Pathogenesis and transmission of lymphocystis disease virus (LCDV) in gilthead seabream (Sparus aurata L.)

AUTOR: Estefanía Jiménez Valverde http://orcid.org/0000-0003-1406-1181 EDITA: Publicaciones y Divulgación Científica. Universidad de Málaga Esta obra está bajo una licencia de Creative Commons Reconocimiento-NoComercialSinObraDerivada 4.0 Internacional: http://creativecommons.org/licenses/by-nc-nd/4.0/legalcode Cualquier parte de esta obra se puede reproducir sin autorización pero con el reconocimiento y atribución de los autores. No se puede hacer uso comercial de la obra y no se puede alterar, transformar o hacer obras derivadas. Esta Tesis Doctoral está depositada en el Repositorio Institucional de la Universidad de Málaga (RIUMA): riuma.uma.es FACULTAD DE CIENCIAS PROGRAMA DE DOCTORADO EN BIOLOGÍA CELULAR Y MOLECULAR “Pathogenesis and transmission of lymphocystis disease virus (LCDV) in gilthead seabream (Sparus aurata L.)” TESIS DOCTORAL Estefanía Jiménez Valverde -2016- FACULTAD DE CIENCIAS PROGRAMA DE DOCTORADO EN BIOLOGÍA CELULAR Y MOLECULAR “Pathogenesis and transmission of lymphocystis disease virus (LCDV) in gilthead seabream (Sparus aurata L.)” Memoria presentada por Dª. Estefanía Jiménez Valverde para optar al grado de Doctor en Biología con mención internacional Fdo. La doctoranda FACULTAD DE CIENCIAS PROGRAMA DE DOCTORADO EN BIOLOGÍA CELULAR Y MOLECULAR D. JUAN JOSÉ BORREGO GARCÍA Catedrático del Departamento de Microbiología de la Universidad de Málaga, como TUTOR dentro del programa de doctorado “BIOLOGIA CELULAR Y MOLECULAR” y DIRECTOR del trabajo de tesis realizado por la doctoranda. Dña. Mª DOLORES CASTRO LÓPEZ, Profesora Titular del Departamento de Microbiología de la Universidad de Málaga y DIRECTORA del trabajo de tesis realizado por la doctoranda. INFORMAN: Que la doctoranda ESTEFANÍA JIMÉNEZ VALVERDE ha realizado de forma satisfactoria y bajo nuestra supervisión las actividades de formación dentro del programa de doctorado arriba mencionado así como el trabajo investigador que se presenta y que lleva por título “Pathogenesis and transmission of lymphocystis disease virus (LCDV) in gilthead seabream (Sparus aurata L.)”. Este trabajo constituye su proyecto de Tesis para aspirar al Título de Doctora en Biología con mención internacional. Y para que así conste, y tenga los efectos que correspondan, en cumplimiento de la legislación vigente, extendemos el presente informe en Málaga, a 14 de septiembre de 2016. Fdo. El Tutor: Juan José Borrego García Fdo. Los directores: Juan José Borrego García Mª Dolores Castro López Los estudios realizados en la presente Tesis Doctoral han sido financiados por los siguientes proyectos y becas: Proyecto Plan Nacional “Patogénesis, transmisión y control de infecciones por el virus de la enfermedad de linfocistis en el cultivo de la dorada” (Ref. AGL 2010-17880). Proyecto de Excelencia de la Junta de Andalucía “Infecciones por linfocistivirus en doradas cultivadas: patogénesis e implicación del sistema inmune en el curso de la infección. Desarrollo de medidas de profilaxis y control” (Ref. P12-RNM-2261). Beca F.P.I. del Ministerio de Ciencia e Innovación de España (Ref. BES2011-043607). i INDICE/INDEX RESUMEN ……………………………………………………………….............................. 1 INTRODUCCIÓN ............................................................................................... 3 CAPÍTULO 1: Métodos de diagnóstico para el virus de la enfermedad de linfocistis ...................................................................................................... 9 1.1. Detección y cuantificación del virus de la enfermedad de linfocistis mediante PCR a tiempo real ………………………………. 10 1.2. Ensayo LAMP (loop-mediated isothermal amplification) para la detección rápida del virus de la enfermedad de linfocistis …….. 11 1.3. Ensayo ICC RT-PCR (integrated cell culture RT-PCR) para la detección y cuantificación de LCDV infectivos …………………… 13 CAPÍTULO 2: Patogénesis y transmisión del virus de la enfermedad de linfocistis …………………………………………………………………………… 14 2.1. Determinación de los órganos diana para la multiplicación del virus de la enfermedad de linfocistis en dorada …………………………… 15 2.2. Transmisión del virus de la enfermedad de linfocistis a larvas y alevines de dorada ………………………………………………………………. 16 2.3. Artemia sp. como hospedador del virus de la enfermedad de linfocistis …………………………………………………………………………….. 17 INTRODUCTION ……………………………………………............................................ 19 1. Current state of aquaculture of gilthead seabream (Sparus aurata L.) …….. 21 2. Major viral diseases affecting farmed fish ......................................................... 22 3. Lymphocystis disease virus …………………………………………........................ 26 3.1. Taxonomy and genetic diversity …………………….......................... 26 3.2. Virion structure …………………………………………........................... 28 3.3. Chemical composition ………………………………............................ 30 3.4. Genome organization ………………………………............................. 31 3.5. Viral multiplication …………………………………................................ 33 4. Lymphocystis disease …………………………………........................................... 35 4.1. Disease features .................................................................................. 36 4.2. Viral pathogenesis and host immunity ……………………………….. 38 4.3. Diagnostic methods …………………………………………………….... 40 4.3.1. Virus isolation in cell cultures ……………................................ 41 4.3.2. Serological techniques ……………………….......................... 42 4.3.3. PCR-based techniques ……………………….......................... 43 4.3.4. Loop-mediated isothermal amplification (LAMP) …........... 44 4.4. Lymphocystis disease virus transmission ……………………………… 45 4.5. Disease control and prevention ………………………....................... 47 5. Artemia as live food in aquaculture ……………………………………………….. 48 5.1. Morphology and life cycle ………………………................................ 49 OBJETIVOS ……………………………….........……………………................................ 53 ii AIMS ……………………………….........……………………........................................... 57 CHAPTER 1: DIAGNOSTIC METHODS FOR LYMPHOCYSTIS DISEASE VIRUS ... 61 1. Real-time PCR assay for lymphocystis disease virus genotype VII detection and quantification ………………………......................................... 63 1.1. INTRODUCTION ……………………………….........…………………....... 63 1.2. MATERIALS & METHODS ………………………………............................ 64 1.2.1. Sample collection and DNA extraction ............................... 64 1.2.2. PCR-hybridization .................................................................... 65 1.2.3. Cloning a fragment of MCP gene ........................................ 66 1.2.4. Real-time PCR assay ............................................................... 66 1.2.5. Standard curve for LCDV quantification .............................. 67 1.2.6. Assay repeatability and reproducibility ................................ 68 1.3. RESULTS ……………………………….........……………………………….. 68 1.3.1. Evaluation of the real-time PCR assay .................................. 68 1.3.2. Surveillance study of LCDV in gilthead seabream farms … 70 2. Rapid and sensitive detection of lymphocystis disease virus genotype VII by loop-mediated isothermal amplification ……….......... 74 2.1. INTRODUCTION ……………………………….........……......................... 74 2.2. MATERIALS & METHODS …………………….……….............................. 75 2.2.1. Virus isolates and fish samples ............................................... 75 2.2.2. LAMP primer design ................................................................ 75 2.2.3. Construction of recombinant plasmid ................................. 77 2.2.4. Optimization of LAMP reaction temperature ...................... 77 2.2.5. Specificity and sensitivity of the LAMP assay ....................... 77 2.2.6. Application of the LAMP assay for LCDV detection in fish samples ..................................................................................... 78 2.3. RESULTS …………………………….........…............................................. 79 2.3.1. Optimization of LAMP reaction temperature ...................... 79 2.3.2. Specificity and sensitivity of the LCDV LAMP assay ............. 79 2.3.3. LCDV diagnosis by LAMP ........................................................ 82 3. Evaluation of an integrated cell culture RT-PCR assay to detect and quantify infectious lymphocystis disease virus ……………........................ 85 3.1. INTRODUCTION ……………………………….........……………………... 85 3.2. MATERIALS & METHODS ………….………………….............................. 86 3.2.1. Virus and cell lines ................................................................... 86 3.2.2. ICC-RT-PCR-hybridization assay ............................................. 87 3.2.3. Sensitivity of the ICC-RT-PCR assay ........................................ 88 3.2.4. ICC-RT-PCR assay for viral quantification ............................. 88 iii INDICE/INDEX 3.3. RESULTS ……………………………….........….......................................... 89 3.3.1. Specificity and sensitivity of the ICC-RT-PCR assay ……...... 89 3.3.2. Viral quantification by ICC-RT-PCR ....................................... 91 CHAPTER 2: PATHOGENESIS AND TRANSMISSION OF LYMPHOCYSTIS DISEASE VIRUS ........................................................................................................... 93 1. Target organs for lymphocystis disease virus replication in gilthead seabream .………………………………................................................................ 95 1.1. INTRODUCTION ……………………………….........…............................. 95 1.2. MATERIALS & METHODS ………………….………….............................. 96 1.2.1. Fish samples .............................................................................. 96 1.2.2. DNA and RNA extraction and cDNA synthesis .................... 97 1.2.3. LCDV DNA quantification and gene expression ................. 97 1.2.4. RNA in situ hybridization and histopathology ....................... 98 1.3. RESULTS ……………………………….........….......................................... 99 1.3.1. Viral load and gene expression ............................................. 99 1.3.2. RNA in situ hybridization .......................................................... 101 1.3.3. Histopathological study .......................................................... 103 2. Transmission of lymphocystis disease virus to gilthead seabream larvae and fingerlings .………………………………......................................... 108 2.1. INTRODUCTION ………………………………............................…......... 108 2.2. MATERIALS & METHODS ……………………………............................... 109 2.2.1. Transmission studies with gilthead seabream larvae ……... 109 2.2.2. Brine shrimp LCDV inoculation ............................................... 110 2.2.3. Gilthead seabream oral challenge ...................................... 111 2.2.4. LCDV detection by PCR-hybridization .................................. 111 2.2.5. Virological analyses ................................................................ 111 2.2.6. LCDV DNA quantification and gene expression ................. 112 2.2.7. Whole-mount in situ hybridization .......................................... 112 2.2.8. Immunohistochemistry ............................................................ 113 2.3. RESULTS ……………………………….........………………….................... 113 2.3.1. LCDV transmission to gilthead seabream larvae ................ 113 2.3.2. LCDV transmission to gilthead seabream fingerlings ……... 117 3. Artemia sp., a susceptible host for lymphocystis disease virus ............ 119 3.1. INTRODUCTION ……………………………….........…............................. 119 3.2. MATERIALS & METHODS …………………………..….............................. 120 3.2.1. Experimental infections ........................................................... 120 3.2.2. LCDV DNA quantification and gene expression ................. 120 3.2.3. Virological analyses ................................................................ 121 iv 3.3. RESULTS ……………………………….........….......................................... 121 DISCUSSION ……………………………….........…....................................................... 125 1. DIAGNOSTIC METHODS FOR LYMPHOCYSTIS DISEASE VIRUS ...................... 127 2. PATHOGENESIS OF LCDV IN GILTHEAD SEABREAM ...................................... 132 3. TRANSMISSION ROUTES FOR LCDV IN GILTHEAD SEABREAM ....................... 136 4. LYMPHOCYSTIS DISEASE VIRUS INFECTION IN BRINE SHRIMP ...................... 139 CONCLUSIONES . ……………………………….........….............................................. 141 CONCLUSIONS ……………………………….........….................................................. 145 REFERENCES ……………………………….........…....................................................... 149 RESUMEN ! ! 3 RESUMEN INTRODUCCIÓN La fuerte demanda a nivel mundial de los productos derivados de la acuicultura ha abocado a que dicho sector haya experimentado un crecimiento exponencial durante las últimas décadas, constituyendo una alternativa suplementaria al sector de la extracción pesquera. En la actualidad es una de las actividades productivas con mayor interés económico, así como un instrumento adecuado para asegurar la sostenibilidad de los recursos naturales. La mayoría de las especies marinas cultivadas presentan alto valor comercial, a veces porque el volumen de las poblaciones salvajes es bajo o se encuentra en descenso. Este es el caso de la dorada (Sparus aurata), tradicionalmente cultivada en el norte de Italia y en el sur de España, que ha acabado convirtiéndose en una de las principales especies de peces cultivados, ocupando el tercer puesto en la Unión Europea y siendo el principal cultivo de peces en la acuicultura española. Uno de los problemas más importantes a los que se enfrenta la acuicultura, y que puede limitar de forma significativa el cultivo de las especies piscícolas, es la aparición de patologías de etiología microbiana que ven favorecidas su propagación por las condiciones derivadas del cultivo intensivo. Dentro de las enfermedades infecciosas, las de origen vírico tienen una especial relevancia debido a diversos factores, como las altas mortalidades que provocan y la capacidad de inducir infecciones persistentes. Las infecciones víricas son, de hecho, un factor limitante para la expansión de la acuicultura debido a las pérdidas directas en la producción de peces, los costes derivados de la reducción de la productividad y la gestión de la enfermedad, sumados a la pérdida de mercados de exportación en relación con las restricciones comerciales. Además, hay que añadir a esta problemática las limitaciones impuestas por los métodos de diagnóstico y la escasez de tratamientos antivirales realmente efectivos. Por otra parte, existen numerosos virus capaces de desarrollar infecciones asintomáticas, con el consiguiente riesgo de liberación continua de partículas víricas al medio que puede pasar desapercibida. Esto podría implicar una rápida transmisión de la enfermedad en ! 4 el sistema de producción, pudiendo llegar a afectar a la totalidad de las fases del cultivo. Debido al riesgo de propagación, muchas de estas enfermedades víricas son de declaración obligatoria, apareciendo en las listas de la Organización Mundial de Sanidad Animal (OIE). En los últimos años, el número de enfermedades de origen vírico descritas en las instalaciones de acuicultura se ha incrementado considerablemente, destacando aquellas que afectan a peces marinos cultivados. Las principales virosis que afectan a la piscicultura son las provocadas por los virus pertenecientes a las familias Birnaviridae, Iridoviridae, Nodaviridae, Reoviridae y Rhabdoviridae. Con respecto a los miembros de la familia Iridoviridae, estos pueden afectar a una gran diversidad de especies animales (vertebrados poiquilotermos y artrópodos) y tienen una amplia distribución geográfica. En el caso de peces cultivados, el virus de la enfermedad de linfocistis (LCDV), incluido en el género Lymphocystivirus, es el representante de la familia con mayor incidencia, afectando tanto a peces marinos como dulceacuícolas, siendo las familias Centrarchidae, Percidae, Scienidae y Pleuronectidae las que incluyen un mayor número de hospedadores susceptibles. El Comité Internacional de Taxonomía Vírica (ICTV), basándose en diferentes características como el tamaño del virión, rango de hospedador, histopatología, perfiles proteicos y análisis de la secuencia del DNA, reconoce actualmente una única especie dentro del género Lymphocystivirus, Lymphocystis disease virus 1 (LCDV-1), que corresponde a virus aislados de platija europea (Platichthys flessus) y de solla (Pleuronectes platessa). Además, se incluyen otros tres virus candidatos en el género: Lymphocystis disease virus 2 (LCDV-2), que corresponde a los aislados de lenguadina (Limanda limanda); Lymphocystis disease virus-China (LCDV-C), aislado de platija japonesa (Paralichthys olivaceus), y Lymphocystis disease rockfish virus (LCDV-RF), aislado a partir de pez roca coreano (Sebastes schlegeli). El genoma del LCDV-C se ha secuenciado recientemente y muestra considerables diferencias con la especie LCDV-1 en cuanto a tamaño, organización genómica e identidad de los productos génicos codificados, además del rango de hospedador. Por ello se ha propuesto que constituya una nueva especie del género Lymphocystivirus. 5 RESUMEN Una de las características distintivas de la familia Iridoviridae es la presencia de la proteína principal de la cápside (MCP), principal componente estructural de las partículas víricas, que representa hasta el 45% de los polipéptidos víricos totales. Los estudios filogenéticos basados en la secuencia del gen que codifica para la MCP han demostrado la existencia de variabilidad genética dentro del género Lymphocystivirus, estableciéndose 9 genotipos que difieren en múltiples propiedades biológicas, incluyendo rango de hospedador, características antigénicas, etc. La especie LCDV-1 constituiría el genotipo I; el genotipo II corresponde a los aislados de platija japonesa (LCDV-C); el genotipo III incluye los aislados del pez roca coreano (LCDV-RF); el genotipo IV, los aislados de cobia (Rachycentron canadum) y lubina japonesa (Lateolabrax japonicus) (LCDV-RC y LCDV-SB, respectivamente); el genotipo V incluye los aislados del pez tetra fantasía (Parambassis baculis) (LCDV-CB); el genotipo VI para los aislados de gourami (Trichopodus leerii y T. Trichopterus) (LCDV-TL); el genotipo VII incluye a los aislados de dorada y de lenguado senegalés (Solea senegalensis) (LCDV-SA y LCDV-SSE, respectivamente); el genotipo VIII corresponde a los aislados de perca americana (Micropterus salmoides) (cepa Leetown HNF); y el genotipo IX, constituido por los aislados de perca amarilla (Perca flavescens). El LCDV es un virus icosaédrico de gran tamaño, que presenta partículas isométricas de 150 a 200 nm, con una cápside bilaminar y un núcleo central o “core” de apariencia filamentosa, separado de la cápside por un material amorfo de 10 a 20 nm de espesor. La estructura del virión está formada por una membrana interna lipoproteica, con un alto contenido en fosfolípidos, y por una cápside proteica constituida fundamentalmente por la MCP. Además, se ha observado la presencia de filamentos externos en la superficie del virión de aproximadamente 2,5 nm de longitud. El genoma del LCDV está constituido por una molécula lineal de DNA de doble cadena altamente metilado que presenta redundancia terminal, permutación circular, y un contenido G-C de aproximadamente el 30%. En cuanto al mecanismo de replicación del LCDV aún no ha sido completamente dilucidado, y sólo se ha podido establecer que la fase final de la morfogénesis se lleva a cabo en el citoplasma celular. No obstante, se ha propuesto como modelo de mecanismo de replicación el del ! 12 hibridación en blot, que los hacen laboriosos y con tiempos de análisis prolongados. Estas dificultades metodológicas son subsanables mediante el uso de la qPCR, pero incluso en este caso podemos encontrar limitaciones debido al coste en reactivos o el equipamiento necesario para realizar dicha técnica. El método LAMP desarrollado en este capítulo es una técnica sensible, específica, rápida y simple, que permite el diagnóstico del LCDV (genotipo VII) tanto en el laboratorio como en ensayos de campo, lo que supone una gran versatilidad, ya que el diagnóstico se puede realizar in situ. La sensibilidad del método se estimó en 10 copias de DNA viral, muy similar a la obtenida por qPCR y significativamente superior a la obtenida mediante PCR convencional. Además, el análisis de la temperatura de disociación de los productos obtenidos permite confirmar la amplificación específica del LCDV. La duración del ensayo, así como la monitorización de la amplificación, puede ser controlada a tiempo real de forma precisa. Así, este ensayo permitió la detección del LCDV en menos de 60 minutos en muestras de dorada y lenguado portadores del virus que cursaban la infección de forma subclínica, en los que la carga viral es extremadamente baja. Por otra parte, la técnica ha sido diseñada utilizando una plataforma portátil (Genie® II), que permite realizar el diagnóstico en condiciones de campo, lo cual es especialmente interesante cuando ocurren brotes de infección en piscifactorías y el tiempo para determinar la presencia del virus, tanto cualitativamente como cuantitativamente, es limitado. La correlación observada entre la cantidad de DNA y los valores de tiempo de positividad (Tp) obtenidos indican que el protocolo LAMP diseñado puede aplicarse para la cuantificación del LCDV sin necesidad de otro equipamiento. Uno de los puntos críticos en el diagnóstico molecular en muestras de campo es el requerimiento de muestras con DNA de alta calidad. Debido a que el método LAMP es menos sensible a sustancias inhibidoras o interferentes que la PCR, y a que su eficiencia no se ve afectada por la presencia de DNA no purificado, el ensayo LAMP diseñado ha podido utilizarse junto a un protocolo de extracción rápido de DNA que no requiere de reactivos tan costosos. Los resultados fueron similares a los obtenidos utilizando DNA purificado mediante un sistema de extracción comercial. Este método de extracción en crudo permite 13 RESUMEN ahorrar tiempo y costes, y es potencialmente adecuado para operadores menos experimentados. 1.3. Ensayo ICC RT-PCR (integrated cell culture RT-PCR) para la detección y cuantificación de LCDV infectivos. El método estándar para el diagnóstico viral en peces está basado en el aislamiento del virus en cultivos celulares, seguido de la confirmación mediante técnicas moleculares o serológicas. Este procedimiento demuestra la infectividad del virus detectado, además de poder utilizarse para determinar el título vírico infectivo. Sin embargo, el LCDV es difícil de propagar en cultivos celulares y, en ocasiones, no produce efectos citopáticos (CPE) consistentes, especialmente en el caso de muestras obtenidas de peces infectados subclínicamente. Además, en dichas muestras, la aparición de CPE requiere al menos 14 días o incluso de un segundo pase ciego. Por ello, en el presente capítulo, se desarrolló un ensayo en cultivos celulares seguido de la detección de transcritos virales mediante RT-PCR (ICCRT-PCR), que se combinó con la hibridación en blot de los productos de RT-PCR, para la detección de partículas infectivas del LCDV. El método está basado en la detección de RNA mensajero vírico mediante RT-PCR a los cinco días postinoculación en cultivos celulares. En el caso de virus DNA como el LCDV, la detección de transcritos en cultivos celulares es indicativo de la presencia de virus infectivos, permitiendo diferenciar estos de virus inactivados también presentes en la muestra. Además, la hibridación en blot permite aumentar considerablemente la sensibilidad en la detección de estos transcritos virales. Por último, este ensayo puede aplicarse para la determinación del título infectivo mediante el método del número más probable, expresándose como número más probable de unidades infectivas por unidad de volumen (MPNIU/ml). El ensayo de ICC RT-PCR diseñado, que puede llevarse a cabo en sólo 7 días, mostró una alta sensibilidad, al menos 100 veces superior a otros métodos de diagnóstico viral basados en el desarrollo de CPE. Además, la ICC RT-PCR permitió la determinación del título infectivo en muestras con muy baja carga ! 14 vírica, incluyendo aquellas de animales portadores asintomáticos, en los que no fue posible obtener CPE después de 14 días de incubación en cultivos celulares. En estos animales asintomáticos, los títulos infectivos obtenidos mediante ICC RTPCR fueron entre 2,2-2,7 logaritmos más bajos que la carga viral estimada mediante qPCR. Por tanto, aunque la qPCR es una técnica adecuada para el diagnóstico del LCDV, la cantidad de virus infectivos puede verse en ocasiones sobreestimada, al menos en peces infectados subclínicamente, lo que estaría relacionado con la detección de virus defectivos. El ensayo de ICC RT-PCR diseñado permitió la detección y cuantificación de partículas infectivas del LCDV de forma rápida, específica y sensible, por lo que puede ser una herramienta de gran utilidad para el estudio de aspectos importantes de la infección por LCDV como la transmisión o la epizootiología. CAPÍTULO 2: Patogénesis y transmisión del virus de la enfermedad de linfocistis. Este capítulo aborda el estudio del tropismo del LCDV en diferentes órganos y tejidos de ejemplares juveniles de dorada, analizándose tanto animales enfermos como animales con infecciones subclínicas o recuperados de la enfermedad. Para ello se han utilizado técnicas de cuantificación vírica mediante qPCR y de detección de transcritos virales mediante RT-qPCR o hibridación in situ (ISH). Así mismo, se presenta el estudio histopatológico realizado en estos animales. También se han estudiado las vías de transmisión del LCDV a larvas y alevines de dorada, estableciéndose la existencia de múltiples rutas para la transmisión horizontal del virus. Por último, se presentan los resultados obtenidos en infecciones experimentales realizadas con artemias con el fin de establecer el carácter de hospedador para el LCDV de este crustáceo. ! 15 RESUMEN 2.1. Determinación de los órganos diana para la multiplicación del virus de la enfermedad de linfocistis en dorada. El principal objetivo de este apartado fue el estudio de los mecanismos de patogénesis del LCDV en alevines de dorada, y en particular la determinación de los órganos diana para la multiplicación vírica. Para ello, se realizaron muestreos en poblaciones de dorada de piscifactorías, incluyendo ejemplares asintomáticos, ejemplares enfermos (es decir, que mostraban la sintomatología típica de la LCD), y ejemplares recuperados de la enfermedad. Los órganos que se analizaron fueron aleta caudal, intestino, hígado, bazo, riñón y cerebro. La carga viral en los distintos órganos de alevines de dorada se determinó mediante qPCR, mientras que la cuantificación relativa de la expresión del gen que codifica la MCP viral se realizó mediante RT-qPCR. Para determinar las células diana para la replicación del LCDV en los distintos órganos analizados, se procedió a la detección de RNA mensajero viral mediante ISH. Por otra parte, también se estudiaron las alteraciones histopatológicas asociadas a la LCD en los grupos de peces antes mencionados, empleándose dos técnicas tintoriales generales: hematoxilinaeosina y hematoxilina-V.O.F. Los resultados obtenidos indican que la infección por LCDV en alevines de dorada presenta un carácter sistémico, tanto cuando estos muestran la sintomatología característica de la enfermedad como en el caso de infecciones subclínicas o asintomáticas. Estas infecciones subclínicas son además productivas, como lo demuestra el hecho de que fue posible detectar transcritos virales en todos los órganos analizados en ambos grupos de peces asintomáticos. Además, aunque la enfermedad es auto-limitada, el virus no se elimina después de que los animales se recuperen de la enfermedad, por lo que la infección es persistente, si bien con cargas virales sólo detectables mediante qPCR. Los análisis mediante ISH mostraron que el LCDV presenta un tropismo tisular amplio, replicándose al menos en fibroblastos de la dermis (donde la infección conlleva su transformación en linfocistes), hepatocitos, células del tejido hematopoyético de bazo y riñón, y en el cerebro. En cuanto a los estudios histológicos realizados, en animales enfermos se observaron alteraciones histopatológicas de distinta consideración según el ! 16 órgano considerado, mientras que en peces recuperados la mayoría de los órganos presentaron características histológicas similares a los de los animales sanos. En peces recuperados sólo se detectaron cambios patológicos en intestino e hígado, aunque estos daños son menos severos que los observados en peces enfermos. Esto indicaría que dichos daños están asociados a una elevada carga vírica y que son reversibles, es decir, desaparecen cuando el animal se recupera de la enfermedad. 2.2. Transmisión del virus de la enfermedad de linfocistis a larvas y alevines de dorada. Los análisis realizados en diversas piscifactorías españolas con historial de LCD demostraron que tanto larvas como alevines de dorada pueden ser portadores del virus, lo cual suele estar asociado con la transmisión vertical de los virus a partir de los reproductores. Los reproductores portadores de virus pueden excretarlos en sus fluidos reproductivos, infectando así a los óvulos en el momento de la fecundación o poco después. En el presente estudio, se analizaron muestras de sangre de reproductores de dorada utilizando PCRhibridación para el diagnóstico del LCDV. El genoma viral se detectó en un 17,5% de los animales analizados, y los huevos fecundados obtenidos a partir de estos reproductores también resultaron positivos para el virus, así como la mayoría de las larvas nacidas de ellos, lo que sugiere una transmisión vertical del LCDV. Para abordar la cuestión de si el LCDV se transmite en la superficie del huevo o intra-óvulo, los huevos fertilizados se desinfectaron con yodo. Las larvas desarrolladas a partir de estos huevos desinfectados resultaron negativas tanto por PCR-hibridación como por ISH, lo que demostraría la localización superficial del virus en los huevos. Las larvas obtenidas a partir de huevos LCDV-positivos presentaron DNA y antígenos víricos en la epidermis, lo que sugiere que los virus presentes en la superficie del huevo son capaces de infectar a dichas larvas. Sin embargo, cuando comienza la fase exotrófica, también se detectaron antígenos víricos en el tracto digestivo de algunas larvas, incluso antes de la introducción de ! 17 RESUMEN alimento vivo, lo que sugiere una transmisión a través del agua de virus excretados por larvas infectadas. Los rotíferos y los nauplios de artemia constituyen el alimento vivo más frecuentemente utilizado en el cultivo larvario y post-larvario de dorada. Sin embargo, estos invertebrados se han considerado como posibles vectores para la introducción de diferentes patógenos microbianos en las instalaciones de acuicultura, entre ellos, algunos patógenos víricos. Las larvas LCDV-negativas alimentadas con rotíferos contaminados mostraron antígenos víricos en el tracto digestivo a los dos días de iniciarse la alimentación, detectándose también el virus en la epidermis unos días después. Por tanto, los rotíferos actúan como vector para la transmisión del LCDV a larvas de dorada. Además, se detectaron antígenos víricos en cerebro e hígado de algunos animales, lo que sugiere que en larvas de dorada la infección por LCDV puede ser también sistémica. La diseminación del virus hacia órganos internos desde la epidermis y/o el tracto digestivo podría producirse por vía sanguínea. Por otra parte, se analizó la carga y la expresión génica virales en alevines de dorada de 0,5-1 g a los que se alimentó con metanauplios de artemia contaminados experimentalmente con LCDV. Los resultados obtenidos demostraron que las artemias pueden participar también en la transmisión vía alimentaria del LCDV. 2.3. Artemia sp. como hospedador del virus de la enfermedad de linfocistis. Artemia sp. es un crustáceo braquiópodo del orden Anostraca que constituye el alimento vivo más importante utilizado en acuicultura. Estudios previos realizados por nuestro grupo de investigación demostraron la presencia de partículas infectivas del LCDV en quistes comerciales de artemia, si bien pudo demostrarse que se trataba de una contaminación externa de los mismos, ya que no fue posible detectar el virus tras someterlos a un tratamiento de desinfección. Sin embargo, en los nauplios eclosionados a partir de estos quistes LCDV-positivos, al igual que los inoculados por baño con un aislado de LCDV procedente de dorada, fue posible la detección específica de genoma viral, localizándose este en el tracto digestivo de los nauplios. La detección del LCDV mediante ISH en nauplios de artemia infectadas por baño podría indicar que ! 18 estos actúan como bioacumuladores de patógenos desde el agua de cultivo. Además, como hemos indicado anteriormente, los metanauplios de artemia actúan como vector del LCDV, lo que implicaría la existencia de una transmisión horizontal del virus desde el agua a la cadena trófica. Sin embargo, estos resultados no permiten determinar si el artrópodo es sólo un vector mecánico o, por el contrario, es susceptible al LCDV. Por ello, en este apartado se realizaron infecciones experimentales en artemias (metanauplios, juveniles y adultos) con LCDV, analizándose en paralelo la carga y la expresión génica virales. Los resultados obtenidos sugieren que el LCDV es capaz de establecer una infección productiva en artemias, al menos en condiciones experimentales, si bien dicha infección cursa de forma subclínica. La carga viral (tanto en copias de DNA vírico como en título infectivo) en artemias es superior a la detectada en doradas asintomáticas infectadas por LCDV. Este es el primer caso descrito de un virus de peces que también infecta invertebrados, y amplía el rango de hospedador del LCDV a crustáceos. INTRODUCTION ! ! 21 INTRODUCTION 1. Current state of aquaculture of gilthead seabream (Sparus aurata L.) Global production of fish from aquaculture has grown substantially in the past five decades, being one of the fastest-growing animal-food-producing sectors and it currently represents more than 66% of global food fish consumption (FAO, 2014). Aquaculture is oriented not only to quantitative production but also to improve the quality of products, and its success is based on the control of species reproduction, technological innovation applied to facilities, and the development of specific foods. Most cultured marine fish species are of relatively high commercial value, sometimes because wild stocks are small or declining. Currently, the aquaculture production of some fish species is substantially higher than fish produced by capture fisheries. This is the case of gilthead seabream (Sparus aurata L.), traditionally cultured in Northern Italy, as well as in Southern Spain, that has become one of the main products of the European aquaculture (FAO, 2014). The gilthead seabream, belonging to the family Sparidae, is a common inhabitant in the Mediterranean Sea and the Eastern coastal areas of the Atlantic Ocean from the United Kingdom to the Canary Islands. It is a euryhaline fish present in both marine and brackish water environments, such as coastal lagoons and estuarine zones, in particular during the initial stages of its life cycle. This fish is gastronomically very appreciated and it is marketed both fresh and frozen. Large-scale production of gilthead seabream juveniles was definitively achieved in 1988-1989 in Spain, Italy and Greece. This species has demonstrated a high adaptability to intensive rearing conditions, both in ponds and cages, and its annual production increases regularly every year, with an estimated global production of about 173,024 tonnes in 2014 (APROMAR, 2015). At present, aquaculture production of gilthead seabream is recorded in more than 20 countries, with Greece, Turkey and Spain being the major producers. Its farming is also carried out in Egypt, Tunisia, Italy, Cyprus, Croatia, Malta, Israel, France and ! ! 28 Palmer et al., 2012). The genetic diversity of LCDV has been related to the host fish species (Kitamura et al., 2006a;b; Hossain et al., 2008). However, when studying the evolutionary relationship of LCDV and its hosts, Yan et al. (2011) did not obtain significant evidence of co-speciation between LCDV genotypes and their host fish species. Figure 2. Phylogenetic tree showing the relationship between viral isolates belonging to the nine different LCDV genotypes. These relationships are based on the amino acid sequence of the major capsid protein (MCP) gene. The neighbour-joining tree was obtained using MEGA4 (modified from Cano et al., 2010). 3.2. Virion structure LCDV are large icosahedral viral particles that, depending on the host fish species, may vary in size from 120 to 340 nm in diameter (Tidona & Darai, 1999; Paperna et al., 2001). The virus consists of a bilaminar capsid and a core that appears filamentous, displaying helicoidal symmetry (Madeley et al., 1978; Samalecos, 1986; Heppell & Berthiaume, 1992) (Figs. 3 and 4). The core is surrounded by a membranous structure that is clearly demonstrated in decaying virus (Smail & Munro, 2001). Negative staining electron images of decaying viruses show that the outer electron-lucent layer of the capsid is composed of knobs, possibly attached to the inner capsid layer by a fringe of fibril-like external ! ! 29 INTRODUCTION protrusions of 2.5 nm in length (Jancovich et al., 2012). The treatment of lymphocystis disease virions with papain before staining revealed a capsomer lattice structure, presumably because the papain removed the outer capsid (Samalecos, 1986). Figure 3. Diagram of the structure and organization of a virion of LCDV (obtained from ViralZone, Swiss Institute of Bioinformatics). Figure 4. Transmission electron micrograph of LCDV particles in the cytoplasm of a lymphocystis cell in the caudal fin of Sparus aurata. Scale bar = 200 nm. ! ! 30 Virions are heat labile and can be inactivated by ether, glycerol, 5iododeoxyuridine and UV-treatments (Wolf, 1988; Iwamoto et al., 2002). Freezingthawing cycles at -20°C may provoke a decrease in viral infectivity (Wolf, 1962). In contrast, the virions show stability to pH 6-9 and are resistant to ultrasonic treatment (Walker & Hill, 1980). 3.3. Chemical composition Lymphocystis disease virions are composed of 42% proteins, 17% lipids and 1.6% nucleic acids, with sugars most likely representing a major portion of the remaining unidentified components (Robin et al., 1983). SDS-PAGE analysis revealed the presence of 33 structural polypeptides, ranging from 4 to 220 kDa, in LCDV-1 virions isolated directly from fish tumours (Flügel et al., 1982). However, purified virions obtained from other fish species showed a different electrophoretic pattern of 23 to 31 polypeptides ranging from 30 to 210 kDa (Robin et al., 1984; Garcia-Rosado et al., 2004). A common characteristic of all LCDV particles is the presence of a MCP of approximately 50 kDa composed of 459 amino acids, which represents up to 45% of the total protein content (Flügel et al., 1982; Robin et al., 1986; Heppell & Berthiaume, 1992). The MCP is one of the antigenic proteins identified in LCDV that immuno-reacted with Japanese flounder antisera from diseased fish and also from formalin-inactivated LCDV vaccinated fish (Jang et al., 2011). The enzymatic activities associated with purified virions include a viral encoded ATP hydrolase, a protein kinase and a thymidine kinase (Flügel et al., 1982; Darai et al., 1983). Several authors have reported the presence of carbohydrates in LCDV. Robin et al., (1986) showed the presence of 10 glycoproteins in highly purified virus particles of an LCDV strain originally isolated from largemouth bass. In addition, Garcia-Rosado et al. (2004) reported the existence of 8 glycoproteins, with molecular weights ranging from 76 to 210 kDa, in viral particles isolated from gilthead seabream. Six of these glycoproteins presented a high content of mannose, and the other two contained a high proportion of sialic acid and Nacetylglucosamine, respectively. ! ! 31 INTRODUCTION Although LCDV is a non-enveloped particle, it may contain 5-17% lipids that are readily digested by a treatment with phospholipase that has been described for other iridoviruses (Robin et al., 1983; Chinchar et al., 2005). These phospholipids constitute an internal lipid membrane that lies between the DNA core and the viral capsid. The origin of the internal lipid membrane is unclear. The composition of the internal lipid membrane suggests that this membrane is not derived from host membranes but is rather produced de novo. However, it has been suggested that the internal lipid membrane is derived from fragments of the endoplasmic reticulum and plays a key role in virion assembly (Jancovich et al., 2012). The LCDV genome is a single lineal double-stranded DNA molecule of 102.6 kbp for LCDV-1 and 186.2 kbp for LCDV-C (Jancovich et al., 2012). This genome is circularly permuted, terminally redundant, and heavily methylated (22%), with a GC content of 29.9% for LCDV-1 and 27.2% for LCDV-C (Darai et al., 1983; Wagner et al., 1985; Tidona & Darai, 1997a; Jancovich et al., 2012). In addition, LCDV-1 DNA contains numerous short-direct, inverted and palindromic repetitive sequence elements (Schnitzler et al., 1987; Schnitzler & Darai, 1989; Jancovich et al., 2012). 3.4. Genome organization Complete DNA sequences of LCDV-1 and LCDV-C have been determined. The former encoded 195 potential open reading frames (ORFs), whereas LCDV-C possesses 240 potential ORFs (Tidona & Darai, 1997b; Zhang et al., 2004; Jancovich et al., 2012). In LCDV-1, 108 largely non-overlapping ORFs are likely to represent viral genes, and 38 show significant homology to proteins related to virus replication and transcription, such as DNA polymerase (ORF 135R), DNA polymerase processing factor (ORF 003L), DNA-dependent RNA polymerases (ORF 016L, ORF 025L and ORF 171R), DNA methyltransferase (ORF 005L), methyl-sensitive restriction endonuclease with specificity for CCGG target sites (ORF 178L), structure-specific endonuclease (ORF 191R), DNA-dependent ATPase (ORF 054R), DNA puff protein homolog (ORF 108L), proteins homologous to an early transcription factor subunit (ORF 132L), late promoter transactivator ! ! 32 protein (ORF 032R), dsRNA-specific ribonuclease (ORF 137R), thymidine kinase (ORF 136R) and ribonucleoside-diphosphate reductases (ORF 027R and ORF 176L). In addition, other putative gene products showed significant homology to proteins involved in the virus-host interaction, including an insulin-like growth factor, a tumour necrosis factor receptor family, thioredoxin, cysteine proteinase, several protein kinases, a tissue differentiation factor, a collagen type IX homolog, b-hydroxy steroid dehydrogenase and ATPase, to name a few (ORFs 010L, 022R, 035L, 036R, 043R, 047L, 063L, 080R, 088R, 093R, 094R, 095L, 122R, 125R, 128L, 153L, 158L and 167L, respectively) (Flügel et al., 1982; Koonin, 1993; Müller et al., 1995; Tidona et al., 1996; Tidona & Darai, 1997b; Sudthongkong et al., 2002; Essbauer et al., 2004; Kim & Lee, 2007; Pontejo et al., 2013). In the case of LCDV-C, Zhang et al. (2004) reported the presence of 240 potential ORFs and 176 non-overlapping putative viral genes. A search of the GenBank database using the 176 individual putative genes revealed 103 homologues to the corresponding ORFs of LCDV-1 and 73 potential genes that were not found in LCDV-1 or in other iridoviruses. Among these 73 genes, 8 genes contain coding sequences of conserved domains of cellular proteins, such as the caspase recruitment domain involved in apoptotic signalling (ORF 002L), thymidylate synthase (ORF 011L), the tumour necrosis factor receptor domain (ORF 016L), site-specific recombinase (ORF 047R), reverse transcriptase (ORF 051L), 7 transmembrane receptor (ORF 058L), the N-terminal domain of cell division protein 48 (ORF 209R) and collagen triple-helix repeat (ORF 216L). The remaining 67 novel genes do not show any significant homology with sequences in the public database. Recently, López-Bueno et al. (2016), using 454/Roche GS-FLX Titanium sequencing system and Illumina assembled contigs, together with PCR and Sanger sequencing reactions, were able to obtain the full-length genome sequence of a LCDV isolate from gilthead seabream (LCDV-Sa). The genome is 208.5 kbp in length, significantly longer than that of the two LCDV previously sequenced, and it is therefore the longest known vertebrate iridovirus genome. The GC content of the LCDV-Sa genome (33%) is higher than those of LCDV-1 (29.1%) or LCDV-C (27.2%). Using PASC, these authors calculated that LCDV-Sa shared only 54.7% identity with LCDV-C or 38.9% identity with LCDV-1. In addition, ! ! 33 INTRODUCTION LCDV-Sa showed evidence of heavy genomic rearrangements as compared to LCDV-C, and an almost complete absence of co-linearity stretches with LCDV-1. Annotation of the full length LCDV-Sa genome indicated the presence of 183 putative ORFs encoding proteins larger than 30 amino acids. These included all 26 conserved iridovirus core genes (Eaton, 2007). 3.5. Viral multiplication The replication mechanism of LCDV has not been investigated, but a model for frog virus 3 (FV-3), a member of the genus Ranavirus, has been proposed (Chinchar et al., 2009; Jancovich et al., 2012) (Fig. 5). The cellular receptor(s) for FV3 is unknown but viral entry is achieved by clathrin-mediated endocytosis. In the case of LCDV-C, a 27.8 kDa protein associated with betaactin in the plasma membrane of flounder gill cells has been identified as the virus receptor (Wang et al., 2011a). Following uncoating, viral cores enter the nucleus where first-stage DNA synthesis, and the synthesis of immediate early (IE) and delayed early (DE) viral transcripts, occurs. One or more virion-associated proteins act as transactivators and re-direct host RNA polymerase II to synthesize IE and DE viral mRNAs using the methylated viral genome as a template. The gene products encoded by the IE and DE viral transcripts include both regulatory and catalytic proteins. The viral DNA polymerase is involved in the first round of viral DNA synthesis. The newly synthesized viral DNA may serve as the template for additional rounds of DNA replication and early transcription, or it may be transported to the cytoplasm where the second stage of viral DNA synthesis occurs. In the cytoplasm, viral DNA is replicated as large, branched concatemers that are processed to mature DNA during DNA packaging. Viral DNA methylation also occurs in the cytoplasm of the host cell; although its precise role is uncertain, it is hypothesised to protect viral DNA from endonucleolytic attack. The transcription of late (L) viral genes occurs in the cytoplasm, and full L gene transcription requires prior DNA synthesis. Homologues of the two largest subunits of RNA polymerase II are encoded by all iridoviruses. Whether this viral DNA-dependent RNA polymerase functions only in the cytoplasm to transcribe L viral genes or ! ! 34 whether it also plays a role in continued early transcription has not yet been determined. Virion formation occurs in the cytoplasm within morphologically distinct areas named viral assembly sites. Within these assembly sites, concatemeric viral DNA is packaged into virions by a “headful” mechanism that results in the generation of circularly permutedand terminally-redundant genomes, similar to those reported in the T-even Enterobacteria phages of the family Myoviridae. Following assembly, virions accumulate in the cytoplasm within large paracrystalline arrays or acquire an envelope by budding from the plasma membrane. Figure 5. Iridovirus replication cycle. The multiplication cycle of frog virus 3 (FV3) is illustrated as proposed model for LCDV replication (based on Williams et al., 2005). 35 INTRODUCTION 4. Lymphocystis disease Lymphocystis disease (LCD) is a well-known fish viral infection that is characterized by hypertrophy of fibroblastic cells in the dermis connective tissue of affected fish, occasionally proliferating as true epithelial tumours (Samalecos, 1986). This viral disease affects a wide variety of freshwater, brackish and marine fish species. LCD was one of the first fish viral diseases reported in the 19th century (Wolf, 1988), and its viral aetiology was demonstrated by electron microscopy by Walker (1962) and the subsequent virus isolation on BF-2 cell line by Wolf (1962). Although this disease is rarely fatal, fish showing the characteristic symptoms cannot be commercialized, causing important economic losses (Masoero et al., 1986). Lymphocystis disease (LCD) has been described in more than 150 species of fish from both marine and freshwater environments (Anders, 1989; Marcogliese et al., 2001; Paperna et al., 2001; Bunkley-Williams et al., 2002; Sheng et al., 2007a; Hossain et al., 2008; Xu et al., 2014; Huang et al., 2015). The affected species belong to evolutionarily advanced orders of bony fish (teleosts), mainly including the families Cichlidae, Osphronemidae, Centrarchidae, Gobiidae, Chaetodontidae, Pomacentridae, Sciaenidae, Serranidae and Pleuronectidae. To date, LCD has not been reported in less-advanced fish orders, such as siluriformes, cyprinids and salmonids. The disease is cosmopolitan, being widely distributed in all continents (Plumb, 1993). In Europe, LCD is an endemic disease in the North Sea and Mediterranean zones, affecting both wild and cultured fish species, such as European flounder, common dab, European plaice, grey gurnard (Eutrigla gurnardus), gilthead seabream, blackspot seabream (Pagellus bogaraveo) and Senegalese sole (Paperna et al., 1982; Anders, 1989; Basurco et al., 1990; Moate et al., 1992; Garcia-Rosado et al., 1999; Dethlefsen et al., 2000; Alonso et al., 2005). LCD is also a common fish disease in Asian aquaculture, particularly affecting Japanese flounder, black rockfish, cobia, Japanese seabass, Japanese amberjack (Seriola quinqueradiata) groupers (orange-spotted grouper, Epinephelus coioides, brown-marbled grouper, E. fuscoguttatus, and giant ! ! 36 grouper, E. lanceolatus), and red seabream (Pagrus major), as well as ornamental aquarium fish species (Matsusato, 1975; Tanaka et al., 1984; Chen, 1996; Park & Sohn, 1996; Muroga, 1997; Chun, 1998; Xu et al., 2000; Zhang, 2002; Xing et al., 2006; Hossain et al., 2008; Xu et al., 2014; Huang et al., 2015). 4.1. Disease features The main characteristic of LCD is the appearance of small creamcoloured nodular lesions on the fish skin and fins (Colorni & Diamant, 1995; Sarasquete et al., 1998) (Fig. 6). Each nodule consists of an LCDV-infected cell, named lymphocyst or lymphocystis cell, of up to 1 mm in diameter (Paperna et al., 1982). These hypertrophied cells may occur singly or grouped in raspberry-like clusters of tumour appearance. These cellular aggregates are usually whitish in colour, but when they cover epithelial tissue that is rich in chromatophores, the chromatophores may render them greyish or darker (Wolf, 1988; Smail & Munro, 2001). In heavily affected fish, lymphocysts may cover the entire body, spreading from the gills to the fins (Paperna et al., 1982; Flügel, 1985; Le Deuff & Renault, 1993; Xing et al., 2006). Less frequently, they have also been described on eyes, causing exophthalmia, and internally over the mesenteries, peritoneum and several internal organs (Huizinga & Cosgrove, 1973; Russell, 1974; Dukes & Lawler, 1975; Howse et al., 1977; Wolf, 1988; Colorni & Diamant, 1995; Xing et al., 2006). Diseased fish show low growth rates, which may be caused by the anaemia generally associated with this disease (Nishida et al., 1998; Iwamoto et al., 2002). Mortalities are typically limited to those individuals whose swimming, breathing, or feeding is severely impaired by particularly large and cumbersome growths of infected cells (Colorni & Padros, 2011). In fish farms, LCD outbreaks may favour secondary bacterial infections, cannibalism and/or parasitic infestations, factors that may increase mortality rates (Williams et al., 2005; Colorni & Padros, 2011; Dezfuli et al., 2012; Haddad-Boubaker et al., 2013). LCD is a chronic and self-limiting disease that, depending on the host fish species and environmental conditions, may persist for a variable period of time (Williams, 1996). Thus, the LCD-associated lesions may be evident for 1 year in ! ! 37 INTRODUCTION cold-water fish, whereas they disappear after several weeks in warm-water species (Paperna et al., 1982; Gonzalez de Canales et al., 1996). Figure 6. Specimens of juvenile gilthead seabream showing lymphocystis nodules grossly visible on fins a body. LCDV infection has been described in bluegill (Lepomis macrochirus) (Dunbar & Wolf, 1966) and European plaice (Roberts, 1976). Although the time course for the development and regression of lymphocysts is quite different in both fish species (28 days at 25°C in bluegill compared to 3 months at 10°C in plaice), certain definitive stages can be recognized: ! ! 44 Polymerase chain reaction (PCR) is a rapid, sensitive and highly specific technique for detecting iridoviral infections (Mao et al., 1997; Grizzle et al., 2003). In the case of LCDV, several PCR techniques based on the sequences of MCP coding genes have been developed in recent years. Using this technique, Cano et al. (2007) successfully detected LCDV from different marine fish species (European flounder, common dab, European plaice and gilthead seabream) collected from both Northern and Southern Europe. PCR combined with blot hybridization demonstrated to be adequate for virus detection in tissue homogenates of asymptomatic gilthead seabream carriers (Cano et al., 2007; 2009a). Similar PCR-based assays have been developed by other authors to detect LCDV in other fish species, such as the Japanese flounder, black rockfish, turbot, redwing sea robin (Lepidotrigla microptera) and white-spotted puffer (Arothrom hispidus) (Xing et al., 2006; Hossain et al., 2007; Sheng et al., 2007a; Zhan et al., 2010). The aforementioned studies have demonstrated the applicability of the PCR-based methods to detect LCDV in asymptomatic carriers, but they do not provide quantitative results that can be useful in epidemiological and pathological studies. Based on competitive PCR technology, Zan et al. (2007) established a semi-quantitative method for LCDV detection in Japanese flounder tissues. Real-time PCR is a powerful technique that has been used for the detection and quantification of several viral fish pathogens, including different iridoviruses (Wang et al., 2006; Pallister et al., 2007; Gias et al., 2011), showing better sensitivity than conventional PCR. Regarding LCDV, Palmer et al. (2012) developed a real-time PCR assay using fluorogenic primers, which proved to be reliable in the detection and quantification of subclinical infected yellow perch. More recently, a new real-time PCR assay has been developed and applied for viral quantification in diseased and asymptomatic gilthead seabream (Ciulli et al., 2015). 4.3.4. Loop-mediated isothermal amplification (LAMP) LAMP is a technique in which DNA is quickly amplified under isothermal conditions with high specificity and sensitivity (Notomi et al. 2000). LAMP- ! ! 45 INTRODUCTION mediated diagnosis has been successfully used for the detection of viral pathogens in the aquaculture industry, including several iridoviruses (Caipang et al., 2004; Mao et al., 2008; Zhang et al., 2009; Ding et al., 2010; Sung et al., 2010; Min et al., 2013). Li et al. (2010) developed and evaluated a LAMP assay for the rapid detection of LCDV from both diseased and apparently healthy Japanese founders. The assay was found to be very specific because no cross-reactivity was obtained using other iridoviruses, and its detection limit was similar to that of real-time quantitative PCR. Due to LAMP amplifies under isothermal conditions (between 63 and 65 ºC) a thermal cycler is not required. In addition, LAMP products can be detected visually using several fluorescent dyes that bind to dsDNA, such as SYBR Green, calcein or ethidium bromide, or by the formation of a white precipitate, magnesium pyrophosphate, as a by-product of the amplification reaction. Therefore, LAMP can be widely used for viral diagnosis, particularly in resource-limited settings. 4.4. Lymphocistis disease virus transmission Only a few studies have been conducted on the fate of LCDV outside the host and whether it is able to remain viable for an extended period of time in water or sediments. However, it is classically assumed that viral transmission occurs through the skin and gills of fish by direct contact or by waterborne exposure (Wolf, 1988; Bowser et al., 1999; Kvitt et al., 2008). Trauma of the skin via handling or netting, mating, parasitism and aggressive behaviour favour viral transmission among fish (Wolf, 1988; Plumb, 1993; Smail & Munro, 2001). Sheng et al. (2007b) and Cano et al. (2009a) reported the possible transmission of LCDV by feeding in aquaculture facilities. Artemia nauplii have been considered as possible vector for the introduction of bacteria, viruses and protozoa into aquaculture facilities for decades, mainly because of its ability to filter and accumulate particles from aquatic environment, making this crustacean a potential candidate for the transmission of several pathogenic microorganisms. According to some authors, Artemia could act as reservoir or mechanical vector of pathogenic bacteria, such as Bacillus, Erwinia, Micrococcus, Staphylococcus and Vibrio (Austin et al., ! ! 46 1982; Tatani et al., 1985; Muroga et al., 1987; Nicolas et al., 1989). Mortensen et al. (1993) highlighted the potential role of Artemia as a vector for infectious pancreatic necrosis virus (IPNV) in larvae of turbot, while Skliris et al. (1998) concluded that both rotifers and Artemia could act as mechanical vector for viral nervous necrosis virus (VNNV). Previous studies by Cano et al. (2009b) showed the presence of infectious LCDV in commercial Artemia cysts, although it was an external contamination, since the virus could not be detected after a decapsulation treatment. Viral genome and antigens were detected in the digestive tract of nauplii hatched from contaminated cysts, but not at the umbrella and instar I stages. This might indicate that nauplii can accumulate virus from water contaminated with LCDV particles derived from breaking cyst. Moreover, nauplii also became LCDV-contaminated after bath challenge. These results suggest that Artemia should be consider a key factor in the introduction of LCDV in fish hatcheries, and therefore, it is critical to perform further studies to establish its role in the transmission of LCDV to cultured fish. In aquaculture facilities, a high percentage of the fish population could be infected by LCDV, likely reflecting the ease of horizontal transmission, and viral infection incidences up to 70% have been described (Paperna et al., 1982; Sano, 1988; Matsuoka, 1995; Xing et al., 2006). The prevalence of LCD is affected by fish density, human manipulation, low salinity, water temperature, reduced oxygen conditions, nutritional deficiencies and chemical and biological water pollution (Paperna et al., 1982; Bowser et al., 1988; Berthiaume et al., 1993; Sindermann, 1996; Vethaak & Jol, 1996; Mellergaard & Nielsen, 1997; Austin, 1999; Bowser et al., 1999; Grygiel, 1999; Kitamura et al., 2007). Hossain et al. (2009) demonstrated the importance of temperature on the persistence of LCDV in Japanese flounder epidermal tissues. These authors found that lymphocystis cells appeared on the skin and fins at 35 days post-challenge at 20°C, but no clinical signs were observed in the fish reared at 10° and 30°C, although LCDV could be detected by PCR. They concluded that at low temperatures, LCDV is able to persist over a long period of time in the fish epidermis, producing a subclinical infection. ! ! 47 INTRODUCTION 4.5. Disease control and prevention Viral disease prevention and control rely on the application of specific prophylactic measures (i.e., vaccination), or, alternatively, on the use of general control strategies, such as improved husbandry and water quality, better nutrition and lower stocking densities (OIE, 2014). Control of LCDV in intensive culture operations would demand scrupulous disinfection procedures at all stages of production, screening and quarantine of each fish lot to be introduced, and treatment of raw seawater used in the fish facility (Bowden et al., 1995). However, only a few studies have been performed on physical and chemical treatments against LCDV. Havikrishnan et al. (2010c) used a bath treatment with formalin, hydrogen peroxide and Jenoclean for LCDV-infected Japanese flounder. The authors concluded that these chemical agents enhanced the fish innate immune response and increased the fish resistance to the disease. However, these treatments cannot be systematically applied in aquaculture practice. For this reason, these authors evaluated the effect of herbal extracts and probiotics added to the fish diet in the course of LCDV infection in Japanese flounder, concluding that they act as immunostimulants that reduce the incidence of LCD (Havikrishnan et al., 2010a;b). Although there is no commercial vaccine available for LCDV infection, both inactivated and genetically engineered vaccines targeting LCDV have been designed and evaluated in recent years. Formalinand heat-inactivated LCDV were used as vaccines, and proved to have a protective effect in Japanese flounder (Yoshimizu & Iwamoto, 2001; Xu et al., 2011b). Nevertheless, its use is hampered by the necessity to obtain large amounts of purified virus particles directly from diseased fish lesions. DNA vaccination is based on the administration of plasmid DNA (pDNA) encoding a protective antigen, rather than the antigen itself. The subsequent expression of the antigen by cells in the vaccinated hosts triggers the host immune response. A single intramuscular injection of low amounts of DNA induces rapid and long protection in fish against economically important viruses affecting aquaculture production (Lorenzen & LaPatra, 2005). Zheng et al. (2006) ! ! 48 designed a DNA vaccine against LCDV composed of a plasmid containing a 0.6-kbp fragment of the MCP gene of LCDV-C. The expression of several immunerelated genes significantly increased after vaccination, and specific anti-LCDV immunoglobulins were also detected in the sera of vaccinated fish (Zheng et al., 2010). In addition, this vaccine induced effective protection against LCD in Japanese flounder after intramuscular injection (Zheng et al., 2011). Oral DNA-based immunotherapy is a new strategy for fish immunization in intensive culture. However, the rate of degradation of DNA vaccines by nucleases and acidic conditions in the fish gastrointestinal tract may reduce vaccine efficiency. To avoid this, the vaccine DNA can be delivered encapsulated in microor nanoparticles that prevent its degradation. Microspheres of alginate, chitosan, and poly/DL-lactide-co-glycolide (PLGA) were tested by Tian et al. (2008a;b;c) for oral delivery of the LCDV pDNA vaccine cited above. Following immunization, the authors detected transgene expression in several organs from fish vaccinated with encapsulated pDNA. The encapsulated vaccine also induced higher levels of antibodies compared to control fish vaccinated with naked pDNA. Later, Tian & Yu (2011) demonstrated a significant increase in resistance to LCDV infection after oral administration of the pDNA vaccine encapsulated into PLGA nanoparticles. 5. Artemia as live food in aquaculture Fish larval rearing is generally carried out under controlled hatchery conditions and requires specific culture techniques. The incomplete development of the larval digestive system during the early first-feeding has become one of the major bottlenecks preventing the successful production of many farmed fish. Usually, formulated feeds do not fulfil all the nutritional requirements and, therefore, result in poor growth and low survival of fish larvae. However, live food organisms seem to provide essential nutrient during larviculture, such as fatty acids, free amino acids, vitamin C and carotenoids. In addition, live food has a triggering effect by their continuous movement, allowing an enhanced perception, while the swimming activity assures a good ! ! 49 INTRODUCTION distribution of food items in the water column, facilitating more frequent encounters with the developing larvae which in most cases have a low mobility (Lavens & Sorgeloos, 1996). Brine shrimps of the genera Artemia (Crustacea, Branchiopoda, Anostraca) constitute the most important live food for aquaculture (Sorgeloos et al., 1986). Easy management, development characteristics, small size and high nutritional value, makes them suitable for both larvae and juvenile fish culture (Tizol, 1994). At present, it is the best suitable live food and, in many cases, the only one appropriate for many aquatic species in their early stages of life. They are usually supplied as Artemia nauplii or metanauplii, depending on the fish size at feeding. 5.1. Morphology and life cycle In its natural environment, under determined stressful conditions, Artemia produces cysts that are metabolically inactive and do not further develop as long as they are kept dry. Artemia cyst consisted mainly of three external structures (Fig. 7) that enclosed the resting embryo (Morris & Afzelius, 1967; FAO, 1986): (i) Corium or alveolar layer: the external envelope of the cyst composed of lipoproteins impregnated with chitin and haematin. The haematin concentration determines the colour of the shell. Its main function is to provide protection for the embryo against mechanical disruption and UV radiation. (ii) Outer cuticular membrane: multilayer membrane that acts as a permeability barrier and protects the embryo from penetration by molecules larger than the CO2 molecule. (iii) Embryonic cuticle: a transparent and highly elastic layer separated from the embryo by the inner cuticular membrane (this becomes the hatching membrane during hatching incubation). ! ! 50 Figure 7. Diagram showing the structure of Artemia cyst. (A) Corium or alveolar layer; (B) outer cuticular membrane; (C) embryonic cuticle; (D) inner cuticular membrane; (E) embryo. Upon immersion in seawater, the biconcave-shaped cysts hydrate and become spherical (Fig. 8), and inside the shell, the embryo resumes its interrupted metabolism. Approximately 20 h later, the outer membrane bursts (this is called “breaking” stage) and the embryo appears, surrounded by the hatching membrane. While the embryo hangs underneath the empty shell (“umbrella” stage), the development of the nauplius is completed and within a short period of time the hatching membrane is ruptured (“hatching”) and the free-swimming nauplius is released (FAO, 1986). Figure 8. Artemia cysts dehydrated (A), hydrated (B), and decapsulated (C). ! ! 51 INTRODUCTION The first naupliar stage (also named instar I) has an average size of 400 to 500 µm in length, is orange-brown (for accumulation of yolk reserves) and has three pairs of appendages: two pairs of antennae and jaws. In this first stage, the digestive system is not yet functional, since the mouth and anus still remain closed. About 24 h later, a second moult occurs and the nauplius goes to a second larval stage, named instar II, which has a fully functional digestive system and is now able to eat small food particles between 1 and 40 µm. Brine shrimps grow and develop by moulting, going through a series of 14 to 17 different stages (nauplii, metanauplii and juveniles) in which major changes, both morphological and functional, occur. Adult Artemia (10-12 mm in length) has an elongated body with two stalked complex eyes, a linear digestive tract, sensorial antennulae and 11 pairs of functional thoracopods. The male has a paired penis in the posterior part of the trunk region and two hooked graspers in the head. The female can easily be recognized by the brood sac or uterus situated just behind the 11th pair of thoracopods (Fig. 9). Figure 9. Schematic representation of adult brine shrimps. Female are slightly larger than males and their brood sac are easy visible to the naked eye (adapted from http://learn.genetics.utah.edu). ! ! 52 Eggs develop in two tubular ovaries in the abdomen and migrate via two oviducts into the uterus. Fertilized eggs usually develop into free-swimming nauplii (ovoviviparous reproduction) that are released by the mother. In extreme conditions (e.g., high salinity, low oxygen levels), the embryos only develop up to the gastrula stage. At this moment, they get surrounded by a thick shell (secreted by the brown shell glands located in the uterus), enter a state of metabolic cessation or dormancy (diapause), and are then released by the female as cysts (oviparous reproduction). The cysts usually float in the high salinity waters and are blown ashore where they accumulate and dry. Cysts are in a state of quiescence and can resume their further embryonic development when hydrated in optimal hatching conditions (FAO, 1986; Polanco, 2000). ! OBJETIVOS ! ! CHAPTER 1: DIAGNOSTIC METHODS FOR LYMPHOCYSTIS DISEASE VIRUS 63 CHAPTER 1 1. Real-time PCR assay for lymphocystis disease virus genotype VII detection and quantification 1.1. INTRODUCTION The only adequate measure for LCD prevention in the aquaculture systems is the use of general prophylactic practices, such as good husbandry practices, reduced stocking density and the virological control of fish to be introduced in the farm facilities in order to detect carrier fish (Anders, 1989). These animals may pose a risk for the introduction of LCDV in fish farms, as direct contact between fish specimens is considered the main route of LCDV spreading (Wolf, 1988). Moreover, asymptomatic carrier breeders may also be involved in LCDV transmission to fish larvae (Chapter 2, section 2). The detection of subclinical viral infections in carrier fish requires the use of sensitive diagnostic methods (Sanz et al., 1992). In this context, PCR-based methods have proved to be adequate for LCDV detection in apparently healthy fish (Cano et al., 2007; Zan et al., 2007; Kvitt et al., 2008; Hossain et al., 2009; Haddad-Boubaker et al., 2013). The PCR-hybridization assay developed by Cano et al. (2007) not only allowed the detection of LCDV in carrier gilthead seabream, but also in rotifer and artemia used as live food for larval stages, which makes it a valuable tool for the detection of other potential LCDV foci in fish farms (Cano et al., 2009). Nevertheless, this assay is relatively time-consuming, not readily applied to screening large sample numbers, and does not provide quantitative results, that can be useful in epidemiological and pathological studies of LCDV. Real-time PCR (qPCR) has been used for the detection and quantification of numerous viral fish pathogens, including LCDV in yellow perch (Palmer et al., 2012), and has been proved to be useful to overcome the disadvantages of conventional PCR above mentioned (Pallister et al., 2007; Pepin et al., 2008; Cutrín et al., 2009). Recently, a qPCR assay has been developed and applied for the detection and quantification of LCDV in a low number of samples of diseased and recovered gilthead seabream. Nevertheless, the use of the assay 64 for LCDV monitoring in different stages of the fish production cycle was not investigated (Ciulli et al., 2015). The objective of this study was to establish the applicability of a new qPCR assay for LCDV genotype VII diagnosis in surveillance studies. In addition, this assay has been evaluated using samples from a gilthead seabream hatchery, in order to demonstrate its utility to detect several sources of LCDV in the fish farm. 1.2. MATERIALS & METHODS 1.2.1. Sample collection and DNA extraction Juvenile gilthead seabream specimens were sampled at four aquaculture farms. In two of these farms (farms A and B), a total of 11 diseased and 25 asymptomatic (i.e., without signs of LCD) fish were collected during an outbreak of LCD, whereas in the other two (farms C and D), only asymptomatic fish were observed, and 24 of them were collected. Juvenile fish were euthanized by anaesthetic overdose (MS-222) (Sigma-Aldrich) before sampling. Samples of caudal fin (approximately 1 cm2 in size) were aseptically cut off, frozen immediately at -20 °C in dry ice and sent to the lab. In addition, 9 gilthead seabream samples were collected at the hatchery in another farm suffering LCD outbreaks over several years (farm E). Samples consisted of fertilized eggs (one sample of 100 mg), larvae (4 pools of 10–15 animals), and fingerlings (up to 2 g of weight) (4 pools of 5–10 animals). No LCD clinical signs were observed in any of the sampled fingerlings. Larvae and fingerlings were also euthanized by anaesthetic overdose. One sample of each rotifers (100 mg), commercial artemia cysts (100 mg), decapsulated artemia cysts (100 mg) and artemia metanauplii (100 mg) used as live food for larvae were also collected. These samples were washed with sterile seawater, gently dried and fresh frozen for shipment to the lab. In this farm, 50 broodstock were also analysed. These animals neither showed symptoms of LCD nor had LCD history, as stated by the farm records. For sampling purposes, these specimens were anaesthetized with MS-222 in seawater at a final concentration 65 CHAPTER 1 of 30 mg/ml. For each specimen, samples of caudal fin and blood were obtained. Blood samples were collected from the branchial arches using SMonovette 4.5 ml LH (Sarstedt), chilled to 4 °C and sent to the lab for analysis within 24 h, whereas samples of caudal fin were obtained and stored as described above. In each aquaculture farm, specialized personnel carried out the sampling procedures described. Fish used in this study have been treated in compliance with the Spanish legislation (RD 53/2013, BOE no. 34). Samples were homogenized in Leibovitz’s L-15 medium (Gibco) (10%, w/v) as described previously (Alonso et al., 2005), except samples of eggs, rotifers and artemia, which were ground in liquid nitrogen. Total DNA was extracted from 200 µl of tissue homogenates or 50 mg of tissue powder using the QIAamp DNA Minikit (Qiagen) according to the manufacturer’s instructions. Finally, DNA was extracted from 200 µl of heparinised blood using the ReliaPrep Blood gDNA Miniprep System (Promega). In all cases, DNA was eluted in a final volume of 100 µl and stored at -20 °C until use as template for PCR-hybridization and qPCR. Prior to PCR assays, purified DNA was quantified spectrophotometrically using a NanoDrop 1000 spectrophotometer (Thermo Scientific), and DNA was diluted to achieve a final concentration of 20 ng/µl. 1.2.2. PCR-hybridization LCDV genome was detected using the PCR-hybridization assay described by Cano et al. (2007). Briefly, a 270-bp fragment of the MCP gene of LCDV was amplified by PCR. PCR products were denatured and blotted onto a Hybond-N nylon membrane (GE Healthcare), and hybridization was carried out using an LCDV-specific digoxigenin (DIG)-labelled probe and the chemiluminescent substrate CSPD (Roche). 66 1.2.3. Cloning a fragment of MCP gene LCDV isolate SA9 was used as source of viral DNA (Cano et al., 2006). A fragment of the viral MCP gene was amplified by PCR using the primers LCDVs-F and LCDVs-R described by Kitamura et al. (2006b). PCR was performed in a 50-µl reaction volume containing 10 µl of 5X Colorless GoTaq Flexi Buffer (Promega), 3 mM MgCl2 (Promega), 5 µl of 0.2 mM dNTP (Roche), 1.25 U of GoTaq DNA polymerase (Promega), and 2 µl of each primer at 15 pmol/µl. DNA was amplified by the use of one denaturation step at 95 °C for 5 min, followed by 35 cycles of denaturation (95 °C for 1 min), annealing (50 °C for 30 s), and extension (72 °C for 1 min), and a final extension step at 72 °C for 10 min. PCR products were run in 2% agarose gels, purified using the High-Pure PCR Product Purification kit (Roche), and cloned into the pCR4-TOPO vector (TOPO TA cloning kit) (Invitrogen) for subsequent transformation in Escherichia coli (One Shot competent cells, Invitrogen) following manufacturer’s instructions. Recombinant plasmid DNA was purified from E. coli cells with a commercial kit (High Pure Plasmid Isolation kit, Roche), and insert size was verified by PCR and sequencing, using the M13 primers provided in the cloning kit. The cloned MCP gene fragment was 609 bp in length, corresponding with nucleotide positions 99 to 707 of the LCDV SA9 MCP gene (GenBank accession no. GU320728). 1.2.4. Real-time PCR assay Primers for qPCR (RT-LCDV-F: 5′-ACGTTTCTCGAGGCGGAGAT-3′, and RTLCDV-R: 5′-CGGACGTTTGCTTGACCAA-3′) were designed to target the MCP gene of LCDV genotype VII, using Primer Express Software v3.0 (Applied Biosystems). This primer set generates a 150-bp amplicon within the 609-bp cloned fragment of MCP gene (nucleotide positions 173 to 322 of the LCDV SA9 MCP gene). Real-time PCR reactions were carried out in 96-well plates (Applied Biosystems), in a final volume of 50 µl containing 25 µl of 2x Power SYBR Green PCR Master Mix (Applied Biosystems), 3 µl of each primer at 15 pmol/µl, and 10 µl of DNA. PCR amplifications were performed in a 7500 Real-Time PCR System 67 CHAPTER 1 (Applied Biosystems). The thermal profile was: 50 °C for 2 min; 95 °C for 10 min, and 40 cycles at 95 °C for 15 s and 60 °C for 1 min. Finally, dissociation curve analysis was carried out automatically in order to allow detection of non-specific amplification products. 1.2.5. Standard curve for LCDV quantification To quantify the amount of viral DNA in different samples, a standard curve was generated using the recombinant plasmid described above. The concentration of the purified plasmid was determined by spectrophotometry as described above, and the plasmid stock was diluted to serve as template for qPCR (10-fold serial dilutions ranging from 106 to 102 copies, and then two-fold dilutions from 50 to 25 copies, and from 16 copies to 1 copy). The sensitivity of the qPCR assay was also determined in terms of infectious viral particles. LCDV SA9 stock was titrated in SAF-1 cells, using the 50% cell culture infectious dose (TCID50) endpoint dilution assay as described previously (Alonso et al., 2005). An aliquot of 1 ml of the viral stock with a titre adjusted to 1 x 106 TCID50/ml was subjected to a 10-fold serial dilution in Leibovitz’s L-15 medium. The DNA of each dilution was extracted as previously specified, and used for qPCR. The amount of infective virus analysed per reaction was 1 x 105 to 1 x 100 TCID50. Milli-Q water and DNA from one LCDV-negative gilthead seabream sample (Cano et al., 2007) were included within each qPCR run as no-template and negative controls, respectively. In each 96-well, plasmid dilutions for the standard curve were run along with the samples and controls, using three technical replicates. The number of copies of LCDV DNA in each well was calculated from its cycle threshold (Ct) value by interpolation in the standard curve (Ct versus log copy number), using the SDS Software v1.3 (Applied Biosystem). The amplification efficiency (E) was calculated from standard curves using the formula E = (10-1/S - 1) x 100 (S being the slope of the linear fit). Viral loads in samples were calculated as the mean of the three replicates, and expressed as viral DNA copies per milligram of tissue (per µl in blood samples). 68 1.2.6. Assay repeatability and reproducibility To evaluate the precision of the qPCR, the intraand inter-assay variability was determined using the recombinant plasmid. To assess intra-assay variation, four plasmid DNA dilution series (from 105 to 2 copies per reaction) were prepared and tested simultaneously in the same plate. Four separate PCR runs were carried out to assess inter-assay variation, using also seven dilutions of plasmid DNA. The mean, standard deviation (SD) and coefficient of variation (CV) were calculated independently for each DNA dilution. 1.3. RESULTS 1.3.1. Evaluation of the real-time PCR assay Specificity of the qPCR was determined by analysis of the dissociation curves generated in each experiment. Standard and positive samples gave a single PCR product with a melting temperature of 77.7 ± 0.5 °C, which correspond with that deduced from the sequence of the expected fragment. The size of amplicons was monitored by agarose gel electrophoresis, and bands were observed at the expected size (150 bp). The linear dynamic range, efficiency and precision of the qPCR assay were evaluated using a recombinant plasmid containing a 609-bp fragment of the LCDV MCP gene. Standard curves generated from four independent assays demonstrated a linear relationship between the amount of plasmid DNA and Ct values over a wide range of concentration, from 106 copies (Ct = 16.25 ± 0.48) to 2 copies (Ct = 34.09 ± 1.14) per reaction (Fig. 10A). The regression analysis yielded a correlation coefficient (r) ≥ 0.994 and an amplification efficiency of 101.89 ± 5.11%. The mean intra-assay variation was 1.38 ± 0.87% when analysing four replicates of plasmid dilutions, whilst the mean inter-assay variation among four experiments was 2.63 ± 0.48% (Table 2). These CV values were considered acceptable to validate the repeatability and reproducibility of the assay. 69 CHAPTER 1 A linear relationship between the infective titre of a viral suspension and Ct values was also observed for viral amounts ranging from 1 x 104 to 1 x 100 TCID50 per reaction (r = 0.999) (Fig. 10B), with 1 x 100 TCID50 yielding a mean viral DNA copy number of 1.2 x 102. Figure 10. Dynamic range and sensitivity of the real-time PCR assay for LCDV detection. (A) Standard curve obtained using dilutions of plasmid DNA ranging from 106 to 2 copies per reaction. Linear regression was performed on mean data from four separate runs. The logarithm to base 10 (log) of the plasmid copy number versus the cycle threshold (Ct) value is represented. (B) Standard curve showing a linear relationship between the log of the amount of infective virus (expressed in TCID50) per reaction and their corresponding Ct values. 76 Loop-B). Sequences and location of the primers within the MCP gene are shown in Table 6 and Fig. 11, respectively. Table 6. Sequences of primers designed for the LCDV LAMP assay. Primers Sequences (5’ – 3’) F3 GTTCTGGATCGCCCAATT B3 TATCATGTTGTCGTAACCTACG FIP (F1c + F2) AACTTAACAGCGGGTATACGCATCTCGAGGCGGAGATTAC BIP (B1c + B2) CGGCACGATTCGATGGTGTAATGAGCCACCAAATCGTTAA Loop-F CCGTCATCCAGGCATTGA Loop-B CAAGCAAACGTCCGTTCAAT Figure 11. Primers for the detection of LCDV by LAMP. Partial DNA sequence of the MCP gene of LCDV (accession no. GU320735.1) showing target region of the primers. The suitability of the LAMP primers for LCDV genotype VII detection was tested in silico through ClustalW alignment analysis. The analysis was performed using all the MCP sequences belonging to genotype VII (from isolates of both gilthead seabream and Senegalese sole), available at GenBank (accession nos. 77 CHAPTER 1 GU320724.1 to GU320739.1, EF184306.1, HE650105.1, and HE650106.1). The specificity of the outer primers was also analysed by means of a virtual PCR using Serial Cloner v2.6.1 (http://serialbasics.free.fr/Serial_Cloner.html). 2.2.3. Construction of recombinant plasmid A fragment of 1356 bp of the viral MCP gene of LCDV SA25 was amplified by PCR using the primers LC1-F and LC1-R described by Kitamura et al. (2006a), and cloned into the pCR4-TOPO vector as previously specified (section 1.2.3). The concentration of the recombinant plasmid was spectrophotometrically determined, and the plasmid stock was diluted to serve as template for LAMP reaction. 2.2.4. Optimization of LAMP reaction temperature LAMP reactions were carried out in a final volume of 25 µl containing 15 µl of Isothermal Master Mix (OptiGene), 1 µM each of FIP, BIP, F3 and B3 primers, and 0.5 µM each of Loop-B and Loop-F primers. Aliquots of 5 µl of plasmid DNA (equivalent to 105 copies) were added as template. Isothermal amplifications were performed in a Genie® II system (OptiGene) that allows real-time monitoring of LAMP reactions. The reaction temperature was optimized using a block gradient from 62-69 ºC. Reactions were incubated at chosen temperature for 40 min, and then subjected to a slow annealing step (0.05 ºC/s) from 95 ºC to 75 ºC to identify specific amplification. The amplification ratio (change in fluorescence over time), annealing/melting temperature (Ta) of the product, and time of positivity (Tp) were obtained from the Genie® II software (OptiGene). 2.2.5. Specificity and sensitivity of the LAMP assay The specificity of the developed LCDV LAMP assay was evaluated by testing the six fish viruses mentioned above. DNA from LCDV SA25 and SAF-1 cells were used as positive and negative controls, respectively. Viral and cellular DNAs 78 were extracted using the EZ1 Virus Mini Kit (Qiagen), according to the manufacturer’s instructions. To determine the sensitivity of the assay, ten-fold serial dilutions (106 to 100 copies) of the recombinant plasmid were used as template for LAMP. Plasmid DNA was diluted in Milli-Q water supplemented with DNA extracted from SAF-1 cells (100 ng per reaction). The optimized LAMP protocol was performed for three independent assays. Standard curves were generated by plotting the number of copies of plasmid DNA against Tp for each particular concentration. 2.2.6. Application of the LAMP assay for LCDV detection in fish samples The suitability of the LAMP assay for the detection of LCDV was evaluated by testing samples from both LC-diseased and asymptomatic juvenile fish. Total DNA was extracted from 30 mg of caudal fin by using the E.Z.N.A. Tissue DNA Kit (Omega Bio-tek), following the protocol provided in the kit. DNA was eluted in a final volume of 50 µl and stored at -20 ºC until used as template for LAMP and qPCR. The LAMP assay was performed as specified above, but the amplification was monitored for up to 50 min. The qPCR analysis was carried out using the methodology described in section 1.2. Finally, a modified Hot-SHOT protocol (Meeker et al., 2007) was evaluated as a quick DNA extraction method, using caudal fin samples obtained from one LC-diseased gilthead seabream specimen. Briefly, a sample of 100 mg of caudal fin was ground using a sterile mortar and pestle, and transferred to a microcentrifuge tube containing 200 µl of 50 mM NaOH. After brief vortexing, the mixture was incubated at 95 °C for 20 min and then cooled to 4 °C, followed by the addition of 1/10 volume of 1 M Tris-HCl (pH 8.0) to neutralize the alkaline solution. Three µl of the supernatant obtained was immediately used for the LAMP assay. In parallel, DNA from another 100 mg of the same sample was extracted by using the EZ1 Virus Mini Kit, and 3 µl was used as template for LAMP. As positive control, 2 µl of LCDV SA25 DNA was combined with 3 µl of DNA from the Hot-SHOT extraction method to assess possible inhibition of the LAMP assay by the Hot-SHOT supernatant. DNA from SAF-1 cells was used as negative control. 79 CHAPTER 1 2.3. RESULTS 2.3.1. Optimization of LAMP reaction temperature The LAMP assay was evaluated at 62-69 ºC using as template 105 copies of the recombinant plasmid containing a fragment of the LCDV MCP gene. Products were amplified across this range of temperature with varying rapidity, with Tp values lower than 18 min. A temperature of 64 ºC was chosen as the optimal amplification temperature (Tp = 12 min, 57 s), and was applied in the subsequent analysis. 2.3.2. Specificity and sensitivity of the LCDV LAMP assay No amplification was observed in LAMP reactions performed with DNA from SAF-1 cells (negative control), neither with DFV, FV3 or CyHV3 DNAs (Fig. 12A). When DNA from EHNV and ESV were used as template, some fluorescence could be observed but only after fluorescence normalization (Fig. 12A). With nonnormalized data, fluorescence signal was lower compared to that showed by negative control (data not shown). Nevertheless, the annealing curves of the amplified products showed a single peak in the range of 84-86 ºC, only for LCDV DNA (Fig. 12B). These results indicate that the LAMP primers were specific to isolates of LCDV, and no cross-reactions have been observed with the other fish DNA viruses analysed. Pairwise comparisons among MCP gene sequences belonging to LCDV genotype VII showed nucleotide identities ≥ 98%, and no mismatches were recorded in the regions corresponding to LAMP primers, except for two sequences (GU320724.1 and GU320734.1, with 100% identity) that showed two transitions at nucleotides 4th and 16th in primer Loop-B region. No amplifications were obtained while running in silico PCR with F3/B3 primers, using as template MCP sequences belonging to different LCDV genotypes except for genotype VII. 80 Figure 12. Specificity of the LCDV LAMP assay. (A) Genie® II software screenshot showing real-time isothermal amplification of DNA isolated from different viruses. Fluorescence was normalized to the background for 0-180 s. (B) Anneal derivative of isothermal amplified products. Negative control: DNA from SAF-1 cells. 81 CHAPTER 1 Figure 13. Sensitivity of the LAMP assay for LCDV detection. (A) Genie® II software screenshot showing real-time isothermal amplification of ten-fold serial dilutions of plasmid DNA ranging from 106 to 1 copies per reaction. Fluorescence was normalized to the background for 0-180 s. (B) Anneal derivative of isothermal amplified products. Negative control: DNA from SAF-1 cells. Ten-fold serial dilutions of the recombinant plasmid ranging from 106 to 1 copies were used as template for the LAMP assay. The analytical sensitivity of the assay was estimated to be 10 copies per reaction (Fig. 13). Data analysis of three 82 independent assays revealed a significant correlation between the number of DNA copies and the Tp values (r ≥ 0.983) (Fig. 14). Figure 14. Correlation between the logarithm to base 10 (log) of the plasmid copy number and the time of positivity (Tp) value performed on mean data from three independent assays. 2.3.3. LCDV diagnosis by LAMP The applicability of the LAMP assay for detection of LCDV in fish farms was evaluated by comparing its diagnostic capability with that of the qPCR assay (Table 7). LCDV was detected by LAMP in all samples from diseased gilthead seabream analysed, and even in those samples collected from asymptomatic carrier fish, with qPCR-estimated viral loads ranged between 1.9 and 2.7 x 101 copies of viral DNA per mg of tissue. Furthermore, no isothermal amplification was observed in samples where no LCDV genome was detected by qPCR. Tp values depend on the disease status, ranging from 3 min, 34 s and 11 min, 25 s for LCdiseased fish, and between 17 min, 32 s and 44 min, 26 s for asymptomatic carrier fish (Table 7). A correlation between the viral load estimated by qPCR and the Tp value (r ≥ 0.963) was obtained for viral loads above the analytical sensitivity of the assay (using the protocol as described, this correspond to > 3.3 copies of viral DNA/mg). 83 CHAPTER 1 Table 7. Evaluation of the LCDV LAMP assay using caudal fin samples from juvenile fish. a Typical lymphocystis disease signs: +, presence; -, absence. b Copies of viral DNA per mg of tissue. c Absence of isothermal amplification. Tp: Time of positivity. Sample LCD signsa Tp (LAMP assay) Viral loadb (qPCR assay) Gilthead seabream No. 1 + 9 min, 59 s 1.1 x 105 No. 2 + 9 min, 32 s 1.3 x 105 No. 3 + 3 min, 34 s 1.2 x 106 No. 4 + 3 min, 42 s 1.1 x 106 No. 5 + 7 min, 26 s 3.5 x 105 No. 6 + 11 min, 5 s 2.8 x 104 No. 7 + 9 min, 27 s 1.9 x 105 No. 8 - 17 min, 32 s 2.7 x 101 No. 9 - -c <1 No. 10 - - <1 No. 11 - 26 min, 59 s 8.5 x 100 No. 12 - 29 min, 31 s 7.2 x 100 No. 13 - - <1 Senegalese sole No. 1 - 34 min, 45 s 2.6 x 100 No. 2 - 23 min, 46 s 9.4 x 100 No. 3 - 44 min, 26 s 1.9 x 100 84 Finally, an analysis was performed to determine if the modified Hot-SHOT DNA extraction protocol chosen as an easy field protocol could negatively affect or inhibit the isothermal polymerase. The LAMP assay detected LCDV genome in DNA extracted using the Hot-SHOT method with a relatively high fluorescence signal (Fig. 15) and a shorter Tp value (7 min, 43 s vs 12 min, 41 s) compared to DNA extracted using the commercial kit from the same sample. Figure 15. Comparison of DNA extraction method on LCDV detection by LAMP. DNA extracted from LC-diseased gilthead seabream by a modified Hot-SHOT extraction method and a commercial extraction kit. Fluorescence was normalized to the background for 0-180 s. Positive control: DNA from LCDV SA25 combined with DNA from the modified Hot-SHOT extraction method. Negative control: DNA from SAF-1 cells. 85 CHAPTER 1 3. Evaluation of an integrated cell culture RT-PCR assay to detect and quantify infectious lymphocystis disease virus 3.1. INTRODUCTION Viral disease prevention requires the use of rapid and sensitive diagnostic tools to avoid the spread of viruses in fish farms. In the case of LCDV, PCR-based methods have proved to be adequate for viral detection and quantification in several samples, including carrier fish and live food (Cano et al., 2007; Ciulli et al., 2015; section 1). However, these molecular methods do not provide any information on infectivity, which it is essential to establish the actual risk of viral transmission. The standard method for diagnosis of fish viruses is based on virus isolation in cell culture and further confirmation by serological or molecular techniques (OIE, 2015). This procedure demonstrated virus infectivity and can also be used to determine viral infectious titres. However, LCDV is not easily propagated in cell culture, and virus isolation usually requires homologous cell lines (Iwamoto et al., 2002). LCDV isolated from gilthead seabream can be cultured in the SAF-1 cell line, but often does not provoke clear and consistent cytopathic effects (CPE). This is particularly true when analysing samples with low viral loads, such as those collected from subclinically infected animals (Cano et al., 2006). In addition, in these samples, CPE expression requires at least 14 d or even a blind passage step (Cano et al., 2007). The integrated cell culture (ICC)-PCR method has been applied for the detection of human viruses in environmental samples and has proved to be faster and more sensitive than methods based on CPE expression, minimizing the occurrence of false negatives when the viral load is low or with viruses that do not produce apparent CPE (Reynolds et al., 2001; Lee et al., 2005; Dong et al., 2010; Li et al., 2010). However, PCR sensitivity may lead to the detection of nucleic acids from inactivated viruses present in the samples. In the case of DNA viruses such as LCDV, the detection of viral mRNA by reverse transcription (RT)- 92 In viral stocks obtained directly from diseased fish, the infectious titres estimated by the ICC-RT-PCR assay were nearly one order of magnitude higher than those obtained by the TCID50 method. Furthermore, after a first passage on SAF-1 cells, viral titre dropped below the detection limit of the CPE assay, but it was still possible to titrate those viral stocks using ICC-RT-PCR (Table 11). Finally, the detection and quantification of infectious LCDV was possible in all the gilthead seabream carriers analysed, although no CPE could be observed in cell cultures inoculated in parallel with those homogenates and maintained up to 14 dpi. Infectious titres estimated by the ICC-RT-PCR assay were 2.2-2.7 log10 lower than viral loads obtained by qPCR (Table 12). Table 12. LCDV quantification in samples from asymptomatic gilthead seabream juveniles using ICC-RT-PCR and qPCR assays. Fish samplea ICC-RT-PCR assayb qPCR assayc 1 1.8 x 101 3.1 x 103 2 3.7 x 101 2.0 x 104 3 1.8 x 101 8.6 x 103 4 3.7 x 101 1.9 x 104 5 3.7 x 101 1.5 x 104 a Samples consisted of caudal fin homogenates. b Viral infectious titres expressed as MPNIU/ml. c Viral loads expressed as copies of viral DNA/ml. CHAPTER 2: PATHOGENESIS AND TRANSMISSION OF LYMPHOCYSTIS DISEASE VIRUS ! ! 95 CHAPTER 2 1. Target organs for lymphocystis disease virus replication in gilthead seabream 1.1. INTRODUCTION LCD is a self-limiting disease characterized by the hypertrophy of fibroblastic cells in the connective tissue of fish (Samalecos, 1986). These hypertrophied cells, named lymphocysts or lymphocystis cells, are usually observed on the skin and fins, although they have also been described in several internal organs (such as the stomach, spleen, liver, kidney and heart) (Rusell, 1974; Dukes & Lawler, 1975; Howse et al., 1977; Wolf, 1988; Sindermann, 1990; Colorni & Diamant, 1995). In gilthead seabream, LCD-associated lesions have been described only in the fish skin and fins, and usually disappear after 20-45 days depending on water temperature (Paperna et al., 1982; Gonzalez de Canales et al., 1996; Kvitt et al., 2008). Data on lymphocystis pathogenesis are very scarce and generally limited to histopathological studies of skin lesions (Gonzalez de Canales et al., 1996; Sheng & Zhan, 2004; Sheng et al., 2007b). However, several studies have shown that viral antigens can be detected in a number of organs of infected fish, not only in lymphocystis lesions (Xing et al., 2006; Sheng et al., 2007b; Cano et al., 2009a). In gilthead seabream, DNA-DNA hybridization and immunohistochemistry were used to detect LCDV in diseased and recovered juveniles. Viral genomes and antigens were detected in the different organs analysed, including the caudal fin, gills, intestine, liver, spleen and kidney, suggesting that LCDV establishes a systemic and persistent infection in this fish species (Cano et al., 2009a). In addition, LCDV is frequently detected by PCR-based methods in apparently healthy seabream (Cano et al., 2007; Ciulli et al., 2015; Chapter 1, section 1), which indicates that they may be subclinically infected. However, further studies are necessary to confirm these results and to establish if these infections are productive. Thus, the objective of the present study was to determine the target organs and cells that support LCDV replication in gilthead seabream juveniles, ! 96 both LC-diseased and subclinically infected. In addition, a histopathological study of LCD was also conducted. 1.2. MATERIALS & METHODS 1.2.1. Fish samples Gilthead seabream specimens were obtained from two fish farms located in southwestern Spain. In the first farm, fish without signs of LCD (6-10 g in weight) were collected, and constituted the group named “asymptomatic”. This farm had no record of an LCD outbreak in more than 15 years. In the second farm, diseased individuals (6-10 g) showing typical external signs of LCD were collected, and two months after disease signs disappeared in the fish population, another group of fish (15-20 g) was sampled. These fish constituted the “diseased” and the “recovered” groups, respectively. Fish used in this study were treated according to the Spanish directive (RD 53/2013, BOE no. 34), and were euthanized by anaesthetic overdose (Chapter 1, section 1.2.1). Samples of the caudal fin, intestine, liver, spleen, kidney and brain of nine individuals from each experimental group were aseptically collected and individually processed for subsequent homogenization and nucleic acid extraction. In addition, the same organs were collected from three fish in each group for in situ hybridization (ISH) and histological examination. Fish samples were fixed in 4% paraformaldehyde (Sigma-Aldrich) in DEPC-treated PBS (pH 7.2) for 24 h at 4 ºC, and embedded in paraffin using standard histological procedures. Fixed caudal fin samples were decalcified with a solution containing 10% EDTA (Sigma-Aldrich) and 4% paraformaldehyde in DEPC-treated water at pH 7.0 for 10-15 days at 4 °C. Tissue sections (5-7 µm) were mounted on TESPA (3triethoxysilylpropylamine)-treated slides. 97 CHAPTER 2 1.2.2. DNA and RNA extraction and cDNA synthesis Total DNA and RNA were extracted using the E.Z.N.A. Tissue DNA Kit and the E.Z.N.A. Total RNA Kit I (Omega Bio-tek), respectively, following the manufacturer’s instructions. Total RNA was treated with RNase-free DNase I (Roche) for 30 min at 37 ºC. RNA purity and quantity was determined using NanoDrop 1000 (Thermo Scientific). After DNase treatment, total RNA was used in the qPCR reaction in order to control for the absence of viral genomic DNA. Firststrand DNA synthesis was carried out with 1 µg of total RNA and random hexamer primers using the Transcriptor First Strand cDNA Synthesis Kit (Roche). DNA and cDNA were stored at -20 ºC until used as template for qPCR. 1.2.3. LCDV DNA quantification and gene expression The qPCR protocol described in Chapter 1 (section 1.2.) was used to quantify the amount of viral DNA in the samples. The number of copies of LCDV DNA was calculated by interpolation in a standard curve, and viral load expressed as viral DNA copies per milligram of tissue. Relative quantification of MCP gene expression was carried out by RTqPCR, following the protocol mentioned above but using 20-µl final volume reactions and 2 µl of cDNA (at a 1/30 dilution). Normalized relative MCP expression levels were calculated for each sample applying the formula: F = log10 [(E + 1)40-Ct/N] (Segarra et al., 2014), where E is the amplification efficiency of the qPCR, Ct (threshold cycle) corresponds to the PCR cycle number, N is the maximal number of viral DNA copies/mg of tissue detected minus the number of viral DNA copies/mg of tissue determined by absolute qPCR for the sample, and Ct of 40 arbitrarily corresponds to “no Ct” by qPCR. Results obtained for viral DNA quantification and relative gene expression were statistically analysed using a Mann-Whitney U-test followed by a HolmBonferroni correction for multiple comparisons. ! 98 1.2.4. RNA in situ hybridization and histopathology! DIG-labelled RNA probes were synthesized by in vitro transcription with the DIG RNA Labelling Kit (Roche) using a 150-bp fragment of the viral MCP gene (nucleotide positions 173 to 322 of the LCDV SA9 MCP gene, GenBank accession no. GU320728) cloned into the pCRII Dual Promoter vector (Invitrogen) as the template. The RNA probes were produced from 1 µg of linearized plasmid using T7 (antisense) or SP6 (sense) polymerases. Deparaffinised and rehydrated tissue sections were permeabilized for 30 min with 10 µg/ml proteinase K in a buffer containing Tris-HCl 0.05 M pH 7.6 (40%, v/v) and CaCl2 1 M (4%, v/v) in DEPC-treated water at 37 °C. Sections were prehybridized with formamide (50%, v/v), 20x SSC (25%, v/v), torula yeast RNA (50 mg/ml), heparin sodium salt (5 mg/ml), Denhardt’s solution (2%, v/v), CHAPS (2%, w/v) and Tween 20 (0.5%, v/v) in DEPC-treated water for 3 h in a 2x SSC saturated atmosphere at 55 °C. Sense and antisense probes were denatured at 85 ºC for 5 min, and hybridization was performed overnight at 60 °C. After hybridization, sections were washed as previously described (Ortiz-Delgado et al., 2006) and treated with 1% blocking reagent (Roche) in maleic acid buffer (0.1 M maleic acid, 0.15 M NaCl, pH 7.5) for 1 h at room temperature. Then, the slides were incubated with anti-digoxigenin-AP (Roche) overnight at 4 ºC, and the hybridization signals were detected using NBT/BCIP (Roche) according to the manufacturer's instructions. All reagents were supplied by Sigma-Aldrich unless otherwise stated. Finally, sections were dehydrated and mounted in Eukitt® quick-hardening mounting medium (Sigma-Aldrich). In parallel, tissue sections were stained with haematoxylin-eosin (HE) and haematoxylin-V.O.F. (Sarasquete & Gutierrez, 2005) for histological examination.! ! ! 99 CHAPTER 2 1.3. RESULTS! 1.3.1. Viral load and gene expression LCDV was detected by qPCR in all the samples analysed. Viral load in different organs from the three experimental groups are shown in Fig. 18. In diseased fish the highest viral loads were detected in the caudal fin (3.5 ± 2.4 x 105 copies of viral DNA/mg of tissue), followed by the kidney (1.2 ± 0.2 x 104 copies of viral DNA/mg) and brain (2.4 ± 0.9 x 103 copies of viral DNA/mg). In fish from the asymptomatic and recovered groups, low-titre infections were observed, with estimated viral loads between 0.4 and 27.5 copies of viral DNA per mg. No significant differences (p < 0.05) were observed between the asymptomatic and the recovered groups, except for liver samples. In fish from the asymptomatic group, the number of LCDV DNA copies in the brain was significantly higher than in the caudal fin (p < 0.01). MCP gene expression was analysed as an indicator of viral productive infection. In diseased fish, relative viral gene expression values were similar to viral loads in the different organs analysed (Fig. 19). Thus, the highest relative expression values were detected in the caudal fin, followed by those in the kidney and brain. F values were significantly higher (p < 0.01) in these organs than in the other internal organs analysed. Viral gene expression was observed in all organs collected from fish from the asymptomatic and recovered groups, with relative values significantly lower (p < 0.01) than those obtained in samples from diseased fish. In asymptomatic fish, F values in the caudal fin were significantly higher (p < 0.01) than those in other tested organs, with the exception of the liver. No significant differences (p < 0.01) were observed between both experimental groups except for the caudal fin and brain samples, with F values significantly higher in the asymptomatic and recovered groups, respectively. ! 100 Figure 18. Viral loads in samples from different organs of gilthead seabream determined by qPCR. Experimental groups: diseased, orange; asymptomatic, green; recovered, blue. Different letters indicate significant differences between organs in the same experimental group. *Significant differences between groups. Significant level p < 0.01 (Mann-Whitney U-test, Holm-Bonferroni correction). Error bars represent ± standard deviation (n = 9). Figure 19. Relative MCP gene expression values in samples of gilthead seabream juveniles. Experimental groups: diseased, orange; asymptomatic, green; recovered, blue. Different letters indicate significant differences between organs in the same experimental group. *Significant differences between groups. Significant level p < 0.01 (Mann-Whitney U-test, Holm-Bonferroni correction). Error bars represent ± standard deviation (n = 9). ! 101 CHAPTER 2 1.3.2. RNA in situ hybridization Viral transcripts were detected by ISH in all organs from the diseased fish, whereas no signal was observed in the sections from fish belonging to the asymptomatic and recovered groups (results not shown). No labelling was observed in the negative controls using sense probe for ISH. In sections of the caudal fin, the hybridization signal was strong, and labelling was observed as cytoplasmic inclusions in the lymphocysts (Fig. 20A) and in some cells in the surrounding connective tissue. In liver sections, numerous hepatocytes showed marked labelling in their cytoplasm (Fig. 20B). The hybridization signal was widely distributed in the splenic pulp, although in some areas the signal appeared concentrated around melanomacrophage centres (MMC) and ellipsoids (Fig. 20C). In the kidney, the hybridization signal was mostly confined to the haematopoietic tissue (Fig. 20D). In sections from the brain, ISH labelling was observed in the cytoplasm of cells in the granular layer (Fig. 20E). Finally, tissue sections from the intestine were also ISH-positives, but the tissue was so damaged by the ISH protocol that it was not possible to distinguish the localization of the labelled cells. ! 108 2. Transmission of lymphocystis disease virus to gilthead seabream larvae and fingerlings 2.1. INTRODUCTION Transmission of LCDV has not yet been completely elucidated but direct contact is the route classically accepted for the spreading of LCDV infection, the skin and gills being the main portals of entry (Wolf, 1988; Bowser et al., 1999). Viral transmission by cohabitation or by contaminated water has been demonstrated for several fish viruses, such as iridoviruses (Drennan et al., 2006; Kurobe et al., 2011), rhabdoviruses (Traxler et al., 1993; Muroga et al., 2004), aquabirnaviruses (Rodriguez Saint-Jean et al., 2003), betanodaviruses (Johansen et al., 2003) and orthomyxoviruses (Totland et al., 1996). Some studies suggest the transmission of LCDV via the alimentary canal (Sheng et al., 2007b; Cano et al., 2009a), similar to that established for other waterborne viral fish diseases (Wolf, 1988; Peducasse et al., 1999; Woodland et al., 2002; Kurobe et al., 2011). There are no previous data supporting vertical transmission of LCDV (i.e. from broodstock to progeny via the eggs) as has been described for other fish viral pathogens, such as infectious pancreatic necrosis virus (IPNV), infectious haematopoietic necrosis virus (IHNV), viral nervous necrosis virus (VNNV), infectious salmon anaemia virus (ISAV) or white sturgeon iridovirus (WSIV) (Mulcahy & Pascho, 1985; Bootland et al., 1991; MacAllister et al., 1993; Mushiake et al., 1994; Georgiadis et al., 2001; Breuil et al., 2002; Nylund et al., 2007). With the exception of IPNV, which is transmitted intra-ovum, fish viruses are generally transmitted on the egg surface (Brock & Bullis, 2001). Consequently, disinfection of the egg surface can be used as a control strategy to prevent viral transmission, reducing disease outbreaks caused by these infectious agents (Grotmol & Totland, 2000; Buchan et al., 2006). Routine virological analyses carried out in gilthead seabream hatcheries from farms suffering recurrent LCD outbreaks in juveniles have demonstrated that larvae, post-larvae and fingerlings were usually infected by LCDV, and also ! 109 CHAPTER 2 rotifers and Artemia naupliar/metanaupliar stages used as live food (Chapter 1, section 1; unpublished results). Based on these results, the aim of the present study was to establish the possible source of LCDV infection in gilthead seabream larvae and fingerlings in an effort to shed light on LCDV transmission routes. In addition, iodine-based disinfection has been tested on eggs as a preventive treatment for LCDV infection. 2.2. MATERIALS & METHODS 2.2.1. Transmission studies with gilthead seabream larvae Gilthead seabream eggs were obtained by natural spawning from a broodstock held in a farm with previous reports of LCD. The broodstock were composed of 42 fish, with a sex ratio of two males per female (Moretti et al., 1999). Before stocking in the spawning tank, heparinised blood samples were collected from the fish (Chapter 1, section 1.2.1), sent to the laboratory within 24 h and analysed for LCDV detection by PCR-hybridization (see below). Fertilized eggs were collected from the spawning tank via an overflow egg collector and placed in a 20-l bucket with clean, running, seawater. Eggs were divided into two batches (50,000 eggs each). Three samples of eggs (500 µl) from each batch were randomly collected and analysed for LCDV detection using both classical virological techniques and PCR-hybridization. Afterwards, one egg batch was stocked at a density of approximately 550 eggs per litre in a 120-l cylindro-conical tank at 19 ± 0.5 °C and 34 g/l salinity. In the other batch, eggs were disinfected by dipping in an active iodine (50 mg/l) solution for 10 min (Moretti et al., 1999). Briefly, eggs were taken out of the water using a filter and quickly placed in a bucket containing 300 ml of aerated seawater supplemented with the disinfectant. After 10 min, eggs were rinsed with clean seawater, sampled for LCDV detection and stocked as described above. Both egg batches were incubated directly into the larval rearing tanks. After hatching, ! 110 rearing tanks operated in a flow through seawater system (Moretti et al., 1999). Temperature and salinity were kept as indicated above. From the fourth day post-hatch (dph), the larvae were fed with rotifers, Brachionus plicatilis (Bs and S-1 strains), at a concentration of 10 rotifers/ml. Rotifers were cultured in the hatchery, harvested as described elsewhere (Moretti et al., 1999) and tested for LCDV before being fed to the fish. Three samples of harvested and rinsed rotifers (100 µl each), collected before feeding fish at 4, 6 and 8 dph, were analysed. 2.2.2. Brine shrimp LCDV inoculation Brine shrimp Artemia cysts (AF, INVE) were decapsulated using a mixture of sodium hypochlorite (0.5 g active chlorine per gram of cysts) and sodium hydroxide (0.15 g/g cysts), following the procedure specified by Moretti et al. (1999). Residual hypochlorite was neutralized with sodium thiosulfate (0.1%, w/v, 5 min). Decapsulated cysts were hatched in sterile seawater (33 g/l salinity) at 26 ºC (Lavens & Sorgeloos, 1996). After a 48-h incubation, hatched instar II nauplii were separated from the unhatched and empty cysts and transferred to aquaria with fresh sterile seawater. Nauplii were reared to naupliar stage (4 d post hatching) at 26 ºC, with continuous aeration and a 24-h photoperiod, and fed with Mikrozell (Hobby). Nauplii were inoculated by immersion with LCDV isolate SA25 (102 TCID50/ml) (Cano et al., 2009b), whereas animals inoculated with Leibovitz’s L-15 medium (Gibco) were used as control group. After 24 h, nauplii were filtered through a synthetic net, washed and transferred to aquaria with fresh sterile seawater, and maintained as specified above. Eight days later, metanauplii from both experimental groups were filtered, washed and used to feed gilthead seabream fingerlings. The presence of LCDV on brine shrimp (two samples of 100 mg per experimental group) was determined by PCR-hybridization. ! 111 CHAPTER 2 2.2.3. Gilthead seabream oral challenge! Gilthead seabream fingerlings (0.5-1 g) were obtained from a research marine aquaculture facility with no record of LCD. Five fish were randomly collected and analysed by PCR-hybridization to ensure that they were LCVDfree. Fish were divided into two groups (50 individuals per group) and stocked at a density of 2 g/l in aquaria with filtered seawater. Fish were maintained at 20-22 ºC and a 12-h photoperiod, and fed an artificial diet (500-800 µm) at a feeding rate of approximately 5% fish body weight per day. For oral challenge, fingerlings were fed once with Artemia metanauplii inoculated with LCDV or L-15 medium (challenged and control groups, respectively) at a concentration of 0.2 g/l. The following day, an artificial diet was resumed, and fish were maintained at the conditions indicated above for 30 d. 2.2.4. LCDV detection by PCR-hybridization DNA was extracted from fertilized egg samples, larvae 2 and 10 dph (two pools of 30 animals), fingerling samples, rotifer samples and artemia samples using DNAzol (Invitrogen), according to the manufacturer’s instructions. Larvae and fingerlings were euthanized by anaesthetic overdose as previously specified (section 1.2.1). In the case of gilthead seabream fingerlings, samples consisted of the caudal part of the body (approximately the posterior one third of the fish body). In addition, DNA was extracted from heparinised blood samples collected from the broodstock using the QIAamp DNA blood kit (Qiagen). A specific PCR combined with dot-blot hybridization was used to detect the LCDV genome as described in Chapter 1 (section 1.2.2). 2.2.5. Virological analyses In parallel to viral DNA detection, eggs, pooled larvae (2 and 10 dph) and rotifer samples were homogenized (20%, w/v) in Leibovitz’s L-15 medium supplemented with 2% L-glutamine, 1% penicillin-streptomycin and 2% FBS. 112 Homogenates were centrifuged at 5,000 x g for 10 min at 4 °C, filtered (0.45-µm pore-size filter) and used to inoculate SAF-1 cells. Cell cultures were maintained at 20 °C until the appearance of CPE (up to 15 dpi). 2.2.6. LCDV DNA quantification and gene expression Seven gilthead seabream fingerlings from both the oral challenged and control groups (i.e. fed on metanauplii inoculated with LCDV and L-15 medium, respectively) were randomly sampled at 7, 12 and 24 dpi. Samples, obtained as specified in section 2.2.4, were homogenized (10%, w/v), and DNA and RNA were extracted from 200 µl of each homogenate using the Illustra triplePrep Kit (GE Healthcare), following the manufacturer’s instructions. DNase treatment and cDNA synthesis were carried out as specified in section 1.2.2. Viral DNA quantification and MCP gene expression were achieved by qPCR and RT-qPCR, respectively, following the protocols described in section 1.2.3. 2.2.7. Whole-mount in situ hybridization Larvae were processed for whole-mount ISH following the protocol described by Cano et al. (2009b). Briefly, 10 animals were randomly sampled at 2 and 3 dph, and fixed with 4% neutral buffered formalin (NBF; Merck) at 4 °C overnight. After washing with PBS supplemented with Tween-80, the animals were rehydrated and permeated by sonication and proteinase K treatment. Hybridization was carried out at 42 °C overnight using the DIG-labelled probe mentioned in Chapter 1 (section 1.2.2). The endogenous phosphatase activity was blocked with 1 mM levamisole (Sigma-Aldrich), and the hybridization signal was detected using NBT/BCIP (Roche) for 1–2 h at room temperature. Animals were mounted using Aquatex (Merck). ! 113 CHAPTER 2 2.2.8. Immunohistochemistry! NBF-fixed larvae sampled from 4 to 9 dph (10 animals per sample point) were dehydrated and embedded into paraffin following standard histological protocols, and sections (6 µm) were mounted on silane-treated slides (SigmaAldrich). Deparaffinised and rehydrated sections were analysed by immunohistochemistry (IHC) according to previously reported procedures (Cano et al., 2009a;b). After Triton X-100 permeabilization, sections were treated with levamisole or H2O2 (Merck) for endogenous phosphatase or peroxidase blocking, respectively. An anti-LCDV serum immunoabsorbed onto a monolayer of SAF-1 cells was used as primary antibody (Garcia-Rosado et al., 2002), and preimmunized rabbit serum as a negative control. Anti-rabbit IgG conjugated with alkaline phosphatase or with peroxidase (Sigma-Aldrich) was employed as secondary antibody. Alkaline phosphatase activity was developed with NBT/BCIP, whilst a solution of diaminobenzidine (Sigma-Aldrich) and H2O2 was used for peroxidase detection. Tissue sections were mounted with a coverslip using Entellan (Merck) or Aquatex, respectively. In parallel, some sections were stained with HE for histological studies. 2.3. RESULTS 2.3.1. LCDV transmission to gilthead seabream larvae LCDV genome was detected by PCR-hybridization in the blood from 7 out of 40 gilthead seabream specimens analysed. Using the same methodology, eggs spawned by these animals were also found to be LCDV positive, as well as 2-day-old larvae hatched from them (in both cases, 100% of the analysed samples were positive). Conversely, all the analyses performed to detect viral DNA from iodine-treated eggs, or from their larvae, generated negative results. Larvae analysed at 10 dph (from both experimental groups, i.e. non-disinfected 114 and disinfected eggs) showed positive results for LCDV detection by using PCRhybridization. The inoculation of SAF-1 cells with homogenates from PCR-hybridization positive eggs and larvae resulted in the appearance of LCDV-typical CPE (i.e. rounded and enlarged cells showing cytoplasmic inclusions) on the SAF-1 cell line. No CPE were observed on cells inoculated with homogenates from disinfected eggs or from their newly hatched larvae. Figure 24. Lymphocystis disease virus (LCDV) detection in gilthead seabream larvae hatched from LCDV-positive (A-D) and negative (i.e. iodine-disinfected) (E-H) eggs, demonstrated by in situ hybridization (ISH) and immunohistochemistry. Hybridization signal is observed microscopically as dark blue staining (A, E), whereas immunolabelling appears either as dark blue (B-D) or brown (F-H) in colour, depending on secondary antibody used (conjugated with alkaline phosphatase or peroxidase, respectively). (A) Viral genome detection by whole-mount ISH in a 2-day-old larva. The hybridization signal is located in the epidermis. (B) Immunopositive primordial fin of a larva 4 dph. (C) 4-dayold larva showing a weak immunostaining in the digestive border. (D,H) LCDV antigens in the digestive tract of 6-day-old larvae. (E) ISH-negative epidermis in a larva 3 dph. (F,G) Immunonegative epidermis and digestive tract, respectively, of a larva analysed at 5 dph. Scale bars: 20 µm. ! 115 CHAPTER 2 116 The whole-mount ISH technique showed the presence of viral DNA in the epidermis of larvae (2 and 3 dph) hatched from LCDV-positive eggs (Fig. 24A); 92.9% of the larvae examined were positive. In addition, IHC analyses also showed the presence of viral antigens in these animals. In all the larvae analysed at 4 dph, immunostaining was observed in the epidermis of body skin and primordial fins (Fig. 24B). In some sections (33.3% of the observed larvae), the digestive borders appeared weakly positive (Fig. 24C). No immunolabelling was observed when preimmune serum was used as primary antibody. Twoto fiveday-old larvae hatched from disinfected eggs did not exhibit LCDV specific signal when they were analysed either by whole-mount ISH or IHC (Fig. 24E–G). Four days after hatching, larvae deriving from both groups of eggs (disinfected and non-disinfected) were fed rotifers daily. The rotifer cultures were LCDV positive, as demonstrated by PCR-hybridization and CPE development on SAF-1 cells. In larvae hatched from LCDV-positive eggs, an increase in immunolabelling in the skin, digestive tract and digestive content was observed after feeding began (Fig. 24D). More than 97% of these animals showed viral antigens in the skin, and 88.9% of them also showed immunopositive digestive tracts. In addition, more than 92% of larvae from disinfected eggs analysed by IHC showed LCDV antigens in the epidermis and digestive tract from 6 dph (Fig. 24H). Microscopic examination of IHC processed larvae was systematically focused on the skin and digestive tract, although in some sections of larvae 6 dph onwards (from both experimental groups), immunolabelling was observed in brain and liver (no data on prevalence of viral antigens were recorded). Microscopic examination of HE-stained tissue sections of larvae did not reveal histological alterations in LCDV-positive animals compared with negative ones. ! 117 CHAPTER 2 ! 2.3.2. LCDV transmission to gilthead seabream fingerlings Artemia metanauplii were effectively contaminated by LCDV, as demonstrated by PCR-hybridization, whereas metanauplii in the control group remained LCDV-negative. In fingerlings fed on the LCDV-positive metanauplii (challenged group) LCDV was detected by qPCR in all fish and at all time points analysed. At 7 dpi, the estimated viral load ranged between 10.6 and 26.8 copies of viral DNA per mg of tissue. Five days later, viral loads were significantly higher (p < 0.01), and they remained at similar values at 24 dpi (Fig. 25A). No LCD symptoms were observed in these fish at the end of the experiment (30 dpi). MCP gene expression was also detected in all fish analysed, with the highest F values observed at 12 dpi (Fig. 25B). Neither LCDV genomes nor mRNA were detected in fish from the control group (i.e. fed on metanauplii inoculated with L-15 medium). 124 Figure 27. Temporal evolution of viral loads (A) and relative MCP gene expression values (B) in Artemia metanauplii (8 d post-hatching) inoculated with LCDV ATCC VR-342. ! DISCUSSION ! !! ! 127 DISCUSSION 1. DIAGNOSTIC METHODS FOR LYMPHOCYSTIS DISEASE VIRUS LCD outbreaks are frequently observed in the Mediterranean gilthead seabream aquaculture (Borrego et al., 2001). Although it is usually described as a self-limiting disease, there are several reports on mortalities up to 45 % in juvenile fish, which may be related to secondary bacterial infections or with particularly large growth of lymphocysts, which severely impaired fish breathing or feeding (Colorni & Padros, 2011; Dezfuli et al., 2012; Haddad-Boubaker et al., 2013). As no effective treatments or commercially available vaccines currently exist, LCD prevention in hatcheries must rely on the selection of LCDV-free broodstock, the use of effective decontamination methods to prevent viral transmission from asymptomatic broodstock to larvae, and the supply of virus-free live food (Cano et al., 2009b; Yoshimizu, 2009). During the growing period, the selection of noninfected fish is also advisable, as LCDV-positive juveniles may become symptomatic under stress conditions, such as transport to on-growing facilities. Moreover, little information is available on a number of epidemiological questions concerning LCDV infections, as the number of virus particles required to induce the disease, the number of genome copies present in asymptomatic and diseased fish, and the kinetics of viral replication. Previous studies have demonstrated the applicability of a PCR-based method to detect LCDV in asymptomatic gilthead seabream carriers as well as in Artemia cultures (Cano et al., 2007; Cano et al., 2009b). Nevertheless, this method is relatively time-consuming, as an additional step of blot-hybridization of the PCR products is required to detect LCDV-positive samples. Furthermore, this assay is not quantitative and applicable for routine diagnosis. In the present study, a qPCR assay has been developed and applied to detect and quantify LCDV in different samples. The assay was specific for LCDV, as demonstrated by analysis of the melting curves generated from each sample. Its analytical sensitivity, determined as the smallest copy number of the plasmid standard reliably detected, was 2 copies of DNA per reaction. The qPCR assay also showed a wide linear dynamic range, extending to 6 log10 concentrations of plasmid DNA, and infectious titres from 104 to 1 TCID50. In addition, the precision 128 of the assay was supported by the high correlation coefficients obtained for the standard curves, and the intraand inter-assay variation of Ct values. Using the protocol as described, the qPCR assay allows the detection of the virus at levels as low as 1 copy of viral DNA per mg of fish tissue. This high sensitivity, combined with its wide dynamic range, makes the qPCR assay suitable to detect low viral loads in subclinical LCDV infections, and, at the same time, to quantify variable viral loads in the course of infection. In addition, it could be useful in searching for potential LCDV reservoirs. The application of the qPCR assay to LCDV surveillance in fish farms has shown that monitoring the infection in individual fish, both diseased and subclinically infected, is possible by sampling caudal fin as reported previously (Cano et al., 2007; Kvitt et al., 2008). The prevalence of LCDV infection in the asymptomatic gilthead seabream populations analysed varied from 30 to 100 %, even in one farm with no previous LCD records. In these fish, estimated viral load in caudal fin was two to five orders of magnitude lower than in diseased fish. Thus viral load seems to correlate with disease manifestation. The low viral loads detected in subclinical infections may represent a status associated with viral replication as it has been established for juvenile fish during the studies on viral pathogenesis carried out in the present Thesis. In addition, the qPCR assay developed could be a valuable tool to study the correlation between viral multiplication and the onset of symptoms in experimental LCDV infections. Palmer et al. (2012) developed a real-time PCR method using fluorogenic primers, specific for LCDV genotype IX sequences, which proved to be reliable in the!detection of subclinically infected yellow perch, although its sensitivity was 5 x 102 copies of DNA per mg/l. LCDV quantification by qPCR has also been carried out in a reduced number of samples from diseased and recovered gilthead seabream (Ciulli et al., 2015). The analytical sensitivity of the former assay was 5.2 copies of DNA per reaction that is more than twice the value reported in the present study. Furthermore, although the authors reported some quantitative data, viral loads were not expressed in terms of viral DNA copies per amount of tissue, which prevents further comparisons (Palmer et al., 2012; Ciulli et al., 2015). !! ! 129 DISCUSSION Carrier fish were also identified in the broodstock from a farm with LCD records by analysing caudal fin samples by qPCR. The assay was applied in parallel to blood samples, and although LCDV could be detected, estimated viral load, and also clinical sensitivity, was lower than that obtained in caudal fin analysis. In this farm, the q-PCR assay allowed the quantitative detection of LCDV in all samples collected in the hatchery, including fertilized eggs, larvae and fingerlings, and also rotifer cultures, and Artemia metanauplii and cysts used for larval rearing. In these samples, as well as in caudal fin samples from asymptomatic juvenile fish, the qPCR assay showed the same clinical sensitivity than the PCR-hybridization protocol described by Cano et al. (2007), but is completed in 130 min, including melting curve generation, which considerably reduces the time required for LCDV diagnosis. In addition, the results of this study support the existence of multiple reservoirs of LCDV in the farm facilities, and the importance of proper application of effective disinfection treatments, as those recommended by the FAO (Moretti et al., 1999), to!avoid viral transmission through fish eggs or live food. The qPCR assay developed in this study has proved to be a rapid, sensitive and reliable method for LCDV diagnosis in surveillance studies. Nevertheless, this technique has limitations as a result of the expensive reagents and equipment required that could prevent its routine use in fish farms. The LAMP method, due to its sensitivity, specificity, efficiency, rapidity and simplicity is a promising molecular technique for identifying some infectious diseases, and can be developed in the laboratory for subsequent deployment in the field, thus making it a suitable method for rapid diagnosis (Boonham et al., 2014). LAMP assays have been applied to detect several fish viruses of high economic relevance (Savan et al., 2005), with a sensitivity 10to 100-times higher compared to conventional PCR (Caipang et al., 2004; Mao et al., 2008; Zhang et al., 2009; Min et al., 2013), and comparable to that of qPCR assays (Li et al., 2010; He et al., 2013). In this study, a LAMP assay was designed for the detection of LCDV genotype VII. The assay has proved to be specific for this LCDV genotype, with no cross-reactions observed with the other iridoviruses tested. The analytical sensitivity of the LAMP assay, determined using ten-fold serial dilutions of the plasmid standard, was 10 copies of viral DNA, which is similar to that obtained by !! ! 130 qPCR, and significantly higher compared to conventional PCR (Cano et al., 2007). When used on fish samples, the LAMP assay showed the same clinical sensitivity as qPCR, allowing the detection of LCDV in subclinically infected gilthead seabream and Senegalese sole. In addition, the apparatus used to perform the LAMP reactions provides information on the annealing/melting temperature of the products, which confirms the specific amplification of LCDV genotype VII in the samples. The LAMP assay is completed in less than 60 min (the duration of the assay can be precisely controlled as the amplification is monitored in real-time), even in samples from asymptomatic carriers where the viral load can be extremely low, which reduces the time required for LCDV diagnosis in comparison to the qPCR assay. Since the Genie® II apparatus allows isothermal amplification on a low power portable platform, the LAMP protocol developed could be used in field conditions for the diagnosis of LCDV infection. In viral outbreaks occurring on fish farms, it is important to determine the presence of the pathogenic fish virus not only qualitatively, but also quantitatively. The correlation observed between the amount of template DNA, obtained from both recombinant plasmid or fish tissues, and the corresponding Tp values indicates that the LAMP assay might be useful for at least semi-quantifying LCDV without the need of expensive equipment. One critical point of the application of diagnostic molecular methods to field samples is the requirement of high-quality DNA as template. Some authors have pointed out that LAMP is less prone to interference or inhibition by biological substances than PCR (Kaneko et al., 2007; Francois et al., 2011), and its efficiency does not seem to be affected by the presence of non-target genomic DNA in the reaction mixture (Notomi et al., 2000). In the present study, a modified Hot-SHOT protocol has been used for crude DNA extraction, and resulted in an isothermal amplification similar to that obtained using DNA purified with a commercial extraction system. This protocol omits complex DNA purification processing; saves time, labour, and cost in the assay, and is potentially suitable for less experienced operators. Molecular methods such as qPCR above discussed are adequate for viral detection and quantification, but they cannot be used if infectious viruses need !! ! 131 DISCUSSION to be detected and/or quantified, which requires virus isolation on cell culture. LCDV is difficult to propagate in cell culture, and does not produce clear and consistent CPE, especially in samples collected from subclinically infected fish. In the present study, an ICC-RT-PCR assay, followed by dot-blot hybridization of the RT-PCR products, was developed to improve the detection of infectious LCDV. The sensitivity of the ICC-RT-PCR assay was at least 100-fold higher than viral diagnosis obtained by CPE development. This could be partially due to the hybridization step that certainly increased the sensitivity of RT-PCR compared to agarose gel detection, as reported by other authors (Phromjai et al., 2002; Cano et al., 2007). In addition, the ICC RT-PCR assay could be completed in 7 d, including the dot-blot hybridization step, which considerably reduces the time required for LCDV titration when compared to the TCID50 method (between 14 and 21 d) (Walker & Hill, 1980; Garcia-Rosado et al., 1999; Cano et al., 2007). The sensitivity of the developed assay enabled the quantification of infectious LCDV in samples with low viral loads, including those from asymptomatic carrier fish, in which no CPE was recorded after a 14-d incubation period. The results obtained for viral stocks after a first passage on SAF-1 cells indicate that gilthead seabream LCDV isolates do not replicate efficiently in vitro, as had been suggested previously (Walker & Hill, 1980; Cuilli et al., 2015). In gilthead seabream carriers, infectious titres were more than 2.2 log10 lower than viral loads obtained by qPCR. Thus, although qPCR is a useful technique for routine LCDV diagnosis, the actual amount of infectious virus may be overestimated when used for viral quantification, at least in subclinically infected fish. This low viral productivity in fish tissues may be related to partial replication or defective virus assembly (Walker & Hill, 1980; Peters & Schmidt, 1995). 132 2. PATHOGENESIS OF LCDV IN GILTHEAD SEABREAM The pathognomonic signs of LCD are the appearance of small pearl-like nodules on the skin and fins. The nodules are usually grouped in clusters, are papillomatous in appearance and can cover the entire body surface of the fish (Wolf, 1988). These nodules consist of LCDV-infected hypertrophied dermal fibroblasts (up to 1 mm in diameter), named lymphocysts or lymphocystis cells (Paperna et al., 1982; 1987; Bowden et al., 1995). LCDV is considered a dermatropic virus (Wolf, 1988); however, in some fish species, lymphocysts have also been observed in the mesenteries, peritoneum, and several internal organs, which could indicate that the infection can become systemic under certain conditions (Howse et al., 1977; Sinderman, 1990; Colorni & Diamant, 1995; Smail & Munro, 2001). Moreover, using sensitive immunological and molecular diagnostic methods, LCDV has been detected in different organs of fish without internal lesions (Sun et al., 2003; Xing et al., 2006; Sheng et al., 2007b; Kvitt et al., 2008; Ciulli et al., 2015). These findings suggest a systemic condition for LCD although it has not been shown if viruses detected in different organs proceed from productive infections or are actually the result of an underlying viraemia. In the present study, lymphocystis cells were exclusively observed in the dermis of the skin and fins of diseased juvenile gilthead seabream. Nevertheless, viral genomes were detected by qPCR in all of the organs analysed. Viral gene expression was also detected in all the samples, with the highest relative expression values recorded in the caudal fin, followed by those in the kidney and brain. Accordingly, the highest viral loads were detected in the fins, and the amount of viral genomes in the kidney and brain were significantly (p < 0.01) higher than in other internal organs analysed. These results support that LCDV establishes a systemic infection in gilthead seabream, similar to infections reported for other iridoviruses, such as ranaviruses and megalocytiviruses (Weber et al., 2009; Whittington et al., 2010). Recent studies carried out in Japanese flounder and turbot have shown that LCDV genome copy numbers increased in all organs analysed during the !! ! 133 DISCUSSION course of experimental infections, and that the extensive range of viral target tissues is, at least partially, the result of the wide distribution of the LCDV-C receptor (Sheng et al., 2015; Wu et al., 2015). Whether this receptor, a membrane protein of 27.8 kDa first identify in FG cells (Wang et al., 2011a; Sheng et al., 2012b), is present in gilthead seabream cells, and also a receptor for LCDV-Sa attachment, needs to be investigated. Viral MCP transcripts were detected by ISH in order to identify susceptible cells supporting LCDV productive infection. As expected, LCDV expression was observed on lymphocysts located on the caudal fin but also in some cells in the surrounding connective tissue. Viral transcripts were also detected in hepatocytes, and in cells of the splenic pulp, the kidney interstitium, and the brain granular layer. This distribution of viral mRNA is similar to results of previous work that detected viral genomes and antigens in several organs of juvenile gilthead seabream (Cano et al., 2009a). In the present study, it was not possible to determine which cell type contained viral transcripts in the intestine. Nevertheless, Cano et al. (2009a) detected LCDV-positive cells in the connective tissue of the lamina propia. Furthermore, other authors also described the detection of LCDV genomes and/or antigens in the gill lamella of LC-diseased Japanese flounder, black rockfish, and gilthead seabream (Xing et al., 2006; Sheng et al., 2007; Cano et al., 2009a). Together, these results support a broad range tissue tropism for LCDV, similar to that established for megalocytiviruses, which were described to be mesotheliotropic (Gibson-Kueh et al., 2003; MarcosLopez et al., 2011). On the basis of the results obtained, the permissive cells for LCDV replication seem to be fibroblasts, hepatocytes, and cells of the mononuclear phagocyte system, as previously suggested (Garcia-Rosado et al., 2002; Cano et al., 2009a). The LCDV-C receptor has been detected in the membrane of a small portion of turbot peripheral leucocytes, which could indicate that they are susceptible to LCDV infection, resulting in LCDV spreading to different host tissues via the bloodstream (Sheng et al., 2015). In the gilthead seabream brain, viral transcripts were detected in cells of the granular layer, which suggests that microglial cells or infiltrating macrophages may be susceptible to LCDV, although neurons cannot be ruled out as a susceptible cell type. Further