Tion and Characterization of Novel Serine Proteases in the Bark Beetle Tomicus Yunnanensis

D. Fco J. Sánchez García

T E S I S

D UNIVERSIDAD DE MURCIA O C FACULTAD DE BIOLOGÍA T O R A L Phylogeography, genomics and biosemiotics of bark beetles (Coleoptera: Scolytinae)

Filogeografía, genómica y biosemiótica de escarabajos de corteza (Coleoptera: Scolytinae)

2015 D. Francisco Javier Sánchez García 2015

D. Fco J. Sánchez García

T E S I S

D UNIVERSIDAD DE MURCIA O C FACULTAD DE BIOLOGÍA T O R A L Phylogeography, genomics and biosemiotics of bark beetles (Coleoptera: Scolytinae)

Filogeografía, genómica y biosemiótica de escarabajos de corteza (Coleoptera: Scolytinae)

2015 D. Francisco Javier Sánchez García 2015

Supervised by:

José Galián Albaladejo Diego Gallego Cambronero Vilmar Machado

1 Table of contents

1 RESUMEN GENERAL...... 1 1.1 INTRODUCIÓN...... 2 1.2 OBJECTIVOS...... 4 1.3 METODOLOGÍA...... 5 1.4 RESULTADOS...... 6 2 INTRODUCTION...... 10 2.1 TAXONOMY OF THE GENUS TOMICUS...... 11 2.1.1 Tomicus destruens...... 13 2.1.2 Tomicus yunnanensis...... 14 2.2 SERINE PROTEASES...... 15 2.3 NON-CODING RNA TYPES AND FUNCTIONS...... 16 2.4 MICRORNAS...... 17 2.5 NICHE THEORY AND MODELLING...... 18 2.5.1 The importance of niche in the delimitation of intraspecific and interspecific entities...... 19 2.6 BIOSEMIOTICS...... 19 2.7 REFERENCES...... 21 3 PLANNING, OBJECTIVES AND HYPOTHESES...... 31 3.1 CHAPTER 1. DISTRIBUTION OF TOMICUS DESTRUENS (COLEOPTERA: SCOLYTINAE) MITOCHONDRIAL LINEAGES: PHYLOGEOGRAPHIC INSIGHTS AND NICHE MODELLING...... 33 3.2 CHAPTER 2. IN SILICO PREDICTION AND CHARACTERIZATION OF MICRORNAS FROM THE BARK BEETLE TOMICUS YUNNANENSIS AND VALIDATION IN T. DESTRUENS (COLEOPTERA, SCOLYTINAE)...... 33 3.3 CHAPTER 3. TRANSCRIPTOME ANALYSIS AND IN SILICO IDENTIFICATION AND CHARACTERIZATION OF SERINE PROTEASES IN THE BARK BEETLE TOMICUS YUNNANENSIS...... 34 3.4 CHAPTER 4. APPLYING ECO-FIELD AND THE GENERAL THEORY OF RESOURCES TO BARK BEETLES: BEYOND THE NICHE CONSTRUCTION THEORY...... 35 4 PLANTEAMIENTO HIPÓTESIS Y OBJETIVOS...... 37 4.1 CAPÍTULO 1. DISTRIBUCIÓN DE LINAJES MITOCONDRIALES DE TOMICUS DESTRUENS (COLEOPTERA: SCOLYTINAE): ENFOQUE FILOGEOGRÁFICO Y MODELADO DE NICHO...... 39 4.2 CAPÍTULO 2. PREDICCIÓN IN SILICO Y CARACTERIZACIÓN DE MICRORNAS DEL ESCARABAJO DE CORTEZA TOMICUS YUNNANENSIS Y VALIDACIÓN EN T. DESTRUENS (COLEOPTERA, SCOLYTINAE)...... 39 4.3 CAPÍTULO 3. ANÁLISIS IN SILICO DEL TRANSCRIPTOMA Y IDENTIFICACIÓN Y CARACTERIZACIÓN DE PROTEASAS DE SERINA EN EL ESCARABAJO DE LA CORTEZA TOMICUS YUNNANENSIS...... 40 4.4 CHAPTER 4. APPLYING ECO-FIELD AND THE GENERAL THEORY OF RESOURCES TO BARK BEETLES: BEYOND THE NICHE CONSTRUCTION THEORY...... 41 5 RESÚMENES...... 44 5.1 CAPÍTULO 1. DISTRIBUCIÓN DE LINAJES MITOCONDRIALES DE TOMICUS DESTRUENS (COLEOPTERA: SCOLYTINAE): ENFOQUE FILOGEOGRÁFICO Y MODELADO DE NICHO...... 45 5.2 CAPÍTULO 2. PREDICCIÓN IN SILICO Y CARACTERIZACIÓN DE MICRORNAS DEL ESCARABAJO DE CORTEZA TOMICUS YUNNANENSIS Y VALIDACIÓN EN T. DESTRUENS (COLEOPTERA, SCOLYTINAE)...... 47 5.3 CAPÍTULO 3. ANÁLISIS IN SILICO DEL TRANSCRIPTOMA Y IDENTIFICACIÓN Y CARACTERIZACIÓN DE PROTEASAS DE SERINA EN EL ESCARABAJO DE CORTEZA TOMICUS YUNNANENSIS...... 49 5.4 CAPÍTULO 4. APLICANDO EL ECO-FIELD Y LA TEORÍA GENERAL DE RECURSOS A LOS ESCARABAJOS DE CORTEZA: MÁS ALLÁ DE LA TEORÍA DE LA CONSTRUCCIÓN DE NICHO...... 51 6 CHAPTER 1...... 53 6.1 ABSTRACT...... 55 6.2 INTRODUCTION...... 56 6.3 METHODS...... 58 6.4 RESULTS...... 63 6.5 DISCUSSION...... 72 7 CHAPTER 2...... 97 7.1 ABSTRACT...... 99 7.2 INTRODUCTION...... 99 7.3 MATERIALS AND METHODS...... 101 7.4 RESULTS...... 105 7.5 DISCUSSION...... 110 8 CHAPTER 3...... 121 8.1 ABSTRACT...... 124 8.2 INTRODUCTION...... 124 8.3 MATERIALS AND METHODS...... 126 8.4 RESULTS AND DISCUSSION...... 128 8.5 FINAL REMARKS...... 137 9 CHAPTER 4...... 145 9.1 ABSTRACT...... 147 9.2 INTRODUCTION...... 147 9.3 ECO-FIELD OF HOST SEARCH AND MATING IN BARK BEETLES. TRIADICS OF EAVESDROPPING EACH OTHER...... 151 9.4 SEMIOTIC NICHE AND THE ROLE OF ECO-FIELD...... 156 9.5 ECO-FIELDS NETWORKS AS THE BEGINNING OF NICHE CONSTRUCTION...... 159 9.6 ECO-FIELD IN TOMICUS SPECIES. A BIOSEMIOTIC APPROACH IN THE EVOLUTION OF T. DESTRUENS AND T. PINIPERDA..162 10 COMENTARIOS GENERALES Y CONCLUSIONES...... 171 10.1 CAPÍTULO 1. DISTRIBUCIÓN DE LINAJES MITOCONDRIALES DE TOMICUS DESTRUENS (COLEOPTERA: SCOLYTINAE): PERSPECTIVAS FILOGEOGRÁFICAS Y MODELADO DE NICHO...... 172 10.2 CAPÍTULO 2. PREDICCIÓN IN SILICO Y CARACTERIZACIÓN DE MICROARNS DEL ESCARABAJO DE LA CORTEZA TOMICUS YUNNANENSIS Y SU VALIDACIÓN EN T. DESTRUENS (COLEOPTERA, CURCULIONIDAE, SCOLYTINAE)...... 173 10.3 CAPÍTULO 3. ANÁLISIS DE UN TRANSCRIPTOMA, IDENTIFICACIÓN Y CARACTERIZACIÓN IN SILICO DE NUEVAS SERINAS PROTEASAS DEL ESCARABAJO DE LA CORTEZA TOMICUS YUNNANENSIS...... 174 10.4 CAPÍTULO 4. APLICANDO EL ECO-FIELD Y LA TEORÍA GENERAL DE RECURSOS A LOS ESCARABAJOS DE CORTEZA: MÁS ALLÁ DE LA TEORÍA DE LA CONSTRUCCIÓN DE NICHO...... 175 11 GENERAL COMMENTS AND CONCLUSIONS...... 177 11.1 CHAPTER 1. DISTRIBUTION OF TOMICUS DESTRUENS (COLEOPTERA: SCOLYTINAE) MITOCHONDRIAL LINEAGES: PHYLOGEOGRAPHIC INSIGHTS AND NICHE MODELLING...... 178 11.2 CHAPTER 2 IN SILICO PREDICTION AND CHARACTERIZATION OF MICRORNAS FROM THE BARK BEETLE TOMICUS YUNNANENSIS AND VALIDATION IN T. DESTRUENS (COLEOPTERA, CURCULIONIDAE, SCOLYTINAE)...... 179 11.3 CHAPTER 3. TRANSCRIPTOME ANALYSIS AND IN SILICO IDENTIFICATION AND CHARACTERIZATION OF NOVEL SERINE PROTEASES IN THE BARK BEETLE TOMICUS YUNNANENSIS...... 180 11.4 CHAPTER 4. APPLYING ECO-FIELD AND THE GENERAL THEORY OF RESOURCES TO BARK BEETLES: BEYOND THE NICHE CONSTRUCTION THEORY...... 181 12 APPENDIX...... 183 2 Index figures

Figure 1. Representation of semiosis...... 20 Figure 2. Schematic view of the Umwelt...... 20 Graphical abstract Chapter 1 Resumen...... 45 Graphical abstract Chapter 3 Resumen...... 49 Graphical abstract Chapter 4 Resumen...... 51 Graphical abstract Chapter 1...... 54 Figure 1. Study area...... 61 Figure 2. Haplotype network...... 65 Figure 3. Spatial distribution of haplotypes and clades...... 68 Figure 4. Potential distribution of the analysed lineages...... 69 Figure 5. Tests for niche identity...... 70 Figure 6. Tests for niche similarity...... 71 Figure 7. Partial response curve results of GAM of diversity haplotype...... 72 Figure S1. Correlation cluster of bioclimatic and topographic variables...... 90 Figure 1. Scheme of the methodological steps...... 103 Figure 2. The origin of the pre-miRNAs...... 106 Figure 3. Length of pre-miRNAs...... 107 Figure 4. The detected size of miRNAs...... 108 Figure 5. Consensus sequences of miRNAs...... 109 Figure 6. Expression of mirRNA 2c and miRNA 4944...... 109 Figure 7. Expression of mirRNA 2c (A, B) and miRNA 4944...... 110 Figure S1. Consensus sequences of pre-miRNAs...... 113 Graphical abstract Chapter 3...... 123 Figure 1. Characterization of the transcriptome...... 128 Figure 2. E-value distribution of the top BLASTx hits...... 129 Figure 3. Species distribution of the top BLAST hits...... 129 Figure 4. GO distribution by level 2 of the transcriptome...... 130 Figure 4. Phylogenetic tree built with RaxML...... 136 Graphical abstract Chapter 4...... 146 Figure 1. The triadic structure of the sign process...... 156 Figure 2. Resource-Ecofield Venn diagram...... 158 Figure 3. Representation of the eco-field networks...... 160 Resumen general

1 ResumenResumen generalgeneral

1 Resumen general

1.1 Introdución

Los escarabajos de corteza son insectos unidos a las plantas leñosas. Estos escarabajos están encuadrados dentro de los curculiónidos en la subfamilia Scolytinae. Esta subfamilia está compuesta solamente por insectos fitófagos. Su ciclo vita es crítico en los ciclos de materia y energia de los bosques templados. Además estos insectos pueden afectar a los intereses ecómicos en las zonas donde se producen fenómenos de plaga, las cuales pueden acabar afectando a miles de kilómetros cuadrados de bosque.

El género de escolítidos aquí estudiado es el género Tomicus. Este género incluye siete especies que se desarrollan principalmente en coníferas del género Pinus. De acuerdo con Duan et al. (2004), es el este de Asia donde se originó el género, ya que allí se encuentran cuatro especies que no se encuentran en el resto de la distribución del taxón: T. puellus (Reitter, 1894), T. pilifer (Spessivtsew, 1919) T. brevipilosus (Eggers, 1929), T. yunanensis (Kirkendall y Faccoli, 2008) y T. armandii (Li & Zhang, 2010). Además de ese área, T. piniperda (Linnaneus, 1758) y T. minor (Hartig, 1834) se encuentran ampliamente distribuidos por Eurasia. La distribución de T. piniperda se ha visto ampliada debido a la translocación de ejemplares a los bosques de los Grandes Lagos en Norteamérica. La única especie del género que no habita en el este de Asia es T. destruens, con areal mediterráneo y macaronésico.

Los microARNs (miRNAs) son secuencias de ARN de doble cadena que tienen entre 18 y 22 nucleótidos presentes en los genomas de plantas, animales y virus y regu- lan la expresión de genes diana. Estas moléculas son evolutivamente conservadas y se pueden encontrar en intrones, exones y /o regiones intergénicas. Los miRNAs están implicados en diversos procesos fisiológicos en animales y plantas, como la difer- enciación de diversos tejidos, control de la muerte celular, la secreción de insulina y también pueden estar asociada con procesos que permiten la adaptación a situaciones de estrés (referencias en Leclercq et al 2013; Valencia-Sánchez et al 2015). Los estudios con estas moléculas han modificado de forma significativa la visión de los biólo- gos moleculares sobre la importancia de los miRNAs (Sontheimer y Carthew 2009).

2 Resumen general

Hay más de 5000 miARNs descritos en una variedad de organismos depositados en miRBase (Kozomara y Griffiths-Jones, 2014).

Las proteasas son enzimas que catalizan la digestión de las proteínas de la dieta lib- erando los aminoácidos esenciales para el crecimiento de los insectos. Se clasifican como serina proteasas; cisteína proteinasas; aspártico proteinasas y metaloproteinasas (Bode & Huber, 1992). Las serina proteasas son las más abundantes y corresponde a aproximadamente el 95% de las enzimas digestivas presentes en el intestino medio insectos (Srinivasan et al., 2006). Además de en la función digestiva, las serina proteasas están involucrados en otros procesos metabólicos importantes, tales como la respuesta inmune (Jiang et al, 2003a, 2003b; Kanost et al, 2004) y los procesos de muda de los insectos (Wei et al, 2007; He et al., 2009).

Al principio del s. XX fueron varios los autores que intentaron explicar el porqué de las distribuciones de los distintos organismos. Fueron Grinnell (1924) and Elton (1927) los que comenzaron una tentativa de definición de la entidad abstracta llamada nicho. Mientras Grinnell lo resumió como “los requerimientos que necesita una pobla- ción para desarrollarse”, la cual es espacialmente explícita y medible desde su distribu- ción. La definición de Elton se centra en las interacciones bióticas y sus impactos en las comunidades. Más tarde, fue con Hutchinson (1947) cuando se unificaron sendos puntos de vista. Por un lado definió el nicho fundamental como el hipervolumen n-dimensional definido por n-variables. Dentro de ese espacio se encuentra el nicho realizado, el cual es un subconjunto del nicho fundamental más las interacciones bióticas (Peterson et al., 2011). En la última década la Teoría del Nicho (TN) ha experimentado un amplio avance debido a la disponibilidad de grandes bases de datos (ecológicas y genéticas) y la facilidad de su uso. La potencia de cálculo de los ordenadores ha per- mitido poner a prueba las bases teóricas en amplias distribuciones de organismos y en las filogenias de especies.

La biosemiotica es el estudio de los signos (signs) en los sistemas vivos. Antes de avanzar conviene aclarar algunos aspectos en los cuales se basa la semiótica. La semiosis es el proceso por el cual un signo obtiene un significado mediante una interpreta-

3 Resumen general ción. La biosemiótica es una disciplina que usa la metodología de la semiótica, con la cual describe y analiza la estructura de procesos semióticos independientemente de su material base o de las condiciones bajo las cuales estos procesos son observados. Por ejemplo, dentro de células, citosemiosis, entre tejidos y poblaciones de células, en comunicación animal zoosemiosis. Esta definición de semiotica está basada en los tra- bajos del fundador de la moderna semiótica además de filósofo y científico Charles Sanders Peirce (1839 – 1914). Según Peirce la descripción de la semiosis debe ser tratada como una relación constituida por tres términos irreductibles contados entre sí (signo-objeto-interpretante, S-O-I). El signo (representamen) es un medio para la comunicación de una forma o de un hábito embebido en un objeto hacia el interpretante (Queiroz y El-Hani, 2008).

*La bibliografía de este apartado se encuentra en el apartado Introduction.

1.2 Objectivos

El objetivo general de esta tesis es realizar una aproximación al estudio de los escarabajos de corteza desde diferentes perspectivas. Los aspectos estudiados han per- mitido identificar si existen patrones de nicho diferenciados en las poblaciones de Tomicus destruens de la cuenca mediterránea, analizar las diferencias en la expresión de microRNAs en T. yunnanensis y T. destruens, caracterizar posibles serina proteasas en T. yunnanensis y analizar las redes ecológicas en las que participan los escarabajos de corteza.

Para conseguir este objetivo se ha estructurado el trabajo en 4 capítulos, uno de ellos basado en análisis filogeográfico y aspectos ecológicos, dos analizando aspectos genómicos y uno desde una perspectiva biosemiótica.

En el primer capítulo, se pretende identificar algunos factores ambientales del nicho ecológico que afecta a la distribución de los linajes mitocondriales (mtLs) y la diversidad genética de T. destruens a lo largo de toda la cuenca mediterránea.

4 Resumen general

El objetivo del segundo capítulo es identificar y caracterizar in silico microRNAs de Tomicus yunnanensis y validar su expresión en T. yunnanensis y T. destruens, mediante el uso de herramientas bioinfomáticas y la utilización de métodos moleculares mediante stem-loop pcrs.

En el tercer cápitulo el objetivo principal de este estudio fue analizar in silico el transcriptoma del escarabajo de corteza T. yunnanensis mediante herramientas bioin- formáticas e identificar posibles genes candidatos para la síntesis de serina proteasas.

El trabajo de los capítulos dos y tres intenta proporcionar información sobre marcadores específicos (enzimas de digestión y microRNAs) que puedan ser utilizados en futuros estudios, con el objetivo de desarrollar nuevas estrategias de control de esta especie de insectos plaga y otras especies del mismo género.

El objetivo del cuarto capítulo es identificar las señales del nicho semiótico que afectan a la comunicación intraespecífica de los escarabajos de corteza y también la comunicación con otros organismos, utilizando el concepto de “eco-field” junto con la Teoría General de Recursos (GTR) con el fin de detectar los procesos de ampliación del nicho semiótico en escarabajos de corteza a lo largo de la matriz del paisaje forestal.

1.3 Metodología

Para cumplir estos objetivos se utilizaron diversos métodos que abarcan herramientas ecológicas, bioinformáticas y semióticas. En el primer capítulo se aplicó una metodología combinada para la distribución de modelado y exploramos el nicho ecoló- gico de los linajes jerárquicos de T. destruens en una escala mediterránea. Para ello se utilizó el análisis de clados anidado (NCA), junto con la el algoritmo de máxima en- tropía (MaxEnt) (NCA-MaxEnt) a fin de detectar procesos de segregación de nicho ecológico de los linajes del ADN mitocondrial (ADNmt).

5 Resumen general

En el segundo capítulo, se han identificado nuevos microRNAs (miRNAs) en la biblioteca genómica del transcriptoma de un conjunto de individuos de T. yunnannensis y se validaron 7 de los miRNAs identificados en esta especie y en la especie con- genérica Tomicus destruens. Este estudio proporciona una base para la evaluación de estos miRNA para ser utilizada en estrategias de control de plagas en el género Tomicus y otros escarabajos de corteza.

Para lograr el objetivo del capítulo tercero, se han identificado nuevas serina proteasas en la biblioteca genómica del transcriptoma de un conjunto de individuos de T. yunnannensis. Cuatro proteasas semejantes a tripsina y cinco proteasas semejantes a quimotripsina fueron anotados in silico; además se anotaron otros cuatro contigs que mostraban un dominio clip. Este estudio proporciona información básica para una posterior evaluación de las serina proteasas como herramientas de control de plagas en el género Tomicus y otros escarabajos de corteza.

Para alcanzar el objetivo del capítulo cuarto, aplicamos una metodología semiótica para encontrar las señales principales y explorar el nicho semiótico en el ciclo de vida de las especies de escarabajos de corteza. Para ello se identican el procesos de semiosis implicado en la búsqueda de un árbol hospedador adecuado y además de la búsqueda de pareja para la reproducción por parte de los escarabajos de corteza.

1.4 Resultados

En el capítulo 1 se presenta un nuevo enfoque de genética de poblaciones junto con datos ambientales y biogeográficos que conduce a la obtención de inferencias para el modelado de nicho ecológico. Utilizamos linajes jerárquicos obtenidos mediante análi- sis cladístico anidado (NCA) de los haplotipos de ADN mitocondrial (ADNmt) del escarabajo de corteza Tomicus destruens, para modelar la distribución por máxima entro- pía, usando variables ambientales y de hospedador a lo largo de toda la cuenca medite- rránea. Además se realizaron pruebas de identidad y similitud para verificar los linajes intraespecíficos (clados NCA y haplotipos) con el fin de determinar un cambio o con-

6 Resumen general servadurismo de nicho entre ellos. Además, se calcularon cuatro índices de nueve zonas geográficas de la cuenca del Mediterráneo para evaluar la variabilidad de los factores ambientales que determinan la distribución de la diversidad de haplotipos a gran escala geográfica. Los modelos ecológicos desarrollados indican que las linajes de ADN mitocondrial orientales minoritarios de T. destruens difieren en su nicho ecológi- co potencial de acuerdo a su relación con variables climáticas extremas. Por el contrario, los linajes occidentales más ampliamente distribuidos muestran una estrecha rela- ción con Pinus, su árbol hospedador. Además, la predicción de niveles más altos de haplotipos exclusivos y endémicas del Mediterráneo se da en las zonas con alta variabilidad de temperatura en el período húmedo. El nicho del grupo oriental parece estar incluido en parte del rango del espacio ecológico de los dos principales clados occidentales.

Este resultado sugiere que el cambio de nicho podría haber comenzado, aunque existiendo todavía una relación temprana con su árbol hospedador Pinus brutia. Alter- nativamente, la variabilidad de la temperatura en el período más húmedo parece estar relacionada con una alta proporción de haplotipos endémicas de T. destruens, posible- mente implicando un equilibrio entre la duración del período reproductivo de vuelo de T. destruens y el estado de vigor del árbol hospedador y su estadio de crecimiento. Este estudio ilustra un buen ejemplo de los beneficios que el modelado de nicho ecológico ofrece para entender los patrones de la genética de poblaciones y la filogeografía.

En el capítulo dos, siete microRNAs fueron validados en ambas especies por PCR cuantitativa en tiempo real (RT-qPCR), de los cuales mir-2c-3p y mir-4944-5p mostraron expresión en ambas especies. La expresión de tyu-mir-2c-3p fue mayor en T. destruens que en T. yunnanensis, tanto en machos como en hembras. Sin embargo, se ob- servó la máxima expresión de tyu-mir-4944-5p en las hembras de T. destruens, seguido por los machos de T. yunnanensis y T. destruens. Las hembras de T. destruens mostraron expresión de mir-2c-3p más baja que los machos, lo que implica que este miRNA podría estar relacionado con la ovogénesis y vitelogénesis. Este estudio proporciona bases para una posterior evaluación de miARNs que podrán ser utilizados en las estrategias de control de plagas en el género Tomicus y otros escarabajos de corteza.

7 Resumen general

En el capítulo tres se presentan el anáisis del transcriptoma de Tomicus yunnanensis, el cual fue analizado por medio de herramientas bioinformáticas, obteniendo 209934 contigs. Más de la mitad de los contigs (55.23 %) no mostraban significante similaridad de las secuencias a nivel de proteínas. Esto sugiere que estas proteínas qui- zás representas genes de rápida evolución o genes taxonómicamente restrictivos que podrían estar relacionados con procesos de diversificación de especies. La composi- ción de diferentes clases de enzimas digestivas fueron anotadas. Las enzimas digestivas más expresadas fueron las serina proteasas, lipasas y β-glucosidasas. Por lo tanto, estas diferencias observadas para T. yunnanensis en relación a otras especies es quizás explicada por la adaptación de este insecto a dieta basadas en coníferas con alto niveles de compuestos terpenos. Entre las serina proteasas identificadas, cuatro proteasas semejantes a tripsinas y cinco proteasas semejantes a quimotripsinas fueron anotadas in silico. Estas enzimas fueron clasificadas dentro de la familia SA1 y mostraba todas las características de serina proteasas digestivas. Además, otros cuatro contigs fueron identificados y anotados como serina proteasas pero con la falta de la región de unión a substrato y sólo dos de las regiones de la triada catalítica (Histidina-Ácido aspártico). Estos contigs tenían un dominio Clip, el cual está involucrado en la respuesta inmune innata y del desarrollo embrionario. Una análisis filogenético junto con serina proteasas representativas de varias especies de insectos fue realizada para discernir sus relaciones evolutivas. Serina proteasas de T. yunnanensis no formaban un clado congruente con la taxonomía, sino que presentaba homologías con otras especies de insectos, que esto es esperable si ellas han evolucionado por duplicaciones de genes, seguidas de divergencia debido a presiones evolutivas selectivas.

Y finalmente en el capítulo cuatro se analizada desde una nueva perspectiva en la ecología del paisaje con la aplicación del concepto Eco-field junto con la Teoría Gene- ral de Recursos. En este artículo, nosotros describimos la existencia de un eco-field en los escarabajos de corteza como una configuración espacial con una específica portadora de significado para cada interacción de cada organismo-recurso. Los escarabajos de la corteza son insectos involucrados en cambios significativos de los paisajes forestales cuando los episodios de plaga ocurren. Estos escarabajos de la corteza son trans-

8 Resumen general portados por el reconocimiento de señales semióticas hacia los árboles hospedadores usando un específico eco-field. Estas señales son principalmente un grupo de olores, los cuales han sido llamado odourtope. Estas interacciones con otros organismos (hongos, bacterias, nematodos, predadores, etc.) toma lugar a través de la compartición de información relevante de las redes de eco-field (redes de representamen) en la matriz forestal. Las redes de eco-field permiten una expansión del nicho semiótico realizado de los escarabajos de la corteza hacia el nicho semiótico potencial. Además, si diferentes organismos terminan mostrando una interdependencia del eco-field, el proceso de construcción de nicho puede ser iniciado. Posteriormente, si esta interdependencia se vuelve crucial el proceso puede llevar al establecimiento de relaciones mutualistas. Este es un ejemplo de cómo los procesos evolutivos son iniciados por el reconocimien- to de las señales por medio de una red de eco-field. Además, nosotros mostramos una descripción de los eco-field de búsqueda de hospedador y de maduración en Tomicus destruens y Tomicus piniperda y una hipotética explicación de la evolución de su coexistencia simpátrica.

Resumen general

22 IntroductionIntroduction

10 Introduction

2.1 Taxonomy of the genus Tomicus

Bark beetles (Coleoptera: Curculionidae, Scolytinae) are insects linked to woody plants. This subfamily is exclusively composed by phytophagous species, and considering the part of the plan where it feeds, several eating habits have been recorded: floe- mofagous (bark beetles), xilofagous (usually immature), xilomicetofagous (ambrosia bark beetles), browsers (a few species feed on stems of different herbaceous or semi- woody plants), seed feeders and feeders from medulla of some plants. In temperate forest ecosystems, their life cycle is critical in the matter and energy cycles of forests, but sometimes come to disrupt the economic interests and produce outbreaks episodes. The importance of these insects is that they are the first to arrive and complete their life cycle in susceptible hosts, facilitating the penetration of other wood-eating organisms, like other insects (Hymenoptera and other Coleoptera) and especially fungi. Bark beetles affect mainly host trees with very little defence capability, either by weakening by environmental conditions, injured or affected by fire. But there are a few species called primary species to be responsible for breaking the defence system of the tree, leaving it weakened for subsequent incursions by bark beetles or other secondary species (Gil and Pajares, 1986).

Some authors consider bark beetles not as family Scolytidae (Wood and Bright, 1992) but as a subfamily, Scolytinae, included within the family Curculionidae. Ac- cording to Marvaldi et al. (2002) the bark beetles are paraphyletic, while in a deeper analysis of 1880 taxa, Hunt et al. (2007) concluded that the bark beetle would be a polyphyletic group. In this paper bark beetles are considered as a subfamily, Scolyt- inae, considering that this consideration may be provisional until conclusive future work determine its taxonomic identity.

The taxonomy of Scolytinae includes approximately 6,000 species distributed into two tribes: Hylesini, with 10 tribes with 62 genera, and Scolytini with 14 subtribes and 113 genera. In the Iberian Peninsula these two groups are present and 31 genera di- vided into 15 tribes have been cited. In Spain, in the lack of updated information, the

11 Introduction number of species present is considered to be around 150 (Lombardero and Novoa, 1994).

The genus Tomicus includes seven species mainly living in coniferous of the genus Pinus. According to Duan et al. (2004), the genus originated in East Asia, as it contains five species that are not found in the rest of their distribution: T. puellus (Reitter, 1894), T. pilifer (Spessivtsew, 1919) T. brevipilosus (Eggers, 1929), T. yunanensis (Kirkendall and Faccoli, 2008) and T. armandii, Li et al., Zhang 2010. In addition to this area, T. piniperda (Linnaneus, 1758) and T. minor (Hartig, 1834) are widely distributed in Eurasia. T. piniperda distribution has been extended due to specimens’ translocation to the Great Lakes forests in North America. The only species of the genus that does not live in East Asia is T. destruens, showing Mediterranean and Ma- caronesian areal.

The genus is morphologically very homogeneous, being insects ranging from 3 to 5.5 mm in size, elongated, with short face, front carinated, pronotum longer than wide with an anterior narrowing, elytra with striae formed by gross point and flat intervals with uniform pilosity with short hairs. Antennal club is conical, formed by 4 segments, with the first more developed, antennal funiculus with 6 segments. The pronotum has an anterior decline compared to elytral flat, finely dotted with small and spaced dots. Basal edge of the elytra curved, striae formed by regularly spaced dots and plane flat intervals (Faccoli, 2007). They are morphologically uniform insects, although with some variation in colour patterns, with just little sexual dimorphism and scarce interspecific variability.

The genus Tomicus Latreille, 1812, stands out as a major pest of Eurasian forests (Schroeder 1987; Bouhot et al 1988, 1991 Hui.). The iberian Peninsula has 3 species, T. minor (Harting, 1834), T. piniperda (Linnaeus, 1758) and T. destruens (Wollaston, 1865) (Wood et al., Bright, 1992). T. piniperda and T. destruens are considered to be important pest of European and Mediterranean pine forests respectively (Guerrero et al 1997, 1983 Långström Schroeder, 1987; Bouhot et al 1988, Hui 1991, Nanni and Tiberi 1997; Monleón et al.1996) while T. minor is considered to be a much less ag-

12 Introduction gressive species. T. destruens and T. piniperda are morphologically very similar species, having been long considered as ecotypes of the same species (Carle 1973 Schwer- dtfger 1981) or as different species (Lekander 1971 Wood et al., Bright, 1992, Pfeffer 1995). Molecular studies have shown that there are enough differences between them to be considered different species (Gallego et al., Galián, 2001). French and Austrian research groups later confirmed those differences (Kerdelhué et al. 2002 Kohlmayr et al. 2002, Ritzerow et al. 2004). Studies on the distribution of these species in the Iberian Peninsula have been addressed by Gil and Pajares (1986) and Lombardero (1994), indicating that T. piniperda is present throughout the Peninsula, while T. minor has a more fragmented distribution. However these authors did not consider T. destruens as a species being therefore included within T. piniperda. Gallego et al. (2004) found that in wetlands of north-central Spain (and possibly high altitudes in the South) where T. piniperda inhabits, T. destruens has its climate distribution limit. Dis- tributions of both species overlap on the northern border of this area, where some co- occurrences have been detected. T. minor potentially inhabit high and humid fragmented areas.

2.1.1 Tomicus destruens

Tomicus destruens (Coleoptera, Scolytidae) is a circummediterranean and Macar- onesian species (Wood and Bright, 1992) that attacks Mediterranean pines (Pinus halepensis, P. pinaster, P. pinea and P. brutia), P. canariensis and occasionally P. nigra (Gallego et al. 2004; Vasconcellos et al. 2006). This bark beetle plays a major role in the decline of pine forests in the Mediterranean countries (Faccoli et al., 2005; Chakali, 2005). The wide distribution of this species is mainly influenced by i) climatic factors (Gallego et al. 2004) that could impose a physiological limitation in the host tree, related to temperature and water availability, and ii) host factors (Kerdelhué et al. 2002; Horn et al. 2006), that limit the distribution of the bark beetle to that of the host pine trees.

13 Introduction

Distribution models for Tomicus have been proposed at the Iberian Peninsula scale, using molecular tools and models (Gallego et al. 2004). These models indicate that T. destruens is the most common species, occupying the entire southern half, the Medi- terranean coast, the Ebro Valley and appearing occasionally on the Atlantic and Can- tabrian coast, even penetrating the North Plateau through the Duero valley, being ex- cluded from the large elevations of the Betic Mountains. Horn et al. (2012) reported that two species of Tomicus, T. destruens and T. piniperda presented parapatric distributions and opposite climate demands, with T. destruens occurring in locations with warmer temperatures, whereas T. piniperda occurs under a colder climate.

Gallego et al., Galián (2008) investigated the relation of environmental variables with the distribution of mitochondrial lineages of T. destruens combining phylogeography and regression models to study the role of five environmental predictors at fine scale in the distribution of a local genealogy. The analysis revealed a high genetic diversity, with 52 haplotypes present in Sierra Espuña forest (SE Spain) and 21 haplotypes in the other 14 Spanish populations, all included in the western clade of the Mediterranean phylogeography of the species. The authors found a micro-distribution of the species related to altitude and putative niche segregation between lineages associated with the micro-environmental conditions of their host pine trees. They demon- strated a relation between the environmental heterogeneity and the haplotypic diversity at Mediterranean basin scale.

2.1.2 Tomicus yunnanensis

Tomicus yunnanensis is an Asian species which reproduces in Pinus yunnanensis, and feeds in shoots of P. yunnannensis, P. armandii, P. kesiya and P. densata. Its potential distribution is the distribution of its host P. yunnanensis in the southwestern Chinese provinces, (Yunnan, Guangxi, Guizhou Sichuan, Xizang) and Burma, at 400- 3100 m elevation (Lieutier et al., 2015). This bark beetle had provoked rapid decline and death of the Chinese forests in several episodic outbreaks (Ye and Ding, 1999).

14 Introduction

This beetle was considered to be T. piniperda for a long time, however, molecular and taxonomic studies had determined its entity as a species (Duan et al., 2004; Kierkedall et al., 2008). Also T. yunnanesis life cycle is more similar to that of T. destruens than to T. piniperda. Thus, T. yunnanensis flight period may start in November but the main flight occurs in February–March and then there is a sister brood flighting from April to May (Ye, 1991; Li et al., 1993). T. yunnanensis shared the host pine with T. minor. Thus, T. minor lives in the lower parts of the tree, while T. yunnanensis lives in the upper parts. Molecular studies have shown that T. yunnanensis had undergone several translocations mediated by humans (Jun et al., 2014).

2.2 Serine proteases

Proteases are enzymes which catalyze the digestion of proteins from the diet releas- ing the essential amino acids for growth of insects. They are classified as serine proteases, cysteine proteinases, aspartic proteinases and metalloproteinases (Bode and Huber, 1992).

Serine proteases are the most abundant and corresponds to approximately 95% of digestive enzymes present in the insect midgut (Srinivasan et al., 2006). In addition to digestive function, serine proteases are involved in other important metabolic processes such as the immune response (Jiang et al., 2003a, 2003b; Kanost et al., 2004) and insect molting processes (Wei et al, 2007; He et al, 2009).

These enzymes are important in polyphagous insects for flexibility and adaptation to different host plants species. This process occurs through changes in the rate of gene expression in response to the presence of several protease inhibitors produced by the infected plants (Srinivasan 2006; Chikate et al 2013). The change in gene expression may involve gene overexpression or protein production insensitive to inhibitors present in the plant (Brioschi et al, 2007; De Oliveira et al, 2013).

Serine proteases are classified based on the amino acids presents at sites which de-

15 Introduction termine the substrate specificity: in trypsin (Lys / Arg), in chymotrypsin (Phe / Tyr / Leu) and elastase (Ala / Val). Furthermore, the central catalytic site of these proteins is composed of three residues: aspartic acid, histidine and serine forming the "catalytic triad" (Tyndall et al, 2005; Page and Di Cera 2008). Studies in the past decade have shown a wide range of serine proteases that are expressed spatially and temporally in the midgut in different insect species (Vinokurov et al., 2006; Simpson et al., 2007; Marshall et al, 2008 ; Ge et al, 2012).. Several studies based on analysis of cDNA lib- raries have been used to identify and characterize these proteins in different groups of insects, especially those with the potential to cause economic damage, as Lepidoptera (Zavala et al 2008; Senthilkumar et al. 2010; Gatehouse 2011).

Metabolic importance of serine proteases makes them targets of studies for the development of different strategies to control insect pests. The development of such strategies depends on the characterization of the enzymes present in the midgut of the insects and also on the analysis of expression patterns of these genes (Zhang et al. 2,010 Haq et al., 2004; Zhu-Salzman and Zeng, 2015).

The serine proteases are well characterized in Lepidoptera, but in Coleoptera their characterization is restricted to a few species of the families Curculionidae, Chrysomelidae, Tenebrionidae and Lucanidae (Gruden et al 2003; Vinokurov et al 2006; Broehan et al, 2010; Alahmadi et al 2012. Zibaee and Malagoli, 2014).

2.3 Non-coding RNA types and functions

A non-coding RNA (ncRNA) is an RNA molecule that is not translated into a protein. Non-coding RNA include several types, among which are those over 200 bp in lengths, called large non coding RNA (lncRNA) and those with less than 30 bp. The latter include highly abundant and functionally important RNAs such as microRNAs (miRNAs), RNA interference (siRNAs), Piwi-associated RNA (piRNAs), nucleolar RNA (snRNA), transfer RNA (tRNA) and ribosomal RNA (rRNA).

16 Introduction

The most studied ncRNA in the last decade and that have been characterized in a wide variety of organisms are the siRNAs and miRNAs. From a functional perspective, the miRNAs are involved in the regulation of endogenous genes while siRNAs act in response to the presence of exogenous nucleic acids, such as those from viruses and transposons. Regulation processes mediated by these RNAs are called interference RNA (iRNA) or gene silencing RNA (Valencia-Sanchez et al. 2006).

2.4 MicroRNAs

MicroRNAs are sequences of double-stranded RNA between 18 and 22 nucleotides present in the genomes of plants, animals and viruses and regulate the expression of target genes. These molecules are evolutionarily conserved and can be found in in- trons, exons and/or intergenic regions. Micro RNAs are involved in diverse physiological processes in animals and plants, differentiation of various tissues, control of cell death, and insulin secretion can also be associated with processes that allow adaptation to stress (references in Leclercq et al. 2013; Valencia-Sanchez et al. 2015). Studies with these molecules have significantly changed the vision of molecular biologists on the importance of miRNAs (Sontheimer and Carthew 2009). There are more than 5000 miRNAs described in a variety of organisms deposited in miRBase (Kozomara and Griffiths-Jones, 2014).

Micro RNAs derive from primary transcripts which are longer and are processed to generate the mature molecule. Alike coding genes, the primary transcripts present CAP in the 5 'end and poly A tail to the 3' end. This molecule in the cytoplasm will go through a final processing, generating a molecule of about 22 base pairs. Mature miRNAs are incorporated into a silencing complex (RISC) in the cytoplasm and the process of splitting or breaking of the target mRNA depends upon its pairing with the miRNA (Gomes et al. 2013).

Different authors have performed studies in different organisms on the possible applications of these regulatory molecules as potential control agents of insect pests. The

17 Introduction results show that there is great potential for using this strategy (Terenius et al 2011;. Zhu 2013; Yu et al 2013; Li et al 2013; Gu and Knipple 2013; Burand and Hunter 2013). Recently, new sequencing technologies have allowed the identification of many new miRNAs. For example, in 2012 over 21,000 miRNAs were deposited in the database miRBase (Wu et al. 2013). MicroRNAs are identified by different algorithms based on structural and thermodynamic features.

2.5 Niche theory and modelling

There are severeals concepts around the niche term. One of them is the fundamental niche, which is the n-dimensional hypervolume defined by n-variables. The other term is the realized niche, which is a subset of fundamental niche plus the biotic interactions (Peterson et al., 2011).In the last decade the Niche Theory (NT) has experienced a broad advance because of the availability of large and ease of use databases (both environmental and genetics). The computing power has tested the theoretical bases in broad distributions of organisms and species phylogenies.

When addressing the problem there are two methods (mechanistic and correlative). The mechanistic methods are carried out by ecophysiological models of a given organism, so that, it manages for more information and therefore are more difficult to ex- ecute. Correlative methods are easier to implement since only need the distribution of samples and layers of environmental variables available in the databases (Peterson et al 2011. Alvarado-Serrano and Knowles, 2014).

Correlative methods use different algorithms depending on the organisation of presence and absence. Some methods only use presences as BIOCLIM or HABITAT. Other procedures use real presences and absences, such as generalized linear models (GLM), Generalized Additive Models (GAM) and Classification and Regression Trees (CART). Other approaches use presences of the organism and the environment ("background") throughout the study area, incorporating information on environmental variation when developing the appropriate model variables, as in the case of maximum en-

18 Introduction tropy algorithms (MAXENT) and factor analysis of ecological niche (ENFA). Finally, some protocols use presences and false absences when we do not have true absences. In this case, the program uses a genetic algorithm (GARP) that makes sampling in areas where they are not presences and are considered as absences (Peterson et al., 2011).

2.5.1 The importance of niche in the delimitation of intraspecific and interspecific entities

The study of niche can be useful to infer speciation processes and diversity patterns. It is in these cases where the terms of niche conservation (NC) and niche shift (NS) are defined. Niche conservation is defined as the tendency of species or clades to retain their niches and ecological characteristics over time. However, niche shifting is the situation in which a species or clade modifies its niche into other ecological values. Both NC and NS can occur at different spatial and/or temporary scales (Wiens et al., 2010). These concepts serve to better assess phenomena such as invasive species, food webs and species richness patterns.

2.6 Biosemiotics

Biosemiotics is the study of signs in living systems. Before moving forward, it is necessary to clarify some aspects in which semiotics is based. Semiosis is the process by which a sign gets a meaning by an interpretation. Biosemiotics is a discipline that uses the methodology of semiotics, with which describes and analyses the structure of semiotic processes regardless of base material or the conditions under which these processes are observed. For example, within cells is called citosemiosis, among tissues and cell populations in animal communication is called zoosemiosis. This definition of semiotics is based on the work of the founder of modern semiotics as well as philo- sopher and scientist, Charles Sanders Peirce (1839-1914). According to Peirce's description of semiosis, it should be treated as a relationship constituted by three irredu- cible terms counted together (sign-object-interpretant, SOI). The sign (representamen)

19 Introduction is a means of communication in one form or habit embedded in an object to the inter- preter (Queiroz and El-Hani, 2008).

Figure 1. Representation of semiosis as the relation between three terms connected (sign-object-interpretant, S-0-I). This triadic relationship communicates or carries a habit from the object to the interpretant through the sign.

Among the different authors who have contributed to the field of biosemiotics it's worthy to distinguish Thomas A. Sebeok (1920-2001). It was Sebeok who began applying semiotics to the study of living beings. Thanks to him, the writings of Jacob Von Ueküll (1864-1944) were also recovered. Besides, he initiated the process to bring together all researchers, to finally come together at the end of XX century and beginning of XXI century, through congresses, various books, an international society (Inter- national Society for Biosemiotic Studies, ISBS) and a journal (Biosemiotics).

Ueküll adopted the word Umwelt to indicate the subjective world of an organism. This Umwelt is the combination of the world perceived plus operating world or engine, with which it is associated by a functional circuit. For that, Ueküll is considered a pre- cursor in cybernetics and animal ethology (Barbieri, 2008a)

20 Introduction

Figure 2.Schematic view of the Umwelt as a functional cycle that connects the inner world with the outer world.

The signs can be classified into icons, indices or symbols. A sign is an icon when the association with the object is established by means of a relation of similarity. For instance, for an insect, the presence of an individual of the same species of the opposite sex, is synonymous of reproduction. A sign is an index when the association with the object is established through a relationship of physical union (correlation). For example, when an insect perceives a chemical compound, it is synonymous of a suitable host tree for oviposition. A sign is a symbol when the association with the object is established by means of a rule or convention. An example of this is the alarm system used by monkeys with the presence of various predators. Even when the alarm oper- ates in the absence of a predator, the monkeys interpret that the predator is around and they initiate the escape behaviour (Barbieri, 2008b).

2.7 References

Alahmadi, S. S., Ouf, S. A., Ibrahim, R. A., and El-Shaikh, K. A. (2012). Possible control of date palm stag beetle, Lucanus cervus (L.) (Coleoptera: Lu-

21 Introduction

canidae), using gut protease inhibitors of different bio-control agents. Egyp- tian Journal of Biological Pest Control, 22(2), 93–101.

Alvarado-Serrano, D. F., and Knowles, L. L. (2014). Ecological niche models in phylogeographic studies: applications, advances and precautions. Molecular Ecology Resources, 14(2), 233–248.

Barbieri, M. (2008a). Has Biosemiotics Come Of Age? and Postscript. In M. Bar- bieri (Ed.), Introduction to Biosemiotics (pp. 101–114). Springer Nether- lands.

Barbieri, M. (2008b). Introduction to Biosemiotics: The New Biological Synthesis (1st ed.). Springer Publishing Company, Incorporated.

Bode, W., and Huber, R. (1993). Natural protein proteinase inhibitors and their interaction with proteinases. In EJB Reviews (pp. 43–61). Springer Berlin Hei- delberg.

Bouhot, L., Lieutier, F., and Debouzie, D. (1988). Spatial and temporal distribution of attacks by Tomicus piniperda L. and Ips sexdentatus Boern. (Col., Scolyti- dae) on Pinus sylvestris. Journal of Applied Entomology, 106(1-5), 356–371.

Brioschi, D., Nadalini, L. D., Bengtson, M. H., Sogayar, M. C., Moura, D. S., and Silva-Filho, M. C. (2007). General up regulation of Spodoptera frugiperda trypsins and chymotrypsins allows its adaptation to soybean proteinase inhibitor. Insect Biochemistry and Molecular Biology, 37(12), 1283–1290.

Broehan, G., Arakane, Y., Beeman, R. W., Kramer, K. J., Muthukrishnan, S., and Merzendorfer, H. (2010). Chymotrypsin-like peptidases from Tribolium castaneum: A role in molting revealed by RNA interference. Insect Biochemistry and Molecular Biology, 40(3), 274–283.

Burand, J. P., and Hunter, W. B. (2013). RNAi: Future in insect management. Jour- nal of Invertebrate Pathology, 112, Supplement 1, S68–S74.

Carthew, R. W., and Sontheimer, E. J. (2009). Origins and Mechanisms of miR- NAs and siRNAs. Cell, 136(4), 642–655.

22 Introduction

Chakali, G. (2005). A Hilésina do Pinheiro, Tomicus destruens Wollaston 1865 (Coleoptera-Scolytidae) em Zonas Semi-Áridas. Silva Lusitana, 13(1), 113– 124.

Chikate, Y. R., Tamhane, V. A., Joshi, R. S., Gupta, V. S., and Giri, A. P. (2013). Differential protease activity augments polyphagy in Helicoverpa armigera. Insect Molecular Biology, 22(3), 258–272. de Oliveira, C. F. R., de Paula Souza, T., Parra, J. R. P., Marangoni, S., de Castro Silva-Filho, M., and Macedo, M. L. R. (2013). Insensitive trypsins are differentially transcribed during Spodoptera frugiperda adaptation against plant protease inhibitors. Comparative Biochemistry and Physiology Part B: Bio- chemistry and Molecular Biology, 165(1), 19–25.

Faccoli, M., Piscedda, A., Salvato, P., MauroSimonato, Masutti, L., and Battisti, A. (2005). Genetic structure and phylogeography of pine shoot beetle populations (Tomicus destruens and T. piniperda, Coleoptera Scolytidae) in Italy. Annals of Forest Science, 62(4), 361-368 .

Faccoli, M. (2006). Morphological separation of Tomicus piniperda and T. destruens (Coleoptera: Curculionidae: Scolytinae): new and old characters. Euro- pean Journal of Entomology, 103(2), 433–442.

Gallego, D., Cánovas, F., Esteve, M. A., and Galián, J. (2004). Descriptive biogeography of Tomicus (Coleoptera: Scolytidae) species in Spain. Journal of Biogeography, 31(12), 2011–2024.

Gallego, D., and Galián, J. (2001). The internal transcribed spacers (ITS1 and ITS2) of the rDNA differentiates the bark beetle forest pests Tomicus destruens and T. piniperda. Insect Molecular Biology, 10(5), 415–420.

Gallego, D., and Galián, J. (2008). Hierarchical structure of mitochondrial lineages of Tomicus destruens (Coleoptera, Scolytidae) related to environmental variables. Journal of Zoological Systematics and Evolutionary Research, 46(4), 331–339.

23 Introduction

Gatehouse, J. A. (2011). Prospects for using proteinase inhibitors to protect transgenic plants against attack by herbivorous insects. Current Protein and Pep- tide Science, 12(5), 409–416.

Ge, Z.-Y., Wan, P.-J., and Han, Z.-J. (2012). Cloning and characterization of trypsin- and chymotrypsin-like genes in the striped rice stem borer, Chilo suppressalis. Genome, 55(4), 281–288.

Gomes, C. P. C., Cho, J.-H., Hood, L., Franco, O. L., Pereira, R. W., and Wang, K. (2013). A Review of Computational Tools in microRNA Discovery. Fron- tiers in Genetics, 4.

Gruden, K., Popovič, T., Cimerman, N., Križaj, I., and Štrukelj, B. (2005). Diverse Enzymatic Specificities of Digestive Proteases, “Intestains”, Enable Col- orado Potato Beetle Larvae to Counteract the Potato Defence Mechanism. Biological Chemistry, 384(2), 305–310.

Gu, L., and Knipple, D. C. (2013). Recent advances in RNA interference research in insects: Implications for future insect pest management strategies. Crop Protection, 45, 36–40.

Haack, R. A., and Lawrence, R. K. (n.d.). Tomicus Piniperda (Coleoptera: Scolyti- dae) Reproduction and Behavior on Scotch Pine Christmas Trees taken In- doors. The Great Lakes Entomologist, 30(1), 19–31.

Haq, S. K., Atif, S. M., and Khan, R. H. (2004). Protein proteinase inhibitor genes in combat against insects, pests, and pathogens: natural and engineered phy- toprotection. Archives of Biochemistry and Biophysics, 431(1), 145–159.

Barbieri, (2008) Has biosemiotics come of age? (2006). Semiotica, 2002(139), 283–295.

He, W.-Y., Zheng, Y.-P., Tang, L., Zheng, S.-C., Béliveau, C., Doucet, D., … Feng, Q.-L. (2009). Cloning, expression and localization of a trypsin-like serine protease in the spruce budworm, Choristoneura fumiferana. Insect Science, 16(6), 455–464.

Horn, A., Roux-Morabito, G., Lieutier, F., and Kerdelhue, C. (2006). Phylogeo- graphic structure and past history of the circum-Mediterranean species Tomi-

24 Introduction

cus destruens Woll. (Coleoptera: Scolytinae). Molecular Ecology, 15(6), 1603–1615.

Horn, A., Kerdelhué, C., Lieutier, F., and Rossi, J.-P. (2012). Predicting the distribution of the two bark beetles Tomicus destruens and Tomicus piniperda in Europe and the Mediterranean region. Agricultural and Forest Entomology, 14(4), 358–366.

Hui, Y. (1991a). On the bionomy of Tomicus piniperda (L.) (Col., Scolytidae) in the Kunming region of China. Journal of Applied Entomology, 112(1-5), 366–369.

Hui, Y., and Longshu, L. (1993). The distribution of Tomicus piniperda (L.) population in the crown of Yunnan pine during the shoot feeding period. Kun chong xue bao. Acta

Jin, S., Kalkum, M., Overholtzer, M., Stoffel, A., Chait, B. T., and Levine, A. J. (2003). CIAP1 and the serine protease HTRA2 are involved in a novel p53- dependent apoptosis pathway in mammals. Genes and Development, 17(3), 359–367.

Kanost, M. R., Jiang, H., and Yu, X.-Q. (2004). Innate immune responses of a lepidopteran insect, Manduca sexta. Immunological Reviews, 198(1), 97–105.

Kerdelhué, C., Roux-Morabito, G., Forichon, J., Chambon, J.-M., Robert, A., and Lieutier, F. (2002). Population genetic structure of Tomicus piniperda L. (Curculionidae: Scolytinae) on different pine species and validation of T. destruens (Woll.). Molecular Ecology, 11(3), 483–494.

Kirkendall, L. R. (2008). Description of the Yunnan shoot borer, Tomicus yunnanensis Kirkendall and Faccoli sp. n.(Curculionidae, Scolytinae), an unusually aggressive pine shoot beetle from southern China, with a key to the species of Tomicus. Zootaxa, 1819, 25–39.

Kohlmayr, B., Riegler, M., Wegensteiner, R., and Stauffer, C. (2002). Morphologi- cal and genetic identification of the three pine pests of the genus Tomicus (Coleoptera, Scolytidae) in Europe. Agricultural and Forest Entomology, 4(2), 151–157.

25 Introduction

Kozomara, A., and Griffiths-Jones, S. (2014). miRBase: annotating high confidence microRNAs using deep sequencing data. Nucleic Acids Research, 42(D1), D68–D73.

Långström, B. (1982). Life cycles and shoot-feeding of the pine shoot beetles (Re- port No. 163). Uppsala.

Lekander, B. (1971). On Blastophagus destruens woll. and a description of its larva (Coleoptera, Scolytidae). Entomologisk Tidskrift.

Lieutier, F., Ye, H., and Yart, A. (2003). Shoot damage by Tomicus sp. (Coleoptera: Scolytidae) and effect on Pinus yunnanensis resistance to subsequent reproductive attacks in the stem. Agricultural and Forest Entomology, 5(3), 227– 233.

Li, J., Wang, X.-P., Wang, M.-Q., Ma, W.-H., and Hua, H.-X. (2013). Advances in the use of the RNA interference technique in Hemiptera. Insect Science, 20(1), 31–39.

Marshall, S. D. G., Gatehouse, L. N., Becher, S. A., Christeller, J. T., Gatehouse, H. S., Hurst, M. R. H., … Jackson, T. A. (2008). Serine proteases identified from a Costelytra zealandica (White) (Coleoptera: Scarabaeidae) midgut EST library and their expression through insect development. Insect Molecular Biology, 17(3), 247–259.

Monleón, A., Blas, M., and Riba, J. M. (1996). Biology of Tomicus destruens (Wollaston, 1865) (Coleoptera: Scolytidae) in the Mediterranean forest. Elytron, 10, 161–167.

Page, M. J., and Cera, E. D. (2008). Serine peptidases: Classification, structure and function. Cellular and Molecular Life Sciences, 65(7-8), 1220–1236.

Peterson, A. T., Soberón, J., Pearson, R. G., Anderson, R. P., Martínez-Meyer, E., Nakamura, M., and Araújo, M. B. (2011). Ecological Niches and Geo- graphic Distributions (MPB-49). Princeton University Press.

Pfeffer, A. (1995). Bark and Ambrosia beetles from the central and west palaearctic region (Coleoptera, Scolytidae, Platypodidae). Entomologica Basiliensia, 17(1994), 5–310.

26 Introduction

Ritzerow, S., Konrad, H., and Stauffer, C. (2004). Phylogeography of the Eurasian pine shoot beetle Tomicus piniperda (Coleoptera: Scolytidae). European Journal of Entomology, 101(1), 13–19.

Schroeder, L. M. (1987). Attraction of the bark beetle Tomicus piniperda to Scots pine trees in relation to tree vigor and attack density. Entomologia Experi- mentalis et Applicata, 44(1), 53–58.

Senthilkumar, R., Cheng, C.-P., and Yeh, K.-W. (2010). Genetically pyramiding protease-inhibitor genes for dual broad-spectrum resistance against insect and phytopathogens in transgenic tobacco. Plant Biotechnology Journal, 8(1), 65–75.

Simpson, R. M., Newcomb, R. D., Gatehouse, H. S., Crowhurst, R. N., Chagné, D., Gatehouse, L. N., … Christeller, J. T. (2007). Expressed sequence tags from the midgut of Epiphyas postvittana (Walker) (Lepidoptera: Tortricidae). Insect Molecular Biology, 16(6), 675–690.

Soberón, J. (2007). Grinnellian and Eltonian niches and geographic distributions of species. Ecology Letters, 10(12), 1115–1123.

Srinivasan, A., Giri, A. P., and Gupta, V. S. (2006). Structural and functional diver- sities in lepidopteran serine proteases. Cellular and Molecular Biology Let- ters, 11(1), 132–154.

Terenius, O., Papanicolaou, A., Garbutt, J. S., Eleftherianos, I., Huvenne, H., Kanginakudru, S., … Smagghe, G. (2011). RNA interference in Lepidoptera: An overview of successful and unsuccessful studies and implications for experimental design. Journal of Insect Physiology, 57(2), 231–245.

Tyndall, J. D. A., Nall, T., and Fairlie, D. P. (2005). Proteases Universally Recog- nize Beta Strands In Their Active Sites. Chemical Reviews, 105(3), 973– 1000.

Valencia-Sanchez, M. A., Liu, J., Hannon, G. J., and Parker, R. (2006). Control of translation and mRNA degradation by miRNAs and siRNAs. Genes and De- velopment, 20(5), 515–524.

27 Introduction

Vasconcelos, T., Horn, A., Lieutier, F., Branco, M., and Kerdelhué, C. (2006). Dis- tribution and population genetic structure of the Mediterranean pine shoot beetle Tomicus destruens in the Iberian Peninsula and Southern France. Agri- cultural and Forest Entomology, 8(2), 103–111

Vinokurov, K. S., Elpidina, E. N., Oppert, B., Prabhakar, S., Zhuzhikov, D. P., Dunaevsky, Y. E., and Belozersky, M. A. (2006). Diversity of digestive proteinases in Tenebrio molitor (Coleoptera: Tenebrionidae) larvae. Compara- tive Biochemistry and Physiology Part B: Biochemistry and Molecular Biol- ogy, 145(2), 126–137.

Wei, Z., Yin, Y., Zhang, B., Wang, Z., Peng, G., Cao, Y., and Xia, Y. (2007). Cloning of a novel protease required for the molting of Locusta migratoria manilensis. Development, Growth and Differentiation, 49(7), 611–621.

Wiens, J. J., Ackerly, D. D., Allen, A. P., Anacker, B. L., Buckley, L. B., Cornell, H. V., … Stephens, P. R. (2010). Niche conservatism as an emerging principle in ecology and conservation biology. Ecology Letters, 13(10), 1310–1324.

Wu, W., Ren, Q., Li, C., Wang, Y., Sang, M., Zhang, Y., and Li, B. (2013). Charac- terization and Comparative Profiling of MicroRNAs in a Sexual Dimor- phism Insect, Eupolyphaga sinensis Walker. PLoS ONE, 8(4).

Xia, L., Zhen, Z., Hongbin, W., Wei, W., Peng, C., PeiYi, Z., and others. (2010). Tomicus armandii Li and Zhang (Curculionidae, Scolytinae), a new pine shoot borer from China. Zootaxa, 2572, 57–64.

Yu, N., Christiaens, O., Liu, J., Niu, J., Cappelle, K., Caccia, S., … Smagghe, G. (2013). Delivery of dsRNA for RNAi in insects: an overview and future di- rections: Delivery of dsRNA for RNAi in insects. Insect Science, 20(1), 4–14.

Zavala, J. A., Giri, A. P., Jongsma, M. A., and Baldwin, I. T. (2008). Digestive Duet: Midgut Digestive Proteinases of Manduca sexta Ingesting Nicotiana attenuata with Manipulated Trypsin Proteinase Inhibitor Expression. PLoS ONE, 3(4), e2008.

Zhang, C., Zhou, D., Zheng, S., Liu, L., Tao, S., Yang, L., … Feng, Q. (2010). A chymotrypsin-like serine protease cDNA involved in food protein digestion

28 Introduction

in the common cutworm, Spodoptera litura: Cloning, characterization, developmental and induced expression patterns, and localization. Journal of In- sect Physiology, 56(7), 788–799.

Zhu, K. Y. (2013). RNA interference: A powerful tool in entomological research and a novel approach for insect pest management. Insect Science, 20(1), 1–3.

Zhu-Salzman, K., and Zeng, R. (2015). Insect Response to Plant Defensive Pro- tease Inhibitors. Annual Review of Entomology, 60(1), 233–252.

Zibaee, A., and Malagoli, D. (2014). Immune response of Chilo suppressalis Walker (Lepidoptera: Crambidae) larvae to different entomopathogenic fungi. Bulletin of Entomological Research, 104(02), 155–163.

29 Introduction

Introduction

33 Planning,Planning, objectivesobjectives andand hypotheseshypotheses

31 Planning, objectives and hypotheses

The overall objective of this thesis is to make an approach to the study of bark beetles from different perspectives. The study has allow to identify patterns of differen- tiated niche in populations of Tomicus destruens in the Mediterranean basin, to analyze differences in the expression of microRNAs in T. yunnanensis and T. destruens, to characterize possible serine proteases in T. yunnanensis and to analyze ecological networks where bark beetles participate.

To achieve this, the study is structured in 4 chapters, one based on phylogeographic analysis and ecological aspects, two chapters analyzing genomic aspects and one chapter focussed on a biosemiotic perspective.

In the first chapter, factors of environmental niche affecting the distribution of genetic diversity of T. destruens using phylogeographic analysis and modelling to detect niche segregation processes along the entire Mediterranean basin are identified. In the second chapter, two microRNAs are identified in T. yunnanensis and their expression is validated in this species and in T. destruens. In chapter three, possible serine proteases, not known to date, are identified and characterized in the transcriptome of T. yunnanensis. The identified microRNAs and the serine proteases characterized may be used in pest control strategies of the genus Tomicus and other bark beetles. Finally, in chapter four, signals involved in the intraspecific communication of forestry beetles, and with other forestry species, are analysed from the point of view of biosemiotics and the concepts of eco-field and the semiotic niche.

The results of the thesis is structured in four chapters, which correspond to an article published in the journal Organisms Diversity and Evolution (Impact Factor 2.888 and located in Q1 the area of Zoology) and three articles in preparation to be sent to journals also included in the Journal Citation Report. The objectives and hypotheses of each chapter are detailed below.

32 Planning, objectives and hypotheses

3.1 Chapter 1. Distribution of Tomicus destruens (Coleoptera: Scolytinae) mitochondrial lineages: phylogeographic insights and niche modelling.

The objective of this chapter was to identify some environmental factors of the ecological niche that affect the distribution of the mitochondrial lineages (mtLs) and the genetic diversity of T. destruens using nested clade analysis (NCA) together with the maximun entropy (MaxEnt) algorithm (NCA-MaxEnt) in order to detect processes of ecological niche segregation of the mitochondrial DNA (mtDNA) lineages along the entire Mediterranean Basin.

To reach this objective, we applied a combined methodology for modelling distribution and explored the ecological niche of hierarchical lineages of T. destruens on a Mediterranean scale.

The hypothesis indicates that habitat heterogeneity allows the fixation of mtLs in T. destruens. The predictions of the hypothesis suggests that mitochondrial haplotypes are structured geographically according to environmental variables (climate, topography and host) and therefore the regions that share the same values of those variables will be occupied by the same haplotypes or clades.

3.2 Chapter 2. In Silico Prediction and Characterization of MicroRNAs from the bark beetle Tomicus yunnanensis and validation in T. destruens (Coleoptera, Scolytinae)

The main objective of this chapter is to identify and characterize in silico microR- NAs from Tomicus yunnanensis and to validate their expression in T. yunnanensis and T. destruens.

To achieve this goal, we have identified new microRNAs (miRNAs) in the genomic library of the transcriptome of a pool of individuals of T. yunnannensis and to validate

33 Planning, objectives and hypotheses

7 of the identified miRNAs in this species and in the congeneric Tomicus destruens. This study provides a foundation for a further assessment of miRNA to be used in pest control strategies in the genus Tomicus and other bark beetles.

The hypothesis suggests that the miRNAs identified in T. yunnanensis through bioinformatics tools are also present in the congeneric species T. destruens. The hypothesis predicts that the primers designed to amplify in T. yunnanensis allow the amplification of the same miRNA in T. destruens. The underlying hypothesis is that genes that regulate physiological processes associated with reproduction, have a sex-biased expression and therefore are expected to show differences between males and females.

3.3 Chapter 3. Transcriptome analysis and in silico identification and characterization of serine proteases in the bark beetle Tomicus yunnanensis

The main goal of this study was to analyse in silico the transcriptome of the bark beetle T. yunnanensis by bioinformatic tools, and identify potential candidate genes for the synthesis of serine proteases. We intended to provide information on specific markers (digestion enzymes) to be used in future studies, aiming to develop new control strategies of this insect pest and its congeneric species.

To achieve this goal, we have identified new serine proteases in the genomic library of the transcriptome of a pool of individuals of T. yunnannensis. Four trypsin-like proteases and five chymotrypsin-like proteases were annotated in silico and other four contigs had clip domain. This study provides a foundation for a further assessment of serine proteases to be used in pest control strategies in the genus Tomicus and other bark beetles.

The hypothesis suggests that the candidate serine proteases identified in the T. yunnanensis cDNA library through bioinformatic tools performing similarity searches against annotated transcriptomes in the database, also meet the requirements already

34 Planning, objectives and hypotheses known for other serine proteases in insects. These characteristics are having a conserved RIVGG propeptide-processing site, a conserved catalytic triad region in the protein sequence and a substrate binding pocket region. A subjacent hypothesis postu- lates that serine proteases of T. yunnanensis are homologues to serine proteases of other beetles, deriving from a common ancestor.

3.4 Chapter 4. Applying eco-field and the General Theory of Resources to bark beetles: Beyond the Niche Construction Theory

The objective of this chapter was to identify the signals of the semiotic niche that affect intraspecific communication of bark beetles and with other organisms using the eco-field concept together with the General Theory of Resources (GTR) in order to detect processes of expansion of the semiotic niche in bark beetles along the entire landscape forestry matrix.

To reach this objective, we applied a semiotic methodology to find the main signals and to explore the semiotic niche of the life cycle of bark beetles species.

The hypothesis indicates that bark beetles are the centre of biodiversity changes in the forestry landscape. The predictions of the hypothesis suggest that bark beetles are a keystone in maintaining the community assembly by the interchange of energy, matter and information over time through the forestry landscape matrix. The alternative hypothesis explains that there are several species that let the community assembly in a forestry landscape. For that, the interchange of signals allows a strong network over time by sharing eco-field networks among different species.

35 Planning, objectives and hypotheses

36 Planning, objectives and hypotheses

44 PlanteamientoPlanteamiento hipótesishipótesis yy objetivosobjetivos

37 Planteamiento hipótesis y objetivos

En el primer capítulo se identifican factores ambientales del nicho ecológico que afectan a la distribución de la diversidad genética de T. destruens utilizando análisis filogeográficos y de modelado de nicho para detectar procesos de segregación a lo largo de toda la cuenca mediterránea. En el segundo capítulo se identifican dos mi- croRNA en T. yunnanensis cuya expresión es validada en esta especie y en T. destruens. En el capítulo tres se identifican y caracterizan posibles serinas proteasas no conocidas hasta el momento, presentes en el transcriptoma de T. yunnanensis. Los mi- croRNAs y serina proteasas identificadas podrán ser utilizados en estrategias de control de plagas en el género Tomicus y otros escarabajos de corteza. Y finalmente, en el capítulo cuatro se analizan las señales que intervienen en la comunicación de los escarabajos forestales, entre ellos y con otras especies forestales desde el punto de vista de la biosemiótica y aplicando los conceptos de eco-field y nicho semiótico.

Los resultados de la tesis están estructurados en 4 capítulos, los cuales se corres- ponden con un artículo publicado en la revista Organisms Diversity and Evolution (Factor de Impacto 2.888 y situada en Q1 del área de Zoología) y tres en preparación para ser enviados a revistas también incluidas en el Journal Citation Report. Los objetivos e hipótesis de cada capítulo se detallan a continuación.

38 Planteamiento hipótesis y objetivos

4.1 Capítulo 1. Distribución de linajes mitocondriales de Tomicus destruens (Coleoptera: Scolytinae): enfoque filogeográfico y modelado de nicho.

El objetivo de este capítulo fue identificar algunos factores ambientales del nicho ecológico que afecta a la distribución de los linajes mitocondriales (mtLs) y la diversidad genética de T. destruens utilizando análisis de clados anidado (NCA), junto con la el algoritmo de máxima entropía (MaxEnt) (NCA-MaxEnt) a fin de detectar procesos de segregación de nicho ecológico de los linajes del ADN mitocondrial (AD- Nmt) a lo largo de toda la cuenca mediterránea.

Para alcanzar este objetivo, se aplicó una metodología combinada para la distribu- ción de modelado y exploramos el nicho ecológico de los linajes jerárquicos de T. destruens en una escala mediterránea.

La hipótesis indica que la heterogeneidad de hábitat permite la fijación de los linajes mitocondriales (mtLs) en T. destruens. Las predicciones de la hipótesis indican que los haplotipos mitocondriales están estructurados geográficamente de acuerdo a variables ambientales (climáticas, topográficas y de hospedador) y por tanto las regiones que comparten los mismos valores de dichas variables estarán ocupados por los mismos haplotipos o clados.

4.2 Capítulo 2. Predicción in silico y caracterización de microRNAs del escarabajo de corteza Tomicus yunnanensis y validación en T. destruens (Coleoptera, Scolytinae)

El objetivo principal de este capítulo es identificar y caracterizar in silico microR- NAs de Tomicus yunnanensis y validar su expresión en T. yunnanensis y T. destruens.

Para alcanzar este objetivo, se han identificado nuevos microRNAs (miRNAs) en la biblioteca genómica del transcriptoma de un conjunto de individuos de T. yunnannen-

39 Planteamiento hipótesis y objetivos sis y se validaron 7 de los miRNAs identificados en esta especie y en la especie con- genérica Tomicus destruens. Este estudio proporciona una base para la evaluación de estos miRNA para ser utilizada en estrategias de control de plagas en el género Tomicus y otros escarabajos de corteza.

La hipótesis indica que los miRNAs identificados en T. yunnanensis mediante herramientas bioinformáticas también están presentes en la especie congenérica T. destruens. La hipótesis predice que los cebadores diseñados para T. yunnanensis permiten amplificar los mismos miRNA en T. destruens. La hipótesis subyacente indica que al regular genes que controlan procesos fisiológicos relacionados con la reproduc- ción, presentan una expresión condicionada por el sexo y por lo tanto se espera encontrar diferencias entre machos y hembras.

4.3 Capítulo 3. Análisis in silico del transcriptoma y identificación y caracterización de proteasas de serina en el escarabajo de la corteza Tomicus yunnanensis

El objetivo principal de este estudio fue analizar in silico el transcriptoma del escarabajo de corteza T. yunnanensis mediante herramientas bioinformáticas, e identificar posibles genes candidatos para la síntesis de serina proteasas. El trabajo intenta proporcionar información sobre marcadores específicos (enzimas de digestión) que puedan ser utilizados en futuros estudios, con el objetivo de desarrollar nuevas estrategias de control de esta especie de insectos plaga y otras especies del mismo género.

Para lograr este objetivo, se han identificado nuevas serina proteasas en la biblioteca genómica del transcriptoma de un conjunto de individuos de T. yunnannensis. Cuatro proteasas semejantes a tripsina y cinco proteasas semejantes a quimotripsina fueron anotados in silico; además se anotaron otros cuatro contigs que mostraban un dominio clip. Este estudio proporciona información básica para una posterior evalua- ción de las serina proteasas como herramientas de control de plagas en el género

40 Planteamiento hipótesis y objetivos

Tomicus y otros escarabajos de corteza.

La hipótesis sugiere que los genes candidatos identificados como serina proteasas en la biblioteca de ADNc de T. yunnanensis, a través de herramientas bioinformáticas que realizan búsquedas de similitud en transcriptomes anotados de la base de datos, también cumplen con los requisitos ya conocidos de las serina proteasas en insectos. Estas características son la existencia de un sitio conservado RIVGG de procesamiento del propéptido, una región conservada en la secuencia de la proteína conocida como tríada catalítica y una región de unión al sustrato. Una hipótesis subyacente postula que las serina proteasas de T. yunnanensis son homólogas a las serina proteasas de otros coleópteros, y por tanto derivan de un ancestro común.

4.4 Chapter 4. Applying eco-field and the General Theory of Resources to bark beetles: Beyond the Niche Construction Theory

El objetivo de este capítulo es identificar las señales del nicho semiótico que afectan a la comunicación intraespecífica de los escarabajos de corteza y también la comunicación con otros organismos, utilizando el concepto de “eco-field” junto con la Teoría General de Recursos (GTR) con el fin de detectar los procesos de ampliación del nicho semiótico en escarabajos de corteza a lo largo de la matriz del paisaje forestal.

Para alcanzar este objetivo, aplicamos una metodología semiótica para encontrar las señales principales y explorar el nicho semiótico en el ciclo de vida de las especies de escarabajos de corteza.

La hipótesis indica que los escarabajos de corteza son el centro de los cambios de la biodiversidad en el paisaje forestal. Las predicciones de la hipótesis sugieren que los escarabajos de corteza son una pieza clave en el mantenimiento del ensamblaje comunitario mediante el intercambio de energía, materia e información en el tiempo a

41 Planteamiento hipótesis y objetivos través de la matriz del paisaje forestal. La hipótesis alternativa explica que hay varias especies que permiten el ensamblaje comunitario en un paisaje forestal. Para ello, el intercambio de señales permite una red cohesiva a través del tiempo mediante el intercambio de las redes eco-field entre las diferentes especies.

42 Planteamiento hipótesis y objetivos

Planteamiento hipótesis y objetivos

55 ResúmenesResúmenes

44 Resúmenes

5.1 Capítulo 1.

Distribución de linajes mitocondriales de Tomicus destruens (Coleoptera: Scolytinae): enfoque filogeográfico y modelado de nicho.

Cytochrome oxidase Nested gene Niche shift of Cladistic oriental clade Analysis and haplotypes

Tomicus destruens mediterranean Potential populations distribution maps

Este artículo presenta un nuevo enfoque de genética de poblaciones junto con datos ambientales y biogeográficos que conduce a la obtención de inferencias para el modelado de nicho ecológico. Utilizamos linajes jerárquicos obtenidos mediante análisis cladístico anidado (NCA) de los haplotipos de ADN mitocondrial (ADNmt) del escarabajo de corteza Tomicus destruens, para modelar la distribución por máxima entro- pía, usando variables ambientales y de hospedador a lo largo de toda la cuenca medite- rránea.

Se realizaron pruebas de identidad y similitud para verificar los linajes intraespecí- ficos (clados NCA y haplotipos) con el fin de determinar un cambio o conservaduris- mo de nicho entre ellos. Además, se calcularon cuatro índices de nueve zonas geográfi- cas de la cuenca del Mediterráneo para evaluar la variabilidad de los factores ambientales que determinan la distribución de la diversidad de haplotipos a gran escala geo- gráfica. Los modelos ecológicos desarrollados indican que las linajes de ADN mito-

45 Resúmenes condrial orientales minoritarios de T. destruens difieren en su nicho ecológico potencial de acuerdo a su relación con variables climáticas extremas. Por el contrario, los linajes occidentales más ampliamente distribuidos muestran una estrecha relación con Pinus, su árbol hospedador. Además, la predicción de niveles más altos de haplotipos exclusivos y endémicas del Mediterráneo se da en las zonas con alta variabilidad de temperatura en el período húmedo. El nicho del grupo oriental parece estar incluido en parte del rango del espacio ecológico de los dos principales clados occidentales.

46 Resúmenes

5.2 Capítulo 2.

Predicción in silico y caracterización de microRNAs del escarabajo de corteza Tomicus yunnanensis y validación en T. destruens (Coleoptera, Scolytinae)

Los escarabajos de corteza del género Tomicus provocan daños importantes en los bosques de pino. T. yunnanensis es una de las plagas más importantes de al especie Pi- nus yunnanensis distribuida en el suroeste de China, y T. destruens es una plaga muy destructiva de varias especies de Pinus en la cuenca mediterránea. A continuación, pre- sentamos la identificación de microRNAs (miRNAs) a partir de una base de datos de EST (marcadores de secuencias expresadas) de T. yunnanensis.

Siete miRNAs fueron validados en ambas especies por PCR cuantitativa en tiempo real (RT-qPCR), de los cuales mir-2c-3p y mir-4944-5p mostraron expresión en ambas especies. La expresión de tyu-mir-2c-3p fue mayor en T. destruens que en T. yunnanensis, tanto en machos como en hembras. Sin embargo, se observó la máxima expre- sión de tyu-mir-4944-5p en las hembras de T. destruens, seguido por los machos de T.

47 Resúmenes yunnanensis y T. destruens. Las hembras de T. destruens mostraron expresión de mir- 2c-3p más baja que los machos, lo que implica que este miRNA podría estar relacionado con la ovogénesis y vitelogénesis. Este estudio proporciona bases para una posterior evaluación de miARNs que podrán ser utilizados en las estrategias de control de plagas en el género Tomicus y otros escarabajos de corteza.

48 Resúmenes

5.3 Capítulo 3.

Análisis in silico del transcriptoma y identificación y caracterización de proteasas de serina en el escarabajo de corteza Tomicus yunnanensis.

ALIGNMENT

SERINE PROTEASES PEPTIDE SIGNAL

CATALYTIC TRIAD

Tysp2b Tysp2a Tysp3b Tysp3b Tysp12a Tysp4a Tysp12b

Trypsin Chymotrypsin CLIP-domain 4 contigs 5 contigs 4 contigs PHYLOGENY

El transcriptoma de Tomicus yunnanensis fue analizado por medio de herramientas bioinformáticas, obteniendo 209934 contigs. Más de la mitad de los contigs (55.23 %) no mostraban significante similaridad de las secuencias a nivel de proteínas. Esto sugiere que estas proteínas quizás representas genes de rápida evolución o genes taxonó- micamente restrictivos que podrían estar relacionados con procesos de diversificación de especies. La composición de diferentes clases de enzimas digestivas fueron anotadas. Las enzimas digestivas más expresadas fueron las serina proteasas, lipasas y β- glucosidasas. Por lo tanto, estas diferencias observadas para T. yunnanensis en relación a otras especies es quizás explicada por la adaptación de este insecto a dieta basadas en coníferas con alto niveles de compuestos terpenos. Entre las serina proteasas identificadas, cuatro proteasas semejantes a tripsinas y cinco proteasas semejantes a quimotripsinas fueron anotadas in silico. Estas enzimas fueron clasificadas dentro de la familia SA1 y mostraba todas las características de serina proteasas digestivas. Además, otros cuatro contigs fueron identificados y anotados como serina proteasas pero con la falta de la región de unión a substrato y sólo dos de las regiones de la triada catalítica

49 Resúmenes

(Histidina-Ácido aspártico). Estos contigs tenían un dominio Clip, el cual está involucrado en la respuesta inmune innata y del desarrollo embrionario. Una análisis filoge- nético junto con serina proteasas representativas de varias especies de insectos fue realizada para discernir sus relaciones evolutivas. Serina proteasas de T. yunnanensis no formaban un clado congruente con la taxonomía, sino que presentaba homologías con otras especies de insectos, que esto es esperable si ellas han evolucionado por duplicaciones de genes, seguidas de divergencia debido a presiones evolutivas selectivas.

50 Resúmenes

5.4 Capítulo 4.

Aplicando el eco-field y la Teoría General de Recursos a los escarabajos de corteza: Más allá de la Teoría de la Construcción de Nicho.

Una nueva perspectiva in la ecología del paisaje es la aplicación del concepto Eco- field junto con la Teoría General de Recursos. En este artículo, nosotros describimos la existencia de un eco-field en los escarabajos de corteza como una configuración espacial con una específica portadora de significado para cada interacción de cada organismo-recurso. Los escarabajos de la corteza son insectos involucrados en cambios significativos de los paisajes forestales cuando los episodios de plaga ocurren. Estos escarabajos de la corteza son transportados por el reconocimiento de señales semióticas hacia los árboles hospedadores usando un específico eco-field. Estas señales son principalmente un grupo de olores, los cuales han sido llamado odourtope. Estas interacciones con otros organismos (hongos, bacterias, nematodos, predadores, etc.) toma lugar a tra- vés de la compartición de información relevante de las redes de eco-field (redes de representamen) en la matriz forestal. Las redes de eco-field permiten una expansión del nicho semiótico realizado de los escarabajos de la corteza hacia el nicho semiótico potencial. Además, si diferentes organismos terminan mostrando una interdependencia

51 Resúmenes del eco-field, el proceso de construcción de nicho puede ser iniciado. Posteriormente, si esta interdependencia se vuelve crucial el proceso puede llevar al establecimiento de relaciones mutualistas. Este es un ejemplo de cómo los procesos evolutivos son iniciados por el reconocimiento de las señales por medio de una red de eco-field. Además, nosotros mostramos una descripción de los eco-field de búsqueda de hospedador y de maduración en Tomicus destruens y Tomicus piniperda y una hipotética explicación de la evolución de su coexistencia simpátrica.

52 Resúmenes

6 Chapter 1

Distribution of Tomicus destruens (Coleoptera: Scolytinae) mitochondrial lineages: phylogeographic insights and niche modelling

53 Chapter 1

Cytochrome oxidase Nested gene Niche shift of Cladistic oriental clade Analysis and haplotypes

Tomicus destruens mediterranean Potential populations distribution maps

Article published in Organisms Diversity and Evolution:

Sánchez-García, F. J.1 , Galián, J.1 , and Gallego, D.1 (2015). Distribution of Tomicus destruens (Coleoptera: Scolytinae) mitochondrial lineages: phylogeographic insights and niche modelling. Organisms Diversity & Evolution, 15(1), 101-113.

1 Departamento de Zoología y Antropología Física, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain

54 Chapter 1

6.1 Abstract

This paper presents a novel approach of population genetics together with environmental and biogeographic data leading to inferences for ecological niche modelling. We used hierarchical lineages obtained using the nested cladistic analysis (NCA) of the mitochondrial DNA (mtDNA) haplotypes of the bark beetle species Tomicus destruens, for modelling the distribution by maximum entropy using environmental and host variables along the whole Mediterranean Basin. The identity and similarity tests were checked in the intraspecific lineages (NCA clades and haplotypes) in order to determine a shift or conservatism niche between them. Also, four indices from nine geographical areas in the Mediterranean Basin were calculated to assess the variability of environmental factors shaping the distribution of haplotype diversity on a large geographic scale. The ecological models developed indicate that minority eastern mtDNA lineages of T. destruens differ in their potential ecological niche according to their relation to extreme climatic variables. By contrast, the most widespread western lineages display a close relationship with their Pinus host tree. Also, higher levels of exclusive and endemic haplotypes were predicted in areas with high temperature variability in the Mediterranean wet period. The eastern group niche seems to be included in part of the range of the ecological space of the two major western clades. This result suggests that a niche shift might have started, being still an early relationship with its host tree Pinus brutia. Alternatively, the temperature variability in the wettest period appears to be related to a high proportion of endemic haplotypes of T. destruens, possibly by involving a balance between the length of the flight reproductive period of T. destruens and the status of the host tree vigour and growth stage. This study illustrates a good example of the benefits that ecological niche modelling provides to understand population genetic and phylogeographic patterns.

Keywords: Ecological niche modelling, Intraspecific lineages, NCA-MaxEnt, Scolytinae, Coleoptera, Pinus

55 Chapter 1

6.2 Introduction

Although necessarily gene-focused, phylogeographic studies provide keys to strengthen our knowledge of the associations between DNA lineages and environmental predictors, improving our understanding of the relationship between habitat diversity and competitive exclusion of phylogenetic lineages within a species (Newman et al. 2011). This approach contributes to the characterization of the ecological niche of a studied species by recording a set of representative biotic and abiotic variables (e.g. type of habitat, climatic variables, etc.) with an important role in understanding the biology and natural history of organisms (Buckley 2009) and large-scale biogeographic patterns (Wiens 2011). Recently, several studies have initiated this approach, linking ecological niche modelling with phylogeographic analysis in an attempt to answer questions related to evolutionary processes within sister species or intraspecific lineages (Graham et al. 2004; Jakob et al. 2007; Gallego and Galián 2008; McCormack et al. 2010; Hundsdoerfer et al. 2011; Pearman et al. 2010; Oney et al. 2013; Schulte et al. 2012). In these studies, correlative methods have been used to estimate the environmental conditions that are suitable for a particular species or lineage by associating known occurrence records of the species with a set of environmental variables (Pear- son 2007). However, restrictions due to sample size could limit the use of regression methods with phylogeographic datasets. Methods less dependent on sample size, such as maximum entropy modelling, have become available (Phillips et al. 2006). The widely used MaxEnt algorithm extrapolates a set of georeferenced occurrence locations of a species and a set of layers of biotic and abiotic variables, for which, maximizing the information entropy subjected to the constraints which are imposed by the prior knowledge of the occurrence points and the environmental variables (Phillips et al. 2004), producing a model of the range distribution of the given species. Using the distribution of mitochondrial lineages (mtLs) within a species instead of merely occurrence locations of the species as input, the method has great potential for its application in spatial modelling of phylogenetic datasets.

Several studies focused on the bark beetle Tomicus destruens (Wollaston, 1865)

56 Chapter 1 were developed in the last decade as an outcome of the last taxonomic separation of its sister species Tomicus piniperda (Linnaeus 1758)(Gallego and Galián 2001; Kerdelhué et al. 2002; Kohlmayr et al. 2002). T. destruens (Wollaston, 1865) exhibits a circum- Mediterranean and Macaronesian distribution (Wood and Bright 1992) and attacks severalPinus species, including Pinus halepensis, Pinus pinaster, Pinus pinea, Pinus brutia and P. canariensis and occasionally Pinus nigra (Gallego et al. 2004; Vascon- celos et al. 2006), being responsible for part of the decline of pine forests (Chakali 2005; Faccoli et al. 2005a, b). The distribution of this species is influenced mainly by host distribution (Kerdelhué et al. 2002; Horn et al. 2006), in which climatic factors related to temperature and water availability can impose a dual limitation on the scolytid species and the host trees (Gallego et al. 2004).

Phylogeographic approaches have facilitated the clarification of the origin and structure of T. destruenspopulations in the Mediterranean Basin (Faccoli et al. 2005a, b; Horn et al. 2006; Vasconcelos et al. 2006). According to Horn et al. (2006), the populations of T. destruens are phylogeographically structured into western and eastern groups, with a contact zone on the Adriatic coast of Italy. Probably, two glacial refugia existed in the western Mediterranean area (the Iberian and Italian peninsulas), where a high diversity of haplotypes and a scarce spatial structure in haplotype distribution have been found (Horn et al. 2006; Vasconcelos et al. 2006). Conversely, the eastern group was characterized by a significant phylogeographic pattern and low levels of gene flow (Horn et al. 2006). The relationship between environmental variables and the distribution of mtLs of T. destruens has been studied on a fine scale in the pine forests of Sierra Espuña (south-eastern Spain) by Gallego and Galián (2008). A phylogeographic method, such as nested clade analysis (NCA) (Templeton et al. 1995), combined with a statistical model, such as the generalized additive model (GAM), in an NCA-GAM approach was implemented to investigate in detail the role of five environmental variables in the distribution of the local genealogy. The analysis revealed a high level of genetic diversity in the forest of Sierra Espuña and a microdistribution of the mtLs related to altitude and putative niche shifts between lineages associated with the micro-environmental conditions of their host pine trees (Gallego and Galián 2008). Those authors hypothesised that habitat heterogeneity allows the fixation of mtLs in T.

57 Chapter 1 destruens.

That hypothesis, erected on the basis of a fine-scale approach, is worth testing on a broader scale. To do that, we applied a combined methodology for modelling distribution and explored the ecological niche of hierarchical lineages of T. destruens on a Mediterranean scale. Starting from this idea, the aim of this work was to identify some environmental factors of the ecological niche that affect the distribution of the mtLs and the genetic diversity of T. destruens using NCA together with the MaxEnt algorithm (NCA-MaxEnt) in order to detect processes of ecological niche segregation of the mitochondrial DNA (mtDNA) lineages along the entire Mediterranean Basin.

6.3 Methods

Haplotype dataset and beetle sampling

Sequences from two different sources were used to cover the entire Mediterranean Basin: (i) haplotype sequences from GenBank accession numbers AF457827, AF457831–AF457852, AF457854, AF457859–AF457861, DQ182709–DQ182712, DQ182714, DQ182716–DQ182731, DQ182733, DQ182734 and DQ295748– DQ295777 (Kerdelhué et al. 2002; Vasconcelos et al. 2006; Horn et al. 2006); and (ii) haplotype sequences from freshly collected insects from not sampled areas before located in central and south-eastern Iberia and the Anatolian Peninsula (Table 1). The Turkish locality fills the Anatolian sampling gap at the eastern Mediterranean area. Ad- ditionally, some samples were added from the Sierra Espuña Natural Park as a zone of high haplotypic variability (Gallego and Galián 2008). Iberian adult beetles were collected using Crosstrap® (Econex, Murcia, Spain). Adult male and female beetles were attracted using a commercial lure of α-pinene and ethanol (Econex S.L., Murcia, Spain). A cup with 100 ml of pure propylene glycol for DNA preservation (Vink et al. 2005) was attached to the bottom of the trap funnel. The Turkish samples were obtained directly by extensive collection from infested trees or wood traps. After collection, insects were immersed in absolute ethanol for subsequent DNA analysis and

58 Chapter 1 long-term preservation. Taxonomic identification of the 48 beetles was confirmed molecularly as T. destruens by the proposed method (Gallego and Galián 2001).

Table 1 List of the new populations molecularly analysed of T. destruens

DNA extraction and amplification

Because high numbers of nematodes are often found in the abdomen, only the head, thorax and legs of theTomicus beetles were used. DNA isolation was done with the Dneasy Tissue Kit (QIAGEN®, Valencia, CA, USA), following the manufacturer’s recommendations. We amplified a fragment of the mitochondrial genes cytochrome c oxidase I (COI) and II (COII) that flank the tRNleu gene, by PCR. Specific primers for T. destruens (primer pair 5′-CCTCATCATTATGAGCTATTGG-3′, 5′-TCATAGGAT- CAATATCATTG-3′; second pair 5′-TCAATAGGAGCAGTATTTGCTA-3′, 5′- AAGTAATCGTAAAGACGGAAGA-3′) were used as described (Kerdelhué et al. 2002). PCR was done in 12.5 μl reactions using Ready-To-Go PCR Beads (GE Health- care, Bucks, UK) following the PCR cycling programme described by Kerdelhué et al. (2002). PCR products were purified using the standard isopropanol/ammonium acetate method (Sambrook et al. 1989). The PCR products were sequenced directly by the dideoxy chain terminator method with the Big Dye Terminator Cycle Sequencing Read Kit and an ABI PRISM 3130 DNA sequencer (Applied Biosystems®, Carlsbad, CA, USA).

59 Chapter 1

Environmental data

In this study, we consider the entire Mediterranean Basin, comprising the areal of T. destruens, excluding the Macaronesian region, where it is considered to be an alien species (Sauvard et al. 2010). We have considered a total of 45 sites where T. destruens was present, in accordance with Horn et al. (2006) and our samples (Fig. 1). We used three environmental datasets: climatic, topographic and host data. The 19 climatic environmental variables were taken from the database WorldClim (http://www. worldclim.org) version 1.4 (Hijmans et al. 2005) (Supplementary material 1, Tables S1 and S2).

The topographic data were considered as a single variable, altitude above sea level, obtained from GTOPO30 (USGS 1996). All raster data have been used in a pixel of 10 km grid (cell size ∼ 5 arcmin), as working scale. Host variables were calculated from distribution maps of the database EUFORGEN (freely available at http://www. bioversityinternational.org/networks/euforgen (Alia and Martin 2003; Fady et al. 2003; Isajev et al. 2004; Mátyás et al. 2004) for P. halepensis, P. brutia, P. pinaster and P. sylvestris). These host distributions were taken in vectorial format and rasterized as the coverage ratio of host pine species in a pixel of working scale of 100 km2 using ArcGis v9.2 (ESRI, Redlands, CA, USA). In order to reduce the data processing time, we masked the study area by the presence distribution of all considered host tree species. The predictor dataset was included in a matrix of 26 columns (independent or predictor variables) and 54,667 rows. We used cluster analysis for correlated variables applying the Raster package (Hijmans and van Etten 2012) with R software (R Core Team 2012) to avoid collinearity. Then, highly correlated climatic and topographic variables (>0.75) with a biological meaning (Warren and Seifert 2010; Dormann et al. 2012) were chosen.

60 Chapter 1

Figure. 1. Study area and sampling plots showing the presence of Tomicus destruens

All the sequences obtained in this work and those downloaded from GenBank were aligned using Clustal W version 1.4 (Thompson et al. 1994) as implemented in Mega 4.0 (Tamura et al. 2007). The hierarchical lineages were obtained by applying an NCA on the mtDNA haplotypes obtained (Rassmann et al. 1997) by computing a statistical parsimony network that estimates gene genealogies from DNA sequences (Templeton et al. 1995), using TCS v1.21 (Clement et al. 2000). Although the nested clade phylogeographic analysis (NCPA) method is widely criticized because of the emergence of false positives using the phylogeographic inference key (Knowles and Maddison 2002; Petit and Grivet 2002; Knowles 2004; Panchal and Beaumont 2007; Petit 2008), this work used the NCA network only for defining the set of hierarchical groups of lineages (from haplotypes to clades). We used topological and frequency criteria to solve cladogram loop ambiguities (Posada and Crandall 2001). The network obtained was adjusted to that reported by Horn et al. (2006). Finally, the haplotypes were joined in different hierarchical levels to obtain the total cladogram (Templeton et al. 1995). We used the hierarchical structure defined by NCA of the 618 bp alignment to define the mtLs according to Rassmann et al. (1997), in order to investigate the significance of the spatial and environmental ecogeographical variables in their large-scale distribution.

61 Chapter 1

Modelling geographical distributions of lineages

Distribution models of mtLs were done with the MaxEnt v. 3.3.3 software (Phillips et al. 2006), which uses the maximum entropy algorithm and is very suitable for limited sample sizes (Hernandez et al. 2006; Papeş and Gaubert 2007; Pearson 2007; Wisz et al. 2008); therefore, sample units (lineages or haplotypes) present in four or fewer localities were discarded, due to the presence of strong deviations and artefacts when included in preliminary trials. The execution of the model was measured by the cross-validation method implemented in MaxEnt. Four replicates were used in all runs, which creates a different random data partition (∼ 25 % test/75 % training) for each run as described (Araújo et al. 2005). Every model was run under default options using a maximum of 2000 interactions. We selected the logistic output format, showing suitability values from 0 (unsuitable) to 1 (optimal). The jack-knife test, which is a resampling test to estimate bias in a dataset, was used to explore the primary environmental factors restricting the T. destruens geographic distribution.

Ecological comparisons between lineages

Two measures of niche overlap were used to assess similarity and equivalence: Schoener’s D and the Hellinger distance I, implemented in ENMTools software v. 1.3. (Warren et al. 2008, 2010). D and I are quantitative measures of differences in habitat suitability between two potential distribution models with values ranging from 0 (completely different niche models) to 1 (identical niche models). Firstly, niche overlap values were calculated using ecological niche modelling (ENMs) for each mtL pair.

The identity test was verified, with the null hypothesis referring to a mtL pair having equivalent ecological niches and, therefore, subject to exactly the same environmental conditions. Later, the background test checked the null hypothesis that the niches are dissimilar compared to background environments. The observed niche overlap values were compared to a null distribution of 100 pseudoreplicate values created by comparing the ENM of one mtL to an ENM created from random points drawn

62 Chapter 1 from the geographic range of the other mtLs (Warren et al. 2008). In order to determine different responses to several scales, buffered background areas (20 and 100 km radius) were developed using the Buffer function of ArcGIS around each known occurrence locality (Couvreur et al. 2011; Nakazato et al. 2010).

Relationship between haplotype diversity and environmental diversity on a large scale

Finally, in order to confirm that the variability of environmental factors influences the distribution of haplotypes on a large scale, as shown by Gallego and Galián (2008) on a fine scale, we calculated four indices of haplotype diversity (Supplementary material 1) from nine geographical areas. We studied the relation between these indices from each zone and the standard error value of each environmental predictor, using the GAM methodology explained in Supplementary material 1.

6.4 Results

Phylogeographic structure

We have characterized 23 haplotypes from our data, of which 15 (BS, BT, BU, BV, BW, ZE, ZF, ZG, ZH, ZI, ZJ, ZK, ZL, ZM and ZN) are new (Table 2) and have been deposited in the GenBank Nucleotide Sequence Database under accession numbers JN676037–JN676051. We worked with a total of 69 haplotypes in this study.

Table 2. Haplotypes (HT) found in the studied locations. The new haplotypes are shown in bold

63 Chapter 1

The reconstruction of a parsimonious haplotype network (Fig. 2) displayed a topology similar to that reported by Horn et al. (2006) but with some differences. For example, we found an alternative way to link clade Z with clade AC (connected through a new Turkish haplotype) separated also by five mutational steps. However, we prefer to keep the link between clade Z with clade AA in accord with Horn et al. (2006), who did this on the basis of the higher haplotype frequencies in ZV and AB compared to ZJ and AQ. We maintain the link between the BR haplotype and BQ instead of linking it to the AY haplotype, on the basis of the shorter geographical distance (Posada and Crandall 2001) between BR and BQ, according to the criteria used by Templeton et al. (1987) and Templeton and Sing (1993). We were not able to elaborate the fourth hierarchical level among the third level clades (option 1, 3-1 with 3-2, or option 2, 3-2 with 3-3). Then, our level 5 corresponds to the entire cladogram, which can be considered as the entire distribution of the speciesT. destruens.

64 Chapter 1

Figure 2. a Statistical parsimony network of haplotypes of Tomicus destruens from the Mediterranean samples, of a 618 bp fragment of the mitochondrial genes cytochrome c oxidase I (COI) and II (COII) that flank the tRNAleu gene. Each line corresponds to a mutational step and each empty circle to a missing intermediate; boxes indicate nested clades of increasing steps. The new haplotypes have been shaded. Dashed line shows the alternative link way of eastern and western clades. b Nested clade analysis of the clades of level 1 and above. The colored haplotypes and cladograms were used in the ecological niche analysis

The new haplotypes found in the samples from Spain belong to the western group, in particular group A, which comprises haplotypes AA and AC, both widely distributed. The Sierra Espuña site has four new haplotypes (BS, BT, BU and BW) and the AY haplotype (found so far only in northern Corsica), corroborating its highly variable

65 Chapter 1 pattern reported earlier (Gallego and Galián 2008). The Turkish samples belong to the eastern group Z, with haplotypes ZV and ZZ. The Yuntdað locality showed ten unique haplotypes (ZE, ZF, ZG, ZH, ZI, ZJ, ZK, ZL, ZM and ZN). Overall, the new haplotypes reported here are included in the hierarchical clade levels 2-1, 2-5, 2-7 and 2-8.

Distribution models of mtLs

The analysis of the contribution of environmental variables showed that the variable “coverage of Pinus pinaster” was of greatest importance, explaining the distribution of the mtL clade 3-1 as well as clade 3-2, (Supplementary material 2; Table S3). The best variable to explain the mtL clade 3-3 distribution, however, was “coverage of Pinus sylvestris”. “Coverage of Pinus sylvestris” was the second most important predictor in the models of mtL clades 3-1 and 3-2 in contrast to the “annual mean temperature” (BIO1) in clade 3-3. The third and fourth most important were “minimum temperature of coldest month” (BIO6) and “coverage of Pinus halepensis” for both mtL clades 3-1 and 3-2. Instead, “precipitation of coldest quarter” (BIO19) and “mean temperature of driest quarter” (BIO9) were the most important for mtL clade 3-3.

The distribution models of level 3 mtLs show western mtLs (clades 3-1 and 3-2) overlap in a wide area in the west Mediterranean Basin, mainly in the central Iberian Peninsula. A few spots of potential distribution of these lineages are seen on the eastern basin (Greek Peninsula and Middle East), where these western mtLs are not found. The eastern mtL clade 3-3 has a dominant presence in the eastern end of the Mediter- ranean Basin, but its potential distribution area extends to the Italian Peninsula, southwest of the Iberian Peninsula, the islands of Sicily and Sardinia and northern Africa. The overlap areas between the three lineages are scarce, focusing on the edges of contact between their respective distributions (Fig. 4a).

At the more basic mtL level (haplotypes), we found that the variable “coverage of Pinus sylvestris” had a large contribution in all haplotypes analysed. In widely distributed haplotypes (AA and AC), however, the host tree variable “coverage of Pinus pinaster” was the main contributor, followed by “coverage of Pinus halepensis”. The

66 Chapter 1 climatic variables “isothermality” (BIO3), BIO6, “mean temperature of wettest quarter” (BIO8) and “precipitation of warmest quarter” (BIO18) made only a minor contribution. By contrast, in rare haplotype models (AD, AS and ZZ), the climate variables contributed at a higher rate than host tree variables, except for haplotype AS, where the variable “coverage of Pinus halepensis” was the largest contributor. Note that this AS haplotype occurs in France, Italy and Algeria, thus, being distributed across a wide environmental range (Fig. 3a). We found extreme climatic variables BIO6 and BIO9 as well as average climatic variables BIO1, “temperature seasonality” (BIO4) and “precipitation seasonality” (BIO15) in these models. The distribution models of western rare haplotypes, AD and AS, show a wide distribution in the eastern Mediterranean Basin. In addition, a large overlap with the ZZ haplotype is shown in the Italian Peninsula, the islands of Sicily and Sardinia, North Africa and the southern Iberian Peninsula (Fig. 4b).

67 Chapter 1

Figure 3. a, b Spatial distribution of haplotypes and level 3 clades used in the maximum entropy methodology (MaxEnt)

68 Chapter 1

Figure 4. Potential distribution of the analysed lineages with a probability ≥0.40. a Clades 3 level. b Clades haplotype level

Ecological niche modelling

We found the overlap index I was always larger than D in all mtL levels analysed. For the identity test, the clade 3-3 index was significantly different from both 3-2 and 3-3 indices (Table 3, Fig. 5). The oriental haplotype ZZ was significantly different from the western haplotypes AA, AC and AD (Table 3) at the haplotype level. There- fore, the null hypothesis was rejected, and no niche equivalence was identified for clade 3-3 and the ZZ haplotype with respect to their peers on the western side.

69 Chapter 1

Table 3 Tests of niche overlap, niche equivalency (identity) and niche similarity (background) for Tomicus destruens lineages **P ≤ 0.01; *P ≤ 0.05; ns not significant (P > 0.05)

Figure 5. Tests for niche identity applied to level 3 clades about the I and D metric. a Clade 3-1 vs clade 3-3. b Clade 3-1 vs clade 3-2. c Clade 3-2 vs clade 3-3. Black lines indicate the I observed overlap values andblack dashed lines specify the D value

As for the background test, we noted that clade 3-3 was always significantly different when projected into the regions of clades 3-1 and 3-2 (Table 3, Fig. 6). Therefore, we rejected the alternative hypothesis of niche similarity between them. Actually,

70 Chapter 1 clades 3-2 and 3-1 were not significant when projected into the regions of clade 3-3. Thus, the null hypothesis of non-similarity in niches was accepted.

Figure 6. Tests for niche similarity applied to level 3 clades about the D metric. The first row shows the background test with 100 km buffer, and the second row illustrates the background test with 20 km buffer. Black dashed lines indicate the observed overlap values

The 20 km buffer showed more ecological similarity than the 100 km buffer at the mtL level. Both major mtLs (AA and AC) and rare mtLs (AD, AS and ZZ) were non- similar at any buffer level (Table 3). Even with the addition of new haplotypes in the eastern clade, the structure was maintained between the mtL clades 2-7 and 2-8 proposed by Horn et al. (2006). We found significant differences in the identity test, although it did not show differences in the indices I and D. Besides, we detected no similarity niche at any scale of buffer (P > 0.01) for the similarity test (Supplementary material 2, Table S4).

Relationship between haplotype diversity and environmental diversity on a large scale

The best variable to explain the zonal distribution of the four haplotype diversity indices was the standard error of the “mean temperature of the wettest quartet” (Fig. 7).

71 Chapter 1

From the four indices used, the model obtained for E ind and H ex was the most explicative (explained deviance = 40 and 32 %, AIC decrease = 62 and 60 %, respectively). Both models predict higher levels of exclusive and endemic haplotypes in zones with high variability of temperatures in the wet period. By contrast, the spatial distribution of H ind and H st indices was barely explained for their models.

Figure 7. Partial response curve results of GAM of diversity haplotype indices using mean values on areas of the environmental mean temperature of wettest quartet variable. a Exclusive haplotype ratio (H ex). b Standardized haplotype (H st). c Haplotypic index (H ind). d Endemicity index (E ind).

6.5 Discussion

The haplotype network obtained in this study adds new haplotypes from the eastern

72 Chapter 1 side of the Mediterranean Basin, which was less sampled in earlier studies. Moreover, samples from the Turkish locality have a high level of haplotype diversity, which increases the distribution of haplotypes ZZ and ZV. The finding of haplotypes ZV and ZZ together in western Turkey confirms the prediction made by Horn et al. (2006) of a contact zone between the two main subclades in the eastern Mediterranean area. Unex- pectedly, the new haplotype ZE is linked to haplotype ZT, which is in the eastern subclade. Taking into account its geographical location, however, it should be connected to the so-called Balkan subclade according to Horn et al. (2006). Likewise, the finding of the four new haplotypes (BS, BT, BU and BW) in Sierra Espuña corroborates the high level of diversity of this locality reported earlier (Vasconcelos et al. 2006; Gallego and Galián 2008). In Yuntdag (Turkey), 45 % of haplotypes were unreported. Further- more, South Anatolia has been considered to be a glacial refuge for several organisms (Médail and Diadema 2009; Habel et al. 2010). Thus, the high level of haplotype diversity (H ex) found in this and other areas, including the Helenic Peninsula, the Middle East, the Iberian Peninsula and North Africa (Table 4), suggests that this might well be the case for T. destruens.

Table 4. Indices of haplotype diversity on the different areas of the Mediterranean Basin

The use of highly polymorphic genetic markers allows the recognition of phylogenetic and ecologic associations, using the multiple regression model or a heuristic model (Jakob et al. 2007). Cytochrome coxidase I (COI) and II (COII) genes are sufficiently polymorphic in T. destruens (Kerdelhué et al. 2002; Vasconcelos et al. 2006;

73 Chapter 1

Horn et al. 2006; Gallego and Galián 2008) to allow the employment NCA-MaxEnt methodology for identifying relationships with ecogeographic variables. P. sylvestris coverage is presented in all models as negative biotic contribution. That is, this variable delimited the central and northern European distributions, where T. destruens is not able to develop its life cycle and where the sister species T. piniperda occurs, in accord with the distribution models proposed by Horn et al. (2012), although those authors did not use any host–tree predictor variable. The coverage of host tree species P. pinaster and P. halepensisoccurred in the models as a major contribution in the western clades 3-1 and 3-2. The same happened in all non-rare mtLs throughout all hierarchic levels (clades 1-1, 1-10, 2-1 and 2-5 and haplotypes AA and AC). Remarkably, P. brutia coverage did not have a major contribution with the occurrence of the clade 3-3 group, its natural host tree. This was in accord with Horn et al. (2006), who did not detect a significant effect of the genetic variance and the host tree, at least qualitatively. However, any of the extreme climatic variables, such as BIO19 and BIO9, works in the mtLs as the eastern clade 3-3, the subclades 2-7, 2-8, 1-18 and 1-15 and the ZZ haplotype.

The two buffer differences found in the similarity test could be explained by the increase of environmental heterogeneity when the geographic scale is increased, resulting in a higher probability of non-similarity, in accord with Nazakato et al. (2010). Our models have identified the potential niche, but never the fundamental niche, which strictly should be set according to the physiological and biophysical principles of the organism (Kearney et al. 2010; Araújo and Peterson 2012).

Our results support the suggestion that the niche of the eastern lineage is different compared to the western lineages; however, niches of western lineages were not different compared to the eastern lineage. The eastern group (3-3) has a potential niche related to extreme weather conditions, i.e. a type of specialization. In turn, the eastern niche (or a major part of it) would be encompassed within the range of the ecological space of the two major clades (3-1 and 3-2).

The question arises as to why this poor association exists between P. brutia and lin-

74 Chapter 1 eage 3-3, when exactly the opposite was expected. One hypothetic explanation would be that the eastern clade was split from the western group. This could be understood under the framework of the ecological niche theory (Peterson et al.2011), as a process of niche shifting among basal lineages (AD, AS and ZZ) vs major lineages, by competitive exclusion (AD and AS) or vicariance (ZZ). These lineages would be adapted to extreme temperature and precipitation, or chemical conditions imposed by their host trees (Blanch et al. 2009). Therefore, they were able to colonize the eastern Mediter- ranean. According to Horn et al. (2006), some lineages of T. destruens were excised from the western basin and spread to the eastern Mediterranean Basin after the Pleisto- cene, something which happened with other organisms (Hewitt 1996). This could be the case of ZZ lineage, which diverged over time from the western group. Our results suggest that a niche shift might have started, being still at an early relationship with its host tree P. brutia, with insufficient time to develop a host-dependent genetic structure. However, the only suitable host tree species in the Middle eastern area is P. brutia. Taking into account that P. halepensis and P. brutia were a vicariant species pair, with several ecological and genetic similarities (Barbéro et al. 1998 and references therein), the absence of an alternative host could have induced a low level of host specialization. That is, the hypothetical ancestral ZZ group was pre-adapted to attack the vicariant host tree, P. brutia, in the new eastern environment. The alternative hypothesis would be that the differences between the clades are due to differences in the ecogeographic space available (Godsoe 2010), and niche shifts result from the anisotropy of ecologic space (Soberón and Peterson 2011). The major contribution of the distribution models of mtLs, however, depends on prior distribution of samples (haplotypes in our case). The major haplotypes (AA and AC) cover a broader environmental spectrum in their niche than the rare lineage ZZ, thereby increasing the sampling in the geographic area overlapping the three lineages (clades 3-1, 3-2 and 3-3), i.e. the Italian Peninsula could diminish the similarity significant values found in the niche tests.

Furthermore, the Iberian Peninsula constitutes the largest and most continuous distribution area for T. destruens, containing all host pine taxa except P. brutia and a large environmental variability that permits generating a high level of haplotype diversity. Other areas with this tendency were North Africa, the Peloponnese and Anatolian Pen-

75 Chapter 1 insula and the Middle East. Our results indicate a high level of variability in the areas with the highest temperatures of the wettest month that is related directly with the higher rates of endemic haplotypes of T. destruens. Accordingly, we hypothesise that T. destruens lineages tend to reduce the overlapping of their realised niche on a large scale by selection of different temperature ranges in the wetter quartet. In the Mediter- ranean climate, the most important rainfall occurs in spring and autumn, and the smooth temperature allows the host pines to develop their vegetative period, which is interrupted in the coldest weeks of winter. This coincides over time with the reproductive flight of T. destruens as has been widely reported (Gallego et al. 2004; Vasconcelos et al. 2006; Chakali 2005; Faccoli et al. 2005a, b). Therefore, the reproductive attacks of T. destruens overlap with the vegetative period of their host trees. Accordingly, high temperatures in autumn permit the extension of the reproductive behaviour of T. destruens, but high temperatures in spring tend to shorten their flight. High temperatures in autumn and winter, however, enable a long growth period of the host trees. So, in areas with highly variable temperature, the selection pressure of these processes could allow the fixation of endemic haplotypes, possibly by reaching a balance between the length of the flight reproductive period of T. destruens and the status of host tree vigour during their growth stage. In conclusion, temperature variability in the wettest period appears to be related to a high proportion of endemic haplotypes of T. destruens in the Mediterranean Basin.

Our proposed methodology (NCA-MaxEnt) emerges from the phylogeographic framework proposed by Horn et al. (2006) as a new approach for modelling the geographic lineage distribution from an NCA on a meso or large geographic scale (the Mediterranean Basin). This represents an improvement of the methodology proposed by Gallego and Galián (2008), which includes analysing T. destruens on a fine scale. By contrast, our work improves the work reported by Horn et al. (2012) using general linear models GAM, for two reasons: (1) we introduce the biotic variables in the modelling instead of using the intersection between distribution maps of the beetles and the geographic distribution of the host pine, which gives a more accurate prediction; and (2) we analyse the contribution of bioclimatic variables to the distribution of intraspecific clades and haplotypes. Finally, paleodistribution modelling should be explored in

76 Chapter 1 the future to detect whether lineages of the eastern group have had a more widespread past distribution. Comparing the results of past and present distributions and time-calibrated phylogenies would help to determine the scope of niche shifts of T. destruens lineages over time (Wiens 2011).

Acknowledgments

We thank Ö. Toprak, José Luís Lencina and Eudaldo Gómez-Rosa for collecting samples and Carmelo Andújar and Carlos Ruiz for their assistance with the script and MaxEnt methodology. Finally, Obdulia S. Sanchez-Domingo and Ana I. Asensio are also thanked for their help in the laboratory. This work was financed by the Fundación Séneca Project (reference 12023/PI/09) of the Murcia Regional Government.

References

Alia, R., & Martin, S. (2003). EUFORGEN technical guidelines for genetic conservation and use for maritime pine (Pinus pinaster). Rome: International Plant Genetic Resources Institute.

Araújo, M. B., & Peterson, A. T. (2012). Uses and misuses of bioclimatic envelope modeling. Ecology, 93, 1527–1539.

Araújo, M. B., Pearson, R. G., Thuiller, W., & Erhard, M. (2005). Validation of species–climate impact models under climate change. Global Change Bio- logy, 11, 1504–1513.

Barbéro, M., Loisel, R., P. Quézel, Richardson, D. M., & Romane, F. Pines of the Mediterranean basin. In D. M. Richardson (Ed.), Ecology and biogeography of Pinus (pp. 153–170). Cam- bridge: Cambridge University Press.

Blanch, J.-S., Peñuelas, J., Sardans, J., & Llusià, J. (2009). Drought, warming and soil fertilization effects on leaf volatile terpene concentrations in Pinus halepensis and Quercus ilex. Acta Physiologiae Plantarum, 31, 207–218.

77 Chapter 1

Buckley, D. (2009). Toward an organismal, integrative, and iterative phylogeography. BioEssays, 31, 784–793.

Chakali, G. (2005). A Hilésina do Pinheiro, Tomicus destruens Wollaston 1865 (Coleoptera-Scolytidae) em Zonas Semi-Áridas. Silva Lusitana, 13, 113– 124.

Clement, M., Posada, D., & Crandall, K. A. (2000). TCS: a computer program to estimate gene genealogies.Molecular Ecology, 9, 1657–1659.

Couvreur, T. L., Porter-Morgan, H., Wieringa, J. J., & Chatrou, L. W. (2011). Little ecological divergence associated with speciation in two African rain forest tree genera. BMC Evolutionary Biology, 11, 296.

Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J. R. G., Gruber, B., Lafourcade, B., Leitão, P. J., Münkemüller, T., McClean, C., Osborne, P. E., Reineking, B., Schröder, B., Skidmore, A. K., Zurell, D., & Lautenbach, S. (2012). Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography, 35, 1–20.

Faccoli, M., Piscedda, A., Salvato, P., Simonato, M., Masutti, L., & Battisti, A. (2005a). Genetic structure and phylogeography of pine shoot beetle populations (Tomicus destruens and T. piniperda, Coleoptera Scolytidae) in Italy. Annals of Forest Science, 62, 361–368.

Faccoli, M., Battisti, A., & Masutti, L. (2005b). Phenology of Tomicus destruens (Wollaston) in northern Italian pine stands. In F. Lieutier & D. Ghaioule (Eds.), Entomological research in Mediterranean forest ecosystems(pp. 185– 193). Paris: Institut National Reserche Agronomique Editions.

Fady, B., Semerci, H., & Vendramin, G. (2003). EUFORGEN technical guidelines for genetic conservation and use for Aleppo pine (Pinus halepensis) and brutia pine (Pinus brutia). Rome: International Plant Genetic Resources Insti- tute.

Gallego, D., & Galián, J. (2001). The internal transcribed spacers (ITS1 and ITS2) of the rDNA differentiates the bark beetle forest pests Tomicus destruens and T. piniperda. Insect Molecular Biology, 10, 415–420.

78 Chapter 1

Gallego, D., & Galián, J. (2008). Hierarchical structure of mitochondrial lineages of Tomicus destruens(Coleoptera, Scolytidae) related to environmental variables. Journal of Zoological Systematics and Evolutionary Research, 46, 331–339.

Gallego, D., Cánovas, F., Esteve, M. A., & Galián, J. (2004). Descriptive biogeography of Tomicus (Coleoptera: Scolytidae) species in Spain. Journal of Biogeography, 31(12), 2011–2024.

Godsoe, W. (2010). Regional variation exaggerates ecological divergence in niche models. Systematic Biology, 59, 298–306.

Graham, C. H., Ron, S. R., Santos, J. C., Schneider, C. J., & Moritz, C. (2004). In- tegrating phylogenetics and environmental niche models to explore speciation mechanisms in dendrobatid frogs. Evolution, 58, 1781–1793.

Habel, J. C., Drees, C., Schmitt, T., & Assmann, T. (2010). Review refugial areas and postglacial colonizations in the western Palearctic. In D. J. C. Habel & P. D. T. Assmann (Eds.), Relict species (pp. 189–197). Berlin: Springer.

Hernandez, P. A., Graham, C. H., Master, L. L., & Albert, D. L. (2006). The effect of sample size and species characteristics on performance of different species distribution modeling methods. Ecography, 29, 773–785.

Hewitt, G. M. (1996). Some genetic consequences of ice ages, and their role in divergence and speciation.Biological Journal of the Linnean Society, 58, 247– 276.

Hijmans, R. J., & Van Etten, J. (2012). Geographic analysis and modeling with raster data. url http://cran.r-project.org/web/packages/raster/raster.pdf.

Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G., & Jarvis, A. (2005). Very high resolution interpolated climate surfaces for global land areas. Interna- tional Journal of Climatology, 25, 1965–1978.

Horn, A., Roux-Morabito, G., Lieutier, F., & Kerdelhue, C. (2006). Phylogeo- graphic structure and past history of the circum-Mediterranean species Tomicus destruens Woll. (Coleoptera: Scolytinae). Molecular Ecology, 15, 1603–1615.

79 Chapter 1

Horn, A., Kerdelhué, C., Lieutier, F., & Rossi, J.-P. (2012). Predicting the distribution of the two bark beetlesTomicus destruens and Tomicus piniperda in Europe and the Mediterranean region. Agricultural and Forest Entomology, 14, 358–366.

Hundsdoerfer, A. K., Mende, M. B., Kitching, I. J., & Cordellier, M. (2011). Tax- onomy, phylogeography and climate relations of the Western Palaearctic spurge hawkmoth (Lepidoptera, Sphingidae, Macroglossinae).Zoologica Scripta, 40, 403–417.

Isajev, V., Fady, B., Semerci, H., & Andonovski, V. (2004). EUFROGEN technical guidelines for genetic conservation and use of European black pine (Pinus nigra). Rome: International Plant Genetic Resources Institute.

Jakob, S. S., Ihlow, A., & Blattner, F. R. (2007). Combined ecological niche modelling and molecular phylogeography revealed the evolutionary history of Hordeum marinum (Poaceae)—niche differentiation, loss of genetic diversity, and speciation in Mediterranean Quaternary refugia. Molecular Eco- logy, 16, 1713–1727.

Kearney, M., Simpson, S. J., Raubenheimer, D., & Helmuth, B. (2010). Modelling the ecological niche from functional traits. Philosophical Transactions of the Royal Society, B: Biological Sciences, 365, 3469–3483.

Kerdelhué, C., Roux-Morabito, G., Forichon, J., Chambon, J.-M., Robert, A., & Lieutier, F. (2002). Population genetic structure of Tomicus piniperda L. (Curculionidae: Scolytinae) on different pine species and validation ofT. destruens (Woll.). Molecular Ecology, 11, 483–494.

Knowles, L. L. (2004). The burgeoning field of statistical phylogeography. Journal of Evolutionary Biology, 17, 1–10.

Knowles, L. L., & Maddison, W. P. (2002). Statistical phylogeography. Molecular Ecology, 11, 2623–2635.

Kohlmayr, B., Riegler, M., Wegensteiner, R., & Stauffer, C. (2002). Morphological and genetic identification of the three pine pests of the genus Tomicus (Cole-

80 Chapter 1

optera, Scolytidae) in Europe. Agricultural and Forest Entomology, 4, 151– 157.

Mátyás, C., Ackzell, L., & Samuel, C. J. A. (2004). EUFORGEN technical guidelines for genetic conservation and use for Scots pine (Pinus sylvestris). Rome: International Plant Genetic Resources Institute.

McCormack, J. E., Zellmer, A. J., & Knowles, L. L. (2010). Does niche divergence accompany allopatric divergence in Aphelocoma jays as predicted under ecological speciation?: insights from tests with niche models.Evolution, 64, 1231–1244.

Médail, F., & Diadema, K. (2009). Glacial refugia influence plant diversity patterns in the Mediterranean Basin.Journal of Biogeography, 36, 1333–1345.

Nakazato, T., Warren, D. L., & Moyle, L. C. (2010). Ecological and geographic modes of species divergence in wild tomatoes. American Journal of Botany, 97, 680–693.

Newman, C. E., & Rissler, L. J. (2011). Phylogeographic analyses of the southern leopard frog: the impact of geography and climate on the distribution of genetic lineages vs. subspecies. Molecular Ecology, 20, 5295–5312.

Oney, B., Reineking, B., O’Neill, G., & Kreyling, J. (2013). Intraspecific variation buffers projected climate change impacts on Pinus contorta. Ecology and Evolution, 3, 437–449.

Panchal, M., & Beaumont, M. A. (2007). The automation and evaluation of nested clade phylogeographic analysis. Evolution, 61, 1466–1480.

Papeş, M., & Gaubert, P. (2007). Modelling ecological niches from low numbers of occurrences: assessment of the conservation status of poorly known viverrids (Mammalia, Carnivora) across two continents. Diversity and Distributions, 13, 890–902.

Pearman, P. B., D’Amen, M., Graham, C. H., Thuiller, W., & Zimmermann, N. E. (2010). Within-taxon niche structure: niche conservatism, divergence and predicted effects of climate change. Ecography, 33, 990–1003.

81 Chapter 1

Pearson, R. (2007). Species’ distribution modeling for conservation educators and practitioners. Center for Biodiversity and Conservation, American Museum of Natural History.

Peterson, A.T., Soberón, J., Pearson, R.G., Anderson, R.P., Martínez-Meyer, E., Nakamura, M., Araújo, M.B.(2011). Ecological Niches and Geographic Dis- tributions. Princeton University Press, Princeton.

Petit, R. J. (2008). The coup de grâce for the nested clade phylogeographic analysis? Molecular Ecology, 17, 516–518.

Petit, R. J., & Grivet, D. (2002). Optimal randomization strategies when testing the existence of a phylogeographic structure. Genetics, 161, 469–471.

Phillips, S. J., Dudík, M., & Schapire, R. E. (2004). A maximum entropy approach to species distribution modeling. In Proceedings of the twenty-first international conference on machine learning (p. 83). New York: ACM.

Phillips, S. J., Anderson, R. P., & Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190, 231– 259.

Posada, D., & Crandall, K. A. (2001). Intraspecific gene genealogies: trees grafting into networks. Trends in Ecology & Evolution, 16, 37–45.

Rassmann, K., Tautz, D., Trillmich, F., & Gliddon, C. (1997). The microevolution of the Galápagos marine iguana Amblyrhynchus cristatus assessed by nuclear and mitochondrial genetic analyses. Molecular Ecology, 6, 437–452.

R Core Team (2012). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0 http://www.R-project.org/

Sambrook, J., Fritsch, E. F., & Maniatis, R. (1989). Molecular cloning: a laboratory manual (Vol. 3). New York: Cold Spring Harbour Press.

Sauvard, D., Branco, M., Lakatos, F., Faccoli, M., & Kirkendall, L. (2010). Weevils and bark beetles (Coleoptera, Curculionoidea). Chapter 8.2. BIOR- ISK – Biodiversity and Ecosystem Risk Assessment, 4.

82 Chapter 1

Schulte, U., Hochkirch, A., Lötters, S., Rödder, D., Schweiger, S., Weimann, T., & Veith, M. (2012). Cryptic niche conservatism among evolutionary lineages of an invasive lizard. Global Ecology and Biogeography, 21, 198–211.

Soberón, J., & Peterson, A. T. (2011). Ecological niche shifts and environmental space anisotropy: a cautionary note. Revista Mexicana de Biodiversidad, 82, 1348–1355.

Tamura, K., Dudley, J., Nei, M., & Kumar, S. (2007). MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution, 24, 1596–1599.

Templeton, A. R., Boerwinkle, E., & Sing, C. F. (1987). A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease mapping I. Basic theory and an analysis of alcohol dehydrogenase activity in Drosophila. Genetics, 117, 343–351.

Templeton, A. R., & Sing, C. F. (1993). A Cladistic Analysis of Phenotypic Associ- ations with Haplotypes Inferred from Restriction Endonuclease Mapping. IV. Nested Analyses with Cladogram Uncertainty and Recombination.Genetics, 134, 659–669.

Templeton, A. R., Routman, E., & Phillips, C. A. (1995). Separating population structure from population history: a cladistic analysis of the geographical distribution of mitochondrial DNA haplotypes in the tiger salamander, Ambystoma tigrinum. Genetics, 140, 767–782.

Thompson, J. D., Higgins, D. G., & Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research, 22, 4673–4680.

USGS (1996) US Geological Survey, Available at: http://eros.usgs.gov/#/Find_ Data/Products_and_Data_Available/gtopo30_info. Accessed 12 Mar 2010

Vasconcelos, T., Horn, A., Lieutier, F., Branco, M., & Kerdelhué, C. (2006). Distri- bution and population genetic structure of the Mediterranean pine shoot

83 Chapter 1

beetle Tomicus destruens in the Iberian Peninsula and Southern France. Agri- cultural and Forest Entomology, 8, 103–111.

Vink, C. J., Thomas, S. M., Paquin, P., Hayashi, C. Y., & Hedin, M. (2005). The effects of preservatives and temperatures on arachnid DNA. Invertebrate Sys- tematics, 19, 99–104.

Warren, D. L., & Seifert, S. N. (2010). Ecological niche modeling in MaxEnt: the importance of model complexity and the performance of model selection criteria. Ecological Applications, 21, 335–342.

Warren, D. L., Glor, R. E., & Turelli, M. (2008). Environmental niche equivalency versus conservatism: quantitative approaches to niche evolution. Evolution, 62, 2868–2883.

Warren, D. L., Glor, R. E., & Turelli, M. (2010). ENMTools: a toolbox for comparative studies of environmental niche models. Ecography, 33, 607–611.

Wiens, J. J. (2011). The niche, biogeography and species interactions. Philosoph- ical Transactions of the Royal Society, B: Biological Sciences, 366, 2336– 2350.

Wisz, M. S., Hijmans, R. J., Li, J., Peterson, A. T., Graham, C. H., Guisan, A., & Group, N. P. S. D. W. (2008). Effects of sample size on the performance of species distribution models. Diversity and Distributions, 14, 763–773.

Wood, S. L., & Bright, D. E. (1992). A catalog of Scolytidae and Platypodidae (Co- leoptera), part 2: taxonomic index. Great Basin Naturalist Memoirs, 13, 1– 1553.

84 Chapter 1

Supplementary material 1

Relationship between haplotype diversity and environmental diversity at great scale

The four indexes of haplotype diversity was used for analyzing as follows:

Exclusive haplotype ratio (Hex), an index of percentage of exclusive haplotypes from a zone

HT ex H ex= ×100 (1) HT z

where HTex= No of exclusive haplotypes of the zone, HTz= Total amount of haplotypes of the zone.

Standardized haplotype rate (Hst), percentage to the rate of haplotypes occurring in a zone in relation to the total amount of haplotypes, standardized to sample effort.

HT z×100 ( HT ) H = (2) st N ( P )

where HT= Total No of haplotypes, N= No of samples, P= No of site samples.

Haplotypic index (Hind), as a rate of each zone of haplotypes per sample effort.

HT H = z ind N (3) ( P )

Endemicity index (Eind), descriptor of the rate of exclusive haplotypes by the haplotypes found in the zone, in relation to the sample effort.

Eex

( HT z ) E = (4) ind N ( P )

85 Chapter 1

We have created a tool in the R software (R Development Core Team, 2013) that calculates and automatically assesses the quality of distribution models using modelling methodologies based on GAM (General Additive Models). The following packages were used: gam (Hastie, 1992), MASS (Veneables & Ripley, 2002), maptools (Lewin-Koh & Bivand, 2011), ROCR (Sing et al., 2005), deldir (Turner, 2009), sp (Bivand, 2008), spdep (Bivand et al. 2011), available at www.cran.r-project.org. The bidirectional Akaike’s Information Criterion (AIC) was used as selection method for the best model. For avoiding the problems of estimating responses, the uncorrelated predictors were used, by fixation of an 80% correlation. For verifying the results of the GAM, a verification was performed by the receiver operating characteristic (ROC) test (Fielding & Bell 1997). The stability of the model was calculated using a cross-validation ROC routine (Fielding & Bell, 1997), by an iterative process that subsampled the 25% of the samples and calculated the AUC value, repeated the procedure 10 times and calculated the AUC mean value. After that, it obtained de difference between the mean AUC with the AUC value obtained in the verification, considering the model as stable if the difference was lower than 10%.

References

Bivand, R. S., Pebesma, E. J. & Gómez-Rubio, V. (2008) Applied spatial data analysis with R. Springer, New York. http://www.asdar-book.org/

Bivand, R. (2011) spdep: spatial dependence: weighting schemes, statistics and models. R package version 0.5-31, URL http://CRAN. R-project. org/package= spdep.

Fielding, A. H., & Bell, J. F. (1997). A Review of Methods for the Assessment of Prediction Errors in Conservation Presence/Absence Models. Environmental Conservation, 24, 38–49.

Hastie, T. J. (1992) Generalized additive models. Statistical Models in S. (ed. by J. M. Chambers and T. J. Hastie), Wadsworth & Brooks/Cole.

86 Chapter 1

Lewin-Koh, N. J. & Bivand, R. (2010) maptools: tools for reading and handling shapefiles. R package version 0.5-2, URL http://CRAN. R-project. org/package= maptools

R Development Core Team, (2012) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Avail- able at: http://www.R-project.org.

Sing T., Sander, O., Beerenwinkel, N. & Lengauer, T. (2005) ROCR: visualizing classifier performance in R. Bioinformatics, 21, 3940–3941

Veneables, W. N. & Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth Edition. Springer, New York.

Table S1 Sample plots of Tomicus destruens species used in the analysis.

87 Chapter 1

88 Chapter 1

Table S2 Wordclim variables from Hijmans et al. (2005).

89 Chapter 1

Figure S1 Correlation cluster of bioclimatic and topographic variables. The squared variables are finally used .

90 Table S3 A heuristic estimate of the contributions of the bioclimatic variables used for modelling. Results of the jackknife analysis of variable importance are given as ranks (1 to 16) for all variables. Isolation: rank of the variable's training gain when used in isolation. Omission: rank of the variable in decreasing the total regularized training gain when omitted. Table S3 (cont.) Table S3 (cont.) Table S3 (cont.) Table S4 Tests of niche overlap, niche equivalency (identity) and niche similarity (background) for Tomicus destruens lineages. ** P≤0.01, * P≤0.05, ns, P>0.05 not significant 7 Chapter 2

In Silico prediction and characterization of mi- croRNAs from the bark beetle Tomicus yunnanensis and validation in T. destruens (Coleoptera, Curculionidae, Scolytinae) Chapter 2

Article in preparation for International Journal of Genomics:

F.J. Sánchez-García1, J-Y. Zhu2, V. Machado1, D. Gallego1 , B. Yang2, and J. Galián1, “In Silico Prediction and Characterization of MicroRNAs from the bark beetle Tomicus yunnanensis and validation in T. destruens (Coleoptera, Curculionidae, Scolytinae)” International Journal of Genomics.

1 Departamento de Zoología y Antropología Física, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain

2 Key Laboratory of Forest Disaster Warning and Control of Yunnan Province, Southwest Forestry University, Kunming 650224, China.

98 Chapter 2

7.1 Abstract

The bark beetles of the genus Tomicus provoke important damage in pine forests. T. yunnanensis is one of the notorious pest of Pinus yunnanensis distributed in southwestern China, and T. destruens is a very destructive pest of several Pinus species in the Mediterranean basin. Here, we present the identification of microRNAs (miRNAs) from a T. yunnanensis EST (Expressed sequence tags) database. Seven miRNAs were validated in both species by quantitative real time PCR (RT-qPCR), of which mir-2c- 3p and mir-4944-5p showed expression in both species. The expression of tyu-mir-2c- 3p was higher in T. destruens than in T. yunnanensis, in both males and females. How- ever, the highest expression of tyu-mir-4944-5p was observed in females of T. destruens, followed by males of T. yunnanensis and T. destruens. Females of T. destruens displayed lower expression than males of mir-2c-3p implying that this miRNA could be related to oogenesis and vitellogenesis. This study provides ground- work for a further assessment of miRNA to be used in pest control strategies in the genus Tomicus and other bark beetles.

Keywords: miRNAs, coleoptera, genomics expression, scolytinae,

7.2 Introduction

The bark beetles of the genus Tomicus live into dead or live bark from conifer trees. Tomicus yunnanensis is a newly described [1] and highly aggressive species of Pinus yunnannensis distributed only in Yunnan, Sichuan, and Guizhou Provinces in southwestern China. The congeneric species Tomicus destruens is a very destructive pest of several Pinus species in Mediterranean pine forests [2]. Recently, there have been two attempts for unravelling some genomic aspects in T. yunnanensis. The first of these [3] was a systematic bioinformatics strategy to perform functional annotation of the transcriptome data and identification of important homologues of genes involved in insecticide resistance and heat shock protein (HSP) genes associated with environmental

99 Chapter 2 stress. The second study [4] constituted a large-scale identification of the main olfactory genes from the head transcriptome.

MicroRNAs (miRNAs) are short single stranded endogenous non-coding RNAs (ncRNAs) that extensively regulate messenger RNAs (mRNAs), or others long non- coding RNAs (lcRNAs) in animals, plants, and protozoans [5]. In general, miRNAs add a new layer of control to gene expression in the post-transcriptional phase. One characteristic in common is their pleiotropic role in regulating transcripts among different places and times during the development. In animals, miRNAs are often 22-25 nucleotides (nt) in length. They regulate gene expression post-transcriptionally by forming imperfect hybrids with target sequences, mainly in the 3' – untranslated regions (UTRs). Also, it has been shown that 5' UTR and coding domain sequence (CDS) have an active role [6,7].

Attempts to discover new ways to protect crops and forest from pests have been made in the last decade. One new of these is involved in the identification of new miRNAs, followed by their oral administration to inhibit developmental genes. It has been evidenced in control of the lepidopteran pest species, Helicoverpa armigera [8]. A step forward is the creation of crops with an over-expression of the miRNAs as an alternative of the use of the Bt-toxin technology. Agrawal et al. [9] reported the design of a vector used as a transgene in tobacco plants that produced artificial microRNA tar- geted the chitinase gene of H. armigera. There is a lack of data for the identification of miRNAs in Scolytinae, a subfamily of polyphagous beetles of the family Curculionidae. This family includes 6,000 species, many of them are serious pests to conifer trees because their larvae develop under bark, and attack the vascular tissues of the plant.

Detection of new miRNAs provides a powerful tool for unravelling cellular processes involving particular gene expression of about 60 % of all protein-coding genes [10]. MicroRNA can be searched in databases of next generation sequencing (NGS) using approaches based on structural or thermodynamic characteristics, as machine learning, evolutionary conservation, phylogenetic shadowing, homology search, min-

100 Chapter 2 imal folding free energy index and statistical methods [11]. Application of this theoretical (in silico) knowledge allows identifying useful miRNAs, which will help to man- age pest species using target genes [12,13]. Some of these genes are those involved in the mevalonate pathway. This biochemical route is essential for the production of the juvenile hormone (JH) that regulates insect growth, development, metamorphosis and reproduction. Thus, manipulation of these genes could expectedly be used in pest management.

The aim of this study was to identify miRNAs in the EST database of T. yunnannensis and to validate seven of them in this species and its congeneric T. destruens. This study provides a foundation for a further assessment of miRNA to be used in pest control strategies in the genus Tomicus and other bark beetles.

7.3 Materials and Methods

Detection of miRNAs

T. yunnanensis ESTs library described by Zhu et al. [3] was downloaded from DDBJ Short Read Archive (accession number: SRA047283). Sequences were trimmed by FastQC [14] considering all position with a Phred score >20 (Fig. 1).

Sequences were assembled using the default settings by SOAP2novo (available in the Japan Supercomputer (DDBJ website, [15])). The k-mer options were calculated with kmergenie software [16]. A more restrictive K mer value (55) than that recom- mended by software (45) was used to cut down chimerical assembled contigs [17].

Assembled sequences (contigs) were subjected to Non-coding RNAs (ncRNAs) identification using Infernal v1.1.1. [18] against the Rfam database version 12 [19]. The blastn with the e-value cutoff of 10-5 in BLAST 2.2.31+ [20] was also performed against RepBase (release 19.09) [21] to delete positive hits.

101 Chapter 2

We used the ensemble machine learning method miR-Bag [22] which is based in the classifiers as Support Vector Machine (SVM), Naive Bayes (NB) and Best First Decision Trees (BFTree). The pre-miRNAs identified in at least two classifiers using Drosophila melanogaster as model were chosen.

102 Chapter 2

Figure 1. Scheme of the methodological steps followed in the detection and validation of the miRNAs in Tomicus yunnanensis and Tomicus destruens.

103 Chapter 2

To confirm the origin of these obtained pre-miRNAs, blastn and blastx analyses against the databases were performed (Blast-NCBI, [23]). Firstly, we performed blastn and blastx against the class Hexapoda. The contigs with low score (less than 49) were used for an additional analysis against the phylum Nematoda and Viruses group. Fi- nally those contigs without hits in those databases were double checked by submitting them to blastx against all database of NCBI excluding insects, viruses and nematodes to be sure of their absence in other groups of organisms.

In this paper we use the software miRdup [24] and MiRDuplexSVM [25]. These software work by identifying the most likely miRNA within a given pre-miRNA. The miRdup software is based on the random forest algorithm, which is trained with the characteristics of the duplex miRNAs (3p-5p) in the data base miRBase, which have been experimentally validated. MiRDuplexSVM works by statistical approaches that evidence that both chains (3p-5p) are complementary around a RNA loop. The algorithm support vector machine (SVM) calculates the probability of the union of both strands of the same miRNA.

Targeting

To obtain the different families of the selected miRNAs and pre-miRNA, sequences were confronted against Drosophila melanogaster miRNAs (mirBase database v21 [26]) by BLAST 2.2.31+ using the following parameters:-task blastn-short, -word size 4, -penalty -4, and -evalue 10. MicroRNAs sequences that meet established criteria, were selected and submitted to DIANA-miRPath [27] to identify those involved in the mevalonate pathway. In addition we performed the targeting using miRecords mirror- website [28] and we chose those miRNAs that were positive in at least three targeting algorithms against FPPS and HMG-S genes.

Validation of miRNAs

T. destruens individuals were collected in Spain using Crosstrap® (Econex, Spain) baited with alcohol and α-pinene (Econex, Spain) or manually in tender shoots or bark

104 Chapter 2 in forests of P. halepensis. T.yunnanensis individuals were collected from P. yunnanensis in China using the manual method (Table 1). Individuals were gender determined using the shape of the last terguite [30]. Prior dissection, all beetles were maintained at room temperature at least 24 h with tender shoots of their pine host. Dissection of male and female tissues was performed into phosphate buffer saline (PBS) buffer treated with diethypyrocarbonate (DECP). RNA was extracted with TRIzol® reagent according to the manufacturer's instructions (Invitrogen) with an additional chloroform step.

RNA was reverse-transcribed (using First Strand cDNA Synthesis Kit (Thermo Sci- entific®) following the protocol described by Chen et al. [31] and the recommendations for primers design made by Kramer [32]. We mixed 1.3μL of RNA, 1 μL stem- loop primer and 3.7 μL water during 5 min at 65ºC in the first step in order to allow dissociation of secondary structures. The chosen controls genes (Arthropod 7SK and 5,5S) were obtained from non-coding RNA analysis with RFAM [19] (Primers listed in Table S1). The RT-qPCR technique was performed in triplicate for every sample in a Rotor-Gene thermal cycler (QIAGEN®) under the following conditions: an initial hold at 95 ºC for 10 min, followed by 50 cycles of 95 ºC for 15 s, 60 ºC 1 min, followed by the melting curve (50-90 ºc) for each sample in order to check the specificity of amplification. The RT-qPCR reaction consisted of 1 μL of diluted cDNA, 7.5 μL of SYBR Green Master Mix (Applied Biosystems®) and 1 μL of 10 μM of forward and reverse primer in 15 μL total volume. The relative miRNA expression was calculated using the method of 2-∆∆Ct [33] with Arthropod 7SK and 5,5S as endogenous control and T. yunnanensis female or T. destruens females as reference sample.

7.4 Results

Bioinformatic analysis

We obtained 934795 contigs of which 99229 had a length over 200 nt, which is the ideal size for detecting pre-miRNAs. After filtering the 99229 contigs with RFAM and RepBase, 91866 contigs were used to look for pre-miRNA. We obtained 228 pre-

105 Chapter 2 miRNAs with miR-BAG software, of which 84.21 % resulted to be homologous to insect sequences after blast searches (Fig. 2). There were 71 pre-miRNAs with positive detection of miRNAs for both softwares, mirDup, and DuplexSVM (Table S1). Al- though some of them has a size longer than 150 bp, their average size was around 80 bp, (Fig. 3). The analysis of these 71 miRNAs with the two above mentioned softwares yielded different results. DuplexSVM showed a normal distribution of the size in both strands (5p and 3p) around 22 bp (varied from 17 bp to 24 bp). In contrast, mirDup tended to show a larger number of miRNAs of longer size, mainly in 3p, ranging from 13 to 30 bp (Fig. 4). Among the 71 miRNAs detected, seven that have low e-value after homology searches with D. melanogaster, and that were involved in the mevalonate pathway (FPPS and HMG-S genes) have been chosen for RT-qPCR validation.

Figure 2. The origin of the 228 pre-miRNAs of Tomicus yunnanensis using a score over 49 in blastn and blastp.

106 Chapter 2

Figure 3. Length of pre-miRNAs of Tomicus yunnanensis which a positive detection of miRNAs.

Figure 4. The detected size of Tomicus yunnanensis miRNAs in both strands (5p and 3p). A, B detected by DuplexSVM and C, D by mirDup.

107 Chapter 2

Expression analysis

Positive expression in two of the seven chosen miRNAs was found in both species of Tomicus, T. yunnanensis and T. destruens. These two miRNAs were named as tyu- mir-2c-3p and tyu-mir-4944-5p, following the nomenclature used in the best match obtained in Blast (dme-mir-2c-3p and dme-mir-4944-5p, form D. melanogaster) (Fig. 5). The expression of tyu-mir-2c-3p was higher in T. destruens than in T. yunnanensis, in both sexes (Figs. 6 and 7). However, the highest expression of tyu-mir-4944-5p was observed in females of T. destruens, followed by males of T. yunnanensis, then males of T. destruens (Fig. 7).

Figure 5. Consensus sequences of miRNAs 2 family (A) and 4944 family (B) of Tomicus yunnanensis together with miRBase. The figure was created by Weblogo v3.

108 Chapter 2

Figure 6. Expression of mirRNA 2c (A, B) and miRNA 4944 (C, D) in females and males of T. destruens and T. yunnanensis using the 5,8S as controls (T. yunnanesis female is scaled as reference).

7.5 Discussion

We have obtained 228 pre-miRNAs from the library of T. yunnanensis of which 5.26 % were of non-insect origin. This preliminary step using blastn and blastx analyses is necessary to detect and exclude false positives, or cross contamination with symbionts (nematodes, bacteria or fungi). The nature of our pre-miRNAs of viral origin could be from the viruses integrated in the insect genome as retro-transposons or lysogenic viruses. On the other hand, there are 24 (10.09 %) unknown pre-miRNAs with not enough score to be assigned to a taxonomic group, as i) they could actually be chimerical contigs without a tangible entity, or ii) they could be real pre-miRNAs of Tomicus that have not yet been described and thus, analyses were unable to find out any homology in the databases. Further analyses of these 24 pre-miRNAs will allow to discriminate between these two alternatives.

109 Chapter 2

Figure 7. Expression of mirRNA 2c (A, B) and miRNA 4944 (C, D) in females and males of T. destruens and T. yunnanensis using the 5,8S gene as controls (T. destruens female is scaled as reference).

The use of several algorithms is necessary to increase the accuracy of miRNAs predictions. The mirDup method shows high variation in the size of the predicted miRNAs (mainly 3p) (Fig. 4D). This high variation can be explained as a bias of the algorithm to detect the true end of a miRNA [34]. It also could be due to the lack of diversity of the insect miRNAs present in the arthropod database, compared to the high diversity present in mammals and other vertebrates [24, and supplementary material therein].

Our trials showed that both miRNAs, tyu-mir-2c-3p and tyu-mir-4944-5p, are expressed in both species of Tomicus. The high differences between T. destruens and T. yunnanensis, may be due to the fact that T. destruens were collected from different sources including bark (still pre-imaginal stages), shoots (mainly immature) and traps (mature individuals). The mature individuals were flying towards a host-tree, searching for a partner to mate. In contrast, T. yunnanensis samples were only collected directly

110 Chapter 2 from shoots (immatures) and bark (pre-imaginal stages), and therefore might not be sexually mature.

Regarding the expression of mir-2c-3p in T. destruens, females displayed lower expression than males. A low expression of this miRNA in females in relation to males was found in different stages of the life cycle of Anopheles stephensi [35]. Addition- ally, its miRNA was down-regulated after infection by Chikungunya Virus in a cell line of the mosquito Aedes albopictus, [36], although no differences between males and females were evaluated. Whether the expression of this miRNA in Tomicus is modulated by viruses infection or not remains to be elucidated. Sex-mediated differential expression has also been observed for esi-miR-2b and esi-miR-2b-3p, belonging to the same family as mir-2c-3p, in Eupolyphaga sinensis of the order Blattodea. These two miRNAs showed higher expression in males than in females [37]. It is suggested that down-regulation of mir-2 family might be involved in oogenesis and vitellogenesis mediated by juvenile hormone [38]. This explanation is congruent with the observed lower level of tyu-mir-2c-3p in T. destruens females compared to males, as female's physiology prepares for ovoposition.

In contrast to mir-2c, expression of mir-4944 has been scarcely studied. In the case of mir-4944, it was detected in D. melanogaster [39] and in E sinensis [37] by bioinformatic analysis, while no expression analyses were performed. In contrast, mir-4944- 3p expression analyses were performed by Lyons et al. [40], who found that miR-4944 was the most significantly down-regulated miRNA measured at -15 ºC exposed larvae in the freeze-avoiding Lepidoptera Epiblema scudderiana, thus indicating a miRNA signature for subzero survival of this insect. In the case of T. destruens in natural environmental conditions in Spain, we found a high expression in males and a much higher expression in females (134-fold). On the other hand, the expression of mir-4944-5p in T. yunnanensis collected in natural environmental conditions in China showed a higher expression in males (74-fold) than in females. Unfortunately, the lack of data on the characterization of this miRNA, does not allow deeper conclusions, but further evaluation of miR-4944 targets would be warranted to explore the relevance of its modulation (probably sex-biased) in T. destruens and T. yunnanensis.

111 Chapter 2

In conclusion, we have identified two miRNAs that were differentially expressed in males and females of two bark beetle species from different geographical areas. The data gathered via an EST database of T. yunnanensis and tested in RNA pools of T. destruens and T. yunnanensis represent the first attempt to understand how miRNA modulate physiological processes in Scolytinae. The exact mRNA transcript targets regulated by the identified miRNAs of these two species remain to be further characterized.

Acknowledgements This work was financed by Special Fund for Forest Scientific Research in the Pub- lic Interest of China (201004067), Fund of Reserve Talents for Young and Middle- Aged Academic and Technological Leaders of Yunnan Province (2013HB077) and Fundación Séneca (12023/PI/09) of the Murcia Regional Government (Spain). Thanks also to CNPq (Brasil) for granting a postdoctoral scholarship (PDE) to V.M.

Figure S1. Consensus sequences of pre- miRNAs 2 family (A) and 4944 family (B) of Tomicus yunnanensis together with miRBase. The figure was created by Weblogo v3.

112 Chapter 2

Table S1. Primers sequences used in qPCR analysis.

113 Chapter 2

Table S1. Primers sequences used in qPCR analysis (cont.) miRNA Primer Sequence 5'- 6-RT GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCA 6 CTGGATACGACATGAAG-3' 6-Forward primer 5'-CACGCACATCTTGATGGA-3' 6-Reverse primer 5'-CCAGTGCAGGGTCCGAGGTA-3' 5'- 7-RT GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCA 7 CTGGATACGACGTTCCA-3' 7-Forward primer 5'-CACGCATCTTTGGTCGA-3' 7-Reverse primer 5'-CCAGTGCAGGGTCCGAGGTA-3' Control 5_8s-Reverse primer 5'-CCGAGCAGTCTTTGAAATTGACA-3' 5,8S 5_8s-Forward primer 5'-GAACTGCAGGACACATGAACATC-3' A7sk-Forward primer 5'-AGGCTTTTTAGCTGATCTGCTCA-3' A7SK A7sk-Reverse primer 5'-CAAGAAACCACTGAATGTCACCG-3'

114 Chapter 2

Bibliography

1. L. R. Kirkendall, M. Faccoli and H. Ye “Description of the Yunnan shoot borer, Tomicus yunnanensis Kirkendall & Faccoli sp. n.(Curculionidae, Scolytinae), an unusually aggressive pine shoot beetle from southern China, with a key to the species of Tomicus,” Zootaxa, vol. 1819, pp. 25–39, 2008.

2. D. Gallego, and J. Galián, “The internal transcribed spacers (ITS1 and ITS2) of the rDNA differentiates the bark beetle forest pests Tomicus destruens and T. piniperda,” Insect Molecular Biology, vol. 10, no. 5, pp. 415–420, 2001.

3. J.-Y. Zhu, N. Zhao, and B. Yang, “Global Transcriptional Analysis of Olfactory Genes in the Head of Pine Shoot Beetle, Tomicus yunnanensis,” Interna- tional Journal of Genomics, vol. 2012, pp. E491748, 2012.

4. J.-Y. Zhu, N. Zhao, and B. Yang, “Global Transcriptome Profiling of the Pine Shoot Beetle, Tomicus yunnanensis (Coleoptera: Scolytinae),” PLoS ONE, vol. 7, no. 2, pp. e32291, 2012.

5. C. de Santi, and C. M. Greene, “The Biology of MicroRNA,” in: C. M. Greene, ed., MicroRNAs and Other Non-Coding RNAs in Inflammation, Springer In- ternational Publishing, ISBN 978-3-319-13688-2, pp. 3–19, 2015.

6. P. S. Pendergrast, and T. Volpe, “MicroRNA rules: Made to be broken,” Fron- teirs in Biology, vol. 8, no. 5, pp. 468–474, 2013.

7. E. Doxakis, “Principles of miRNA-Target Regulation in Metazoan Models,” In- ternational Journal of Molecular Sciences, vol. 14, no. 8, pp. 16280–16302, 2013.

8. B. Jayachandran, M. Hussain, and S. Asgari, “An insect trypsin-like serine protease as a target of microRNA: Utilization of microRNA mimics and inhibitors by oral feeding,” Insect Biochemistry and Molecular Biology vol. 43, no. 4, pp. 398–406, 2013.

9. A. Agrawal, V. Rajamani, V. S. Reddy, S. K. Mukherjee, and R. K. Bhatnagar, “Transgenic plants over-expressing insect-specific microRNA acquire insect-

115 Chapter 2

icidal activity against Helicoverpa armigera: an alternative to Bt-toxin technology,” Transgenic Research, vol. 24, no. 5, pp. 791–801, 2015.

10. R. C. Friedman, K. K.-H. Farh, C. B. Burge, and D. P. Bartel, “Most mammalian mRNAs are conserved targets of microRNAs,” Genome Research, vol. 19, no. 1, pp. 92–105, 2009.

11. S. Pundhir, P. Poirazi, and J. Gorodkin, “Emerging applications of read profiles towards the functional annotation of the genome,” Fronteir in Genetics, vol. 6, pp. 188, 2015.

12. Z. Rao, W. He, L. Liu, S. Zheng, L. Huang, and Q. Feng, “Identification, Ex- pression and Target Gene Analyses of MicroRNAs in Spodoptera litura,” PLoS ONE vol. 7, no. 5, pp. e37730, 2012.

13. S. Asgari, “MicroRNA functions in insects,” Insect Biochemistry and Molecu- lar Biology, vol. 43, no. 4, pp. 388–397, 2013.

14. S. Andrews, “FastQC A Quality Control tool for High Throughput Sequence Data,” Babraham Bioinformatics, http://www.bioinformatics.babraham.ac.uk/projects/fastqc/, April. 2015.

15. H. Nagasaki, T. Mochizuki, Y. Kodama, S. Saruhashi, S. Morizaki, H. Sug- awara, H. Ohyanagi, N. Kurata, K. Okubo, T. Takagi, E. Kaminuma, and Y. Nakamura, “DDBJ Read Annotation Pipeline: A Cloud Computing-Based Pipeline for High-Throughput Analysis of Next-Generation Sequencing Data,” DNA Research, vol. 20 no. 4 pp. 383–390, 2013.

16. R. Chikhi, and P. Medvedev, “Informed and automated k-mer size selection for genome assembly,” Bioinformatics, vol. 30, no. 1, pp. 31–37, 2014.

17. Y. He, Z. Zhang, X. Peng, F. Wu, and J. Wang, “De novo assembly methods for next generation sequencing data.,” Tsinghua Science and Technology, vol. 18, no. 5, pp. 500–514, 2013.

18. E. P. Nawrocki, and S. R. Eddy, “Infernal 1.1: 100-fold faster RNA homology searches,” Bioinformatics, vol. 29, no. 22, pp. 2933–2935, 2013.

116 Chapter 2

19. S. W. Burge, J. Daub, R. Eberhardt, J. Tate, L. Barquist, E. P. Nawrocki, S. R. Eddy, P. P. Gardner, and A. Bateman, “Rfam 11.0: 10 years of RNA families,” Nucleic Acids Research, vol. 41, no. D1, pp. D226–D232, 2013.

20. C. Camacho, G. Coulouris, V. Avagyan, N. Ma, J. Papadopoulos, K. Bealer, and T. L. Madden, “BLAST+: architecture and applications,” BMC Bioin- formatics, vol. 10, no. 1, pp. 421, 2009.

21. J. Jurka, V. V. Kapitonov, A. Pavlicek, P. Klonowski, O. Kohany, and J. Wa- lichiewicz, , “Repbase Update, a database of eukaryotic repetitive elements,” Cytogenetic and Genome Research vol. 110, no. 1-4, pp. 462–467, 2005.

22. A. Jha, R. Chauhan, M. Mehra, H. R. Singh, and R. Shankar, “miR-BAG: Bag- ging Based Identification of MicroRNA Precursors,” PLoS ONE vol. 7, no. 9, e45782, 2012.

23. G. M. Boratyn, C. Camacho, P. S. Cooper, G. Coulouris, A. Fong, N. Ma, T. L. Madden, W. T. Matten, S. D. McGinnis, Y. Merezhuk, Y. Raytselis, E. W. Sayers, T. Tao, J. Ye, and I. Zaretskaya, “BLAST: a more efficient report with usability improvements,” Nucleic Acids Research, vol. 41, no. W1, pp. W29– W33, 2013.

24. M. Leclercq, A. B. Diallo, and M. Blanchette, “Computational prediction of the localization of microRNAs within their pre-miRNA,” Nucleic Acids Re- search, vol. 41, no. 15, pp. 7200–7211, 2013.

25. N. Karathanasis, I. Tsamardinos, and P. Poirazi, “MiRduplexSVM: A High-Per- forming MiRNA-Duplex Prediction and Evaluation Methodology,” PLoS ONE, vol. 10, no. 5, e0126151, 2015.

26. A. Kozomara, and S. Griffiths-Jones, “miRBase: annotating high confidence microRNAs using deep sequencing data,” Nucleic Acids Research, vol. 42 no. D1 pp. D68–D73, 2014.

27. I. S. Vlachos, K. Zagganas, M. D. Paraskevopoulou, G. Georgakilas, D. Karagkouni, T. Vergoulis, T. Dalamagas, and A. G. Hatzigeorgiou, “DIANA- miRPath v3.0: deciphering microRNA function with experimental support,” Nucleic. Acids Research, vol. 43, no. W1, pp. W460–W466, 2015.

117 Chapter 2

28. F. Xiao, Z. Zuo, G. Cai, S. Kang, X. Gao, and T. Li,“miRecords: an integrated resource for microRNA–target interactions,” Nucleic Acids Research vol. 37, no. suppl 1, pp. D105–D110, 2009.

29. D. Gallego, J. Galián, J.J. Diez, and J.A. Pajares. “Kairomonal responses of Tomicus destruens (Col., Scolytidae) to host volatiles α-pinene and ethanol”. Journal of Applied Entomology vol. 132, pp. 654–662, 2008.

30. M. Faccoli, “Morphological separation of Tomicus piniperda and T. destruens (Coleoptera: Curculionidae: Scolytinae): new and old characters,” European Journal of Entomology, vol. 103, no. 2, pp. 433–442, 2006.

31. C. Chen, D. A. Ridzon, A. J. Broomer, Z. Zhou, D. H. Lee, J. T. Nguyen, M. Barbisin, N. L. Xu, V. R. Mahuvakar, M. R. Andersen, K. Q. Lao, K. J. Livak, and K. J. Guegler, “Real-time quantification of microRNAs by stem– loop RT–PCR,” Nucleic. Acids Research, vol. 33, no. 20, pp. e179–e179, 2005.

32. M. F. Kramer, “Stem-Loop RT-qPCR for miRNAs,” in: John Wiley & Sons, ed. Current Protocols in Molecular Biology, Inc., ISBN 978-0-471-14272-0, pp. 1-15, 2011.

33. K. J. Livak, and T. D. Schmittgen, “Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2−ΔΔCT Method,” Methods vol. 25, no. 4, pp. 402–408, 2001.

34. N. Karathanasis, I. Tsamardinos, and P. Poirazi, “Don’t use a cannon to kill the … miRNA mosquito,” Bioinformatics, vol. 30, no. 7, pp. 1047–1048, 2014.

35. S. Jain, V. Rana, A. Tridibes, S. Sunil, and R. K. Bhatnagar, “Dynamic expression of miRNAs across immature and adult stages of the malaria mosquito Anopheles stephensi,” Parasites & Vectors vol. 8, no. 1, pp. 179, 2015.

36. J. Shrinet, S. Jain, J. Jain, R. K. Bhatnagar, and S. Sunil, “Next Generation Se- quencing Reveals Regulation of Distinct Aedes microRNAs during Chikun- gunya Virus Development,” PLOS Neglected Tropical Diseases, vol. 8, no. 1, pp. e2616, 2014.

118 Chapter 2

37. W. Wu, Q. Ren, C. Li, Y. Wang, M. Sang, Y. Zhang, and B. Li, “Characteriza- tion and Comparative Profiling of MicroRNAs in a Sexual Dimorphism In- sect, Eupolyphaga sinensis Walker,” PLoS ONE vol. 8, no. 4, pp. e59016, 2013.

38. J. Song, W. Guo, F. Jiang, L. Kang, and S. Zhou, “Argonaute 1 is indispensable for juvenile hormone mediated oogenesis in the migratory locust, Locusta migratoria,” Insect Biochemistry and Molecular Biology, vol. 43, no. 9, pp. 879–887, 2013.

39. E. Berezikov, N. Robine, A. Samsonova, J. O. Westholm, A. Naqvi, J.-H. Hung, K. Okamura, Q. Dai, D. Bortolamiol-Becet, R. Martin, Y. Zhao, P. D. Zamore, G. J. Hannon, M. A. Marra, Z. Weng, N. Perrimon, and E. C. Lai, “Deep annotation of Drosophila melanogaster microRNAs yields insights into their processing, modification, and emergence,” Genome Research, vol. 21, no. 2, pp. 203–215, 2011.

40. P. J. Lyons, N. Crapoulet, K. B. Storey, and P. J. Morin, “Identification and profiling of miRNAs in the freeze-avoiding gall moth Epiblema scudderiana via next-generation sequencing,” Molecular and Cellular Biochemistry pp. 1–9, 2015.

119 Chapter 2

120 Chapter 2

8 Chapter 3

Transcriptome analysis and in silico identification and characterization of novel serine proteases in the bark beetle Tomicus yunnanensis

121 ALIGNMENT

SERINE PROTEASES PEPTIDE SIGNAL

CATALYTIC TRIAD

Tysp2b Tysp2a Tysp3b Tysp3b Tysp12a Tysp4a Tysp12b

Trypsin Chymotrypsin CLIP-domain 4 contigs 5 contigs 4 contigs PHYLOGENY

In preparation for Insect Science

Sánchez-García, F.J.1, Zhu , J-Y., Machado, V.1, Gallego, D.1, Liu, N-Y.2, Yang, B.2 and

Galián, J.1 (2015).Transcriptome analysis and in silico identification and characterization of novel serine proteases in the bark beetle Tomicus yunnanensis. Insect Science.

1 Departamento de Zoología y Antropología Física, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain

2 Key Laboratory of Forest Disaster Warning and Control of Yunnan Province, Southwest Forestry University, Kunming 650224, China.

122 8.1 Abstract

The transcriptome of Tomicus yunnanensis was analysed by bioinformatic tools, ob- taining 209934 contigs. More than half of the contigs (55.23 %) did not show significant sequence similarity at protein-level. It is suggested that these proteins may represent rapid evolution or taxonomically restricted genes that could be related to species diversification processes. The composition of different classes of digestive enzymes was annotated. The most expressed digestive enzymes were serine proteases, lipases and β-glucosidases. Thus, these differences observed for T. yunnanensis in relation to other species might be explained by the adaptation of this insect to a conifer-based diet with high levels of terpenic compounds. Among the serine proteases identified, four trypsin-like proteases and five chymotrypsin-like proteases were annotated in silico. These enzymes were classified into the SA1 family and showed all the characteristics of digestive serine proteases. Besides, other four contigs were identified and annotated as serine proteases but lacking the substrate binding pocket region and with only two of the regions of the triadic catalytic (Histidine-Aspartic acid). These contigs have the clip domain, which is involved in the innate immune response and the embryonic development. A phylogenetic analysis together with serine proteases representative of several insect species was performed to discern their evolutionary relationships. Serine proteases from T. yunnanensis do not form a clade congruent to the taxonomy, but rather present homologies with other insect species, what is expected if they have evolved by gene duplications, followed by divergence due to selective evolutionary pressure.

Key words Coleoptera , Serine protease homologe, Trypsin-like serine protease, Chymotrypsin-like serine protease, Transcriptome;

8.2 Introduction

The impact of bark beetles on forest resources may be increasing in the next years due to climate change and global warming. This increment of drought conditions and wildfires over the conifer trees will likely result in population increases and more beetle

123 generations per year. This would provide a strong economic shock in the rural areas around the conifer trees (Waring et al., 2009). The bark beetle Tomicus yunnanensis (Coleoptera: Scolytinae) is a pest species in Southwest China, which can provoke the death of thousands km2 of Pinus yunnanesis (Lieutier et al., 2003). The management and control of this bark beetle outbreaks in the near future are crucial. Several attempts to study genes involved in insecticide resistance and heat shock protein genes (HSP) associated with environmental stress, as well as olfactory genes (Zhu et al., 2012a, b) which can be used in pest control, have been made.

The genera Tomicus, Phloeosinus, and some species of Scolytus are peculiar among bark beetles, as the maturation (feeding) phase does not take place under the bark, but on tender shoots. Serine proteases (SPs) are key enzymes for the nourishment of all animals, as digestion of proteins provides the amino acids required for growth and development. Herbivorous insects obtain their nutrients from target plants, which often use chemical defence strategies that interfere with the insect digestion process. One of these strategies is the production of protease inhibitors (Jamal et al., 2012; Mithöfer and Bo- land, 2012; Bode et al., 2013; Joshi et al., 2013). The success of herbivorous insects is partially associated with their abilities to respond effectively to the diversity of toxic compounds produced by target plants, especially the protease inhibitors (Zhu-Salzman et al., 2003; Zhu-Salzman and Zeng, 2008). Several studies have showed the presence of protease inhibitors in plant species attacked by insects, often showing a coevolution- ary relationship (Jongsma & Beekwilder, 2011). Kovalchuk et al. (2015) provided an important insight into the transcriptional response of the scot pine in response to attacks by the pine weevil Hylobius abietis. The conifer transcriptome showed 18 genes up-regulated, showing similarity to various types of protease inhibitors, common between conifers and angiosperms.

The functional importance of the serine proteases makes these proteins study targets in search of new strategies for the control of agricultural and forestry pests (Zhu- Salzman & Zeng, 2015). The development of studies aimed to produce transgenic plants expressing protease inhibitors enhanced the study and characterization of these proteins, especially in agricultural pests (Zavala et al., 2008; Senthilkumar et al., 2010). Addi- tionally, advances in the study of the mechanisms of RNA interference (RNAi) associ-

124 ated with genome sequencing of many species of pest insects generated information needed to search for new control strategies for insect pests based on RNAi technology (Zhu et al., 2011; Han et al., 2014; Sapountzis et al., 2014). These enzymes are particularly well characterized in Lepidoptera, but their knowledge in Coleoptera is restricted to a few species of the families Curculionidae, Chrysomelidae, Tenebrionidae and Lu- canidae (Gruden et al., 2004; Vinokurov et al., 2006; Broehan et al., 2010; Alahmadi et al., 2012, Darvishzadeh et al., 2015).

Bioinformatic analysis has become a relevant tool to provide a way for the identification and characterization of protein of interest. Information from in silico investigations is badly needed particularly when there is lack of in vivo information, as it will provide the foundation for approaches based on experimental work (Christie et al., 2014; Nesbit & Christie, 2014). The main goal of this study was to analyse in silico the transcriptome of the bark beetle T. yunnanensis by bioinformatic tools, and identify potential candidate genes for the synthesis of serine proteases. We intended to provide information on specific markers (digestion enzymes) to be used in future studies, aiming to develop new control strategies of this insect pest and its congeneric species.

8.3 Materials and methods

Library Construction and EST Assembly A cDNA library was constructed after pooling RNA extracted from larvae, pupae and adults of T. yunnanensis from Southwest China. The library was constructed by the research group of Dr. Jia-Ying Zhu of College of Forestry, Southwest Forestry University, Yunnan, China. Data was pre-processed by FastQC (Andrews et al., 2014) used for quality control of the sequences, which were trimmed such that all the positions had a Phred score greater than 20.

125 Assembly, Annotation and Gene Ontology

The contigs were de novo assembled using Trinity software (Hass et al., 2013) available in the Galaxy platform online (Hillman-Jackson et al., 2002). Non-coding RNAs (ncRNAs) were identified in the whole contigs using Infernal v1.1.1. (Nawrocki & Eddy, 2013) against the Rfam database version 12 (Burge et al., 2013). Functional annotation by GO (Gene Ontology) terms (http://www.geneontology.org), InterPro entries (InterProScan; http://www.ebi.ac.uk/Tools/pfa/iprscan/), enzyme classification codes (EC) and metabolic pathways (KEGG, Kyoto Encyclopedia of Genes and Genomes: http://www.genome.jp/kegg/) were determined using the Blast2GO software suite v3.1 (Conesa et al., 2005). Sequences were submitted to the NCBI protein nr database via BLASTx, with 1e-3 e-value threshold. GO terms were improved with the ANNEX tool (Myhre et al., 2006), followed by GOSlim tool available at Blast2GO (goslim_gener- ic.obo) (Conesa & Gotz, 2008). Combined graphs were constructed at level 2, for Biolo- gical Process, Molecular Function and Cellular Component categories. Enzymatic classification codes and KEGG metabolic pathways were generated by direct mapping of GO terms, with their respective ECs. InterPro searches were performed remotely from Blast2GO on InterProEBI server.

Protease Sequence Analysis

The amino acid sequences were obtained by in silico translation using the contigs annotated as serine proteases or serine proteinases. Subsequently, the largest open reading frame (ORF) from each contig was obtained. The functional analysis was based on tools that provide information about family, domains and important protein sites, such as Sig- nalP (Petersen et al., 2011), and Smart (Letunic et al., 2015) for prediction of signal peptides. Smart was also used to look for transcripts with a Tryp-Spc domain. The MEROPS peptidase database was performed to determine the different serine protease families and the position of the triadic pocket (http://merops.sanger.ac.uk) (Rawlings et al., 2013). Compute pI/MW was carried out using online tools at ExPASy SIB Bioin- formatics Resource Portal (http://www.expasy.org/tools/)(Artimo et al., 2012). Finally, the disulphide bonds were predicted with the EDBCP tool (http://biomedical.c- tust.edu.tw/edbcp) (Lin et al., 2013).

126 A phylogeny with insect sequences with the Tryp-Spc domain, available in the SMART database was built. All the Coleoptera sequences with this annotation were selected, but fragments or predictive sequences were discarded. Sequences of other insect families were also downloaded to test for paralogy. Those sequences with more than 400 amino acids were deleted from the matrix. The sequences were aligned with MUSCLE algorithm (Edgar , 2004) using the Geneious v7 software (Kearse et al., 2012). Finally, a tree was built using the RAxML (Stamatakis, 2014) plugin available in the Geneious software using “Rapid hill-climbing”.

8.4 Results and Discussion

Sequence Analysis, de novo Assembly and Annotation

The assembled sequences (209934 contigs) were submitted to the BLASTx program of the BLAST tools at NCBI, to compare with known sequences of the nr protein database. At the superior cut-off threshold for blast search set to 1e-3, 93978 contigs (44.77 %) returned hits against this database (Fig. 1). About 115.942 contigs (55.23 %) did not show significant sequence similarity at protein-level. This high percentage of novel proteins not yet described or annotated, that may represent rapid evolution genes or taxonomically restricted genes, put forward the necessity of investigating their structure and function that might be related with species diversification processes (Wissler et al., 2013).

Figure 1. Characterization of the transcriptome from Tomicus yunnanensis.

127 Figure 2. E-value distribution of the top BLASTx hits of the transcriptome of Tomicus yunnanensis. Sequences with e-value equal to 0 are represented in the right peak. The cut-off used was 1e-3.

The BLASTx hits distribution, according to the adopted e-value of 1e-3, was shown in Fig. 2. To determine the coverage of our library, we grouped the contigs according to the most frequent species similarities. The highest number of sequence hits occurred as expected with the Coleoptera and in particular with protein sequences of the bark beetle Dendroctonus ponderosae (Fig. 3) which have partially sequenced genome (Reddy et al., 2015). The second number of hits was the tenebrionid beetle Tribolium castaneum with completely sequenced genome as the Genome OnLine Database (Reddy et al., 2015) (Fig. 3).

128 Figure 3. Species distribution of the top BLAST hits for each unique sequence of Tomicus yunnanensis. A great number of contigs matched insect genes, mainly others coleopteran, Dendroctonus ponderosae and Tribolium castaneum. E-value cut-off is 1e-3.

GO and Function Classification

GO analysis was performed to classify the functions of the predicted proteins (Fig. 4). We observed a dominance of Biological Process GO terms for metabolic (~ 21 %) and cellular (~ 20 %) processes (Fig. 4, BP). For Molecular Function, it was observed a high percentage of terms for binding (~44 %) and catalytic activity (~39 %) (Fig. 4B, MF). For Cellular Components, a high percentage of GO terms were predicted for cell components (~33 %), organelle (~22 %) and membrane (~20 %) (Fig. 4, CC). The same pattern of GO classification was observed for other scolitid transcriptomes what is an evidence that a broad spectrum of the transcriptome was sampled (Aw et al., 2010, Firmino et al., 2013).

129 Figure 4. GO distribution by level 2 of the transcriptome of Tomicus yunnanensis. Biological Processes (BP), Molecular Functions (MP), and Cellular Components (CC).

Concerning the different classes of digestive enzymes, there are high proportions of serine proteases (314), lipases (170) and β-glucosidases (100) contigs annotated via In- terPro, however, there have been detected any aminopeptidases and only one cystein protease (Table 1). These differences (more lipases and β-glucosidases) observed for T. yunnanensis may explain the adaptation of the insect to a conifer-based diet (high proportion of cellulose) (Lazarevic and Janković-Tomanić, 2015) and to overwinter (Bon- net et al., 2012).

130 Table 1. Digestive enzymes found in the Tomicus yunnanensis transcriptome.

Serine Proteases

As most insects, Coleoptera are adapted to feed on a variety of nutrient sources that are digested by proteases present in the gut (Terra & Ferreira, 1994). According to Klein et al. (1996), the major class of proteases found in the Coleoptera gut is serine protease, particularly trypsin (EC 3.4.21.4) and chymotrypsin (EC 3.4.21.1), both characterized by the catalytic triad His 57, Asp 102, and Ser 195 (number refers to bovine chymotrypsin). Chymotrypsin cuts protein chains on the carboxyl side of the aromatic amino acids phenylalanine and tyrosine (Terra and Ferreira, 1994), whereas trypsin cuts specially protein chains on the carboxyl side of basic amino acids arginine and lysine. On the other hand, the specificity of serine proteases is determined by the assembly of the substrate to the binding pocket (S1) (Perona et al., 1995).

In the T. yunnanensis library, 346 contigs for serine proteases were identified by In- terPro analysis. But there are only 13 contigs with a signal peptide. When the sequence alignment was carried out to identify conserved sites, four trypsin-like proteases and five chymotrypsin-like proteases present complete protein coding sequences, where Asp 189, Gly 216 and Gly 226 residues form the trypsin S1 pocket and Gly/Ser 189, and Gly 216 and Gly 226 residues form the chymotrypsin S1 pocket (Perona et al., 1995, 1997) (Table 2). According to the MEROPS peptidase database, all the sequences were classi-

131 fied into the SA1 family and many preserved all the characteristics of digestive serine proteases.

The sequences contained 5' UTR and 3' UTR regions and an ORF varying from 1100 to 5400 bp. The trypsin-like transcripts coded for predicted proteins of 267–293 amino acids with predicted isoelectric points ranging from 4.62 to 7.80 and theoretical molecular weights of 25.6 to 35.5 kDa. The chymotrypsin-like transcripts coded for predicted proteins of 280–327 amino acids with isoelectric points ranging from 4.86 to 5.89 and theoretical molecular weights of 28.0 to 35.8 kDa (Table 2).

The trypsin-like transcripts were named Tysp1-Tysp4 and contained an N-terminal signal peptide. A conserved RIVGG propeptide-processing site and the catalytic triad were shown in all sequences. Also the binding pocket is conserved (D-G-G) , although the adjacent amino acids are different (Table 2).

Chymotrypsin-like transcripts were named Tysp5-Tysp9 and they preserved the N- terminal signal peptide. A conserved RIVGG propeptide-processing site is shown in the Tysp-5,9. However, this site is different from the contigs of Tysp6 (a-e), Tysp7 and Tysp8, RXXGG where X means any amino acid. The catalytic triad is similar to other chymotrypsins in Tysp7, Tysp8 and Tysp9. However, slight differences in DIAL region were observed. For instance, the adjacent amino acids of Asp are DIAV in spliced transcripts in Tysp6 (a-e) and DVAL in Tysp5. Concerning the binding pocket of chymotrypsin proteins (G-G-S), it is conserved in Tysp5, Tysp6, Tysp7 and Tysp9, although there are differences in the adjacent amino acids. The protein Tysp7 shows eight cysteine residues in contrast to most chymotrypsins which have six cysteines. In addition, Tysp8 shows a difference in the binding pocket, where the serine is replaced by a threonine (Table 2). These variations in the substrate binding sites of chymotrypsins in T. yunnanensis may led to a plasticity of chymotrypsin to accept different substrates, and also to avoid protease inhibitors of pine trees (Reynolds & Fischer, 2015).

132 Table 2. Characteristic of nine putative serine proteases of Tomicus yunnanensis deduced from library contigs: size (in number of amino acids) of mature peptide (MT), conserved cleavage site (CS), catalytic triad and substrate-binding sequences, number of cysteine residue, Smart Domain, Isoelectric point / Molecular weight (pI/MW), family, code, region and e value form the Peptidase database Merops.

Table 2. (cont.)

Other sequences (Tysp10, Tysp11, Tysp 12 and Tysp13) with a domain Tryp_spc but without the substrate binding pocket region and with only two of the regions of the triadic catalytic (histidine-aspartic acid) were identified. These kinds of molecules are classified as serine protease homologs (SPH). The spliced transcripts of Tysp13 (a-d)

133 shows the Trypsin or Tryp_spc domain and the loss of a full catalitic domain can be explained as a paralogous pseudoprotease protein (Reynolds & Fischer, 2015). However, Tysp10, Tysp11 and Tysp12 also carry a Clip-domain, which is involved in the innate immune response and the embryonic development (Kanost & Jiang, 2015; Veillard et al., 2015).

To better characterize the serine proteases in T. yunnanensis, a phylogenetic analysis was performed to discern their evolutionary relationships among representative SPs of several Coleoptera species (Fig. 4). The sequences were fully distributed into three phylogenetic classes: trypsins, chymotrypsins and clip-domains. Within these clades the T. yunnanensis sequences do not form a clade congruent to the taxonomy, but rather present homologies with other insect species what it is expected if they have evolved by gene duplications, followed by divergence due to selective evolutionary pressure, to produce a diverse set of paralogues (Lazarevic & Janković-Tomanić, 2015)

134 Figure 4. Phylogenetic tree built with RaxML. Tomicus yunnanensis samples are coloured in red. Blue: Chymiotrypsin, Red: Trypsin, Green: Clip-domain.

135 8.5 Final remarks

Bioinformatic tools have been used in this study to identify, annotate and characterize in silico nine genes coding for digestive enzymes of a transcriptome of T. yunnanensis, of which five are chymotrypsin and four are trypsin. Other four contigs were identified as serine protease homologs (SPH) possibly related to immunity and development. These enzymes could eventually be used for molecular investigation, like gene expression quantitation, differential gene expression in specific insect stages or submitted to certain conditions and silencing via RNAi, in order to validate genes to be used in pest management strategies in the bark beetles of the genus Tomicus.

Acknowledgements

This work was financed by Special Fund for Forest Scientific Research in the Public Interest of China (201004067), Fund of Reserve Talents for Young and Middle-Aged Academic and Technological Leaders of Yunnan Province (2013HB077) and Fundación Séneca (12023/PI/09) of the Murcia Regional Government (Spain). Thanks also to CNPq (Brasil) for granting a postdoctoral scholarship (PDE) to V.M..

References

Alahmadi, S. S., Ouf, S. A., Ibrahim, R. A., and El-Shaikh, K. A. (2012). Possible control of date palm stag beetle, Lucanus cervus (L.) (Coleoptera: Lucanidae), using gut protease inhibitors of different bio-control agents. Egyptian Journal of Biological Pest Control, 22(2), 93–101.

Andrews, S. (2014). FastQC A quality control tool for high throughput sequence data. retrieved October 17, 2014, from http://www.bioinformatics.babraham.ac.uk/projects/fastqc/

Artimo, P., Jonnalagedda, M., Arnold, K., Baratin, D., Csardi, G., Castro, E. de ,Duvaud, S, Flegel, V, Fortier, A, Gasteiger, E, Grosdidier, A, Hernandez, C, Ioannidis, V, Kuznetsov, D, Liechti, R, Moretti, S, Mostaguir, K, Redaschi,

136 N, Rossier, G, Xenarios, I, Stockinger, H. (2012). ExPASy: SIB bioinformatics resource portal. Nucleic Acids Research, 40(W1), W597–W603.

Aw, T., Schlauch, K., Keeling, C. I., Young, S., Bearfield, J. C., Blomquist, G. J., and Tittiger, C. (2010). Functional genomics of mountain pine beetle (Dendroctonus ponderosae) midguts and fat bodies. BMC Genomics, 11(1), 215. 15

Bode, R. F., Halitschke, R., and Kessler, A. (2013). Herbivore damage-induced production and specific anti-digestive function of serine and cysteine protease inhibitors in tall goldenrod, Solidago altissima L. (Asteraceae). Planta, 237(5), 1287–1296.

Bonnett, T. R., Robert, J. A., Pitt, C., Fraser, J. D., Keeling, C. I., Bohlmann, J., and Huber, D. P. W. (2012). Global and comparative proteomic profiling of overwintering and developing mountain pine beetle, Dendroctonus ponderosae (Coleoptera: Curculionidae), larvae. Insect Biochemistry and Molecular Bio- logy, 42(12), 890–901.

Broehan, G., Arakane, Y., Beeman, R. W., Kramer, K. J., Muthukrishnan, S. and Merzendorfer, H. (2010). Chymotrypsin-like peptidases from Tribolium castaneum: A role in molting revealed by RNA interference. Insect Biochem- istry and Molecular Biology, 40(3), 274–283.

Burge, S. W., Daub, J., Eberhardt, R., Tate, J., Barquist, L., Nawrocki, E. P., Eddy,S.R., Gardner,P.P and Bateman, A. (2013). Rfam 11.0: 10 years of RNA families. Nucleic Acids Research, 41(D1), D226–D232.

Christie, A. E., Fontanilla, T. M., Roncalli, V., Cieslak, M. C., and Lenz, P. H. (2014). Identification and developmental expression of the enzymes responsible for dopamine, histamine, octopamine and serotonin biosynthesis in the copepod crustacean Calanus finmarchicus. General and Comparative Endo- crinology, 195, 28–39.

Conesa, A., and Götz, S. (2008). Blast2GO: A Comprehensive Suite for Functional Analysis in Plant Genomics. International Journal of Plant Genomics, 2008, e619832.

137 Conesa, A., Götz, S., García-Gómez, J. M., Terol, J., Talón, M., and Robles, M. (2005). Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics, 21(18), 3674–3676.

Darvishzadeh, A., Bandani, A., Amiri, A., and Mousavi, S. Q. (2015). Serine and cysteine proteases of Rhynchophorus ferrugineus (Coleoptera: Curculionidae) larvae raised on date palms (Phoenix dactylifera). Journal of Asia-Pacific En- tomology, 18(3), 547–552.

Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32(5), 1792–1797.

Firmino, A. A. P., Fonseca, F. C. de A., de Macedo, L. L. P., Coelho, R. R., Antonino de Souza Jr, J. D., Togawa, R. C. Silva-Junior, O. B., Pappas-Jr, G. J., Mattar da Silva, M. C., Engler, G., and Grossi-de-Sa, M. F. (2013). Transcriptome Analysis in Cotton Boll Weevil (Anthonomus grandis) and RNA Interference in Insect Pests. PLoS ONE, 8(12), e85079.

Gruden, K., Kuipers, A. G. J., Gunčar, G., Slapar, N., Štrukelj, B., and Jongsma, M. A. (2004). Molecular basis of Colorado potato beetle adaptation to potato plant defence at the level of digestive cysteine proteinases. Insect Biochem- istry and Molecular Biology, 34(4), 365–375.

Haas, B. J., Papanicolaou, A., Yassour, M., Grabherr, M., Blood, P. D., Bowden, J.,Couger, M. B., Eccles, D., Li, B., Lieber, M., MacManes, M, D,, Ott, M., Orvis, J., Pochet, N., Strozzi, F., Weeks, N., Westerman, R., William, T., Dewey, C. N., Henschel, R., Leduc, R. D., Friedman, N. and Regev, A. (2013). De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nature Protocols, 8(8), 1494– 1512.

Han, P., Fan, J., Liu, Y., Cuthbertson, A. G. S., Yan, S., Qiu, B.-L., and Ren, S. (2014). RNAi-Mediated Knockdown of Serine Protease Inhibitor Genes In- creases the Mortality of Plutella xylostella Challenged by Destruxin A. PLoS ONE, 9(5), e97863.

138 Hillman-Jackson, J., Clements, D., Blankenberg, D., Taylor, J., Nekrutenko, A., and Team, G. (2002). Using Galaxy to Perform Large-Scale Interactive Data Ana- lyses. In Current Protocols in Bioinformatics. John Wiley & Sons, Inc.

Jamal, F., Pandey, P. K., Singh, D., and Khan, M. Y. (2012). Serine protease inhibitors in plants: nature’s arsenal crafted for insect predators. Phytochemistry Re- views, 12(1), 1–34.

Jongsma, A. M., and Beekwilder, J. (2011). Co-Evolution of Insect Proteases and Plant Protease Inhibitors. Current Protein and Peptide Science, 12(5), 437– 447.

Joshi, R. S., Mishra, M., Suresh, C. G., Gupta, V. S., and Giri, A. P. (2013). Comple- mentation of intramolecular interactions for structural–functional stability of plant serine proteinase inhibitors. Biochimica et Biophysica Acta (BBA) - Gen- eral Subjects, 1830(11), 5087–5094.

Kanost, M. R., and Jiang, H. (2015.). Clip-domain serine proteases as immune factors in insect hemolymph. Current Opinion in Insect Science.

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S.,Bux- ton, S, Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P. and Drummond, A. (2012). Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics, 28(12), 1647–1649.

Klein, B., Le Moullac, G., Sellos, D., and Van Wormhoudt, A. (1996). Molecular cloning and sequencing of trypsin cDNAs from Penaeus vannamei (Crustacea, Decapoda): Use in assessing gene expression during the moult cycle. The In- ternational Journal of Biochemistry & Cell Biology, 28(5), 551–563.

Kovalchuk, A., Raffaello, T., Jaber, E., Keriö, S., Ghimire, R., Lorenz, W. W., Dean, J. F. D., Holopainen, J. K., and Asiegbu, F. O. (2015). Activation of defence pathways in Scots pine bark after feeding by pine weevil (Hylobius abietis). BMC Genomics, 16(1), 352.

Lazarević, J., and Janković-Tomanić, M. (2015). Dietary and phylogenetic correlates of digestive trypsin activity in insect pests. Entomologia Experimentalis et Ap- plicata, 157(2), 123–151.

139 Letunic, I., Doerks, T., and Bork, P. (2015). SMART: recent updates, new develop- ments and status in 2015. Nucleic Acids Research, 43(D1), D257–D260.

Lin, H.-H., Hsu, J.-C., Hsu, Y.-N., Pan, R.-H., Chen, Y.-F., and Tseng, L.-Y. (2013). Disulfide connectivity prediction based on structural information without a prior knowledge of the bonding state of cysteines. Computers in Biology and Medicine, 43(11), 1941–1948.

Mithöfer, A., and Boland, W. (2012). Plant Defense Against Herbivores: Chemical Aspects. Annual Review of Plant Biology, 63(1), 431–450.

Myhre, S., Tveit, H., Mollestad, T., and Lægreid, A. (2006). Additional Gene Onto- logy structure for improved biological reasoning. Bioinformatics, 22(16), 2020–2027.

Nawrocki, E. P., and Eddy, S. R. (2013). Infernal 1.1: 100-fold faster RNA homology searches. Bioinformatics, 29(22), 2933–2935.

Nesbit, K. T., and Christie, A. E. (2014). Identification of the molecular components of a Tigriopus californicus (Crustacea, Copepoda) circadian clock. Comparat- ive Biochemistry and Physiology Part D: Genomics and Proteomics, 12, 16– 44.

Perona, J. J., and Craik, C. S. (1995). Structural basis of substrate specificity in the serine proteases. Protein Science : A Publication of the Protein Society, 4(3), 337–360.

Perona, J. J., and Craik, C. S. (1997). Evolutionary divergence of substrate specificity within the chymotrypsin-like serine protease fold. The Journal of Bio- logical Chemistry, 272(48), 29987–29990.

Petersen, T. N., Brunak, S., von Heijne, G., and Nielsen, H. (2011). SignalP 4.0: discriminating signal peptides from transmembrane regions. Nature Methods, 8(10), 785–786.

140 Rawlings, N. D., Waller, M., Barrett, A. J., and Bateman, A. (2014). MEROPS: the database of proteolytic enzymes, their substrates and inhibitors. Nucleic Acids Research, 42(D1), D503–D509.

Reddy, T. B. K., Thomas, A. D., Stamatis, D., Bertsch, J., Isbandi, M., Jansson, J., Mallajosyula, J., Pagani, I., Lobos, E. A. and Kyrpides, N. C. (2015). The Gen- omes OnLine Database (GOLD) v.5: a metadata management system based on a four level (meta)genome project classification. Nucleic Acids Research, 43(Database issue), D1099–1106.

Reynolds, S. L., and Fischer, K. (2015). Pseudoproteases: mechanisms and function. Biochemical Journal, 468(1), 17–24.

Sapountzis, P., Duport, G., Balmand, S., Gaget, K., Jaubert-Possamai, S., Febvay, G., … Calevro, F. (2014). New insight into the RNA interference response against cathepsin-L gene in the pea aphid, Acyrthosiphon pisum: Molting or gut phenotypes specifically induced by injection or feeding treatments. Insect Biochemistry and Molecular Biology, 51, 20–32.

Senthilkumar, R., Cheng, C.-P., and Yeh, K.-W. (2010). Genetically pyramiding protease-inhibitor genes for dual broad-spectrum resistance against insect and phytopathogens in transgenic tobacco. Plant Biotechnology Journal, 8(1), 65– 75.

Stamatakis, A. (2014). RAxML version 8: a tool for phylogenetic analysis and post- analysis of large phylogenies. Bioinformatics, 30(9), 1312–1313.

Terra, W. R., and Ferreira, C. (1994). Insect digestive enzymes: properties, compart- mentalization and function. Comparative Biochemistry and Physiology Part B: Comparative Biochemistry, 109(1), 1–62.

Veillard, F., Troxler, L., and Reichhart, J.-M. (2015). Drosophila melanogaster clip- domain serine proteases: Structure, function and regulation. Biochimie, in press.

141 Biochemistry and Physiology Part B: Biochemistry and Molecular Biology, 145(2), 126–137.

Waring, K. M., Reboletti, D. M., Mork, L. A., Huang, C.-H., Hofstetter, R. W., Gar- cia, A. M., Fule, P. Z. and Davis, T. S. (2009). Modeling the impacts of two bark beetle species under a warming climate in the southwestern USA: Ecolo- gical and Economic Consequences. Environmental Management, 44(4), 824– 835.

Zhu, F., Xu, J., Palli, R., Ferguson, J., and Palli, S. R. (2011). Ingested RNA interference for managing the populations of the Colorado potato beetle, Leptinotarsa decemlineata. Pest Management Science, 67(2), 175–182.

Zhu, J.-Y., Zhao, N. and Yang, B. (2012a). Global Transcriptional Analysis of Ol- factory Genes in the Head of Pine Shoot Beetle, Tomicus yunnanensis. Inter- national Journal of Genomics, 2012, e491748.

Zhu, J.-Y., Zhao, N. and Yang, B. (2012b). Global Transcriptome Profiling of the Pine Shoot Beetle, Tomicus yunnanensis (Coleoptera: Scolytinae). PLoS ONE, 7(2), e32291.

Zhu-Salzman, K., Koiwa, H., Salzman, R. A., Shade, R. E., and Ahn, J.-E. (2003). Cowpea bruchid Callosobruchus maculatus uses a three-component strategy to overcome a plant defensive cysteine protease inhibitor. Insect Molecular Biology, 12(2), 135–145.

Zhu-Salzman, K., and Zeng, R. (2015). Insect Response to Plant Defensive Protease Inhibitors. Annual Review of Entomology, 60(1), 233–252.

Zhu-Salzman, K., and Zeng, R. S. (2008). Molecular mechanisms of insect adaptation to plant defense: Lessons learned from a Bruchid beetle. Insect Science, 15(6), 477–481.

142 143 144 9 Chapter 4

Applying eco-field and the General Theory of Resources to bark beetles: Beyond the Niche Construction Theory

145 Article in preparation for Biosemiotics

Sánchez-García, F. J.1, Machado, V1, Galián, J.1 and Gallego, D.1 (2015). Applying eco-field and the General Theory of Resources to bark beetles: Beyond the Niche Construction Theory. Biosemiotics.

1 Departamento de Zoología y Antropología Física, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain

146 9.1 Abstract

A new perspective in landscape ecology is the application of the term Eco-Field together with the General Theory of Resources. In this paper, we describe the putative eco-field in bark beetles as a spatial configuration with a specific meaning-carrier for every organism-resource interaction. Bark beetles are insects with a relevant role in matter and energy cycles in conifer forests and they are involved in significant changes of forestry landscapes when plague outbreaks occur. Bark beetles are conveyed by the recognition of semiotic signals towards host trees using a specific eco-field. These signals are mainly a group of scents, which have been called odourtope. The interactions with other organisms (fungi, bacteria, nematodes, predators, etc.) takes place through sharing relevant information of the eco-field networks (representamen networks) in the forestry ecosystem. The eco-field networks let an expansion of the realized semiotic niche of the bark beetle towards the potential semiotic niche. Also if different organisms end up showing an interdependence of the eco-field, the niche construction process can be initiated. Subsequently, if this interdependence becomes crucial the process can led to the establishment of mutualistic relationships. This is an example on how evolutionary processes are initiated by the recognition of signals by a network of eco-fields. Besides, we showed a description of host search and maturation eco-fields in Tomicus destruens and Tomicus piniperda and a hypothetical explanation of the evolution of their sympatric co- existance.

Keywords: semiotic niche, eco-field networks, niche construction, bark beetles, mutualistic relationships

9.2 Introduction

The landscape studies have undergone several novel approaches in recent years. These new perspectives has been based on Pateson (1978) ideas, who was inspired by the pioneer work of Von Uexküll (1940) with the definition of Umwelt as a set of relations an organism has in an ecosystem (Kull, 2010). This new interpretation of living

147 systems is being studied by several researchers (linguistics, biologists, psychologists, informatics, philosophers, etc.) establishing the field of biosemiotics. The compartmental- ization of knowledge in biological entities according to the object of study had led to many sub-disciplines of semiotics, such as ecosemiotics, zoosemiotics, phytosemiotics, mycosemiotics, endosemiotics, microsemiotics. Ecosemiotics is a branch of integrative biology and semiotics that considers ecosystems as communication systems and its components as a relation signs in space (Maran & Kull, 2014). Some attempts treated the ecosystems as interpreting identities, called environs (Nielsen, 2007), where ecological networks are treated from the perspective of thermodynamics and information fluxes. These environs take inputs and produce outputs (which are defined by the context) in which the importance is given to the quality of information rather than to the amount of information. Each element of the system can act as compartment, connector or control. Therefore, the full information of the system is the summation of the different elements (ITotal = Icomp + Iconn + Ictrl). Nielsen (2007) explains the concept of exergy information to explain the compartment information (Icomp) and the concept of connecty (also known as connectance) to explain the connector information (Iconn). But control information (Icrtl), which must be at the core of semiotic and non-mechanical processes in the ecosystem, has not been explained yet.

A new perspective in landscape ecology is the application of the term eco-field (Farina & Belgrano, 2004, 2006) together with the General Theory of Resources (GTR) (Farina, 2008), where the eco-field is a spatial configuration with a specific meaning- carrier for every semethic organism-resource interaction. The summation of all eco- fields constitute the habitat where this organism find all the resources that are necessary to stay alive and to perform the functions of living (Farina & Napoletano, 2010, Farina, 2010, Farina, 2011). Several works have tried to use this paradigm in their landscape ecology studies in order to find for example ecological similarities in the landscape (Pizzolotto, 2009) using carabids beetles, the spatial and temporal dimensions of a bird soundscape (Farina et al. 2011, Farina & Peretti, 2013, Farina & Peretti, 2014, Farina et al. 2014, Malavasi et al. 2013) or the herbal landscape (Söukand & Kalle, 2010). This semiotic point of view is focused into the qualitative rather than the quantitative relations, such as matter and energy of the ecosystem components. To do so, a consortium, as the group of organisms connected via sign relations (icons, indices and symbols) that

148 we want to analyze, have to be identified. These sign relations are ecologically inherit- able and are performed by ecological codes (Kull, 2010). The ecological codes are distributed in the ecosystems and are open to new species to become involved. They are built upon the consistencies, constraints and habits in a particular ecological community. Also they are based on indexical relations and uses different types of memories (Maran, 2012).

Organisms can use different types of codes to communicate, such as smells, colors, sounds, or movements. The term ecomone includes those non-trophic molecules which contribute to insure a “flux of information” between organisms in an ecosystem. This term was adopted by Florkin (1974) in order to include the diversity of chemical compounds present in ecosystems (Pasteels, 1982; Bruni, 2013). This term ceased in use and was replaced by the synonymy infochemical and semiochemical in the literature. These last terms diversified into different nomenclature (pheromone, allomone, kairomone, synomone) depending on the context (interspecies or intraspecies), the emitter and the recipient of these signals (Dicke & Sabelis, 1988). However, before that, Florkin (1965, 1967) suggested a new terminology that did not have much use. This author had described the molecules as coactones that determine the relationship between the coactor (active and directing organism) and the coactee (passive and receiving organism). He added the exo- prefix if the molecules are liberated by the coactor in the environment to reach the coactee (exocoactones). In contrast if the molecules are secreted inside the organism the endo- prefix is added (endocoactones). To harmonize these old and new concepts, pheromones correspond to exocoactones where both coactor and coactee belong to the same species and, allomones are molecules where the coactor and coactee belong to different species, and hormones are endocoactones (Favareau, 2009 and references therein)

A group of scientists have proposed a new framework into evolutionary sciences called Extended Evolutionary Synthesis (EES) (Laland et al. 2015), not without generating controversy with supporters of the Standard Evolutionary Theory (SET) (Laland et al. 2014). EES points out those proximate causes (developmental bias, developmental plasticity, inclusive inheritance, niche construction) are themselves often also evolution-

149 ary causes. Matthews et al. (2013) states that “the niche construction concept refers to the process whereby the metabolism, activities and choices of organisms modify or sta- bilize environmental states, and thereby affect selection acting on themselves and other species”. This conceptual framework of niche construction has appeared in recent years as a cohesive attempt between the fields of ecology and evolution. It is focused in the knowledge gap between Evolutionary Genetics and Population Ecology as these disciplines idealize the component (environmental or genetic) outside its scope as a fixed variable. In turn, both disciplines shows an aloofness from ecosystem ecology, the latter does not distinguish between the organism and its environment, but the ecosystem is manifold, both biotic and abiotic interconnected functional components whose interactions are understood in terms of flows of energy and matter (Barker & Odlling-Smee, 2014).

Holobionts, as units of biological organization, are complex multicellular eukaryotes that are not and have never been autonomous organisms, but rather are biological units organized from numerous microbial symbionts and their genomes (Bordenstein & Theis, 2015). There was one attempt to point out that the niche construction and development scaffolding led to understand how different species in the birth holobiont are affected each other in their development and evolution among mother, child and symbiont mi- croorganism (Chiu & Gilbert, 2015).

Forests share the properties of complex systems such as heterogeneity, hierarchy, self-organization, openness, adaptation, memory, non-linearity, and uncertainty (Filotas et al. 2014). The forest ecosystem have many multitrophic interactions (Plant–Herbi- vore–Predator-Parasitoid–Pathogen) which should be studied under a context-dependent view (Bruni, 2013). For example, the saproxylic organisms (fungi, beetles and others) have considerable importance in the recycling of the dead wood in the forests. In this way, it lets the flux of matter and energy on the ecosystem network (Yang & Gratton, 2014). Bark beetles (Curculionidae, Scolytinae) are important components of the forest ecosystems, so that disentangle the semiotic network which emerge around them would be of interest. Bark beetles comprise several thousands of species which live in alive and dead tissues of forest trees. Functionally these beetle species can be classified as primary or secondary species depending of the fitness of attacked target tree. The

150 primary species are specialized to inhabit live tissues. However, the secondary species prefer to colonize dead or highly injured trees (Vega & Hofstetter, 2014). The aim of this study is to describe the different types of meaningful information (Barbieri, 2009) surrounding around bark beetles. That is, whereas the relation of the flux of energy and matter has been the only way to quantify the information in a landscape matrix, recently a new perspective has arisen that consider the quality rather than the quantity of information that flows into the ecosystem (Farina et al. 2011, Farina & Peretti, 2013, Farina et al. 2014, Malavasi et al. 2014). Although there are several reviews that link the different chemical compounds related to the search of host trees as part of the bark beetles behaviour (Gitau et al., 2013, Byers & Zhang, 2011), our purpose is looking at this from a different but complementary angle. We would like to pay attention to the different mechanisms in the semiotic niche of the bark beetles. That is, we aim to describe the eco-field of searching the host tree and mating by a hypothetical bark beetle, using the different examples of particular bark beetle species in the literature. Also we want to analyse evidences that point out that the union of different eco-fields of the different forestry species may have the capacity to initiate the niche construction process. That is, how these organisms can interrelate different eco-fields with others species to create new niches where the species can explore and to evolve.

9.3 Eco-field of host search and mating in bark beetles. Triadics of eavesdropping each other.

Firstly, every bark beetle individual must find their resource (host tree) among of specific eco-field into the landscape matrix where the beetle develops its life cycle. For that, it must decide what suitable host tree (i.e. conifer trees) to go to, among the rest of plants (non-hosts or not suitable host) by smells and visual signals. The bark beetle can fly from hundred to thousand meters (long range) following the odorant signals created by the ecomones (pheromones from conspecifics insects and/or kairomones from host volatiles). On the way, substances such as green volatiles from angiosperm trees provoke a repellence (or non-attraction) in the beetle flight. When the beetle is getting closer to host tree (around a few meters), the beetle’s eco-field is fitted by feedback of visual signals. These signals represent “index” which resemble the host tree, towards the

151 beetle orientate constituting a narrow dark silhouette with low reflectance (Vasechko, 1978, Campbell & Borden, 2006a, Strom et al. 2001). Finally, the insect lands into the suitable host due to a combination of odorants and visual cues. This has been checked in bark beetles such as the mountain pine beetle Dendroctonus ponderosae Hopkins, 1902 (Campbell & Borden, 2006b). The researchers used this trait in order to collect beetles in the forest using dark plastic traps baited with different attractants compounds from bark beetle’s ecofield

Detailed works about the structure of the antennal club of the species Ips typo- graphus shows highly variable receptors for host compounds and little variable receptors for discriminating non-host substances. In order to avoid the redundancy in the repellent message from non-host trees, the organism use few types of receptors for non- hosts angiosperm trees. However, for an effective landing in the host tree, the bark beetle needs more detailed information for what it uses around ten different receptors, which can discriminate between pheromones and host compounds (Andersson, 2012, and references therein).

The secondary species Pityogenes bidentatus (Herbst, 1783) was observed to have a dual behaviour in relation to the host monoterpenes. These substances are emitted from the resin of the host conifer tree as a defense mechanism. A high resin exudation by a tree can stop insect attacks and, so that, it is indication of healthy status and therefore insects fly away avoiding these unsuitable tree. However if this insect have already found an appropriate tree and it has landed on it the insect does not avoid this odour (Byers, 2012), and a walking behaviour towards resin sources was observed. This signal might be indicative of a potential injured tree where the beetle can dig into the host. Interest- ingly, the same signal causes different behavioural patterns depending on the surrounding context, that is to say, that the beetle fitted the eco-field to its needs.

In the evolution of the beetle-host system with others surrounded symbiont microorganisms as fungi or bacteria, called holobionts, an openness of new semiotic niche seems to be present. For example, the bark beetle, Dendroctonus ponderosae has big progeny in galleries colonized by symbiotic fungi. These fungi species concentrate the wood with nitrogen and ergosterol improving its nutritional quality for the beetles

152 (Goodsman et al., 2012). Besides, the beetles may acquire new enzymes pathways in order to detoxificate the toxic host compounds or to resist the attack of other pathogens under the bark (Six, 2013 and references therein). Therefore, when the pioneer beetles start to create a new gallery inside the bark, they have to eat the phloem-cambium tissues that include toxic compounds of the plant defense system and also they have to overtake structural defenses of the vascular tissues (Kolosova & Bohlmann, 2012). Af- terwards, microbiote which lives inside in the beetle's gut or in the beetle's galleries starts to synthetize proteins which are detoxifying enzymes (Xu et al. 2015, Boone et al., 2013, Adam et al., 2013). The transformation of toxic compounds into new excreted substances results in new odors clues themselves, transmitted to their conspecifics as sign of host-acceptance or aggregation pheromone (Seybold et al., 2006). These substances excreted by the beetle can act as attraction signs for other species, such as symbiotic mites (Tarsonemus spp), which collaborate in the colonization eating fungi that might be harmful for the beetles (Machingambi et al., 2014, Hofstetter & Moser, 2014). In turn, these excreted substances might act as attractant for predators (Reeve et al. 2009; Costa & Reeve, 2011a). That can explain that in the symbiont relationships there are a connection of chains of triadic meaning, similar to the triadic described at molecular scale. In those cases, the cell transfers information from the extracellular environment into the cell through cascades of proteins and other metabolites when the presence of antigens are detected in the outside (El-Hani et al. 2007). Applying this to bark beetles, there is a resource (Object) represented by the host tree and a bark beetle species (Subject) that considers this tree as a fitting host. This choice is based on a certain spatial configuration that is the sum of odors and visual signals (like host-compounds, aggregation pheromones and dark silhouette among others) that we call eco-field (Inter- pretant). Afterwards, the presence of the beetle (initially the Subject) becomes the Ob- ject for other species, as the presence of the bark beetle creates a different spatial configuration of a new eco-field of pheromones and/or excretion products (Intertrepant). This new eco-field will attract to the host tree different species that act as predators (Thanas- imus spp, Temnochila spp) or parasitoids (wasps -Heydenia spp, Rhopalicus spp) (Boone et al. 2008; Xiaoyi & Zhongqi, 2008) of the beetles. Besides, the growth of symbiotic fungus as Ophiostoma, Ceratocystiopsis, Grosmannia or Ceratocystis spp (Six, 2012), different nematodes species (Susoy & Herrmann, 2014) or mites inside the gallery will be favored as the beetle will contribute to their mechanical transport (phoresis)

153 to other trees. All these new organisms become themselves the Subjects of the new created triadics (Figure 1,2). This new triadics will create eco-field networks which its evolution's implications will explain in the next section.

Even these eco-fields of host search and/or defense against predators can change in- traspecifically depending on the geographic location and the biodiversity of scolytid species in a particular habitat (Miller et al. 1997). For instance, other herbivore insects species that coexist with the bark beetles may use other chemical compounds in order to colonize different parts of the host tree or in different seasons (Ayres et al. 2001; Lu et al., 2012; Lu et al. 2012) and therefore, predators vary their diet according to the most common bark beetles (Reeve et al., 2009; Costa & Reeve, 2011b). Organisms in general and bark beetles in particular use the codes which let an intrinsic flexibility (semiotic freedom), for example, the chemical compounds (ecomones) which can permute in different combinatorial stereoisomers and also can modulate in different concentrations or mix between compounds. It lets to avoid predator pressure changing the ratios of different compounds from point of view of the bark beetle population (Raffa et al. 2007).

After the host choice, the bark beetle digs a hole (generally the female in monogamus species as Dendroctonus or Tomicus species) through the bark. While the female is wait- ing in the nuptial chamber the male arrives to the entrance hole. The courtship is a mix of odors, sounds, tactile signals between the male and female. For example, the male of Dendroctonus valens LeConte, 1860 can identify what female is the better for generating progeny by ecomones (pheromones) cues. This choice is important in terms of reproductive efficiency as it has been observed that these monogamous couples create a higher number of egg galleries with a higher number of eggs and larvae offspring than couples experimentally forced consisted in males and non-preferred females (Chen et al., 2012). Likewise, the D. valens female choose the male mate by different signals, including sounds (Lindeman & Yack, 2015) as males produce chirp trains by stridulatory structure in the abdomen upon approaching the entrance of a female’s gallery. There are two types of chirps, simple and interrupted. Females let to enter the galleries to a male which produces interrupted chirps, that seems to be indicative of having more motor skills in order to create the gallery (Lindeman & Yack, 2015). So that, the sound (and vibrations) has importance in close-range communication. There are several contexts in

154 which the bark beetles produce different types of chirps. For example, the stress signals are produced when there a predator nearby to alert conspecific individuals. Also intraspecific signals may be produced by a female in order to indicate to a male that the hole is inhabited by her, but not by a predator (Fleming et al. 2013). All these interaction questions, male-male, female-male, larvae, predators and the medium substrate questions (airborne or vibrational) will be solved in the next years in order to understand the “grammar” and “syntaxes” of this bark beetle language.

We can suppose that an odourtope emerged into the holobionts relations of the consortium of bark beetles in the forestry matrix. Just like the study of the soundtope in birds (Farina, 2014; Malavasi et al., 2014). This is a non-random coordinated distribution of the odors into the forestry matrix. Smell cues are the dominant traits in their search host behavior. According to the eco-field concept, the bark beetles use the chemical information that they perceive from their surrounding as an odour carrier of meaning. This eco-field is a real sign system with its own grammar rules. We can defined the odourtope like the set of chemical signal around the umwelt of the bark beetle which has a signification by the eco-field. The odour pattern can be described as a distribution of different volatile chemical compounds in the forestry composed by combinational mix, evapotranspiration rates, modulation of the dosage and time. This grammar could be considered as simple in contrast to that of mammalian and bird communications systems. Due to the nature of the chemical compounds which spread passively, the answers between conspecifics can be delayed in hours or days. In our case these simple grammar rules have the aims to differentiate the host trees from non-host trees, the over-crowed tree with peers from a free tree, the damage tree from a healthy tree. But even these odor codes can contribute to fit the perception of different chemical compounds in a habitat occupied by different species of bark beetles or predators in order to reorganize the ecological code seasonally (Raffa et al. 2007). Also that the odour signals, there are other signals like tactiles, sounds, vibrational that has similar or more importance during the mating into the gallery.

155 9.4 Semiotic niche and the role of eco-field

The organisms are characterized by an ecological niche. The first definition was proposed by Grinnell (1924) who specified it as the sum of all necessary requirements for a population to develop. That is, it is spatially explicit and measurable from the organism distribution. In contrast, Elton (1927) focused his niche definition on the biotic interactions and their impacts on others species. Later, Hutchinson (1957) unified both point of views and defined the fundamental niche as the n-dimensional hypervolumen defined by n-variables, “every point of which corresponds to a state of the environment which

156 would permit the species to exist indefinitely”. On the other hand, a subset of the fundamental niche plus the biotic interactions constitute the realized niche (Peterson et al., 2011).

A heuristic tool appeared in the 2000s decade, which helped to clarify the niche concepts (Pulliam, 2000; Sorberón and Peterson, 2005; Peterson et al., 2011) in the field of niche modelling in ecology. This tool is a Venn diagram which let to define the realized niche or potential niche as merging points of biotic factors, abiotic factors and geography. We try to bring this scheme to the biosemiotic study of the semiotic niche in the landscape. The semiotic niche was defined by Hoffmeyer (2008) as “the signs or cues in the surroundings of an organism that it must be able to meaningfully interpret to ensure its survival and welfare”. There are three concepts (Organism, Resource and Eco-field) that we want to join in the Venn diagram (Fig 2). Thus, the intersection of the three concepts is the Realized (Landscape) Semiotic Niche (NSr). It is the place of the graph where the organisms found all the signs to interpret the eco-field to get resources. How- ever, there are other parts of the graph where either the organism or the resource is ab- sent. These parts are called Potential (Landscape) Semiotic Niche. As an example it can be mention the case of an alien or invasive species that has not yet arrived to a landscape but it could live perfectly there (SNp1). Another example could be a landscape which offers an ideal configuration for a predator to hunt, but the prey is not present (SNp2). There are other situations where the organism has available resources, but it cannot use them due to the lack of eco-field. In this case the organism cannot interpret the surroundings of this landscape, that is, there is not Semiotic Niche (nSN).

157 Figure 2.. Resource-Ecofield Venn diagram. SNr: Semiotic Niche Realized SNp: Semiotic Niche Potential nSN: non-Semiotic Niche.

An example of expansion of the Realized Semiotic Niche (Snr) is the case of the Red Turpentine Beetle (RTB) D. valens. This species is a secondary species in its native region in North America, in contrast, RTB is a primary species with high capacity of killing trees in the region of China where it lives as an invasive species (Sun et al., 2013). This novel behaviour is facilitated by the assemblage with two species, a fungus and other bark beetle. The fungus Leptographium procerum produces more amount of 3-carene than other fungi, so that RTB is more attracted to a host tree occupied by this fungus (Lu et al., 2010). Besides, RTB is attracted by chemicals compounds released by a secondary bark beetle Hylastes parallelus Chapuis, 1875, which in turn is attracted to 3-carene, also abundant in hindgut volatiles of RTB (Lu et al., 2007). While H. parallelus colonizes the roots, RTB habits the upper side of the tree. Both can overcome the tree defences. It has been showed that the NSr can expand towards the adjacent NSp if the organism can track the new signals of eco-field in the new invasive area. This leads to a further point of niche modelling combined with semiotics. Models of bark beetle outbreaks can be based in agent-based models with swarm intelligence algorithms. These models use the probability of the pheromone distribution (carrier meaning) plus

158 other variables which are associated to beetle density. This can determinate the future scope of insect outbreaks in this way (Pérez et al., 2013). The union of the biosemiotics and the modelling ecology can bring a fruitful future. The use of models based on the interchange of signals (ecomones, sounds, etc.) between organisms could be necessary in order to get more accurate predictions.

9.5 Eco-fields Networks as the beginning of Niche Construction

The Niche Construction paradigm is a new view of the ecological and evolution processes. The representamen that organisms use is the eco-field. This eco-field let to distinguish the resources in the surroundings. These resources can be the metabolic products, structural pieces, behavior movements or other kind of signs in the environment. Organisms can track resources of other counterpart organism with their respective eco-fields in the niche construction. But this kind of relationship can be more complex than only two species interacting in each other evolution. Thus, in this cascade of triadic relations, the Organism (Subject) change to → resource (Object) towards other organisms cohabitating the same territory. This creates an eco-field network (representamen network) between the members of one proto-holobiont where this network let the interchange of information, matter and energy. This network promotes an initial weak symbiotic relation which can evolve into a narrow mutualist relationship. It is important to point that the holobiont is the union of microorganisms which generally live in a euka- ryote (in this case the bark beetle) (Bordenstein & Theis, 2015). Besides the holobiont cannot be seen as a niche construction process, but the result of that process. Also the holobiont (sensu stricto) cannot be considered as a supra-organism, as there are a fuzzy line between interdependence of the different genetics networks. Then you only can see niche construction process in the relation of the representamen network among the different holobionts which are present in a specific place. For example, the bark-beetle- holobiont (formed by the insect and all microorganisms living inside or over it) with the cohabitant organisms such as fungi, mites, nematodes and bacteria which live inside the bark gallery but not inside the bark beetle. That is, the niche construction process it fin- ished when the different organisms created a common holobiont (ecofield network). This rigid and interdependent eco-field network is called mutualism relationship (+, +).

159 If there are few, sporadic weak or unidirectional links among the eco-field networks, a stable and interdependent relation cannot be created over time, so that would be the case of other types of symbiosis, such as amensalism (+,0), comensalism (+,-), predation (+,-) or parasitic (+,-).

Figure 3. Representation of the eco-field networks around a bark beetle (red square).

160 9.6 Eco-field in Tomicus species. A biosemiotic approach in the evolution of T. destruens and T. piniperda.

Experiments with T. destruens and T. piniperda have shown that these species are attracted to a combination of two substances (α-pinene and ethanol) released by the host tree. The substance α-pinene is usually produced by the host under regular conditions and ethanol is emitted when the tree is undergoing hydric stress conditions (Gallego et al., 2008, González-Rosa et al., 2013, Kelsey et al., 2013). However, other chemical compounds, such as the non-host allomone benzil-alcohol (Guerrero et al., 1997), produced by angiosperm trees, acts as repellent and provokes a repulsion response in this insect. The odourtope of these species is a combination of attractants and repellents, which let to find a suitable tree. Moreover, in the American genus Dendroctonus it has been shown that several species of fungi let to colonise the pine trees. Although Tomicus has not been as profusely studied as Dendroctonus, initial data indicate that a similar situation may occur in Tomicus with the fungi Lepthographium spp., Ophiostoma spp, Fusarium circinatum or Grossmania spp (Peverieri et al., 2006; Ben Jamaa et al., 2007; Romón et al., 2007; Bezos et al., 2015; Silva et al., 2015). It represents a scheme of the host tree search eco-field in Tomicus species.

The genus Tomicus is peculiar among bark beetles, as the maturation (feeding) phase does not take place under the bark, but on tender shoots. The maturation eco-field is the spatial configuration which has meaning-carrier to find a suitable resource (food in tender shoots). Hypothetically, an expansion of semiotic niche toward the use of tender shoots as resource occurred in ancestral Tomicus before species divergence. Whether this change of behaviour that let to explore and to eat the tender shoots, represented a nutritional advantage or a way to avoid competence with other bark beetles is a question that remains open. This semiotic niche expansion may have occurred in parallel to the ability to detect odours (or visual signals) from tender shoots. This is an example that a semiotic niche can provoke a change in the ecological niche that could culminate in a tangible evolutionary change and perhaps in species diversification in the genus Tomicus.

161 T. destruens and T. piniperda colonise different pine tree species, however, there are circumstances where both species live in sympatry in which they can be found together in the same host tree (P. pinaster and P. radiata). The competence is avoided through a temporal partition of the tender shoot resources, as T. destruens has its maturation flight from autumn to spring, in contrast to T. piniperda that flies on late spring and summer. That is, its semiotic niche was fitted to the climatic characteristic that let to optimise the use of the pine host tree. T. destruens tolerates high temperatures on Mediterranean pines and T. piniperda lives on continental pines with hard autumn-winter seasons. Mul- tiple factors may have modelled adaptations that have led to the avoidance of competence. Whether the modification of niche was a cause or a consequence of present distribution still remains an open question.

This paper has showed situations on how the bark beetle eco-field let interactions with others organisms. The relations of the semiotic realized niche with the organisms, resources and eco-field bring a new approach to study the interchange of signals among members of the landscape. Besides, the presence of new signals allows to infer the expansion of the realized semiotic niche towards potential semiotic niche. Also we explain a hypothetical origin of the mutualistic relations based on eco-field networks. Besides, we analyse the hypothetical expansion of semiotic niche that let to genus Tomicus to feed in tender shoots. The view of the bark-beetle consortium is an example on how the biosemiotic approach can help to analyze the biological patterns and processes from the signals point of view.

References

Adams, A. S., Aylward, F. O., Adams, S. M., Erbilgin, N., Aukema, B. H., Currie, C. R., Suen, G., & Raffa, K. F. (2013). Mountain pine beetles colonizing historical and naïve host trees are associated with a bacterial community highly enriched in genes con- tributing to terpene metabolism. Applied and Environmental Microbiology, 79(11), 3468–3475. Andersson, M. N. (2012). Mechanisms of odor coding in coniferous bark beetles: from neuron to behavior and application. Psyche: A Journal of Entomology, 2012, e149572. doi:10.1155/2012/149572

162 Ayres, B. D., Ayres, M. P., Abrahamson, M. D., & Teale, S. A. (2001). Resource parti- tioning and overlap in three sympatric species of Ips bark beetles (Coleoptera: Sco- lytidae). Oecologia, 128(3), 443–453. Barbieri, M. (2009). A short history of biosemiotics. Biosemiotics, 2(2), 221–245. Barker, G., & Odling-Smee, J. (2014). Integrating ecology and evolution: niche construction and ecological engineering. In G. Barker, E. Desjardins, and T. Pearce (Eds.), Entangled Life (pp. 187–211). Springer Netherlands. Ben Jamaa, M. L., Lieutier, F., Yart, A., Jerraya, A., & Khouja, M. L. (2007). The vir- ulence of phytopathogenic fungi associated with the bark beetles Tomicus piniperda and Orthotomicus erosus in Tunisia. Forest Pathology, 37(1), 51–63. Bezos, D., Martínez-Álvarez, P., Diez, J. J., & Fernández, M. M. (2015). The pine shoot beetle Tomicus piniperda as a plausible vector of Fusarium circinatum in northern Spain. Annals of Forest Science, 1–10. Boone, C. K., Keefover-Ring, K., Mapes, A. C., Adams, A. S., Bohlmann, J., & Raffa, K. F. (2013). Bacteria associated with a tree-killing insect reduce concentrations of plant defense compounds. Journal of Chemical Ecology, 39(7), 1003–1006. Boone, C. K., Six, D. L., Zheng, Y., & Raffa, K. F. (2008). Parasitoids and dipteran predators exploit volatiles from microbial symbionts to locate bark beetles. Environ- mental Entomology, 37(1), 150–161. Bordenstein, S. R., & Theis, K. R. (2015). Host biology in light of the microbiome: ten principles of holobionts and hologenomes. PLoS Bioliogy, 13(8), e1002226. Byers, J. A. (2012). Bark Beetles, Pityogenes bidentatus, orienting to aggregation pheromone avoid conifer monoterpene odors when flying but not when walking. Psyche: A Journal of Entomology, 2012, e940962. Byers, J. A., & Zhang, Q. (2011). Chemical ecology of bark beetles in regard to search and selection of host trees. In Recent Advances in Entomological Research (pp. 150–190). Springer. 5 Campbell, S. A., & Borden, J. H. (2006a). Close-range, in-flight integration of olfactory and visual information by a host-seeking bark beetle. Entomologia Experimentalis et Applicata, 120(2), 91–98. Campbell, S. A., & Borden, J. H. (2006b). Integration of visual and olfactory cues of hosts and non-hosts by three bark beetles (Coleoptera: Scolytidae). Ecological Entomo- logy, 31(5), 437–449.

163 Chen, H.-F., Salcedo, C., & Sun, J.-H. (2012). Male mate choice by chemical cues leads to higher reproductive success in a bark beetle. Animal Behaviour, 83(2), 421–427. Chiu, L., & Gilbert, S. F. (2015). The birth of the holobiont: multi-species birthing through mutual scaffolding and niche construction. Biosemiotics, 8(2), 191–210. Costa, A., & Reeve, J. D. (2011a). Upwind flight response of the bark beetle predator Thanasimus dubius towards olfactory and visual cues in a wind tunnel. Agricultural and Forest Entomology, 13(3), 283–290. Costa, A., & Reeve, J. D. (2011b). Olfactory experience modifies semiochemical responses in a bark beetle predator. Journal of Chemical Ecology, 37(11), 1166–1176. Dicke, M., & Sabelis, M. W. (1988). Infochemical terminology: based on cost-benefit analysis rather than origin of compounds? Functional Ecology, 2, 131–139. El-Hani, C. N., Arnellos, A., & Queiroza, J. (2007). Modeling a semiotic process in the immune system: signal transduction in B-cells activation. tripleC, 5(2), 24–36. Emilio Bruni, L. (2011). The multitrophic Plant–Herbivore–Parasitoid–Pathogen system: a biosemiotic perspective. In Towards a Semiotic Biology (pp. 143–166). Imperial College Press. Eudaldo González-Rosa, Sabah-Mazzetta, S., & Gallego. (2013). Atracción de Tomicus piniperda a compuestos cairomonales. 6o Congreso Forestal Español. Farina, A. (2008). The landscape as a semiotic interface between organisms and resources. Biosemiotics, 1(1), 75–83. Farina, A. (2010). Ecology, cognition and landscape: linking natural and social systems (Vol. 11). Springer Science & Business Media. Farina, A. (2011). A biosemiotic perspective of the resource criterion: toward a Gen- eral Theory of Resources. Biosemiotics, 5(1), 17–32. Farina, A. (2014). Soundscape and landscape ecology. In Soundscape Ecology (pp. 1– 28). Springer Netherlands. Farina, A., & Belgrano, A. (2004). The eco-field: a new paradigm for landscape ecology. Ecological Research, 19(1), 107–110. Farina, A., & Belgrano, A. (2006). The eco-field hypothesis: toward a cognitive landscape. Landscape Ecology, 21(1), 5–17. Farina, A., Lattanzi, E., Malavasi, R., Pieretti, N., & Piccioli, L. (2011). Avian soundscapes and cognitive landscapes: theory, application and ecological perspectives. Land- scape Ecology, 26(9), 1257–1267.

164 Farina, A., & Napoletano, B. (2010). Rethinking the landscape: new theoretical perspectives for a powerful agency. Biosemiotics, 3(2), 177–187. Farina, A., & Pieretti, N. (2013). From umwelt to soundtope: an epistemological es- say on cognitive ecology. Biosemiotics, 7(1), 1–10. Farina, A., & Pieretti, N. (2014). Acoustic codes in action in a soundscape context. Biosemiotics, 7(2), 321–328. Farina, A., Pieretti, N., & Malavasi, R. (2014). Patterns and dynamics of (bird) soundscapes: A biosemiotic interpretation. Semiotica, 2014(198).9 Favareau, D. D. (2009). Concepts of molecular biosemiotics. In Essential Readings in Biosemiotics (pp. 463–500). Springer Netherlands. Filotas, E., Parrott, L., Burton, P. J., Chazdon, R. L., Coates, K. D., Coll, L., Haeussier, S., Martin, K., Nocentini, S., Puettmann, K. J., Putz, F. E., Simard, S. W., & Messier, C. (2014). Viewing forests through the lens of complex systems science. Eco- sphere, 5(1), art1. Fleming, A. J., Lindeman, A. A., Carroll, A. L., & Yack, J. E. (2013). Acoustics of the mountain pine beetle (Dendroctonus ponderosae) (Curculionidae, Scolytinae): sonic, ul- trasonic, and vibration characteristics. Canadian Journal of Zoology, 91(4), 235–244. Gallego, D., Galián, J., Diez, J. J., & Pajares, J. A. (2008). Kairomonal responses of Tomicus destruens (Col., Scolytidae) to host volatiles α-pinene and ethanol. Journal of Applied Entomology, 132(8), 654–662. Gitau, C. W., Bashford, R., Carnegie, A. J., & Gurr, G. M. (2013). A review of semio- chemicals associated with bark beetle (Coleoptera: Curculionidae: Scolytinae) pests of coniferous trees: A focus on beetle interactions with other pests and their associates. Forest Ecology and Management, 297, 1–14. Goodsman, D. W., Erbilgin, N., & Lieffers, V. J. (2012). The Impact of phloem nutrients on overwintering mountain pine beetles and their fungal symbionts. Environmental Entomology, 41(3), 478–486. Guerrero, A., Feixas, J., Pajares, J., Wadhams, L. J., Pickett, J. A., & Woodcock, C. M. (1997). Semiochemically induced inhibition of behaviour of Tomicus destruens (Woll.) (Coleoptera: Scolytidae). Naturwissenschaften, 84(4), 155–157. Hoffmeyer, J. (2008). The semiotic niche. Journal of Mediterranean Ecology, 9, 5– 30. Hofstetter, R. W., & Moser, J. C. (2014). The role of mites in insect-fungus associ-

165 ations. Annual Review of Entomology, 59(1), 537–557. Kelsey, R. G., Gallego, D., Sánchez-García, F. J., & Pajares, J. A. (2014). Ethanol ac- cumulation during severe drought may signal tree vulnerability to detection and attack by bark beetles. Canadian Journal of Forest Research, 44(6), 554–561. Kolosova, N., & Bohlmann, J. (2012). Conifer defense against insects and fungal pathogens. In R. Matyssek, H. Schnyder, W. Oßwald, D. Ernst, J. C. Munch, & H. Pretz- sch (Eds.), Growth and Defence in Plants (pp. 85–109). Springer Berlin Heidelberg. Kull, K. (2010). Ecosystems are made of semiosic bonds: consortia, umwelten, bio- phony and ecological codes. Biosemiotics, 3(3), 347–357. Laland, K. N., Uller, T., Feldman, M. W., Sterelny, K., Müller, G. B., Moczek, A., Jablonka, E., & Odling-Smee (2015). The extended evolutionary synthesis: its structure, assumptions and predictions. Proccedings of the Royal Society B, 282(1813), 20151019. Laland, K., Uller, T., Feldman, M., Sterelny, K., Müller, G. B., Moczek, A., Jablonka, E., Odling-Smee, J., Wray, G. A., Hoekstra, H. E., Futuyma, D. J., Lenski, R. E., Mackay, T. F. C., Schluter, D., & Strassmann, J. E. (2014). Does evolutionary theory need a rethink? Nature, 514(7521), 161–164. Lindeman, A. A., & Yack, J. E. (2015). What is the password? Female bark beetles (Scolytinae) grant males access to their galleries based on courtship song. Behavioural Processes, 115, 123–131. Lu, M., Miller, D. R., & Sun, J.-H. (2007). Cross-attraction between an exotic and a native pine bark beetle: a novel invasion mechanism? PLoS ONE, 2(12), e1302. Lu, M., Wingfield, M. J., Gillette, N. E., Mori, S. R., & Sun, J.-H. (2010). Complex interactions among host pines and fungi vectored by an invasive bark beetle. New Phytologist, 187(3), 859–866. Lu, R. C., Wang, H. B., Zhang, Z., Byers, J. A., Jin, Y. J., Wen, H. F., & Shi, W. J. (2012). Coexistence and competition between Tomicus yunnanensis and T. minor (Cole- optera: Scolytinae) in Yunnan Pine. Psyche: A Journal of Entomology, 2012, e185312. Machingambi, N. M., Roux, J., Dreyer, L. L., & Roets, F. (2014). Bark and ambrosia beetles (Curculionidae: Scolytinae), their phoretic mites (Acari) and associated Geo- smithia species (Ascomycota: Hypocreales) from Virgilia trees in South Africa. Fungal Biology, 118(5–6), 472–483. Malavasi, R., Kull, K., & Farina, A. (2014). The acoustic codes: how animal sign processes create sound-topes and consortia via conflict avoidance. Biosemiotics, 7(1),

166 89–95. Maran, T. (2012). Are ecological codes archetypal structures? Semiotics in the wild, 147–156. Maran, T., & Kull, K. (2014). Ecosemiotics: main principles and current develop- ments. Geografiska Annaler: Series B, Human Geography, 96(1), 41–50. Matthews, B., De Meester, L., Jones, C. G., Ibelings, B. W., Bouma, T. J., Nuutinen, V., van de Koppel, J. And Odling-Smee, J. (2013). Under niche construction: an opera- tional bridge between ecology, evolution, and ecosystem science. Ecological Mono- graphs, 84(2), 245–263. Miller, D. R., Gibson, K. E., Raffa, K. F., Seybold, S. J., Teale, S. A., & Wood, D. L. (1997). Geographic variation in response of pine engraver, Ips pini, and associated species to pheromone, lanierone. Journal of Chemical Ecology, 23(8), 2013–2031. Nielsen, S. N. (2007). Towards an ecosystem semiotics: some basic aspects for a new research programme. Ecological Complexity, 4(3), 93–101. 1 Pasteels, J. M. (1982). Is kairomone a valid and useful term? Journal of Chemical Ecology, 8(7), 1079–1081. Pérez, L., Dragićević, S., & White, R. (2013). Model testing and assessment: Per- spectives from a swarm intelligence, agent-based model of forest insect infestations. Computers, Environment and Urban Systems, 39, 121–135. Peterson, A. T., Soberón, J., Pearson, R. G., Anderson, R. P., Martínez-Meyer, E., Na- kamura, M., & Araújo, M. B. (2011). Ecological niches and geographic distributions (MPB-49). Princeton University Press. Peverieri, G. S., Capretti, P., & Tiberi, R. (2006). Associations between Tomicus destruens and Leptographium spp. in Pinus pinea and P. pinaster stands in Tuscany, central Italy. Forest Pathology, 36(1), 14–20. Pizzolotto, R. (2009). Characterization of different habitats on the basis of the species traits and eco-field approach. Acta Oecologica, 35(1), 142–148. Pulliam, H. r. (2000). On the relationship between niche and distribution. Ecology Letters, 3(4), 349–361. Raffa, K. F., Hobson, K. R., LaFontaine, S., & Aukema, B. H. (2007). Can chemical communication be cryptic? Adaptations by herbivores to natural enemies exploiting prey semiochemistry. Oecologia, 153(4), 1009–1019. Reeve, J. D., Strom, B. L., Rieske, L. K., Ayres, B. D., & Costa, A. (2009). Geo-

167 graphic variation in prey preference in bark beetle predators. Ecological Entomology, 34(2), 183–192. Romón, P., Iturrondobeitia, J. C., Gibson, K., Lindgren, B. S., & Goldarazena, A. (2007). Quantitative association of bark beetles with pitch canker fungus and effects of verbenone on their semiochemical communication in monterey pine forests in northern Spain. Environmental Entomology, 36(4), 743–750. Seybold, S. J., Huber, D. P. W., Lee, J. C., Graves, A. D., & Bohlmann, J. (2006). Pine monoterpenes and pine bark beetles: a marriage of convenience for defense and chemical communication. Phytochemistry Reviews, 5(1), 143–178. Silva, X., Terhonen, E., Sun, H., Kasanen, R., Heliövaara, K., Jalkanen, R., & Asiegbu, F. O. (2015). Comparative analyses of fungal biota carried by the pine shoot beetle (Tomicus piniperda L.) in northern and southern Finland. Scandinavian Journal of Forest Research, 30(6), 497–506. Six, D. L. (2012). Ecological and evolutionary determinants of bark beetle —fungus symbioses. Insects, 3(1), 339–366. Six, D. L. (2013). The bark beetle holobiont: why microbes matter. Journal of Chem- ical Ecology, 39(7), 989–1002. Sorberón, J. (2005). Interpretation of models of fundamental ecological niches and species’ distributional areas. Biodiversity Informatics, 2, 1–10. Sõukand, R., & Kalle, R. (2010). Plant as object within herbal landscape: different kinds of perception. Biosemiotics, 3(3), 299–313. Strom, B. L., & Goyer, R. A. (2001). Effect of silhouette color on trap catches of Dendroctonus frontalis (Coleoptera: Scolytidae). Annals of the Entomological Society of America, 94(6), 948–953. Sun, J., Lu, M., Gillette, N. E., & Wingfield, M. J. (2013). Red turpentine beetle: in- nocuous native becomes invasive tree killer in China. Annual Review of Entomology, 58(1), 293–311. Susoy, V., & Herrmann, M. (2014). Preferential host switching and codivergence shaped radiation of bark beetle symbionts, nematodes of Micoletzkya (Nematoda: Diplo- gastridae). Journal of Evolutionary Biology, 27(5), 889–898. Vasechko, G. I. (1978). Host selection by some bark beetles (Col., Scolytidae). Zeits- chrift für Angewandte Entomologie, 85(1-4), 141–153. Vega, F. E., & Hofstetter, R. W. (2014). Bark Beetles: Biology and Ecology of Native

168 and Invasive Species. Academic Press. Xiaoyi, W., & Zhongqi, Y. (2008). Behavioral mechanisms of parasitic wasps for searching concealed insect hosts. Acta Ecologica Sinica, 28(3), 1257–1269. Xu, L., Lou, Q., Cheng, C., Lu, M., & Sun, J. (2015). Gut-associated bacteria of Dendroctonus valens and their Involvement in verbenone production. Microbial Eco- logy, 1–12. Yang, L. H., & Gratton, C. (2014). Insects as drivers of ecosystem processes. Current Opinion in Insect Science, 2, 26–32.

169

10 ComentariosComentarios generalesgenerales yy conclusionesconclusiones

170 En esta tesis se han abordado aspectos filogeográficos y genómicos de los escarabajos de corteza para entender mejor sus relaciones evolutivas y su ecología. Además se ha realizado una aproximación biosemiótica para analizar las implicaciones de la comunic- ación de los escolítidos en las redes ecológicas. De acuerdo a los aspectos ecólogicos, se ha analizado una red haplotípica de poblaciones mediterráneas del escolítido Tomicus destruens. De dicha red, se ha determinado que ocurrió un hipotético cambio de nicho de los haplotipos adaptados a las variables climáticas más extremas. En relación a los aspectos genómicos, dos microRNAs previamente detectados en una genoteca de la especie T. yunnanensis, (mir-2c-3p and mir-4944-5p) se han validado en esta especie y en la congenérica T. destruens. También se han identificado y caracterizado cuatro proteasas semejantes a trypsina, cinco proteasas semejantes a quimotrypsina, encargadas de procesos de digestión extracelular y cuatro proteasas con domino clip relacionadas con procesos del sistema inmune y del desarrollo embrionario. En el último capítulo se de- scriben las relaciones interespecificas e intraespecificas de los escolítidos en términos biosemióticos, determinando a su vez el nicho semiótico de estos coleópteros y sus implicaciones evolutivas en las redes ecológicas de un paisaje forestal.

10.1 Capítulo 1. Distribución de linajes mitocondriales de Tomicus destruens (Coleoptera: Scolytinae): perspectivas filogeográficas y modelado de nicho.

1. Los modelos ecológicos desarrollados indican que los linajes minoritarios de ADN mitocondrial de T. destruens difieren en su nicho ecológico potencial de acuerdo a su relación con las variables climáticas extremas. Por el contrario, los linajes occidentales, más extendidos, muestran una estrecha relación con su árbol hospedador Pinus.

2. Los modelos de distribución de los clados occidentales (3-2, 3-1) y los haplotipos mayoritarios (AA, AC) presentan una mayor contribución de las variables de cobertura de Pinus halepensis y Pinus pinaster.

3. El modelo de distribución del clado oriental (3-3) y de los haplotipos minoritarios (AD, AS, ZZ) se caracteriza con una mayor contribución de las variables climáticas me-

171 dias (temperatura media anual, isotermalidad, estacionalidad de la precipitación y de la temperatura) y variables climáticas extremas (precipitación del cuatrimestre más frio y media de la temperatura del cuatrimestre más seco).

4. La escasa asociación entre Pinus brutia y el linaje 3-3 explicaría que un proceso de cambio de nicho está en progreso, y que aún es insuficiente para mostrar una estructura genética de la población oriental asociada a su árbol hospedador, P. brutia.

5. Los haplotipos minoritarios (AD, AS, ZZ) hipotéticamente se escindieron de los haplotipos mayoritarios (AA, AC). Los linajes AD y AS quedaron en zonas con variables climáticas más extremas por medio de exclusión competitiva. Y el haplotipo ancestral ZZ estaba pre-adaptado a atacar a P. brutia en los ambientes del este de la cuenca del mediterráneo.

6. El alto índice de diversidad de haplotipos en el sur de Anatolia, península helén- ica, península ibérica, norte de África y Oriente Medio sugieren que ha sido un refugio glacial para T. destruens.

10.2 Capítulo 2. Predicción in silico y caracterización de microARNs del escarabajo de la corteza Tomicus yunnanensis y su validación en T. destruens (Coleoptera, Curculionidae, Scolytinae).

1. Se han identificado 228 pre-miRNAs utilizando parámetros estructurales y ter- modinámicos de un transcriptoma de T. yunnanensis. La mayoría de los pre-miRNAs (84,21 %) son homólogos de insectos, habiendo un pequeño porcentaje (5,26 %) cuya homología era con virus o nematodos.

2. Se detectaron 71 miRNAs por medio de aproximaciones estadísticas y de un proto- colo de elaboración propia. Dos microRNAs seleccionados (tyu-mir-2c-3p and tyu-mir- 4944-5p) que tenían homología con la base de datos de miRNAs de Drosophila melanogaster, obtuvieron expresión positiva en T. destruens y T. yunnanensis.

172 3. Tyun-mir-2c-3p presenta una alta expresión en T. destruens respecto T. yunnanensis. Las hembras muestra una expresión más baja que en machos en la especie T. destruens. Tyu-mir-2c-3p estaría relacionado en la regulación de los procesos de ovoo- génesis y vitelogénesis.

4. Tyun-mir-4944-5p presenta una alta expresión en hembras respeto a machos en la especie T. destruens.

10.3 Capítulo 3. Análisis de un transcriptoma, identificación y caracterización in silico de nuevas serinas proteasas del escarabajo de la corteza Tomicus yunnanensis.

1. Se han obtenido 346 contigs con la anotación de serina proteasas. Solamente 13 contigs presentaban el péptido señal y la presencia de un dominio de serina proteasa.

2. Entre las serina proteasas identificadas, cuatro fueron caracterizadas como tripsinas y cinco como quimotripsinas por la presencia del dominio de unión al substrato por los aminoácidos Ácido Aspártico-Glicina-Glicina o Glicina-Glicina-Glicina, respectiva- mente.

3. Se detectó en las quimiotripsinas variaciones en los aminoacidos adyacentes al domino de unión a substrato. Estas diferencias podrían estar relacionadas con la plasti- cidad de las quimiotripsinas a unirse a diferentes substratos además de ser un mecan- ismo para evitar la acción de los inhibidores de proteasas presentes en los pinos.

4. Otras cuatro serina proteasas presentaban un dominio clip, el cual está relacionado con el sistema inmune y el desarrollo embrionario. Estas proteínas no presentaban una triada catalítica completa y podrían ser clasificadas como pseudoproteasas.

5. Las relaciones filogenéticas de las proteasas identificadas se clasifican en tres cla-

173 dos formados por tripsinas, quimotripsinas y domino clip. Dentro de cada clado, las pro- teínas no presentan una congruencia taxonómica, sino que presentan homologias con otras especies de insectos, tal y como cabría esperar si han evolucionado por duplicación de genes para producir distintos parálogos.

6. Las enzimas digestivas más expresadas son serina proteasas, lipasas y beta-glucosidasas. Estas diferencias observadas para T. yunnanensis en relación con otras especies podrían explicarse por la adaptación de este insecto a una dieta a base de coníferas

7. Se han obtenido 115.942 contigs (55,23 %) que no mostraban similaridad con pro- teinas hómologas en las bases de datos. Estos contigs quizás representan genes de evolu- ción rápida o genes taxonómicamente restringidos

8. Nuestros contigs mostraban una alta similitud con secuencias de coleópteros, como el escolítido Dendroctonus ponderosae, también perteneciente a la tribu Tomicini).

10.4 Capítulo 4. Aplicando el eco-field y la Teoría General de Recursos a los escarabajos de corteza: Más allá de la Teoría de la Construcción de Nicho.

1. Se ha descrito el eco-field que permite a los escarabajos forestales encontrar su ár- bol hospedador y encontrar a su pareja para la reproducción. Para ello estos escolítidos se guían por diversas señales semióticas (olores, vibraciones) usando un eco-field es- pecífico para cada función (alimentación, reproducción).

2. Varias especies (escolítidos, ácaros, bacterias, nematodos y hongos) permiten el ensamblaje de la comunidad forestal, mediante el intercambio de señales que permiten una red ecológica fuerte en el tiempo. Este ensamblaje está mantenido por medio de la compartición de redes de eco-fields. 3. Se ha definido la entidad de nicho semiótico realizado y nicho semiótico potencial, además de las relaciones entre los conceptos organismo, recurso y eco-field. Se propone la hipótesis que plantea que cambios en la configuración del eco-field permiten al organ-

174 ismo ampliar su nicho semiótico realizado hacia el nicho semiótico potencial.

4. Se propone la hipótesis que plantea que la interdependencia de los eco-field puede iniciar el proceso de construcción de nicho. Si esta interdependencia es crucial el proceso puede finalizar en relaciones mutualistas.

5. El comportamiento de búsqueda de recursos alimenticios para madurar, por parte de los Tomicus ancestrales, es una novedad evolutiva en los escolitidos que puede haberse iniciado mediante la expansión del nicho semiótico. La coexistencia simpátrida de Tomicus destruens con Tomicus piniperda se ha mostrado como el resultado de una partición de recursos alimenticios. Dicha partición es mantenida por que ambas especies tienen un eco-field de búsqueda de recursos para la maduración con diferente estacionalidad.

175 11 GeneralGeneral commentscomments andand conclusionsconclusions

176 This thesis has addressed phylogeographic and genomic aspects of bark beetles to better understand their ecology and evolutionary relationships. In addition a biosemiotic approach was performed to analyse the implications of the communication of bark beetles in ecological networks. In terms of ecological aspects, in the first chapter a haplotype network of the Mediterranean populations of the scolytid beetle Tomicus destruens have been analysed. From the network obtained, it has been inferred that haplotypes adapted to extreme climatic variables have undergone a hypothetical niche shift. Concerning the genomic aspects, in the second chapter, two microRNAs previously detected in the species T. yunnanensis (miR-2c-3p and miR-4944-5p) have been validated, in this species and in its congeneric T. destruens. In chapter three, four trypsin-like proteases and five quimotrypsin-like proteases responsible for extracellular digestion processes have been identified and characterized. Additionally, four proteases with clip domain related to the immune system and processes of embryonic development have also been identified and characterized. In the fourth chapter, interspecific and intraspecific relations are described in biosemiotics terms, determining the semiotic niche of these beetles and their evolutionary implications in ecological networks of a forest landscape.

11.1 Chapter 1. Distribution of Tomicus destruens (Coleoptera: Scolytinae) mitochondrial lineages: phylogeographic insights and niche modelling

1. The ecological models developed indicate that minority eastern mtDNA lineages of T. destruens differ in their potential ecological niche according to their relation to extreme climatic variables. By contrast, the most widespread western lineages display a close relationship with their Pinus host tree.

2. Coverage of host tree species P. pinaster and P. halepensis occurred in the models as the variables with a major contribution in the western clades 3-1 and 3-2.

3. The distribution model of the oriental clade (3-3) and the minority haplotypes

177 (AD, AS, ZZ) is characterized by a greater contribution from average climatic variables (annual mean temperature, isothermality, temperature seasonality and precipitation seasonality) and extreme climatic variables (precipitation of coldest quarter and mean temperature of driest quarter).

4. The poor association between Pinus brutia and lineage 3-3 suggests that a niche shifting process is in progress, but it is still insufficient to show a genetic structure of the eastern population associated with its host tree, P. brutia.

5. Minor haplotypes (AD, AS, ZZ) were likely excised from the major haplotypes (AA, AC). AD and AS lineages were likely segregated in areas with extreme climatic variables through competitive exclusion. The ancestral haplotype ZZ was pre-adapted to attack P. brutia in environments East of the Mediterranean basin.

6. The high level of haplotype diversity in Southern Anatolia, Helenic Peninsula, Iberian Peninsula, North Africa and the Middle East suggests that those areas acted as a glacial refuge for T. destruens.

11.2 Chapter 2 In Silico prediction and characterization of microRNAs from the bark beetle Tomicus yunnanensis and validation in T. destruens (Coleoptera, Curculionidae, Scolytinae)

1. We have obtained 228 pre-miRNAs from the library of T. yunnanensis, of which 84.21 % resulted to be homologous to insect sequences, having a small percentage (5.26 %) whose homology was to viruses or nematodes.

2. Seventy one miRNAs were detected through statistical approaches and following a newly designed protocol by the authors. Two microRNAs selected (tyu-mir-2c-3p and tyu-mir-4944-5p) having homology with the miRNAs database of Drosophila melanogaster, was shown to have positive expression in T. destruens and T. yunnanensis.

3. Tyun-mir-2c-3p shows high expression in T. destruens compared to T. yunnanen-

178 sis. Females of T. destruens show a lower expression than males. Tyu-mir-2c-3p would seems to be related to the regulation of oogenesis and vitellogenesis processes.

4. Tyun-mir-4944-5p shows higher expression in females than in males in the species T. destruens.

11.3 Chapter 3. Transcriptome analysis and in silico identification and characterization of novel serine proteases in the bark beetle Tomicus yunnanensis.

1. We obtained 346 contigs annotated as serine proteases. Only 13 contigs had the signal peptide and the presence of a serine protease domain.

2. Four of the serine proteases identified, were characterized as trypsin and five as chymotrypsin by the presence of the substrate binding domain with the amino acids Gly- cine-Glycine-Aspartic Acid and Glycine-Glycine-Glycine respectively.

3. Chymotrypsins had variations in the amino acids adjacent to the substrate binding domain. These differences could be related to the plasticity of chymotrypsins to join different substrates and/or being a mechanism to prevent the action of the protease inhibitors present in pines.

4. Four other serine proteases showed a clip domain, which is related to the immune system and to the embryonic development. These proteins did not have a complete catalytic triad and could be classified as pseudoproteases.

5. The phylogenetic relationships of the identified proteases are classified into three clades formed by trypsin, chymotrypsin and clip domain. Within each clade, protein groups do not show a taxonomic consistency, but rather show homologies with other insect species, as would be expected if they have evolved by gene duplication to produce different paralogs.

179 6. The more expressed digestive enzymes are serine proteases, lipases and beta-glucosidases. These differences observed for T. yunnanensis in relation to other species could be explained by adaptations to a conifers diet in this insects.

7. A high number of the contigs obtained (115942; 55.23%) showed no similarity to homologous proteins in the databases. These contigs may represent rapidly evolving or taxonomically restricted genes.

8. Our contigs showed high similarity to beetle sequences such as the scolytid Dendroctonus ponderosae, also belonging to the tribe Tomicini.

11.4 Chapter 4. Applying eco-field and the General Theory of Resources to bark beetles: Beyond the Niche Construction Theory

1. We describe the eco-field that allows forestry beetles to find their host tree and to find a partner for reproduction. These bark beetles guide themselves towards different semiotic signals (odours, vibrations) using an specific eco-field for each function (feeding, reproduction).

2. Several species (bark beetles, mites, bacteria, nematodes and fungi) allow the assembly of the forest community, through the exchange of signals that allows a strong ecological network through time. This assembly is maintained through sharing eco-field networks.

3. We define the realized semiotic niche and potential semiotic niche in addition to the relationships of organisms, resources and eco-fields. It has been hypothesized that changes in the eco-field configuration allows the expansion of the realized semiotic niche made towards the potential semiotic niche.

4. It has been hypothesized that the interdependence of eco-fields can initiate niche construction processes. Also, if this interdependence is essential, it can end up in mutualistic relationships.

180 5. Seeking resources behaviour by the ancestral Tomicus to allow maturation is an evolutionary novelty in bark beetles, which may have been initiated by the expansion of the semiotic niche. The sympatric coexistence of T. destruens with T. piniperda has proven to be the result of a partition of food resources. This partition is maintained as both species have a searching resources eco-field with different seasonality.

181 12 AppendixAppendix

182 Org Divers Evol DOI 10.1007/s13127-014-0186-2

ORIGINAL ARTICLE

Distribution of Tomicus destruens (Coleoptera: Scolytinae) mitochondrial lineages: phylogeographic insights and niche modelling

F. Javier Sánchez-García & José Galián & Diego Gallego

Received: 29 October 2013 /Accepted: 6 October 2014 # Gesellschaft für Biologische Systematik 2014

Abstract This paper presents a novel approach of population haplotypes of T. destruens, possibly by involving a balance genetics together with environmental and biogeographic data between the length of the flight reproductive period of leading to inferences for ecological niche modelling. We used T. destruens and the status of the host tree vigour and growth hierarchical lineages obtained using the nested cladistic anal- stage. This study illustrates a good example of the benefits that ysis (NCA) of the mitochondrial DNA (mtDNA) haplotypes ecological niche modelling provides to understand population of the bark beetle species Tomicus destruens, for modelling genetic and phylogeographic patterns. the distribution by maximum entropy using environmental and host variables along the whole Mediterranean Basin. Keywords Ecological niche modelling . Intraspecific The identity and similarity tests were checked in the intraspe- lineages . NCA-MaxEnt . Scolytinae . Coleoptera . Pinus cific lineages (NCA clades and haplotypes) in order to determine a shift or conservatism niche between them. Also, four indices from nine geographical areas in the Mediterranean Introduction Basin were calculated to assess the variability of environmental factors shaping the distribution of haplotype diversity on a Although necessarily gene-focused, phylogeographic studies large geographic scale. The ecological models developed provide keys to strengthen our knowledge of the associations indicate that minority eastern mtDNA lineages of between DNA lineages and environmental predictors, im- T. destruens differ in their potential ecological niche according proving our understanding of the relationship between habitat to their relation to extreme climatic variables. By contrast, the diversity and competitive exclusion of phylogenetic lineages most widespread western lineages display a close relationship within a species (Newman et al. 2011). This approach con- with their Pinus host tree. Also, higher levels of exclusive and tributes to the characterization of the ecological niche of a endemic haplotypes were predicted in areas with high tem- studied species by recording a set of representative biotic and perature variability in the Mediterranean wet period. The abiotic variables (e.g. type of habitat, climatic variables, etc.) eastern group niche seems to be included in part of the range with an important role in understanding the biology and of the ecological space of the two major western clades. This natural history of organisms (Buckley 2009) and large-scale result suggests that a niche shift might have started, being still biogeographic patterns (Wiens 2011). Recently, several stud- an early relationship with its host tree Pinus brutia. ies have initiated this approach, linking ecological niche Alternatively, the temperature variability in the wettest period modelling with phylogeographic analysis in an attempt to appears to be related to a high proportion of endemic answer questions related to evolutionary processes within sister species or intraspecific lineages (Graham et al. 2004; Jakob et al. 2007; Gallego and Galián 2008; McCormack et al. Electronic supplementary material The online version of this article 2010; Hundsdoerfer et al. 2011; Pearman et al. 2010; Oney (doi:10.1007/s13127-014-0186-2) contains supplementary material, which is available to authorized users. et al. 2013; Schulte et al. 2012). In these studies, correlative methods have been used to estimate the environmental condi- * : : F. J. Sánchez-García ( ) J. Galián D. Gallego tions that are suitable for a particular species or lineage by Departamento de Zoología y Antropología Física, Universidad de Murcia, Campus de Espinardo, 30100 Murcia, Spain associating known occurrence records of the species with a set e-mail: [email protected] of environmental variables (Pearson 2007). However, F.J. Sánchez-García et al. restrictions due to sample size could limit the use of regression variables in the distribution of the local genealogy. The analysis methods with phylogeographic datasets. Methods less depen- revealed a high level of genetic diversity in the forest of Sierra dent on sample size, such as maximum entropy modelling, Espuña and a microdistribution of the mtLs related to altitude have become available (Phillips et al. 2006). The widely used and putative niche shifts between lineages associated with the MaxEnt algorithm extrapolates a set of georeferenced occur- micro-environmental conditions of their host pine trees rence locations of a species and a set of layers of biotic and (Gallego and Galián 2008). Those authors hypothesised that abiotic variables, for which, maximizing the information en- habitat heterogeneity allows the fixation of mtLs in tropy subjected to the constraints which are imposed by the T. destruens. prior knowledge of the occurrence points and the environmen- That hypothesis, erected on the basis of a fine-scale ap- tal variables (Phillips et al. 2004), producing a model of the proach, is worth testing on a broader scale. To do that, we range distribution of the given species. Using the distribution applied a combined methodology for modelling distribution of mitochondrial lineages (mtLs) within a species instead of and explored the ecological niche of hierarchical lineages of merely occurrence locations of the species as input, the meth- T. destruens on a Mediterranean scale. Starting from this idea, od has great potential for its application in spatial modelling of the aim of this work was to identify some environmental phylogenetic datasets. factors of the ecological niche that affect the distribution of Several studies focused on the bark beetle Tomicus the mtLs and the genetic diversity of T. destruens using NCA destruens (Wollaston, 1865) were developed in the last decade together with the MaxEnt algorithm (NCA-MaxEnt) in order as an outcome of the last taxonomic separation of its sister to detect processes of ecological niche segregation of the species Tomicus piniperda (Linnaeus 1758)(Gallego and mitochondrial DNA (mtDNA) lineages along the entire Galián 2001; Kerdelhué et al. 2002;Kohlmayretal.2002). Mediterranean Basin. T. destruens (Wollaston, 1865) exhibits a circum- Mediterranean and Macaronesian distribution (Wood and Bright 1992) and attacks several Pinus species, including Pinus halepensis, Pinus pinaster, Pinus pinea, Pinus brutia Methods and P.canariensis and occasionally Pinus nigra (Gallego et al. 2004; Vasconcelos et al. 2006), being responsible for part of Haplotype dataset and beetle sampling the decline of pine forests (Chakali 2005; Faccoli et al. 2005a, b). The distribution of this species is influenced mainly by Sequences from two different sources were used to cover the host distribution (Kerdelhué et al. 2002;Hornetal.2006), in entire Mediterranean Basin: (i) haplotype sequences from which climatic factors related to temperature and water avail- GenBank accession numbers AF457827, AF457831– ability can impose a dual limitation on the scolytid species and AF457852, AF457854, AF457859–AF457861, DQ182709– the host trees (Gallego et al. 2004). DQ182712, DQ182714, DQ182716–DQ182731, DQ182733, Phylogeographic approaches have facilitated the clarifica- DQ182734 and DQ295748–DQ295777 (Kerdelhué et al. tion of the origin and structure of T. destruens populations in 2002; Vasconcelos et al. 2006;Hornetal.2006); and (ii) the Mediterranean Basin (Faccoli et al. 2005a, b;Hornetal. haplotype sequences from freshly collected insects from not 2006; Vasconcelos et al. 2006). According to Horn et al. sampled areas before located in central and south-eastern (2006), the populations of T. destruens are phylogeographically Iberia and the Anatolian Peninsula (Table 1). The Turkish structured into western and eastern groups, with a contact zone locality fills the Anatolian sampling gap at the eastern on the Adriatic coast of Italy. Probably, two glacial refugia Mediterranean area. Additionally, some samples were added existed in the western Mediterranean area (the Iberian and from the Sierra Espuña Natural Park as a zone of high Italian peninsulas), where a high diversity of haplotypes and haplotypic variability (Gallego and Galián 2008). Iberian adult a scarce spatial structure in haplotype distribution have been beetles were collected using Crosstrap® (Econex, Murcia, found (Horn et al. 2006; Vasconcelos et al. 2006). Conversely, Spain). Adult male and female beetles were attracted using a the eastern group was characterized by a significant phylogeo- commercial lure of α-pinene and ethanol (Econex S.L., Murcia, graphic pattern and low levels of gene flow (Horn et al. 2006). Spain). A cup with 100 ml of pure propylene glycol for DNA The relationship between environmental variables and the dis- preservation (Vink et al. 2005) was attached to the bottom of the tribution of mtLs of T. destruens has been studied on a fine trap funnel. The Turkish samples were obtained directly by scale in the pine forests of Sierra Espuña (south-eastern Spain) extensive collection from infested trees or wood traps. After by Gallego and Galián (2008). A phylogeographic method, collection, insects were immersed in absolute ethanol such as nested clade analysis (NCA) (Templeton et al. 1995), for subsequent DNA analysis and long-term preserva- combined with a statistical model, such as the generalized tion. Taxonomic identification of the 48 beetles was con- additive model (GAM), in an NCA-GAM approach was im- firmed molecularly as T. destruens by the proposed method plemented to investigate in detail the role of five environmental (Gallego and Galián 2001). Distribution of Tomicus destruens mitochondrial lineages

Table 1 List of the new populations molecularly analysed of T. destruens

Code Country Locality Pine host Date Collector Latitude Longitude Altitude (m)

Esp-Spa Spain Espuña P. halepensis 05 October 2005 D. Gallego 37° 51′ N1°29′ W807 Mon-Spa Spain San Martín de Montalbán P. halepensis 17 October 2010 E. Gómez 39° 45′ N4°25′ W571 Yeb-Spa Spain Los Yébenes P. halepensis 19 October 2005 E. Gómez 39° 25′ N4°04′ W884 Yun-Tur Turkey Yuntdag P. brutia 26 October 2010 Ö. Toprak 38° 42′ N27°16′ W130

DNA extraction and amplification pinaster andP.sylvestris. These host distributions were taken in vectorial format and rasterized as the coverage ratio of host Because high numbers of nematodes are often found in the pine species in a pixel of working scale of 100 km2 using abdomen, only the head, thorax and legs of the Tomicus ArcGis v9.2 (ESRI, Redlands, CA, USA). In order to reduce beetles were used. DNA isolation was done with the Dneasy the data processing time, we masked the study area by the Tissue Kit (QIAGEN®, Valencia, CA, USA), following the presence distribution of all considered host tree species. The manufacturer’s recommendations. We amplified a fragment of predictor dataset was included in a matrix of 26 columns the mitochondrial genes cytochrome c oxidase I (COI) and II (independent or predictor variables) and 54,667 rows. We (COII) that flank the tRNleu gene, by PCR. Specific primers used cluster analysis for correlated variables applying the for T. destruens (primer pair 5′-CCTCATCATTATGAGCTA Raster package (Hijmans and van Etten 2012) with R software TTGG-3′,5′-TCATAGGATCAATATCATTG-3′;secondpair (R Core Team 2012) to avoid collinearity. Then, highly cor- 5′-TCAATAGGAGCAGTATTTGCTA-3′,5′-AAGTAATC related climatic and topographic variables (>0.75) with a GTAAAGACGGAAGA-3′) were used as described biological meaning (Warren and Seifert 2010;Dormann (Kerdelhué et al. 2002). PCR was done in 12.5 μlreactions et al. 2012) were chosen. using Ready-To-Go PCR Beads (GE Healthcare, Bucks, UK) All the sequences obtained in this work and those following the PCR cycling programme described by downloaded from GenBank were aligned using Clustal W Kerdelhué et al. (2002). PCR products were purified using version 1.4 (Thompson et al. 1994) as implemented in Mega the standard isopropanol/ammonium acetate method 4.0 (Tamura et al. 2007). The hierarchical lineages were (Sambrook et al. 1989). The PCR products were sequenced obtained by applying an NCA on the mtDNA haplotypes directly by the dideoxy chain terminator method with the Big obtained (Rassmann et al. 1997) by computing a statistical Dye Terminator Cycle Sequencing Read Kit and an ABI parsimony network that estimates gene genealogies from PRISM 3130 DNA sequencer (Applied Biosystems®, DNA sequences (Templeton et al. 1995), using TCS v1.21 Carlsbad, CA, USA). (Clement et al. 2000). Although the nested clade phylogeographic analysis (NCPA) method is widely criticized because Environmental data of the emergence of false positives using the phylogeographic inference key (Knowles and Maddison 2002; Petit and Grivet In this study, we consider the entire Mediterranean Basin, 2002; Knowles 2004; Panchal and Beaumont 2007; Petit comprising the areal of T. destruens, excluding the 2008), this work used the NCA network only for defining Macaronesian region, where it is considered to be an alien the set of hierarchical groups of lineages (from haplotypes to species (Sauvard et al. 2010). We have considered a total of 45 clades). We used topological and frequency criteria to solve sites where T. destruens was present, in accordance with Horn cladogram loop ambiguities (Posada and Crandall 2001). The et al. (2006) and our samples (Fig. 1). We used three environ- network obtained was adjusted to that reported by Horn et al. mental datasets: climatic, topographic and host data. The 19 (2006). Finally, the haplotypes were joined in different hier- climatic environmental variables were taken from the database archical levels to obtain the total cladogram (Templeton et al. WorldClim (http://www.worldclim.org) version 1.4 (Hijmans 1995). We used the hierarchical structure defined by NCA of et al. 2005) (Online resource 1, Tables S1 and S2). The the 618 bp alignment to define the mtLs according to topographic data were considered as a single variable, altitude Rassmann et al. (1997), in order to investigate the significance above sea level, obtained from GTOPO30 (USGS 1996). All of the spatial and environmental ecogeographical variables in raster data have been used in a pixel of 10 km grid (cell size their large-scale distribution. ∼5 arcmin), as working scale. Host variables were calculated from distribution maps of the database EUFORGEN (freely Modelling geographical distributions of lineages available at http://www.bioversityinternational.org/networks/ euforgen (Alia and Martin 2003;Fadyetal.2003; Isajev et al. Distribution models of mtLs were done with the MaxEnt v. 2004; Mátyás et al. 2004) for P. halepensis, P. brutia, P. 3.3.3 software (Phillips et al. 2006), which uses the maximum F.J. Sánchez-García et al.

Fig. 1 Study area and sampling plots showing the presence of Tomicus destruens

entropy algorithm and is very suitable for limited sample sizes different responses to several scales, buffered background (Hernandez et al. 2006; Papeş and Gaubert 2007;Pearson areas (20 and 100 km radius) were developed using the 2007;Wiszetal.2008); therefore, sample units (lineages or Buffer function of ArcGIS around each known occurrence haplotypes) present in four or fewer localities were discarded, locality (Couvreur et al. 2011; Nakazato et al. 2010). due to the presence of strong deviations and artefacts when included in preliminary trials. The execution of the model was Relationship between haplotype diversity and environmental measured by the cross-validation method implemented in diversity on a large scale MaxEnt. Four replicates were used in all runs, which creates ∼ a different random data partition ( 25%test/75%training) Finally, in order to confirm that the variability of environmen- for each run as described (Araújo et al. 2005). Every model tal factors influences the distribution of haplotypes on a large was run under default options using a maximum of 2000 scale, as shown by Gallego and Galián (2008) on a fine scale, interactions. We selected the logistic output format, showing we calculated four indices of haplotype diversity (Online suitability values from 0 (unsuitable) to 1 (optimal). The jack- resource 1) from nine geographical areas. We studied the knife test, which is a resampling test to estimate bias in a relation between these indices from each zone and the stan- dataset, was used to explore the primary environmental factors dard error value of each environmental predictor, using the restricting the T. destruens geographic distribution. GAM methodology explained in Online Resource 1.

Ecological comparisons between lineages

Two measures of niche overlap were used to assess similarity Results and equivalence: Schoener’s D and the Hellinger distance I, implemented in ENMTools software v. 1.3. (Warren et al. Phylogeographic structure 2008, 2010). D and I are quantitative measures of differences in habitat suitability between two potential distribution models We have characterized 23 haplotypes from our data, of which with values ranging from 0 (completely different niche 15 (BS, BT, BU, BV, BW, ZE, ZF, ZG, ZH, ZI, ZJ, ZK, ZL, models) to 1 (identical niche models). Firstly, niche overlap ZM and ZN) are new (Table 2) and have been deposited in the values were calculated using ecological niche modelling GenBank Nucleotide Sequence Database under accession (ENMs) for each mtL pair. numbers JN676037–JN676051. We worked with a total of The identity test was verified, with the null hypothesis 69 haplotypes in this study. referring to a mtL pair having equivalent ecological niches The reconstruction of a parsimonious haplotype network and, therefore, subject to exactly the same environmental (Fig. 2) displayed a topology similar to that reported by Horn conditions. Later, the background test checked the null hy- et al. (2006) but with some differences. For example, we pothesis that the niches are dissimilar compared to back- found an alternative way to link clade Z with clade AC ground environments. The observed niche overlap values (connected through a new Turkish haplotype) separated also were compared to a null distribution of 100 pseudoreplicate by five mutational steps. However, we prefer to keep the link values created by comparing the ENM of one mtL to an ENM between clade Z with clade AA in accord with Horn et al. created from random points drawn from the geographic range (2006), who did this on the basis of the higher haplotype of the other mtLs (Warren et al. 2008). In order to determine frequencies in ZV and AB compared to ZJ and AQ. We Distribution of Tomicus destruens mitochondrial lineages

Table 2 Haplotypes (HT) found in the studied locations. The new Code Number HT number Haplotypes haplotypes are shown in italics Esp-Spa 10 8 AA, AC(3), AD, AY, BS, BT, BU, BW Mon-Spa 1 1 AA Yeb-Spa 8 3 AA, AC(6), BV Yun-Tur 29 12 ZE, ZF, ZG, ZH, ZI, ZJ, ZK, ZL, ZM, ZN,ZV(8),ZZ(11) maintain the link between the BR haplotype and BQ instead of elaborate the fourth hierarchical level among the third level linking it to the AY haplotype, on the basis of the shorter clades (option 1, 3-1 with 3-2, or option 2, 3-2 with 3-3). geographical distance (Posada and Crandall 2001)between Then, our level 5 corresponds to the entire cladogram, which BR and BQ, according to the criteria used by Templeton et al. can be considered as the entire distribution of the species (1987) and Templeton and Sing (1993). We were not able to T. destruens.

Fig. 2 a Statistical parsimony network of haplotypes of Tomicus of increasing steps. The new haplotypes have been shaded. Dashed line destruens from the Mediterranean samples, of a 618 bp fragment of the shows the alternative link way of eastern and western clades. b Nested mitochondrial genes cytochrome c oxidase I (COI) and II (COII) that clade analysis of the clades of level 1 and above. The colored haplotypes flank the tRNAleu gene. Each line corresponds to a mutational step and and cladograms were used in the ecological niche analysis each empty circle to a missing intermediate; boxes indicate nested clades F.J. Sánchez-García et al.

The new haplotypes found in the samples from Spain contributed at a higher rate than host tree variables, except belong to the western group, in particular group A, which for haplotype AS, where the variable “coverage of Pinus comprises haplotypes AA and AC, both widely distributed. halepensis” was the largest contributor. Note that this AS The Sierra Espuña site has four new haplotypes (BS, BT, BU haplotype occurs in France, Italy and Algeria, thus, being and BW) and the AY haplotype (found so far only in northern distributed across a wide environmental range (Fig. 3a). We Corsica), corroborating its highly variable pattern reported found extreme climatic variables BIO6 and BIO9 as well as earlier (Gallego and Galián 2008). The Turkish samples be- average climatic variables BIO1, “temperature seasonality” long to the eastern group Z, with haplotypes ZV and ZZ. The (BIO4) and “precipitation seasonality” (BIO15) in these Yuntdað locality showed ten unique haplotypes (ZE, ZF, ZG, models. The distribution models of western rare haplotypes, ZH, ZI, ZJ, ZK, ZL, ZM and ZN). Overall, the new haplotypes AD and AS, show a wide distribution in the eastern reported here are included in the hierarchical clade levels 2-1, Mediterranean Basin. In addition, a large overlap with the 2-5, 2-7 and 2-8. ZZ haplotype is shown in the Italian Peninsula, the islands of Sicily and Sardinia, North Africa and the southern Iberian Distribution models of mtLs Peninsula (Fig. 4b).

The analysis of the contribution of environmental variables Ecological niche modelling showed that the variable “coverage of Pinus pinaster” was of greatest importance, explaining the distribution of the mtL We found the overlap index I was always larger than D in all clade 3-1 as well as clade 3-2, (Online Resource 2; mtL levels analysed. For the identity test, the clade 3-3 index Table S3). The best variable to explain the mtL clade 3-3 was significantly different from both 3-2 and 3-3 indices distribution, however, was “coverage of Pinus sylvestris”. (Table 3,Fig.5). The oriental haplotype ZZ was significantly “Coverage of Pinus sylvestris” was the second most important different from the western haplotypes AA, AC and AD predictor in the models of mtL clades 3-1 and 3-2 in contrast (Table 3) at the haplotype level. Therefore, the null hypothesis to the “annual mean temperature” (BIO1) in clade 3-3. The was rejected, and no niche equivalence was identified for third and fourth most important were “minimum temperature clade 3-3 and the ZZ haplotype with respect to their peers of coldest month” (BIO6) and “coverage of Pinus on the western side. halepensis” for both mtL clades 3-1 and 3-2. Instead, As for the background test, we noted that clade 3-3 “precipitation of coldest quarter” (BIO19) and “mean was always significantly different when projected into temperature of driest quarter” (BIO9) were the most the regions of clades 3-1 and 3-2 (Table 3,Fig.6). important for mtL clade 3-3. Therefore, we rejected the alternative hypothesis of The distribution models of level 3 mtLs show western mtLs niche similarity between them. Actually, clades 3-2 (clades 3-1 and 3-2) overlap in a wide area in the west and 3-1 were not significant when projected into the Mediterranean Basin, mainly in the central Iberian regions of clade 3-3. Thus, the null hypothesis of non- Peninsula. A few spots of potential distribution of these line- similarity in niches was accepted. ages are seen on the eastern basin (Greek Peninsula and The 20 km buffer showed more ecological similarity than Middle East), where these western mtLs are not found. The the 100 km buffer at the mtL level. Both major mtLs (AA and eastern mtL clade 3-3 has a dominant presence in the eastern AC) and rare mtLs (AD, AS and ZZ) were non-similar at any end of the Mediterranean Basin, but its potential distribution buffer level (Table 3). Even with the addition of new haplo- area extends to the Italian Peninsula, southwest of the Iberian types in the eastern clade, the structure was maintained be- Peninsula, the islands of Sicily and Sardinia and northern tween the mtL clades 2-7 and 2-8 proposed by Horn et al. Africa. The overlap areas between the three lineages are (2006). We found significant differences in the identity test, scarce, focusing on the edges of contact between their respec- although it did not show differences in the indices I and tive distributions (Fig. 4a). D. Besides, we detected no similarity niche at any scale At the more basic mtL level (haplotypes), we found that the of buffer (P >0.01) for the similarity test (Online variable “coverage of Pinus sylvestris” had a large contribu- Resource 2,TableS4). tion in all haplotypes analysed. In widely distributed haplotypes (AA and AC), however, the host tree variable “coverage Relationship between haplotype diversity and environmental of Pinus pinaster” was the main contributor, followed by diversity on a large scale “coverage of Pinus halepensis”. The climatic variables “isothermality” (BIO3), BIO6, “mean temperature of wettest The best variable to explain the zonal distribution of the four quarter” (BIO8) and “precipitation of warmest quarter” haplotype diversity indices was the standard error of the (BIO18) made only a minor contribution. By contrast, in rare “mean temperature of the wettest quartet” (Fig. 7). From the haplotype models (AD, AS and ZZ), the climate variables four indices used, the model obtained for Eind and Hex was the Distribution of Tomicus destruens mitochondrial lineages

Fig. 3 a, b Spatial distribution of haplotypes and level 3 clades used in the maximum entropy methodology (MaxEnt)

Fig. 4 Potential distribution of the analysed lineages with a probability ≥0.40. a Clades 3 level. b Clades haplotype level F.J. Sánchez-García et al.

Table 3 Tests of niche overlap, niche equivalency (identity) and niche similarity (background) for Tomicus destruens lineages

Lineages Niche Identity test Similarity test overlap A → BA→ B

IDI D I D I D

20 km 100 km 20 km 100 km 20 km 100 km 20 km 100 km

Clade_3_3 vs clade_3_1 0.603 0.387 0.813** 0.575** 0.664** 0.672** 0.429** 0.463** 0.628 ns 0.554 ns 0.364** 0.487 ns Clade_3_1 vs clade_3_2 0.944 0.756 0.908 ns 0.694 ns 0.928 ns 0.939** 0.715 ns 0.608 ns 0.984 ns 0.963 ns 0.867 ns 0.797 ns Clade_3_2 vs clade_3_3 0.606 0.380 0.813** 0.575** 0.657 ns 0.582 ns 0.398 ns 0.496 ns 0.677** 0.648** 0.455** 0.430** Haplotype _AA vs haplotype _AC 0.921 0.710 0.873 ns 0.598 ns 0.856 ns 0.948 ns 0.592 ns 0.565 ns 0.966 ns 0.940 ns 0.799 ns 0.738 ns Haplotype _AD vs haplotype _AA 0.879 0.617 0.768 ns 0.489 ns 0.807** 0.692 ns 0.586 ns 0.451 ns 0.854 ns 0.909 ns 0.611** 0.699 ns Haplotype _AS vs haplotype _AA 0.801 0.517 0.703 ns 0.447 ns 0.625** 0.652 ns 0.325 ns 0.378 ns 0.639 ns 0.762** 0.364** 0.503** Haplotype _ZZ vs haplotype _AA 0.571 0.328 0.772** 0.530** 0.601** 0.643** 0.341** 0.414** 0.532** 0.484 ns 0.300** 0.428 ns Haplotype _AD vs haplotype _AC 0.807 0.551 0.833** 0.563** 0.815** 0.707 ns 0.593** 0.471 ns 0.789** 0.876 ns 0.528** 0.638 ns Haplotype _AS vs haplotype _AC 0.828 0.570 0.703 ns 0.448 ns 0.640** 0.692 ns 0.363 ns 0.401 ns 0.713 ns 0.818** 0.426 0.556* Haplotype _ZZ vs haplotype _AC 0.606 0.359 0.751** 0.506** 0.634** 0.676** 0.407 ns 0.435 ns 0.577** 0.543 ns 0.337** 0.516 ns Haplotype _AS vs haplotype _AD 0.883 0.613 0.837 ns 0.631* 0.657 ns 0.633 ns 0.362 ns 0.356 ns 0.896 ns 0.904 ns 0.676 ns 0.708 ns Haplotype _ZZ vs haplotype _AD 0.781 0.533 0.860** 0.645** 0.871 ns 0.890 ns 0.675 ns 0.697 ns 0.666 ns 0.572 ns 0.421 ns 0.361 ns Haplotype _ZZ vs haplotype _AS 0.750 0.552 0.701 ns 0.459 ns 0.782 ns 0.846 ns 0.566 ns 0.615 ns 0.535 ns 0.609 ns 0.325 ns 0.364 ns

**P≤0.01; *P≤0.05; ns not significant (P>0.05) most explicative (explained deviance=40 and 32 %, AIC Discussion decrease=62 and 60 %, respectively). Both models predict higher levels of exclusive and endemic haplotypes in zones The haplotype network obtained in this study adds new hap- with high variability of temperatures in the wet period. By lotypes from the eastern side of the Mediterranean Basin, contrast, the spatial distribution of Hind and Hst indices was which was less sampled in earlier studies. Moreover, samples barely explained for their models from the Turkish locality have a high level of haplotype

Fig. 5 Tests for niche identity applied to level 3 clades about the I and D metric. a Clade 3-1 vs clade 3-3. b Clade 3-1 vs clade 3-2. c Clade 3-2 vs clade 3-3. Black lines indicate the I observed overlap values and black dashed lines specify the D value Distribution of Tomicus destruens mitochondrial lineages

Fig. 6 Tests for niche similarity applied to level 3 clades about the D metric. The first row shows the background test with 100 km buffer, and the second row illustrates the background test with 20 km buffer. Black dashed lines indicate the observed overlap values diversity, which increases the distribution of haplotypes ZZ Yuntdag (Turkey), 45 % of haplotypes were unreported. and ZV. The finding of haplotypes ZV and ZZ together in Furthermore, South Anatolia has been considered to be a western Turkey confirms the prediction made by Horn et al. glacial refuge for several organisms (Médail and Diadema (2006) of a contact zone between the two main subclades in 2009; Habel et al. 2010). Thus, the high level of haplotype the eastern Mediterranean area. Unexpectedly, the new hap- diversity (Hex) found in this and other areas, including the lotype ZE is linked to haplotype ZT, which is in the eastern Helenic Peninsula, the Middle East, the Iberian Peninsula and subclade. Taking into account its geographical location, how- North Africa (Table 4), suggests that this might well be the ever, it should be connected to the so-called Balkan subclade case for T. destruens. according to Horn et al. (2006). Likewise, the finding of the The use of highly polymorphic genetic markers allows the four new haplotypes (BS, BT, BU and BW) in Sierra Espuña recognition of phylogenetic and ecologic associations, using corroborates the high level of diversity of this locality reported the multiple regression model or a heuristic model (Jakob earlier (Vasconcelos et al. 2006; Gallego and Galián 2008). In et al. 2007). Cytochrome c oxidase I (COI) and II (COII)

Fig. 7 Partial response curve results of GAM of diversity haplotype indices using mean values on areas of the environmental mean temperature of wettest quartet variable. a Exclusive haplotype ratio (Hex). b Standardized haplotype (Hst). c Haplotypic index (Hind). d Endemicity index (Eind). F.J. Sánchez-García et al.

Table 4 Indices of haplotype diversity on the different areas of the part of it) would be encompassed within the range of the Mediterranean Basin ecological space of the two major clades (3-1 and 3-2).

Area Hex Hst Hind Eind The question arises as to why this poor association exists between P. brutia and lineage 3-3, when exactly the opposite Northern Africa 69.57 3.3 2.56 7.73 was expected. One hypothetic explanation would be that the Iberian Peninsula 70 0.38 3.64 12.73 eastern clade was split from the western group. This could be Western France 44.44 0.45 1.86 9.2 understood under the framework of the ecological niche the- Italian Peninsula 25 0.46 1.6 5 ory (Peterson et al. 2011), as a process of niche shifting among Helenic Peninsula 75 0.39 1.07 20 basal lineages (AD, AS and ZZ) vs major lineages, by com- Anatolia 78.57 0.6 0.82 4.62 petitive exclusion (AD and AS) or vicariance (ZZ). These Middle East 71.43 0.78 1.62 16.48 lineages would be adapted to extreme temperature and pre- Islands 16.67 0.68 1.38 3.85 cipitation, or chemical conditions imposed by their host trees (Blanch et al. 2009). Therefore, they were able to colonize the eastern Mediterranean. According to Horn et al. (2006), some genes are sufficiently polymorphic in T. destruens (Kerdelhué lineages of T. destruens were excised from the western basin et al. 2002; Vasconcelos et al. 2006;Hornetal.2006;Gallego and spread to the eastern Mediterranean Basin after the and Galián 2008) to allow the employment NCA-MaxEnt Pleistocene, something which happened with other organisms methodology for identifying relationships with ecogeographic (Hewitt 1996). This could be the case of ZZ lineage, which variables. P. sylvestris coverage is presented in all models as diverged over time from the western group. Our results sug- negative biotic contribution. That is, this variable delimited gest that a niche shift might have started, being still at an early the central and northern European distributions, where relationship with its host tree P. brutia, with insufficient time T. destruens is not able to develop its life cycle and where to develop a host-dependent genetic structure. However, the the sister species T. piniperda occurs, in accord with the only suitable host tree species in the Middle eastern area is distribution models proposed by Horn et al. (2012), although P. brutia. Taking into account that P. halepensis and P. brutia those authors did not use any host–tree predictor variable. The were a vicariant species pair, with several ecological and coverage of host tree species P. pinaster and P. halepensis genetic similarities (Barbéro et al. 1998 and references there- occurred in the models as a major contribution in the western in), the absence of an alternative host could have induced a clades 3-1 and 3-2. The same happened in all non-rare mtLs low level of host specialization. That is, the hypothetical throughout all hierarchic levels (clades 1-1, 1-10, 2-1 and 2-5 ancestral ZZ group was pre-adapted to attack the vicariant and haplotypes AA and AC). Remarkably, P. brutia coverage host tree, P. brutia, in the new eastern environment. The did not have a major contribution with the occurrence of the alternative hypothesis would be that the differences between clade 3-3 group, its natural host tree. This was in accord with the clades are due to differences in the ecogeographic space Horn et al. (2006), who did not detect a significant effect of the available (Godsoe 2010), and niche shifts result from the genetic variance and the host tree, at least qualitatively. anisotropy of ecologic space (Soberón and Peterson 2011). However, any of the extreme climatic variables, such The major contribution of the distribution models of mtLs, as BIO19 and BIO9, works in the mtLs as the eastern however, depends on prior distribution of samples (hap- clade 3-3, the subclades 2-7, 2-8, 1-18 and 1-15 and the lotypes in our case). The major haplotypes (AA and ZZ haplotype. AC) cover a broader environmental spectrum in their The two buffer differences found in the similarity test could niche than the rare lineage ZZ, thereby increasing the be explained by the increase of environmental heterogeneity sampling in the geographic area overlapping the three when the geographic scale is increased, resulting in a higher lineages (clades 3-1, 3-2 and 3-3), i.e. the Italian probability of non-similarity, in accord with Nazakato et al. Peninsula could diminish the similarity significant (2010). Our models have identified the potential niche, but values found in the niche tests. never the fundamental niche, which strictly should be Furthermore, the Iberian Peninsula constitutes the largest set according to the physiological and biophysical prin- and most continuous distribution area for T. destruens,con- ciples of the organism (Kearney et al. 2010; Araújo and taining all host pine taxa except P. brutia and a large environ- Peterson 2012). mental variability that permits generating a high level of Our results support the suggestion that the niche of the haplotype diversity. Other areas with this tendency were eastern lineage is different compared to the western lineages; North Africa, the Peloponnese and Anatolian Peninsula and however, niches of western lineages were not different com- the Middle East. Our results indicate a high level of variability pared to the eastern lineage. The eastern group (3-3) has a in the areas with the highest temperatures of the wettest month potential niche related to extreme weather conditions, i.e. a that is related directly with the higher rates of endemic haplotype of specialization. In turn, the eastern niche (or a major types of T. destruens. Accordingly, we hypothesise that Distribution of Tomicus destruens mitochondrial lineages

T. destruens lineages tend to reduce the overlapping of their References realised niche on a large scale by selection of different temperature ranges in the wetter quartet. In the Mediterranean Alia, R., & Martin, S. (2003). EUFORGEN technical guidelines for climate, the most important rainfall occurs in spring and genetic conservation and use for maritime pine (Pinus pinaster). autumn, and the smooth temperature allows the host pines to Rome: International Plant Genetic Resources Institute. develop their vegetative period, which is interrupted in the Araújo, M. B., & Peterson, A. T. (2012). Uses and misuses of bioclimatic envelope modeling. Ecology, 93,1527–1539. coldest weeks of winter. This coincides over time with the Araújo, M. B., Pearson, R. G., Thuiller, W., & Erhard, M. (2005). reproductive flight of T. destruens as has been widely reported Validation of species–climate impact models under climate change. (Gallego et al. 2004; Vasconcelos et al. 2006; Chakali 2005; Global Change Biology, 11,1504–1513. Faccoli et al. 2005a, b). Therefore, the reproductive attacks of Barbéro, M., Loisel, R., P. Quézel, Richardson, D. M., & Romane, F. Pines of the Mediterranean basin. In D. M. Richardson (Ed.), T. destruens overlap with the vegetative period of their host Ecology and biogeography of Pinus (pp. 153–170). Cambridge: trees. Accordingly, high temperatures in autumn permit the Cambridge University Press. extension of the reproductive behaviour of T. destruens, Blanch, J.-S., Peñuelas, J., Sardans, J., & Llusià, J. (2009). Drought, but high temperatures in spring tend to shorten their warming and soil fertilization effects on leaf volatile terpene concentrations in Pinus halepensis and Quercus ilex. Acta Physiologiae flight.Hightemperaturesinautumnandwinter,howev- Plantarum, 31,207–218. er, enable a long growth period of the host trees. So, in Buckley, D. (2009). Toward an organismal, integrative, and iterative areas with highly variable temperature, the selection phylogeography. BioEssays, 31,784–793. pressure of these processes could allow the fixation of Chakali, G. (2005). A Hilésina do Pinheiro, Tomicus destruens Wollaston 1865 (Coleoptera-Scolytidae) em Zonas Semi-Áridas. Silva endemic haplotypes, possibly by reaching a balance Lusitana, 13,113–124. between the length of the flight reproductive period of Clement, M., Posada, D., & Crandall, K. A. (2000). TCS: a computer T. destruens and the status of host tree vigour during program to estimate gene genealogies. Molecular Ecology, 9, 1657– their growth stage. In conclusion, temperature variability 1659. Couvreur, T. L., Porter-Morgan, H., Wieringa, J. J., & Chatrou, L. W. in the wettest period appears to be related to a high (2011). Little ecological divergence associated with speciation in proportion of endemic haplotypes of T. destruensinthe two African rain forest tree genera. BMC Evolutionary Biology, 11, Mediterranean Basin. 296. Our proposed methodology (NCA-MaxEnt) emerges from Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J. R. G., Gruber, B., Lafourcade, B., Leitão, P. J., the phylogeographic framework proposed by Horn et al. Münkemüller, T., McClean, C., Osborne, P. E., Reineking, B., (2006) as a new approach for modelling the geographic line- Schröder, B., Skidmore, A. K., Zurell, D., & Lautenbach, S. age distribution from an NCA on a meso or large geographic (2012). Collinearity: a review of methods to deal with it and a – scale (the Mediterranean Basin). This represents an improve- simulation study evaluating their performance. Ecography, 35,1 20. Faccoli, M., Piscedda, A., Salvato, P., Simonato, M., Masutti, L., & ment of the methodology proposed by Gallego and Galián Battisti, A. (2005a). Genetic structure and phylogeography of pine (2008), which includes analysing T. destruens on a fine scale. shoot beetle populations (Tomicus destruens and T. piniperda, By contrast, our work improves the work reported by Horn Coleoptera Scolytidae) in Italy. Annals of Forest Science, 62,361– et al. (2012) using general linear models GAM, for two 368. Faccoli, M., Battisti, A., & Masutti, L. (2005b). Phenology of Tomicus reasons: (1) we introduce the biotic variables in the modelling destruens (Wollaston) in northern Italian pine stands. In F. Lieutier instead of using the intersection between distribution maps of & D. Ghaioule (Eds.), Entomological research in Mediterranean the beetles and the geographic distribution of the host pine, forest ecosystems (pp. 185–193). Paris: Institut National Reserche which gives a more accurate prediction; and (2) we analyse Agronomique Editions. Fady, B., Semerci, H., & Vendramin, G. (2003). EUFORGEN technical the contribution of bioclimatic variables to the distribution of guidelines for genetic conservation and use for Aleppo pine (Pinus intraspecific clades and haplotypes. Finally, paleodistribution halepensis) and brutia pine (Pinus brutia). Rome: International modelling should be explored in the future to detect whether Plant Genetic Resources Institute. lineages of the eastern group have had a more widespread past Gallego, D., & Galián, J. (2001). The internal transcribed spacers (ITS1 and ITS2) of the rDNA differentiates the bark beetle forest pests distribution. Comparing the results of past and present distri- Tomicus destruens and T. piniperda. Insect Molecular Biology, 10, butions and time-calibrated phylogenies would help to deter- 415–420. mine the scope of niche shifts of T. destruens lineages over Gallego, D., & Galián, J. (2008). Hierarchical structure of mitochondrial time (Wiens 2011). lineages of Tomicus destruens (Coleoptera, Scolytidae) related to environmental variables. Journal of Zoological Systematics and Evolutionary Research, 46,331–339. Acknowledgments We thank Ö. Toprak, José Luís Lencina and Gallego, D., Cánovas, F., Esteve, M. A., & Galián, J. (2004). Descriptive Eudaldo Gómez-Rosa for collecting samples and Carmelo Andújar and biogeography of Tomicus (Coleoptera: Scolytidae) species in Spain. Carlos Ruiz for their assistance with the script and MaxEnt methodology. Journal of Biogeography, 31(12), 2011–2024. Finally, Obdulia S. Sanchez-Domingo and Ana I. Asensio are also Godsoe, W. (2010). Regional variation exaggerates ecological divergence thanked for their help in the laboratory. This work was financed by the in niche models. Systematic Biology, 59,298–306. Fundación Séneca Project (reference 12023/PI/09) of the Murcia Region- Graham, C. H., Ron, S. R., Santos, J. C., Schneider, C. J., & Moritz, C. al Government. (2004). Integrating phylogenetics and environmental niche models F.J. Sánchez-García et al.

to explore speciation mechanisms in dendrobatid frogs. Evolution, Médail, F., & Diadema, K. (2009). Glacial refugia influence plant diver- 58, 1781–1793. sity patterns in the Mediterranean Basin. Journal of Biogeography, Habel, J. C., Drees, C., Schmitt, T., & Assmann, T. (2010). Review 36, 1333–1345. refugial areas and postglacial colonizations in the western Nakazato, T., Warren, D. L., & Moyle, L. C. (2010). Ecological and Palearctic. In D. J. C. Habel & P. D. T. Assmann (Eds.), Relict geographic modes of species divergence in wild tomatoes. American species (pp. 189–197). Berlin: Springer. Journal of Botany, 97,680–693. Hernandez, P. A., Graham, C. H., Master, L. L., & Albert, D. L. (2006). Newman, C. E., & Rissler, L. J. (2011). Phylogeographic analyses of the The effect of sample size and species characteristics on performance southern leopard frog: the impact of geography and climate on the of different species distribution modeling methods. Ecography, 29, distribution of genetic lineages vs. subspecies. Molecular Ecology, 773–785. 20, 5295–5312. Hewitt, G. M. (1996). Some genetic consequences of ice ages, and their Oney, B., Reineking, B., O’Neill, G., & Kreyling, J. (2013). Intraspecific role in divergence and speciation. Biological Journal of the Linnean variation buffers projected climate change impacts on Pinus Society, 58,247–276. contorta. Ecology and Evolution, 3,437–449. Hijmans, R. J., & Van Etten, J. (2012). Geographic analysis and modeling Panchal, M., & Beaumont, M. A. (2007). The automation and evaluation with raster data. url http://cran.r-project.org/web/packages/raster/ of nested clade phylogeographic analysis. Evolution, 61,1466– raster.pdf. 1480. Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G., & Jarvis, A. Papeş, M., & Gaubert, P. (2007). Modelling ecological niches from low (2005). Very high resolution interpolated climate surfaces for numbers of occurrences: assessment of the conservation status of global land areas. International Journal of Climatology, 25, poorly known viverrids (Mammalia, Carnivora) across two conti- 1965–1978. nents. Diversity and Distributions, 13,890–902. Horn, A., Roux-Morabito, G., Lieutier, F., & Kerdelhue, C. (2006). Pearman, P. B., D’Amen, M., Graham, C. H., Thuiller, W., & Phylogeographic structure and past history of the circum- Zimmermann, N. E. (2010). Within-taxon niche structure: niche Mediterranean species Tomicus destruens Woll. (Coleoptera: conservatism, divergence and predicted effects of climate change. Scolytinae). Molecular Ecology, 15,1603–1615. Ecography, 33,990–1003. Horn, A., Kerdelhué, C., Lieutier, F., & Rossi, J.-P. (2012). Predicting the Pearson, R. (2007). Species’ distribution modeling for conservation edu- distribution of the two bark beetles Tomicus destruens and Tomicus cators and practitioners. Center for Biodiversity and Conservation, piniperda in Europe and the Mediterranean region. Agricultural and American Museum of Natural History. Forest Entomology, 14,358–366. Peterson, A.T., Soberón, J., Pearson, R.G., Anderson, R.P., Martínez- Hundsdoerfer, A. K., Mende, M. B., Kitching, I. J., & Cordellier, M. Meyer, E., Nakamura, M., Araújo, M.B.(2011). Ecological Niches (2011). Taxonomy, phylogeography and climate relations of the and Geographic Distributions. Princeton University Press, Western Palaearctic spurge hawkmoth (Lepidoptera, Sphingidae, Princeton. Macroglossinae). Zoologica Scripta, 40,403–417. Petit, R. J. (2008). The coup de grâce for the nested clade phylogeograph- Isajev, V., Fady, B., Semerci, H., & Andonovski, V. (2004). EUFROGEN ic analysis? Molecular Ecology, 17,516–518. technical guidelines for genetic conservation and use of European Petit, R. J., & Grivet, D. (2002). Optimal randomization strategies when black pine (Pinus nigra). Rome: International Plant Genetic testing the existence of a phylogeographic structure. Genetics, 161, Resources Institute. 469–471. Jakob, S. S., Ihlow, A., & Blattner, F. R. (2007). Combined ecological Phillips, S. J., Dudík, M., & Schapire, R. E. (2004). A maximum entropy niche modelling and molecular phylogeography revealed the evolu- approach to species distribution modeling. In Proceedings of the tionary history of Hordeum marinum (Poaceae)—niche differentia- twenty-first international conference on machine learning (p. 83). tion, loss of genetic diversity, and speciation in Mediterranean New York: ACM. Quaternary refugia. Molecular Ecology, 16,1713–1727. Phillips, S. J., Anderson, R. P., & Schapire, R. E. (2006). Maximum Kearney, M., Simpson, S. J., Raubenheimer, D., & Helmuth, B. (2010). entropy modeling of species geographic distributions. Ecological Modelling the ecological niche from functional traits. Philosophical Modelling, 190,231–259. Transactions of the Royal Society, B: Biological Sciences, 365, Posada, D., & Crandall, K. A. (2001). Intraspecific gene genealogies: 3469–3483. trees grafting into networks. Trends in Ecology & Evolution, 16,37– Kerdelhué, C., Roux-Morabito, G., Forichon, J., Chambon, J.-M., 45. Robert, A., & Lieutier, F. (2002). Population genetic structure of Rassmann, K., Tautz, D., Trillmich, F., & Gliddon, C. (1997). The Tomicus piniperda L. (Curculionidae: Scolytinae) on different pine microevolution of the Galápagos marine iguana Amblyrhynchus species and validation of T. destruens (Woll.). Molecular Ecology, cristatus assessed by nuclear and mitochondrial genetic analyses. 11,483–494. Molecular Ecology, 6,437–452. Knowles, L. L. (2004). The burgeoning field of statistical R Core Team (2012). R: A language and environment for statistical phylogeography. Journal of Evolutionary Biology, 17,1–10. computing. R Foundation for Statistical Computing, Vienna, Knowles, L. L., & Maddison, W. P. (2002). Statistical phylogeography. Austria. ISBN 3-900051-07-0 http://www.R-project.org/ Molecular Ecology, 11,2623–2635. Sambrook, J., Fritsch, E. F., & Maniatis, R. (1989). Molecular cloning: a Kohlmayr, B., Riegler, M., Wegensteiner, R., & Stauffer, C. (2002). laboratory manual (Vol. 3). New York: Cold Spring Harbour Press. Morphological and genetic identification of the three pine pests of Sauvard, D., Branco, M., Lakatos, F., Faccoli, M., & Kirkendall, L. the genus Tomicus (Coleoptera, Scolytidae) in Europe. Agricultural (2010). Weevils and bark beetles (Coleoptera, Curculionoidea). and Forest Entomology, 4,151–157. Chapter 8.2. BIORISK – Biodiversity and Ecosystem Risk Mátyás, C., Ackzell, L., & Samuel, C. J. A. (2004). EUFORGEN tech- Assessment, 4. nical guidelines for genetic conservation and use for Scots pine Schulte, U., Hochkirch, A., Lötters, S., Rödder, D., Schweiger, S., (Pinus sylvestris). Rome: International Plant Genetic Resources Weimann, T., & Veith, M. (2012). Cryptic niche conservatism Institute. among evolutionary lineages of an invasive lizard. Global Ecology McCormack, J. E., Zellmer, A. J., & Knowles, L. L. (2010). Does niche and Biogeography, 21,198–211. divergence accompany allopatric divergence in Aphelocoma jays as Soberón, J., & Peterson, A. T. (2011). Ecological niche shifts and envi- predicted under ecological speciation?: insights from tests with ronmental space anisotropy: a cautionary note. Revista Mexicana de niche models. Evolution, 64,1231–1244. Biodiversidad, 82,1348–1355. Distribution of Tomicus destruens mitochondrial lineages

Tamura, K., Dudley, J., Nei, M., & Kumar, S. (2007). MEGA4: molecular Mediterranean pine shoot beetle Tomicus destruens in the Iberian evolutionary genetics analysis (MEGA) software version 4.0. Peninsula and Southern France. Agricultural and Forest Molecular Biology and Evolution, 24,1596–1599. Entomology, 8,103–111. Templeton, A. R., Boerwinkle, E., & Sing, C. F. (1987). A cladistic analysis Vink, C. J., Thomas, S. M., Paquin, P., Hayashi, C. Y., & Hedin, M. of phenotypic associations with haplotypes inferred from restriction (2005). The effects of preservatives and temperatures on arachnid endonuclease mapping I. Basic theory and an analysis of alcohol DNA. Invertebrate Systematics, 19,99–104. dehydrogenase activity in Drosophila. Genetics, 117,343–351. Warren, D. L., & Seifert, S. N. (2010). Ecological niche modeling in Templeton,A.R.,&Sing,C.F.(1993). A Cladistic Analysis of MaxEnt: the importance of model complexity and the performance Phenotypic Associations with Haplotypes Inferred from of model selection criteria. Ecological Applications, 21,335–342. Restriction Endonuclease Mapping. IV. Nested Analyses with Warren, D. L., Glor, R. E., & Turelli, M. (2008). Environmental niche Cladogram Uncertainty and Recombination. Genetics, 134, 659– equivalency versus conservatism: quantitative approaches to niche 669. evolution. Evolution, 62,2868–2883. Templeton, A. R., Routman, E., & Phillips, C. A. (1995). Separating Warren, D. L., Glor, R. E., & Turelli, M. (2010). ENMTools: a toolbox for population structure from population history: a cladistic analysis of comparative studies of environmental niche models. Ecography, 33, the geographical distribution of mitochondrial DNA haplotypes in 607–611. doi:10.1111/j.1600-0587.2009.06142.x. the tiger salamander, Ambystoma tigrinum. Genetics, 140,767–782. Wiens, J. J. (2011). The niche, biogeography and species interactions. Thompson, J. D., Higgins, D. G., & Gibson, T. J. (1994). CLUSTAL W: Philosophical Transactions of the Royal Society, B: Biological improving the sensitivity of progressive multiple sequence align- Sciences, 366,2336–2350. ment through sequence weighting, position-specific gap penalties Wisz, M. S., Hijmans, R. J., Li, J., Peterson, A. T., Graham, C. and weight matrix choice. Nucleic Acids Research, 22, 4673–4680. H., Guisan, A., & Group, N. P. S. D. W. (2008). Effects of doi:10.1093/nar/22.22.467 sample size on the performance of species distribution USGS (1996) US Geological Survey, Available at: http://eros.usgs.gov/#/ models. Diversity and Distributions, 14, 763–773. doi:10. Find_Data/Products_and_Data_Available/gtopo30_info. Accessed 1111/j.1472-4642.2008.00482 12 Mar 2010 Wood, S. L., & Bright, D. E. (1992). A catalog of Scolytidae and Vasconcelos, T., Horn, A., Lieutier, F., Branco, M., & Kerdelhué, C. Platypodidae (Coleoptera), part 2: taxonomic index. Great Basin (2006). Distribution and population genetic structure of the Naturalist Memoirs, 13,1–1553.