Immunogenetic variation along the latitudinal gradient in Scandinavian anuran species: Evolutionary processes, demography and infection

ACTA

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1811

Immunogenetic variation along the latitudinal gradient in Scandinavian anuran species

Evolutionary processes, demography and infection

MARIA CORTAZAR-CHINARRO

Dissertation presented at Uppsala University to be publicly examined in Zootisalen, Norbyvägen 14-18, Uppsala, Friday, 14 June 2019 at 13:00 for the degree of Doctor of Philosophy. The examination will be conducted in English. Faculty examiner: Senior lecturer Helena Westerdahl (Lund Universitet).

Abstract

Cortazar-Chinarro, M. 2019. Immunogenetic variation along the latitudinal gradient in Scandinavian anuran species. Evolutionary processes, demography and infection. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 1811. 45 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-513-0662-9.

The evolutionary and demographic processes affecting how genetic variation is partitioned and distributed over large geographical scales is of fundamental importance for our understanding of how organisms may adapt to their environments. Northern peripheral populations generally have lower genetic variation and individuals in these populations may therefore face difficulties adapting to their local environment. At northern latitudes lack of genetic variation could be detrimental in face of newly emerging diseases as a result of anthropogenic actions and warmer climate in these areas. In this thesis, I explore genetic variation and the contemporary evolutionary processes affecting genes involved in the adaptive immune defense (Major Histocompatibility Complex; MHC) and the innate immune defense (AMP; Antimicrobial Peptides) over a large geographical gradient in anuran species (paper I, II and IV). I study signatures of historical selection on the MHC class II exon 2 and AMP (Temporin, Brevinin and Palustrin) sequences in the Signal Peptide and the Acidic Propiece domains (paper II and III).

Finally, I investigate potential associations between specific MHC class II exon 2 alleles and a chytrid fungus infection (Bd) in common toads (Bufo bufo) (paper IV). The results reveal that genetic variation of MHC class II exon 2 decreases towards northern latitudes in R. arvalis and B. bufo and have been shaped by complex evolutionary processes (drift, selection, migration) affected by different demographic scenarios. On the other hand, AMP nucleotide variation is divergent among geographical areas, but there is no clear geographical pattern along the same gradient, suggesting diversifying selection as the main force shaping genetic variation. Finally, I found an effect of two specific MHC class II exon 2 alleles on survival in juvenile B. bufo when infected with Bd. In summary, my thesis unravels the complex patterns shaping genetic diversity at large scales. My results may guide conservation practices aiming to prevent amphibian mass mortality events on-going all over the world.

Maria Cortazar-Chinarro, Department of Ecology and Genetics, Animal ecology, Norbyvägen 18 D, Uppsala University, SE-752 36 Uppsala, Sweden.

© Maria Cortazar-Chinarro 2019 ISSN 1651-6214

ISBN 978-91-513-0662-9

urn:nbn:se:uu:diva-382092 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-382092)

To Alberto, Axel, my

parents and my family

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Cortázar-Chinarro, M., Lattenkamp, E.Z., Meyer-Lucht, Y., Luquet, E., Laurila, A., Höglund, J. (2017) Drift, selection, or migration? Processes affecting genetic differentiation and varia- tion along a latitudinal gradient in an amphibian. BMC Evolu- tionary Biology, 17:89

II Cortázar-Chinarro, M., Meyer-Lucht, Y., Van der Valk, T., Richter-Boix, A., Laurila, A., Höglund, J. Antimicrobial peptide variation along a latitudinal gradient in two Ranid species: AMP genetic variation is not associated with demographic processes.

(Manuscript)

III Cortázar-Chinarro, M., Meyer-Lucht, Y., Laurila, A., Höglund, J (2018). Signatures of historical selection on MHC re- veal different selection patterns in moor frog (Rana arvalis). Im- munogenetics, 70:6

IV Cortázar-Chinarro, M., Meurling, S., Schroyens, L., Siljestam, M., Laurila, A, Höglund, J. Latitudinal MHC variation and hap- lotype associated differential survival in response to experi- mental infection of two Bd-GPL strains in common toads. (Man- uscript)

Reprints are made with permission from the respective publishers.

Contents

Introduction ... 9 

Research aims ... 11 

Materials and Methods ... 12 

Study species and sampling collection ... 12 

Rana arvalis (Paper I, II and III) ... 12 

Rana temporaria (Paper II) ... 12 

Bufo bufo (Paper IV) ... 13 

Sample collection ... 13 

Molecular Methods ... 16 

DNA extraction (Paper I, II and IV) ... 16 

Amplicon sequencing and library preparation (Paper I, II and IV) ... 16 

Other genetic markers (Paper I) ... 16 

Experimental infections (paper IV) ... 17 

Data Analyses ... 17 

MHC and AMPs (Paper I, II, IV) ... 17 

Statistical methods for measuring genetic variation and detecting selection (Paper I, II, III) ... 18 

Phylogenetic trees (Paper II, III) ... 19 

Infection analyses (Paper IV) ... 19 

Other analyses (Paper I) ... 19 

Results and discussion ... 20 

Genetic variation along latitudinal gradients (Paper I, II and IV) ... 20 

Diversity and Selection at the sequence level (Paper II and III) ... 24 

MHC Alleles associated with Chytrid fungus infection (Paper IV) ... 25 

Conclusions and future directions ... 27 

Svensk sammanfattning ... 28 

Resumen en Español ... 31 

Gracias, tack, thank you ... 35 

References ... 41 

Abbreviations

MHC; major histocompatibility complex AMPs; Antimicrobial Peptides

PBR; Peptide binding region

Bd; Batrachochytrium dendrobatidis

Introduction

Decreasing richness in biological diversity along latitudinal gradients from tropical to extratropical areas is one of longest recognized patterns in ecology (Willig, Kaufman, & Stevens, 2003). Like most research in ecology and evo- lutionary biology, understanding patterns and mechanisms related to diversity along large-scale latitudinal gradients is quite challenging. Studies along en- vironmental gradients provide ample evidence for adaptive evolution, inferred from genetic divergence (Talarico et al., 2019). However, disentangling whether current patterns of genetic divergence are caused by drift, selection and/or migration is complicated by demographic history. One approach to study the role of evolutionary processes shaping contemporary genetic varia- tion contrasts diversity patterns of neutral markers with that at genes under targeted selection (Li et al., 2016; DeCandia et al., 2019).

Vertebrates fight pathogenic infections using both the adaptive and innate im- mune system. The adaptive immune system, also known as the acquired im- mune system, is composed of highly specialized organs, tissues and systemic cells (Wood, 2006). The main cells of the specific immune system (e.g. lym- phocytes B and T) are able to synthetize glycoproteins (antibodies) and rec- ognize infinite diversity of antigens of single specificity tissue by entering through the bloodstream. Adaptive immune system, in contrast to innate im- mune system, usually clears infection and protects the host against reinfection with the same pathogen. One of the best understood adaptive molecules is the major histocompatibility complex (MHC class I and II). MHC molecules bind to specific antigens, which are presented to the TCR antigen-binding site of lymphocyted T cells (CD8+; MHC class I, CD4+; MHC class II) (Klein et al.

1989). Once the lymphocyted T cells are bound to a specific MHC molecule, the adaptive immune response is activated (Janeway et al., 2001; Travers, et al., 2001; Wood, 2006).

The MHC is one of the most important model genes in a multitude of evolu- tionary studies ( Burri, et al, 2008; Edwards et al., 2018; Kaufman et al., 2018).

Moreover, MHC has been cloned from representatives of all vertebrates with

the exception of jawless fish (Agnatha) (Klein et al., 1993) and it has been

extensively studied in humans (e.g. Bondinas et al., 2007). This extensive

background knowledge allows the MHC to be used in codon model studies in

order to detect signatures of historical selection within the sequences (Meyer- Lucht, et al., 2008; Luo et al., 2012; Kiemnec-Tyburczy et al., 2018).

In addition to the adaptive immune system, vertebrates are protected by innate mechanisms including macrophages and neutrophils (Cosaro et al., 2000), nat- ural killer cells (Horton et al., 2003) and antimicrobial peptides (AMPs), the latter being secreted from the granular glands on the dermal layer on the skin (Duda et al.,2002). The cells of the innate system provide a non-specific re- sponse to the presence of a pathogen and act immediately with an induced inflammatory responses recognized by non-specific effectors (Janeway et al., 2001). AMPs are an important part of the innate immune system. They are generally short (15-45 aminoacid residues) (Tennessen, 2005) and present in a great variety of taxa including amphibians. Typical organization of an AMP gene consists of Signal Peptide, Acidic Propiece and Mature Peptide domains (Tenessen, 2005, Duda et al., 2002). During the last decades, more than 5000 AMPs have been identified in amphibians (Vanhoye,et al., 2003), but very little is known about AMPs in an evolutionary perspective.

Amphibian populations all over the world are currently suffering massive de- clines. One of the main causes of populations declines are the newly emerging diseases, most significantly the chytrid fungus Batrachochytrium dendroba- tidis (Bd) and Ranavirus. Certain MHC alleles and, to a lesser extent, AMPs have been associated with disease resistance (Woodhams et al., 2006; Savage

& Zamudio, 2011, Bataille et al., 2015, Flechas et al., 2019, Savage, et al., 2019). For intance, Pseudomonas fluorescens together with AMPs secreted from the skin work in synergy to effectively inhibit Bd growth in Rana mus- cosa (Myers et al., 2012).

At present, emerging infectious diseases and climate change represent two

crucial challenges for amphibian persistence. An important question is

whether or not the two are connected to each other. This is uncertain due to

the lack of concrete experimental evidence so far. The study of immune gene

variation and its relation to infection status is important, especially at higher

latitudes, which coincide with the edge of distribution in many species and

consequently the depletion of genetic variation (Eckert et al., 2008, Lewan-

dowska‐Sabat et al., 2010). At higher latitudes, anuran populations are more

likely to suffer infectious diseases outbreaks under warming environmental

conditions as these populations are locally adapted to cooler climates (Cohen

et al., 2017; 2018). In this thesis, I investigate immune genetic variation in

three anuran species as an important first step to generate insights into the

links between newly emerging diseases, climate change and its potential ef-

fects on amphibian populations.

Research aims

The specific aims of this thesis are the following:

 To study how genetic variation in MHC and AMP is distributed along latitudinal gradients in three anurans (Rana arvalis, Rana temporaria and Bufo bufo).

 To investigate which evolutionary processes have shaped MHC and AMPs genetic variation and to study the influence of demographic history on genetic variation.

 To explore the effects of contemporary and historical selection over the latitudinal gradient on within and among-site gene diversity.

 To investigate the role of specific MHC alleles on Bufo. bufo survival

when experimentally infected with a chytrid fungal disease.

Materials and Methods

Study species and sampling collection

Rana arvalis (Paper I, II and III)

The moor frog (R. arvalis) is commonly distributed through most of the northern, central and eastern parts of Europe, eastwards to Siberia. It is ex- tinct in Switzerland and de- clining in China (Kuzmin et al.

2009). Rana arvalis occurs in

a great variety of habitats: tundra, forest, peatlands, moorlands and agricul- tural landscapes (Vershinin, 1997, Kuzmin et al. 2009).

Rana temporaria (Paper II)

The common frog (R. tempo- raria) is found throughout much of Europe, up to Polar sea in the north It is absent from southern and central Iberia, southern Italy and Caucasus.

Rana temporaria is present in a variety of terrestrial and aquatic habitats, breeding in both temporary and permanent

ponds (Kuzmin et al., 2009). Rana temporaria is the most common amphibian species in Scandinavia.

In Fennoscandia, Rana frogs breed from mid-March to mid-May depending on the latitude, a bit earlier for R. temporaria than for R. arvalis. Both species have the capacity for a faster development at northern latitudes (R.arvalis:

(Luquet et al., 2019); R. temporaria: (Laugen et al., Merila, 2003). Also, R.

temporaria is able to respond to pond-drying adaptively by speeding up their development (Laurila & Kujasalo, 1999).

Bufo bufo (Paper IV)

The common toad species group is widesly distributed in Europe and Asia. Accord- ing to Arntzen et al., (2013) this group consists of four species: Bufo eichwaldi, B.

spinosus, B. verrucosissimus and my study species, B.

bufo. Depending on latitude, the breeding season in Fen- noscandia occurs between

April and late May, being later at higher latitudes (Robertson and Höglund 1987). Bufo bufo is an adaptable species present in a great variety of natural habitats (coniferous, mixed and deciduous forests, groves, bushlands, mead- ows, arid areas) and in many modified habitats (Agasyan et al., 2009).

Sample collection

I collected R. arvalis eggs (Paper I, II and III) from 14 populations in five different regions along a 1800 km latitudinal gradient from northern Germany (Hannover) to northern Sweden (Luleå: Fig. 1). Eggs from R. temporaria (Pa- per II) were collected from 17 populations in six regions from northern Ger- many to north of Sweden and Finland (Fig. 2). At each site, I collected ca. ten eggs from each of 20 freshly laid clutches. The eggs were transported to Upp- sala and kept in a climate-controlled room at 16 °C. After hatching, the tad- poles were euthanized (at stage 25, (Gosner, 1960)) with an overdose of MS222, preserved in 96% ethanol and stored at 4 °C until DNA extraction.

For B. bufo MHC genotyping along the gradient (Paper IV), I collected toe- clipped tissues in situ in the field from 20 adult individuals per population from a total of 12 populations (Fig. 3). The tissue was preserved and stored in 70% ethanol at 4 °C until DNA extraction. For the experimental infection study (Paper IV), I carefully collected ca. 10 B. bufo eggs from several clutches per pond in order to reduce the risk of sampling related individuals.

Eggs were collected from four ponds, two ponds both in the northern (LU1

and LU2) and in the southern part (PM and PH) of Sweden. Eggs were brought

to the laboratory in Uppsala and reared to metamorphosed toadlets before ex-

perimental infections and MHC genotyping was carried out.

Fig 1. Sampling site locations of Rana arvalis.

Fig 2. Sampling site locations of Rana temporaria.

Fig 3. Sampling site locations of Bufo bufo (MHC-gradient) are represented with red circles.

Blue triangles represent Bufo bufo locations chosen for sampling the eggs for the experimental

infection study (Paper IV). These locations were also part of the MHC-gradient sampling.

Molecular Methods

DNA extraction (Paper I, II and IV)

Genomic DNA was extracted from 240 R. arvalis tadpoles (one individ- ual/clutch) (Paper I, II) using a high salt extraction precipitation protocol (modified from Paxton et al., 1996). DNA from 170 R. temporaria tadpoles (Paper II) and 460 B. bufo tissue samples (Paper IV) was extracted by using Qiagen DNeasy Tissue Extraction Kit for Blood and Tissue (Qiagen) accord- ing to the manufacturer’s protocol. Quality control and quantification for DNA were done with a Nanodrop spectrophotometer (Thermo Scientific).

Species verification (Paper I, II) was carried out by mtDNA cytochrome b amplification followed by the addition of HaeIII restriction enzyme (Palo &

Merila, 2003). Digestions by HaeIII produce different, easily distinguishable banding patterns between R. arvalis and R. temporaria.

Amplicon sequencing and library preparation (Paper I, II and IV)

Target-selected genes included the MHC class II exon 2, which contains the largest part of the peptide binding region, and the AMP group of genes (Brevinin, Temporin and Palustrin). These regions were amplified following standard PCR protocols as described in paper I, IV and II, respectively. For- ward and reverse primers were modified with an 8-bp barcode and a sequence of three N s that allows cluster identification for Illumina Miseq sequencing.

Purified samples were quantified and combined into 4 (Paper I) and to 8 (Pa- per II and IV) equimolar pools for library preparation. Pools were sequenced on an Illumina Miseq at the National Genomic infrastructure (NGI), hosted in SciLifeLab, Uppsala, Sweden (Lampa et al., 2013).

Other genetic markers (Paper I)

In paper I, I compared evolutionary patterns between target-selected genes

(MHC) and neutral markers (microsatellite markers). I genotyped 28 microsat-

ellites: 15 were included in the study but only nine were identified as neutral

markers (R1atCa17, Rtempμ4, Rtempμ5, RCIDII, EU_06, EU_12, EU_15,

EU_19, EU_24). All microsatellites had been isolated from different Rana species. PCR amplification and genotyping are described in detail in chapter I (Paper I). Samples were genotyped using GeneMapper® Software 5 (Life Technologies ^TM ).

Experimental infections (paper IV)

I used two different Bd isolates: one from the United Kingdom (GPL-UK) and the other from Sweden (GPL-SWE). Each of these was used to infect 50 B.

bufo post-metamorphosed toadlets; 51 toadlets were treated with a sham con- trol. Experimental individuals were exposed to 12.10 ⁶ zoospores of Bd in 30 ml of reconstituted soft water (RSW) for 5 h. Metamorphs in the control group were exposed to an equivalent volume of sterile media and RSW for the same time period, at the same stage of development. Experiments were conducted and the animals were housed in the sealed facilities at the Swedish Institute for Veterinary Science, Uppsala.

Data Analyses

MHC and AMPs (Paper I, II, IV)

MHC and AMP sequences were, respectively, combined into single forward reads using FLASH (Magoč & Salzberg, 2011). I used jMHC (Stuglik et al., 2001) to demultiplex, to generate alignments of all variants per amplicon and to filter out PCR artifacts and chimeras. I removed amplicons with < 300 reads from the analysis due to poor quality. Only sequences with open reading frames were considered as valid alleles. In paper I, I assigned the two most frequent variants within each amplicon as valid MHC alleles that occurred in at least 3% of the reads suggested by previous studies (Babik et al., 2009;

Galan et al., 2010), whereas in paper II and IV I generally assigned more than two variants as valid alleles. In addition to the 3% rule, I followed the degree of change method (DOC) in paper I, II and IV. The DOC method criterion calculates cumulative amplicon sequencing depth among the variants in every amplicon. In this procedure, a calculation of the coverage break point (DOC statistic) around each variant is estimated. The variant with the highest DOC value is assumed to be the last true allele (Lighten et al., 2014).

Valid MHC and AMP alleles were imported into MEGA v.7.0 (Kumar et al.,

2006) and compared to other anuran sequences (accessed from Genbank) by

multiple sequence alignment using ClustalW/MUSCLE. Verified variants

were labelled following the nomenclature suggested by Klein, (1975): four

digits for the species abbreviation, followed by gene*numeration (e.g the MHC allele Bubu_DAB*01; the Brevinin allele Raar_Brev*01).

Statistical methods for measuring genetic variation and detecting selection (Paper I, II, III)

In paper I, I tested for outliers from neutrality in microsatellite markers and MHC by using a hierarchical Bayesian method in Bayescan (Foll & Gaggiotti, 2008) and Lositan softwares (Antao et al., 2008).

Disentangling the effects of evolutionary forces and demography acting on specific genotypes, both of which can affect genetic structure, is a very com- plex task. In paper I, I compared patterns of diversity and population structure at the MHC class II exon 2 with neutral expectations. To examine population structure and population differentiation in detail, global and pairwise F ST be- tween populations was implemented in the R-package hierfstat (Goudet, 2005) with a previous data adjustment (F ST /(1-F ST )) based on Slatkin´s (1995) method. Then, I tested for isolation by distance (IBD) by comparing F ST and log-distance (km) for every population pair. Mantel test correlations were per- formed in R using the Adegenet package (Jombart, 2008). Genetic clustering of microsatellite markers was performed by STRUCTURE (Pritchard, Stephens, & Donnelly, 2000) to find the most likely number of clusters (K) and to assign individuals to these defined clusters. I visualized special struc- ture of microsatellite marker genetic data by using discriminant analyses of principal components (DAPC) in the R package Adegenet (Jombart, Devillard, & Balloux, 2010).

In paper II and III, I measured nucleotide diversity (П), number of segregating sites (S) and average number of pairwise nucleotide differences (theta K) within MHC class II exon 2 and AMPs sequences in DNAsp v5.0 (Librado &

Rozas, 2009). In paper I and III, observerd heterozygosity (H O ), expected het- erozygosity (H E ) and allelic richness (AR) were estimated in Fstat (Goudet, 1995) and deviation from Hardy-Weinberg equilibrium were tested in ARLE- QUIN v. 3.5 (Excoffier & Lischer, 2010) in paper I and in Genepop (Rousset, 2008), in paper III.

In order to detect the signal of positive selection within the MHC class II exon

2 nucleotide sequences (Paper III), I calculated the average synonymous (dS)

and non-synonymous (dN) subtitutions and their overall ratio (dN/dS, ω) in

MEGA v.7.0 (Kumar et al., 2016). I used Z-tests (incorporated in MEGA) to

estimate deviations from neutrality (dN≠dS) on the peptide binding region

(PBR) compared to the non-PBR sites. Several additional codon models meth-

ods were used to test signatures of historical selection in paper II and III. First,

I used OmegaMap (Wilson and McVean2006) to detect codons under positive selection dN/dS (ω) and signatures of recombination (Rho;ρ), based on the detections of patterns of linkage disequilibrium (LD) by Bayesian interference (Paper III). A second method used in papers II and III to detect signatures of positive selection was the CodeML, implemented in PAML package (Yang, 2007). This method is based on phylogenetic analyses of protein sequences constructed using a maximum likelihood approach. In paper III, selected co- dons in the amphibians were visually compared to the putatively selected co- dons of the MHC class II in humans (Bondinas et al., 2007) in every case. In paper II, I additionally implemented two methods within the (HyPhy) package (Pond, Frost, & Muse, 2005): 1) the random effect likelihood (REL) and 2) the effect likelihood (FEL) to detect codons subjected to positive selection in AMPs to contrast with the results obtained by CodeML.

Phylogenetic trees (Paper II, III)

I used the neighbor joining methods and unrooted phylogenetic networks to illustrate the phylogenetic relationship among R. arvalis MHC sequences in paper III and the minimum spanning network to show the phylogenetic rela- tionships among R. arvalis and R. temporaria AMP sequences in paper II.

Infection analyses (Paper IV)

In paper IV, I tested the potential differences in survival (as a binary trait) between individuals infected with different fungal isolates (GPL-UK, GPL- SWE, control). The effects of the geographic location (region and population the presence of specific MHC haplotypes were analysed in a generalized linear model run in the R packages car (Fox et al., 2012) and nlme (Pinheiro et al., 2017).

Other analyses (Paper I)

In paper I, I compared standardized F´ ST for differentially selected loci (MHC

class II and RCO8640) with standardized G´ ST for neutral markers (microsat-

ellites) between populations in order to detect population divergence between

a northern and southern cluster in R. arvalis. I computed a restricted major

axis regression (RMA) in the R package “lmodel2” (Legendre, 2014).

Results and discussion

Genetic variation along latitudinal gradients (Paper I, II and IV)

Despite decades of studies on how genetic variation is shaped by selection, drift and migration, and how demography affects these processes among and within natural populations, their relative roles are still a contentious issue in evolutionary biology. In paper I, I assessed genetic variation in 12 R. arvalis populations along a 1700 km latitudinal gradient from northern Germany to northern Sweden. I contrasted patterns of genetic variation in MHC class II exon 2 and RCO8640 target selected genes and 9 neutral microsatellites mark- ers. Outlier analyses showed that MHC class II exon 2 and RCO8640 were subjected to diversifying selection in the south and drift in the north. This was directly related to the smaller effective population sizes in the north (Fig. 4).

Fig 4. F

vs expected heterozygosity for the15 microsatellites and the MHC II exon 2 locus. Black dashed lines show the upper and lower 99% confidence intervals with 10 000 simulations from a stepwise mutation model (SMM), loci under neutrality expectations are colored in grey, loci under differential selection are colored in yellow and loci under diversifying selection are colored in red. Figure a) represent the plot for the southern cluster, figure b) the northern cluster, respectively.

I observed a decreasing pattern of genetic variation and unique MHC class II

exon 2 variants at northern latitudes (Fig. 5). I also found clear patterns of

differentiation between northern and southern populations in both selected and

neutral genetic markers. These results emphasize the great importance of his-

torical demographic colonization processes in R. arvalis (Babik et al., 2004;

Knopp & Merila, 2009), described in detail in paper I. Combined with differ- ent selective regimes, a complex pattern of differentiation along the geograph- ical gradient has emerged. In summary, in paper I, I described genetic varia- tion and estimated contemporary selection based on allele frequency data. We found less variation and unique MHC class II alleles at northern and at inter- mediate latitudes.

Fig 5. Allelic distribution of MHC Class II alleles in 12 R arvalis populations (B: Besbyn (Luleå); F: Ernäs (Luleå); N: Nydalasjön (Umeå); H: Holmsjön (Umeå); ÖA: Österbybruk (Uppsala); V: Valsbrunna (Upp- sala); R: Räften (Skåne); S: Sjöhusen (Skåne); T: Tvedöra (Skåne), M: Mardorf (Germany), Se: Seebeck- wiesen (Germany). A: Altwarmbüchen (Germany). Colour coding scheme for MHC alleles is given in the Supplementary Material (Chapter I).

In paper II, I studied genetic variation at AMPs along the same latitudinal gradient in two closely related anuran species: R. arvalis and R. temporaria.

AMP genetic variation was high in all three groups of genes (Temporin,

Brevinin and Palustrin). I found that AMP genetic variation was widespread

and distributed equally along the geographical gradient, without a clear latitu-

dinal pattern (Fig. 6). I translated all the nucleotide sequences into amino-acid

sequences in order to investigate potential functional diversity patterns along

the gradient. Amino acid variant frequency showed a structured north to south

pattern, with relatively lower peptide variation at northern latitudes. This

structural pattern differed between group of genes and species. For instance, I

found a structured pattern in amino acid variation in Temporin for R. tempo- raria and in Brevinin for R. arvalis. The different genetic diversity patterns depending on the group of genes might suggest between-species variation in importance of specific AMPs in the defence against infectious diseases.

In paper IV, I studied genetic variation at MHC class II exon 2 in B. bufo along a similar latitudinal gradient as described previously (see Fig. 1 in paper IV).

I found a decline in MHC genetic variation at northern latitudes in the direc- tion away from putative glacial refugia. This has been observed earlier in a variety of species (Babik et al., 2008; Babik et al., 2009; Wielstra et al., 2015;

Talarico et al., 2019). The lower MHC genetic variation at northern latitudes (Paper I and IV) might have negative consequences for population survival, especially with emerging infectious disease, since immunogenetic diversity is predicted to be shaped by selection and/or drift at northern latitudes as sug- gested in paper III.

Amplification of a single MHC locus greatly simplified my work in paper I. I assigned from one to two true alleles per amplicon in R. arvalis. In B. bufo however, I could assign up to four MHC class II valid alleles (Paper IV), and from 1 to 11 valid alleles for AMP genes in R. temporaria and R. arvalis, (Paper II). Therefore, MHC class II loci (B.bufo) as well as the AMP loci (R.

arvalis and R. temporaria) have been duplicated several times within the ge- nomes of these species, indicating multi-locus systems. One of the main chal- lenges of studying a multi-locus system is the difficulty in estimating contem- porary selection within and among populations based on allele frequency data.

It is not straightforward to assign one or two alleles per locus regardless to the

number of loci present in your system. Therefore, in these specific cases, func-

tional selection was inferred directly from the nucleotide sequences as de-

scribed in the section below.

23

. Allele fr equency distr ibution of t he T em por in, B revinin and Palustr in g roup of genes in 14 R, arvalis p opulations (A: A ltwarm büchen; M : Mar dof; Se: Seebeckwiesen; S : T : T vedör a; R: Räften; AÖ: Öster by br uk; V : Valsbr unna ; C: C ray fish/Alm by ; H : Holm sj ön; N y: N ydalasjön; B: Besbyn; E : E rn äs; G: G em tr äsket) and 17 R .temporaria ns (B: A ltwar m büchen; K : Schneer en – K uhteich; W : O ste rl oh – W ienhausen; HO: H öör ; SF: S jöbo S; S L : Ö str a O dar slöv; G rä : Gränby ; KO: K olvia; Ö : Ös ter by br uk; af te å; G ro ss : G ro ss jö n; L T 1: B esbyn; LT2: Mock träsket; L T 3: G em tr äsket; Ga: Gälliva re; L e: L eipojärvi; F: K ilpisjärvi). Colour c oding schem e for the alleles is given in the m entary M aterial Ch ap te r II )

Diversity and Selection at the sequence level (Paper II and III)

I used ratios of synonymous (dS) to non-synonymous substitutions (dN) and calculation of different diversity indices within the sequences to make infer- ences about selection acting on immune genes (AMPs and MHC) in popula- tions along latitudinal gradients.

In paper II, I studied diversity at the sequence level in AMPs. I found higher nucleotide diversity (ᴨ), number of segregating sites (S), and pairwise nucle- otides differences (Theta k) within the Mature Peptide domain compare to the Acidic Propiece domain. These results suggest that the mutation rate and se- quence divergence are higher within the Mature Peptide compared to the Acidic Propiece domain. Moreover, I detected signatures of positive selection acting on the Mature Peptide domain within the Temporin-Brevinin and Pal- ustrin AMP genes. In fact, 12 out 14 in Temporin-Brevinin and four amino- acid codon sites revealed signatures of positive selection (dN/dS>1), suggest- ing that the Mature Peptide domain is subjected to a strong selection pressure.

I also investigated diversity and signatures of historical selection in paper III

using the same MHC class II amplicon sequencing set used in Paper I. In paper

III, I assigned R. arvalis populations into two different clusters: a northern

and a southern cluster, respectively. Grouping the population by clusters to

infer historical selection was in line with results obtained in paper I and two

previous studies supporting a dual post-glacial recolonization history of R. ar-

valis (Babik et al., 2004; Knopp & Merila, 2009). In paper III, I found lower

nucleotide diversity (ᴨ) and nucleotide pairwise differences (Theta k) in the

northern cluster compared to the southern cluster. I also estimated ratios of

synonymous and non-synomous sutbitutions (dN/dS) and used a Z-test to an-

alyse differences between northern and southern clusters when comparing

PBR and non-PBR sites. I found significant positive selection in both clusters,

but weaker selection acting in the north. By running codon model evolution

approaches, I also detected fewer amino-acid codons under positive selection

at northern latitudes (Fig 7). The low sequence divergence and weaker signs

of selection in the northern compared to the southern cluster are most likely

explained by purifying selection acting on northern latitudes, and/or the gen-

eral effect of demography associated with stronger drift acting on northern

latitudes while divergent selection might be more likely acting at southern lat-

itudes. In summary, I found different selective pressures acting at northern and

southern latitudes. This may be related to the diversity of parasites, the histor-

ical distribution of genetic variation along the geographical gradient and the

complex interactions between hosts, parasites and the environment.

Fig 7. Sliding window dN/dS, for the 272 bp exon 2 fragments of the MHC II exon 2 (window size 5b, step size 15) in orange for the northern group and in blue for the southern group. PBR position from Bondinas et al. 2007 are represented with a star below the x axes.

MHC Alleles associated with Chytrid fungus infection (Paper IV)

During the past decades, amphibian species have suffered enormous world- wide declines from habitat loss (Greenberg et al., 2018), habitat fragmentation (Belasen et al., 2018) and emerging infectious diseases, such as chytrid fungi and Ranaviruses (O’hanlon et al., 2018, Scheele et al., 2019) .

The main focus of paper IV was to investigate the link between specific MHC alleles screened and individual survival in a chytrid fungus infection chal- lenge. I genotyped MHC Class II exon 2 in 300 B. bufo wild-sampled individ- uals and in 160 B. bufo from experimental infections. Individuals were in- fected with one of two different Bd isolates, from Britain (GPL-UK) and Swe- den (GPL-SWE) (Fig 8a).

Previous studies have shown differences in survival related to MHC class II

haplotype in Lithobates yavapaiensis (Savage & Zamudio, 2011, 2016). I ob-

served higher mortality among the experimentally infected northern B. bufo

individuals. I also found significant differences in survival depending on the

Bd isolate used to infect individuals. I also detected two MHC-haplotypes with

different susceptibility to Bd. One was linked with higher mortality for indi-

viduals infected by the Swedish isolate (Bubu_DAB*9), but not when indi-

viduals were infected by the UK Bd isolate. Conversely, Bubu_DAB*2 pro- vided protection against the Swedish Bd isolate but not against the UK isolate.

This study is the first study reporting specific MHC class II variants that con- fer a survival advantage or disadvantage depending on the Bd-isolate, see Fig 8b) and c).

Understanding how genetic variation is distributed among and between popu- lations along extensive geographical gradients, measuring contemporary se- lection and inferring signatures of historical selection to elucidate current and past evolutionary mechanisms shaping genetic variation has been discussed in the previous sections of the thesis (paper I, II and III). Finally, paper IV is an applied study where the evolutionary knowledge accumulated from previous chapters was put in context with experimental infection studies to ultimately understand and try to mitigate pathogen and co-infections that threaten am- phibian biodiversity.

Fig 8. Allele frequency distribution of MHC class II exon 2 from individuals infected with Bd is represented

in figure a). The predicted effect of the two alleles with potential Bd-strain interactions. b) The

Bubu_DAB*2 haplotype provided protection against the Swedish Bd-strain, but was detrimental for indi-

viduals infected with the UK Bd-strain. The predicted values assume using a background of carrying the

Bubu_DAB*6 and Bubu_DAB*7 haplotype as this was the case for all three individuals carrying the

Bubu_DAB*2 haplotype. None of the three individuals with Bubu_DAB*2 were present in the control

treatment, which is thereby excluded. c) Bubu_DAB*9 haplotype has a detrimental effect on survival for

Bd-infected individuals, especially for the UK-strain. All predicted values assume the region south as this

was the case for all individuals carrying these MHC-alleles.

Conclusions and future directions

My thesis provides important insights to our understanding of immune genetic variation at large geographic scale. My thesis shows that genetic variation among regions and populations along a latitudinal gradient can be explained by complex patterns of selection, drift and migration, which shape variation at the MHC and AMPs in the anuran species. Additionally, my thesis reveals signs of contemporary and historical selection acting at MHC and AMPs along the geographical gradient (Paper I, II, and III). I found specific MHC variants in B. bufo that seem to confer resistance or susceptibility in to the chytrid fun- gus. Also, I found a general trend of decreasing genetic variation towards northern latitudes at MHC, but not in AMPs. I hypothesize that smaller north- ern populations, which have less MHC variation, might be at greater risk of extinction, as they may not harbour immunogenetic variants which can recog- nize pathogens and initiate immune responses. This fact is important under warming/extreme climate conditions where amphibians might become more susceptible to infections, especially when individuals are locally adapted to cooler climates.

In the future, there is a need for further studies to improve our understanding

on how genetic variation is geographically structured and how evolutionary

forces shape the genetic variation in amphibian species. Amphibians are fac-

ing strong declines, and emerging diseases are a major contributor to these

declines. Therefore, to improve our understanding on how genetic variation is

distributed within and between populations, it is crucial to take into consider-

ation in future conservation programs the evolutionary processes and the pres-

ence of advantageous and detrimental immune genetic variants. Implementing

this knowledge might increase the persistence of amphibian populations and

help preventing mass extinctions in the future.

Svensk sammanfattning

Genetisk variabilitet är det grundläggande kravet för att populationer ska kunna anpassa sig till nya förutsättningar då deras livsmiljöer förändras. Ge- netisk variabilitet avser variationen i det genetiska materialet hos en populat- ion eller art. Gener består av nukleotidsekvenser belägna i en specifik region av en DNA-molekyl. Dessa nukleotidsekvenser som hör till samma DNA-om- råde koda för olika proteiner. Variationen i sekvensen av aminosyror, som bildar ett protein, kan resultera i en förändring i proteinets funktionalitet vilket orsakar skillnader till exempel i reproduktionshastigheter, överlevnad eller be- teendet hos individerna av en art. Stora populationer har generellt en hög ge- netisk variation, medan den genetiska variationen i små populationerär lägre.

Arter med små populationer löper därför större risk att dö ut.

Variationen har påverkats av historiska processer och genetisk variation när minskar vi närmar oss en arts utbredningsgräns. Nordeuropa var täckt av stora istäcken under Pleistocene och de flesta arter begränsades då till södra Europa.

Under Holocene började isen tina och arterna började återkolonisera de tidi- gare nedisade områdena i norr och förlorade genetisk mångfald i denna pro- cess. Det är viktigt att notera att inte bara är koloniseringsprocesserna som förklarar fördelningen av genetisk variation globalt och lokalt. Det finns även andra processe och evolutionära mekanismer som naturligt urval, genetisk drift och genflöden som har och haft betydlese för distributionen av genetisk mångfald.

I bevarandebiologin studeras mönster av genetisk mångfald och associerade

evolutionära processer med hjälp av två typer av markörer: 1) "neutrala" mar-

körer, (dvs DNA-sekvenser som inte är och varit föremål för naturligt urval),

och 2) adaptiva markörer, som tvärtom är under urval i naturliga miljöer. De

mest kända och använda "neutrala" markörerna, är sk. mikrosatelliter och och

enkla nukleotidpolymorfismer (sk. SNPs). Den mest adaptiva markörerna är

sekvenser som kodar för för immunförsvarsgnere i histokompatibilitetskom-

plexet (MHC-komplexet eller HLA hos människor). I min avhandling har jag

också för första gången använt antimikrobiella peptider (AMP) i studien av

mångfaldsmönster. MHC-komplexet består av en grupp gener som studeras

mycket väl hos människor och finns hos alla ryggradsdjur. Det kodar för en

grupp molekyler (glykoproteiner) som deltar i presentationen av antigener till

T-lymfocyter, vilket möjliggör aktiveringen av förvärvat immunsvar. MHC

anses vara den mest polymorfa gruppen av gener och är kopplad till sjukdoms- resistens. Antimikrobiella peptider proteiner har också antibiotiska egen- skaper och är grundläggande komponenter i det medfödda eller inflammato- riska immunsvaret. Dessa peptider fungerar som ett medel för försvar mot sjukdomar som orsakas av olika mikroorganismer, såsom bakterier, svampar eller virus.

Cirka 501 amfibiearter har försvunnit under de senaste decennierna. Denna nedgång beror delvis på fragmenteringen av deras livsmiljöer som en följd av antropogen stress och förorening. Dessutom är denna drastiska minskning av antalet amfibier associerad med framväxten av många patogener, såsom ra- navirus eller chytrid-svamp (Batrachochytrium dendrobatidis), känd som Bd- infektion. Den senare har blivit den patogen som har gjort mest skada på den biologiska mångfalden. Allt pekar på att de dödande svamparnas ursprung är asiatiskt, där arterna inte påverkas. Ett problem är att svampen är känslig mot temperaturen och föredrar relativt kallt vatten och globalt växer i regioner som ligger mer norrut eller där temperaturen är gynnsam för svampen. På den skan- dinaviska halvön är förekomsten av den svampen hög i södra Sverige men den har ännu inte hittats i de nordligare regionerna. Därför tror vi att undersök- ningen av genetisk mångfald är nödvändig, särskilt i de nordligaste region- erna, eftersom vi förväntar oss att mångfalden är lägre och det kan innebära en större mottaglighet av svampen hos amfibier om denna typ av sjukdomar sprids norrut genom den globala uppvärmningen. För närvarande finns det inte många studier som associerar närvaron av en viss genetisk variant av MHC eller antimikrobiella peptider med känsligheten eller sårbarheten att infekte- ras. Det är också väldigt lite känt om variationen i genetisk mångfald som i allmänhet, och i speciellt med avseende på MHC och antimikrobiella peptider, och hur det varierar med breddgrad hos amfibier.

I denna doktorsavhandling studerar jag evolutionära mönster och mångfalds-

processer hos immungener genom att använda molekylära tekniker i olika

grodarter (åkergroda, Rana arvalis, vanlig groda, R. temporaria och vanlig

padda, Bufo bufo) längs latitudgradienter från norra Tyskland (Hannover) till

norra Sverige (Luleå / Kiruna) över mer än 1700 km. I kapitel I studerar jag

både de evolutionära och demografiska processerna som påverkar den gene-

tiska mångfalden i exon 2 i histokompatibilitetskomplexet (MHC). Resultaten

indikerar att det finns en tydlig minskning av mångfalden i både neutrala mar-

körer och i MHC klass II och att genetisk variation mellan regioner och popu-

lationer kan förklaras av kombinationen av komplexa urvalsmönster i söder

och gendrift i norr. Det är troligt att detta är resultat av mer fragmenterade

populationer som resultat av den övergripande effekten av efterglacial rekolo-

nisering över gradienten. I kapitel II studerar jag den genetiska variationen

längs latitudgradienten, karakteriseringen av tre olika mikrobiella peptider

(Temporin, Brevinin och Palustrin) och de därmed sammanhängande evolut-

ionära processerna i R. arvalis och R. temporaria. Huvudresultatet i detta av-

snitt är att vi hittade en mycket hög genetisk variation längs hela gradienten i

de tre grupperna av antimikrobiella peptider. Överraskande och i motsats till

våra förväntningar distribueras de genetiska varianterna utan någon form av

demografiskt mönster längs gradienten. Vi observerade också tecken på se-

lektion bland nukleotidsekvenserna av dessa antimikrobiella peptider och fö-

reslog att det observerade mönstret kan förstås som som lokal anpassning till

olika populationer av parasiter. I kapitel III undersöker jag historiska ur-

valsmönster i MHC hos R. arvalis. Denna studie tyder på ett divergerande

urval som verkar på populationerna söder om gradienten samtidigt som det

föreslås ett historiskt uniformt urval i de nordliga populationerna. Slutligen i

kapitel IV studerar jag, som i kapitel I och II, mönster av genetisk mångfald

längs en latitudgradient, men den här gången hos B. bufo eller vanlig padda

och vi finner att den genetiska mångfalden minskar från söder till norr som

hos R. arvalis. I detta kapitel studerar jag också den genetiska mångfalden i

MHC klass II hos individer som infekterats med 2 isolat (GPL-UK och GPL-

SWE) av chytridsvampen (Bd) i laboratoriet. Syftet var att leta efter en kopp-

ling mellan genetiska varianter av MHC och sannolikheten att överleva infekt-

ion. Resultaten tyder på att överlevnaden är lägre i nordliga populationer jäm-

fört med populationer från södra Sverige, vilket skulle bekräfta vår initiala

hypotes om en högre mottaglighet i nordliga populationer av chytrid-svamp

och i förhållande till låg genetisk mångfald i nordliga populationer. Dessutom

har vi funnit att det specifikt finns skillnader i överlevnad när individer smittas

av olika isolat. Detta indikerar att Bubu * 2-allelen och Bubu * 9-allelen är

direkt relaterade till den infekterade individens överlevnad.

Resumen en Español

La variabilidad genética es el requisito fundamental para que las poblaciones tengan la capacidad de adaptarse a nuevas condiciones producidas por un cam- bio ambiental o de hábitat. Variabilidad genética referida a la variación en el material genético de una población o especie.

Los genes están compuestos por secuencias nucleotídicas localizadas en una región concreta de una molécula de ADN, que pertenecientes a la misma re- gión del ADN pueden codificar diferentes proteínas. Una variación en la se- cuencia de los aminoácidos, que forman una proteína, puede dar lugar a un cambio en la funcionalidad de la proteína, causando diferencias por ejemplo en las tasas de reproductividad, en la supervivencia o en el comportamiento de los individuos de una especie. Las poblaciones grandes generalmente po- seen una variación genética elevada mientras que en las poblaciones pequeñas la variación genética es menor, por eso las especies con poblaciones reducidas son un foco de preocupación en conservación.

De manera natural e influenciada por procesos históricos de re-colonización, la variación genética disminuye al acercarnos al límite de distribución de las especies, siendo más elevada en el centro de distribución. El norte de Europa estuvo cubierto por inmensas capas de hielo durante el Pleistoceno y la mayo- ría de las especies se desplazaron e instalaron en el sur de Europa en lo que conocemos como refugios glaciares. Durante el Holoceno el hielo comenzó a descongelarse y las especies comenzaron a colonizar las tierras desde el sur de Europa hasta el norte, perdiendo diversidad genética en este proceso. Es importante resaltar que no sólo los procesos de colonización son los encarga- dos de explicar la distribución de la variación genética tanto a nivel global como local, existen otros procesos conocidos como procesos o mecanismos evolutivos que también contribuyen a ello, como la selección natural, la deriva génica o la migración de las especies. Estos procesos juegan un papel funda- mental en el modelado y distribución de la diversidad genética.

En biología de la conservación, los patrones de diversidad genética y procesos

evolutivos asociados se estudian a través de dos tipos de marcadores: 1) mar-

cadores moleculares “neutrales”, no sometidos a presiones selectivos, 2) los

genes adaptativos, que por el contrario si suelen estar bajo selección en el me-

los microsatelites y los polimorfismos de nucleótido único o SNP (Single Nu- cleotide Polymorphism). Los segundos, son por ejemplo el complejo mayor de histocompabilidad (CMH o MHC y HLA en humanos). En esta tesis se han utilizado, por primera vez, péptidos antimicrobianos (AMPs) en el estudio de patrones de diversidad, rara y ocasionalmente estudiados con este objetivo con anterioridad.

El complejo CMH, está formado por un grupo de genes muy bien estudiados en humanos y hallado en todos los vertebrados, codifican un grupo de molé- culas (las glicoproteínas) y son denominados antígenos de histocompatibili- dad, que participan en la presentación de antígenos a los linfocitos T, permi- tiendo la activación de la respuesta inmunitaria adquirida. Asimismo, el CMH es considerado como el grupo de genes más polimórfico y están estrechamente relacionados con la resistencia a enfermedades. Por otro lado, los péptidos antimicrobianos son proteínas con propiedades antibióticas y son componen- tes fundamentales de la respuesta inmunitaria innata o antiinflamatoria. Estos péptidos actúan como medio de defensa contra enfermedades producidas por diversos microorganismos como, bacterias, hongos o virus.

En la actualidad cerca de 501 especie de anfibios han desaparecido en las úl- timas décadas. Este declive se debe, en parte, a la fragmentación de sus hábi- tats, como consecuencia del estrés antropogénico y la contaminación. Esta disminución drástica del número de anfibios está asociado a la aparición emer- gente de numerosos patógenos como los ranavirus o el hongo quítrido (Ba- trachochytrium dendrobatidis), conocido como infección por Bd. Este último se ha convertido en el patógeno que más daño has hecho a biodiversidad de la historia. Es muy posible que el origen del hongo quítrido sea asiático de ahí que en ese continente las especies no se estén viendo afectadas por esta enfer- medad exterminadora de los anfibios.

El principal problema es que el hongo es sensible a la temperatura y prefiere aguas relativamente frías por lo que está proliferando a nivel mundial en re- giones localizadas cada vez más hacia el norte. En la península escandinava, la prevalencia del hongo quítrido es elevada en la región del sur de Suecia y aún no se ha encontrado en las regiones más septentrionales, por eso se piensa que es primordial el estudio de la diversidad genética sobre todo en las regio- nes más septentrionales, puesto que sería probable que la diversidad sea me- nor, por lo explicado con anterioridad y esto podría suponer una mayor sus- ceptibilidad de las especies de anfibios para hacer frente a este tipo de enfer- medades, ante un calentamiento global.

En la actualidad hay multitud de estudios que asocian la presencia de una de-

terminada variante genética de CMH o péptido antimicrobiano y la suscepti-

bilidad o vulnerabilidad a ser infectado. Sin embargo, apenas se conoce cómo

es la diversidad genética en general, y en CMH y péptidos antimicrobianos en particular y como varía con la latitud en anfibios.

En esta tesis doctoral se estudian procesos y patrones evolutivos y de diversi- dad en genes inmunes, en concreto el exón II del complejo mayor de histo- compatibilidad y en tres péptidos antimicrobianos, utilizando técnicas mole- culares de secuenciación masiva principalmente en diferentes especies de anuros (Rana arvalis, Rana común; Rana temporaria y Sapo Común; Bufo bufo) a lo largo de un gradiente latitudinal, desde el norte de Alemania (Ha- nover) hasta el norte de Suecia (Luleå/Kiruna) de más de 1700 km.

En el capítulo I se analizan tanto los procesos evolutivos como demográficos que afectan a la diversidad genética del exón 2 del complejo mayor de histo- compatibilidad (CMH). Los resultados indican que hay una clara disminución de la diversidad tanto en marcadores neutrales como en CMH clase II y que la variación genética entre regiones y población se puede explicar mediante la combinación de complejos patrones de selección en el sur, deriva génica en el norte, como resultado de la fragmentación de las poblaciones y el efecto en general de recolonización post glacial en todo el gradiente.

En el capítulo II, se comprueba la variación genética a lo largo del gradiente latitudinal, la caracterización de tres péptidos microbianos diferentes (Tempo- rin, Brevinin y Palustrin) y los procesos evolutivos asociados en Rana arvalis y Rana temporaria. El principal resultado de esta sección es que existe una variación genética muy alta a lo largo de todo el gradiente en los tres grupos de péptidos antimicrobianos. Sorprendentemente y al contrario que nuestras expectativas, las variantes genéticas están distribuidas, sin sugerir ningún tipo de patrón demográfico (de sur a norte) a lo largo del gradiente. Se observaron también signos de selección entre las secuencias nucleotidícas de estos pépti- dos antimicrobianos lo que sugiere que la selección disruptiva predomina, de- bido a una adaptación local de las poblaciones a los parásitos de la zona.

En el capítulo III, se investigaron los patrones históricos de selección en CMH en Rana arvalis. Este estudio sugiere una selección divergente cuando se trata de las poblaciones del sur del gradiente, mientras que parece tratarse de una selección histórica positiva, posiblemente direccional en las poblacio- nes del norte, o una fuerte disminución de la diversidad asociada a cambios demográficos o deriva génica.

Finalmente, en el capítulo IV, al igual que en el capítulo I y II se estudiaron

patrones de diversidad genética a lo largo de un gradiente latitudinal, pero esta

vez en Bufo bufo o sapo común y efectivamente se encontró que la diversidad

genética disminuye de sur a norte al igual que en Rana arvalis. En este capí-

infectados previamente con 2 cepas (GPL-UK and GPL-SWE) del hongo quí- trido (Bd) en el laboratorio. El objetivo fundamental era buscar una asociación entre variantes genéticas de CMH y probabilidad de supervivencia a la infec- ción. Los resultados sugieren que la supervivencia de sapo común es menor en poblaciones del norte comparadas con las poblaciones del sur de Suecia, lo que confirmaría nuestra hipótesis inicial de una mayor susceptibilidad de las poblaciones del norte ante la presencia de hongo quítrido como agente pató- geno y la relación con la baja diversidad genética en las poblaciones del norte.

Además, se comprobó concretamente que hay diferencias en la supervivencia,

cuando los individuos son infectados por las diferentes cepas. Por último, los

resultados indican que el alelo Bubu*2 y el Bubu*9 están directamente rela-

cionados con la supervivencia del individuo infectado según la cepa usada en

el experimento.

Gracias, tack, thank you

Probably I do not have either space or words in this thesis to thanks all the people who I have shared with big or small moments during these so many years. I have a bunch of mixed feeling what it makes me feel quite sentimental right now. There are so many people who has helped in this process of becom- ing and independent researcher. I am deeply grateful to have been surrounded by open-minded researchers, good friends and fantastic family.

I am very grateful to my main supervisor, Jacob Höglund for believing in me and for giving me this amazing opportunity as a PhD student. I love your pos- itivism, your always good advice and your eternal encouragement to do eve- rything I wanted to do. I have always felt that you truly have understood all my points during this trip and you perfectly have known how to guide me at every moment, I have really have learned a lot from you. Thank you for all of this and to take so much care of me. A huge thanks to my second supervisor Anssi Laurila. Anssi, thank you for being always there behind the door every time I have need your input and support. I admire you great capacity of listen- ing and your honesty, thank also for believing in me.

I also eternal grateful to one of the main roles of this game, to you Yvonne Meyer-Lutch, I know that I am not going learn ever how to type your surname but I keep on trying. I just want to say to you that without you all this work would not have been possible. Thanks for all the time you have spent with me even when you did not have time, really. You always have found a time for me despite of all the difficulties in your day life. Thank you for your perfection and precision and all transmitted knowledge with the most caring and affec- tion I have ever seen. But also, I would like to thank you for all your care, guidance and preparation to become a mother. All of your advices helped me a lot and you always have quick solutions and answers for all questions related to maternity. Currently you continue helping in this and other aspects. Thanks for all, especially for your supervision and friendship.

Thanks to Alexei to introduced the amazing Nematode world and give me the

opportunity to learn so much about artificial selection by using this fantastic

model system. It was such a great experience working together with Martyna,

Martin and Ana. Martyna, thanks for all the amazing moments in the lab,

working “hand in hand” and having extra time to learn “Maria del Monte”.

patient and the calm you transmit, I would never forget. Ana, we have been a perfect working team, you know that!!!. I really missed your spirit and our

“Candela playlist” and all fantastic moments we have live together at univer- sity and outside university “Cuando nos comemos el primer helado de la pri- mavera?” You have been always and unconditional friend.

Thanks Frank who gave an enormous opportunity to work in your group, be- ing more close to damselflies, those fantastic insects. David, it was such a great experience being your companion in the field. I really have a fantastic time and I learn a lot. I also want to thank you for always being there for sharing good and bad news with me, for understand me when I was very frus- trated and calm me down, for complaining together y “cotillear mucho”. I re- ally miss you my favorite neighbor office mate.

Thanks to Mats, Göran, Ingrid, Ana, Anders B, Lars, Anders Ö, Violeta for introducing me to the Animal Ecology department, first being my teachers during my masters and then thought the amazing cozy environment you were maintaining at the department together with current and past people: Katja, Claus, Richard, Simon K, Emilien, David O., Ahmed, Masahito, David W, Rado Elina, David B,Magnus, Karl, Brian, Murielle, Andre, Paula, Julieta, Foteini, Elena Rosa, Jaelle, Josefine, Jelena, Leah, Will, Suzana, Carolina, Julian, Ivain and others.

I also would like to thank to all the frog/grouse group. Patrik, I will be always grateful to you for all the hard nights we have spent in the night collecting and sampling frogs even knowing that you prefer doing computer stuffs, it does say so much from you, thanks also for all your supportive words. To me, this is just the start of so many things we can do together. Sara, thanks for the time in the field and at the department. Also, I would like to thank you to all the field assistants (David, Catia and Miariam) who came with us and help us a lot and in special to Alex Ritcher-Boix. Alex, I will never forget all the trips to the north with you, all the hard work at night time and all the scientific conversations and non-scientific conversations we have had in our way up north or in the Apadana. Thank you for your patient in the field when all of us were pretty exhausted and frustrated because the frogs were not coping. You have inspired me a lot with your talented ideas. Kai, I know that you are not a member of the froggy group but it is like you are an important part as well.

Thanks for being always smiling, I promise I would learn how to make the best dumpling you have ever seen. Many thanks to all students I have super- vised in one way or another (Ella, Anna, Gesa, Laurens and Filip). I have learned so much from you guys.

Gunilla, I would need a whole book to be grateful to you. I am very glad that

you have been all this year’s very close to me giving me always good advice

and support. You have always trying to help me as much as you can specially

in the stressing moments. You have also been a mentor in the lab, a mum at a times and also a very good friend. Thank you for everything.

Johanna, thank you for always being so kind with me and count me for the Christmas market. I remember the day I got officially the PhD and you were one of the first ones to come and congratulate me with so much love. I would never forget.

Eryn, my almost 4 year’s office mate. Thank you for all special moments, for the dancing to overcome all the stress and frustration. You have been just awe- some. I wish you all the best for your new period in your life as a family of 5.

Germán, thank you for all the time we have spent together at the department, for all the conversations and for the non-success field trips that we did to the east coast to collect P. lessonae adult’s/eggs despite of the bad weather in two consecutive years and the frustration I had inside those ponds. Literarily,

“Tengo una espinita clavada”, we will achieve it one day.

Peter, thank you for being always there in all frustrating moments. You have always solved all kind of problems related to computer and analyses when I have asked for that. I am impressed by your resolute skills. Thank you for the fox collaboration and help, it would not have been possible without you help.

Mara, I would never forget the time we spent together at the department and outside. You are one of the funniest person I have ever meet. I would never forget the intense and dramatic conversations about your PhD interviews, and then you suddenly got it:D. Thank you for all the time together, axo.

Mattias, we initiated this journey more less at the same time. We organized a great Revolution together. I would never forget your organization skills and the updating “sticky notes” on your computer screen. Thanks for your sup- portive discussions about everything, you have become a real friend since the beginning.

Pablo, thank for the intensive moments in the lab. They have been such a great fun. I love your sense of humor and our intense and non-stop conversations about whatever while we were pipetting. ¿Pendiente queda una colaboración para ganar el premio Nobel eh?

Tom, thanks for sharing all wonderful moments while we were teaching and

thank you very much for all the time you spent helping me out with the chapter

4 of this thesis and downloading the data from Uppmax, oh god! I really wish

you all the best for your thesis preparation and defense.