylene, acenaphthene,
anthracene, B[c]F, B[a]A, chrysene, B[b]F, B[k]F, B[j]F, D[a,h]A, PCB 153
and 180, diphenylamine, tebuconazole, ethoxyquin dimer, boscalid and
azoxystrobin); 0.1–25 ng·mL−1 (D[a,l] P and D[a,e]P, HCB and p,p′-DDE);
0.5–25 ng·mL−1 (D[a,i]P, D[a,h]P, ethoxyquin, fluazinam, imazalil, p,p′-DDD
and p,p′-DDT).
Trueness and precision data for salmon and sea bream fillets are shown in
Table 2 which shows satisfactory recoveries and precision for the wide
majority of compounds at the two concentrations studied, 2 and 50 ng·g−1. A
LOQ objective of 2 ng·g−1 was empirically demonstrated for most compounds
in both matrices, using samples spiked at this level and subjected to the overall
analytical procedure. A statistical LOQ of 5 ng·g−1 (S/N=10) was estimated
for naphthalene, fluorene, fluoranthene and pyrene, as the high concentrations
present in the “blank” samples did not allow the calculation of recoveries and
precision below that concentration. Dibenzopyrenes have poor response, and
therefore the LOQ was established in 50 ng·g−1. The same occurred for HCB,
fluazinam, imazalil, p,p′-DDE, p,p′-DDD and p,p′- DDT which could not be
detected at the low level of 2 ng·g−1 being 50 ng·g−1 still the normally applied
MRL for these compounds (as commented above, these compounds were
negatively affected by the introduction of water as modifier in the source).
Similarly to some light PAHs, ETQ and ETQ-D were found at quite high
concentrations in the samples impeding validation at 2 ng·g−1 level. Thus,
LOQs were statistically calculated for S/N=10. LODs were found to be
between 0.5–2 ng·g−1 for 80% of the PAHs, all the PCBs and 50% of the

pesticides studied. The rest of compounds showed LODs between 5 and 50

ng·g−1.
In relation to the remaining 17 matrices, a total of 48 QC samples, spiked at 5

ng·g−1 and 50 ng·g−1, were analyzed. Fig.4 shows the boxplot diagrams
representing the recovery values for PAHs in all matrices at the two
concentrations assayed. Some recoveries could not be calculated (e.g.
napthalene, fluorene, phenanthrene, pyrene, D[a,e]P, D[a,i]P and D[a,h]P) at
5 ng·g-1 spiking level due to the high analyte concentration present in “blank”
sample. It can be emphasized that all QC recoveries were in the range 60–
140%, with median values between 88% and 115%, at 5 ng·g−1, and between
92% and 116% at 50 ng·g−1. Among the studied matrices, PAPs showed
higher complexity giving individual recoveries out of 70–120% range without
a clear tendency for specific compounds. Data from above illustrate that the
methodology applied was robust and satisfactory for the determination of very
low PAH concentrations in different sample matrices, some of them of high
complexity. The excellent sensitivity of GC-APCI-MS/MS allowed the extra
dilution (x20) of the sample extracts, minimizing matrix effects but still
reaching low LOQs, comparable and, in some cases, even lower than those
reported in the literature [21,22] with the great advantage of using solvent
calibration for quantification of the compounds in all matrices tested.

Table 2: Mean recoveries (%) and precision (RSD, %) for PAHs, PCBs and
pesticides after application of overall analytical procedure to salmon fillet and
gilthead fillet (n =6), LOQs objective of the method.
Compounds Salmon fillet Sea bream fillet
LOQ LOQ
Spiking levels Spiking levels
objective objective
(ng·g-1) (ng·g-1)
(ng·g-1) (ng·g-1)
2 50 2 50
a
Naphthalene 98 (18) 5b a 98 (17) 5b
Acenaphthylene 119 (11) 101 (7) 2 106 (8) 103 (2) 2
Acenaphthene 74 (19) 72 (6) 2 93 (12) 73 (3) 2
a
Fluorene 95 (11) 5b a 108 (6) 5b
a
Phenanthrene 116 (9) 2b a 113 (9) 2b
Anthracene 91 (8) 104 (5) 2 101 (8) 107 (6) 2
a
Fluoranthene 103 (7) 5b a 96 (11) 5b
a
Pyrene 106 (7) 5b a 99 (14) 5b
B(c)F 98 (11) 94 (8) 2 90 (3) 90 (2) 2
B(a)A 114 (5) 101 (5) 2 113 (9) 73 (2) 2
C(c,d)P c 108 (7) 115 (4) 1 112 (19) 112 (7) 1
Chrysene 107 (12) 101 (4) 2 116 (8) 76 (6) 2
5-MC 117 (5) 102 (5) 2 108 (7) 75 (3) 2
B(b)F 80 (13) 104 (7) 2 106 (20) 96 (4) 2
B(k)F 71 (8) 103 (6) 2 78 (7) 86 (4) 2
B(j)F 104 (17) 76 (4) 2 115 (18) 102 (5) 2
B(a)P 104 (3) 88 (7) 2 113 (14) 86 (5) 2
I(1,2,3,cd)P 83 (12) 104 (5) 2 112 (9) 102 (3) 2
D(a,h)A 113 (11) 100 (5) 2 109 (11) 81 (6) 2
B(g,h,i)Pe 84 (8) 101 (4) 2 98 (20) 101 (3) 2
D(a,l)P – 97 (4) 50 – 77 (6) 50
D(a,e)P – 95 (8) 50 – 72 (9) 50
D(a,i)P – 86 (13) 50 – 95 (13) 50
D(a,h)P – 81 (10) 50 – 73 (6) 50
Diphenylamine 119 (20) 100 (17) 2 – 98 (17) 50
HCB – 112 (11) 50 – 116 (17) 50
a a
Ethoxyquin 15b a a 75b
Fluazinam – 76 (4) 50 – 107 (15) 50
Chlorpiriphos-
77 (10) 100 (6) 2 113 (13) 80 (7) 2
methyl
Pirimiphos-methyl 92 (12) 82 (4) 2 115 (13) 110 (9) 2
Malathion 95 (20) 94 (13) 2 90 (11) 104 (10) 2
4,4′-DDE – 95 (19) 50 – 76 (19) 50
Imazalil – 87 (8) 50 – 112 (14) 50
4,4′-DDD – 112 (9) 50 – 114 (18) 50
4,4′-DDT – 109 (9) 50 – 115 (11) 50
Tebuconazole 103 (19) 109 (11) 2 99 (16) 113 (12) 2
a a
Ethoxyquin dimer 5b a a 15b
Boscalid 115 (7) 103 (17) 2 103 (4) 106 (11) 2
Azoxystrobin 95 (16) 102 (14) 2 119 (10) 109 (9) 2

Compounds Salmon fillet Sea bream fillet
LOQ LOQ
Spiking levels Spiking levels
objective objective
(ng·g-1) (ng·g-1)
(ng ·g-1) (ng·g-1)
2 50 2 50
PCB-28 106 (13) 117 (7) 2 110 (11) 120 (8) 2
PCB-52 107 (19) 100 (12) 2 98 (20) 97 (12) 2
PCB-101 106 (17) 120 (3) 2 109 (15) 118 (3) 2
PCB-118 110 (21) 105 (3) 2 98 (19) 121 (3) 2
PCB-153 95 (20) 120 (14) 2 87 (15) 115 (16) 2
PCB-138 115 (6) 118 (13) 2 117 (8) 116 (13) 2
PCB-180 105 (11) 111 (13) 2 81 (18) 108 (3) 2
a
Recoveries could not be calculated due to the high concentration in the blank
samples.
b
LOQ was estimated as the analyte concentration that produced a peak signal
of ten times the background noise in the chromatogram at the lowest
fortification level studied for each compound.
c
C[c,d]P spiked at 1 ng·g-1 and 25 ng·g-1.

Fig. 4. Box plots that shows the recoveries of the 48 QCs (corresponding to
17 different matrices) spiked at 5 (up) and 50 ng·g−1 (down).

3.6. Analysis of samples
In order to demonstrate the applicability of the methodology developed in this

paper, it was applied to the analysis of a notable number of samples from
aquaculture production: 12 diets, 12 fish fillets, 19 terrestrial animal
ingredients, 6 marine ingredients, 19 plant ingredients and 8 plant oils. A
reagent blank and two spiked samples (5 and 50 ng·g-1) for each matrix
analyzed were included in each batch. The results, in Table S2, are expressed
in a wet weight basis. The qi/Q ratios obtained for all positive samples were
within the range of the tolerance accepted (30%) around the experimental qi/Q
value obtained from reference standards.
LMW-PAHs together with HMW-PAHs as fluoranthene and pyrene were

found in most of the samples. Regarding diets, fish fillets, plant ingredients
and PAPs, concentrations ranged from < 5 ng·g−1 for acenaphthene to 126.6,
77.1, 88.6 and 177.5 ng·g−1 for naphthalene, respectively. In marine
ingredients, they ranged from < 5 ng·g−1 for acenaphthene and acenaphthylene
to 129.0 ng·g−1 for phenanthrene. For plant oil ingredients, concentrations
ranged from < 5 ng·g−1 for acenaphthylene to 186.9 ng·g−1 for phenanthrene.
In the present study only two fish oils were analyzed, and clearly a large
variation exist in the environmental background levels of PAHs in fish oils
[13,14]. Fig. S4 shows the identification of 5-MC in one fish meal. This PAH
was below the LOQ in all samples analyzed. As can be seen, three transitions
were acquired and qi/Q ratios were in agreement with SANTE/11945/2015
identification criteria followed in the current research [33].
Regarding pesticides, the non-organochlorine pesticides that are currently

used on crop such as chlorpiriphos-methyl and pirimiphos-methyl were only

detected in the novel plant ingredients and not in marine feed ingredients such
as fish oil and meal [25]. Malathion, imazalil and tebuconazole were only
found in some ingredients while boscalid and azoxystrobin were indeed found
in feed. It is worth to mention that studied PCBs, DDTs and pesticides were
not found in fillets from salmon and sea bream. The anti-oxidant ethoxyquin
and its main metabolite ethoxyquin dimer are deposited to marine feed
ingredients such as fish meal to prevent lipid oxidation and spontaneous
combustion during overseas transport and storage (Lundebye et al., 2010). The
ethoxyquin levels were highest in fish meal and krill (32,462 and 40,076
ng·g−1, respectively), but surprisingly high levels were also found in fish oil
(37,137 ng·g−1) where normally other antioxidant such as BHT and BHA are
used [37]. All other feed ingredients, including terrestrial and plant, also
contained ethoxyquin, albeit at l00-150 fold lower levels than the marine
ingredients (Table S2). PCB congeners were not detected in any sample.

142
Capítulo 2
Table S2. Analysis of samples. Concentrations found in ingredients, feed and fish.
Marine ingredients (ng/g) Plant oil ingredients (ng/g) Plant meal ingredients (ng/g)
Krill Rapeseed Palm Palm Linseed Soya Corn Wheat Pea Rapeseed Sunflower
PAHs Fish meal Fish Oil Wheat Peas
meal oil Olein oil oil protein gluten gluten protein cake Meal
Naphthalene 57.9-69.4 67.3-70.7 84.6 93.9-135.1 130.4 106.5 77.8 55.1-62.0 49.9-72.2 59.2-73.4 51.8-72.1 60.9-88.6 57.7 70.7 45.1
Acenaphthylene ND ND-9.6 d d-41.0 d d d ND-d ND-d ND-d d ND-d ND d ND
Acenaphthene d-5.2 5.6-10.8 8.0 6.3-15.6 6.9 7.6 11.5 d-7.9 d-6.7 d-6.0 d-7.2 d-8.4 d 5.9 6.4
Fluorene 9.5-15.9 20.3-54.2 18.7 19.3-46.6 14.1 19.9 29.6 6.4-17.7 6.7-16.7 8.7-14.1 d-18.7 8.0-13.6 12.9 10.3 13.7
Phenanthrene 24.2-46.4 56.0-129.0 56.6 34.7-186.9 38.9 78.6 116.1 21.8-52.8 11.9-63.4 17.9-51.4 20.1-56.6 29.9-43.5 19.2 43.0 60.3
Fluoranthene 6.0-10.3 18.3-27.9 12.8 11.4-50.2 16.0 18.0 27.3 6.0-9.5 4.8-11.1 8.3-12.2 5.2-11.6 5.6-10.0 7.8 11.5 8.3
Pyrene 25.5-38.9 81.8-85.6 49.5 47.3-125.1 67.5 69.8 99.1 25.8-33.8 13.9-28.9 33.8-44.8 27.8-35.9 23.9-32.3 31.4 41.5 34.5
B(a)A ND ND ND ND-5.2 ND ND ND ND ND ND-2 ND ND ND ND ND
C(c.d)P ND ND ND ND-5.6 ND ND ND ND ND ND ND ND ND ND ND
Chrysene ND-2 ND ND ND-8.0 ND ND ND ND ND ND-3.1 ND ND ND ND ND
5-MC ND-d ND ND ND ND ND ND ND ND ND ND ND ND ND ND
I(1.2.3.cd)P ND-7.5 ND ND ND ND ND ND ND ND ND ND ND ND ND ND
B(g.h.i)Pe ND-5.6 ND ND ND ND ND ND ND ND ND ND ND ND ND ND
Pesticides
Ethoxyquin* 476-32462 441-37137 40076 57.4-12117 740 239 339 187-599 174-506 165-411 221-431 42.6-148 439 306 212
Chlorpiriphos-methyl ND ND ND ND-d ND ND ND ND ND ND-36.9 ND ND ND ND ND
Pirimiphos-methyl ND ND ND 6.0-40.6 ND ND d ND-d ND-44.4 ND-191 ND-64.2 d ND ND ND
Malathion ND ND ND ND-d ND ND ND ND ND ND ND ND ND ND ND
Imazalil ND-d ND ND ND-d ND ND ND ND ND ND ND ND ND ND ND
Tebuconazole ND-16.3 ND ND ND-10.4 ND ND ND ND ND ND-d ND ND-d ND ND ND
Ethoxyquin dimer 25.9-35071 8.7-263 152 ND-38.7 ND d ND ND-12.3 ND ND ND ND 22.9 ND ND
Boscalid ND-d ND ND d–7.2 ND ND ND ND ND ND ND-d ND ND ND ND
Azoxystrobin ND-d ND ND ND-d ND ND ND ND ND ND ND ND ND ND ND
*Included as preservative
T. Portolés et al., Talanta 172 (2017) 109-119

Feed (ng/g) Fillet (ng/g) Terrestrial animal ingredients (ng/g)
Capítulo 2
Sea bream Sea bream Salmon Sea bream Sea bream Salmon Poultry Poultry Feather Pork blood Pork Pork
PAHs
feeda feedb feed filleta filletb fillet blood meal meal meal meal meal greaves
Naphtalene 53.9-69.1 49.8-64.1 75.7-126.6 67.2-70.9 52.8-77.1 60.9-72.0 8.7-68.0 54.5-177.5 62.7-82.4 57.3-66.3 66.7-81.6 70.6-75.9
Acenaphthylene ND ND ND ND ND ND ND ND ND ND ND ND
Acenaphthene d d d-6.9 d d d d-8.0 6.4-14.6 7.0-10.0 d-6.8 6.9-9.2 6.3-9
Fluorene 8.9-13 7.9-18.5 11.1-21.9 10.8-13.3 11.7-17.8 8-14.4 d-14.1 11.1-20.2 11.2-26.4 5.1-10.8 10.7-13.1 13.3-13.4
Phenanthrene 17-38.7 21.3-43.4 26-38.9 29.5-36.3 30.7-67 17.5-34.8 6.3-39.3 34.8-48.3 26.7-40.4 13.3-33.2 16.9-41.8 26.8-27.8
Fluoranthene 7-11.2 6.3-9.2 8.8-11.5 5.7-9.5 5.4-9.3 d-6.1 d-10.2 7.2-9.1 11.2-14.4 d-6.4 5.7-8.8 ND-9.8
Pyrene 30.1-44.4 39.4-45 36.8-56.9 29.5-42.5 24.8-36.2 20.2-33.3 22.5-37.3 32.8-49.9 40.8-57.2 20.1-28 19.9-28.7 35.5-47.7
B(a)A ND ND ND-3.9 ND ND ND ND ND ND ND ND ND-d
C(c.d)P ND ND ND ND-d ND ND ND ND ND ND ND ND
Chrysene ND ND-2 ND-4.7 ND ND ND ND ND ND ND-d ND ND
5-MC ND ND ND-d ND ND ND ND ND ND ND ND ND
B(b)F ND ND ND-5.1 ND ND ND ND-3.1 ND ND ND ND ND
B(k)F ND ND ND-6.3 ND ND ND ND-3.9 ND ND ND ND ND
B(a)P ND ND ND-7.8 ND ND ND ND ND ND ND ND ND
I(1.2.3.cd)P ND ND ND-7 ND ND ND ND ND ND ND ND ND
D(a.h)A ND ND ND-16 ND ND ND ND ND ND ND ND ND
T. Portolés et al., Talanta 172 (2017) 109-119

B(g.h.i)Pe ND ND ND-17.6 ND ND ND-d ND ND-d ND-d ND ND ND
Pesticides
Ethoxyquin* 6385-9314 6578-8169 5450-6437 375-512 349-463 126-418 256-375 187-692 316-351 369-477 256-317 379-441
Chlorpiriphos-methyl d-7.0 ND ND-8.8 ND ND ND ND ND ND ND ND ND
Pirimiphos-methyl 15.7-28.5 ND-d 5.8-49.0 ND ND ND ND ND ND ND ND ND
Tebuconazole ND ND ND ND ND ND ND ND ND ND ND ND
6881- 11697-
Ethoxyquin dimer 1318-1429 261-480 635-782 29.1-158 ND ND ND-d ND ND ND
17529 16535
Boscalid ND ND-d d-8.6 ND ND ND ND-6.7 ND ND ND ND ND
Azoxystrobin ND ND d-9.0 ND ND ND ND-7.5 ND ND ND ND ND
*Included as preservative
aFeed (1.5-4.5 mm) used and fish harvested after the first 8 months of the production cycle
bFeed (6-9 mm) used and fish harvested after the 18th month of the production cycle
143
Figure S4. Positive finding of 5-MC in fish meal (<5 ng·g-1) determined and
identified by applying GC-APCI-MS/MS in SRM mode. Q: Quantification
ion; qi: confirmation ion. , q/Q ratio within tolerance limits; ×, q/Q ratio out
of tolerance limits.

4. Conclusions
The use of APCI has been evaluated for GC-MS/MS analysis of PAHs. The
high sensitivity of this technique allowed the simultaneous quantification of
19 different complex matrices (from aquaculture field) using solvent
calibration. The key aspect for this aim was the elimination of matrix effect,
without the need of time-consuming purification steps, only by a 20-fold
dilution factor of the final QuEChERs extract, being this aspect a great
contribution of the present work. Despite this, LOQs of the developed method
were 2 ng·g−1 for most analytes, in the same order or better than those reported
in previously published methods for similar matrices [21,22] showing higher
efficiency [19]. Also, the use of both M+• or [M+H]+ gave an additional value
to the selectivity in the determination/identification capabilities.
Acknowledgment
This work has been funded by the EU Seventh Framework Programme

ARRAINA Project 288925 (Advanced Research Initiatives for Nutrition and
Aquaculture), and the Norwegian Research Council SAFETY-PAP project
(227387) from the Sustainable Innovation in Food and Bio-based Industries
(Bionær) program. Additional funding was obtained from Generalitat
Valenciana (research group of excellence PROMETEOII/2014/085;
PROMETEOII/2014/023; ISIC 2012/016). This work has been developed
within the framework of the Research Unit of Marine Ecotoxicology (IATS
(CSIC)-IUPA (UJI)). We also thank the Spanish Ministry of Economy and
Competitiveness and European funding from FEDER program (project
AGL2012-37201).

References
[1] L. Benedito-Palos, J.C. Navarro, A. Sitjà-Bobadilla, J.G. Bell, S. Kaushik,

J. Pérez- Sánchez, High levels of vegetable oils in plant protein-rich diets fed
to gilthead sea bream (Sparus aurata L.): growth performance, muscle fatty
acid profiles and histological alterations of target tissues, Br. J. Nutr. 100
(2008) 992–1003. http:// dx.doi.org/10.1017/S0007114508966071.
[2] B.E. Torstensen, J.G. Bell, G. Rosenlund, R.J. Henderson, I.E. Graff, D.R.
Tocher, et al., Tailoring of Atlantic salmon (Salmo salar L.) flesh lipid
composition and sensory quality by replacing fish oil with a vegetable oil
blend, J. Agric. Food Chem. 53 (2005) 10166–10178. http://dx.doi.org/
10.1021/jf051308i.
[3] N.S. Liland, G. Rosenlund, M.H.G. Berntssen, T. Brattelid, L. Madsen,

B.E. Torstensen, Net production of Atlantic salmon (FIFO, Fish in Fish out <
1) with dietary plant proteins and vegetable oils, Aquac. Nutr. 19 (2013) 289–
300. http://dx.doi.org/10.1111/j.1365-2095.2012.00958.x.
[4] A.G.J. Tacon, M. Metian, Feed Matters: satisfying the feed demand of
aquaculture, Rev. Fish. Sci. Aquac. 23 (2015) 1–10. http://dx.doi.org/10.
1080/ 23308249.2014.987209.
[5] T. Ytrestøyl, T.S. Aas, T. Åsgård, Utilisation of feed resources in

production of Atlantic salmon (Salmo salar) in Norway, Aquaculture 448
(2015) 365–374. http:// dx.doi.org/10.1016/j.aquaculture.2015.06.023.
[6] M. Benísek, L. Bláha, K. Hilscherová, Interference of PAHs and their N-

heterocyclic analogs with signaling of retinoids in vitro, Toxicol. Vitr. 22
(2008) 1909–1917. http://dx.doi.org/10.1016/j.tiv.2008.09.009.

[7] M.J. Dennis, R.C. Massey, G. Cripps, I. Venn, N. Howarth, G. Lee, Factors
affecting the polycyclic aromatic hydrocarbon content of cereals, fats and
other food products, Food Addit. Contam. 8 (1991) 517–530.
http://dx.doi.org/10.1080/ 02652039109374004.
[8] A. Fromberg, A. Højgård, L. Duedahl-Olesen, Analysis of polycyclic

aromatic hydrocarbons in vegetable oils combining gel permeation
chromatography with solid-phase extraction clean-up, Food Addit. Contam.
24 (2007) 758–767. http:// dx.doi.org/10.1080/02652030601150505.
[9] S. Moret, G. Purcaro, L.S. Conte, Polycyclic aromatic hydrocarbons in

vegetable oils from canned foods, Eur. J. Lipid Sci. Technol. 107 (2005) 488–
496. http:// dx.doi.org/10.1002/ejlt.200501060.
[10] V.H. Teixeira, S. Casal, M.B.P.P. Oliveira, PAHs content in sunflower,

soybean and virgin olive oils: evaluation in commercial samples and during
refining process, Food Chem. 104 (2007) 106–112. http://dx.doi.org/10.1016
/j.foodchem.2006.11.007.I.
[11] Yebra-Pimentel, R. Fernández-González, E. Martínez Carballo, J. Simal-

Gándara, Searching ingredients polluted by polycyclic aromatic hydrocarbons
in feeds due to atmospheric or pyrolytic sources, Food Chem. 135 (2012)
2043–2051. http:// dx.doi.org/10.1016/j.foodchem.2012.06.069.
[12] E.A. Pena, L.M. Ridley, W.R. Murphy, J.R. Sowa, C.S. Bentivegna,
Detection of polycyclic aromatic hydrocarbons (PAHs) in raw menhaden fish
oil using fluorescence spectroscopy: method development, Environ. Toxicol.
Chem. 34 (2015) 1946–1958. http://dx.doi.org/10.1002/etc.3015.
[13] M.H.G. Berntssen, K. Julshamn, A.-K. Lundebye, Chemical


use of traditional- versus alternative feed ingredients, Chemosphere 78 (2010)
637–646. http://dx.doi.org/ 10.1016/j.chemosphere.2009.12.021.
[14] J. Nácher-Mestre, R. Serrano, L. Benedito-Palos, J.C. Navarro, F.J.

López, S. Kaushik, et al., Bioaccumulation of polycyclic aromatic
hydrocarbons in gilthead sea bream (Sparus aurata L.) exposed to long term
feeding trials with different experimental diets, Arch. Environ. Contam.
Toxicol. 59 (2010) 137–146. http:// dx.doi.org/10.1007/s00244-009-9445-1.
[15] EC, Commission regulation (EU) No 835/2011 of 19 August 2011

amendingRegulation (EC) No 1881/2006 as regards maximum levels for
polycyclic aromatic hydrocarbons in foodstuffs. Off. J. Eur. Union L215, 4-8.,
n.d, 2011.
[16] M. Ciecierska, M.W. Obiedziński, Polycyclic aromatic hydrocarbons in

vegetable oils from unconventional sources, Food Control. 30 (2013) 556–
562. http:// dx.doi.org/10.1016/j.foodcont.2012.07.046.
[17] M.D.L. Easton, D. Luszniak, E. Von der Geest, Preliminary examination

of contaminant loadings in farmed salmon, wild salmon and commercial
salmon feed, Chemosphere 46 (2002) 1053–1074. http://dx.doi.org/10.1016/
S0045-6535(01) 00136-9.
[18] J. Hellou, K. Haya, S. Steller, L. Burridge, Presence and distribution of

PAHs, PCBs and DDE in feed and sediments under salmon aquaculture cages
in the Bay of Fundy, New Brunswick, Canada, Aquat. Conserv. Mar. Freshw.
Ecosyst. 15 (2005) 349–365. http://dx.doi.org/10.1002/aqc.678.
[19] J. Nácher-Mestre, R. Serrano, T. Portolés-Nicolau, F. Hernández, L.

Benedito- Palos, J. Pérez-Sánchez, A reliable analytical approach based on

gas chromatography coupled to triple quadrupole and time-of-flight mass
analyzers for the determination and confirmation of polycyclic aromatic
hydrocarbons in complex matrices from aquaculture activities, Rapid
Commun. Mass Spectrom. 23 (2009) 2075–2086. http://dx.doi.org/10.1002
/rcm.4120.
[20] P. Plaza-Bolaños, A.G. Frenich, J.L.M. Vidal, Polycyclic aromatic

hydrocarbons in food and beverages. Analytical methods and trends, J.
Chromatogr. A. 1217 (2010) 6303–6326. http://dx.doi.org/10.1016/j.chroma.
2010.07.079.
[21] H. Ziarrusta, M. Olivares, a. Delgado, O. Posada-Ureta, O. Zuloaga, N.

Etxebarria, Multiscreening determination of organic pollutants in molluscs
using matrix solid phase dispersion, J. Chromatogr. A. 1391 (2014) 18–30.
http://dx.doi.org/ 10.1016/j.chroma.2015.02.072.
[22] Y. Sapozhnikova, S.J. Lehotay, Multi-class, multi-residue analysis of

pesticides, polychlorinated biphenyls, polycyclic aromatic hydrocarbons,
polybrominated diphenyl ethers and novel flame retardants in fish using fast,
low-pressure gas chromatography-tandem mass spectrometry, Anal. Chim.
Acta 758 (2013) 80–92. http://dx.doi.org/10.1016/j.aca.2012.10.034.
[23] M. Jánská, M. Tomaniová, J. Hajšlová, V. Kocourek, Appraisal of

“classic” and “novel” extraction procedure efficiencies for the isolation of
polycyclic aromatic hydrocarbons and their derivatives from biotic matrices,
Anal. Chim. Acta 520 (2004) 93–103. http://dx.doi.org/10.1016/j.aca.
2004.05.073.

[24] J. Djinovic, A. Popovic, W. Jira, Polycyclic aromatic hydrocarbons

(PAHs) in traditional and industrial smoked beef and pork ham from Serbia,
Eur. Food Res.Technol. 227 (2008) 1191–1198. http://dx.doi.org/10.1007/
s00217-008-0836-8.
[25] J. Nácher-Mestre, R. Serrano, T. Portolés, M.H.G. Berntssen, J. Pérez-

Sánchez, F. Hernández, Screening of pesticides and polycyclic aromatic
hydrocarbons in feeds and fish tissues by gas chromatography coupled to high-
resolution mass spectrometry using atmospheric pressure chemical ionization,
J. Agric. Food Chem. 62 (2014) 2165–2174. http://dx.doi.org/10.
1021/jf405366n.
[26] Y.S. Johnson, Determination of polycyclic aromatic hydrocarbons in

edible seafood by QuEChERS-based extraction and gas chromatography-
tandem mass spectrometry, J. Food Sci. 77 (2012). http://dx.doi.org/10.1111
/j.1750- 3841.2012.02758.x.
[27] L. Cherta, T. Portolés, J. Beltran, E. Pitarch, J.G.J. Mol, F. Hernández,

Application of gas chromatography-(triple quadrupole) mass spectrometry
with atmospheric pressure chemical ionization for the determination of
multiclass pesticides in fruits and vegetables, J. Chromatogr. A. 1314 (2013)
224–240. http://dx.doi.org/ 10.1016/j.chroma.2013.09.029.
[28] M. Raro, T. Portolés, J.V. Sancho, E. Pitarch, F. Hernández, J. Marcos,

et al., Mass spectrometric behavior of anabolic androgenic steroids using gas
chromatography coupled to atmospheric pressure chemical ionization source.
Part I: ionization, J. Mass Spectrom. 49 (2014) 509–521 http://dx.doi.org/
10.1002/jms.3367.

[29] T. Portolés, L.E. Rosales, J.V. Sancho, F.J. Santos, E. Moyano, Gas
chromatography-tandem mass spectrometry with atmospheric pressure
chemical ionization for fluorotelomer alcohols and perfluorinated
sulfonamides determination, J. Chromatogr. A. 1413 (2015) 107–116.
http://dx.doi.org/10.1016/j.chroma.2015.08.016.
[30] T. Portolés, J.G.J. Mol, J.V. Sancho, F. Hernández, Advantages of

atmospheric pressure chemical ionization in gas chromatography tandem mass
spectrometry: pyrethroid insecticides as a case study, Anal. Chem. 84 (2012)
9802–9810. http:// dx.doi.org/10.1021/ac301699c.
[31] J. Nácher-Mestre, R. Serrano, E. Beltrán, J. Pérez-Sánchez, J. Silva, V.

Karalazos, et al., Occurrence and potential transfer of mycotoxins in gilthead
sea bream and Atlantic salmon by use of novel alternative feed ingredients,
Chemosphere 128 (2015) 314–320. http://dx.doi.org/10.1016/j.chemosphere
.2015.02.021.
[32] A.M. Botero-Coy, J.M. Marín, R. Serrano, J.V. Sancho, F. Hernández,

Exploring matrix effects in liquid chromatography-tandem mass spectrometry
determination of pesticide residues in tropical fruits, Anal. Bioanal. Chem.
407 (2015) 3667–3681. http://dx.doi.org/10.1007/s00216-015-8586-5.
[33] SANTE/11945/2015. Guidance document on analytical quality control

and method validation procedures for pesticides residues analysis in food and
feed.
[34] T. Portolés, J.G.J. Mol, J.V. Sancho, F. Hernández, Use of electron

ionization and atmospheric pressure chemical ionization in gas
chromatography coupled to time- of-flight mass spectrometry for screening

and identification of organic pollutants in waters, J. Chromatogr. A. 1339

(2014) 145–153. http://dx.doi.org/10.1016/ j.chroma.2014.03.001.
[35] T. Portolés, J.V. Sancho, F. Hernández, A. Newton, P. Hancock, Potential

of atmospheric pressure chemical ionization source in GC-QTOF MS for
pesticide residue analysis, J. Mass Spectrom. 45 (2010) 926–936.
http://dx.doi.org/ 10.1002/jms.1784.
[36] K. Ziegenhals, H.-J. Hübschmann, K. Speer, W. Jira, Fast-GC/HRMS to

quatify the EU priority PAH, J. Sep. Sci. 31 (2008). http://dx.doi.org/10.1002/
jssc.200700641.
[37] A.-K. Lundebye, H. Hove, A. Mage, V.J. Bohne, K. Hamre, Levels of

synthetic antioxidants (ethoxyquin, butylated hydroxytoluene and butylated
hydroxyanisole) in fish feed and commercially farmed fish, Food Addit.
Contam. 27 (2010) 1652–1657.

En el ámbito de la investigación de contaminantes orgánicos resulta de enorme

importancia diseñar estrategias analíticas integrales capaces de abordar el
análisis de gran variedad de matrices y compuestos de una manera fiable, y lo
más económica, simple y verde posible; para dar respuesta a nuevas
problemáticas medioambientales. Dada la gran cantidad de contaminantes con
propiedades físico-químicas diferentes, resulta interesante desarrollar
estrategias como la descrita en este capítulo, que combine LC y GC para
abarcar el mayor número de analitos posible.
En primer lugar, y como objetivo principal del artículo científico I, se validó

y aplicó un método de screening de pesticidas en ingredientes, piensos y peces
de acuicultura mediante GC-APCI-QTOF MS y LC-QTOF MS. Primero, se
amplió con 39 nuevos pesticidas el screening por GC-APCI-QTOF MS,
validado previamente por Nácher-Mestre et al [15], que contenía 131
pesticidas en las principales matrices de acuicultura. Para ello, al tratarse de
una ampliación del método, se analizaron 10 matrices diferentes fortificadas a
0.01 y 0.05 mg·kg-1 con los 170 pesticidas, tanto bajo condiciones de
transferencia de carga como de protonación. Siguiendo un criterio del 0% de
falsos negativos, dado el reducido número de QCs, se confirmaron los SDLs
de la validación previa y se establecieron SDLs provisionales para los nuevos
39 compuestos. La validación del screening por LC-QTOF MS se llevó a cabo
en veinte matrices diferentes para 125 pesticidas siguiendo la máxima de un
5% de falsos negativos como indica la guía SANCO/12571/2013 [16].
Surgido el interés de incrementar el número de compuestos, esta se amplió en
36 nuevos pesticidas de la misma manera que en la validación por GC. En
conjunto, se establecieron SDLs o SDLs provisionales para 207 y 46
153
compuestos, respectivamente. A efectos prácticos los SDLs provisionales se

trataron como SDLs puramente validados, ya que el criterio que se tomó no
permitía ningún falso negativo. Además, un tercio de los nuevos compuestos
añadidos a la validación se analizaron mediante las dos técnicas, mejorando
así la confianza en la detección.
Tras la validación, ambos métodos de screening se aplicaron en el análisis de

muestras reales. El tratamiento de datos se abordó desde las aproximaciones
target y suspect screening. Para el tratamiento target, se utilizó una base de
datos experimental creada por nuestro grupo de investigación que contenía
información cromatográfica i espectrométrica de los compuestos incluidos en
la validación. Para la aproximación suspect, se utilizó otra base de datos que
únicamente contenía información molecular de contaminantes para los cuales
no se disponía de estándares de referencia. Independientemente del modo de
tratamiento de los datos, se siguieron los criterios de identificación de la guía
SANCO/12571/2013 a excepción de la q/Q ratio, para la cual se adoptó cierta
laxitud, ya que la selectividad instrumental del QTOF se consideró suficiente
para identificar correctamente los compuestos. De hecho, como puede
observarse en la Tabla 1, la última versión de la guía ya no contempla un valor
orientativo de q/Q ratio. De los aproximadamente 800 pesticidas incluidos en
ambas bases de datos, se detectaron 16 de ellos, 12 de los cuales eran
analizables por GC y comprendían 162 de los 167 casos reportados como
identificados o detectados.
A fin de confirmar y cuantificar los compuestos detectados e identificados en

el screening, se desarrolló y validó un método basado en GC-APCI-MS/MS
con analizador QqQ. Además, el método se amplió a PAHs y PCBs debido al
interés en su monitorización para evaluar el cambio de los niveles de
154
concentración en función de la composición de las dietas.
Tabla 1: Resumen de los criterios de identificación para analizadores HRMS

en modo de adquisición full spectrum según las diferentes versiones de la guía
SANCO/SANTE.
SANCO/12571/2013 SANTE/11945/2015 SANTE/11813/2017
2 (incluyendo, al 2 (incluyendo, al 2 (incluyendo, al
Iones
menos, 1 ion menos, 1 ion menos, 1 ion
diagnóstico
fragmento) fragmento) fragmento)
Exactitud de
< 5 ppm ≤ 5 ppm ≤ 5 ppm
masa
Sin valor orientativo
q/Q ratio ± 30 % ±30% a
a
La variabilidad de la q/Q ratio no solo se ve afectada por la relación S/N de los picos en
los XICs, sino que también puede verse afectada por la forma en que se generan los iones
fragmento y por la matriz. Debido al valor añadido de la HRMS, las q/Q ratio son menos
críticas.
Tras una primera etapa de optimización de las condiciones espectrométricas,

se llevó a cabo el estudio de la ionización bajo condiciones de transferencia
de carga y de protonación. La mayoría de los compuestos incluidos en el
método tendían a la protonación de la molécula, excepto los pesticidas
organoclorados y los PCBs, por lo que adquiriendo un compromiso entre
sensibilidad y repetitividad de la señal se decidió trabajar en condiciones
favorables a este mecanismo de ionización. Además, bajo estas condiciones
se observó una mayor selectividad de las transiciones para el C[c,d]P y el
D(a,h)A, resolviendo la problemática en el proceso de cuantificación
comúnmente reportada en métodos de EI [17]. Relativo a la cromatografía, se
decidió utilizar una columna DB-17, de mayor polaridad que la 5-MS utilizada
en el screening, para poder separar cromatográficamente el triplete B(b)F,
B(k)F y B(j)F. Sin embargo, para esta columna se observó una mayor
retención de los PAHs menos volátiles, especialmente los dibenzopirenos. Por
este motivo, y gracias a la posibilidad de trabajar a flujos más altos en sistemas
155
GC-APCI-MS, se estableció una rampa de flujo, manteniendo 2 mL·min-1 en

los dos primeros tercios del cromatograma y subiéndolo a 6 mL·min-1 en el
tercio final, obteniendo así picos cromatográficos más estrechos. Otra
problemática que se observó fue la degradación de la columna en la línea de
transferencia provocando distorsiones cromatográficas a lo largo de la
secuencia (ver figura 1, izquierda). A diferencia de los sistemas de EI, la línea
de transferencia en GC-APCI-MS es más larga, lo que puede provocar la
degradación de una porción significativa de la columna. Este problema se
resolvió utilizando una post-columna desactivada (ver figura 1, derecha).
Figura 1: Efecto de la degradación de la columna sobre la resolución

cromatográfica del triplete B(b)F, B(k)F y B(j)F. A la izquierda, sin post-
columna desactivada; a la derecha, con post-columna desactivada.
156
Una vez optimizadas las condiciones cromatográficas, se llevó a cabo el

estudio del efecto matriz. Como se ha comentado en la introducción general
de la tesis, el efecto matriz en GC se debe principalmente a la interacción de
la matriz con los puntos activos del liner y de la columna; y requiere de
estrategias de corrección para cuantificar correctamente los analitos. Para este
propósito, únicamente se contempló la dilución del extracto final y el uso de
SIL-IS, ya que la intención de la estrategia desarrollada era poder utilizar los
mismos extractos del screening para su posterior cuantificación. En términos
generales, no se observó un efecto matriz significativo en la mayoría de
combinaciones matriz/analito para los extractos diluidos 20 veces, lo que dio
pie a la cuantificación usando curvas de calibración en solvente relativas a
SIL-IS. Este hecho pone en valor el potencial de la fuente APCI y supone un
avance en las aplicaciones analíticas en campos como la acuicultura, donde la
simplificación del método facilita enormemente el análisis de matrices
complejas muy variadas.
Sin embargo, la cuantificación aplicada en este trabajo plantea varias

cuestiones. Por un lado, las relacionadas con la corrección del efecto matriz y,
por otro, las relacionadas con la corrección de las recuperaciones al utilizar
los SIL-IS como surrogate. Idealmente, el uso de SIL-IS para la corrección
del efecto matriz requiere que el compuesto a corregir y el compuesto marcado
corrector tengan un comportamiento similar en el sistema instrumental.
Además, si el SIL-IS se utiliza como surrogate, el compuesto a corregir
también se debe comportar idénticamente durante el tratamiento de muestra.
No obstante, la situación descrita anteriormente es poco habitual ya que, para
que esta ocurra, es prácticamente necesario el uso de SIL-IS de cada
157
compuesto nativo incluido en el método, lo que lo hace poco viable

económicamente. Así pues ¿sería correcto el uso de un compuesto marcado
que no se comporta idénticamente al compuesto que corrige, pero que en
términos de recuperación global, es decir, teniendo en cuenta la suma del
efecto matriz y del tratamiento de muestra, lo corrige correctamente de manera
aparente? A primera vista, parece incorrecto ya que esta corrección podría ser
azarosa, es decir, poco robusta y repetible, aplicable únicamente para una
matriz homogénea concreta. Pero, ¿y si se valida en diferentes matrices,
heterogéneas, y en diferentes días?
Para resolver esta cuestión, se procedió a validar el método. Debido a la gran

cantidad de matrices diferentes a analizar, se decidió validarlo siguiendo los
criterios de la guía SANTE/11945/2015 [18], a 2 y 50 ng·g-1, en salmón y
dorada. La decisión de utilizar solo estas dos matrices se fundamentó en el
hecho de ser representativas de los productos finales del proceso de
producción acuícola. El resto de matrices fueron validadas mediante QCs
individuales a 5 y 50 ng·g-1 inyectados en cada secuencia. El método cumplió
los requisitos de validación en salmón y dorada para la mayoría de compuestos
a ambos niveles. Referente a los QCs individuales, la tabla 2 resume el total
de QCs inyectados en función de su clasificación según la guía
SANTE/11945/2015 [18]. Como puede observarse, pese a la gran variedad de
matrices analizadas, el método se mostró adecuado para la mayoría de
compuestos, demostrando, además, la robustez del uso de señales relativas a
los SIL-IS para su cuantificación.
Finalmente, el método se aplicó al análisis de las muestras. Por lo que respeta

al índice de falsos positivos reportados en el método de screening, todos los
compuestos identificados se confirmaron por GC-APCI-MS/MS, excepto 5
158
casos que no se pudieron confirmar por incompatibilidad cromatográfica o

inadecuación del método. Además, cabe destacar que de los 37 casos
reportados como “detectados” se confirmó la identidad de 22 de ellos.
Únicamente 4 de los detectados por LC no se pudieron identificar por
incompatibilidad con la cromatografía. Referente a los falsos negativos,
gracias a la mayor sensibilidad del acoplamiento GC-APCI-MS/MS se
reportaron 47 nuevos positivos, todos ellos a concentraciones inferiores a los
SDL de los compuestos detectados en el screening. Así pues, la estrategia
desarrollada demostró ser adecuada y de gran utilidad para el análisis de
muestras de acuicultura.
159
160
161
2.5. Referencias
1. Berntssen MHG, Olsvik PA, Torstensen BE, Julshamn K, Midtun T,

Goksøyr A, Johansen J, Sigholt T, Joerum N, Jakobsen J V., Lundebye
AK, Lock EJ (2010) Reducing persistent organic pollutants while
maintaining long chain omega-3 fatty acid in farmed Atlantic salmon
using decontaminated fish oils for an entire production cycle.

Pérez-Sánchez J (2009) Effects of fish oil replacement and re-feeding
on the bioaccumulation of organochlorine compounds in gilthead sea
bream (Sparus aurata L.) of market size. Chemosphere 76:811–817.
doi: 10.1016/j.chemosphere.2009.04.046
3. Berntssen MHG, Julshamn K, Lundebye A-K (2010) Chemical

contaminants in aquafeeds and Atlantic salmon (Salmo salar)
following the use of traditional- versus alternative feed ingredients.

aromatic hydrocarbons in feeds and fish tissues by gas
chromatography coupled to high-resolution mass spectrometry using
atmospheric pressure chemical ionization. J Agric Food Chem
62:2165–74. doi: 10.1021/jf405366n

Kaushik S, Pérez-Sánchez J (2010) Bioaccumulation of polycyclic
163
aromatic hydrocarbons in gilthead sea bream (Sparus aurata L.)

exposed to long term feeding trials with different experimental diets.
Arch Environ Contam Toxicol 59:137–146. doi: 10.1007/s00244-009-
9445-1
6. FAO (2018) El estado mundial de la pesca y la acuicultura 2018.

Cumplir los objetivos de desarrollo sostenible. Roma. Licencia: CC
BY-NC-SA 3.0 IGO
7. Hernández F, Pozo OJ, Sancho J V., Bijlsma L, Barreda M, Pitarch E

(2006) Multiresidue liquid chromatography tandem mass spectrometry
determination of 52 non gas chromatography-amenable pesticides and
metabolites in different food commodities. Journal of Chromatography
A
8. Moreno JLF, Frenich AG, Bolaños PP, Vidal JLM (2008) Multiresidue
method for the analysis of more than 140 pesticide residues in fruits
and vegetables by gas chromatography coupled to triple quadrupole
mass spectrometry. J Mass Spectrom. doi: 10.1002/jms.1400
9. Careri M, Bianchi F, Corradini C (2002) Recent advances in the

application of mass spectrometry in food-related analysis. J.
Chromatogr. A
10. Sapozhnikova Y, Lehotay SJ (2013) Multi-class, multi-residue

analysis of pesticides, polychlorinated biphenyls, polycyclic aromatic
hydrocarbons, polybrominated diphenyl ethers and novel flame
retardants in fish using fast, low-pressure gas chromatography-tandem
mass spectrometry. Anal Chim Acta 758:80–92. doi:
164
10.1016/j.aca.2012.10.034
11. Comisión Europea (2002) Decisión de la Comisión Europea

2002/657/CE, por la que se aplica la Directiva 96/23/CE del Consejo
en cuanto al funcionamiento de los métodos analíticos y la
interpretación de los resultados. D Of las Comunidades Eur 29. doi:
10.7326/L14-5033
12. SANTE/11813/2017 (2017) Guidance document on analytical quality

control and validation procedures for pesticide residues analysis in
food and feed. Eur Comm Heal Consum Prot Dir 2–44. doi:
10.13140/RG.2.2.33021.77283
13. Hernández F, Ibáñez M, Portolés T, Cervera MI, Sancho J V., López

FJ (2014) Advancing Towards Universal Screening for Organic
Pollutants in Waters. J Hazard Mater 282:86–95. doi:
10.1016/j.jhazmat.2014.08.006
14. Lehotay SJ, Sapozhnikova Y, Mol HGJ (2015) Current issues

involving screening and identification of chemical contaminants in
foods by mass spectrometry. TrAC - Trends Anal Chem 69:62–75. doi:
10.1016/j.trac.2015.02.012

aromatic hydrocarbons in feeds and fish tissues by gas
chromatography coupled to high-resolution mass spectrometry using
atmospheric pressure chemical ionization. J Agric Food Chem
62:2165–2174. doi: 10.1021/jf405366n
165

document no. SANCO/12571/2013
17. Veyrand B, Brosseaud A, Sarcher L, Varlet V, Monteau F, Marchand

P, Andre F, Le Bizec B (2007) Innovative method for determination of
19 polycyclic aromatic hydrocarbons in food and oil samples using gas
chromatography coupled to tandem mass spectrometry based on an
isotope dilution approach. J Chromatogr A 1149:333–344. doi:
10.1016/j.chroma.2007.03.043

document no. SANTE/11945/2015
166
CAPÍTULO 3
ESTUDIO DE LA TRANSFERENCIA
DE CONTAMINANTES ORGÁNICOS
A TRAVÉS DE LA DIETA EN
SALMÓN ATLÁNTICO MEDIANTE
CROMATOGRAFÍA DE LÍQUIDOS
ACOPLADA A ESPECTROMETRÍA
DE MASAS EN TÁNDEM
Capítulo 3 Transferencia de contaminantes a través de la dieta
CAPÍTULO 3: ESTUDIO DE LA TRANSFERENCIA DE

CONTAMINANTES ORGÁNICOS A TRAVÉS DE LA DIETA EN
SALMÓN ATLÁNTICO MEDIANTE CROMATOGRAFÍA DE LÍQUIDOS
ACOPLADA A ESPECTROMETRÍA DE MASAS EN TÁNDEM
3.1. Introducción
3.2. Artículo científico III
“LC-MS/MS method for the determination of organophosphorus pesticides

and their metabolites in salmon and zebrafish fed with plant-based feed
ingredients”
Analytical and Bioanalytical Chemistry, 2017.
3.4. Referencias
169
3.1. Introducción
Ante el nuevo escenario generado en el ámbito de la acuicultura en relación a

la composición de los piensos, los trabajos expuestos en el capítulo 2 de la
presente tesis confirmaron la tendencia descrita en estudios anteriores [1,2],
reportando PAHs y pesticidas como los principales contaminantes orgánicos
presentes en ingredientes vegetales y piensos, si bien la transferencia a la parte
comestible del pescado era baja o no detectable. De los aproximadamente 800
pesticidas incluidos en las bases de datos para realizar el screening del
artículo científico I, el pirimiphos-methyl y el chlorpyrifos-methyl fueron los
principales principios activos detectados en ingredientes vegetales y piensos.
De hecho, el 100% de los piensos contenían pirimiphos-methyl.
Pese a estar regulados por la UE en la mayoría de productos alimenticios, sus

MRLs en pescado, piensos e ingredientes de acuicultura son aún inexistentes,
por lo que resulta imprescindible conocer el efecto de estos y de sus
metabolitos en peces de cría para establecer límites apropiados que garanticen
la salud de los peces y la seguridad alimentaria al consumidor. De hecho, el
informe elaborado por la UE “Draft Renewal Assessment Report prepared
according to the Commission Regulation (EU) N° 1107/2009” de septiembre
de 2017 sobre el pirimiphos-methyl [3], no dispone de información sobre
metabolismo y estudios de alimentación para peces, mientras que sí está
disponible para rumiantes, cerdos y aves de corral. Así pues, fruto de los
resultados extraídos de proyectos como ARRAINA y la falta de conocimiento
sobre la transferencia de pesticidas del pienso al pescado [4], en el año 2016
se inició el proyecto “AQUASAFE (254807)”: un proyecto multidisciplinar
que pretende contribuir a la regulación del pirimiphos-methyl mediante su
evaluación toxicológica in vitro e in vivo en salmón atlántico.
171
En el ámbito de la toxicología, la química analítica juega un papel

fundamental, especialmente en los experimentos de exposición in vivo. Estos
requieren de métodos analíticos sensibles y fiables que permitan detectar
compuestos activos y sus metabolitos a bajas concentraciones en matrices
habitualmente complejas para, así, generar datos que permitan crear modelos
toxicológicos de bioacumulación y eliminación. Los metabolitos pueden
presentar una toxicidad semejante al compuesto activo y pueden contribuir a
la toxicidad general, razón por la cual se recomienda su análisis [5]. Además,
también pueden tener propiedades físico-químicas muy diferentes al
compuesto precursor aunque, por lo general, suelen ser más polares. Por este
motivo, la técnica analítica ampliamente usada para su determinación es la
LC.
En el presente capítulo se discute el desarrollo, validación y aplicación de un

método analítico basado en el acoplamiento LC-MS/MS con analizador QqQ
para la determinación de chlorpyrifos-methyl, pirimiphos-methyl y sus
principales metabolitos en muestras de pez cebra y diferentes tejidos de
salmón. El trabajo descrito en el artículo científico III se presenta como una
herramienta para evaluar la cinética de bioacumulación y eliminación del
pirimiphos-methyl que contribuye a ampliar el conocimiento sobre el
comportamiento de este principio activo en salmón sometido a una exposición
dietaria prolongada.
El trabajo realizado en este capítulo se ha desarrollado en el marco del

proyecto AQUASAFE (Evaluation of contaminant interactions using
integrative high-throughput technology for aquafeed safety).
172
3.2. Artículo científico III
Analytical and Bioanalytical Chemistry (2019)
LC-MS/MS method for the determination of organophosphorus

pesticides and their metabolites in salmon and zebrafish fed with plant-
based feed ingredients
Borja Garlito†, María Ibáñez†, Tania Portolés†, Roque Serrano†, Heidi

Amlund#, Anne-Katrine Lundebye§, Monica Sanden§, Marc H.G.
Berntssen§*, Félix Hernández†*
†
#
Technical University of Denmark (DTU), Anker Engelunds Vej 1, 2800
Kgs. Lyngby, Denmark
§
Abstract
The composition of Atlantic salmon feed has changed considerably over the
last two decades from being marine-based (fishmeal and fish oil) to mainly
containing plant ingredients. Consequently, concern related to traditional
persistent contaminants typically associated with fish-based feed has been
replaced by other potential contaminants not previously associated with
salmon farming. This is the case for many pesticides, which are used
worldwide to increase food production, and may be present in plant
ingredients. Earlier studies have identified two organophosphorus pesticides,
B. Garlito et al., Anal. Bioanal. Chem (2019) 173

chlorpyrifos-methyl and pirimiphos-methyl, in plant ingredients used for

aquafeed production. In the present study, we developed a reliable and
sensitive analytical method, based on liquid chromatography coupled to
tandem mass spectrometry, for the determination of these pesticides and their
main metabolites in warm-water (zebrafish) and cold-water (Atlantic salmon)
species, where possible differences in metabolites could be expected. The
method was tested in whole zebrafish and in different salmon tissues, such as
muscle, bile, kidney, fat and liver. The final objective of this work was to
assess kinetics of chlorpyrifos-methyl and pirimiphos-methyl and their main
metabolites in fish tissue, in order to fill the knowledge gaps on these
metabolites in fish tissues when fed over prolonged time.
Keywords
Chlorpyrifos-methyl; Pirimiphos-methyl; Metabolites; Atlantic salmon,

Zebrafish; LC-MS/MS
174 B. Garlito et al., Anal. Bioanal. Chem (2019)

1. Introduction
Plant ingredients are the main substitutes for fish oil and fishmeal and
currently typically constitute about 70% of the ingredients in commercial
salmon feed in Norway (1,2). The use of plant ingredients, together with
commercial decontamination techniques, decreases the content of persistent
organic pollutants (POPs) traditionally associated with fish oil and other
marine ingredients (3,4). However, plant ingredients may introduce novel
contaminants not previously associated with salmon farming (4). Among
them, pesticides are the group of major concern (5). Earlier studies in the EU
projects “AQUAMAX” and “ARRAINA” identified novel feed contaminants,
such as polyaromatic hydrocarbons (PAHs), mycotoxins and none
organochlorine pesticides (OCP) in plant ingredients and fish feed with low
or non-detectable transfer of the parent compounds to the edible part of the
fish (4,5).
Until recently, research has focused on the analysis of organochlorine

compounds in fish, and less information has been available concerning other
groups of pesticides. In the last five years the number of scientific articles
related to pesticides in fish matrices has notably increased reflecting growing
concern regarding these contaminants (6–8). Most recent literature dealing
with pesticide residue analysis is based on the use of liquid chromatography
(LC) coupled to tandem mass spectrometry (MS/MS) with triple quadrupole
(QqQ) (6–10). This technique is applicable for currently used pesticides,
mostly polar in nature, and is especially suitable for metabolites and
transformation products (TPs), which are usually more polar than the parent
compound. LC-MS/MS is a powerful technique in this field due to its excellent
sensitivity and selectivity, as well as robustness and less sample treatment

required (e.g. in comparison with GC-MS methods).
Our previous work indicated that from all new compounds screened,
pesticides were the major contaminants present in novel fish feed (11). Among
more than 400 pesticides investigated, chlorpyrifos-methyl and pirimiphos-
methyl were found in several vegetable feed ingredients as well as in salmon
feed. Further surveillance of commercially produced Norwegian salmon feed
and feed ingredients showed that 55% of the analyzed rapeseed oils contained
pirimiphos-methyl. For most food products, maximum residue levels (MRLs)
for none OCP pesticides have been established in the EU; however, no specific
MRLs have been defined yet for fish or seafood and default precautionary
MRLs are currently applied. Knowledge on the effect of dietary plant-derived
pesticides and their metabolites in Atlantic salmon (Salmo salar) is needed to
set appropriate limits for pesticides to ensure good fish health and food safety.
In a benefit-risk assessment of fish and fish products, it was highlighted that

knowledge on the feed-to-fillet transfer of plant-derived pesticides from feed
to fish is lacking (12). With regards to chlorpyrifos-methyl and pirimiphos-
methyl, one might expect that bioavailability and accumulation are high due
to their lipophilic nature and relatively small molecular size. Bioaccumulation
of chlorpyrifos-ethyl has been reported in body, head and viscera of tilapia
(Oreochromis mossambicus) (13). However, biotransformation may be
crucial in the process of accumulation of the parent compound. Particularly,
for non-persistent pesticides metabolism plays an important role in the
bioavailability and potential transfer to edible parts of fish. It is known that
3,5,6-trichloropyridinol (TCP free and conjugated) is the major metabolite of
both chlorpyrifos-ethyl and chlorpyrifos-methyl in products of animal and
plant origin (14,15), while pirimiphos-methyl is mainly metabolized into 2-

(diethylamino)-6-methyl-4-pyrimidinol (2-DAMP), O-[2-(ethylamino)-6-

methylpyrimidin-4-yl]O,O-dimethylphosphorothioate (N-Des-PM) and 2-
amino-6-methyl-4-pyrimidinol (2-AMP). The first two metabolites are
considered of toxicological significance by the European Food Safety
Authority (EFSA) (16).
In the project “Aquasafe” the main objective was to investigate the

bioaccumulation, biotransformation and elimination kinetics of dietary
chlorpyrifos-methyl and pirimiphos-methyl in whole zebrafish (Danio rerio)
and Atlantic salmon tissues. For this purpose, the present study is aimed to
develop a modern, fast and sensitive analytical method, based on LC-MS/MS
with QqQ, for the quantification of these two pesticides and their main
metabolites in zebrafish, and also in salmon muscle, bile, kidney, fat and liver.
2. Materials and methods
2.1. Chemicals and reagents
Pirimiphos-methyl (PM), chlorpyrifos-methyl (CLP-M), 3,5,6-trichloro-2-

pyridinol (TCP), chlorpyrifos-methyl-oxon (CLP-M-oxon), N-desethyl-
pirimiphos-methyl (N-Deset-PM) and 2-diethylamino-6-methyl-4-
pyrimidinol (2-DAMP) were purchased from Sigma-Aldrich (Pestanal ®
analytical standard, St Louis, MO, USA). Stock standard solutions (around
500 mg·L-1) were prepared in acetone. Working standard solutions containing
all compounds were prepared by dilution of mixtures with acetonitrile. Both
stock standard solutions and working solutions were stored in a freezer at -27
ºC.

Stable Isotopic Labelled Internal Standards (SIL-IS) CLP-M D6, PM D6 and

TCP 13C3 were purchased from Dr. Ehrenstorfer (Augsburg, Germany).
HPLC-grade water was obtained from water passed through a Milli-Q water
purification system (Millipore LTD, Bedford, MA, USA). LC-MS grade
acetonitrile (ACN) and methanol (MeOH), residue analysis grade acetone,
extra pure anhydrous magnesium sulphate (MgSO4), sodium hydroxyde and
LC-MS grade formic acid (FA) were obtained from Scharlau (Barcelona,
Spain). MgSO4 was dried overnight at 300ºC before its use. Leucine
enkephaline was provided by Sigma-Aldrich.
2.2. Instrumentation
UHPLC-MS/MS.
A UPLCTM system (Acquity, Waters, Milford, MA, USA) was interfaced to a

triple quadrupole mass spectrometer (Xevo TQ-S, Waters Corporation,
Manchester, UK). LC separation was performed with a 50 x 2.1mm, 1.7 μm
particle size Acquity UPLC BEH C18 analytical column (Waters). The mobile
phases employed consisted on water (A) and acetonitrile (B) both with
0.0025% HCOOH, at a flow rate of 0.3 mL·min-1. The gradient program
started with 50% B, increased linearly to 90% of B for 1.5 min and maintained
during 1.5 min. Finally the gradient was held to initial conditions in order to
re-equilibrate the column. Temperature column was set to 25ºC. 2 μL were
selected as injection volume.
In the Selected Reaction Monitoring (SRM) method applied, dwell time

values ranging from 5 to 90 ms were used in order to obtain 12 points per

peak. Source temperature was set to 150 ºC. Drying and nebulising gas was
nitrogen (Praxair, Valencia, Spain). Desolvation gas flow was set to 1200 L·h-
1
and the cone gas to 250 L·h-1. For operating in MS/MS mode, argon
(99.995%; Praxair, Valencia, Spain) was used as collision gas at 0.25 mL·min-
1).
Capillary voltage and desolvation gas temperature were set at 3.2 kV (1.9
kV in ESI- mode) and 650ºC respectively. TargetLynx (MassLynx v. 4.1,
Waters, Manchester, UK) software was used to process the quantitative data.
UHPLC-(Q)TOF MS.
UHPLC-(Q)TOF MS analysis was performed following the conditions used

by Portolés et al. [42]. A UPLCTM system (Acquity, Waters) was coupled to a
hybrid QTOF mass spectrometer (XEVO G2, Waters Micromass, Manchester,
UK) with an orthogonal Z-spray electrospray ionization interface. The
chromatographic separation was performed using a Cortecs C18 (Waters)
(100 × 2.1 mm, 2.7 μm) analytical column at a flow rate of 0.3 mL·min-1. The
column temperature was set to 40°C. The mobile phases used were H2O with
0.01% HCOOH (A) and MeOH with 0.01% HCOOH (B) performing a phase
gradient as follows: 10% of B at 0 min, 90% of B at 14 min linearly increased,
90% of B at 16 min, and finally 10% B at 18 min to return to initial conditions.
The injection volume was 20 μL.
For MSE experiments, two acquisition functions with different collision

energies were created and applied sequentialy in each sample injection: the
low energy function (LE), selecting a collision energy of 4 eV, and the second
one, the high energy function (HE), with a collision energy ramp ranging from
15 to 40 eV. The TOF resolution was 20.000 at FWHM at m/z 556,2771.

2.3. Samples
Muscle, liver, kidney, bile and fat tissue samples were obtained from seawater
adapted Atlantic salmon, that was fed with pirimiphos-methyl spiked diets to
a level of 15.2 mg·kg-1 for 81 days. The pirimiphos-methyl was vacuum top
coated to commercially produced (Skretting ARC, Stavanger, Norway)
salmon feed pellets with 2% fish oil at an ambient temperature of 15ºC. No
pirimiphos-methyl was detected in the unspiked feed pellets. Post-smolt
Atlantic salmon (Salmo salar L.) of both genders (SalmoBreed strain) were
distributed among fifteen flow-through fiberglass tanks (100L; 0.80m x 0.95m
x 0.5m). Initial weight and length (fork-tail) were respectively 132 ± 25 g and
18 ± 2 cm (mean ± standard deviation; n = 375). The experiment complied
with the guidelines of the Norwegian Regulation on Animal Experimentation
and EC Directive 86/609/EEC. The experiment was ethically approved by the
Norwegian Animal Research Authority (now the Norwegian Food Safety
Authority; approval number 12091) and performed according to national and
international ethical standards.
2.4. Sample treatment
LC-QTOF MS screening of salmon liver, kidney and muscle
For each matrix, a control sample (not exposed to contaminants) and the most
exposed one to contaminated diets were subjected to a screening analysis, in
order to identify potential metabolites of the pesticides under study To this
aim, 1 g of muscle (0.5 g for liver and kidney) was accurately weighed into a
15 mL Falcon tube and 2 mL of ACN:acetone (80:20) with 1% HCOOH (1
mL in the case of liver and kidney) was added, and the tube was vigorously
shaken by Vortex for 1 min. After that, 0.5 g of MgSO4 per gram of sample

were added and the tube was immediately shaken for 1 min. Subsequently, the
tube was centrifuged at 6,000 rcf for 5 min, and 200 L of the supernatant
were evaporated to dryness at 30ºC under a gentle stream of nitrogen. The
residue was dissolved in 200 L of water and filtered through 0.45 m nylon
filters (Phenomenex, Torrance, CA, USA). Finally, 20 L of the extract was
injected into the LC-QTOF MS system.
LC-MS/MS QqQ analysis (see Figure 1A)
For LC-MS/MS analysis, 1 g of zebrafish or salmon muscle (0.5 g for liver

and kidney, and 0.1 g for fat) was accurately weighed into a 15 mL Falcon
tube (2 mL Eppendorf tube for fat). Then, 2 mL per gram of ACN:acetone
(80:20) with 1% HCOOH were added for zebrafish, muscle, liver and kidney
(5 mL per gram for fat), and the tube was vigorously shaken by Vortex for 1
min. After that, 0.5 g of MgSO4 per gram of sample was added and the tube
was immediately shaken for 1 min. Subsequently, the tube was centrifuged at
6000 rcf for 5 min and 100 L of the supernatant was diluted with 800 L of
water and 100 L of 25 ng·mL-1 SIL-IS solution. Finally, the diluted extracts
were filtered through 0.45 m nylon filters and 2 L were injected into the
LC-MS/MS system.
For the analysis of bile samples, 400 L of ACN:acetone (80:20) with 1%

HCOOH were added to 100 L of bile in a 2 mL Eppendorf tube. The tube
was shaken by Vortex for 1 min and centrifruged at 12600 rcf for 5 min. Then,
250 L of the extract were 4-fold diluted with 650 L of water and 100 L of
25 ng·mL-1 SIL-IS solution. Finally, the diluted extract was filtered through
0.45 m nylon filters and 2 L were injected into the LC-MS/MS system.

The procedure for feed samples was as follows: 1 g of feed was accurately
weighed into a 15 mL Falcon tube. Then, 10 mL of ACN:acetone (80:20) with
1% HCOOH were added and the tube was vigorously shaken by Vortex for 1
min. After that, 0.5 g of MgSO4 was added and the tube was immediately
shaken for 1 min. Subsequently, the tube was centrifuged at 6000 rcf for 5
min. 20 L of the supernatant were diluted with 880 L of water and 100 L
of 25 ng·mL-1 SIL-IS solution. Finally, the diluted extracts were filtered
through 0.45 m nylon filters and 2 L were injected into the LC-MS/MS
system.
Thermal stability experiment (see Figure 1B)
1 g of feed was accurately weighed into a 15 mL Falcon tube (in

quadruplicate). Then, 150 L of a 20 ng·μL-1 standard solution containing
CLP-M and PM were added in each tube and kept aging for 30 min (spiking
level, 3 mg·kg-1). Then, 2 mL of ACN:acetone (80:20) with 1% HCOOH were
added to one tube (QC tube) which was vigorously shaken by Vortex for 1
min. The other three tubes were subjected to the simulated conditions of the
feed production process. To this aim, samples were heated in an oven at 50ºC
for one hour. After that, they were extracted identically to the QC tube. The
samples were centrifuged at 6000 rcf for 5 min and 100 L of the extract were
diluted to 100 mL with water. Finally, 100 L of 25 ng·mL-1 SIL-IS solution
were added to 1 mL of the diluted extract, which was filtered through 0.45 m
nylon filters, and 2 L were injected into the LC-MS/MS QqQ system.

Capítulo 3
B. Garlito et al., Anal. Bioanal. Chem (2019)

Fig. 1 Scheme of the analytical procedure for quantification of pesticides and TPs in solid and bile samples (A) and
183
thermal stability experiment for CLP-M and PM in feed
Transferencia de contaminantes a través de la dieta
2.5. Validation study
Quantitative validation of the method was performed by evaluating the

following parameters:
-Linearity: The calibration curves were obtained by injecting ten reference

standards in solvent (except for bile, where matrix-matched calibration was
applied) in the range 0.025-25 ng·mL-1 at the beginning and the end of the
validation batch. Linearity was assumed when the regression coefficient was
higher than 0.99 with residuals lower than 20% and the difference between
initial and final calibration curves did not exceed 30% (RSD ≤ 30% of the
SIL-IS signal for those compounds whose quantification was carried out using
relative areas).
-Trueness and precision: Trueness was evaluated by means of recovery

experiments, analyzing zebrafish, muscle, liver and bile matrices in
sextuplicates at three concentrations: 1, 10 and 100 µg·kg-1 (ng·mL-1 for bile).
Blank matrices were not available for kidney and fat tissue hence validation
was performed by the analysis of the lowest contaminated samples spiked at
10 and 100 µg·kg-1, and 500 and 5000 µg·kg-1, respectively. No replicates
could be performed for fat tissue, due to the small amount of sample available.
Feed matrix was validated at 500 and 5000 µg·kg-1 due to the characteristics
of the samples. Precision, expressed as the repeatability of the method, was
evaluated in terms of relative standard deviation (RSD) from recovery
experiments at each fortification level (n=6). Quantification was performed by
means of calibration curves in solvent using relative responses to the selected
SIL-IS (see Table 1), except for bile which was quantified using matrix-
matched calibration curves. Recoveries (between 70-120%) and RSDs (below

20%) were considered as satisfactory, according to SANTE/11813/2017

guideline (17).
-Limit of quantification (LOQ) was defined as the lowest concentration

satisfactorily validated, following the SANTE/11813/2017 guideline criteria
(recoveries 70-120 and RSDs < 20%) (17).
-Limit of detection (LOD) was estimated, from the quantification transition,

as the analyte concentration that produced a peak signal with a signal-to-noise
ratio of 3 from the chromatogram at the lowest fortification level.
-Specificity was evaluated by verification of the absence of interfering peaks

at the retention times of each compound in blank samples. To this aim, the
response of a potential peak in the blank sample should be lower than 30% of
the lowest level validated.
3.1. Screening of salmon liver, kidney and muscle using LC-(Q)TOF MS
As stated in the “Reasoned opinion on the review of the existing maximum

residue levels (MRLs) for pirimiphos-methyl according to Article 12 of
Regulation (EC) Nº 396/2005” (16), PM is mainly metabolized into the
metabolites shown in Table S1 in the case of lactating goat milk, muscle, liver,
kidney and fat. However, to our knowledge, information regarding dietary
pirimiphos-methyl metabolism in teleost fish is currently lacking. Therefore,
LC-(Q)TOF MS screening was applied in order to investigate the potential
PM metabolites present in our samples. The Extracted Ion Chromatograms

(XICs) at LE (0.005 Da mass window) were obtained for the theoretical

masses of the (de)protonated molecules of the expected metabolites. As shown
in Figure 2, PM seemed to be metabolized mainly into 2-DAMP (R46382) and
N-Des-PM (R36341) in muscle, liver and kidney. Both metabolites have been
reported to be toxicologically significant by the EFSA in order to generate
appropriate MRLs (16). The other hydroxypyrimidine metabolites reported
in warm-blooded animals were not detected in the fish samples. The identity
of the compounds was determined by comparing the LE and HE spectra with
those of the standards in solvent. Mass errors for the protonated molecules
were in all cases below ±1.5 ppm, and the main fragment ions did not exceeded
±4 ppm mass errors.
Table S1. List of the main pirimiphos-methyl metabolites and their related
structural formula.
Common
IUPAC name Abbreviation Structural formula
name
2-(diethylamino)-4-hydroxy-6-
R46382 2-DAMP
methylpyrimidin
2-(ethylamino)-6-methyl pyrimidin-4-
R35510 2-EAMP
ol
O-[2-(ethylamino)-6-methylpyrimidin-
R36341 N-Des-PM
4- yl] O,O-dimethyl phosphorothioate
R4039 2-amino-4-hydroxy-6-methylpyrimidin 2-AMP
R4041 2-hydroxy-6-methylpyrimidin-4-ol 2-OH-MP

Fig. 2 Screening of salmon liver, kidney and muscle

From the results obtained after screening of metabolites, a LC-MS/MS QqQ

quantitative method was developed for the determination of PM, 2-DAMP and
N-Des-PM in different fish tissues.
3.2. Optimization of LC-MS/MS QqQ conditions
The MS parameters were optimized by direct infusion of 0.1 ng·L-1

individual standard solutions in methanol:water (1:1) 0.01% FA at a flow rate
of 10 μL·min-1 (25 μL·min-1 for CLP and TCP). The optimal cone voltage and
collision energies finally selected are shown in Table 1.
Regarding LC conditions, different mobile phases (H2O, MeOH and ACN)

and additives (HCOOH and NH4OAc) were tested. For most of the
compounds except TCP, sensitivity improved using a mobile phase containing
0.01% HCOOH. Decreasing the HCOOH concentration to 0.0025%,
improved the peak shape for TCP and sensitivity was not substantially
affected. Finally, H2O:ACN with 0.0025% HCOOH was used for the analysis
of samples.
The qi/Q ratio (qi identification transition; Q quantification transition), of the

chromatographic peaks in samples were compared with those of the reference
standard (average value for standard solutions at 1, 5, 10 and 25 ng·mL-1; see
Table 1) for identification of the compounds, with a tolerance in deviations
±30%.

Table 1. Experimental conditions of the optimized UHPLC-ESI-MS/MS

method. Quantification (Q) and identification (qi) ions, collision energy (CE),
cone voltage (CV), qi/Q ratio and linear range.
Linear
Rt Internal Precursor Product CE CV
Compound qi/Q range
(min) standard ion ion (eV) (V)
(ng·mL-1)
0,53 2-DAMP PM D6 Q 182.1 84.1 20 30
q1 109.1 25 0.41
0.025 -
q2 137.1 20 0.38
25
q3 126.1 20 0.21
q4 99.0 20 0.14
1,13 TCP 13C3 Q 203 203 5 20
TCP
1,14 TCP 13C Q 196 196 5 10
3
0.25 - 25
q1 198 197.9 5 0.86
q2 200 199.8 5 0.26
1,24 N-Des-PM PM D6 Q 278,1 125.1 25 40

q1 67.1 35 1.2
0.025 -
q2 108.1 25 1.0
25
q3 246 15 0.62
q4 100.1 20 0.46
2,79 CLP-M D6 Q 330 131 20 20
CLP-M
2,82 CLP-M Q 322 290 15 30
D6
q1 125.1 20 2.6
0.25 - 25
q2 79.1 30 0.62
q3 109.1 20 0.38
q4 212 30 0.21
2,86 PM D6 Q 312,2 284.2 20 10
2,87 PM PM D6 Q 306,1 108.1 30 10

q1 67.1 40 0.90
0.025 -
q2 164.1 20 0.84
25
q3 95.1 25 0.45
q4 136.2 25 0.26
3.3. Sample treatment optimization and matrix effect study
Sample treatment for solid matrices was optimized in order to get the
maximum recovery with the simplest method possible. Different extraction

systems, followed by several clean-up sorbents were tested. Recovery

experiments were carried out at 50 μg·kg-1 in triplicate using salmon fillet as
the reference matrix (spiked samples were aged for 45 min). Quantification
was performed by matrix-matched calibration in each experiment.
The following solvents were firstly tested: ACN, ACN:acetone (80:20) and
ACN:acetone (80:20) containing 1% FA using mechanical agitator for 1 hour.
It was found that CLP-M-oxon was rapidly converted to TCP after spiking the
sample, causing the overestimation of TCP. This instability indicated that
CLP-M-oxon should not be present in the samples, and therefore it was
removed from the analytical method. Using ACN the less polar compounds
(PM and CLP-M) showed low recoveries (68 and 56 %, respectively), which
improved using ACN:acetone (80:20). The addition of 1% FA to the later
solvent mixture improved extraction efficiency (83-103% recoveries) with a
maximum RSD of 11% (see Figure 3.A). Thus, ACN:acetone (80:20) 1% FA
was chosen as the extractant solution in further studies.
Once the extractant was selected, different extraction times and techniques
were evaluated. For this purpose, mechanical agitator (1 hour), vortex (1 min
+ 1 min after adding MgSO4) and ultrasonic assisted extraction (US, 15 min)
were tested, selecting finally 2 min Vortex, as the most suitable and simplest
system (see Figure 3.B). In order to ensure its extraction efficiency, an extra
experiment consisting on the analysis of three spiked samples aged for 2 days
at 7ºC was performed, obtaining recoveries between 83 and 93%, with
RSD<5%.
Several clean-up treatments were also evaluated: Z-Sep, Z-Sep+, freezing and
10-fold dilution. As can be seen in Figure 3.C, 10-fold dilution showed

excellent recoveries (72-108%), with RSD <6%, and was selected for the
analysis of samples.
Prior to the analysis of samples, we performed an evaluation of matrix effects

in the samples under study. To this aim, matrix-matched calibrations were
prepared according to the sample treatment showed in Figure 1, in which 100
µL of the corresponding standard solution in ACN (between 1 and 250
ng·mL−1), instead of 100 µL of the SIL-IS solution, were added to the final
extract, resulting in final analyte concentrations between 0.1 and 25 ng·mL−1.
Matrix effect was evaluated by calculating the relative error between the
slopes of the calibration graphs obtained with standards in solvent and in
matrix (10).
Bile showed strong matrix effects for 2-DAMP and N-Des-PM (77 and 41%
signal suppression, respectively), whereas the rest of the compounds were not
substantially affected (suppression of 7-21%). Despite the notable ionization
suppression observed, the required concentrations were still reached due to the
high sensitivity of the method. In order to compensate matrix effects, the
accurate quantification in bile samples was ensured by using matrix-matched
calibration (with relative responses to SIL-IS only for PM, CLP and TCP).
Regarding salmon fillet and liver, matrix effect ranged 4-28% signal
suppression for 2-DAMP, N-Des-PM, PM and TCP. CLP-M signal was 46
and 50% suppressed in salmon and liver, respectively. Quantification using
calibration in solvent with relative responses to the selected SIL-IS (see Table
1) provided satisfactory results in salmon fillet, liver, fat, kidney and feed.

Fig. 3 Sample treatment optimization (A) extraction solvent, (B) extraction

technique and (C) clean-up treatment. Percentage recoveries are calculated as
means of triplicate experiments at 50 μg·kg-1
3.4. Method validation
Validation of the method was carried out with zebrafish, salmon tissues (fillet,
liver, kidney, bile and fat), and salmon feed. The study of linearity in solvent
revealed that correlation coefficients (R2) were higher than 0.99 with residuals
lower than 20% for 2-DAMP, N-Des-PM and PM in the range 0.025-25
ng·mL-1 and 0.25-25 ng·mL-1 for CLP-M and TCP. Matrix matched
calibration for bile analysis also showed correlation coefficients (R2) higher
than 0.99 with residuals lower than 20% for PM and its TPs in the range 0.025-
25 ng·mL-1, and 0.25-25 ng·mL-1 for CLP-M and TCP.

Blank samples were pre-analyzed (except salmon kidney and salmon fat
which were not available) in order to ensure the absence of interfering peaks
at the retention time of the analytes of study. The method was found to be
highly specific as no relevant signals were observed.
Trueness and precision data are shown in Table 2. For zebrafish, salmon
muscle, liver and bile, recoveries ranged from 72 to 106%, with RSD ≤ 16%,
for PM and its metabolites; and from 71 to 112%, with RSD ≤ 12%, for CLP-
M and TCP. CLP-M and TCP could only be validated at 10 and 100 g·kg-1
(ng·mL-1) spiking levels. Although no EU regulations exist for marine
products, the concentrations tested were lower than the precautional maximum
residue limits (MRLs). Thus, LOQs were established at 1 g·kg-1 (ng·mL-1 in
bile) for PM and its metabolites, and 10 g·kg-1 for CLP-M and TCP. For
these matrices, LODs were in the range 0.1 – 0.6 g·kg-1 (ng·mL-1 in bile) and
2.5 – 8,0 g·kg-1 (ng·mL-1 in bile), respectively.
Blank samples were not available for kidney and fat tissue. Consequently,
analyzed samples with the lowest contamination levels were subsequently
spiked for validation experiments, at a level at least three times the
concentration present. Recoveries were then calculated by subtracting “blank”
concentration. In kidney, the spiking levels were 10 and 100 g·kg-1 as the
concentrations of PM and 2-DAMP in the “blank” sample were 2.4 and 2.9
g·kg-1, respectively. Trueness and precision were estimated in sextuplicates,
obtaining recoveries between 70 – 82% (RSD < 9%) for PM and its
metabolites, and 72 - 87% (RSD < 15%) for CLP-M and TCP. LODs were
calculated from the “blank” samples used. Fat could be validated by a single
QC spiked at 500 and 5000 g·kg-1 due to the low amount of sample available.
The spiking levels were selected based on the concentrations found in the

“blank” samples (666, 56.5 and 102 g·kg-1 for PM, 2-DAMP and N-Des-PM,
respectively). Recoveries ranged 71 to 105%.
Salmon feed was validated at 500 and 5000 µg·kg-1 as the experimental design
of the study established 3000 µg·kg-1 as the approximated concentration of
PM and CLP for feeding trials. Recoveries ranged 74 – 84% with RSD<6%
for parent compounds. N-Des-PM and TCP were not evaluated as they were
not of interest in the analysis.
Table 2. Validation of the analytical method. Mean recoveries (%) and RSD
(%, in brackets) of the overall procedure (n=6). Estimated limits of detection
(LOD).
Zebrafish (g·kg-1) Salmon muscle (g·kg-1)

1 10 100 LOD 1 10 100 LOD
PM 73 (6) 89 (7) 91 (4) 0.1 84 (7) 82(7) 83 (11) 0.1
2-DAMP 96 (8) 83 (9) 83 (6) 0.2 97 (10) 94 (10) 101 (14) 0.2
N-Des-PM 91 (13) 87 (9) 82 (8) 0.3 82 (6) 86 (9) 87 (11) 0.6
CLP-M -a 80 (6) 81 (6) 2.5 -a 90 (12) 86 (8) 2.0
TCP -a 71 (2) 105 (10) 3.3 -a 112 (9) 100 (12) 8.0
Salmon liver (g·kg-1) Salmon bile (ng·mL-1)

1 10 100 LOD 1 10 100 LOD
PM 77 (1) 80 (4) 82 (9) 0.1 90 (6) 94 (9) 106 (5) 0.1
2-DAMP 91 (10) 83 (4) 72 (10) 0.3 89 (16) 89 (10) 101 (15) 0.3
N-Des-PM 107 (10) 95 (2) 92 (6) 0.4 75 (5) 72 (11) 77 (5) 0.3
CLP-M -a 111 (4) 104 (8) 2.0 -a 100 (9) 107 (4) 1.8
TCP -a 96 (3) 105 (7) 7.6 -a 92 (11) 107 (4) 4.5
Salmon kidney (g·kg-1) Salmon fat (µg·kg-1) Feed (µg·kg-1)

10 100 LOD 500 5000 LOD 500 5000 LOD
PM 80 (9) 82 (6) 0.1 -b 87 d 74 (2) 75 (2) d
d d
2-DAMP 73 (3) 70 (5) 0.3 96 105 85 (3) 92 (4)
d
N-Des-PM 80 (8) 78 (5) 0.4 94 89 -c -c d
d d
CLP-M 72 (15) 74 (7) 2.0 81 78 81 (2) 84 (6)
d c c d
TCP 80 (6) 87 (15) 7.3 71 95 - -
a
Limit of detection > lowest spiking level (1 g·kg-1 (ng·mL-1)).
b
Blank concentration > spiking level.
c
Compounds out of interest from the purpose of the analysis.
d
Very high concentrations to calculate LODs

3.5. Thermal stability study of parent compounds
In order to assess the thermal stability of CLP and PM in feed, a trial

simulating the conditions employed in feed production process (1 h, 50ºC) was
carried out. The experiment was performed in triplicate, and results were
compared with a QC that was not subjected to elevated temperature. The
percentage of pesticide degradation was calculated by using Equation 1:
% recovery Trial
% degradation = 100 − × 100
% recovery QC
As shown in Table S2, CLP and PM did not show relevant degradation at the
production temperature conditions, with partial degradation of 15 and 17%,
respectively. It was found that PM was degraded to 2-DAMP, generating a
considerable background in the final diets (see Table 3).
Table S2. Thermal stability results. Numbers in bold are the mean values of
% degradation for PM and CLP with their RSD (in brackets).
Pirimiphos-methyl Chlorpyrifos-methyl
Recovery Degradation Recovery Degradation
mg·kg-1 mg·kg-1
(%) (%) (%) (%)
QC 2.27 75.6 2.47 82.3
Feed 1 1.85 61.6 18 2.07 69.2 16
Feed 2 1.90 63.3 16 2.09 69.6 15
Feed 3 1.88 62.8 17 2.13 71.1 14
17 (6) 15 (8)
3.6. Quantitative analysis of samples in dietary exposed fish
The developed method was applied for the analysis of zebrafish samples,
salmon fillet, salmon liver, salmon kidney, salmon bile and diets. A reagent

blank, a reagent blank spiked with SIL-IS (to evaluate SIL-IS stability), a
blank (non-spiked) sample and 9 spiked samples (3 at each validation level)
were included in each batch. Each matrix was analyzed in different batches.
The results summarized in Table 3 corresponding to the analysis of solid fish
tissues are expressed in a wet weight basis, whereas those which correspond
to bile analysis, are expressed in ng·mL-1. The qi/Q ratios obtained for all
positive samples were in agreement with those of the reference standards with
deviations lower than the maximum tolerance accepted (30%). This data
confirmed the identity of the compounds in samples according to the
SANTE/11813/2017 guideline (17).
In whole zebrafish fed with CLP-M, TCP was the main metabolite and was
present in higher concentrations (approximately two fold higher) than the
parent compound. For PM, both 2-DAMP and N-Des-PM metabolites were
identified in whole zebrafish, but at lower levels than the parent compound
(see Figure 4). As for zebrafish, also for Atlantic salmon the main PM
metabolites were 2-DAMP and N-Des-PM. The distribution of the PM
metabolites showed highest concentrations for 2-DAMP, higher than the
parent compound, in the liver which is likely the main organ of metabolisation.
This is confirmed by the higher concentrations of 2-DAMP in the bile. The
second metabolite, N-Des-PM, was found in all tissues (muscle, liver, kidney)
at concentrations in the same range (3-6 µg·kg-1). The parent compound, PM,
had highest levels in the fat tissue. Similarly, EFSA concludes that PM in
commodities of animal origin is fat soluble and, in goat, parent pirimiphos-
methyl was the main compound, accounting for 55 % of the total radioactive
residue (16).

Fig. 4 UHPLC-(ESI)-MS/MS chromatograms obtained for the quantification

and identification of A) PM (5.6 μg·kg-1), B) 2-DAMP (2.1 μg·kg-1), C) N-
Des-PM (0.6 μg·kg-1), D) CLP-M (5.5 μg·kg-1) and E) TCP (16.0 μg·kg-1); in
zebrafish samples. Q: quantification transition; qi: identification transitions. ✓
q/Q within accepted tolerances

Table 3. Concentration of PM, 2-DAMP, N-Des-PM, CLP-M and TCP in

zebrafish, salmon muscle, liver, kidney, fat and bile.
(g·kg-1) PM 2-DAMP N-Des-PM CLP-M TCP

Trial 1 n.d 0.7 n.d 5.5 16.0
Zebrafish Trial 2 5.6 2.1 0.6 n.d n.d
Trial 4 n.d 0.6 n.d n.d n.d
Trial 1 n.d 200 n.d 1600 n.d

Feed Trial 2 2300 500 n.d n.d n.d
Trial 4 n.d 200 n.d n.d n.d
muscle T81 28.2 18.0 3.4 n.d n.d

liver T81 6.1 28.8 5.9 n.d n.d
kidney T81 22.4 16.8 3.1 n.d n.d
Fat T81 2757 75.5 346 n.d n.d
(ng·mL-1) PM 2-DAMP N-Des-PM CLP-M TCP

Bile T81 48.0 205 89.1 n.d n.d
n.d: not detected. Concentration < LOD
4. Conclusions
A fast, simple and sensitive method for the determination of PM, CLP-M and
their main metabolites in different fish tissues has been developed. Previous
LC-(Q)TOF screening demonstrated that cold-blooded fish show a different
metabolism of PM than in warm-blooded animals, with 2-DAMP and N-Des-
PM being the most abundant metabolites in salmon. This was supported by
analysis performed in the present work. The application of this method to
zebrafish fed with CLP-M also allowed the identification of TCP as the most
abundant metabolite. This work has generated analytical information essential
for developing a kinetic model of accumulation and elimination of PM in
salmon, and will contribute to establish relevant MRLs for fish.

Acknowledgements
The present work was financed by the Norwegian Research Council (NFR)
project “AQUASAFE” (254807). The authors wish to thank the staff at the
Skretting ARC, Lerang Research Station for their excellent help with
sampling and conducting the feeding trial. Tania Portolés acknowledges
Ramon y Cajal Program from Ministry of Economy and Competitiveness
(RYC-2017-22525) for funding her research.
Compliance with ethical standards
Conflict of interest: The authors state that there is no conflict of financial and
non-financial interest.
Research involving animals: The experiment complied with the guidelines

of the Norwegian Regulation on Animal Experimentation and EC Directive
86/609/EEC. The National Animal Research Authority approved the protocol
(ID 12091).
References
1. Shepherd CJ, Jackson AJ. Global fishmeal and fish-oil supply: Inputs,
outputs and marketsa. J Fish Biol. 2013;83(4):1046–66.
2. Ytrestøyl T, Aas TS, Åsgård T. Resource utilisation of Norwegian salmon

farming in 2012 and 2013. Nofima Report 36/2014. 2014. Available from:
https://nofima.no (accessed on 10th September 2019).

3. Berntssen MHG, Lundebye A-K, Torstensen BE. Reducing the levels of

dioxins and dioxin-like PCBs in farmed Atlantic salmon by substitution of fish
oil with vegetable oil in the feed. Aquac Nutr. 2005;11(3):219–31.
4. Berntssen MHG, Julshamn K, Lundebye A-K. Chemical contaminants in

aquafeeds and Atlantic salmon (Salmo salar) following the use of traditional-
versus alternative feed ingredients. Chemosphere. 2010;78(6):637–46.
5. Nácher-Mestre J, Serrano R, Portolés T, Berntssen MHG, Pérez-Sánchez J,

Hernández F. Screening of pesticides and polycyclic aromatic hydrocarbons
in feeds and fish tissues by gas chromatography coupled to high-resolution
mass spectrometry using atmospheric pressure chemical ionization. J Agric
Food Chem. 2014;62(10):2165–74.
6. Kaczyński P, Łozowicka B, Perkowski M, Szabuńko J. Multiclass pesticide

residue analysis in fish muscle and liver on one-step extraction-cleanup
strategy coupled with liquid chromatography tandem mass spectrometry.
Ecotoxicol Environ Saf. 2017;138:179–89.
7. Belenguer V, Martinez-Capel F, Masiá A, Picó Y. Patterns of presence and

concentration of pesticides in fish and waters of the júcar river (eastern spain).
J Hazard Mater. 2014;265:271–9.
8. Zhang H, Wang J, Li L, Wang Y. Determination of 103 Pesticides and Their

Main Metabolites in Animal Origin Food by QuEChERS and Liquid
Chromatography–Tandem Mass Spectrometry. Food Anal Methods.
2017;10(6):1826–43.
9. Gan J, Lv L, Peng J, Li J, Xiong Z, Chen D, et al. Multi-residue method for

the determination of organofluorine pesticides in fish tissue by liquid

chromatography triple quadrupole tandem mass spectrometry. Food Chem.

2016;207:195–204.
10. Botero-Coy AM, Marín JM, Serrano R, Sancho JV, Hernández F.

Exploring matrix effects in liquid chromatography-tandem mass spectrometry
determination of pesticide residues in tropical fruits. Anal Bioanal Chem.
2015 May;407(13):3667–81.
11. Portolés T, Ibáñez M, Garlito B, Nácher-Mestre J, Karalazos V, Silva J, et

al. Comprehensive strategy for pesticide residue analysis through the
production cycle of gilthead sea bream and Atlantic salmon. Chemosphere.
2017;179:242–53.
12. VKM. Benefit-risk assessment of fish and fish products in the Norwegian
diet – an update. Scientific Opinion of the Scientific Steering Committee.
VKM Report 15 [293 pp] 2014. Available online: www.vkm.no (accessed on
10th September 2019)
13. Rao JV, Rani CHS, Kavitha P, Rao RN, Madhavendra SS. Toxicity of
chlorpyrifos to the fish Oreochromis mossambicus. Bull Environ Contam
Toxicol. 2003;70(5):985–92.
14. European Food Safety Authority. Modification of the existing MRLs for
chlorpyrifos-methyl in various crops. EFSA J 2011;9(6):2219.
doi:10.2903/j.efsa.2011.2219. Available online: www.efsa.europa.eu/
efsajournal (accessed on 10th September 2019)
15. European Commission. Refined risk assessment regarding certain

maximum residue levels (MRLs) of concern for the active substance
chlorpyrifos. European Food Safety Authority (EFSA) 2015, Available

online: https://www.efsa.europa.eu/en/efsajournal/pub/4142 (accessed on 10th

September 2019)
16. EFSA (European Food Safety Authority). Reasoned opinion on the review
of the existing maximum residue levels (MRLs) for pirimiphos-methyl
according to Article 12 of Regulation (EC) No 396/2005. EFSA J
2015;13(1):3974. Available from: http://doi.wiley.com/10.2903/j.efsa.2015.
3983 (accessed on 10th September 2019)
17. SANTE/11813/2017. Guidance document on analytical quality control

and validation procedures for pesticide residues analysis in food and feed.
Available online: https://ec.europa.eu/food/sites/food/files/plant/docs/
pesticides_mrl_guidelines_wrkdoc_2017-11813.pdf. (accessed on 10th
September 2019)

El artículo científico III surge a partir de los resultados obtenidos en los

artículos científicos I y II, con la intención de estudiar en profundidad el
metabolismo y la cinética de bioacumulación y eliminación del pirimiphos-
methyl en salmón atlántico y, así, contribuir a la armonización de la
legislación del principio activo en toda la cadena de producción acuícola. Para
ello, se desarrolló un método basado en el acoplamiento LC-MS/MS con
analizador QqQ para el análisis de pez cebra y diferentes tejidos de salmón
sometidos a varios experimentos de exposición dietaria.
En primer lugar, tras optimizar las condiciones espectrométricas y

cromatográficas, se llevó a cabo la optimización del tratamiento de muestra
en filete de salmón para el pirimiphos-methyl, el chlorpyrifos-methyl y los
principales metabolitos reportados por la EFSA que estaban disponibles
comercialmente [6,7]. Como se muestra en la figura 3 del artículo, el mejor
compromiso entre sencillez, tiempo y recuperaciones se obtuvo en la
extracción con ACN:acetona 80:20 al 1% de HCOOH durante dos minutos en
vórtex y una dilución final de diez veces. Llama la atención las bajas
recuperaciones obtenidas mediante sonicación, quizás provocadas por la
degradación de los analitos en el proceso de cavitación [8–10].
Una vez optimizado el tratamiento de muestra, se extrajeron y se realizó un

screening mediante LC-QTOF MS de las muestras más expuestas a los
contaminantes según el diseño experimental, a fin de estudiar los posibles
metabolitos del pirimiphos-methyl presente en filete, hígado y riñones de
salmón dada la ausencia de antecedentes científicos. Como se muestra en la
figura 2 del artículo, los principales metabolitos detectados fueron el 2-DAMP
203
y el N-Des-PM. Cabe mencionar que, pese a ser poco habitual la optimización

del tratamiento de muestra en métodos de screening, la decisión de hacerlo
estuvo motivada por la poca cantidad de riñón e hígado disponible, de modo
que el mismo extracto crudo se pudo aprovechar para fines cualitativos y
cuantitativos.
A partir de los resultados obtenidos, se validó el método cuantitativo para la

determinación de pirimiphos-methyl, 2-DAMP, N-Des-PM, chlorpyrifos-
methyl y TCP en pienso, pez cebra y filete, hígado, riñon, bilis y grasa de
salmón; cumpliendo con los parámetros de validación establecidos por la guía
SANTE/11813/2017 [11]. En el proceso de validación, el estudio previo del
efecto matriz fue determinante para la selección de la estrategia de
cuantificación dada la complejidad de las matrices. Por una parte, la bilis
presentó una fuerte supresión de la señal para el 2-DAMP y el N-Des-PM que
no se pudo corregir con los SIL-IS disponibles, lo que obligó a compensarla
mediante calibrado en matriz. Por el contrario, el resto de matrices, pese a
presentar supresión de señal para algunos compuestos, se pudieron cuantificar
correctamente a los niveles deseados usando calibrado en solvente con
respuestas relativas a los SIL-IS.
Por último, se analizaron las muestras de los experimentos de exposición

dietaria en dos tandas. En la primera de ellas se determinaron todos los analitos
mencionados en el párrafo anterior en muestras de pienso y pez cebra. En la
segunda de ellas, únicamente se determinaron el pirimiphos-methyl y sus
metabolitos en filete, hígado, riñón, bilis y grasa de salmón. La razón principal
se debe al diseño de los experimentos de exposición dietaria en base a criterios
toxicológicos. Por una parte, al experimento con pez cebra se aplicó un
enfoque multicontaminante, es decir, una aproximación más cercana a la
204
exposición ambiental que contempla las posibles interacciones sinérgicas o

antagónicas de la mezcla de pirimiphos-methyl, chlorpyrifos-methyl y los
adyuvantes que contienen los formulados comerciales. Por otra parte, y
siguiendo las directrices de guías como la SANCO/11187/2013 o la
reglamentación 440/2008(EC) para estudios de metabolismo y toxicocinéticos
[12,13], el experimento con salmón se llevó a cabo desde un enfoque
unicontaminante, es decir, expuesto únicamente a pirimiphos-methyl.
Si bien es cierto que, desde el punto de vista analítico, el artículo científico

III puede resultar poco novedoso; no es menos cierto que, en un presente en
el que no se pueden entender las ciencias ambientales fuera de un contexto
multidisciplinar, este trabajo ha generado información analítica - la mayor
parte de ella sin publicar - esencial para desarrollar, en un futuro inmediato,
un modelo cinético de acumulación y eliminación de pirimiphos-methyl en el
salmón que contribuirá a establecer MRLs seguros y objetivos en peces de
acuicultura.
205
3.4. Referencias
[1] J. Nácher-Mestre, R. Serrano, T. Portolés, M.H.G. Berntssen, J. Pérez-

Sánchez, F. Hernández, Screening of pesticides and polycyclic aromatic
hydrocarbons in feeds and fish tissues by gas chromatography coupled to high-
resolution mass spectrometry using atmospheric pressure chemical
ionization., J. Agric. Food Chem. 62 (2014) 2165–74. doi:10.1021/jf405366n.
[2] M.H.G. Berntssen, K. Julshamn, A.-K. Lundebye, Chemical

use of traditional- versus alternative feed ingredients, Chemosphere. 78 (2010)
637–646. doi:10.1016/j.chemosphere.2009.12.021.
[3] European Commission, Draft Renewal Assessment Report prepared

according to the Commission Regulation (EU) N° 1107/2009 - Pirimiphos-
methyl. List of endpoints, 2009.
[4] VKM. Benefit-risk assessment of fish and fish products in the

Norwegian diet – an update. Scientific Opinion of the Scientific Steering
Committee. VKM Report 15 [293 pp] 2014. Available online: www.vkm.no
(accessed on 10th September 2019).
[5] EFSA (European Food Safety Authority), Scientific Opinion on

Evaluation of the Toxicological Relevance of Pesticide Metabolites for
Dietary Risk Assessment, EFSA J. 10 (2012) 1–187. doi:10.2903/j.efsa.
2012.2799.
[6] EFSA (European Food Safety Authority), Conclusion on the peer

review of the pesticide human health risk assessment of the active substance
chlorpyrifos, EFSA J. 12 (2016) 1–34. doi:10.2903/j.efsa.2014.3640.
[7] EFSA (European Food Safety Authority). Reasoned opinion on the
207
review of the existing maximum residue levels (MRLs) for pirimiphos-methyl

according to Article 12 of Regulation (EC) No 396/2005. EFSA J
2015;13(1):3974. Available from: http://doi.wiley.com/10.2903/j.efsa.2015.
3983 (accessed on 10th September 2019).
[8] V. Andruch, M. Burdel, L. Kocúrová, J. Šandrejová, I.S. Balogh,

Application of ultrasonic irradiation and vortex agitation in solvent
microextraction, TrAC - Trends Anal. Chem. 49 (2013) 1–19. doi:10.1016/
j.trac.2013.02.006.
[9] B. Miljevic, F. Hedayat, S. Stevanovic, K.E. Fairfull-Smith, S.E.

Bottle, Z.D. Ristovski, To sonicate or not to sonicate PM filters: Reactive
oxygen species generation upon ultrasonic irradiation, Aerosol Sci. Technol.
48 (2014) 1276–1284. doi:10.1080/02786826.2014.981330.
[10] Y.G. Adewuyi, Sonochemistry: Environmental Science and

Engineering Applications, Ind. Eng. Chem. Res. 40 (2001) 4681–4715.
[11] SANTE/11813/2017. Guidance document on analytical quality control

and validation procedures for pesticide residues analysis in food and feed.
Available online: https://ec.europa.eu/food/sites/food/files/plant/docs/
pesticides_mrl_guidelines_wrkdoc_2017-11813.pdf. (accessed on 10th
September 2019).
[12] Comisión Europea, REGLAMENTO (CE) 440/2008, D. Of. La Unión

Eur. 142 (2008) 1–739.
[13] European Commission, SANCO/11187/2013. Working document on

the nature of pesticide residues in fish, (2013) 1–35.
208
CAPÍTULO 4
DETERMINACIÓN DE COMPUESTOS
ENDÓGENOS MEDIANTE EL USO DE
ESPECTROMETRÍA DE MASAS CON
ANALIZADOR DE CUADRUPOLO-
TIEMPO DE VUELO Y FUENTE DE
IONIZACIÓN APCI
Capítulo 4 Determinación de compuestos endógenos
CAPÍTULO 4: DETERMINACIÓN DE COMPUESTOS ENDÓGENOS

ANALIZADOR DE CUADRUPOLO-TIEMPO DE VUELO Y FUENTE DE
IONIZACIÓN APCI
4.1. Introducción
4.2. Artículo científico IV
“Identification of very long-chain (>C24) fatty acid methyl esters using gas
chromatography coupled to quadrupole/time-of-flight mass spectrometry with
atmospheric pressure chemical ionization source”
Analytica Chimica Acta, 1051, 103-109, 2019
4.4. Referencias
211
4.1. Introducción
Actualmente, el pescado representa la principal fuente de ácidos grasos

poliinsaturados de cadena larga de la serie ω3 (LC-PUFA ω3) para el consumo
humano, cuya importancia ha sido puesta de manifiesto en numerosos estudios
que resaltan el papel fundamental de estos ácidos grasos en diversos procesos
fisiológicos y metabólicos [1]. Esta circunstancia ha motivado, en gran
medida, el estudio del metabolismo de LC-PUFA en peces y sus mecanismos
de regulación con fines aplicados a la acuicultura. De hecho, estos compuestos
son protagonistas ineludiblemente asociados al cultivo de peces puesto que
esta actividad depende, durante las fases tempranas de cultivo, del adecuado
aporte dietario de LC-PUFA ω3 dada la capacidad limitada de biosíntesis que
tienen las especies marinas para satisfacer sus requerimientos fisiológicos [2].
A pesar de la enorme atención recibida por los LC-PUFA, como los ácidos
eicosapentaenoico (20:5ω3, EPA), docosahexaenoico (22:6ω3, DHA) y
araquidónico (20:4ω6, ARA); otros ácidos grasos poliinsaturados
denominados de cadena muy larga (C > 24) (very long-chain polyunsaturated
fatty acids, VLC-PUFA) han sido prácticamente inexplorados en peces, lo
cual es particularmente sorprendente si se tiene en cuenta que, como
postularon Agbaga et al. [3], la síntesis de VLC-PUFA en mamíferos se
produce por elongaciones sucesivas de LC-PUFA como el ácido
docosatetraenoico (22:4ω6) y DHA en las que la elongasa de ácidos grasos de
cadena muy larga 4 (ELOVL4) tiene un papel determinante. Además, varios
estudios llevados a cabo en mamíferos, han demostrado la presencia de VLC-
PUFA en células fotorreceptoras de la retina [4–6], cerebro [7, 8] y testículos
[9, 10], y se cree que estos compuestos están asociados a procesos vitales
clave. De hecho, se ha demostrado que mutaciones genéticas en el gen
213
ELOVL4 reduce el contenido de VLC-PUFA en la retina, provocando una

pérdida de capacidad visual [6]. Pese a que la función de los VLC-PUFA en
estos tejidos aún no se ha determinado, parece tener que ver con el hecho de
que comparten propiedades de los ácidos grasos saturados en el extremo
carboxílico proximal y de los poliinsaturados en la región distal de la molécula
[11].
Relativo al ámbito de la acuicultura, estudios llevados a cabo en el pez cebra

(Danio rerio) [12], en la cobia (Rachycentron canadum) [13], en el salmón
atlántico (Salmo salar) [14] y en el pez conejo (Siganus canaliculatus) [15],
han permitido demostrar que las ELOVL4 de peces también participan en el
metabolismo de VLC-PUFA expresado en tejidos como el ojo, las gónadas y
el cerebro. Estos resultados resaltan la relevancia del estudio de las elongasas
ELOVL4 y de los VLC-PUFA en especies de peces cultivados, donde
alteraciones en la capacidad visual - determinante en animales predadores
como la mayor parte de especies de acuicultura -, problemas de infertilidad en
reproductores, y disrupciones en el funcionamiento cerebral pueden afectar al
rendimiento de la producción acuícola.
Desde un punto de vista meramente analítico, la detección y caracterización

de los VLC-PUFA han sido escasamente exploradas. La mayor parte de las
referencias consultadas aplican el acoplamiento GC-MS con fuente EI para el
análisis de los VLC-PUFA derivatizados [16], principalmente en forma de sus
ésteres metílicos (VLC-PUFA-MEs) [17]. Sin embargo, la alta fragmentación
provocada por la fuente EI implica la pérdida total y/o parcial del M+· y genera
el mismo espectro de masas para PUFA-MEs de las series ω3 y ω6
independientemente de la longitud de cadena [18]. Estos hechos, junto a la
ausencia de patrones de referencia comercialmente disponibles, dificultan el
214
proceso de identificación de los compuestos, lo que ha motivado el estudio de

la aplicación del acoplamiento GC-MS con fuente APCI como alternativa
instrumental para su análisis.
En este capítulo se explora el potencial del acoplamiento GC-APCI-QTOF

MS para la detección e identificación de VLC-PUFA en la fracción de
fosfatidilcolinas (PC) de retinas de diferentes especies cultivadas. Como se
expone a lo largo del artículo científico IV, en primer lugar, se lleva a cabo
un estudio del comportamiento espectrométrico de los ésteres metílicos de
ácidos grasos saturados y LC-PUFA para los cuales se dispone de estándares
de referencia. A partir de este estudio, mediante herramientas estadísticas, se
genera un modelo que permite diferenciar las series ω3 de las ω6 en base a sus
patrones de fragmentación y se aplica al análisis de muestras reales. Por
último, con el fin de confirmar la identidad de los compuestos detectados y
validar la estrategia estadística de identificación, se lleva a cabo el análisis de
diferentes experimentos de expresión heteróloga, es decir, se analizan los
productos de elongación in vitro de diferentes LC-PUFA generados por la
elongasa ELOVL4.
215
4.2. Artículo científico IV
B. Garlito et al., Analytica Chimica Acta 1051 (2019) 103-109 217

Analytica Chimica Acta 1051 (2019) 103-109
Identification of very long-chain (>C24) fatty acid methyl esters using

gas chromatography coupled to quadrupole/time-of-flight mass
spectrometry with atmospheric pressure chemical ionization source
B. Garlito a, T. Portolés a, *, W.M.A. Niessen b, J.C. Navarro c, d,

F. Hontoria c,d, Ó. Monroig e, I. Varó c, d, R. Serrano a, d
a
b
Hyphen MassSpec, Margrietstraat 34, 2215 HJ, Voorhout, the Netherlands
c
Instituto de Acuicultura Torre de La Sal (IATS-CSIC), Ribera de Cabanes,
S/N, 12595, Cabanes, Castellón, Spain
d
Associated Unit of Marine Ecotoxicology (IATS-IUPA), Ribera de
Cabanes, S/N, 12595, Cabanes, Castellón, Spain
e
Institute of Aquaculture, Faculty of Natural Sciences, University of Stirling,
Stirling, FK9 4LA, Scotland, UK
Abstract
Gas chromatography coupled to a quadrupole time-of-flight mass analyzer

(QTOF) with an atmospheric pressure chemical ionization (APCI) source has
been tested to study the ionization and mass spectrometric behavior of long-
chain and very long-chain polyunsaturated fatty acids (LC-PUFAs, C18-24;
VLC-PUFAs, >C24). The protonated molecule ([M+H]+), measured at
accurate mass, became the base peak of the spectrum for all the studied
compounds and was promoted by the addition of water into the source. This
fact overcame the existing difficulties for the identification of VLC-PUFAs
218 B. Garlito et al., Analytica Chimica Acta 1051 (2019) 103-109

when using an electron ionization source (EI). The extensive fragmentation of

PUFAs in this source is the main drawback due to the fact that since reference
standards are not commercially available, final identification relies on
retention time estimation. The application of GC-APCI-QTOF to the
screening of lipid extracts from the eyes of different fish species added
confidence to the identification of several VLC-PUFAs. Further investigation
of ion ratios allowed to predict the position of key double bonds enabling the
classification of VLC-PUFAs as ω3 or ω6 compounds. VLC-PUFAs spectra
found in biological samples were compared to those obtained from
corresponding peaks found in heterologous expression experiments of fish's
Elovl4.
Highlights
► First application of GC-APCI-QTOF MS for determination of VLC-PUFA

methyl esters. ► Accurate mass [M+H]+ in APCI spectrum: additional value
for confident identification. ► Reliable identification of VLC-PUFAs without
the need of reference standards. ► Development of ion ratio procedure to
distinguish among ω3 and ω6 congeners. ► A step forward to the knowledge
of the metabolism of these compounds.
Keywords
Gas chromatography, Quadrupole time-of-flight mass spectrometry,

atmospheric pressure chemical ionization source, Very long-chain
polyunsaturated fatty acids.

Graphical abstract

1. INTRODUCTION
Fatty acids (FAs), besides being a source of energy, are essential compounds
involved in different physiological and metabolic processes, like cell
membrane formation or signal transduction, that ensure the correct cellular
functioning [1]. Among FAs, certain long-chain (C18-24) polyunsaturated fatty
acids (LC-PUFAs) including arachidonic acid (20:4ω6), eicosapentaenoic
acid (20:5ω3) and docosahexaenoic acid (22:6ω3) play pivotal roles in
inflammatory response and function and in development of the central nervous
system [2]. Several studies have demonstrated the influence of LC-PUFAs in
neural development, neurodegenerative diseases, and visual system in
mammals [3,4].
Besides, the very long-chain (>C24) polyunsaturated fatty acids (VLC-

PUFAs), biosynthesized in vertebrates by FA elongase enzymes such as
Elovl4 [5], have been regarded as essential for cognitive development, retinal
function and reproductive processes [6–10]. Consequently, VLC-PUFAs are
typically found in low concentrations in specific lipid classes such as
phosphatidylcholine (PC), sphingomyelin and ceramides in brain, retina and
gonads of vertebrates [11], and their importance has been related to their
structure combining a proximal carboxylic region similar to saturated FAs and
a distal region with PUFA character [8]. In addition to being present in specific
lipid classes from certain tissues in very small amounts compared to their
shorter chain counterparts, the lack of standards and references in mass
spectrometry libraries have involved an analytical challenge for their
determination and characterization.
Liquid and gas chromatography (LC and GC) coupled to mass spectrometry

(MS) have been the techniques chosen to carry out both quantitative and
qualitative analysis. On the one hand, LC-MS techniques based on
electrospray ionization (ESI) or atmospheric pressure chemical ionization
(APCI) have been commonly applied for the analysis of lipid species
containing VLC PUFAs [8]. To achieve this, LC in normal and reversed phase
are used but long chromatographic runs are required to achieve good
chromatographic separation between polar species [12,13]. In order to
facilitate the quantitative analysis of total VLC-PUFAs, hydrolysis must be
performed leading to LC-ESI-MS in negative-ion mode the required
technique. However, the drawbacks of low specificity in the negative-ion MS
detection mode and requirement of the post-column addition of a cationizating
agent after acidic chromatographic separation have hindered a simple way of
FA analysis. Thus, derivatization of FA must be finally applied to facilitate
their analysis by LC-ESI-MS in positive-ion mode [14].
On the other hand, while a derivatization step is also required, GC is the most
frequently used technique in this field. To improve the separation
performance, comprehensive two dimensional gas chromatography (GC×GC)
has also been applied to clarify the identity of fatty acid isomers [15].
Although selective GC stationary phases like polyethylene glycols or
cyanopropyl silicones (medium-high polarity) are widely applied for the
analysis of FAMEs (C < 24), for the study of VLC-PUFA 5%-(phenyl)-
methylpolysiloxane columns are more convenient as the applicable upper
temperature limit (350ºC) is higher than the selective ones (260ºC). Under
these conditions, VLC-PUFA (C > 24) can be analyzed avoiding column
degradation. GC-MS with electron ionization (EI) is frequently applied for the
analysis of VLC-PUFAs [16] since it provides complete information about the
fragmentation patterns of FA derivatives. Among others, methyl esters

(FAMEs) are commonly used in FA analysis [17]. However, the extensive

fragmentation of FAMEs in EI source involves the partial or total loss of the
highly diagnostic molecular ion (M+) and generates the same EI spectra for
unsaturated FAMEs (ω6 or ω3) with different chain length [18]. This fact,
together with the absence of reference standards, hinders the identification
process of VLC-PUFAs. In this context, APCI combined with GC-MS
instruments has revealed as a useful alternative to EI. As it has been
demonstrated in several fields [19–24], the soft and universal ionization in this
source provides abundant M+ and/or [M+H]+ in the mass spectra, thus
substantially improving the identification capability, as well as enhancing
sensitivity and selectivity for the detection of analytes in the samples [25,26].
Under these circumstances, the use of GC-APCI-QTOF MS will add
confidence to the identification of VLC-PUFA.
Since their detection in bovine retinas by Aveldaño et al. in 1987 [27], the
study of VLC-PUFAs has been mainly focused on mammals [9,11,12,28,29].
However, the detection and characterization of these compounds in fish has
been scantily studied despite being the main source of ω3 LC-PUFAs in
human’s diet [30]. Studies in fish have focused on elucidating the molecular
mechanisms underlying the biosynthesis of VLC-PUFAs, particularly the
roles that elongase enzymes such as Elovl4 play in these metabolic pathways
[2,31].
The aim of this work is the development of a reliable analytical methodology

for detection and identification of VLC-PUFAs, for which no commercially
available standards exist, in the PC fraction of lipid extracts from the eyes of
different fish species. The potential of GC-APCI-QTOF MS to provide
accurate mass measurement of intact M+ and/or [M+H]+, was considered to

add confidence in the identification step. For this purpose, a range of

commercially available saturated and polyunsaturated LC-PUFA methyl
esters were analyzed by GC-APCI-QTOF MS and their ionization and
fragmentation behavior was studied. Additionally, relationships between the
fragmentation patterns and FAME structures have been studied in order to
differentiate ω3 from ω6 VLC-PUFAs’ series without the need of reference
standards. As a final step, in order to confirm the identity of detected
compounds, heterologous expression experiments were carried out. To the
best of our knowledge, GC-APCI-QTOF MS has not been applied for the
determination of VLC-PUFA methyl esters yet.
2. EXPERIMENTAL
2.1 Standard compounds, solvents and reagents
Methyl stearate (18:0-ME), methyl arachidate (20:0-ME), methyl behenate

(22:0-ME) and methyl tetracosanoate (24:0-ME) standards were purchased
from Fluka (Zwijndrecht, The Netherlands). Methyl hexacosanoate (26:0-
ME), methyl octacosanoate (28:0-ME) and methyl triacontanoate (30:0-ME)
standards were purchased from Dr. Ehrenstorfer (Augsburg, Germany).
Arachidonic acid methyl ester (20:4ω6-ME), cis-5,8,11,14,17-
eicosapentaenoic acid methyl ester (20:5ω3-ME), cis-7,10,13,16-
docosatetraenoic acid methyl ester (22:4ω6-ME), cis-7,10,13,16,19-
docosapentaenoic acid methyl ester (22:5ω3-ME), cis-4,7,10,13,16,19-
docosahexaenoic acid methyl ester (22:6ω3-ME), 20:4ω3-ME, 22:4ω3-ME,
and 24:5ω3 standards were purchased from Supelco (Bellefonte, PA, USA).

Stock solutions (around 500 g·mL-1 and 1 g·mL-1 for 24:0-ME and 26:0-
ME) were prepared by dissolving solid reference standards in hexane or by
diluting reference standard solutions in hexane and subsequently stored in a
freezer at -20 ºC under a N2 atmosphere. Working solutions were prepared by
diluting stock solutions in hexane.
Hexane (ultra-trace quality) was purchased from Scharlab (Barcelona, Spain).

Chloroform (CHCl3), diethyl ether and toluene were purchased from Merck
Millipore (Darmstadt, Germany). Methanol (MeOH) was purchased from
VWR (Radnor, PA, USA). Sulfuric acid, glacial acetic acid and potassium
chloride (KCl) were purchased from Panreac (Castellar del Vallés, Barcelona,
Spain). Iodine and butylated hydroxytoluene (BHT, content >99%), used as
antioxidant, were purchased from Sigma Aldrich (St. Louis, MO, USA). All
reagents were analytical grade. HPLC-grade water was obtained by purifying
double distilled water in a Milli-Q Gradient A10 system (Millipore, Bedford,
MA, USA).
2.2 Samples
Fish eyes
Twelve eyes from adult specimens of fish species including the European
seabass (2), gilthead seabream (1), Atlantic salmon (1) and Senegalese sole
(8) were dissected and cleaned from other tissues. Next, the crystalline lens
was removed and stored at -20 ºC for further treatment.

Elongated fatty acids by Elovl4 protein
Further biological samples containing VLC-PUFAs were generated by

expressing the coding region of the Atlantic salmon Elovl4 in yeast
Saccharomyces cerevisiae which was grown in the presence of one of the
following polyunsaturated fatty acid (PUFA) substrates: 20:4ω3, 22:4ω6,
22:5ω3, 22:6ω3, 24:5ω3. A previous study confirmed that the Atlantic salmon
Elovl4 could efficiently elongate some of these PUFA substrates to produce
polyenes up to 36 carbons [32]. After 48 h, yeast samples were harvested and
washed twice with double distilled water and kept at -20 ºC until further
treatment [32]. Given that Elovl4 enzymes add 2-carbon units to FA substrates
[33], all these PUFA substrates provided a set of step-wise elongation products
that could be easily identified in the chromatogram since both FA substrates
and products are absent in yeast.
2.3 Sample treatment
Total lipids were extracted according to the method of Folch et al. [34].
Briefly, samples (fish eyes and yeast) were homogenized in CHCl3:MeOH
(2:1, v/v) with BHT (0.01%, w/v). After addition of 0.25 % (v:v) aqueous
solution of KCl (0.88%, w/v), the lower organic phase was filtered (70 m
Whatman filter paper), evaporated to dryness under a stream of nitrogen and
kept in desiccator overnight. The extracts were then weighed to determine the
total amount of lipids, redissolved in CHCl3:MeOH (2:1) (10 mg/mL) and
stored at -20 °C.
Phospholipid classes were separated following the method developed by

Olsen et al. [35] consisting in 20 cm × 20 cm TLC plates (Silica gel G 60

(Merck, Darmstadt, Germany) run in a polar solvent system of methyl
acetate:isopropanol:CHCl3:MeOH:0.25% (w/v) KCl (25:25:25:10:9 by
volume). A well known Artemia nauplii extract [36] run in parallel to the
samples allowed the location of the PC fraction on the plates after using a 1%
(w/v) iodine solution in CHCl3 for lipid dying. PC spots were scrapped off and
submitted to acid-catalyzed transesterification [37]. Resulting fatty acid
methyl esters (FAMEs) were stored at -20 °C in hexane/BHT (0.01%) under
nitrogen, until chromatographic analysis.
2.4 Instrumental
GCAPCI-(Q)TOF MS
An Agilent 7890 N gas chromatograph (Palo Alto, CA, USA) equipped with
an Agilent 7683 autosampler was coupled to a quadrupole orthogonal
acceleration time-of-flight mass spectrometer, Xevo G2 QTOF (Waters
Corporation, Manchester, UK), operated in APCI mode. A fused silica HP-
5MS capillary column with a length of 30 m × 0.25 mm i.d. and a film
thickness of 0.25 µm with a 1 m × 0.25 mm i.d. deactivated post-column (J&W
Scientific, Folson, CA, USA) was used for GC separation. The oven
temperature was programmed as follows: 60 ºC (1 min); 40 ºC/min to 180ºC;
10 ºC/min to 320 ºC (4 min). Pulsed splitless (30 psi) injections of 1 L were
carried out at 280 ºC with a splitless time of 1 min. Helium 99.999% (Praxair,
Spain) was used as carrier gas at a flow of 2 mL·min-1.
The interface and ionization source temperatures were set to 320 ºC and 150

ºC, respectively. N2 was used as auxiliary gas at 175 L·h-1, as cone gas at 25
L·h-1 and as make-up gas at 300 mL·min-1. The APCI corona discharge pin
was operated at 1.2 A and the cone voltage was set to 20 V.
The QTOF was operated at 2.5 spectra·s-1 acquiring the mass range m/z 50–
650. The TOF MS resolution was approximately 15,000 (FWHM) at m/z 264.
Acquisition was done in MSE mode in which two alternating acquisition
functions with different collision energies were generated: the low-energy
(LE) function, selecting a collision energy of 4 eV to avoid or minimize
fragmentation, and the high-energy (HE) function, with a collision energy
ramp ranging from 25 to 40 eV to obtain a greater range of fragment ions.
Perfluorotributylamine (Sigma Aldrich, Madrid, Spain) was used for the daily
mass calibration. Internal calibration was performed using a background ion
from the GC-column bleed as lock mass (protonated molecule of
hexamethylcyclotrisiloxane, m/z 223.0642).
In order to work under proton transfer conditions, an uncapped vial containing

water was placed in a designed holder into the APCI source door to enhance
protonation. MS data were acquired in centroid mode and processed by the
ChromaLynx XS application manager (within MassLynx v4.1; Waters).
Mass-Fragment software (Waters) was used for mass spectra interpretation.
3. RESULTS AND DISCUSSION
3.1 Ionization behavior of FAMEs
The ionization behavior was tested using all commercially available FAME

standards with a focus on their differences in the degree of unsaturation and

position of the last double bond: saturated (18:0-ME, 20:0-ME, 22:0-
ME,24:0-ME, 26:0-ME, 28:0-ME and 30:0-ME) and unsaturated (20:4ω3-
ME, 20:4ω6, 20:5ω3-ME, 22:4ω3-ME, 22:4ω6, 22:5ω3-ME, 22:6ω3-ME,
and 24:5ω3-ME). As it has been described in previous studies, two ionization
mechanisms can take place simultaneously in the APCI source: charge transfer
promoting the formation of M+• and/or protonation [38]. In many cases,
protonation is directly produced by the water traces present in the N2 supply.
However, it can be enhanced using protic solvents as modifier [20]. Some
experiments regarding analyte ionization were made in order to evaluate the
effect of the addition of modifiers in the source.
Despite no protic modifier was initially added into the source, the protonated
molecule [M+H]+ was the base peak in the spectrum for all analytes. For
saturated FAMEs, the molecular ion M+• was also formed from 39 to 86%,
showing a higher tendency of protonation when the number of carbon atoms
decreased (see Figure 1a up). In the case of ω3 and ω6 FAMEs, M+• was
generated from 40 to 75% and 23 to 71%, respectively, showing a higher
tendency of protonation when the number of carbon atoms decreased and the
unsaturation degree increased (see Figure 1b-c up)
Taking into account the proneness of FAMEs to protonation, a subsequent

experiment using water as modifier in the source was carried out. In this way,
the proton transfer mechanism was enhanced, decreasing the relative
abundance of M+• (Figure 1 down). M+• still appeared in the spectra, but only
in a percentage among 7-10% for saturated FAMEs (with the same protonation
tendency as for charge transfer conditions), 8-12% for ω6 and 7-9% for ω3
FAMEs showing higher protonation affinity of the smallest FAMEs with

highest degree of unsaturation. Therefore, because of both the better

sensitivity of protonated molecule and higher spectral purity, it was decided
to use water as modifier for further experiments.
As a consequence of the soft ionization character of the APCI source, low in-
source fragmentation of FAMEs was observed if compared with their
corresponding EI spectra taken from NIST library. As it has been reported in
other studies, EI spectra of saturated FAMEs are dominated by the ion with
m/z 74 [CH2C(OH)OCH3]+• resulting from a McLafferty rearrangement
(Figure 2a), whereas the ion with m/z 79 [C6H7]+ is usually the base peak in
mass spectra of FAMEs with three or more methylene-interrupted double
bonds (Figure 2c) [17]. Exploring the APCI spectra of saturated FAMEs, no
in-source fragment ions were observed. Thus, [M+H]+ is completely
preserved (Figure 2b). Regarding polyunsaturated ω6 and ω3 FAMEs, all of
them exhibited [M+H-CH3OH]+ in-source fragments with relative abundances
between 4-11% and 2-6%, respectively. In addition, less abundant [M+H-
CH3OH-H2O]+ fragments were also found for all polyunsaturated FAMEs
(Figure 2d).
Figure 1: APCI spectra at the LE function of (a) 20:0-ME (saturated); (b)

C22:63-ME; and (c) C20:46-ME under charge transfer (up) and
protonation conditions (down).

Figure 2: Comparison of the in-source fragmentation for arachidic acid ME

(up) and C20:53-ME (down) in EI (left) and APCI (right) sources.
3.2 Fragmentation behavior of FAMEs
FAMEs share a common methyl ester group. The oxygen atoms are the only
structural features with lone electron pairs, thus being the most feasible sites
for protonation. Taking into account the proton affinity of the carbonyl and
methyl ether groups, the expected protonation site would be the oxygen atom
of the carbonyl function. However, the degree of unsaturation plays an
important role in ionization/fragmentation behavior, as is described below.
3.2.1 Group I. Saturated FAMEs
All saturated FAMEs showed [M+H]+ as the base peak of the spectrum at LE
due to the protonation of the carbonyl function. At HE conditions, the spectra
are characterized by the presence of the ion with m/z 103.0759 (C5H11O2+) as
the base peak, with a subsequent neutral loss of HCOOH producing the
fragment ion with m/z 57.0704 (C4H9+). For this class of compounds, although
being unusual, apparently charge-remote fragmentation occurs and series of

C2nH4n losses from the hydrocarbon chain are observed.
3.2.2 Group II. ω6 and ω3 FAMEs
All ω6 and ω3 FAMEs showed a common fragmentation pathway showing

fragment ions with the same m/z, irrespective the ω6 or ω3 character. After
studying it at the HE function, none of the fragment ions by itself was specific
for either the ω3 or the ω6, as in most cases only abundance differences
seemed to be observed. This constitutes an inconvenience in the
characterization of VLC-PUFAs for which no reference standards are
(commercially) available. The presence of the [M+H]+ in the LE spectrum
together with the fragmentation information of the HE spectrum allows to
tentatively identify the VLC-PUFA in terms of carbon chain length and the
degree of unsaturation. At this point, it was considered to use of abundance
ratios to distinguish between the ω6 and ω3 compounds. For that, all the
spectral information available for the commercial reference standards was
used to create a data matrix that contained the abundance for each m/z ion.
Data were normalized to the ion with m/z 121 which was the base peak of
spectrum from the 22:43 and 22:46. Inverse values were then calculated in
order to work with values above 1. Among them, there were standards with
different chain lengths (C20-C24) and different degrees of unsaturation (4, 5
and 6) for ω3 while only C20 and C22 with a degree of unsaturation of 4 were
available for ω6.
Firstly, considering a completely unbiased analysis, a principal component

analysis (PCA) was applied over the training data matrix. Figure 3a illustrates
the PCA obtained in which a rough distinction between ω3 and ω6 can be
appreciated. The next step consisted in performing an orthogonal partial least

squares-discriminant analysis (OPLS-DA), which considers additional

information about the two groups during classification. Thus, an OPLS-DA
was applied to construct a statistical model that highlight differences in ion
ratios between the two groups. As it can be seen, the OPLS-DA analysis
separated the two groups of samples. For this purpose, an S-Plot graphic for
ω3 vs ω6 was obtained (Figure 3b) and studied looking for endpoints.
Theoretically, in an S-Plot the points with a p-corr value between 0.9 and 1
are closely related to the compounds characterizing the samples of one group,
which constitute important class markers. According to this, only one marker
(ion ratio for the ions with m/z 121 and 163) was selected. Returning to the
data, we could observe that 3 compounds used for the model construction
presented ion ratios in the range of 4-10 while 6 were found to be in the
range of 17-24. Thus, this ion-ratio procedure was used for further
experiments.
Figure 3: (a) PCA plot corresponding to the reference standards analyzed to

create the model; (b) SPLOTs -3 versus -6.
3.3 Analysis of adult fish eyes samples
The system (GCAPCI-QTOF MS in MSE mode) was used for the eye
extracts analysis. Data were processed using ChromaLynx XS software in

combination with a homemade compound database that contained the

elemental composition, retention time and fragments information previously
obtained from the study with commercially available reference standards (C20-
C22). Additionally, this database was extended with those FAMEs reported in
the literature, mainly VLC-PUFAs for which reference standards were not
commercially available (C24-C34), and whose reliable identification is
considered an additional value given by GCAPCI-QTOF MS. The
ChromaLynx XS application manager obtained eXtracted Ion Chromatograms
(XICs) at LE and HE functions (0.02 Da window) for m/z values
corresponding to the theoretical exact masses of the protonated molecules and
the fragments (if known), respectively.
Table 1 summarizes the positive findings of FAMEs detected in the fish eye
samples analyzed from the different species. Detection was initially based on
the presence of [M+H]+ in the LE function. Further investigation of fragment
ions from the HE function together with the application of the ion-ratio
procedure developed allowed their tentative identification. As an example,
Figure 4 illustrates the detection and tentative identification of 32:63-ME in
a gilthead sea bream eyes sample by GCAPCI-QTOF MS. The protonated
molecule of 32:6-ME with m/z 483.4197 (C33H55O2+) was detected in the LE
function with a mass error of 1.0 ppm. As the reference standard was not
available, chemical structures for the most abundant fragment ions were
suggested based on their accurate masses, using the Mass-Fragment software.
All structures proposed for the fragments were compatible with the chemical
structure of 32:6-ME making the identification even more reliable. At this
point, the ion-ratio procedure developed was applied in order to decide
whether it was 3 or 6. The ratio of m/z 121 and m/z 163 was calculated to

Capítulo 4
B. Garlito et al., Analytica Chimica Acta 1051 (2019) 103-109

Figure 4: Detection and identification of 32:63-ME in a gilthead sea bream eyes sample by
GC–QTOF MS (the reference standard was not available at our laboratory at the time of the
detection): (a) Extracted-ion chromatograms (0.02 Da window) for protonated molecule in LE
function and different fragment ions in HE function; (b) LE (bottom) and HE (top) spectra of
235
Determinación de compuestos endógenos
be a value of 4.6, which thus indicates to correspond to 3, being 32:63-ME.
the compound eluting at 18.6 min. Proposed elemental composition for fragment ions.
Table 1. List of PUFAs identified in the fish eye samples analyzed from the
different species
European Gilthead Senegalese Atlantic

Compounds
seabass seabream sole salmon
20:46    
20:53    
22:63    
22: 56    
22:53    
24:63    
26:63  
28:63  
30:63   
32:63    
34:63  
3.4 Analysis of heterologous expression experiments and validation of

the ion-ratio procedure
In order to confirm the identity of detected compounds and validate the

developed ion-ratio procedure for the characterization of 3 and 6 VLC-
PUFAs, heterologous expression experiments were carried out in order to
establish retention times for VLC-PUFAs and compare the obtained spectra.
As it has been explained in the introduction, Elovl4 proteins enable the
elongation of PUFAs including, among others, 20:46 (arachidonic acid),
20:53 (eicosapentaenoic acid) and 22:63 (docosahexaenoic acid) [2]. In

order to expand our VLC-PUFA database, 20:43, 22:46, 22:53, 22:63

and 24:53 FAs were individually used as elongation substrates for the
Atlantic salmon Elovl4, a protein that had previously shown elongase capacity
towards all these compounds [32]. Figure 5 shows the XIC for the elongation
products coming from 22:63 (up) and 22:53 (down). All the elongation
products and substrates were evaluated and the information obtained allowed
the final identification of the expected compounds.
Figure 5: Extracted ion chromatograms for the elongation products coming

from 22:63 (up) and 22:53 (down).

4. CONCLUSIONS
The use of an APCI interface in GC–QTOF has been employed for the first
time to study the ionization and fragmentation behavior of FAMEs. The
addition of water in the source as modifier promoted the proton transfer
reaction producing mainly [M+H]+ species. In contrast to EI, low in-source
fragmentation occurred in APCI. This fact together with the preservation of
the protonated molecule measured at accurate mass has given an additional
value to the reliable identification of VLC-PUFAs in real samples. The ratio
between the ions with m/z 121 and m/z 163 has been used for differentiation
of 3 and 6 congeners. In addition, the heterologous expression experiments
have enabled expanding our VLC-PUFA spectral library, and also establishing
retention times for non-commercially available standards, thus improving
characterization, identification and confirmation purposes. Results obtained in
this paper will help to advance into the knowledge of the metabolism of these
compounds.
REFERENCES
[1] P.C. Calder, N-3 Fatty Acids, Inflammation and Immunity: New
Mechanisms To Explain Old Actions, Proc. Nutr. Soc. 72 (2013) 326–36.
doi:10.1017/S0029665113001031.
[2] O. Monroig, J. Rotllant, J.M. Cerdá-Reverter, J.R. Dick, A. Figueras, D.R.

Tocher, Expression and role of Elovl4 elongases in biosynthesis of very long-
chain fatty acids during zebrafish Danio rerio early embryonic development,
Biochim. Biophys. Acta - Mol. Cell Biol. Lipids. 1801 (2010) 1145–1154.

doi:10.1016/j.bbalip.2010.06.005.
[3] C.I.F. Janssen, A.J. Kiliaan, Long-chain polyunsaturated fatty acids

(LCPUFA) from genesis to senescence: The influence of LCPUFA on neural
development, aging, and neurodegeneration, Prog. Lipid Res. 53 (2014) 1–17.
doi:10.1016/j.plipres.2013.10.002.
[4] E.J. Johnson, E.J. Schaefer, Potential role of dietary n-3 fatty acids in the
prevention of dementia and macular degeneration, Am. J. Clin. Nutr. 83
(2006) 1494S–1498S.
[5] G. Schmitz, J. Ecker, The opposing effects of n-3 and n-6 fatty acids, Prog.
Lipid Res. 47 (2008) 147–155. doi:10.1016/j.plipres.2007.12.004.
[6] M.I. Aveldano, A novel group of very long chain polyenoic fatty acids in
dipolyunsaturated phosphatidylcholines from vertebrate retina, J. Biol. Chem.
262 (1987) 1172–1179.
[7] A. Poulos, Very long chain fatty acids in higher animals—A review,
Lipids. 30 (1995) 1–14. doi:10.1007/BF02537036.
[8] M.-P. Agbaga, M.N. a Mandal, R.E. Anderson, Retinal very long-chain
PUFAs: new insights from studies on ELOVL4 protein, J. Lipid Res. 51
(2010) 1624–42. doi:10.1194/jlr.R005025.
[9] P. Barabas, A. Liu, W. Xing, C.-K. Chen, Z. Tong, C.B. Watt, B.W. Jones,
P.S. Bernstein, D. Križaj, Role of ELOVL4 and very long-chain
polyunsaturated fatty acids in mouse models of Stargardt type 3 retinal
degeneration., Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 5181–6. doi:10.1073/
pnas.1214707110.

[10] R. Harkewicz, H. Du, Z. Tong, H. Alkuraya, M. Bedell, W. Sun, X.

Wang, Y.H. Hsu, J. Esteve-Rudd, G. Hughes, Z. Su, M. Zhang, V.S. Lopes,
R.S. Molday, D.S. Williams, E. a. Dennis, K. Zhang, Essential role of
ELOVL4 protein in very long chain fatty acid synthesis and retinal function,
J. Biol. Chem. 287 (2012) 11469–11480. doi:10.1074/jbc.M111.256073.
[11] A. Liu, R. Terry, Y. Lin, K. Nelson, P.S. Bernstein, Comprehensive and

sensitive quantification of long-chain and very long-chain polyunsaturated
fatty acids in small samples of human and mouse retina., J. Chromatogr. A.
1307 (2013) 191–200. doi:10.1016/j.chroma.2013.07.103.
[12] O. Berdeaux, P. Juaneda, L. Martine, S. Cabaret, L. Bretillon, N. Acar,

Identification and quantification of phosphatidylcholines containing very-
long-chain polyunsaturated fatty acid in bovine and human retina using liquid
chromatography/tandem mass spectrometry, J. Chromatogr. A. 1217 (2010)
7738–48. doi:10.1016/j.chroma.2010.10.039.
[13] R. Taguchi, M. Ishikawa, Precise and global identification of

phospholipid molecular species by an Orbitrap mass spectrometer and
automated search engine Lipid Search, J. Chromatogr. A. 1217 (2010) 4229–
4239. doi:10.1016/j.chroma.2010.04.034.
[14] X. Li, A.A. Franke, Improved LC-MS Method for the Determination of
Fatty Acids in Red Blood Cells by LC-Orbitrap MS, (2011) 3192–3198.
[15] L. Tedone, R. Costa, S. De Grazia, S. Ragusa, L. Mondello,

Monodimensional (GC-FID and GC-MS) and comprehensive two-
dimensional gas chromatography for the assessment of volatiles and fatty
acids from Ruta chalepensis aerial parts, Phytochem. Anal. 25 (2014) 468–

475. doi:10.1002/pca.2518.
[16] M. Volpato, J.A. Spencer, A.D. Race, A. Munarini, A. Belluzzi, A.J.

Cockbain, M.A. Hull, P.M. Loadman, A liquid chromatography–tandem mass
spectrometry method to measure fatty acids in biological samples, J.
Chromatogr. B. 1055–1056 (2017) 125–134. doi:10.1016/j.jchromb.2017.04.
030.
[17] J. Pettersen, Determination of trans double bonds in polyunsaturated fatty

acid methyl esters from their electron impact mass spectra. Eur. J. Lipid Sci.
Technol. 105 (2003) 156–164.
[18] D.-P. Kloos, E. Gay, H. Lingeman, F. Bracher, C. Müller, O. a

Mayboroda, A.M. Deelder, W.M. a Niessen, M. Giera, Comprehensive gas
chromatography-electron ionisation mass spectrometric analysis of fatty acids
and sterols using sequential one-pot silylation: quantification and
isotopologue analysis., Rapid Commun. Mass Spectrom. 28 (2014) 1507–14.
doi:10.1002/rcm.6923.
[19] T. Portolés, L. Cherta, J. Beltran, F. Hernández, Improved gas

chromatography-tandem mass spectrometry determination of pesticide
residues making use of atmospheric pressure chemical ionization, J.
Chromatogr. A. 1260 (2012) 183–192. doi:10.1016/j.chroma.2012.08.009.
[20] M. Raro, T. Portolés, J. V Sancho, E. Pitarch, F. Hernández, J. Marcos,

R. Ventura, C. Gómez, J. Segura, O.J. Pozo, Mass spectrometric behavior of
anabolic androgenic steroids using gas chromatography coupled to
atmospheric pressure chemical ionization source. Part I: ionization, J. Mass
Spectrom. 49 (2014) 509–21. doi:10.1002/jms.3367.

[21] E. Hurtado-Fernández, T. Pacchiarotta, E. Longueira-Suárez, O. a

Mayboroda, A. Fernández-Gutiérrez, A. Carrasco-Pancorbo, Evaluation of
gas chromatography-atmospheric pressure chemical ionization-mass
spectrometry as an alternative to gas chromatography-electron ionization-
mass spectrometry: avocado fruit as example., J. Chromatogr. A. 1313 (2013)
228–44. doi:10.1016/j.chroma.2013.08.084.
[22] R. García-Villalba, T. Pacchiarotta, A. Carrasco-Pancorbo, A. Segura-

Carretero, A. Fernández-Gutiérrez, A.M. Deelder, O.A. Mayboroda, Gas
chromatography-atmospheric pressure chemical ionization-time of flight
mass spectrometry for profiling of phenolic compounds in extra virgin olive
oil, J. Chromatogr. A. 1218 (2011) 959–971. doi:10.1016/j.chroma.2010.
12.014.
[23] K. Hennig, J.P. Antignac, E. Bichon, M.-L. Morvan, I. Miran, S.

Delaloge, J. Feunteun, B. Le Bizec, Steroid hormone profiling in human breast
adipose tissue using semi-automated purification and highly sensitive
determination of estrogens by GC-APCI-MS/MS, Anal. Bioanal. Chem. 410
(2018) 259–275. doi:10.1007/s00216-017-0717-8.
[24] E. Bichon, I. Guiffard, A. Vénisseau, E. Lesquin, V. Vaccher, A.

Brosseaud, P. Marchand, B. Le Bizec, Simultaneous determination of 16
brominated flame retardants in food and feed of animal origin by fast gas
chromatography coupled to tandem mass spectrometry using atmospheric
pressure chemical ionisation, J. Chromatogr. A. 1459 (2016) 120–128.
doi:10.1016/j.chroma.2016.07.003.
[25] F. Hernández, M. Ibáñez, T. Portolés, M.I. Cervera, J. V Sancho, F.J.

López, Advancing towards universal screening for organic pollutants in

waters., J. Hazard. Mater. 282 (2015) 86–95. doi:10.1016/j.jhazmat.2014.08.

006.
[26] L. Cherta, T. Portolés, J. Beltran, E. Pitarch, J.G.J. Mol, F. Hernández,

Application of gas chromatography-(triple quadrupole) mass spectrometry
with atmospheric pressure chemical ionization for the determination of
multiclass pesticides in fruits and vegetables, J. Chromatogr. A. 1314 (2013)
224–240. doi:10.1016/j.chroma.2013.09.029.
[27] M.I. Aveldano, H. Sprecher, Very long chain (C24 to C36) polyenoic
fatty acids of the n-3 and n-6 series in dipolyunsaturated phosphatidylcholines
from bovine retina, J. Biol. Chem. 262 (1987) 1180–1186.
[28] A. Liu, Y. Lin, R. Terry, K. Nelson, P.S. Bernstein, Role of long-chain

and very-long-chain polyunsaturated fatty acids in macular degenerations and
dystrophies, Clin. Lipidol. 6 (2011) 593–613. doi:10.2217/clp.11.41.
[29] A. Liu, J. Chang, Y. Lin, Z. Shen, P.S. Bernstein, Long-chain and very
long-chain polyunsaturated fatty acids in ocular aging and age-related macular
degeneration., J. Lipid Res. 51 (2010) 3217–29. doi:10.1194/jlr.M007518.
[30] D.R. Tocher, Omega-3 long-chain polyunsaturated fatty acids and

aquaculture in perspective, Aquaculture. 449 (2015) 94–107. doi:10.1016/
j.aquaculture.2015.01.010.
[31] S. Li, Ó. Monroig, T. Wang, Y. Yuan, J. Carlos Navarro, F. Hontoria, K.

Liao, D.R. Tocher, K. Mai, W. Xu, Q. Ai, Functional characterization and
differential nutritional regulation of putative Elovl5 and Elovl4 elongases in
large yellow croaker (Larimichthys crocea), Sci. Rep. 7 (2017). doi:10.1038/
s41598-017-02646-8.

[32] G. Carmona-Antoñanzas, Ó. Monroig, J.R. Dick, A. Davie, D.R. Tocher,

Biosynthesis of very long-chain fatty acids (C > 24) in Atlantic salmon:
Cloning, functional characterisation, and tissue distribution of an Elovl4
elongase, Comp. Biochem. Physiol. - B Biochem. Mol. Biol. 159 (2011) 122–
129. doi:10.1016/j.cbpb.2011.02.007.
[33] A. Jakobsson, R. Westerberg, A. Jacobsson, Fatty acid elongases in

mammals: Their regulation and roles in metabolism, Prog. Lipid Res. 45
(2006) 237–249. doi:10.1016/j.plipres.2006.01.004.
[34] Folch J.; Lees M.; Stanley G. H. S., J. Folch, M. Lees, G.H.S. Stanley, A
simple method for the isolation and purification of total lipids from animal
tissues, J Biol Chem. 226 (1957) 497–509. doi:10.1007/s10858-011-9570-9.
[35] R.E. Olsen, R.J. Henderson, The rapid analysis of neutral and polar
marine lipids using double-development HPTLC and scanning densitometry,
J. Exp. Mar. Bio. Ecol. 129 (1989) 189–197. doi:10.1016/0022-0981(89)
90056-7.
[36] J.C. Navarro, F. Amat, J.R. Sargent, A study of the variations in lipid
levels, lipid class composition and fatty acid composition in the first stages of
Artemia sp., Mar. Biol. 111 (1991) 461–465. doi:10.1007/BF01319419.
[37] W.W. Christie, Lipid analysis, Pergamon Press Oxford; 2nd edition,
1982.
[38] T. Portolés, C. Sales, B. Gómara, J.V. Sancho, J. Beltrán, L. Herrero, M.J.

González, F. Hernández, Novel analytical approach for brominated flame
retardants based on the use of gas chromatography-atmospheric pressure
chemical ionization-tandem mass spectrometry with emphasis in highly

brominated congeners., Anal. Chem. 87 (2015) 9892–9. doi:10.1021/acs.

analchem.5b02378.

Como se ha comentado en la introducción de este capítulo, el principal

inconveniente del análisis de VLC-PUFA reside en la ausencia de patrones de
referencia comercialmente disponibles, pues implica la ausencia de bases de
datos espectrales fiables y, por tanto, supone una dificultad añadida a la hora
de confirmar la identidad de estos compuestos. Por este motivo, la
identificación de los compuestos recae principalmente en la información
estructural obtenida mediante el análisis espectrométrico. Tradicionalmente,
el análisis de VLC-PUFA-MEs se ha llevado a cabo mediante el acoplamiento
GC-MS con fuente EI, obteniendo información estructural poco específica
dada la pérdida total y/o parcial del M+·. Ante esta problemática, el artículo
científico IV aporta una estrategia novedosa para la identificación de VLC-
PUFA-MEs sin necesidad de patrones analíticos mediante la técnica GC-
APCI-QTOF MS.
En primer lugar, dada la novedad del uso de la fuente APCI en esta aplicación,
ha sido necesaria la creación de una base de datos experimental y teórica que
contenga información espectral y/o cromatográfica. La base de datos teórica
se basa en listados de VLC-PUFA reportados en la literatura para los cuales
no hay patrones analíticos disponibles. Esta base de datos teórica incluye la
fórmula molecular de cada compuesto y la masa exacta de M+· y [M+H]+. Para
crear la base de datos experimental, se llevó a cabo el estudio de la ionización
y fragmentación de quince ésteres metílicos de ácidos grasos (FAMEs) –
saturados e insaturados ω3 y ω6 - disponibles comercialmente poniendo
especial énfasis en sus diferencias estructurales. La adquisición en modo MSE
permitió obtener en una misma inyección los espectros completos de LE y HE
con elevada exactitud de masa.
247
Pese a mantener prácticamente intacta la [M+H]+ en la función LE, los LC-

PUFA-MEs estudiados mostraron fragmentos comunes, independientemente
del carácter ω6 o ω3, en la función HE (ver figura 1). Sin embargo, sí que
presentaban diferencias en las abundancias relativas de algunos fragmentos
estructurales.
Figura 1: Espectros de APCI en la función de HE de los ésteres metílicos

de los LC-PUFA 22:4ω3 (arriba) y 22:4ω6 (abajo)
Con el objetivo de diferenciar los ácidos grasos isoméricos ω6 de los ω3

mediante un modelo estadístico, se creó una matriz de datos con la abundancia
de cada fragmento relativa a la del ion m/z 121, que es el pico base del espectro
de HE de los compuestos 22:4ω3 y 22:4ω6 bajo las condiciones de medida.
Para contemplar también la consistencia de las abundancias relativas, la matriz
248
contenía los datos de cinco inyecciones consecutivas de los estándares.

Considerando un análisis estadístico completamente imparcial, en primer
lugar se aplicó un análisis de componentes principales (PCA) sobre la matriz
de datos, mostrando una distinción aproximada entre ω3 y ω6. A continuación,
se realizó un análisis ortogonal de mínimos cuadrados parciales
discriminantes (OPLS-DA), generando un modelo que destacó la abundancia
relativa entre los iones m/z 121 y 163 como la variable diferenciadora de
ambos grupos.
A fin de validar el modelo y evaluar su alcance en la clasificación de VLC-

PUFA se llevó a cabo el análisis de diferentes experimentos de expresión
heteróloga, es decir, se analizaron los productos de elongación in vitro de
diferentes LC-PUFA ω3 y ω6 generados por la elongasa ELOVL4 en
levadura. Como resultado, se observó que el modelo clasificaba correctamente
los productos de elongación.
Por último, su aplicación en el análisis de muestras reales junto con los

criterios de identificación de métodos de screening descritos a lo largo de esta
tesis, permitió identificar sin error diferentes VLC-PUFA en retinas de
salmón, dorada, lenguado y lubina sin necesidad de patrones analíticos. La
identidad de los compuestos se confirmó, en última instancia, a partir de los
experimentos de expresión heteróloga, reafirmando así la fiabilidad del
método.
Pese a demostrar la fiabilidad de la estrategia propuesta, cabe remarcar que el

modelo estadístico presentado se basa en datos espectrométricos generados
bajo unas condiciones instrumentales concretas. La fragmentación de los
compuestos y, por tanto, la abundancia relativa entre los fragmentos, depende
249
de parámetros como el voltaje de cono, la rampa de energía de colisión y las

condiciones de ionización. Por este motivo, los estudios futuros deberán
evaluar la robustez del modelo al cambiar las condiciones instrumentales, bien
para el mismo equipo o transferido a otros.
250
4.4. Referencias
1. Calder PC (2013) N-3 Fatty Acids, Inflammation and Immunity: New

Mechanisms To Explain Old Actions. Proc Nutr Soc 72:326–36. doi:
10.1017/S0029665113001031
2. Ibeas C, Izquierdo MS, Lorenzo A (1994) Effect of different levels of

n-3 highly unsaturated fatty acids on growth and fatty acid composition
of juvenile gilthead seabream (Sparus aurata). Aquaculture. doi:
10.1016/ 0044-8486(94)90424-3
3. Agbaga M-P, Brush RS, Mandal MNA, Henry K, Elliott MH,

Anderson RE (2008) Role of Stargardt-3 macular dystrophy protein
(ELOVL4) in the biosynthesis of very long chain fatty acids. Proc Natl
Acad Sci USA 105:12843–12848. doi: 10.1073/pnas.0802607105
4. Aveldaño MI, Sprecher H (1987) Very long chain (C24 to C36)

polyenoic fatty acids of the n-3 and n-6 series in dipolyunsaturated
phosphatidylcholines from bovine retina. J Biol Chem 262:1180–6
5. Suh M, Wierzbicki AA, Clandinin MT (2002) Dietary n-3 FA

modulate long and very long chain FA content, rhodopsin content, and
rhodopsin phosphorylation in rat rod outer segment after light
exposure. Lipids. doi: 10.1007/s11745-002-0888-0
6. McMahon A, Jackson SN, Woods AS, Kedzierski W (2007) A

Stargardt disease-3 mutation in the mouse ELOVL4 gene causes retinal
deficiency of C32-C36 acyl phosphatidylcholines. FEBS Lett. doi:
10.1016/j.febslet. 2007.10.050
251
7. Poulos A (1995) Very long chain fatty acids in higher animals—A

review. Lipids 30:1–14. doi: 10.1007/BF02537036
8. Robinson BS, Johnson DW, Poulos A (1990) Unique molecular

species of phosphatidylcholine containing very-long-chain (C24-C38)
polyenoic fatty acids in rat brain. Biochem J. doi: 10.1042/bj2650763
9. Aveldaño MI, Robinson BS, Johnson DW, Poulos A (1993) Long and
very long chain polyunsaturated fatty acids of the n-6 series in rat
seminiferous tubules. J Biol Chem
10. Furland NE, Maldonado EN, Aveldaño MI (2003) Very Long Chain
Polyunsaturated Fatty Acids in Murine Testicular Tri-glycerides and
Cholesterol Esters. 1666:
11. McMahon A, Kedzierski W (2010) Polyunsaturated very-long-chain

C28-C36 fatty acids and retinal physiology. Br. J. Ophthalmol.
12. Monroig O, Rotllant J, Cerdá-Reverter JM, Dick JR, Figueras A,

Tocher DR (2010) Expression and role of Elovl4 elongases in
biosynthesis of very long-chain fatty acids during zebrafish Danio
rerio early embryonic development. Biochim Biophys Acta - Mol Cell
Biol Lipids. doi: 10.1016/j.bbalip.2010.06.005
13. Monroig Ó, Webb K, Ibarra-Castro L, Holt GJ, Tocher DR (2011)

Biosynthesis of long-chain polyunsaturated fatty acids in marine fish:
Characterization of an Elovl4-like elongase from cobia Rachycentron
canadum and activation of the pathway during early life stages.
Aquaculture 312:145–153. doi: 10.1016/j.aquaculture.2010.12.024
252
14. Carmona-Antoñanzas G, Monroig Ó, Dick JR, Davie A, Tocher DR

(2011) Biosynthesis of very long-chain fatty acids (C > 24) in Atlantic
salmon: Cloning, functional characterisation, and tissue distribution of
an Elovl4 elongase. Comp Biochem Physiol Part B Biochem Mol Biol
159:122–129. doi: 10.1016/J.CBPB.2011.02.007
15. Monroig Ó, Wang S, Zhang L, You C, Tocher DR, Li Y (2012)

Elongation of long-chain fatty acids in rabbitfish Siganus
canaliculatus: Cloning, functional characterisation and tissue
distribution of Elovl5- and Elovl4-like elongases. Aquaculture. doi:
10.1016/j.aquaculture.2012.04.017
16. Volpato M, Spencer JA, Race AD, Munarini A, Belluzzi A, Cockbain

AJ, Hull MA, Loadman PM (2017) A liquid chromatography–tandem
mass spectrometry method to measure fatty acids in biological
samples. J Chromatogr B 1055–1056:125–134. doi: 10.1016/j.jchromb
.2017.04.030
17. Pettersen J (2003) Determination of trans double bonds in

polyunsaturated fatty acid methyl esters from their electron impact
mass spectra. Eur. J. Lipid Sci. Technol. 105:156–164.
18. Kloos D-P, Gay E, Lingeman H, Bracher F, Müller C, Mayboroda OA,

Deelder AM, Niessen WMA, Giera M (2014) Comprehensive gas
chromatography-electron ionisation mass spectrometric analysis of
fatty acids and sterols using sequential one-pot silylation:
quantification and isotopologue analysis. Rapid Commun Mass
Spectrom 28:1507–14. doi: 10.1002/rcm.6923
253
CAPÍTULO 5
CONCLUSIONES/CONCLUSIONS
Capítulo 5 Conclusiones
CONCLUSIONES
La conclusion general derivada de los trabajos realizados en la presente Tesis

Doctoral es que el potencial de las técnicas instrumentales disponibles
actualmente, junto con la fiabilidad de los métodos analíticos desarrollados,
hacen de la química analítica una disciplina fundamental en las ciencias
ambientales, demostrando ser una herramienta muy útil para abordar y
resolver problemas de seguridad alimentaria y medioambientales, entre otros.
De los trabajos presentados en esta Tesis Doctoral se pueden extraer las

siguientes conclusiones específicas:
1) El screening mediante LC-QTOF MS y GC-APCI-QTOF MS permite

abordar, de forma rápida, sensible y fiable, el análisis de un gran rango
de compuestos con diferentes propiedades físico-químicas. El uso
conjunto de ambas técnicas es completamente necesario para abordar
analíticamente nuevas problemáticas ambientales.
2) El potencial del acoplamiento GC-APCI-MS/MS con analizador QqQ

permite reducir el efecto matriz en muestras complejas mediante
dilución simple del extracto, posibilitando la cuantificación de las
mismas mediante calibrado en solvente relativo a SIL-IS.
3) La estrategia integral para el análisis de pesticidas, PAHs y PCBs en

muestras de acuicultura se presenta como una herramienta fiable para
la evaluación de riesgos durante el ciclo de producción acuícola.
4) El acoplamiento LC-MS/MS es una técnica poderosa, rápida y fiable

para la determinación de pesticidas y sus metabolitos en matrices
complejas debido a su elevada sensibilidad y selectividad.
257
Capítulo 5 Conclusiones
5) El análisis de muestras de pez cebra y salmón sometidas a la

exposición dietaria de pirimiphos-methyl permitirá crear un modelo
cinético de acumulación y eliminación que contribuirá a establecer
MRLs seguros y objetivos en pescado.
6) El uso de la fuente de ionización APCI supone un avance en la

investigación de VLC-PUFA, incrementando la fiabilidad en el
proceso de identificación de los compuestos.
7) Los experimentos de expresión heteróloga permiten generar

compuestos de referencia para la validación de modelos estadísticos
de identificación de VLC-PUFA.
258
Chapter 5 Conclusions
CONCLUSIONS
The general conclusion of this Doctoral Thesis is that the potential of the
currently available instrumental techniques, together with the reliability of the
analytical methods developed, make analytical chemistry an essential
discipline in environmental sciences, proving to be a very useful tool to
address and solve food safety and environmental problems, among others.
As a result of the research performed in this work, several specific conclusions

can be extracted:
1) The screening by LC-QTOF MS and GC-APCI-QTOF MS allows the fast,

sensitive and reliable analysis of a wide range of compounds with different
physicochemical properties. The complementary use of both techniques is
required to face new environmental problems.
2) The potential of GC-APCI-MS/MS with QqQ analyzer allows reducing the

matrix effect in complex samples by simple dilution of the extract, making it
possible to quantify them by calibration with standards in solvent relative to
SIL-IS.
3) The comprehensive strategy for the analysis of pesticides, PAHs and PCBs
in aquaculture samples is a reliable tool for risk assessment during the
aquaculture production cycle.
4) The LC-MS/MS coupling is a powerful, fast and reliable technique for the
determination of pesticides and their metabolites in complex matrices due to
its high sensitivity and selectivity.
5) The analysis of zebrafish and salmon samples subjected to the dietary
259
Chapter 5 Conclusions
exposure of pirimiphos-methyl will allow the creation of a kinetic model of

accumulation and elimination that will contribute to establish reliable and
objective MRLs in fish.
6) The use of the APCI ionization source represents an advance in the research
of VLC-PUFA, increasing the reliability in the process of identification of the
compounds.
7) The heterologous expression experiments allow generating reference

compounds for the validation of statistical models for the identification of
VLC-PUFA.
260
Sugerencias para futuros trabajos
SUGERENCIAS PARA FUTUROS TRABAJOS
El trabajo realizado en la presente Tesis Doctoral se ha desarrollado en base a

la instrumentación disponible en el Institut Universitari de Plaguicides i
Aigües durante un periodo de tres años. Sin embargo, la metodología
presentada podría enriquecerse, reevaluarse o mejorarse gracias, en parte, a la
aparición y adquisición de nuevos sistemas instrumentales.
Por un lado, la aparición de espectrómetros de masas con movilidad iónica,

como el VION IMS QTOF, supone un avance en el screening de
contaminantes orgánicos. Debido a la naturaleza ortogonal de la separación
por movilidad iónica, este sistema instrumental es capaz de resolver, en
muchos casos, compuestos coeluyentes y permite la detección de
componentes que pueden haber estado ocultos por la matriz o mostrar especies
isoméricas que no pueden separarse por cromatografía. Además, la separación
por movilidad iónica introduce una nueva variable identificativa: la sección
transversal de colisión (CCS), lo que podría traducirse en una mejora en la
capacidad de identificación. Por esta razón, sería interesante explorar en
futuros trabajos el potencial de los acoplamientos LC-IMS QTOF MS y GC-
APCI-IMS QTOF MS para el screening de contaminantes orgánicos en
muestras de acuicultura y, así, comparar la detectabilidad y la capacidad de
identificación con la reportada en el artículo científico I. Además, este
sistema instrumental puede ser de gran utilidad para estudios de metabolismo.
Tal es así que Bijlsma et al (2019) [1] han explorado recientemente el
potencial de esta técnica para investigar la presencia de otros metabolitos del
pirimiphos-methyl no reportados en los extractos de músculo, hígado, riñón,
bilis y grasa de salmón utilizados en el artículo científico III.
261
Sugerencias para futuros trabajos
Por otro lado, algunos de los métodos presentados requieren de una

ampliación del estudio, como puede ser el caso del artículo científico IV que,
pese a haber demostrado la fiabilidad de la estrategia propuesta para la
identificación de VLC-PUFA, sería recomendable evaluar la robustez de la
misma al cambiar las condiciones instrumentales. Además, puede ser de gran
interés buscar aproximaciones para la cuantificación de estos compuestos ya
que, a la dificultad analítica en la identificación, se suma la dificultad en la
cuantificación.
1. Bijlsma L, Berntssen MHG, Merel S (2019) A Refined Nontarget

Workflow for the Investigation of Metabolites through the Prioritization by in
Silico Prediction Tools. Anal Chem. doi: 10.1021/acs.analchem.9b01218
262
Suggestions for future works
SUGGESTIONS FOR FUTURE WORKS
The work performed in this Doctoral Thesis has been developed according to
the instruments available at the Research Institute for Pesticides and Water for
a period of three years. However, the methodology presented could be
enriched, reevaluated or improved thanks, in part, to the availability of new
instrumental systems.
On the one hand, the emergence of ion mobility separation mass

spectrometers, such as the VION IMS QTOF, represents an advance in the
screening of organic pollutants. Due to the orthogonal nature of ion mobility
separation, this instrumental system can resolve, in many cases, coeluting
compounds and allows the detection of components that may have been
hidden by the matrix or show isomeric species that cannot be separated by
chromatography. In addition, ion mobility separation introduces a new
identification factor: the collisional cross section (CCS), which could improve
the identification capacity. For this reason, it would be interesting to explore
in future works the potential of LC-IMS QTOF MS and GC-APCI-IMS QTOF
MS for the screening of organic pollutants in aquaculture samples and, thus,
compare the detectability and identification capacity with the one reported in
the scientific article I. In addition, this instrumental system can be very useful
for metabolism studies. In fact, Bijlsma et al (2019) [1] recently explored the
potential of this technique to investigate the presence of other unreported
pyrimiphos-methyl metabolites in muscle, liver, kidney, bile and salmon fat
extracts used in the scientific article III.
On the other hand, some of the methods require an extension of the study, such
as that presented in the scientific article IV in which, despite having
263
Suggestions for future works
demonstrated the reliability of the proposed strategy for the identification of

VLC-PUFA, the evaluation of the robustness when changing the instrumental
conditions would be recommended. In addition, it may be of great interest to
look for quantitative approches for the determination of these compounds
since, to the analytical difficulty in the identification, the difficulty in the
quantification is added.
1. Bijlsma L, Berntssen MHG, Merel S (2019) A Refined Nontarget

Workflow for the Investigation of Metabolites through the Prioritization by in
Silico Prediction Tools. Anal Chem. doi: 10.1021/acs.analchem.9b01218
264
Artículos relacionados/Related articles
ARTÍCULOS RELACIONADOS/RELATED ARTICLES
Sanden M, Olsvik PA, Søfteland L, Rasinger JD, Rosenlund G, Garlito B,

Ibáñez M, Berntssen MHG (2018) Dietary pesticide chlorpyrifos-methyl
affects arachidonic acid metabolism including phospholipid remodeling in
Atlantic salmon (Salmo salar L.). Aquaculture 484:1–12. doi:
10.1016/j.aquaculture.2017.10.033
Nácher-Mestre J, Ballester-Lozano GF, Garlito B, Portolés T, Calduch-Giner

J, Serrano R, Hernández F, Berntssen MHG, Pérez-Sánchez J (2018)
Comprehensive overview of feed-to-fillet transfer of new and traditional
contaminants in Atlantic salmon and gilthead sea bream fed plant-based diets.
Aquac Nutr 24:1782–1795. doi: 10.1111/anu.12817
265

2019 Tesis Garlito Molina Borja

Cargado por

Copyright:

Formatos disponibles

2019 Tesis Garlito Molina Borja

Cargado por

Información del documento

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

2019 Tesis Garlito Molina Borja

Cargado por

Copyright:

Formatos disponibles

Programa de Doctorado en Ciencias

Escuela de Doctorado de la Universitat Jaume I

ANÁLISIS DE COMPUESTOS ENDÓGENOS Y EXÓGENOS EN

Memoria presentada por Borja Garlito Molina para optar al grado de

Borja Garlito Molina Roque Serrano Gallego Tania Portolés Nicolau

Castelló de la Plana, octubre de 2019

Agencias financiadoras del doctorando

 Universitat Jaume I (I-4663)

Agencias financiadoras del proyecto de investigación o de los recursos

 EU Seventh Framework Programme

I) T. Portolés, M. Ibáñez, B. Garlito, J. Nácher-Mestre, V. Karalazos, J. Silva,

II) T. Portolés, B. Garlito, Jaime Nácher-Mestre, M.H.G. Berntssen, J. Pérez-

III) B. Garlito, M. Ibáñez, T. Portolés, R. Serrano, H. Amlund, A-K.

IV) B. Garlito, T. Portolés, W.M.A. Niessen, J.C. Navarro, F. Hontoria, Ó.

Aquesta Tesi Doctoral ha estat realitzada en l’Institut Universitari de

Ante el estancamiento de la producción de la pesca de captura desde finales

En la presente Tesis Doctoral se ha estudiado el potencial analítico y la

El trabajo presentado se ha estructurado en dos bloques: un primer bloque

El primer bloque está formado por tres trabajos consecutivos y relacionados

acuicultura. En el primero de los trabajos presentados se estudia el potencial

En el segundo trabajo, surgido a partir de los resultados obtenidos en el trabajo

Para finalizar, en el tercer trabajo de este bloque se desarrolla un método

desde una aplicación analítica en el ámbito de la toxicología, a la ausencia de

El segundo bloque aborda el estudio de ácidos grasos poliinsaturados de

In this Doctoral Thesis the analytical potential and applicability of different

APCI-QTOF MS) are studied for the development of a comprehensive

In the second work, a method based on gas chromatography coupled to tandem

CAPÍTULO 1. INTRODUCCIÓN GENERAL ............................................. 9

1.1. Retos actuales de la acuicultura .......................................................... 13

1.2. Seguridad alimentaria y salud animal ................................................. 16

1.3. Técnicas analíticas para el control alimentario ................................... 18

1.3.1. Técnicas cromatográficas............................................................. 19

1.3.2. Espectrometría de masas .............................................................. 23

1.4. Estrategias analíticas para el control alimentario ............................... 33

1.5. Tratamiento de muestra en matrices de acuicultura. .......................... 36

1.6. Estrategias de corrección del efecto matriz. ....................................... 38

1.7. Referencias ......................................................................................... 43

CAPÍTULO 2. DETERMINACIÓN DE COMPUESTOS EXÓGENOS

2.1. Introducción ........................................................................................ 55

2.2. Artículo científico I ............................................................................ 61

Chemosphere, 179, 242-253, 2017

2.3. Artículo científico II ......................................................................... 105

Multi-class determination of undesirables in aquaculture samples by gas

Talanta, 172, 109–119, 2017

2.4. Discusión de los resultados obtenidos .............................................. 153

2.5. Referencias ....................................................................................... 163

CAPÍTULO 3. ESTUDIO DE LA TRANSFERENCIA DE

3.1. Introducción...................................................................................... 171

3.2. Artículo científico III........................................................................ 173

LC-MS/MS method for the determination of organophosphorus pesticides

Analytical and Bioanalytical Chemistry, 2017

3.3. Discusión de los resultados obtenidos .............................................. 203

CAPÍTULO 4. DETERMINACIÓN DE COMPUESTOS ENDÓGENOS

4.1. Introducción ...................................................................................... 213

4.2. Artículo científico IV........................................................................ 217

Analytica Chimica Acta, 1051, 103-109, 2019

4.3. Discusión de los resultados obtenidos .............................................. 247

4.4. Referencias ....................................................................................... 251

CAPÍTULO 5. CONCLUSIONES/CONCLUSIONS ............................... 255

SUGERENCIAS PARA FUTUROS TRABAJOS ..................................... 261