Role of plastid markers in environmental studies on the example of the endangered species Cistus heterophyllus

Abstract

[SPA] Los marcadores moleculares son una herramienta muy poderosa en muchos campos como la filogenia, la biología evolutiva o la conservación. Sin embargo, no es una tarea fácil encontrar marcadores adecuados para especies raras. Los marcadores ideales tienen que ser de carácter informativo dependiendo de la cuestión biológica: la diferenciación entre especies estrechamente relacionadas requiere un marcador para regiones altamente diferenciados, mientras marcadores para la diferenciación entre organismos pertenecientes a familias distantes están seleccionados para la detección de regiones conservadas. Importante es también el tipo de ADN como fuente de nuestro marcador. El ADN de plástidos se prefiere en proyectos sobre filogenia, mientras que la detección de eventos de hibridación demanda el análisis del ADN nuclear. Sin embargo, en el caso de especies raras, los científicos se encuentran con una falta de información sobre secuencias de los genomas bajo estudio. En este caso la única solución es la aplicación de marcadores universales, ya descritos para otros organismos. Este proyecto tiene como objetivo analizar un conjunto de marcadores moleculares para el rastreo de eventos de hibridación en la población de una especie en peligro de extinción de la familia Cistaceae, Cistus heterophyllussubsp. carthaginensis. La distribución de esta subespecie se limita a una sola población natural en el sureste de España, donde co-ocurren individuos con fenotipo silvestre y fenotipos híbridos, lo que sugiere eventos de hibridación entre esta población en peligro de extinción y una especie localmente abundante, Cistus albidus. Estos híbridos se han descrito en África como C. ×clausonis.Se realizaron búsquedas de regiones de ADN que permiten la discriminación de entre los individuos de tipo silvestre e híbridos supuestos. Los datos generados podrían mejorar la estrategia de conservación de las especies con el fin de evitar su extinción. En el capítulo 1, se describe la posible aplicación de marcadores moleculares plastídicos, regiones de marcadores conocidas como “códigos de barras”, para su aplicación de la población mencionado anteriormente. Regiones no codificantes de ADN (rbcL, trnK-matK) no resultaron lo suficientemente variables para ser informativas en individuos estrechamente relacionados. Regiones intra-específicas (trnL-F, trnH-psbA) presentan una alta tasa de cambios evolutivos, indicado por su alto grado de variabilidad. Sin embargo, encontramos que estos marcadores no son suficientemente estables como para proporcionar información fiable para la diferenciación entre individuos silvestres e híbridos. Sorprendentemente, se observó para los genes rpoBy rpoC1 una heteroplasmia en C. heterophyllusy C. × clausonislocal, pero no en C. albidusu otra especie comun a esta región, C. monspeliensis.Encontramos dos alelos distintos de rpoB, uno presente en todas las especies y un segundo presente sólo en C. heterophyllusy C.× clausonislocal. También se detectaron dos alelos de rpoC1, uno común a todas las especies analizadas y un segundo presente sólo en C. × clausonislocal. Nuestros resultados muestran que hay un alelo rpoBdistintivo y común a C. heterophyllus y C. × clausonisde África y Europa. El alelo rpoC1 unicamente encontrado en C.× clausonislocal indíca un origen de esta pequeña población diferente que no resulta de una hibridación entre los C. albiduso C. heterophyllusactualmente presentes en esta ubicación. El capítulo 2 describe la aplicación de regiones internas inter-espaciadas (ITS, internal transcribed spacer) ribosomales. Estos marcadores altamente polimórficos permiten la construcción de árboles filogenéticos moleculares con el objetivo de analizar las relaciones entre poblaciones geograficamente aislados de Cistus heterophyllus, Cistus albidus y posibles híbridos entre estos dos especies, C. × clausonisde África y de Europa. Nuestros datos indican que, depnediendo de individuo o población, C. × clausonisfilogenéticamenteparece más a Cistus heterophylluso Cistus albidus y problamente está relacionado a la homogenización de variación por evolution concertada. En el capítulo 3, se presenta un problema que surgió durante el análisis de los datos de PCR cuantitativa de los marcadores de código de barras. Como resultaron diferencias significativas entre especies en la eficiencia de la PCR aplicando el mismo marcador molecular, decidimos investigar si el sesgo observado podría perturbar la identificación de especies durante el metabarcoding de muestras. Utilizamos seis lociuniversales y 48 especies de plantas y cuantificamos el posible sesgo en cada paso del proceso de identificación desde PCR apunto final hasta la secuenciación. La amplificación a punto final fue significativamente diferente para un solo lociy entre las especies. Análisis por PCR cuantitativa reveló que el umbral Cq para diversos loci, incluso dentro de una sola extracción de ADN, mostró una diferencia de 2000 veces en la cantidad de ADN obtenida después de la amplificación. Experimentos de secuenciación de próxima generación (NGS) en nueve especies mostraron sesgos significativos hacia especies y lociespecíficos utilizando cebadores específicos del adaptador. El sesgo durante la secuenciación NGS se puede predecir en cierta medida por los valores Cq de amplificación en qPCR y depende de la secuencia primaria de ADN. [ENG] Molecular markers are a very powerful tool in many fields such as phylogenetics, evolutionary or conservation biology. However, it is not an easy task to find proper markers for rare species. The perfect marker depends on the biological question: for differentiation among closely related species we need a sensitive marker for highly differentiated region, whereas differentiation among organisms belonging to distant families requires markers for conserved regions. Important is also the type of the DNA as source of our marker. Plastid DNA is preferred in plant phylogenetic projects whereas the analysis of hybridization events requires markers proceeding from nuclear DNA. However, in case of rare species, scientist encounter a lack of sequence information about the genomes studied. In this case the only solution are universal markers already described for other organisms. This project aims to analyse a set of molecular markers for tracing hybridization events in the population of an endangered species from the Cistaceae family, Cistus heterophyllussubsp. carthaginensis. The distribution of this subspecies is limited to only one natural population in the south-eastern Spain where individuals with wild type and hybrid phenotypes co-occure, suggesting hybridization events between the endangered population and the locally abundant Cistus albidus. These hybrids have been described in Africa as C.× clausonis. We searched for DNA regions that allow discrimination between the wild type individuals and putative hybrids. The generated data could improve the species conservation strategy in order to avoid its extinction. In chapter 1, we describe the possible application of plastid markers, regions known as ¨DNA barcodes¨ as markers for the aforementioned population. Noncoding DNA regions (rbcL, trnK-matK) were found as not variable enough to be informative in closely related individuals. Intraspecific regions (trnL-F, trnH-psbA) presented a high rate of the evolutionary changes as indicated by their high variability. However, we found these markers as not sufficiently stable to give reliable information for the identification of wild type and hybrid individuals. Surprisingly, we observed heteroplasmy for rpoBand rpoC1genes in C. heterophyllusand the local C.× clausonis, but not in C. albidusor another species common to this region, C. monspeliensis. We found two distinct alleles of rpoB, one present in all species and a second present only in C. heterophyllusand the local C.× clausonis. We also detected two alleles ofrpoC1, one common to all species analyzed and a second present only in the local C.× clausonis. Our results show that there is a distinctive rpoBallele common to C. heterophyllusand C.× clausonisfrom Africa andEurope. The unique rpoC1allele found in the local C.× clausonis directs to a different origin of this small population, indicating that it is not a hybrid originating from C. albidusor C.heterophylluscurrently present in this location. Chapter 2 describes the application of the highly polymorphic internal transcribed spacer (ITS) region of the ribosomal DNA in the construction of a molecular tree in order to unravel the relationship among geographically isolated populations of Cistus heterophyllus, Cistus albidus and possible hybrids of these two species,C. × clausonisfrom Africa and Europe. Our data indicate that, depending on the individual and population, C. × clausonis phylogenetically resembles more either Cistus heterophyllusor Cistus albidus what might be related to the homogenization of variation between repeat types through concerted evolution. In chapter 3, we present an issue that arose during the analysis of quantitative PCR data of the barcode markers. As we realized that there were significant differences between species in PCR efficiency of the same marker, we decided to investigate if the observed bias may disturb species identification during metabarcoding of samples. We used six universal lociand 48 plant species and quantified the bias at each step of the identification process from end point PCR to next-generation sequencing. End point amplification was significantly different for single lociand between species. Quantitative PCR revealed that the Cq threshold for various loci, even within a single DNA extraction, showed 2,000-fold differences in DNA quantity after amplification. Next generation sequencing (NGS) experiments in nine species showed significant biases towards species and specific lociusing adaptor-specific primers.NGS sequencing bias may be predicted to some extent by the Cq values of qPCR amplification.

[ENG] Molecular markers are a very powerful tool in many fields such as phylogenetics, evolutionary or conservation biology. However, it is not an easy task to find proper markers for rare species. The perfect marker depends on the biological question: for differentiation among closely related species we need a sensitive marker for highly differentiated region, whereas differentiation among organisms belonging to distant families requires markers for conserved regions. Important is also the type of the DNA as source of our marker. Plastid DNA is preferred in plant phylogenetic projects whereas the analysis of hybridization events requires markers proceeding from nuclear DNA. However, in case of rare species, scientist encounter a lack of sequence information about the genomes studied. In this case the only solution are universal markers already described for other organisms. This project aims to analyse a set of molecular markers for tracing hybridization events in the population of an endangered species from the Cistaceae family, Cistus heterophyllussubsp. carthaginensis. The distribution of this subspecies is limited to only one natural population in the south-eastern Spain where individuals with wild type and hybrid phenotypes co-occure, suggesting hybridization events between the endangered population and the locally abundant Cistus albidus. These hybrids have been described in Africa as C.× clausonis. We searched for DNA regions that allow discrimination between the wild type individuals and putative hybrids. The generated data could improve the species conservation strategy in order to avoid its extinction. In chapter 1, we describe the possible application of plastid markers, regions known as ¨DNA barcodes¨ as markers for the aforementioned population. Noncoding DNA regions (rbcL, trnK-matK) were found as not variable enough to be informative in closely related individuals. Intraspecific regions (trnL-F, trnH-psbA) presented a high rate of the evolutionary changes as indicated by their high variability. However, we found these markers as not sufficiently stable to give reliable information for the identification of wild type and hybrid individuals. Surprisingly, we observed heteroplasmy for rpoBand rpoC1genes in C. heterophyllusand the local C.× clausonis, but not in C. albidusor another species common to this region, C. monspeliensis. We found two distinct alleles of rpoB, one present in all species and a second present only in C. heterophyllusand the local C.× clausonis. We also detected two alleles ofrpoC1, one common to all species analyzed and a second present only in the local C.× clausonis. Our results show that there is a distinctive rpoBallele common to C. heterophyllusand C.× clausonisfrom Africa andEurope. The unique rpoC1allele found in the local C.× clausonis directs to a different origin of this small population, indicating that it is not a hybrid originating from C. albidusor C.heterophylluscurrently present in this location. Chapter 2 describes the application of the highly polymorphic internal transcribed spacer (ITS) region of the ribosomal DNA in the construction of a molecular tree in order to unravel the relationship among geographically isolated populations of Cistus heterophyllus, Cistus albidus and possible hybrids of these two species,C. × clausonisfrom Africa and Europe. Our data indicate that, depending on the individual and population, C. × clausonis phylogenetically resembles more either Cistus heterophyllusor Cistus albidus what might be related to the homogenization of variation between repeat types through concerted evolution. In chapter 3, we present an issue that arose during the analysis of quantitative PCR data of the barcode markers. As we realized that there were significant differences between species in PCR efficiency of the same marker, we decided to investigate if the observed bias may disturb species identification during metabarcoding of samples. We used six universal lociand 48 plant species and quantified the bias at each step of the identification process from end point PCR to next-generation sequencing. End point amplification was significantly different for single lociand between species. Quantitative PCR revealed that the Cq threshold for various loci, even within a single DNA extraction, showed 2,000-fold differences in DNA quantity after amplification. Next generation sequencing (NGS) experiments in nine species showed significant biases towards species and specific lociusing adaptor-specific primers.NGS sequencing bias may be predicted to some extent by the Cq values of qPCR amplification.