Comparative Genomics of the Major Histocompatibility Complex in the Family Equidae

MASARYK UNIVERSITY FACULTY OF SCIENCE

Comparative genomics of the major histocompatibility complex in the family Equidae

Ph.D. Dissertation

Marie Klumplerová

Supervisor: prof. MVDr. RNDr. Petr Hořín, CSc Brno 2015

Bibliographic entry

Author: Mgr. Marie Klumplerová

Institute of Animal Genetics

University of Veterinary and Pharmaceutical Sciences Brno

Title of Dissertation: Comparative genomics of the major histocompatibility complex in the family Equidae

Degree Programme: Biology

Field of Study: General and Molecular Genetics

Supervisor: prof. MVDr. RNDr. Petr Hořín, CSc.

Academic year: 2014/2015

Number of pages: 125

Keywords: Horse; Donkey; Equidae; selection; association; genetic diversity

Bibliografická identifikace

Jméno a příjmení autora: Mgr. Marie Klumplerová

Ústav genetiky

Veterinární a farmaceutická univerzita Brno

Název disertační práce: Komparativní genomika hlavního histokompatibilitního komplexu u čeledi Equidae

Studijní program: Biologie

Studijní obor: Obecná a molekulární genetika

Školitel: prof. MVDr. RNDr. Petr Hořín, CSc.

Akademický rok: 2014/2015

Počet stran: 125

Klíčová slova: Kůň; osel; koňovití; selekce; asociace; genetická diverzita

Abstract

Genes of the major histocompatibility complex (MHC) are crucial for the adaptive immune system of jawed vertebrates. Variation at the MHC loci affect many important biological traits, including immune recognition, susceptibility to infectious and other diseases, mating preferences, and/or maternal-fetal interaction and pregnancy outcome. MHC has become a paradigm for how pathogens are shaping the patterns of adaptive genetic variation. It is used, as one of the most preferred marker, in many different research fields ranging from human medicine, molecular evolutionary studies to conservation genetics. The family Equidae is a rapidly evolving mammalian group comprising domesticated as well as free-ranging animals occupying a wide range of habitats differing in climatic conditions and pathogen pressure. In this thesis, genetic diversity, molecular evolution and selection of MHC class II loci in equids were characterized. In addition, an extraordinary feature of the equine MHC, polymorphism of its DRA locus, was used to seek for associations with infection and allergy as well as for analyzing the diversity of a model endangered horse population.

Abstrakt

Geny hlavního histokompatibilitního komplexu (MHC) jsou nezbytné pro správnou funkci získané imunity u čelistnatých obratlovců. Proměnlivost v genech MHC má vliv na řadu důležitých biologických znaků, které zahrnují imunitní rozpoznání vlastního od cizího, vnímavost k infekčním a dalším typům onemocnění, výběr partnera i průběh těhotenství. MHC je považováno za typický příklad adaptivní genetické variability utvářené pod vlivem patogenů. Z těchto důvodů se stal hlavní histokompatibilitní komplex jedním z nevíce preferovaných markerů používaným pro studium genetické proměnlivosti v mnoha vědních disciplínách od humánní medicíny, přes evoluční studie až po ochranářskou genetiku. Koňovití jsou rychle se vyvíjející čeleď, do které patří domestikované i volně žijící druhy obývající různá prostředí lišící se jak klimatickými podmínkami, tak výskytem patogenů. V předkládané disertační práci jsem se zabývala genetickou diverzitou, molekulární evolucí a selekcí u jednotlivých lokusů dvou genů třídy II u koňovitých. Dále jsem se věnovala výjimečnému rysu koňského MHC, kterým je polymorfismus genu DRA, a využila jsem variability v tomto lokusu pro asociační studie s infekcí, alergickým onemocněním a k analýze genetické variability v modelové ohrožené koňské populaci.

Contents

ACKNOWLEDGMENT ...... viii

INTRODUCTION ...... 1

The major histocompatibility complex ...... 1

MHC molecules structure ...... 2

MHC polymorphism, its origins and maintenance in populations...... 3

Signatures of selection ...... 6

Evolution of major histocompatibility complexes ...... 7

MHC and disease ...... 8

The family Equidae as model for immunogenetic studies ...... 9

The equine MHC ...... 11

SPECIFIC OBJECTIVES OF THE DISSERTATION ...... 14

MATERIALS AND METHODS ...... 15

Animals ...... 15

Methods ...... 16

RESULTS ...... 17

List of attached manuscripts and the contribution of the author ...... 17

Synopsis of the results obtained ...... 19

MS I...... 19

MS II ...... 20

MS III ...... 21

MS IV ...... 22

MS V ...... 23

MS VI ...... 24

DISCUSSION ...... 25

GENERAL CONCLUSIONS ...... 29

REFERENCES ...... 30

LIST OF ABREVIATIONS...... 39

APPENDIX ...... 40

MS I ...... 40

MS II ...... 80

MS III ...... 87

MS IV ...... 96

MS V ...... 106

MS VI ...... 115

ACKNOWLEDGMENT

I would like to express my gratitude to my supervisor prof. Petr Hořín. Without the opportunity and trust he gave me, I would not be the person I am today. I would like to thank him for his inspiring, kind yet demanding leadership and many hours of discussions, which were usually spiced with his unique sense of humor. I am thankful to my colleagues for creating a friendly atmosphere during everyday work. Many thanks go to Zuzana Bayerová for our, not always scientific, chats and for her help with routine laboratory tasks at the end of my pregnancy. Last but not least, I wish to sincerely thank to my parents and family for their support and love.

viii

INTRODUCTION

The major histocompatibility complex

The immune system allows living organisms to discriminate between self and non-self and thus contributes to maintaining their homeostasis. By providing them with defense against pathogens, the immune system also represents a critical determinant of the mechanisms of adaptation. Pathogens are one of the major driving forces of evolution leaving signatures of selection in the host genomes. Therefore, studies of immunity-related genes (the immunogenome) are important for our understanding of immunity, susceptibility to disease, adaptation and evolution.

In jawed vertebrates, two main branches of immune responses can be distinguished: the innate and adaptive immunity. Molecules of the major histocompatibility complex (MHC) play an essential role in the activation of adaptive immune responses. These cell surface molecules are trans-membrane proteins which can bind antigens of both intracellular and extracellular origin and present them to T- lymphocytes. MHC molecules are encoded in a large genetic region. In mammals, this region spans almost 4 Mb and contains about 230 translatable highly polymorphic genes. Organization of human MHC is shown in Figure 1.

Genes coding for antigen presenting molecules are assembled into 2 regions, class I and class II, separated by a class III region. Class III genes are more diverse in function, not always connected to immune response. Class I MHC molecules can be found on virtually all nucleated cells. They present intracellular-derived peptides to CD8+ T-lymphocytes or interact with natural killer cell receptors, ensuring protection mainly against intracellular pathogens such as viruses. Class II molecules, expressed on antigen presenting cells, are involved in processing of extracellular peptides, which are presented to CD4+ T-lymphocytes, fighting mainly against bacteria and helminths (Flajnik and Kasahara, 2001; Klein, 1986; Krejsek and Kopecký, 2004). Genes for the transporter associated with antigen processing (TAP) and the TAP-binding protein (TAPBP, tapasin) are localized within the extended MHC region. Molecules encoded by these genes are indispensable for the transport and loading of the antigenic peptide onto MHC class I molecule. In this thesis, I focused solely on the classical MHC class II genes.

Figure 1. Organization of human MHC on chromosome 6. Modified from Marotte and Miossec, 2010 and Ober, 1998.

MHC molecules structure

Both class I and class II MHC molecules are heterodimeric integral surface membrane proteins possessing a cytoplasmic tail, a small transmembrane domain and 4 large extracellular domains. A groove, which can accommodate an antigenic peptide, is localized in the most distal extracellular portion of the MHC molecule. This part of the molecule is called the peptide binding region (PBR). It is a groove-shaped structure composed of 8 strands of anti-parallel β-sheets as a floor and 2 anti-parallel helical regions as sides. Differences in the helical regions of class I and class II molecules determine “closeness” of the class I compared to “openness” of the class II molecule. The MHC class I molecules, therefore, bind primarily short, nonameric peptides, whereas the class II MHC molecules bind longer, 15-24 residue peptides. The sites in direct contact with the antigenic peptide are called antigen binding sites (ABS). The antigen binding sites can be conserved or polymorphic. The conserved residues are suggested to bind the main chain atoms of antigenic peptides and the polymorphic residues create pockets, which can harbor the side chains of the peptides. Due to differences among these polymorphic sites, different MHC molecules can exhibit different binding specificities, i.e. selectively bind peptides of different sequences (Brown et al., 1993; Klein, 1986; Krejsek and Kopecký, 2004; Nikolich-Žugich et al., 2004).

Differences in the structure of class I and class II molecules are illustrated in the Figure 2 and 3. The MHC class I molecule consist of a trans-membrane α chain which is non-covalently associated with an invariant β-microglubulin. Both molecules belong to the immunoglobulin protein family, but only the α chain is encoded within the MHC region and forms the peptide binding region. More specifically, the PBR is composed of α1 and α2 domains of the class I MHC molecule, encoded in second and third exon of the class I gene. In comparison, the class II MHC molecule is composed of two trans- membrane glycoproteins, where both chains contribute to the formation of the PBR with their α1 and β1 domains, both encoded in second exons of α and β class II genes.

Figure 2. Illustration of class I MHC and class II MHC molecules structure. Adapted from Hughes and Yeager (1998).

MHC polymorphism, its origins and maintenance in populations

The MHC genes are known to be the most polymorphic genes in the vertebrate genome (Klein, 1986). Their extreme polymorphism is generated by mutation, recombination and gene conversion, which are accompanied by gene duplications and gene losses. The relative contribution of individual mechanisms to the origin of MHC polymorphism is not known. Two competing hypotheses attempt to explain it (Martinsohn et al., 1999; Nei and Rooney, 2005). The concerted evolution suggests the gene conversion as the main mechanism, whereas the birth-and-death hypothesis deny its major role and emphasize the importance of gene duplication and gene losses (Gu and Nei, 1999). 3

Figure 3. Structure of the extracellular portion of human leucocyte antigen (HLA) molecules. (A) Structure of the peptide binding region of class I HLA with regions encoded by exons 2 (blue), 3 (green) and 4 (red) of α-chain. (B) Structure of the class I HLA with bound peptide. (C) Structure of the peptide binding region of the class II HLA with regions encoded by exon 2 of α-chain (yellow), exon 2 of β-chain (light blue) and exon 3 of β-chain (red). (D) Structure of the class II HLA with bound peptide. Adapted from Warren et al. (2012).

The high level of genetic diversity can be observed between species as well as within species in both number of genes and allelic diversity (Hedrick, 1998; Kelley et al., 2005). The sequence polymorphism is most important in the PBR, especially in those sites in direct contact with the antigenic peptide and with the T-cell receptor. For this reason, it has been widely accepted since a long time that pathogens are the main driving force of the MHC polymorphism. Pathogen-mediated selection can be explained by three different mechanisms: heterozygote advantage, rare-allele advantage and fluctuating selection (Spurgin and Richardson, 2010).

The hypothesis of heterozygote advantage (also called over-dominant selection) suggests that heterozygotes at MHC loci can respond to a greater range of pathogen peptides as compared to homozygotes and are likely to have higher relative fitness. As a result, more MHC alleles will be preserved in the population. The more divergent is the range of binding specificities of particular alleles the more likely will be the heterozygous individual favored by the over-dominant selection (Doherty and Zinkernagel, 1975; Jeffery and Bangham, 2000; Spurgin and Richardson, 2010). Although the heterozygote advantage has received convincing empirical support (Osborne et al., 2015; Penn et al., 2002), a computer simulation of the host-pathogen coevolution has shown that the heterozygote advantage per se is insufficient to explain the high degree of polymorphism at the MHC (Borghans et al., 2004).

Rare-allele advantage hypothesis (also called negative frequency-dependent selection) assumes that pathogens can adapt and avoid being recognized by the most common MHC alleles. A new or rare allele may offer a better protection in such a situation. Such allele may gain a selective advantage and increase in its frequency. However, as the selected allele frequency is increasing, it becomes more likely a target for pathogen adaptation. This dynamic process of race of arms has cyclical character and causes pathogen and MHC frequencies to fluctuate in time (Spurgin and Richardson, 2010; Takahata and Nei, 1990). In a computer simulations, the rare-allele advantage was shown to be more powerful mechanism of maintenance of the MHC polymorphism compared to the heterozygote advantage (Borghans et al., 2004; Ejsmond et al., 2010).

Finally, the fluctuating selection hypothesis proposes that pathogen pressure is changing in space and time causing also the directional selection exerted on MHC genes to fluctuate spatio-temporally. Different MHC alleles are, therefore, favored in different points in space and/or time and the MHC diversity is maintained across different sub-population. According to this hypothesis, the changes in pathogen pressure are not determined by the pathogen-host coevolution but rather by biotic and abiotic environment changes. Fluctuating selection was shown to be able to maintain the MHC polymorphism even in the absence of the heterozygote and the rare-allele advantage (Hedrick, 2002).

The discrimination between different selection mechanisms involved in the maintenance of MHC polymorphism and assessing their relative importance is a rather difficult task. Similar patterns of MHC diversity can be produced by all proposed mechanisms and there are usually multiple explanations for experimental data. The three mechanisms are not mutually exclusive and are likely to interact. Other types of

5 selection, were also suggested to be involved in the maintenance of MHC variation: negative assortative mating and maternal-fetal interaction (Hedrick, 1998). These mechanisms are however considered as “amplifiers”, helping to reach an optimal number of MHC alleles rather than equivalent to pathogen-mediated selection (Spurgin and Richardson, 2010).

Signatures of selection

Selection is expected to produce detectable effects on functionally important genes and its signatures can be identified on molecular and/or population level. Different types of natural selection contribute to the shaping of genetic variation observed among MHC loci. Both positive/diversifying and negative/purifying selection play an important role and determine sites with functional importance. However, more attention is typically being paid to positive selection, because it is associated with adaptation (Biswas and Akey, 2006). Comparative approaches to detect past selection use sequences from multiple different species and, as a major tool, the ratio of non- synonymous (dN) to synonymous (dS) substitutions rates. The rates of synonymous and non-synonymous substitutions are predicted to be similar under neutrality. Higher rate of synonymous substitutions is anticipated in molecules under purifying selection, since majority of non-synonymous substitutions is expected to have deleterious effect. If positive selection favors new beneficial mutations, the non-synonymous changes are predicted to outweigh the synonymous (Hughes and Yeager, 1998). The global selection analysis averaged across all codon sites of the gene may however be unsuccessful in detection of positive selection due to high number of codon sites under purifying selection (Yang and Bielawski, 2000). Site-specific selection analysis which allows the dN/dS ratio to vary among sites was found more suitable for the analysis of MHC genes. The purpose of these methods is the identification of particular sites targeted by positive (or negative) selection (Nielsen, 2005). Contemporary selection in a population can be detected by a variety of methods, including neutrality tests and analysis based on the population differentiation. Neutrality tests determine whether a distortion in the allele frequency distribution compared to neutral expectations is present in the analyzed data (Biswas and Akey, 2006). Higher level of differentiation within MHC loci compared to neutral loci also indicates acting of positive selection.

Balancing selection is believed to maintain advantageous genetic diversity by a number of mechanisms, as described above. Extreme polymorphism of MHC genes, old origin of contemporary alleles, demonstrated by their long coalescence times, and 6 trans-species polymorphism are considered to be supporting evidence of balancing selection (Meyer and Thomson, 2001). Balancing selection is expected to affect allele distribution in contemporary population. Genotype frequencies of MHC alleles may deviate from Hardy-Weinberg expectations and result in a deficit of homozygotes. Effects of balancing selection in a population can also be postulated based on the existence of associations between particular MHC alleles, MHC heterozygosity and/or total number of MHC alleles with various diseases (Spurgin and Richardson, 2010).

Neutral evolutionary processes, such as migration, genetic drift and fluctuating population size, have to be taken into account to avoid misinterpretation. To overcome this problem, MHC diversity is usually compared to theoretical neutral distribution by using neutral markers such as microsatellites. While neutral demographic processes affect all loci, selection is shaping only functional genes, which can be shown on contrasting patterns of variation at MHC and neutral loci.

Evolution of major histocompatibility complexes

An ancestral MHC is supposed to be the origin of all contemporary major histocompatibility complexes found in jawed vertebrates. The ancestral MHC, along with primordial T-cell receptor and immunoglobulin genes arose before the development of adaptive immunity (Flajnik and Kasahara, 2001; Martínez-Borra and López-Larrea, 2012). The organization found nowadays in most non-mammalian species, i.e. class I genes and antigen processing genes localized next to each other, is considered as primordial. A large inversion, which interrupted the class I genes and antigen processing genes, and brought the class III region in the middle, has formed a typical mammalian MHC (Kaufman, 2013). In mammals, adjacent class I, class III and class II region can be found. The evolutionary history of the class I region is rich in gene duplication and gene loss events, resulting in the variation in the number of class I loci observed among species (Jaratlerdsiri et al., 2014; Kelley et al., 2005). The rate of these processes among class II genes is slower compared to class I and many orthologous loci can be identified in different mammals (Takahashi et al., 2000). However, since individual MHC developed individually in different species, they can differ from the general mammalian model. For instance, a complete deletion of the DQ region replaced by an expansion of the DR region was observed in felines (Yuhki et al., 2003). The knowledge of the evolution and organization of MHC in different species can help us elucidate features fundamental for the immune recognition mediated by the MHC.

The polymorphism observed in the contemporary MHC loci was documented to be tens of millions years old (Těšický and Vinkler, 2015). Variation at MHC genes predates the speciation events and is being maintained by balancing selection due to its favorable functional outcome. The longevity of MHC alleles allows them to accumulate more mutations and therefore the proportion of differences within MHC genes is one order of magnitude greater than for the nuclear genes (Meyer and Thomson, 2001). MHC alleles from different species may be found more closely related than distinct alleles from a single species, a situation which is called trans-species polymorphism. Its effect is usually demonstrated on a phylogenetic tree of MHC sequences derived from multiple species. Under trans-species polymorphism, phylogeny of MHC alleles is expected not to follow the taxonomic patterns and sequences do not cluster by species.

MHC and disease

As the MHC molecules have an exceptional role in the discrimination between self and non-self and in triggering of immune response against pathogens, it is not surprising that variation within these loci influence the intensity and efficiency of the host immune response against invading pathogens on one hand and the predisposition to autoimmune and other types of disorders on the other. Above, I have described how pathogens can drive and maintain MHC polymorphism and how different MHC alleles might provide better or worse protection against particular pathogen based on their capability to bind peptides derived from this pathogen. In accordance with this assumption, there are numerous examples of associations of MHC alleles/haplotypes with susceptibility/resistance to infectious disease. Among the first well-supported finding belongs the association between specific chicken MHC haplotypes and resistance to Rous sarcoma virus and Marek’s disease virus (Briles et al., 1977) or associations of individual MHC alleles and different outcome of infectious diseases in inbred strains of mice (Monteyne et al., 1997; Zinkernagel et al., 1985). These early studies in birds and mice were possible mainly due to lower complexity of their model MHCs and allowed the identification of candidate genes that may be applicable to more genetically diverse human population. The human MHC, for historical reasons often called human leucocyte antigen (HLA), is the best characterized MHC. However, it was difficult to provide a clear evidence for associations between HLA genes and infectious disease. Associations found between HLA and infection mostly relate to susceptibility rather than to protection (Jeffery and Bangham, 2000). A protective effect of particular HLA alleles was detected in hepatitis B and C, HTLV and HIV infections (Chapman and 8

Hill, 2012; Migueles et al., 2000; Nikolich-Žugich et al., 2004; Thursz et al., 1999, 1995). With the growing interest about the MHCs in non-model species, the numbers of reports on MHC associations with resistance/susceptibility to infection in these species is increasing. The wide range of species analyzed involves song birds (Bonneaud et al., 2006; Westerdahl et al., 2005), fish (Dionne et al., 2009; Lohm et al., 2002), voles (Deter et al., 2008), wild boar (Acevedo-Whitehouse et al., 2005), cattle (Juliarena et al., 2008; Lewin et al., 1988) and horse (Hořín et al., 2004).

MHC molecules do not bind only molecules derived from pathogens, but also self-derived peptides from the intracellular environment and have therefore an important role in the maintenance of self-tolerance. MHC molecules are also involved in the control of the development of T lymphocytes in the thymus and thus influence the specificity of the T-cell receptor repertoire. The MHC variation can therefore be associated with various autoimmune disorders.

Many diseases show associations with multiple MHC alleles/haplotypes, possibly due to high levels of linkage disequilibrium within the MHC region and to effects of non-MHC genes located within this region. The mechanisms underlying such associations remain unknown. Besides a direct effect on antigen presentation and resulting immune responses (Höglund et al., 1999) or a receptor-mediated effects (Sasaki et al. 2011), epistatic effects of aminopeptidases involved in the generation of peptide ligands for MHC class I molecules (Sollid et al., 2014), and genetic variants modulating MHC class II gene expression (Handunnetthi et al., 2010) represent new findings in this area of MHC research.

The family Equidae as model for immunogenetic studies

The family Equidae consists of a single genus Equus, comprising the African wild asses, E. africanus africanus and E. africanus somaliensis, the Domestic donkey, E. asinus asinus, the Asian wild asses, E. kiang and E. hemionus (with two subspecies, E. hemionus kulan and E. hemionus onager), three species of zebra, E. quagga (with subspecies, E. quagga burchellii, E. quagga boehmi, E. quagga chapmani and E. quagga borensis) E. grevyi and E. zebra (with two subspecies, E. zebra zebra and E. zebra hartmannae), the domestic horse, E. caballus, and the wild horse, E. przewalskii.

Phylogenetic relationship among equid species remained for a long time unclear (Krüger et al., 2005; Vilstrup et al., 2013) and were fully resolved only recently using whole genome sequencing data (Jónsson et al., 2014) (Figure 4). Equids are supposed to evolve rapidly at the karyotype level and they differ remarkably in the diploid number, which range from 2n = 32 in Hartmann’s mountain zebra (Equus zebra hartmannae) to 2n = 66 in Przewalski’s horse (E. przewalskii) (Yang et al., 2003). Despite this extreme divergence in karyotypes, Jónsson et al. (2014) have identified multiple examples of hybridizations throughout the equine tree.

Figure 4. Equine phylogeny. According to Jónsson et al. (2014).

The family includes domestic, captive and free-ranging species occupying a wide range of habitats, differing in climatic conditions and pathogen pressure. Equid populations are subject to different levels of general and health care. This results into various types of adaptation to specific environmental challenges and to specific 10 pathogens. Interactions between equids and their pathogens are expected to leave signatures in the genomes of both organisms involved (Trowsdale and Parham, 2004). This makes the family Equidae suitable model for studying diversity, selection and evolution of the immunity related genes, of which the MHC genes are of undisputed importance.

The equine MHC

The major histocompatibility complex of the horse, Eqca, also called ELA for equine leucocyte antigen, is located on the long arm of chromosome 20 (Ansari et al., 1988; Mäkinen et al., 1989). ELA comprises approximately 4 Mb and appears to be similar to HLA in size, genetic content and organization with adjacent class I, III, and II regions (Gustafson et al., 2003). The extend of ELA class I and class II polymorphism was initially demonstrated by serology (Lazary et al., 1988). Existence of class II genes; DRA, DRB, DQA, DQB and DPB, was proven by restriction fragment length polymorphism (RFLP) and southern blot, using human probes for MHC sequences detection (Guerin et al., 1987; Hänni et al., 1988). During 90’s, the extent of the ELA class II polymorphism was studied mainly by using polymerase chain reaction (PCR) followed by single strand conformation polymorphism (SSCP), molecular cloning and DNA sequencing (Albright et al., 1991; Albright-Fraser et al., 1996; Fraser and Bailey, 1998, 1996; Gustafsson and Andersson, 1994; Szalai et al., 1993).

Class I ELA genes

Seven expressed ELA class I genes, accompanied by eight pseudogenes was identified in the horse (Tallmadge et al., 2005). Total number of expressed MHC class I genes was shown to differ among different ELA haplotypes (Tallmadge et al., 2010).

Class II ELA genes

One DRA locus was identified in the horse (Albright et al., 1991; Gustafson et al., 2003), which is in agreement with other mammalian MHC. However, this locus showed higher level of polymorphism in equids (Albright-Fraser et al., 1996; Brown et al., 2004; Díaz et al., 2008; Kamath and Getz, 2011) than in most other taxa with little or no DRA sequence variation (Chardon et al., 1999; Takada et al., 1998; Wagner et al., 1999; Yuhki et al., 2007). Signatures of positive selection identified in this locus suggest a functional importance of the polymorphism observed (Janova et al., 2009; Kamath and Getz, 2011). On the other hand, a relatively high but still limited level of 11 polymorphism within a single locus makes the DRA gene a suitable candidate for population diversity and associations studies.

The DQA gene polymorphism and existence of multiple loci was first demonstrated by Fraser and Bailey (1998). One DQA homolog was localized to chromosome 5 (Bailey et al., 2000), i.e. to a different chromosome than the rest of ELA genes. Later, existence of at least two DQA homologues within the MHC region was proven (Gustafson et al., 2003) and the current in-silico annotation of the horse genome assembly suggested the presence of three DQA genes on the horse chromosome 20. Effects of positive selection on the DQA genes were observed (Janova et al., 2009; Kamath and Getz, 2011).

Three DRB genes were reported in the horse (Fraser and Bailey, 1996; Gustafson et al., 2003). Their diversity was characterized in the domestic horse (Díaz et al., 2001) and in Przewalski’s horses (Hedrick et al., 1999).

The DQB gene was first amplified and sequenced in homozygous horses (Szalai et al., 1994). In this study, no evidence for multiple DQB loci was found. Later, my colleagues have provided evidence for existence of a second locus (Horín and Matiasovic, 2002), and two DQB loci were confirmed later on in a BAC contig map of the equine MHC (Gustafson et al., 2003). After the horse genome sequence became available (Wade et al., 2009), different annotations in different genome browsers disagreed on both the number and localization of DQB genes. Moreover, a previously described DQB gene fragment (Mashima, 2003) could not be annotated in the currently available full genome sequence assembly EquCab2.0, suggesting a complex organization of the horse DQB region. It is clear, however, that DQB genes are highly polymorphic, representing a substantial part of the horse MHC class II diversity (Horín and Matiasovic, 2002).

In methodological terms, the primers used for amplifying MHC class II genes were usually located within the exon 2 sequence of the particular gene. Therefore, they were usually not locus-specific, although they could capture a significant proportion of the existing variation.

Although the MHC of domestic horses has been studied from different perspectives, including its structure (Gustafson et al., 2003; Gustafsson and Andersson, 1994), expression and immune functions (Rappocciolo et al., 2003), associations with diseases and other physiological processes (Andersson et al., 2012; Eder et al., 2001; Lazary et al., 1994), relatively little attention has been paid so far to comparative genomics of the MHC of the entire family Equidae and to its evolution. The genetic

12 diversity of the DRA locus was characterized in populations of domestic donkeys (Arbanasić et al., 2013), and host and pathogen interactions were studied in free- ranging populations of plain zebras (Kamath et al., 2014). Therefore, the general objective of this thesis was to contribute to a better characterization and understanding of the diversity and evolution of the major histocompatibility complex in the Equidae, an important and model mammalian family.

SPECIFIC OBJECTIVES OF THE DISSERTATION

1. Analysis of the genomic structure of the MHC class II DQB region in the family Equidae.

2. Studies of molecular genetic polymorphism of MHC class II genes in equids.

3. Analysis of molecular evolution and selection of MHC class II genes in equids.

4. Studies of genetic diversity and associations with disease in model populations.

MATERIALS AND METHODS

Animals

Individual animals/populations used for the analyses are described in details in the particular manuscripts. Extracted genomic DNA was used for all analyses. Individual DRB and DQB loci were studied in the entire family Equidae. Association and population diversity studies were performed in model horse and donkey populations.

Equids: the basic panel

The panel of equids consisted of domestic and wild horses, Equus caballus and Equus przewalskii, donkeys, Equus asinus asinus and Equus africanus somaliensis, Asian wild asses, Equus kiang and Equus hemionus kulan, Equus hemionus onager and three species of zebra, Equus grevyi, Equus zebra hartmannae and Equus quagga represented by four sub-species, Equus quagga burchellii, Equus quagga boehmi, Equus quagga chapmanni, Equus quagga borensis. At least two individuals of each species/sub-species were studied.

Donkey populations

Donkeys from 3 different continents sharing a common pathogen, tick- transmitted piroplasmids were analyzed (Manuscript II). The Martina Franca breed is an ancient native breed of Apulia (southern Italy), which was in the past mainly used for production of mules and as a light draught animal. Jordanian donkeys came from several rural localities in western Jordan, characterized by hot semi-arid and arid climate. Jordanian donkeys belong to local breed as well as African donkeys owned by semi-nomadic pastoralists of Turkana and Samburu tribes living in an arid environment in Northern Kenya.

Horse populations

The Icelandic horse, the only horse breed to be found on Iceland, has evolved isolated for more than thousand years. The ancestors of Icelandic horse were probably brought to Iceland by Viking age Scandinavians and later crossed with horses from Ireland and Scotland. The breed underwent strong natural selection due to cold climate and poor pastures – on the other hand there is virtually no selective pressure of

15 parasites, which are quite rare in Iceland. Horses born in Iceland and imported to Europe as adults served as a model population for studying insect bit hypersensitivity.

The Old Kladruby horses are an autochthonous Czech, unique and endangered horse breed of high historical and cultural value, which can be considered as a model population in terms of conservation issues. The population underwent historical bottlenecks and intensive inbreeding and also nowadays, the breeding nucleus is rather small. In the Kladruby horses, the grey and black coat color varieties exist as two sub- populations with different recent breeding history. In Manuscript III, we describe their genetic diversity and in Manuscripts IV and VI, MHC associations with insect bite hypersensitivity and melanoma were analyzed.

Methods

The methods are described in details in the Materials and Methods sections of the particular manuscripts. The laboratory work performed by myself consisted of genomic DNA extraction, primer design, PCR, gel electrophoresis, PCR-RFLP and Sanger and next generation sequencing. For Sanger sequencing and one part of the NGS, commercial services of Macrogen, Inc. (Seoul, South Korea) and of the Eurofins MWG operon (Ebersberg, Germany) were used. I also have done most of the subsequent bioinformatic analysis, including molecular evolutionary and selection analyses, and the intra- and inter-population genetics analyses. My specific contribution to all manuscripts is characterized below.

RESULTS

This thesis consists of five published papers and one manuscript in preparation. The relative contribution of the author of the thesis (maiden name Vranová) is given bellow and approved by given co-authors. In this section, a synopsis of the results obtained is provided. The manuscripts attached bringing complete relevant information represent an inherent part of the thesis.

List of attached manuscripts and the contribution of the author

MS I Klumplerova, M., Splichalova, P., Baumeisterova Kohutova, A., Oppelt, J., Orlando, L., Horin, P. Genetic diversity, evolution and selection in the major histocompatibility complex DRB and DQB genes in the family Equidae. In preparation, to be submitted by October 2015.

Contribution:

I have performed all the laboratory work except cloning and amplification of TNFA. I performed all molecular evolutionary and phylogeny analyses and I wrote the manuscript.

MS II Vranova, M., Alloggio, I., Qablan, M., Vyskocil, M., Baumeisterova, A., Sloboda, M., Putnova, L., Vrtkova, I., Modry, D., Horin, P., 2011. Genetic diversity of the class II major histocompatibility DRA locus in European, Asiatic and African domestic donkeys. Infection, genetics and evolution. 11, 1136–1141.

Contribution:

I have performed the laboratory work concerning genotyping and assessment of genetic markers of the donkey. I have performed all population genetics analyses except the statistical association analyses. I wrote the majority of the manuscript.

MS III Janova, E., Futas, J., Klumplerova, M., Putnova, L., Vrtkova, I., Vyskocil, M., Frolkova, P., Horin, P., 2013. Genetic diversity and conservation in a small endangered horse population. Journal of applied genetics. 54, 285–292.

Contribution:

I have performed laboratory work concerning the DRA genotyping.

MS IV Vychodilova, L., Matiasovic, J., Bobrova, O., Futas, J., Klumplerova, M., Stejskalova, K., Cvanova, M., Janova, E., Osickova, J., Vyskocil, M., Sedlinska, M., Dusek, L., Marti, E., Horin, P., 2013. Immunogenomic analysis of insect bite hypersensitivity in a model horse population. Veterinary immunology and immunopathology. 152, 260–268.

Contribution:

I have performed laboratory work concerning the DRA genotyping and I took part in writing of the manuscript.

MS V Klumplerova, M., Vychodilova, L., Bobrova, O., Cvanova, M., Futas, J., Janova, E., Vyskocil, M., Vrtkova, I., Putnova, L., Dusek, L., Marti, E., Horin, P., 2013. Major histocompatibility complex and other allergy-related candidate genes associated with insect bite hypersensitivity in Icelandic horses. Molecular biology reports. 40, 3333– 3340.

Contribution:

I have performed the laboratory work concerning the DRA genotyping and MHC markers assessment and I took part in writing of the manuscript.

MS VI Futas, J., Vychodilova, L., Hofmanova, B., Vranova, M., Putnova, L., Muzik, J., Vyskocil, M., Vrtkova, I., Dusek, L., Majzlik, I., Horin, P., 2012. Genomic analysis of resistance/susceptibility to melanoma in Old Kladruber greying horses. Tissue Antigens. 79, 247–248.

Contribution:

I have performed laboratory work concerning the DRA genotyping. 18

Synopsis of the results obtained

MS I Genetic diversity, evolution and selection in the major histocompatibility complex DRB and DQB genes in the family Equidae.

Klumplerova, M., Splichalova, P., Baumeisterova Kohutova, A., Oppelt, J., Orlando, L., Horin, P.

Manuscript in preparation, to be submitted by October 2015.

DRB and DQB genes belong to the most diverse genes of the class II region among majority of mammalian species. They show numerous associations with susceptibility to infectious and other types of diseases. Equine DRB and DQB genes have so far been only poorly characterized. No information about the molecular evolution and selection acting on these loci is known. Therefore, the focus of this study was to characterize the DRB and DQB loci in the equid species.

Three DRB and four DQB loci with one DQB pseudogene were identified. The genomic organization of DQB region was shown to be complex and probably not completely resolved. Two of DQB genes were shown to exhibit individual copy number variation in the horse and other equids. Positive selection was detected in all loci, except the DQB3 locus. Individual loci differed intensity of selection and level of trans- species sharing, suggesting their different evolution and different relative contribution to adaptive immune responses.

The phylogenetic trees constructed revealed the presence of trans-species polymorphism among the class II loci, suggesting an important effect of balancing selection. This assumption was further supported by a different pattern of the phylogenetic tree of the class III TNFA gene, located within the MHC region and closely linked to the class II cluster. The immune functions of the tumor necrosis factor alpha are not related to antigen presentation and only traces of positive selection were identified within the coding part of the gene. This tree copied, although not entirely, the standard taxonomic classification. To which extent is for the distorted topology responsible the positive selection or linkage disequilibrium with antigen presenting genes is not known.

MS II Genetic diversity of the class II major histocompatibility DRA locus in European, Asiatic and African domestic donkeys.

Vranova, M., Alloggio, I., Qablan, M., Vyskocil, M., Baumeisterova, A., Sloboda, M., Putnova, L., Vrtkova, I., Modry, D., Horin, P. (2011)

Published in: Infection Genetics and Evolution. 11, 1136–1141.

Impact factor (2011): 3.128

Citations (June 2015): 4

The MHC class II DRA gene in equids shows extraordinary degree of polymorphism compared to the majority of mammalian taxa where the DRA has limited or no variation. In the horse, 5 DRA alleles and in the donkey 7 DRA alleles were described up today. In this paper, we describe a feasible method for the DRA gene exon 2 genotyping in the donkey. This method was used for the assessment of genetic diversity of the DRA gene in three populations of donkeys from three different continents, sharing common protozoan parasites, Theileria equi and Babesia caballi.

Neutral evolutionary processes such as migration, genetic drift and fluctuating population size can affect both neutral and functional loci, while adaptive processes are assumed to occur in functionally important sequences. Therefore, genetic diversity of the class II DRA gene was compared with diversity in 17 potentially neutral microsatellite loci. Although balancing selection is expected to leave detectable effects on allele distribution at loci under selective pressure, no significant departure from Hardy-Weinberg expectations or contrasting population structures at MHC and neutral loci were found. Statistical associations of particular DRA alleles and genotypes with the presence of protozoan parasites were only marginal suggesting that the extent of genetic diversity observed in the three populations reflected effects of selective breeding and different genetic origins of analyzed donkeys rather than pathogen- mediated selection.

MS III Genetic diversity and conservation in a small endangered horse population.

Janova, E., Futas, J., Klumplerova, M., Putnova, L., Vrtkova, I., Vyskocil, M., Frolkova, P., Horin, P. (2013)

Published in: Journal of applied genetics. 54, 285–292.

Impact factor (2013): 1.902

Citations (June 2015): 0

The Old Kladruby horse arose in the 17th century as an autochthonous Czech breed used for ceremonial purposes. The population underwent historical bottlenecks and intensive inbreeding and became specific and isolated from related breeds. Nowadays, it can be considered as a model population in terms of conservation issues. In the Kladruby horses, the grey and black coat color varieties exist as two sub- populations with different recent breeding history. Here, we report on the genetic diversity of these two sub-populations in both, neutral and functionally important loci performed thirteen years after a similar study (Horín et al., 1998). The comparison between 1997 and 2010 did not show differences in the extent of genetic diversity and the Old Kladruby horses were shown to be comparable to other horse breeds, despite its small population size, historical bottlenecks and inbreeding. These two analyses provided a framework for further studies of associations between genetic diversity and particular diseases occurring in this model population.

MS IV Immunogenomic analysis of insect bite hypersensitivity in a model horse population.

Vychodilova, L., Matiasovic, J., Bobrova, O., Futas, J., Klumplerova, M., Stejskalova, K., Cvanova, M., Janova, E., Osickova, J., Vyskocil, M., Sedlinska, M., Dusek, L., Marti, E., Horin, P. (2013)

Published in: Veterinary immunology and immunopathology. 152, 260–268.

Impact factor (2013): 1.748

Citations (June 2015): 5

Here, the Old Kladruby horses served as a model population for association analysis between the MHC region and other candidate allergy-related genes with insect bite hypersensitivity. Forty-six SNPs located in 29 candidate genes were examined together with exon 2 alleles of the ELA class II DRA and DQA genes. Besides the genomic analysis, a long-term follow-up of the population allowed also skin biopsies and analysis of gene expression. Different genes were found to be associated with different clinical manifestations of the disease. Two associated genes also showed differences in the mRNA expression in skin biopsies. This is the first report on non- MHC genes associated with IBH in horses. Selected candidate genes later served as a candidate gene panel in the MS V.

MS V Major histocompatibility complex and other allergy-related candidate genes associated with insect bite hypersensitivity in Icelandic horses.

Klumplerova, M., Vychodilova, L., Bobrova, O., Cvanova, M., Futas, J., Janova, E., Vyskocil, M., Vrtkova, I., Putnova, L., Dusek, L., Marti, E., Horin, P. (2013)

Published in: Molecular biology reports. 40, 3333–3340.

Impact factor (2013): 1.958

Citations (June 2015): 2

Insect bite hypersensitivity (IBH) is an important disease of the horse caused by bites of insects. The mechanisms resulting in susceptibility or resistance to IBH still remain unclear. It was shown that IBH is the result of a complex interplay of environmental and genetic factors. Icelandic horses living in their original habitat, do not suffer from IBH due to absence of responsible insect (Culicoides, Simulium) on Iceland. When Icelandic horses are imported to continental Europe and get exposed to Culicoides, they develop IBH with a much higher prevalence than horses of this breed born in the same environment. In this study, the candidate gene approach was used for analysis of genetic factors contribution to development of the disease. We have analyzed the MHC region and 20 single nucleotide polymorphisms (SNPs) in 17 allergy- related genes. The MHC region was examined by genotyping five microsatellites spanning the MHC region (COR112, COR113, COR114, UM011 and UMN-JH34-2) and exon 2 polymorphism of class II DRA gene, using a newly developed typing method. Horses were classified as IBH-affected and non-affected based on clinical signs and on the results of an in vitro sulfidoleukotriene-release assay. Associations with Eqca-DRA and MHC-linked microsatellite COR113 were identified. Associations of SNPs in allergy-related genes was found only in combined genotypes for genes coding for the CD14 receptor, interleukin 23 receptor, thymic stromal lymphopoietin and transforming growth factor beta 3 molecules. This study confirmed previous findings about the role of the MHC in the horse IBH (Andersson et al., 2012). Our specific contribution was based on the fact that Icelandic horses born in Iceland and not in Europe were used. The results showed that even in this generation of Icelandic horses, the MHC region contributes to the susceptibility to IBH.

MS VI Genomic analysis of resistance/susceptibility to melanoma in Old Kladruber greying horses.

Futas, J., Vychodilova, L., Hofmanova, B., Vranova, M., Putnova, L., Muzik, J., Vyskocil, M., Vrtkova, I., Dusek, L., Majzlik, I., Horin, P. (2012)

Published in: Tissue Antigens. 79, 247–248.

Impact factor (2012): 2.753

Citations (June 2015): 2

Melanoma is a common disease occurring in many horse breeds with higher incidence in the Grey horses. A mutation within the syntaxin 17 gene responsible for greying in horses was shown to be associated with occurrence of melanoma but the association observed did not explain completely the variation observed. Here, we aimed to further characterize genes involved in the melanoma development. In a model population of Old Kladruby horses, we have analyzed SNPs in immunity-related genes and the MHC class II DRA variation. We have confirmed strong association with syntaxin 17 and further identified other genes; TLR4, IL12A to be associated with presence of melanoma. Minor effect of the MHC class II DRA was found only in association of composed genotypes with presence of melanoma.

DISCUSSION

The major histocompatibility complex is a multigene family crucial for the adaptive immune response of all jawed vertebrates studied so far (Klein, 1986). Variation at the MHC genes affect many important biological traits, including immune recognition, susceptibility to infectious and other types of diseases, mate choice, maternal-fetal interaction and pregnancy outcome (Sommer, 2005). MHC has become a paradigm for how pathogens are shaping the patterns of adaptive genetic variation. It is used, as one of the most favored molecular marker, in many different research fields, ranging from human medicine, molecular evolutionary studies to conservation genetics (Sommer et al., 2013). Most of the current knowledge on MHC genes has been obtained with experimental animal and human models. However, the number of studies on MHC in free-ranging and non-model species is increasing, providing an important support for the previously obtained data and helping to understand evolutionary processes that shape the structure and function of major histocompatibility complexes in different species.

The equine MHC was shown to be similar to the human HLA in its size, genetic content and organization (Gustafson et al., 2003). On the other hand, it exhibits several interesting and extraordinary features compared to a model mammalian MHC, such as polymorphism of the DRA locus (Albright-Fraser et al., 1996) and/or an extra-MHC DQA homologue (Fraser and Bailey, 1998). The genomic structure of the equine MHC has not been not fully resolved yet, especially in terms of the number and structure of class I loci (Tallmadge et al., 2010), and of the genomic structure of the DQB region. In the first manuscript presented, we attempted to identify the missing pieces in the overall picture of the equine MHC class II region. By using locus-specific primers, we have identified individual DRB and DQB genes in the entire family Equidae, their molecular variation, evolution and signatures of selection. We have shown that the equine DQB genes do not follow the pattern found in humans, swine and sheep, where one or two copies of the gene are present (Chardon et al., 1999; Doherty et al., 1992; Dukkipati et al., 2006). We have identified at least four DQB genes and one DQB pseudogene and found evidence for individual copy number variation in two of them. A similar pattern of variation in the number and structure of DQ genes was observed in cattle (Ellis and Ballingall, 1999) where the existence up to five BoLA-DQB genes was postulated (He et al., 2014). The analysis of recently published individual genomes of

Mongolian horse Ajinai and Przewalski’s horse Burgud (Huang et al., 2014) showed that the definitive number of the horse DQB genes still remains to be determined as well. These individual genomes were not aligned to the reference assembly EquCab2.0. They thus represent a promising tool for resolving problems resulting from the fact that the reference genome was derived from a horse previously selected for MHC homozygosity (Antczak, 2012). As these individual genomes are so far available only at the scaffold level, the possibilities of annotating individual loci and of identifying potential new loci remain limited.

All loci except the DQB3 locus were proven to be subject to positive selection. Both beta-genes analyzed (DRB, DQB) were shown to be subject to positive selection, which was stronger than for the alpha-genes (DRA, DQA) (Janova et al., 2009; Kamath and Getz, 2011). Similarly, stronger selective pressure among beta-genes was described in grey wolves (H. Arbanasić et al., 2013) and the giant panda (Chen et al., 2010). The typical distribution of alleles in the constructed phylogenetic trees and the presence of trans-species polymorphism were in agreement with putative effects of balancing selection on the class II loci. In agreement with this assumption, the phylogenetic tree of TNFA, a gene distant approximately 1.5 Mb from the class II genes analyzed and with a different role in the host and pathogen interactions, showed completely different phylogenetic patterns. Three DRB and three DQB loci (DQB4 was not analyzed) were shown to be transcribed (Don Miller, personal communication). Therefore, we can assume that all the class II genes analyzed, including those previously described are of functional importance contributing to adaptive immune responses in the family Equidae.

The DRA locus of equids has been analyzed previously in different laboratories (Albright-Fraser et al., 1996; Brown et al., 2004; Díaz et al., 2008; Janova et al., 2009; Kamath and Getz, 2011). As it is more polymorphic than DRA loci in most other mammals (Yuhki et al., 2003), and as its polymorphism could be of functional importance (Janova et al., 2009), it was be used for population analyses. For these purposes, we developed PCR-RFLP-based genotyping methods for horses and donkeys. They showed to be feasible and reliable for population analyses. In the second manuscript, we have used it for analyzing the genetic diversity of three donkey populations from three different continents sharing a common parasite. Along with the results of neutrality tests for the DRA locus, no effects of selection could be identified at the population level. This is in agreement with the assumptions that it is more difficult to reveal effects of selection at this level as compared to neutrality tests at the molecular level (Spurgin and Richardson, 2010), suggesting either that selection on the MHC is not homogenous or that the contemporary neutrality tests do not have sufficient power 26

(Garrigan and Hedrick, 2003). Only marginal associations between Eqas DRA and infection were observed, even though associations between protozoan infections and MHC were repeatedly reported in humans and domestic animals (Gilbert et al., 1998; Gray and Gill, 1993; Stear and Wakelin, 1998). It seems that our results reflected strong effects of selective breeding and different genetic origins of the populations studied rather than subtle effects of pathogen-driven selection.

Associations between the MHC complex variation and large amount of allergic (hypersensitivity) disorders are well established. Genome-wide association studies have demonstrated that the contribution of non-MHC loci is not always comparable to the effect of the MHC region (Sollid et al., 2014). The most common allergies associated with MHC in humans are asthma with COPD (Lara-Marquez et al., 1999; Newby et al., 2011) and atopic dermatitis (Hirota et al., 2012; Lee et al., 2001). Similar clinical conditions are also an important problem in horses. The recurrent airway obstruction (RAO) represent an equine homologue of human asthma and the insect bite hypersensitivity (IBH) manifested by atopic dermatitis resembles human skin allergic reactions (Kehrli et al., 2015). In both conditions, effects of IR genes and of the MHC region were analyzed in both species (Andersson et al., 2012; Kehrli et al., 2015; March et al., 2015; Tantisira et al., 2004). Our contribution was to analyze these associations in two genetically distinct model horse populations.

The Old Kladruby horse breed is a genetically homogenous population, whose molecular diversity including MHC genes was characterized in time (Horín et al., 1998; paper III of this thesis). We have shown that over years, the breed has been subdivided into two genetically distinct sub-populations represented by the two color varieties. The results showed no loss of both neutral and functional genetic diversity in time. However, our unpublished data obtained later on demonstrated reduced numbers of alleles and decreased observed heterozygosity in the DRA gene as compared to other seven horse breeds. In such a well-defined model, a long term (10-year) association study could be performed. While other analyses of associations between IBH and genetic polymorphisms were based on short-term observations and/or even on a single assessment of the clinical disease, our specific approach was based on a 10 year follow- up of the Old Kladruby population, where the IBH phenotyping was based on the recurrence of the disease in time. Despite this, the effect of the MHC region was associated with total IgE levels but not with clinical disease although it has been reported by other studies (Andersson et al., 2012). This may be explained in several ways, the major reasons were probably a combination of weak MHC effects with low numbers of the horses in this population and its genetic stratification. Observations on

27 associations with the IgE markers are in agreement with the fact that IBH is an IgE mediated disease.

The population of Icelandic horses studied by us (MS V) was another model. While other studies analyzed Icelandic horses born and raised in Europe, we have analyzed horses born in Iceland and living in Europe, which are, in terms of IBH, two different situations. In agreement with other studies, the effect of the MHC variation was observed in this model situation as well. Like in the Old Kladruby horses, our contribution consisted also in the identification of non-MHC genes associated with different manifestation of IBH in horses.

Only a minor effect of MHC class II DRA variation was demonstrated in melanoma. Since associations between MHC and melanoma were repeatedly reported in humans and several animal species (Fensterle et al., 2006; Ho et al., 2010; Nagore et al., 2002; Planelles et al., 2006), the explanations of our findings are probably the same like for the IBH. While genes with strong contributions to the overall variation were found to be associated with disease (i.e. IBH or melanoma) associations of other genes with weaker effects passed undetected. This assumption is based on the association between the STX17 duplication and melanoma, which confirmed the original observation of (Rosengren Pielberg et al., 2008) and based on our recent results, when we tested distinct population by using a larger set of MHC-linked microsatellites. With higher numbers of multi-allelic markers, associations with the MHC region could be detected (Futas et al., unpublished results).

For the same reason probably, no associations between the ELA-DRA variation and post-vaccination antibody levels were found in another study that I co-authored (Rusek et al., 2013), although the MHC region was repeatedly reported to be associated with post-vaccination immune-responses in humans and other species (Milich and Leroux-Roels, 2003; Ovsyannikova et al., 2004).

GENERAL CONCLUSIONS

We have analyzed the DRB and DQB genes in all available equids. The genomic organization of DQB gene was shown to be complex and probably, due to limitation in the current reference genome, not completely resolved. Two DQB loci were demonstrated to exhibit inter-individual copy number variation in the horse and other two equid species. The polymorphism, selection and molecular evolution were examined within individual loci of these genes. Analyzed loci differed in level of trans- species polymorphism and intensity of selection, suggesting their different evolutionary history and functionally importance. The MHC class II DRA gene was found suitable for association studies and for the assessment of functional genetic diversity. Associations of the MHC region with infection and insect bite hypersensitivity were found. The data obtained provided us with information on basic evolutionary relationships of MHC genes within the family Equidae and on some signatures of adaptation within this genomic region.

REFERENCES

Acevedo-Whitehouse, K., Vicente, J., Gortazar, C., Höfle, U., Fernández-De-Mera, I.G., Amos, W., 2005. Genetic resistance to bovine tuberculosis in the Iberian wild boar. Mol. Ecol. 14, 3209–3217. doi:10.1111/j.1365-294X.2005.02656.x

Albright, D., Bailey, E., Woodward, J.G., 1991. Nucleotide sequence of a cDNA clone of the horse (Equus caballus) DRA gene. Immunogenetics 34, 136–138.

Albright-Fraser, D.G., Reid, R., Gerber, V., Bailey, E., 1996. Polymorphism of DRA among equids. Immunogenetics 43, 315–317.

Andersson, L.S., Swinburne, J.E., Meadows, J.R.S., Broström, H., Eriksson, S., Fikse, W.F., Frey, R., Sundquist, M., Tseng, C.T., Mikko, S., Lindgren, G., 2012. The same ELA class II risk factors confer equine insect bite hypersensitivity in two distinct populations. Immunogenetics 64, 201–208. doi:10.1007/s00251-011-0573-1

Ansari, H.A., Hediger, R., Fries, R., Stranzinger, G., 1988. Chromosomal localization of the major histocompatibility complex of the horse (ELA) by in situ hybridization. Immunogenetics 28, 362–364.

Antczak, D.F., 2012. A life with horses: it’s been a great ride! Vet. Immunol. Immunopathol. 148, 6–11. doi:10.1016/j.vetimm.2012.05.015

Arbanasić, H., Galov, A., Ambriović-Ristov, A., Grizelj, J., Arsenos, G., Marković, B., Dovenski, T., Vince, S., Curik, I., 2013. Extensive polymorphism of the major histocompatibility complex DRA gene in Balkan donkeys: perspectives on selection and genealogy. Anim. Genet. 44, 711–716. doi:10.1111/age.12054

Arbanasić, H., Huber, Đ., Kusak, J., Gomerčić, T., Hrenović, J., Galov, A., 2013. Extensive polymorphism and evidence of selection pressure on major histocompatibility complex DLA-DRB1, DQA1 and DQB1 class II genes in Croatian grey wolves. Tissue Antigens 81, 19–27. doi:10.1111/tan.12029

Bailey, E., Marti, E., Fraser, D.G., Antczak, D.F., Lazary, S., 2000. Immunogenetics of the horse., in: Bowling, A.T., Ruvinsky, A. (Eds.), The Genetics of the Horse. CABI, Wallingford, pp. 123–155.

Biswas, S., Akey, J.M., 2006. Genomic insights into positive selection. Trends Genet. 22, 437–446. doi:10.1016/j.tig.2006.06.005

Bonneaud, C., Pérez-Tris, J., Federici, P., Chastel, O., Sorci, G., 2006. Major Histocompatibility Alleles Associated with Local Resistance to Malaria in a Passerine. Evolution 60, 383–389. doi:10.1111/j.0014-3820.2006.tb01114.x

Borghans, J.A.M., Beltman, J.B., Boer, R.J.D., 2004. MHC polymorphism under host- pathogen coevolution. Immunogenetics 55, 732–739. doi:10.1007/s00251-003-0630-5

Briles, W.E., Stone, H.A., Cole, R.K., 1977. Marek’s disease: effects of B histocompatibility alloalleles in resistant and susceptible chicken lines. Science 195, 193–195.

Brown, J.H., Jardetzky, T.S., Gorga, J.C., Stern, L.J., Urban, R.G., Strominger, J.L., Wiley, D.C., 1993. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 364, 33–39. doi:10.1038/364033a0 30

Brown, J.J., Thomson, W., Clegg, P., Eyre, S., Kennedy, L.J., Matthews, J., Carter, S., Ollier, W.E.R., 2004. Polymorphisms of the equine major histocompatibility complex class II DRA locus. Tissue Antigens 64, 173–179. doi:10.1111/j.1399-0039.2004.00269.x

Chapman, S.J., Hill, A.V.S., 2012. Human genetic susceptibility to infectious disease. Nat. Rev. Genet. 13, 175–188. doi:10.1038/nrg3114

Chardon, P., Renard, C., Vaiman, M., 1999. The major histocompatibility complex in swine. Immunol. Rev. 167, 179–192. doi:10.1111/j.1600-065X.1999.tb01391.x

Chen, Y.-Y., Zhang, Y.-Y., Zhang, H.-M., Ge, Y.-F., Wan, Q.-H., Fang, S.-G., 2010. Natural selection coupled with intragenic recombination shapes diversity patterns in the major histocompatibility complex class II genes of the giant panda. J. Exp. Zoolog. B Mol. Dev. Evol. 314B, 208–223. doi:10.1002/jez.b.21327

Deter, J., Bryja, J., Chaval, Y., Galan, M., Henttonen, H., Laakkonen, J., Voutilainen, L., Vapalahti, O., Vaheri, A., Salvador, A.R., Morand, S., Cosson, J.-F., Charbonnel, N., 2008. Association between the DQA MHC class II gene and Puumala virus infection in Myodes glareolus, the bank vole. Infect. Genet. Evol., MEEGID VIII 2006 8, 450–458. doi:10.1016/j.meegid.2007.07.003

Díaz, S., Echeverría, M.G., It, V., Posik, D.M., Rogberg-Muñoz, A., Pena, N.L., Peral- García, P., Vega-Pla, J.L., Giovambattista, G., 2008. Development of an ELA-DRA gene typing method based on pyrosequencing technology. Tissue Antigens 72, 464–468. doi:10.1111/j.1399- 0039.2008.01113.x

Díaz, S., Giovambattista, G., Dulout, F.N., Peral-García, P., 2001. Genetic variation of the second exon of ELA-DRB genes in Argentine Creole horses. Anim. Genet. 32, 257–263.

Dionne, M., Miller, K.M., Dodson, J.J., Bernatchez, L., 2009. MHC standing genetic variation and pathogen resistance in wild Atlantic salmon. Philos. Trans. R. Soc. B Biol. Sci. 364, 1555–1565. doi:10.1098/rstb.2009.0011

Doherty, D.G., Vaughan, R.W., Donaldson, P.T., Mowat, A.P., 1992. HLA DQA, DQB, and DRB genotyping by oligonucleotide analysis: distribution of alleles and haplotypes in British caucasoids. Hum. Immunol. 34, 53–63. doi:10.1016/0198-8859(92)90085-2

Doherty, P.C., Zinkernagel, R.M., 1975. Enhanced immunological surveillance in mice heterozygous at the H-2 gene complex. Nature 256, 50–52.

Dukkipati, V.S.R., Blair, H.T., Garrick, D.J., Murray, A., 2006. “Ovar-Mhc” - ovine major histocompatibility complex: structure and gene polymorphisms. Genet. Mol. Res. GMR 5, 581–608.

Eder, C., Curik, I., Brem, G., Crameri, R., Bodo, I., Habe, F., Lazary, S., Sölkner, J., Marti, E., 2001. Influence of environmental and genetic factors on allergen-specific immunoglobulin-E levels in sera from Lipizzan horses. Equine Vet. J. 33, 714–720. doi:10.2746/042516401776249264

Ejsmond, M.J., Babik, W., Radwan, J., 2010. MHC allele frequency distributions under parasite-driven selection: A simulation model. BMC Evol. Biol. 10, 332. doi:10.1186/1471-2148- 10-332

Ellis, S.A., Ballingall, K.T., 1999. Cattle MHC: evolution in action? Immunol. Rev. 167, 159–168.

Fensterle, J., Trefzer, U., Berger, T., Andersen, M.H., Ugurel, S., Becker, J.C., 2006. HLA-B8 association with late-stage melanoma -- an immunological lesson? BMC Med. 4, 5–6. doi:10.1186/1741-7015-4-5 31

Flajnik, M.F., Kasahara, M., 2001. Comparative genomics of the MHC: glimpses into the evolution of the adaptive immune system. Immunity 15, 351–362.

Fraser, D.G., Bailey, E., 1998. Polymorphism and multiple loci for the horse DQA gene. Immunogenetics 47, 487–490.

Fraser, D.G., Bailey, E., 1996. Demonstration of three DRB loci in a domestic horse family. Immunogenetics 44, 441–445.

Garrigan, D., Hedrick, P.W., 2003. Perspective: Detecting Adaptive Molecular Polymorphism: Lessons from the Mhc. Evolution 57, 1707–1722. doi:10.1111/j.0014- 3820.2003.tb00580.x

Gilbert, S.C., Plebanski, M., Gupta, S., Morris, J., Cox, M., Aidoo, M., Kwiatkowski, D., Greenwood, B.M., Whittle, H.C., Hill, A.V.S., 1998. Association of Malaria Parasite Population Structure, HLA, and Immunological Antagonism. Science 279, 1173–1177. doi:10.1126/science.279.5354.1173

Gray, G.D., Gill, H.S., 1993. Host genes, parasites and parasitic infections. Int. J. Parasitol., Special Issue: Proceedings of the Joint Conference of the Australian and New Zealand Societies for Parasitology, 1992 23, 485–494. doi:10.1016/0020-7519(93)90037-Y

Guerin, G., Bertaud, M., Chardon, P., Geffrotin, C., Vaiman, M., Cohen, D., 1987. Molecular genetic analysis of the major histocompatibility complex in an ELA typed horse family. Anim. Genet. 18, 323–336.

Gustafson, A.L., Tallmadge, R.L., Ramlachan, N., Miller, D., Bird, H., Antczak, D.F., Raudsepp, T., Chowdhary, B.P., Skow, L.C., 2003. An ordered BAC contig map of the equine major histocompatibility complex. Cytogenet. Genome Res. 102, 189–195. doi:75747

Gustafsson, K., Andersson, L., 1994. Structure and polymorphism of horse MHC class II DRB genes: convergent evolution in the antigen binding site. Immunogenetics 39, 355–358.

Gu, X., Nei, M., 1999. Locus specificity of polymorphic alleles and evolution by a birth- and-death process in mammalian MHC genes. Mol. Biol. Evol. 16, 147–156.

Handunnetthi, L., Ramagopalan, S.V., Ebers, G.C., Knight, J.C., 2010. Regulation of major histocompatibility complex class II gene expression, genetic variation and disease. Genes Immun. 11, 99–112. doi:10.1038/gene.2009.83

Hänni, K., Hesford, F., Lazary, S., Gerber, H., 1988. Restriction fragment length polymorphisms of horse class II MHC genes observed using various human alpha- and beta- chain cDNA probes. Anim. Genet. 19, 395–408.

Hedrick, P.W., 2002. Pathogen resistance and genetic variation at MHC loci. Evol. Int. J. Org. Evol. 56, 1902–1908.

Hedrick, P.W., 1998. Balancing selection and MHC. Genetica 104, 207–214. doi:10.1023/A:1026494212540

Hedrick, P.W., Parker, K.M., Miller, E.L., Miller, P.S., 1999. Major histocompatibility complex variation in the endangered Przewalski’s horse. Genetics 152, 1701–1710.

He, Y., Xi, D., Leng, J., Qian, T., Jin, D., Chen, T., Yang, C., Hao, T., Yang, Z., Deng, W., 2014. Genetic variability of MHC class II DQB exon 2 alleles in yak (Bos grunniens). Mol. Biol. Rep. 41, 2199–2206. doi:10.1007/s11033-014-3071-3

Hirota, T., Takahashi, A., Kubo, M., Tsunoda, T., Tomita, K., Sakashita, M., Yamada, T., Fujieda, S., Tanaka, S., Doi, S., Miyatake, A., Enomoto, T., Nishiyama, C., Nakano, N., Maeda, K., Okumura, K., Ogawa, H., Ikeda, S., Noguchi, E., Sakamoto, T., 2012. Genome-wide 32 association study identifies eight new susceptibility loci for atopic dermatitis in the Japanese population. Nat. Genet. 44, 1222–1226. doi:10.1038/ng.2438

Ho, C.-S., Martens, G.W., Amoss, M.S., Gomez-Raya, L., Beattie, C.W., Smith, D.M., 2010. Swine leukocyte antigen (SLA) diversity in Sinclair and Hanford swine. Dev. Comp. Immunol. 34, 250–257. doi:10.1016/j.dci.2009.09.006

Höglund, P., Mintern, J., Waltzinger, C., Heath, W., Benoist, C., Mathis, D., 1999. Initiation of Autoimmune Diabetes by Developmentally Regulated Presentation of Islet Cell Antigens in the Pancreatic Lymph Nodes. J. Exp. Med. 189, 331–339. doi:10.1084/jem.189.2.331

Horín, P., Cothran, E.G., Trtková, K., Marti, E., Glasnák, V., Henney, P., Vyskocil, M., Lazary, S., 1998. Polymorphism of Old Kladruber horses, a surviving but endangered baroque breed. Eur. J. Immunogenetics Off. J. Br. Soc. Histocompat. Immunogenetics 25, 357–363.

Horín, P., Matiasovic, J., 2002. A second locus and new alleles in the major histocompatibility complex class II (ELA-DQB) region in the horse. Anim. Genet. 33, 196–200.

Hořín, P., Smola, J., Matiašovic, J., Vyskočil, M., Lukeszová, L., Tomanová, K., Králík, P., Glasnák, V., Schröffelová, D., Knoll, A., Sedlinská, M., Krřenková, L., Jahn, P., 2004. Polymorphisms in equine immune response genes and their associations with infections. Mamm. Genome 15, 843–850. doi:10.1007/s00335-004-2356-6

Huang, J., Zhao, Y., Shiraigol, W., Li, B., Bai, D., Ye, W., Daidiikhuu, D., Yang, L., Jin, B., Zhao, Q., Gao, Y., Wu, J., Bao, W., Li, A., Zhang, Y., Han, H., Bai, H., Bao, Y., Zhao, L., Zhai, Z., Zhao, W., Sun, Z., Zhang, Y., Meng, H., Dugarjaviin, M., 2014. Analysis of horse genomes provides insight into the diversification and adaptive evolution of karyotype. Sci. Rep. 4, 4958. doi:10.1038/srep04958

Hughes, A.L., Yeager, M., 1998. Natural selection at major histocompatibility complex loci of vertebrates. Annu. Rev. Genet. 32, 415–435. doi:10.1146/annurev.genet.32.1.415

Janova, E., Matiasovic, J., Vahala, J., Vodicka, R., Van Dyk, E., Horin, P., 2009. Polymorphism and selection in the major histocompatibility complex DRA and DQA genes in the family Equidae. Immunogenetics 61, 513–527. doi:10.1007/s00251-009-0380-0

Jaratlerdsiri, W., Isberg, S.R., Higgins, D.P., Ho, S.Y.W., Salomonsen, J., Skjodt, K., Miles, L.G., Gongora, J., 2014. Evolution of MHC class I in the order Crocodylia. Immunogenetics 66, 53–65. doi:10.1007/s00251-013-0746-1

Jeffery, K.J., Bangham, C.R., 2000. Do infectious diseases drive MHC diversity? Microbes Infect. Inst. Pasteur 2, 1335–1341.

Jónsson, H., Schubert, M., Seguin-Orlando, A., Ginolhac, A., Petersen, L., Fumagalli, M., Albrechtsen, A., Petersen, B., Korneliussen, T.S., Vilstrup, J.T., Lear, T., Myka, J.L., Lundquist, J., Miller, D.C., Alfarhan, A.H., Alquraishi, S.A., Al-Rasheid, K.A.S., Stagegaard, J., Strauss, G., Bertelsen, M.F., Sicheritz-Ponten, T., Antczak, D.F., Bailey, E., Nielsen, R., Willerslev, E., Orlando, L., 2014. Speciation with gene flow in equids despite extensive chromosomal plasticity. Proc. Natl. Acad. Sci. 111, 18655–18660. doi:10.1073/pnas.1412627111

Juliarena, M.A., Poli, M., Sala, L., Ceriani, C., Gutierrez, S., Dolcini, G., Rodríguez, E.M., Mariño, B., Rodríguez-Dubra, C., Esteban, E.N., 2008. Association of BLV infection profiles with alleles of the BoLA-DRB3.2 gene. Anim. Genet. 39, 432–438. doi:10.1111/j.1365- 2052.2008.01750.x

Kamath, P.L., Getz, W.M., 2011. Adaptive molecular evolution of the Major Histocompatibility Complex genes, DRA and DQA, in the genus Equus. BMC Evol. Biol. 11, 128. doi:10.1186/1471-2148-11-128 33

Kamath, P.L., Turner, W.C., Küsters, M., Getz, W.M., 2014. Parasite-mediated selection drives an immunogenetic trade-off in plains zebras (Equus quagga). Proc. R. Soc. Lond. B Biol. Sci. 281, 20140077. doi:10.1098/rspb.2014.0077

Kaufman, J., 2013. Antigen processing and presentation: evolution from a bird’s eye view. Mol. Immunol. 55, 159–161. doi:10.1016/j.molimm.2012.10.030

Kehrli, D., Jandova, V., Fey, K., Jahn, P., Gerber, V., 2015. Multiple Hypersensitivities Including Recurrent Airway Obstruction, Insect Bite Hypersensitivity, and Urticaria in 2 Warmblood Horse Populations. J. Vet. Intern. Med. 29, 320–326. doi:10.1111/jvim.12473

Kelley, J., Walter, L., Trowsdale, J., 2005. Comparative genomics of major histocompatibility complexes. Immunogenetics 56, 683–695. doi:10.1007/s00251-004-0717-7

Klein, J., 1986. Natural History of the Major Histocompatability Complex, 99th edition. ed. John Wiley & Sons Inc, New York.

Krejsek, J., Kopecký, O., 2004. Klinická imunologie. NUCLEUS HK.

Krüger, K., Gaillard, C., Stranzinger, G., Rieder, S., 2005. Phylogenetic analysis and species allocation of individual equids using microsatellite data. J. Anim. Breed. Genet. 122, 78– 86. doi:10.1111/j.1439-0388.2005.00505.x

Lara-Marquez, Yunis, Layrisse, Ortega, Carvallo-Gil, Montagnani, Makhatadze, Pocino, Granja, Yunis, Lara-Marquez, M.L., 1999. Immunogenetics of atopic asthma: association of DRB1*1101 DQA1*0501 DQB1*0301 haplotype with Dermatophagoides spp.-sensitive asthma in a sample of the Venezuelan population. Clin. Exp. Allergy 29, 60–71. doi:10.1046/j.1365- 2222.1999.00461.x

Lazary, S., Antczak, D.F., Bailey, E., Bell, T.K., Bernoco, D., Byrns, G., McClure, J.J., 1988. Joint Report of the Fifth International Workshop on Lymphocyte Alloantigens of the Horse, Baton Rouge, Louisiana, 31 October-1 November 1987. Anim. Genet. 19, 447–456.

Lazary, S., Marti, E., Szalai, G., Gaillard, C., Gerber, H., 1994. Studies on the frequency and associations of equine leucocyte antigens in sarcoid and summer dermatitis. Anim. Genet. 25, 75–80. doi:10.1111/j.1365-2052.1994.tb00406.x

Lee, H.J., Ha, S.J., Han, H., Kim, J.W., 2001. Distribution of HLA-A, B alleles and polymorphisms of TAP and LMP genes in Korean patients with atopic dermatitis. Clin. Exp. Allergy 31, 1867–1874. doi:10.1046/j.1365-2222.2001.01042.x

Lewin, H.A., Wu, M.-C., Stewart, J.A., Nolan, T.J., 1988. Association between BoLA and subclinical bovine leukemia virus infection in a herd of Holstein-Friesian cows. Immunogenetics 27, 338–344. doi:10.1007/BF00395129

Lohm, J., Grahn, M., Langefors, Å., Andersen, Ø., Storset, A., Schantz, T. von, 2002. Experimental evidence for major histocompatibility complex–allele–specific resistance to a bacterial infection. Proc. R. Soc. Lond. B Biol. Sci. 269, 2029–2033. doi:10.1098/rspb.2002.2114

Mäkinen, A., Chowdhary, B., Mahdy, E., Andersson, L., Gustavsson, I., 1989. Localization of the equine major histocompatibility complex (ELA) to chromosome 20 by in situ hybridization. Hereditas 110, 93–96.

March, M.E., Sleiman, P.M.A., Hakonarson, H., 2015. Genome-wide association studies in asthma: progress and pitfalls. Adv. Genomics Genet. 5, 107–118. doi:10.2147/AGG.S54505

Martínez-Borra, J., López-Larrea, C., 2012. The emergence of the major histocompatilibility complex. Adv. Exp. Med. Biol. 738, 277–289. doi:10.1007/978-1-4614- 1680-7_16

Martinsohn, J.T., Sousa, A.B., Guethlein, L.A., Howard, J.C., 1999. The gene conversion hypothesis of MHC evolution: a review. Immunogenetics 50, 168–200.

Mashima, S., 2003. Comparative sequence analysis of equine and human MHC class II DQB genes. Cytogenet. Genome Res. 102, 196–200. doi:75748

Meyer, D., Thomson, G., 2001. How selection shapes variation of the human major histocompatibility complex: a review. Ann. Hum. Genet. 65, 1–26.

Migueles, S.A., Sabbaghian, M.S., Shupert, W.L., Bettinotti, M.P., Marincola, F.M., Martino, L., Hallahan, C.W., Selig, S.M., Schwartz, D., Sullivan, J., Connors, M., 2000. HLA B*5701 is highly associated with restriction of virus replication in a subgroup of HIV-infected long term nonprogressors. Proc. Natl. Acad. Sci. 97, 2709–2714. doi:10.1073/pnas.050567397

Milich, D.R., Leroux-Roels, G.G., 2003. Immunogenetics of the response to HBsAg vaccination. Autoimmun. Rev. 2, 248–257. doi:10.1016/S1568-9972(03)00031-4

Monteyne, P., Bureau, J.F., Brahic, M., 1997. The infection of mouse by Theiler’s virus: from genetics to immunology. Immunol. Rev. 159, 163–176.

Nagore, E., Planelles, M.D., Ledesma, E., Millán, J.M., Insa, A., Oliver, V., Guillén, C., Fortea, J.M., 2002. Molecular Genetic Analysis of HLA-DR and -DQ Alleles in Spanish Patients with Melanoma. Acta Derm. Venereol. 82, 90–93. doi:10.1080/00015550252948095

Nei, M., Rooney, A.P., 2005. Concerted and birth-and-death evolution of multigene families. Annu. Rev. Genet. 39, 121–152. doi:10.1146/annurev.genet.39.073003.112240

Newby, P.R., Schmutz, C., Buckley, C.R., Stockley, R.A., Wood, A.M., 2011. Differentiation of Monocytes to Pro-Inflammatory Forms Is Influenced by Cigarette Smoke and Hla Type in Copd. Thorax A38–A38. doi:10.1136/thoraxjnl-2011-201054b.78

Nielsen, R., 2005. Molecular Signatures of Natural Selection. Annu. Rev. Genet. 39, 197–218. doi:10.1146/ annurev.genet.39.073003.112420

Nikolich-Žugich, J., Fremont, D.H., Miley, M.J., Messaoudi, I., 2004. The role of mhc polymorphism in anti-microbial resistance. Microbes Infect. 6, 501–512. doi:10.1016/j.micinf.2004.01.006

Osborne, A.J., Pearson, J., Negro, S.S., Chilvers, B.L., Kennedy, M.A., Gemmell, N.J., 2015. Heterozygote advantage at MHC DRB may influence response to infectious disease epizootics. Mol. Ecol. 24, 1419–1432. doi:10.1111/mec.13128

Ovsyannikova, I.G., Jacobson, R.M., Poland, G.A., 2004. Variation in vaccine response in normal populations. Pharmacogenomics 5, 417–427. doi:10.1517/14622416.5.4.417

Penn, D.J., Damjanovich, K., Potts, W.K., 2002. MHC heterozygosity confers a selective advantage against multiple-strain infections. Proc. Natl. Acad. Sci. 99, 11260–11264. doi:10.1073/pnas.162006499

Planelles, D., Nagore, E., Moret, A., Botella-Estrada, R., Vila, E., Guillén, C., Montoro, J.A., 2006. HLA class II polymorphisms in Spanish melanoma patients: homozygosity for HLA- DQA1 locus can be a potential melanoma risk factor. Br. J. Dermatol. 154, 261–266. doi:10.1111/j.1365-2133.2005.06896.x

Rappocciolo, G., Birch, J., Ellis, S.A., 2003. Down-regulation of MHC class I expression by equine herpesvirus-1. J. Gen. Virol. 84, 293–300. doi:10.1099/vir.0.18612-0 35

Rosengren Pielberg, G., Golovko, A., Sundström, E., Curik, I., Lennartsson, J., Seltenhammer, M.H., Druml, T., Binns, M., Fitzsimmons, C., Lindgren, G., Sandberg, K., Baumung, R., Vetterlein, M., Strömberg, S., Grabherr, M., Wade, C., Lindblad-Toh, K., Pontén, F., Heldin, C.-H., Sölkner, J., Andersson, L., 2008. A cis-acting regulatory mutation causes premature hair graying and susceptibility to melanoma in the horse. Nat. Genet. 40, 1004– 1009. doi:10.1038/ng.185

Rusek, J., Klumplerova, M., Molinkova, D., Sedlinska, M., Dusek, L., Muzik, J., Putnova, L., Vrtkova, I., Celer, V., Horin, P., 2013. Genetics of anti-EHV antibody responses in a horse population. Res. Vet. Sci. 95, 137–142. doi:10.1016/j.rvsc.2013.03.011

Sollid, L.M., Pos, W., Wucherpfennig, K.W., 2014. Molecular mechanisms for contribution of MHC molecules to autoimmune diseases. Curr. Opin. Immunol., Autoimmunity * Allergy and hypersensitivity 31, 24–30. doi:10.1016/j.coi.2014.08.005

Sommer, S., 2005. The importance of immune gene variability (MHC) in evolutionary ecology and conservation. Front. Zool. 2, 16. doi:10.1186/1742-9994-2-16

Sommer, S., Courtiol, A., Mazzoni, C.J., 2013. MHC genotyping of non-model organisms using next-generation sequencing: a new methodology to deal with artefacts and allelic dropout. BMC Genomics 14, 542. doi:10.1186/1471-2164-14-542

Spurgin, L.G., Richardson, D.S., 2010. How pathogens drive genetic diversity: MHC, mechanisms and misunderstandings. Proc. Biol. Sci. 277, 979–988. doi:10.1098/rspb.2009.2084

Stear, M.J., Wakelin, D., 1998. Genetic resistance to parasitic infection. Rev. Sci. Tech. Int. Off. Epizoot. 17, 143–153.

Szalai, G., Antczak, D.F., Gerber, H., Lazary, S., 1994. Molecular cloning and characterization of horse DQB cDNA. Immunogenetics 40, 458.

Szalai, G., Bailey, E., Gerber, H., Lazary, S., 1993. DNA sequence analysis of serologically detected ELA class II haplotypes at the equine DQ beta locus. Anim. Genet. 24, 187–190.

Takada, T., Kikkawa, Y., Yonekawa, H., Amano, T., 1998. Analysis of goat MHC class II DRA and DRB genes: identification of the expressed gene and new DRB alleles. Immunogenetics 48, 408–412.

Takahashi, K., Rooney, A.P., Nei, M., 2000. Origins and divergence times of mammalian class II MHC gene clusters. J. Hered. 91, 198–204. doi:10.1093/jhered/91.3.198

Takahata, N., Nei, M., 1990. Allelic Genealogy under Overdominant and Frequency- Dependent Selection and Polymorphism of Major Histocompatibility Complex Loci. Genetics 124, 967–978.

Tallmadge, R.L., Campbell, J.A., Miller, D.C., Antczak, D.F., 2010. Analysis of MHC class I genes across horse MHC haplotypes. Immunogenetics 62, 159–172. doi:10.1007/s00251- 009-0420-9

Tallmadge, R.L., Lear, T.L., Antczak, D.F., 2005. Genomic characterization of MHC class I genes of the horse. Immunogenetics 57, 763–774. doi:10.1007/s00251-005-0034-9

Tantisira, K., Klimecki, W.T., Lazarus, R., Palmer, L.J., Raby, B.A., Kwiatkowski, D.J., Silverman, E., Vercelli, D., Martinez, F.D., Weiss, S.T., 2004. Toll-like receptor 6 gene (TLR6): single-nucleotide polymorphism frequencies and preliminary association with the diagnosis of asthma. Genes Immun. 5, 343–346. doi:10.1038/sj.gene.6364096

Těšický, M., Vinkler, M., 2015. Trans-Species Polymorphism in Immune Genes: General Pattern or MHC-Restricted Phenomenon? J. Immunol. Res. 2015, 838035. doi:10.1155/2015/838035

Thursz, M.R., Kwiatkowski, D., Allsopp, C.E.M., Greenwood, B.M., Thomas, H.C., Hill, A.V.S., 1995. Association between an MHC Class II Allele and Clearance of Hepatitis B Virus in the Gambia. N. Engl. J. Med. 332, 1065–1069. doi:10.1056/NEJM199504203321604

Thursz, M., Yallop, R., Goldin, R., Trepo, C., Thomas, H.C., 1999. Influence of MHC class II genotype on outcome of infection with hepatitis C virus. The Lancet 354, 2119–2124. doi:10.1016/S0140-6736(99)91443-5

Trowsdale, J., Parham, P., 2004. Mini-review: defense strategies and immunity-related genes. Eur. J. Immunol. 34, 7–17. doi:10.1002/eji.200324693

Vilstrup, J.T., Seguin-Orlando, A., Stiller, M., Ginolhac, A., Raghavan, M., Nielsen, S.C.A., Weinstock, J., Froese, D., Vasiliev, S.K., Ovodov, N.D., Clary, J., Helgen, K.M., Fleischer, R.C., Cooper, A., Shapiro, B., Orlando, L., 2013. Mitochondrial phylogenomics of modern and ancient equids. PloS One 8, e55950. doi:10.1371/journal.pone.0055950

Wade, C.M., Giulotto, E., Sigurdsson, S., Zoli, M., Gnerre, S., Imsland, F., Lear, T.L., Adelson, D.L., Bailey, E., Bellone, R.R., Blöcker, H., Distl, O., Edgar, R.C., Garber, M., Leeb, T., Mauceli, E., MacLeod, J.N., Penedo, M.C.T., Raison, J.M., Sharpe, T., Vogel, J., Andersson, L., Antczak, D.F., Biagi, T., Binns, M.M., Chowdhary, B.P., Coleman, S.J., Della Valle, G., Fryc, S., Guérin, G., Hasegawa, T., Hill, E.W., Jurka, J., Kiialainen, A., Lindgren, G., Liu, J., Magnani, E., Mickelson, J.R., Murray, J., Nergadze, S.G., Onofrio, R., Pedroni, S., Piras, M.F., Raudsepp, T., Rocchi, M., Røed, K.H., Ryder, O.A., Searle, S., Skow, L., Swinburne, J.E., Syvänen, A.C., Tozaki, T., Valberg, S.J., Vaudin, M., White, J.R., Zody, M.C., Broad Institute Genome Sequencing Platform, Broad Institute Whole Genome Assembly Team, Lander, E.S., Lindblad-Toh, K., 2009. Genome sequence, comparative analysis, and population genetics of the domestic horse. Science 326, 865–867. doi:10.1126/science.1178158

Wagner, J.L., Burnett, R.C., Storb, R., 1999. Organization of the canine major histocompatibility complex: current perspectives. J. Hered. 90, 35–38.

Warren, E.H., Zhang, X.C., Li, S., Fan, W., Storer, B.E., Chien, J.W., Boeckh, M.J., Zhao, L.P., Martin, P.J., Hansen, J.A., 2012. Effect of MHC and non-MHC donor/recipient genetic disparity on the outcome of allogeneic HCT. Blood 120, 2796–2806. doi:10.1182/blood-2012- 04-347286

Westerdahl, H., Waldenström, J., Hansson, B., Hasselquist, D., Schantz, T. von, Bensch, S., 2005. Associations between malaria and MHC genes in a migratory songbird. Proc. R. Soc. B Biol. Sci. 272, 1511–1518. doi:10.1098/rspb.2005.3113

Yang, F., Fu, B., O’Brien, P.C.M., Robinson, T.J., Ryder, O.A., Ferguson-Smith, M.A., 2003. Karyotypic relationships of horses and zebras: results of cross-species chromosome painting. Cytogenet. Genome Res. 102, 235–243. doi:75755

Yang, Z., Bielawski, J.P., 2000. Statistical methods for detecting molecular adaptation. Trends Ecol. Evol. 15, 496–503. doi:10.1016/S0169-5347(00)01994-7

Yuhki, N., Beck, T., Stephens, R.M., Nishigaki, Y., Newmann, K., O’Brien, S.J., 2003. Comparative Genome Organization of Human, Murine, and Feline MHC Class II Region. Genome Res. 13, 1169–1179. doi:10.1101/gr.976103

Yuhki, N., Beck, T., Stephens, R., Neelam, B., O’Brien, S.J., 2007. Comparative genomic structure of human, dog, and cat MHC: HLA, DLA, and FLA. J. Hered. 98, 390–399. doi:10.1093/jhered/esm056 37

Zinkernagel, R.M., Leist, T., Hengartner, H., Althage, A., 1985. Susceptibility to lymphocytic choriomeningitis virus isolates correlates directly with early and high cytotoxic T cell activity, as well as with footpad swelling reaction, and all three are regulated by H-2D. J. Exp. Med. 162, 2125–2141.

LIST OF ABREVIATIONS

ABS antigen binding site

COPD chronic obstructive pulmonary disease

ELA equine leucocyte antigen

HIV human immunodeficiency virus

HLA human leucocyte antigen

HTLV human T-cell lymphotropic virus

IBH insect bite hypersensitivity

MHC major histocompatibility complex

MS manuscript

NGS next generation sequencing

PBR peptide binding region

PCR polymerase chain reaction

RAO recurrent airway obstruction

RFLP restriction fragment length polymorphism

SNP single nucleotide polymorphisms

SSCP single strand conformation polymorphism

APPENDIX

MS I Genetic diversity, evolution and selection in the major histocompatibility complex DRB and DQB genes in the family Equidae.

MS II Genetic diversity of the class II major histocompatibility DRA locus in European, Asiatic and African domestic donkeys.

80 Infection, Genetics and Evolution 11 (2011) 1136–1141

Contents lists available at ScienceDirect

Infection, Genetics and Evolution

journal homepage: www.elsevier.com/locate/meegid

Genetic diversity of the class II major histocompatibility DRA locus in European, Asiatic and African domestic donkeys

Marie Vranova a, Ingrid Alloggio b, Moneeb Qablan c, Mirko Vyskocil a, Aneta Baumeisterova a, Michal Sloboda c, Lenka Putnova d, Irena Vrtkova d, David Modry c, Petr Horin a,* a Institute of Animal Genetics, Faculty of Veterinary Medicine, University of Veterinary and Pharmaceutical Sciences Brno, Palackeho 1/3, 612 42 Brno, Czech Republic b Department PROGESA, Agronomic Faculty, University of Bari, via Amendola 165/a, 70126 Bari, Italy c Department of Parasitology, Faculty of Veterinary Medicine, University of Veterinary and Pharmaceutical Sciences Brno, Palackeho 1/3, 612 42 Brno, Czech Republic d Laboratory of Agrigenomics, Mendel University Brno, Zemedelska 1/1665, 613 00 Brno, Czech Republic

ARTICLE INFO ABSTRACT

Article history: The major histocompatibility complex (MHC) genes coding for antigen presenting molecules are the Received 15 December 2010 most polymorphic genes in vertebrate genome. The MHC class II DRA gene shows only small variation in Received in revised form 6 April 2011 many mammalian species, but it exhibits relatively high level of polymorphism in Equidae, especially in Accepted 7 April 2011 donkeys. This extraordinary degree of polymorphism together with signatures of selection in speciﬁc Available online 14 April 2011 amino acids sites makes the donkey DRA gene a suitable model for population diversity studies. The objective of this study was to investigate the DRA gene diversity in three different populations of donkeys Keywords: under infectious pressure of protozoan parasites, Theileria equi and Babesia caballi. Three populations of Domestic donkey domestic donkeys from Italy (N = 68), Jordan (N = 43), and Kenya (N = 78) were studied. A method of the Major histocompatibility class II DRA gene Genetic diversity donkey MHC DRA genotyping based on PCR-RFLP and sequencing was designed. In addition to the DRA Theileria equi gene, 12 polymorphic microsatellite loci were genotyped. The presence of Theileria equi and Babesia Babesia caballi caballi parasites in peripheral blood was investigated by PCR. Allele and genotype frequencies, observed

and expected heterozygosities and FIS values were computed as parameters of genetic diversity for all

loci genotyped. Genetic distances between the three populations were estimated based on FST values. Statistical associations between parasite infection and genetic polymorphisms were sought. Extensive DRA locus variation characteristic for Equids was found. The results showed differences between populations both in terms of numbers of alleles and their frequencies as well as variation in expected heterozygosity values. Based on comparisons with neutral microsatellite loci, population sub-structure characteristics and association analysis, convincing evidence of pathogen-driven selection at the population level was not provided. It seems that genetic diversity observed in the three populations reﬂects mostly effects of selective breeding and their different genetic origins. ß 2011 Elsevier B.V. All rights reserved.

1. Introduction and susceptibility to pathogens (Trowsdale and Parham, 2004; Tibayrenc, 2007). The major histocompatibility complex (MHC) is a The amount of genetic diversity has been associated with the cluster of linked genes playing a central role in the presentation of ability to adapt to environmental changes and with the potential to antigenic peptides to T lymphocytes (Klein, 1986). The MHC genes evolve (Reed and Frankham, 2003). Immune functions represent are the most polymorphic genes in the vertebrate genome. Their one of major components of an organism’s fitness and determine high polymorphism seems to be maintained by balancing the potential for evolutionary interactions with pathogens or with selection, predating speciation events and reflecting the co- other species (Lazzaro and Little, 2009). Diversity of genes evolution of hosts with their pathogens (Bernatchez and Landry, important for immune functions may be associated with resistance 2003). The mechanisms maintaining the genetic diversity and the role of pathogens have not yet been completely clarified. Empirical evidence for pathogen-driven selection on MHC genes is based on * Corresponding author. Tel.: +420 541 562 292; fax: +420 549 248 841. the population diversity analysis and on associations with E-mail addresses: [email protected] (M. Vranova), pathogens (Spurgin and Richardson, 2010). For this purpose, [email protected] (I. Alloggio), [email protected] (M. Qablan), specific model populations living in specific areas and exposed to [email protected] (M. Vyskocil), [email protected] (A. Baumeisterova), various pathogens can be studied. [email protected] (M. Sloboda), [email protected] (L. Putnova), [email protected] (I. Vrtkova), [email protected] (D. Modry), The family Equidae is a suitable model for studying diversity, [email protected] (P. Horin). selection and evolution of the MHC genes (Janova et al., 2009). It is

81 M. Vranova et al. / Infection, Genetics and Evolution 11 (2011) 1136–1141 1137 a rapidly evolving and variable group composed of a single genus, were sampled from several rural localities in western Jordan, Equus, with a relatively well-documented history of evolution characterized by hot semi-arid and arid climate; all sampled (Bowling and Ruvinsky, 2000). Domestic, captive and free ranging animals belonged to the local breed. The third population were equid populations are available for different types of studies. local African donkeys owned by semi-nomadic pastoralists of Domestication of wild asses occurred probably 6000 years ago in Turkana and Samburu tribes living in an arid environment in Northeastern Africa (Rossel et al., 2008). Analysis of mitochondrial Northern Kenya (n = 78). All donkeys were under permanent risk of DNA of modern donkeys revealed two highly divergent phylogenetic infection by tick-transmitted piroplasmids. In contrast to Italian groups, suggesting existence of two maternal origins of the domestic donkeys, where basic veterinary care is available, no therapeutic donkeys from two distinct wild populations, the Nubian (Equus and/or prophylactic measures were ever taken in the Asian and africanus africanus) and the Somali (Equus africanus somaliensis)wild African donkeys. asses (Beja-Pereira et al., 2004; Kimura et al., 2011). The domestic donkey is a suitable model equid species for diversity study. It exists 2.2. Assessment of genetic diversity in various populations in different geographical areas, often naturally exposed to infectious pathogens. The MHC DRA genetic diversity was compared to diversity in 14 The horse major histocompatibility complex (ELA or Eqca)is microsatellite loci. Distribution of genotype and allelic frequencies, located on the horse chromosome (ECA) 20. The equine and human expected heterozygosities within populations, population struc- MHCs have a similar genomic organization with class I, II and III ture and associations with a common pathogen were investigated regions (Gustafson et al., 2003). The class II genes of Equidae have for both types of loci, i.e. DRA and microsatellites. been extensively characterized and high level of exon 2 sequence variation was observed (Albright-Fraser et al., 1996; Fraser and 2.3. Genetic diversity within populations, neutrality tests Bailey, 1998; Horin and Matiasovic, 2002; Brown et al., 2004; Janova et al., 2009). For population diversity studies, a reliable method of In all donkeys, individual genotypes were determined for the individual genotyping is needed. Due to the extensive variation in MHC DRA locus and the microsatellite loci. In all loci analyzed, the class II DQA, DRB and DQB genes, individual genotyping of these genotype and allelic frequencies, expected, observed and unbiased genes in Equids is not available or it is of limited value (Fraser and expected heterozygosities and the corresponding P-values were Bailey, 1998; Diaz et al., 2001; Horin and Matiasovic, 2002; Janova et computed using GENETIX v. 4.05 (http://www.genetix.univ- al., 2009). montp2.fr/genetix/genetix.htm) and Arlequin v. 3.11 (Excoffier While exon 2 DRA alleles generally exhibit if ever only small et al., 2005; http://cmpg.unibe.ch/software/arlequin3/). Ewens- variation in mammalian species (e.g. Yuhki et al., 2003), extensive Watterson, Tajima’s D and Fu’s FS tests were used for analyzing polymorphism even of DRA genes has been reported in Equidae. neutrality of the DRA locus by Arlequin. The sequence variations are mainly located in exon 2 coding for the extracellular antigen binding domain. Current knowledge of the 2.4. DRA genotyping donkey MHC is only fragmentary. The donkey MHC (Eqas) contains probably a single DRA locus with seven DRA alleles identified so far Blood for DNA extraction was collected by jugular venipunc- (Albright-Fraser et al., 1996; Brown et al., 2004; GenBank ture. Two methods for genomic DNA extraction were used, due to accession numbers FJ487912, HM165492). Effect of positive different methods of fixation of blood samples collected in selection on exon 2 DRA sequences was reported (Janova et al., different climatic conditions. In Italian donkeys, genomic DNA 2009). was extracted from EDTA-fixed peripheral blood, using the Availability of various donkey populations living in different NucleoSpin blood kit (Macherey-Nagel, Duren, Germany). In climatic conditions and with different levels of general and health Jordanian and Kenyan donkeys, a standard phenol–chloroform care, relatively, but not extremely high level of polymorphism in a extraction from ethanol-fixed blood samples was used. single locus, with signatures of selection in specific amino acid Amplification of the 307 bp long product was carried out with sites, makes the donkey DRA gene a suitable model for population standard primers Be3 and Be4 (Albright-Fraser et al., 1996). The diversity studies. Similarly to other equids, donkeys are affected by extent of exon 2 DRA sequence variation in all populations was pre- plethora of infectious diseases. Among them, the piroplasmids, screened by single strand conformation polymorphism analysis apicomplexan intracellular protists represent valuable model (SSCP) as described previously (Janova et al., 2009). Individual pathogen, as they apparently co-evolved with their hosts and PCR-SSCP patterns were sequenced by 3730xl DNA analyzer exhibit remarkable pathogenicity. Equine and donkey piroplas- (Applied Biosystems, Foster City, CA, USA). The sequences obtained mosis is an often fatal, tick-borne disease of equids caused by were aligned by using the BioEdit sequence alignment editor (Hall, Theileria equi and Babesia caballi (Bruning, 1996). 1999) with known DRA alleles. A new E. asinus DRA allele The objective of this study was to investigate, based on individual (accession number HM165492), submitted to GenBank after we genotyping, the DRA gene diversity in three different populations of had completed the analysis, was not included. Heterozygote donkeys under infectious pressure of equine piroplasms. genotypes recognized based on double peaks were resolved manually by subtracting known alleles identified as specific SSCP 2. Materials and methods patterns. No new allele was identified in the groups studied. The nomenclature suggested by Janova et al. (2009) was used for 2.1. Animals designing the DRA alleles. Based on the exon 2 DRA sequences, a PCR-RFLP genotyping The genetic diversity was studied in three populations of system was developed. Digestion of PCR products with restriction domestic donkeys. Italian donkeys belonged to the Martina Franca enzymes BsaJI, NlaIII, AciI, and Cac8I produced fragments of breed. It is an ancient native breed of Apulia (southern Italy), specific length (Table 1) that could be distinguished by capillary characterized by extraordinary sturdiness, frugality and adapta- electrophoresis (MCE-202 MultiNA, Shimadzu Corporation, Kyoto, tion to rocky ground. The genetic uniqueness of this breed lies in its Japan). The combination of restriction sites allowed identification adaptation to enzootic tick-borne pathogens typically found in of 4 alleles, Eqas-DRA*0101, Eqas-DRA*0201, Eqas-DRA*0401 and Apulia (Rizzi et al., in press). Unrelated donkeys selected from 12 Eqas-DRA*0501. The remaining alleles, Eqas-DRA*0301 and Eqas- farms (n = 68) were used in this study. Jordanian donkeys (n = 43) DRA*0601, could be identified by subsequent sequencing and

82 1138 M. Vranova et al. / Infection, Genetics and Evolution 11 (2011) 1136–1141

Table 1 Table 2 Predicted DNA fragment size (bp) for PCR-RFLP detection of Eqas alleles. Frequencies of DRA alleles.

BsaJI NlaIII AciI Cac8I Na Italy Jordan Kenya

Eqas-DRA*01 73,228,6 128,119,60 157,150 184,123 68 42 78 Eqas-DRA*02 307 247,60 157,150 184,123 f (Eqas-DRA*01) 0.096 0.107 0.032 Eqas-DRA*03 73,228,6 247,60 157,150 184,123 f (Eqas-DRA*02) 0.199 0.381 0.468 Eqas-DRA*04 73,228,6 247,60 157,150 307 f (Eqas-DRA*03) 0.198 0.143 0.083 Eqas-DRA*05 73,228,6 247,60 307 184,123 f (Eqas-DRA*04) 0.007 0.083 0.244 Eqas-DRA*06 73,228,6 247,60 157,150 184,123 f (Eqas-DRA*05) 0.500 0.072 0.077 f (Eqas-DRA*06) 0.000 0.214 0.096

a manual editing of heterozygous sequences. As the recently Number of individuals. reported allele Eqas-DRA*0701 could not be involved in the set- up, it could not be distinguished with this genotyping system. computed was corrected for six/five alleles identified in the respective populations, while for microsatellite loci; corrections 2.5. Microsatellites for 12 polymorphic loci analyzed were made. The associations were analyzed within the three populations, within the whole Fourteen microsatellites from the horse parentage test (Lee and group of donkeys analyzed and within the group of Kenyan and Cho, 2006) were amplified in a standard multiplex PCRs using Jordanian donkeys with the same allele pool. fluorescent-labeled primers and analyzed using automated sequencer ABI Prism310 (Applied Biosystems, Foster City, CA, USA) as 3. Results described (Glowatzki-Mullis et al., 2006). 12 microsatellite loci, including one X-linked marker (Lex003), were polymorphic in the The numbers of DRA alleles and their frequencies, the values of populations analyzed. heterozygosity found in the populations studied are in Tables 2 and 3. In Jordanian and Kenyan donkeys, six DRA alleles were found, 2.6. Population structure while in Italian donkeys the allele Eqas-DRA*0601 was missing. The highest value of expected heterozygosity was found in Jordanian FIS and FST values for the entire donkey group and for its three donkeys, while their observed heterozygosity was lowest from the sub-populations and their P values were computed by Arlequin group. Departure from H–W equilibrium due to the excess of DRA (based on Weir and Hill, 2002) for both types of loci. homozygotes was observed in the Jordanian population. The data on microsatellite diversity are in Table 4. Two loci HMS1 2.7. Model pathogen: blood parasites Theileria equi and Babesia and ASB17 were monomorphic. In the remaining loci, numbers of caballi alleles ranged from 2 (HMS6)to12(AHT4). The numbers of alleles were smaller in Italian donkeys (average number per locus 4.42), The blood parasites Theileria equi and Babesia caballi occurring while in Jordanian and Kenyan donkeys the numbers of alleles per in all three populations were selected as model pathogens for locus were similar (6.83 and 6.75, respectively). association studies. Both pathogens were diagnosed by PCR from The FIS values for DRA and microsatellite loci are also shown in DNA extracted from peripheral blood of all donkeys analyzed. Table 3. All FIS values were positive except the DRA locus in Kenyan Amplification of a specific fragment of the piroplasmid SSU rRNA donkeys. The FST pair-wise comparisons among populations for gene was performed as described elsewhere (Sloboda et al., 2010). DRA and microsatellites are in Table 5. In both types of loci, the Based on the results of PCR, donkeys were classified as ‘‘double genetic distances between the Jordanian and Kenyan populations positive’’, Theileria equi or Babesia caballi positive or negative. The are smaller than distances between either of them and the Italian values of pathogen prevalence were 77.94; 26.19 and 88.46% in the group. The FST values for DRA and microsatellites were comparable. Italian, Jordanian and Kenyan donkeys, respectively. In neutrality tests, only a significant Ewens–Waterson neutrality

test P value for the Jordanian population was found (Fobs = 0.235, 2.8. Association analysis Fexp = 0.417, P = 0.041). Marginal P values were found for associations of Eqas-DRA with Associations between infection with one or both pathogens and Theileria equi infection in the merged group of Jordanian and DRA and microsatellite polymorphisms were investigated using Kenyan donkeys (Table 6). A statistically signiﬁcant difference standard chi-square and/or Fisher’s exact tests with Bonferroni (P < 0.004) in numbers of Theileria equi PCR-positive donkeys corrections for multiple comparisons. For DRA alleles, the P value between Eqas-DRA*0401 and Eqas-DRA*0601 carriers was observed

Table 3 Comparison of genetic diversity of DRA locus and average values of 12 polymorphic microsatellite loci.

a b Nall H exp. H n.b. H obs. FIS (CI 95%) HWE P value/range of P values

Italy DRA 5.00 0.662 0.667 0.603 0.097 (À0.02 to 0.206) 0.294 ( MS 4.42 0.569 0.574 0.555 0.033 (0.004 to 0.052) 0.132–0.933

N = 67) Jordan DRA 6.00 0.765 0.774 0.548 0.295 (0.147 to 0.359) 0.000 (N = 43) MS 6.83 0.697 0.705 0.623 0.119 (0.084 to 0.155) 0.02–0.785 Kenya DRA 6.00 0.699 0.703 0.769 À0.095 (À0.176 to 0.012) 0.572 (N = 78) MS 6.75 0.691 0.695 0.645 0.072 (0.045 to 0.106) 0.010–0.866

H exp., expected heterozygosity; H n.b., non-biased expected heterozygosity; H obs., observed heterozygosity. a Mean number of alleles per loci. b 95% conﬁdence interval.

83 M. Vranova et al. / Infection, Genetics and Evolution 11 (2011) 1136–1141 1139

Table 4 Genetic diversity of 11 autosomal microsatellite loci in three domestic donkey populations.

Number of alleles H exp. H n.b. H obs. P value

Italy N = 67 Jordan N = 43 Kenya N = 78 Italy Jordan Kenya Italy Jordan Kenya Italy Jordan Kenya Italy Jordan Kenya

AHT004 3 12 10 0.609 0.774 0.780 0.614 0.783 0.785 0.671 0.786 0.705 0.381 0.301 0.369 VHL020 3 3 5 0.642 0.605 0.664 0.646 0.612 0.669 0.776 0.605 0.654 0.134 0.785 0.866 AHT005 7 11 11 0.773 0.873 0.825 0.779 0.885 0.831 0.776 0.769 0.859 0.603 0.051 0.628 ASB023 5 5 6 0.662 0.791 0.781 0.667 0.801 0.786 0.657 0.833 0.718 0.132 0.617 0.150 HMS006 2 4 4 0.430 0.447 0.328 0.434 0.452 0.330 0.388 0.349 0.303 0.404 0.254 0.321 HTG006 4 4 4 0.254 0.654 0.687 0.256 0.663 0.691 0.227 0.657 0.718 0.412 0.020 0.232 CA425 4 9 7 0.515 0.704 0.717 0.519 0.713 0.722 0.552 0.595 0.680 0.579 0.136 0.440 HMS002 3 6 6 0.419 0.637 0.661 0.422 0.644 0.665 0.493 0.619 0.590 0.180 0.597 0.010 HMS003 4 4 3 0.640 0.546 0.606 0.645 0.555 0.610 0.609 0.531 0.526 0.487 0.109 0.059 HTG010 6 8 7 0.605 0.773 0.646 0.609 0.783 0.650 0.597 0.833 0.654 0.933 0.621 0.643 HTG007 7 10 10 0.620 0.862 0.875 0.625 0.873 0.881 0.522 0.791 0.885 0.135 0.224 0.491

H exp., expected heterozygosity; H n.b., non-biased expected heterozygosity; H obs., observed heterozygosity.

Table 5 SSCP. However, pre-screening with PCR-SSCP was used for Genetic distances (pairwise FST) determined for MHC-DRA/microsatellite loci. avoiding the risk of loss of non-recognized alleles potentially Italy Jordan Kenya present in the populations studied.

Italy 0 Five DRA alleles seem to be equally common in donkeys Jordan 0.154/0.132 0 analyzed so far. The sixth and seventh allele, Eqas-DRA*0601, Eqas- Kenya 0.191/0.152 0.029/0.041 0 DRA*0701 were only recently added to GenBank (FJ487912, HM165492) and there is no information on their population frequencies. DRA frequencies observed by Brown et al. (2004) in 23 (Table 7). No associations of DRA alleles or genotypes with donkeys (origin not specified) were similar to those found in this homozygosity/heterozygosity were found. No associations for study in the Jordanian group. Eqas-DRA*0601 and Eqas-DRA*0701 microsatellite loci were found. were not known at this time and they were not identified by the authors in the group analyzed. In two populations studied here, all 4. Discussion known DRA alleles were found with exception of Eqas-DRA*0701.It could not be distinguished by the PCR-RFLP system used. However, The extent of Eqas DRA polymorphism was primarily deter- we have observed no unexpected PCR-SSCP patterns in the groups mined by sequencing. In Equids, individual DRA genotyping based analyzed and the sequence data have not suggested existence of on single strand conformation polymorphism analysis (PCR-SSCP) additional alleles. We thus believe that if this allele was present in (Albright-Fraser et al., 1996), reference-strand-mediated confor- our populations, it must be very rare and could not influence the mational analysis (RSCA) (Brown et al., 2004) or pyrosequencing results obtained and their interpretation. (Diaz et al., 2008) was reported. Our approach using PCR-RFLP and In the Italian group, where Eqas-DRA*0601 was not found, sequencing proved to be another feasible approach for assessing observed frequency of another allele, Eqas-DRA*0401, was also very individual variation in the donkey DRA locus. It is rapid and low. This population thus seems to be less diverse in DRA than the efficient; there is no need for reference sequences like in RSCA and two other groups. Similar results were found for expected DRA

Table 6 Associations of Eqas DRA with Theileria equi infection.

Population/allele associated Positive Negative P uncorrected P corrected

Absolute fa Relative fa Absolute fa Relative fa

Jordan/Eqas-DRA*03 6/16 0.375 6/68 0.088 0.009 0.053 Jordan, Kenya/Eqas-DRA*04 34/154 0.221 9/86 0.105 0.010 0.061 Jordan, Kenya/Eqas-DRA*06 15/154 0.097 18/86 0.209 0.010 0.058

Associations of individual DRA alleles with Theileria equi infection were computed for the Jordanian population and for the merged group of Kenyan and Jordanian donkeys. Absolute frequencies show number of Theileria equi positive/negative individuals carrying a speciﬁc allele out of total numbers of Theileria equi positive/negative individuals. Odds ratios could not be calculated due to null frequencies. a Frequency.

Table 7 Distribution of susceptibility- and resistance-associated alleles (Eqas-DRA*04 and Eqas-DRA*06, respectively) in Jordanian and Kenyan donkeys.

Infected Non-infected

Absolute fa Relative fa Absolute fa Relative fa

Eqas-DRA*04 but not Eqas-DRA*06 carriers 24/30 0.80 6/30 0.20 Eqas-DRA*06 but not Eqas-DRA*04 carriers 9/22 0.41 13/22 0.59

P = 0.0038

Absolute frequencies show numbers of animals carrying only resistance- or susceptibility-associated alleles, respectively, among infected and non-infected donkeys. a Frequency.

84 1140 M. Vranova et al. / Infection, Genetics and Evolution 11 (2011) 1136–1141 heterozygosities. The Italian population showed lowest values of studied. However, significant differences in Theileria equi infection expected DRA heterozygosity. Calculations of expected and among Eqas-DRA*0401 and Eqas-DRA*0601 carriers were observed unbiased expected heterozygosities produced similar results due in two populations closely related in terms of genetic distances and to the high numbers of donkeys. The most diverse in this genetic diversity, living in similar arid environment with the same parameter was the Jordanian group, showing at the same time tick vectors (e.g. Rhipicephalus, Hyalomma). Effects of MHC on the lowest values of observed DRA heterozygosity. The values of protozoan infections could be biologically plausible (Luder et al., expected DRA heterozygosity observed in our donkeys (0.66–0.77) 2009), but biases in candidate-gene association studies leading to are higher than values observed for the same MHC class II locus in false positive results should always be considered (Campbell and horses, 0.27–0.65 (Diaz et al., 2008), and roughly correspond to Rudan, 2002). No association with microsatellites was found. This values for human HLA class I loci observed by Prugnolle et al. is in agreement with the assumption of neutrality, although for (2005) in populations living in environments with moderate level some microsatellites, associations with disease due to linkage of pathogen-richness. disequilibrium with other loci could be found, even in Equids The differences observed can be explained by several different (Horin et al., 2004). factors, including different origins of the populations under study, The data collected from these three special populations thus different approaches to selective breeding as well as by pathogen- allowed us to characterize the extent of genetic diversity in the driven selection. They suggest that populations living under Eqas DRA locus. Extensive DRA locus variation characteristic for ‘‘natural’’ conditions, i.e. with no selective breeding, virtually no Equid MHC was found. The results showed differences between veterinary care and no preventive measures, like vaccinations, are populations both in terms of numbers of alleles and their slightly more diverse in this MHC locus than the Italian breed, frequencies as well as variation in expected heterozygosity values. subject to selective breeding and elementary veterinary care. In At the molecular level, we could identify effects of positive Kenyan and Jordanian donkeys, no pedigree information was selection on the DRA locus (Janova et al., 2009). In this study available, and random sample collection could not eliminate the however, despite the fact that the pathogens selected are endemic risk of including relatives into the group analyzed. In Italian horses, in all three areas of origin and their long-term interaction with the pedigree data were available. Due to selective breeding, parental host populations might be assumed, we could not provide half-sibs could not be completely eliminated from the study. convincing evidence for pathogen-driven selection at the popula- However, in terms of parentage, the group analyzed was a tion level. Based on comparisons with neutral microsatellite loci, representative population sample. population sub-structure description and results of association Comparison with microsatellites showed that the Italian studies, it seems that the results reflect effects of selective breeding population was less diverse even in these neutral loci both in and different genetic origins of the populations studied rather than terms of heterozygosity values and especially of mean numbers of effects of pathogen-driven selection. alleles (Table 3). In contrast to Eqas DRA, there was no discrepancy between observed and expected heterozygosities in Jordanian Acknowledgments donkeys. The reasons for the departure from H–W equilibrium and from neutrality observed for the DRA locus in the Jordanian This work was supported by the Czech Science Foundation population remain unclear. Due to the absence of pedigree data, project GA CR 523/09/1972, partly by UVPS IGA 1230-IG101233 excess of homozygotes in this population cannot be interpreted project and by the Apulian Regional Fund ‘‘ATZ 2008’’ administered properly. The lower extent of genetic diversity in both types of loci by the Apulian Regional Office for Livestock Management. seems to be due to selective breeding applied in Italian donkeys with controlled pedigree. The mean expected microsatellite heterozygosity values ranging between 0.57 and 0.71, similar to References those reported in Spanish (0.66) and Croatian (0.66–0.70) domestic donkeys (Aranguren-Mendez et al., 2001; Ivankovic et al., 2002) Albright-Fraser, D.G., Reid, R., Gerber, V., Bailey, E., 1996. Polymorphism of DRA among equids. Immunogenetics 43, 315–317. are in agreement with this assumption. In terms of neutral Aranguren-Mendez, J., Jordana, J., Gomez, M., 2001. Genetic diversity in Spanish variation, genetic diversity of African and Arab donkeys studied donkey breeds using microsatellite DNA markers. Genet. Sel. Evol. 33, 433–442. was not strikingly different from the European domestic donkey Beja-Pereira, A., England, P.R., Ferrand, N., Jordan, S., Bakhiet, A.O., Abdalla, M.A., Mashkour, M., Jordana, J., Taberlet, P., Luikart, G., 2004. African origins of the populations. However, the two non-European populations were domestic donkey. Science 304, 1781–11781. more similar to each other than to Italian donkeys in both types of Bernatchez, L., Landry, C., 2003. MHC studies in nonmodel vertebrates: what have we learned about natural selection in 15 years? J. Evol. Biol. 16, 363–377. loci as expressed by FST values. Bowling, A., Ruvinsky, A. (Eds.), 2000. The Genetics of the Horse. CABI Publishing MHC polymorphism can be maintained by balancing selection Wallingford, New York, p. 527. (Hedrick, 1999). Effects of pathogen-driven selection on the MHC Brown, J.J., Thomson, W., Clegg, P., Eyre, S., Kennedy, L.J., Matthews, J., Carter, S., locus can be observed at the population level as differences Ollier, W.E., 2004. Polymorphisms of the equine major histocompatibility complex class II DRA locus. Tissue Antigens 64, 173–179. between MHC and neutral loci, population substructure and Bruning, A., 1996. Equine piroplasmosis an update on diagnosis, treatment and associations with pathogens (Spurgin and Richardson, 2010). The prevention. Br. Vet. J. 152, 139–151. Campbell, H., Rudan, I., 2002. Interpretation of genetic association studies in population structure estimated based on FST values, similar for complex disease. Pharmacogenomics J. 2, 349–360. microsatellite and MHC loci, did not indicate effects of selection, Diaz, S., Giovambattista, G., Dulout, F.N., Peral-Garcia, P., 2001. Genetic variation of similarly to the results of neutrality tests for the DRA locus. The the second exon of ELA-DRB genes in Argentine Creole horses. Anim. Genet. 32, pathogen richness in the areas concerned could not be estimated 257–263. due to lack of systemic data on disease prevalence in the Diaz, S., Echeverria, M.G., It, V., Posik, D.M., Rogberg-Munoz, A., Pena, N.L., Peral- Garcia, P., Vega-Pla, J.L., Giovambattista, G., 2008. Development of an ELA-DRA populations studied. The only pathogens comparable in the gene typing method based on pyrosequencing technology. Tissue Antigens 72, populations were piroplasmids. Associations between protozoan 464–468. infections and MHC were repeatedly reported in humans and Excoffier, L., Laval, G., Schneider, S., 2005. Arlequin (ver. 3.0): an integrated software package for population genetics data analysis. Evol. Bioinform. Online 1, 47–50. domestic animals (Gray and Gill, 1993; Stear and Wakelin, 1998; Fraser, D.G., Bailey, E., 1998. Polymorphism and multiple loci for the horse DQA Gilbert et al., 1998). Protozoan parasites were also shown to drive gene. Immunogenetics 47, 487–490. human genetic diversity (Pozzoli et al., 2010). Here, only marginal Gilbert, S.C., Plebanski, M., Gupta, S., Morris, J., Cox, M., Aidoo, M., Kwiatkowski, D., Greenwood, B.M., Whittle, H.C., Hill, A.V., 1998. Association of malaria parasite associations between Eqas DRA and infection were observed. This population structure, HLA, and immunological antagonism. Science 279, may be partially due to relatively small numbers of donkeys 1173–1177.

85 M. Vranova et al. / Infection, Genetics and Evolution 11 (2011) 1136–1141 1141

Glowatzki-Mullis, M.L., Muntwyler, J., Pfister, W., Marti, E., Rieder, S., Poncet, P.A., Luder, C.G., Stanway, R.R., Chaussepied, M., Langsley, G., Heussler, V.T., 2009. Gaillard, C., 2006. Genetic diversity among horse populations with a special Intracellular survival of apicomplexan parasites and host cell modification. focus on the Franches–Montagnes breed. Anim. Genet. 37, 33–39. Int. J. Parasitol. 39, 163–173. Gray, G.D., Gill, H.S., 1993. Host genes, parasites and parasitic infections. Int. J. Pozzoli, U., Fumagalli, M., Cagliani, R., Comi, G.P., Bresolin, N., Clerici, M., Sironi, M., Parasitol. 23, 485–494. 2010. The role of protozoan-driven selection in shaping human genetic vari- Gustafson, A.L., Tallmadge, R.L., Ramlachan, N., Miller, D., Bird, H., Antczak, D.F., ability. Trends Genet. 26, 95–99. Raudsepp, T., Chowdhary, B.P., Skow, L.C., 2003. An ordered BAC contig map of Prugnolle, F., Manica, A., Charpentier, M., Guegan, J.F., Guernier, V., Balloux, F., 2005. the equine major histocompatibility complex. Cytogenet. Genome Res. 102, Pathogen-driven selection and worldwide HLA class I diversity. Curr. Biol. 15, 185–195. 1022–1027. Hall, T.A., 1999. BioEdit. A user friendly biological sequence alignment editor Reed, D.H., Frankham, R., 2003. Correlation between fitness and genetic diversity. and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser 41, Conserv. Biol. 17, 230–237. 95–98. Rizzi, R., Tullo, E., Cito, A.M., Caroli, A., Piragostini, E. Monitoring of genetic diversity Hedrick, P.W., 1999. Balancing selection and MHC. Genetica 104, 207–214. in the endangered Martina Franca donkey population. J. Anim. Sci., in press. Horin, P., Matiasovic, J., 2002. A second locus and new alleles in the major Rossel, S., Marshall, F., Peters, J., Pilgram, T., Adams, M.D., O’Connor, D., 2008. histocompatibility complex class II (ELA-DQB) region in the horse. Anim. Genet. Domestication of the donkey: timing, processes and indicators. Proc. Natl. Acad. 33, 196–200. Sci. U.S.A. 105, 3715–3720. Horin, P., Smola, J., Matiasovic, J., Vyskocil, M., Lukeszova, L., Tomanova, K., Kralik, P., Sloboda, M., Jirku, M., Lukesova, D., Qablan, M., Batsukh, Z., Fiala, I., Horin, P., Modry, Glasnak, V., Schroffelova, D., Knoll, A., Sedlinska, M., Krenkova, L., Jahn, P., 2004. D., Lukes, J., 2010. A survey for piroplasmids in horses and Bactrian camels in Polymorphisms in equine immune response genes and their associations with north-eastern Mongolia. Vet. Parasitol. doi:10.1016/j.vetpar.2011.01.064. infections. Mamm. Genome 15, 843–850. Spurgin, L.G., Richardson, D.S., 2010. How pathogens drive genetic diversity: MHC, Ivankovic, A., Kavar, T., Caput, P., Mioc, B., Pavic, V., Dovc, P., 2002. Genetic diversity mechanisms and misunderstandings. Proc. Biol. Sci. 277, 979–988. of three donkey populations in the Croatian coastal region. Anim. Genet. 33, Stear, M.J., Wakelin, D., 1998. Genetic resistance to parasitic infection. Rev. - Off. Int. 169–177. Epizoot. 17, 143–153. Janova, E., Matiasovic, J., Vahala, J., Vodicka, R., Van Dyk, E., Horin, P., 2009. Tibayrenc, M., 2007. Human genetic diversity and the epidemiology of parasitic and Polymorphism and selection in the major histocompatibility complex DRA other transmissible diseases. Adv. Parasitol. 64, 377–422. and DQA genes in the family Equidae. Immunogenetics 61, 513–527. Trowsdale, J., Parham, P., 2004. Mini-review: defense strategies and immunity- Kimura, B., Marshall, F.B., Chen, S., Rosenbom, S., Moehlman, P.D., Tuross, N., related genes. Eur. J. Immunol. 34, 7–17. Sabin, R.C., Peters, J., Barich, B., Yohannes, H., Kebede, F., Teciai, R., Beja- Weir, B.S., Hill, W.G., 2002. Estimating F-statistics. Annu. Rev. Genet. 36, 721–750. Pereira, A., Mulligan, C.J., 2011. Ancient DNA from Nubian and Somali wild ass Yuhki, N., Beck, T., Stephens, R.M., Nishigaki, Y., Newmann, K., O’Brien, S.J., 2003. provides insights into donkey ancestry and domestication. Proc. Biol. Sci. 278, Comparative genome organization of human, murine, and feline MHC class II 50–57. region. Genome Res. 13, 1169–1179. Klein, J., 1986. Natural history of the major histocompatibility complex. Wiley, New York. Lazzaro, B.P., Little, T.J., 2009. Immunity in a variable world. Philos. Trans. R. Soc. Further reading (Web references) Lond., B, Biol. Sci. 364, 15–26. Lee, S.Y., Cho, G.J., 2006. Parentage testing of Thoroughbred horse in Korea using http://cmpg.unibe.ch/software/arlequin3/ (last accessed 14.12.2010). microsatellite DNA typing. J. Vet. Sci. 7, 63–67. http://www.genetix.univ-montp2.fr/genetix/genetix.htm (last accessed 14.12.2010).

86 MS III Genetic diversity and conservation in a small endangered horse population.

87 J Appl Genetics (2013) 54:285–292 DOI 10.1007/s13353-013-0151-3

ANIMAL GENETICS & ORIGINAL PAPER

Genetic diversity and conservation in a small endangered horse population

Eva Janova & Jan Futas & Marie Klumplerova & Lenka Putnova & Irena Vrtkova & Mirko Vyskocil & Petra Frolkova & Petr Horin

Received: 22 October 2012 /Revised: 14 April 2013 /Accepted: 17 April 2013 /Published online: 7 May 2013 # Institute of Plant Genetics, Polish Academy of Sciences, Poznan 2013

Abstract The Old Kladruber horses arose in the 17th cen- Genetic differences identified between the black and grey tury as a breed used for ceremonial purposes. Currently, sub-populations observed 13 years ago persisted. Deviations grey and black coat colour varieties exist as two sub- from the Hardy–Weinberg equilibrium found in 19 micro- populations with different recent breeding history. As the satellite loci and in five SNP loci are probably due to population underwent historical bottlenecks and intensive selective breeding. No differences between neutral and inbreeding, loss of genetic variation is considered as the immunity-related markers were found. No changes in the major threat. Therefore, genetic diversity in neutral and frequencies of markers associated with two diseases, mela- non-neutral molecular markers was examined in the current noma and insect bite hypersensitivity, were observed, due nucleus population. Fifty microsatellites, 13 single nucleo- probably to the short interval of time between comparisons. tide polymorphisms (SNPs) in immunity-related genes, It, thus, seems that, despite its small size, previous bottle- three mutations in coat colour genes and one major histo- necks and inbreeding, the molecular variation of Old compatibility (MHC-DRA) gene were studied for assessing Kladruber horses is comparable to other horse breeds and genetic diversity after 15 years of conservation. The results that the current breeding policy does not compromise ge- were compared to values obtained in a similar study 13 years netic variation of this endangered population. ago. The extent of genetic diversity of the current population was comparable to other breeds, despite its small size and Keywords Genetic diversity . Old Kladruber horses . isolation. The comparison between 1997 and 2010 did not Endangered horse breed . Microsatellites . Single nucleotide show differences in the extent of genetic diversity and no polymorphisms . Immunity-related genes . Conservation loss of allele richness and/or heterozygosity was observed.

Electronic supplementary material The online version of this article Introduction (doi:10.1007/s13353-013-0151-3) contains supplementary material, which is available to authorized users. : : : : The Old Kladruber horses arose in the 17th century as an E. Janova J.: Futas M. Klumplerova M. Vyskocil autochthonous Czech breed used for ceremonial purposes P. Frolkova P. Horin by the Habsburg emperors. Although its major founders Department of Animal Genetics, Faculty of Veterinary Medicine, University of Veterinary and Pharmaceutical Sciences, were old Spanish and Italian horses, important bottlenecks Palackého 1-3, 612 42 Brno, Czech Republic and admixtures of further breeds occurred repeatedly in the Old Kladruber horse breeding history (Porter 2002). Among : : * J. Futas M. Klumplerova P. Horin ( ) other breeds, the contribution of Lipizzaner horses was Ceitec VFU, University of Veterinary and Pharmaceutical Sciences, Palackého 1-3, 612 42 Brno, Czech Republic important. As a carriage horse, the Kladruber horse differed e-mail: [email protected] from Lipizzaners in size and performance capacities. Over : the years, the Kladruber population became specific and L. Putnova I. Vrtkova isolated from related breeds. In the 19th century, it was Laboratory of Agrigenomics, Department of Animal Morphology, Physiology and Genetics, Mendel University, Zemědělská 1, sub-divided into two colour varieties, grey and black, whose 613 00 Brno, Czech Republic effective size was significantly reduced due to further

88 286 J Appl Genetics (2013) 54:285–292 bottlenecks and inbreeding following the two world wars. recurrent process of inbreeding and genetic drift, when allele Currently,5to10stallionsandaround60–80 breeding frequencies tend to change from one generation to the next mares producing 35 to 40 foals of each of the colour vari- simply as a result of gametic sampling error (Arcos-Burgos eties form the breeding nucleus kept in the Kladruby and Muenke 2002). In addition, selective breeding in small National Stud (http://www.nhkladruby.cz). To a much lesser populations may have synergistic effects. extent, breeding horses from private owners can also con- Therefore, a concern about possible undesirable changes tribute to the gene pool. The mean value of the coefficient of in genetic variation of the Old Kladruber breed arose. Loss inbreeding in this population is 0.1 (Volenec et al. 1995; of heterozygosity and loss of specific alleles, especially in Jakubec et al. 2009; Vostrý et al. 2011). Since the 1st of immunity-related genes, are the major threats of this specific January 1996, no other breeds can be used within this population. The objective of this study was to analyse the population and a conservation programme aiming to pre- genetic diversity of the current nucleus population of breed- serve the Old Kladruber gene pool was launched. ing mares after 17 years of conservation in neutral as well as Small numbers of breeding horses, historical bottlenecks, in functionally important molecular markers, to compare specificity and the non-existence of closely related breeds parameters of genetic diversity between these two types of make the Kladruber a unique but endangered horse breed of markers and to make a comparison with the data obtained high historical and cultural value, which can be considered 13 years ago. as a model population in terms of conservation issues. The genetic diversity of an endangered breed, especially if it has been subject to inbreeding, is an important parameter for Materials and methods conservation programmes (Frankham 1995). Polymorphic molecular markers have become one of the fundamental Horses tools in population and conservation genetics. They can be used for determining the extent of variation, as well as for Peripheral blood for DNA analysis was collected in 2010. establishing conservation strategies of specific populations All breeding mares of the grey (n=75) and black (n=70) (O’Brien 1994; Thomson et al. 2010). sub-populations of the Old Kladruber breed sampled were We analysed the genetic diversity of the Old Kladruber analysed. Twenty-one out of 75 (28 %) grey and 6 out of 70 breed 13 years ago by using neutral as well as functionally (9 %) black mares were identical to those analysed 13 years important markers (Horín et al. 1998). Microsatellites, blood ago. The average age of the current grey sub-population was groups, biochemical polymorphisms and major histocompat- 12.4 years, as compared to 10.4 years in 1997, while in the ibility complex genes were characterised by methods avail- black sub-population, it was 11.4 years compared to able at that time. In both colour varieties, the extent of genetic 8.4 years, respectively. diversity was comparable to other horse breeds. Since then, methods of selection for size, conformation and performance Markers and genotyping have been modified (Jakubec et al. 2009) and an intensive programme of selection against insect bite hypersensitivity All breeding mares available during the study were geno- (IBH) was launched in the grey sub-population. In addition, typed by using a set of markers composed of microsatellites new genomic techniques and novel types of molecular (Msats) and molecular markers in expressed genes (SNPs, markers (single nucleotide polymorphisms, SNPs) became duplications, repetitions). available. Applications of SNPs for population genetics re- quire evaluation of the markers used for a specific population. Microsatellites In this context, information obtained based on SNPs in neutral and non-neutral loci may differ (Helyar et al. 2011). SNPs in Fifty microsatellite markers (AHT004, AHT005, AHT031, immunity-related genes may be associated with susceptibility ASB002, ASB017, ASB023, ASB043, COR007, COR018, to various diseases and different populations may display COR022, COR058, COR069, EB2E8, HMS001, HMS002, different patterns of SNP allele frequencies, linkage disequi- HMS003, HMS005, HMS006, HMS007, HTG003, HTG004, librium and haplotypes (Lazarus et al. 2002). Therefore, an HTG006, HTG007, HTG010, I018, LEX003, LEX033, assessment of genetic diversity in these loci may bring impor- LEX054, LEX073, LEX078, SGCV028, TKY287, TKY294, tant new information about a population and its vulnerability. TKY297, TKY301, TKY312, TKY321, TKY325, TKY333, This issue is especially important in small endangered TKY337, TKY341, TKY343, TKY344, TKY374, TKY394, populations. UCDEQ425, UM005, UM011, UM032, VHL020) were ge- Different stochastic and deterministic processes may oc- notyped in four multiplex polymerase chain reaction (PCR) cur in small and isolated populations. Genetic isolation can protocols (Glowatzki-Mullis et al. 2006). Length variations be interpreted in the context of population genetics as the of the microsatellites were determined using an ABI PRISM

89 J Appl Genetics (2013) 54:285–292 287

310 automated sequencer (Applied Biosystems, Foster City, DRA, and microsatellite markers were computed by using CA, USA). the Arlequin 3.1 software (Excoffier et al. 2005).

Single nucleotide polymorphism (SNP) markers in candidate Comparison between colour varieties immunity-related (IR) genes The genetic diversity of both current sub-populations was SNPs in genes coding for the CD14, interleukin 1 beta compared in all types of markers available. Since genetic (IL1B), interleukin 4 (IL4), interleukin 4 receptor (IL4R), differences between them were confirmed, further compar- interleukin 10 (IL10), interleukin 12 p34 sub-unit isons were made within the colour varieties. (IL12p35), interferon gamma (INFG), MxA protein, MYD88, and toll-like receptors 3 and 4 (TLR3, TLR4) were Comparisons between 1997 and 2010 used for the genetic diversity analysis of expressed genes. IBH-associated markers located in the statin-Janus kinase Parameters of genetic diversity computed for each type of (STAT-JAK2) and involucrin (IVL) genes were genotyped marker were compared to values determined 13 years ago only in grey horses. (Horín et al. 1998). The Wilcoxon paired test was used for Horse-specific primers amplifying selected gene regions comparing values of genetic diversity between various groups. were designed. A complete list of primer pairs used in this study is shown in Table S1. The PCR total reaction volume Microsatellites was 12.5 μl with 100 ng of genomic DNA. In general, the PCR mixture comprised 6.25 μl of HotStarTaq Master Mix A direct comparison between the entire grey and black nucle- (Qiagen, Carlsbad, CA, USA) and 25 pmol of each primer. us sub-populations in 1997 and 2010 was possible for 12 The PCR protocol consisted of initial denaturation at 95 °C microsatellites typed in both studies. Genetic diversity param- for 15 min, 35–40 cycles at 94 °C for 20 s, annealing eters based on 12 and 50 microsatellites were used for another temperature (range) for 30 s, 68–72 °C for 90 s, followed comparison between 1997 and 2010, respectively. by final extension at 68–72 °C for 10 min. SNP markers were genotyped using PCR-restriction fragment length Expressed genes polymorphism (RFLP) with appropriate enzymes. The resulting fragments were resolved by using the microchip SNPs were not typed in 1997. Therefore, comparisons be- electrophoresis system for DNA/RNA analysis MCE-202 tween genetic diversity in SNP markers and biochemical MultiNA (Shimadzu, Japan). The CD14/a polymorphism loci were made. Blood group system markers as multi- was tested by PCR-single-strand conformation polymor- allelic systems seemed to be comparable to microsatellites phism (SSCP), as described previously (Vychodilova- rather than to SNPs. Krenkova et al. 2005). MHC DRA MHC ELA class II (DRA) genotyping The results obtained for ELA loci could be compared direct- Based on the sequences available, a simple PCR-RFLP ly. Here, however, additional variation (more DRA alleles) method for DRA genotyping using three enzymes was set could be detected by PCR-RFLP as compared to the previ- up (Vychodilova et al. 2013). ous study (Horín et al. 1998).

Coat colour genes Retrospective genotyping

In addition to the set sampled in 1997, coat colour gene As DNA samples from 31 grey mares from the 1997 study analysis was performed. The genotyping of Agouti, were available, retrospective genotyping of microsatellites, Extension and Grey loci was made based on the original SNPs and ELA-DRA markers allowed direct comparisons of protocols (Marklund et al. 1996;Riederetal.2001; this sub-group with the current grey mare population. This Rosengren Pielberg et al. 2008). was not possible for the black sub-population.

Genetic diversity analysis Allelic frequencies

Gene diversity (Gd), numbers of alleles, observed (Ho) and Besides comparing parameters of genetic diversity, changes expected heterozygosity (He), deviation from the Hardy– in allelic frequencies were tested for specific markers, espe-

Weinberg equilibrium (HWE) and FST values for the SNP, cially those associated with diseases which are important for

90 288 J Appl Genetics (2013) 54:285–292 the population. For this purpose, melanoma and IBH, dis- Table 1 FST values for two sub-populations of Kladruber horses based eases which are important for the grey sub-population, were on microsatellites (set of 12 microsatellites below the diagonal, set of 50 microsatellites above the diagonal) and DRA and SNPs (DRA chosen. Nine microsatellites and five SNPs of IR genes analysis under the diagonal; SNP above the diagonal) associated with these diseases (Futas et al. 2012; Vychodilova et al. 2013) were analysed. The standard Microsatellites Grey 1997 Grey 2010 Black Pearson χ2 test was used for testing changes in allelic Grey 1997 – 0.0002 0.1150 frequencies. Grey 2010 ––0.1040 Black – 0.1110 – Coat colour genes SNPs Grey 1997 Grey 2010 Black Grey 1997 – −0.1930 −0.1000 As coat colour loci are not comparable between colour Grey 2010 0.0060 – 0.1390 varieties, these genes were not included in comparisons of Black 0.0120 0.0330 – diversity. They were analysed separately for assessing the variation underlying this important phenotype. Significant differences (p<0.05) are shown in bold

Results and 2010 are shown in Table 2. Detailed analysis based on 50 markers showed similar values of expected heterozygos- General comparison between colour varieties ities in both current sub-populations (Wilcoxon test z=0.951; p>0.05). However, the observed heterozygosity of black Genetic differences between the grey and black varieties were mares tended to be lower (z=1.185; p<0.05) and the numbers found. The FST values confirmed differences between the of alleles tended to be higher (Wilcoxon test z=1.779; black and grey sub-populations in microsatellites as well as p=0.075) as compared to the grey variety. in SNPs and MHC (Table 1). For further analyses, the two The set of 50 markers showed similar values of observed groups were, therefore, treated as separate sub-populations. (Wilcoxon test z=0.950; p>0.05) and expected heterozygos- Differences between the varieties in coat colour loci ities (z=1.448;p>0.05) in the current grey sub-population and corresponded to the phenotypes. The Grey allele G was pres- in the group of 31 mares from 1997, although the numbers of ent in all grey horses (74 % homozygotes of dominant allele, alleles were higher in 2010 than in 1997 (Wilcoxon test z= 26 % heterozygotes) and was not found in black horses. All 4.026, p<0.05). The new alleles appeared either in the grey black mares were homozygotes for the recessive Agouti allele, group only (15 alleles) or in both colour variants (29 alleles). while the grey mares had all three Agouti genotypes (56 % Altogether, the frequencies of 34 out of 44 new alleles iden- homozygotes of dominant allele, 36 % heterozygotes, 8 % tified in both sub-populations were lower than 0.005 and only homozygotes of recessive allele). Only the dominant E allele for two alleles were they higher than 0.1. Two alleles found in was present in grey horses, while heterozygotes in the 1997 were not retrieved in 2010. Extension locus were observed among black horses (85 % homozygotes of dominant allele, 15 % of heterozygotes). All SNP markers of the coat colour genes tested were in the HWE. The CD14/a marker was monomorphic in both black and Microsatellite diversity grey horses, while the IL1B/a SNP was monomorphic in the black sub-population. In five SNP loci, departure from the All microsatellite loci were polymorphic in this population. HWE was observed. Characteristics of genetic diversity in The comparison between 1997 and 2010 based on 12 loci polymorphic markers of expressed loci and their compari- confirmed that genetic diversity has not diminished; both sons are summarised in Table S3. Similar to microsatellite expected and observed heterozygosities were nearly identical loci, no differences between 1997 and 2010 in the genetic in both sub-populations, and changes in the numbers of alleles diversity of expressed loci were observed (Table 2). Direct concerned mostly rare alleles (Table 2, Table S2). The param- comparisons of horse typed retrospectively for selected IR eters of genetic diversity of the 2010 population for the set of gene SNPs confirmed these findings (Table S3). The ob- 50 markers are summarised in Table S2. Sixty-four alleles at served and expected heterozygosities of SNP markers of 36 microsatellite loci were found only in black mares, while breeding grey mares even increased between 1997 and 31 alleles at 20 loci were identified only in the grey group 2010 (Wilcoxon paired test Ho z=2.482; p<0.05; He z= (Table S2). 2.438; p<0.05). The level of heterozygosity between the In 19 microsatellite loci, departure from the HWE was black and grey sub-populations in 2010 was not different observed. Comparisons between breeding mares of 1997 (Ho z=0.785, p>0.05; He z=0.282, p>0.05).

91 J Appl Genetics (2013) 54:285–292 289

Major histocompatibility complex class II ELA-DRA locus 7 (0.13) 66 (0.14) .64 (0.15) .66 (0.14) The observed and expected heterozygosities between 1997 and 2010 for MHC did not reveal significant loss of variability (Table 2).

IBH and melanoma-associated markers , observed heterozygosity Ho, Changes in allelic frequencies in three markers associated with melanoma were observed (microsatellites TKY343, TLR4/a, TLR4/c), while for other markers associated with melanoma (UM011, HTG004, IL12p35) and for all markers associated with IBH (TKY341, COR069, AHT004, LEX054, STAT-JAK2, IVL), no changes in frequencies were found.

Comparison among different types of markers

The parameters of genetic diversity computed for neutral and expressed loci were similar to each other in multi-allelic systems (Table 2).

Discussion

Only breeding mares were studied, with the numbers of

Black 2010 Grey 1997 Grey 2010 Black 1997 Black 2010 Grey 1997 Grey 2010 Black 1997 Black 2010 stallions being too low to allow statistical analysis. Two sub-populations of Old Kladruber horses were established

1997 based on differences in the hair colour. Following historical bottlenecks, the original population differing primarily only by coat colour genes further differentiated, due mainly to admixtures of different breeds. Genetic differences between the two colour varieties observed in our previous study were, therefore, not surprising. They are accompanied by differences in linear morphological traits (Jakubec et al. DRA MHC(ELA)-DRA 2007), which is in agreement with correlations observed for variation in morphological traits and distances between Black 2010 Grey 1997 Grey 2010 Black microsatellite alleles (Curik et al. 2003). Different types of molecular markers provide different 1997 information on the population analysed. Neutral markers

blood groups loci; like microsatellites inform on genetic structure with little

BG effect of positive and/or negative selection, while the analysis of non-neutral loci may reveal different effects on the levels of variation and population differentiation (Helyar et

GDGrey 1997 Grey 2010 Black MNAal. 2011 Ho). In our previous analysis, we He showed that microsatellite genetic diversity of the small and partially inbred Black 2010 biochemical loci; Old Kladruber population was comparable to other horse

BC breeds, and that genetic distances between the two colour Black 1997 varieties were comparable to differences observed between

Grey 2010 related horse breeds (Horín et al. 1998). Clear genetic differences between the two sub-populations were confirmed Comparison of genetic diversity in Old Kladruber horses assessed by various types of molecular markers (gene diversity GD, mean number of alleles MNA 59 75 66 70 nd 0.63 (0.33) nd 0.64 (0.33) nd 5.33 (1.37) nd 5.25 (0.72) 0.70 (0.14) 0.70 (0.17) 0.69 (0.12) 0.67 (0.16) 0.67 (0.16) 0.65 (0.17) 0.64 (0.17) 0 59 75 66 70 nd 0.61 (0.29) nd 0.62 (0.30) nd 5.52 (1.96) nd 5.94 (1.77) 0.70 (0.14) 0.64 (0.18) 0.69 (0.12) 0.60 (0.17) 0.67 (0.16) 0.63 (0.16) 0.64 (0.17) 0 31 75 059 70 0 0.57 (0.28) 0.61 66 (0.29) nd 70 0.62 (0.30) nd 4.72 (1.50) 5.52 (1.96) nd 0.61 (0.29) 5.94 (1.77) nd 0.66 (0.19) 0.64 0.62 (0.18) (0.30) nd nd 0.60 5.52 (0.17) (1.96) 0.61 nd (0.17) 0.63 (0.16) 5.94 nd (1.77) nd 0. 0.64 (0.18) nd 0.60 (0.17) 0.47 (0.09) 0.63 (0.16) 0.52 (0.09) 0.66 (0.14) n Grey 1997 13 years later by using additional microsatellite markers as microsatellites; well as immunity-related SNP markers. Genetic distances 12/12 12/50 50/50 Msats 50 expected heterozygosity He) Msats Msats Msats SNPBC/SNP 31 59BG/ 75 75 0 66DRA 70 0 31 nd 69 0.11 (0.07) nd 0.28 (0.15) nd 0.28 66 (0.15) nd nd nd 0.36 (0.19) 2 (0) 2 (0) nd 2 (0) 2 nd (2) 2 (0) nd nd 2 (0) nd 0.36 4 (0.11) (0) 0.39 (0.18) 0.34 0.46 (0.18) (0.13) 0.39 4 (0.18) 0.39 (0) (0.17) nd 0.34 (0.09) 0.37 (0.15) nd 0.42 (0.09) 0.3 nd 4 (0) 0.42 0.34 (0.16) 0.37 (0.15) nd 0.48 nd nd 0.59 0.45 0.46 nd 0.52 Table 2 Markers Msats between both sub-populations computed for neutral loci

92 290 J Appl Genetics (2013) 54:285–292 confirmed not only phenotypic (coat colour) but also genetic Fifteen years of conservation and the 13-year interval subdivision of the population. Candidate immunity-related between our two studies represent a rather short period of gene markers showed that this is also the case for function- time. However, allele frequencies can change even from one ally important genetic variation. generation to the next due to possible gametic sampling Taking into consideration the epistatic effect of Grey, errors. It seems that this has happened neither in neutral variation in other coat colour genes is less important for nor in immunity-related loci in Old Kladruber horses. the grey sub-population. However, genetic diversity in coat Similar results were reported by Rendo et al. (2012) for an colour genes is an important issue for the black Kladruber endangered population of Pottoka ponies and its conserva- horses, due to the genetic nature of the black coat colour and tion programme. Having used a set of 17 microsatellites and their higher genetic heterogeneity. The results presented analysing genetic diversity 10 years after the beginning of here showed that, in all of the black horses, the black the conservation programme, they found values for the colouration is caused by homozygosity of the recessive genetic diversity of the breed to be high and stable. Agouti allele a with an 11-bp deletion (Rieder et al. 2001). However, this population is significantly bigger and, there- However, slight differences in the intensity of the black fore, less prone to rapid stochastic changes. colouration observed within this population suggest that It, thus, seems that the breeding policy currently further genes could be involved in the phenotypic variation implemented in the small breeding nucleus of Old of black Kladruber horses. Kladruber breeding mares, split into two even smaller and Despite intensive selective breeding within such a small genetically separated sub-populations, has not caused a de- population, no reduction in genetic diversity over 15 years tectable reduction of genetic diversity. The values of of conservation was observed and the population remained expected and observed heterozygosities for the same set of unchanged in all types of comparable markers during 12 microsatellites remained almost identical in both sub- 13 years of observation. Despite the small population size populations over 13 years. The same result was shown for and a history of bottlenecks and inbreeding, the genetic the whole set of 50 microsatellites. In grey horses, the diversity of the current population of Old Kladruber breed- numbers of alleles increased in time. The low frequencies ing mares assessed by neutral microsatellite loci is compa- of new alleles observed in 2010 indicate that they could rable to the diversity of other domestic horse breeds, ranging have passed undetected during the 1997 study, due to the generally from 0.43 to 0.79 (Aberle et al. 2004; Leroy et al. smaller numbers of horses analysed. The effects of individ- 2009) for different breeds and loci. Local breeds often ual stallions used throughout the period of 13 years and display higher genetic diversity than highly selected differences between sire lines observed in the Kladruber specialised breeds (Aberle et al. 2004; Solis et al. 2005). horses (Vostrý et al. 2011) had no apparent effect on the In Lipizzaner horses, which are closely related to the Old population in the time interval examined. Kladruber breed, the microsatellite diversity is similar to our Although the overall genetic diversity remained values, ranging from 0.66 to 0.67 (Curik et al. 2003; unchanged, specific loci could be affected by various fac- Achmann et al. 2004). Allelic richness was proposed as a tors, including genetic drift and effects of selective breeding. good parameter of genetic diversity and as an indicator of High numbers of microsatellite loci deviated from the HWE, past bottlenecks (Petit et al. 1998; Foulley and Ollivier but no consistent trend in terms of increased or reduced 2006). The mean number of alleles in our population was heterozygosity was observed. In general, deviations could slightly lower than in the Lipizzaner horses (Achmann et al. be explained by effects of inbreeding, selective breeding, 2004), which can be caused by the fact that inbreeding and effects of individual stallions and random effects. Achmann bottlenecks reduce the number of alleles faster than hetero- et al. (2004) and Leroy et al. (2009) observed deviations zygosity (Nei et al. 1975). The higher mean number of from the HWE in microsatellite loci LEX054 and HTG010, alleles in the black sub-population and the high number of which are explained by null alleles. population-specific alleles can be a consequence of admix- Individual genetic variation in expressed, especially non- tures of relatively unrelated Friesian stallions in the 20th neutral, loci does not necessarily correlate with that at micro- century (Vostrý et al. 2011). Changes in allele numbers satellite loci (Bömcke et al. 2011;Helyaretal.2011). The observed in the two sub-populations concerned mainly rare variability at microsatellite loci is affected by the numbers of alleles and they did not influence the overall within-breed alleles and, at the population level, may change more rapidly and between-breed comparisons. Analysis of two sets of in microsatellites than at protein loci (Luís et al. 2007). Only microsatellites (12 vs. 50) confirmed common findings that limited information on genetic variation in immunity-related the overall characteristics of genetic diversity may vary loci is available for domestic horse populations. The genetic according to the markers used. Here, however, this effect diversity of class II major histocompatibility complex loci seems to be rather small, as no major differences between were studied by Luís et al. (2005) and Díaz et al. (2005). In data obtained with the two marker sets were observed. terms of conservation, we observed no substantial differences

93 J Appl Genetics (2013) 54:285–292 291 between the two types of loci. Like in microsatellites, no could be a reasonable addition to the current breeding and reduction of genetic variation in expressed loci, including conservation programme. MHC, was observed in our previous report (Horín et al. 1998). This study confirmed these results and extended them Acknowledgements This study was supported by the project of the to candidate immunity-related gene markers, showing that the National Agency for Agricultural Research of the Ministry of Agriculture of the Czech Republic NAZV QH92277. same is true for potentially functionally important genetic variation. Only one SNP polymorphic in other breeds Conflict of interest The authors declare no conflict of interest. (Vychodilova-Krenkova et al. 2005) was monomorphic in the entire breed (CD14/a), and another one (IL1B/b) was Ethical standards The work performed complied with the current monomorphic in the black sub-population, which is likely to laws and ethical standards of the Czech Republic. be due to general inter-breed differences observed in horses (Leroy et al. 2009), rather than to specifically low variation of the Old Kladruber breed. Further SNPs identified in these two genes were polymorphic in the entire breed. Similar results References were observed for expressed biochemical and blood group loci in the previous study, which suggests that most of the Aberle KS, Hamann H, Drögemüller C, Distl O (2004) Genetic diver- SNPs analysed were neutral, i.e. without important effects on sity in German draught horse breeds compared with a group of fitness. primitive, riding and wild horses by means of microsatellite DNA markers. Anim Genet 35:270–277 No differences between neutral and functionally important Achmann R, Curik I, Dovc P, Kavar T, Bodo I, Habe F, Marti E, Sölkner loci that could be due to selection were observed. Increased J, Brem G (2004) Microsatellite diversity, population subdivision heterozygosity of SNP markers observed in grey breeding and gene flow in the Lipizzan horse. Anim Genet 35:285–292 mares was due probably to the decreasing rate of inbreeding Arcos-Burgos M, Muenke M (2002) Genetics of population isolates. Clin Genet 61:233–247 in the whole breed over time (Jakubec et al. 2007), based on Bömcke E, Gengler N, Cothran EG (2011) Genetic variability in the negative correlations between individual heterozygosity and Skyros pony and its relationship with other Greek and foreign inbreeding coefficient (Aberle et al. 2004). This could also be horse breeds. Genet Mol Biol 34:68–76 Curik I, Zechner P, Sölkner J, Achmann R, Bodo I, Dovc P, Kavar T, the reason for higher FST values observed for SNP markers in Marti E, Brem G (2003) Inbreeding, microsatellite heterozygosity, grey mares of 1997 and 2010. Lower FST values based on and morphological traits in Lipizzan horses. J Hered 94:125–132 potentially non-neutral markers (IR gene SNPs and MHC), Díaz S, Giovambattista G, Peral-García P (2005) Polymorphisms of the indicating smaller genetic distances between the two sub- upstream regulatory region of the major histocompatibility com- populations as compared to neutral microsatellite loci, result plex DRB genes in domestic horses. Int J Immunogenet 32:91–98 Excoffier L, Laval G, Schneider S (2005) Arlequin (version 3.0): an probably from their lower polymorphisms. integrated software package for population genetics data analysis. The deviation from the HWE detected in the IL10, CD14/b, Evol Bioinform Online 1:47–50 TLR4/a and TLR4/c loci is probably caused by the same Foulley J-L, Ollivier L (2006) Estimating allelic richness and its factors as in microsatellite loci. TLR4/a, TLR4/b and TLR4/c diversity. Livest Sci 101:150–158 Frankham R (1995) Conservation genetics. Annu Rev Genet 29:305–327 SNPs were associated with melanoma in the grey sub- Futas J, Vychodilova L, Hofmanova B, Vranova M, Putnova L, Muzik population (Futas et al. 2012). There is, however, no artificial J, Vyskocil M, Vrtkova I, Dusek L, Majzlik I, Horin P (2012) and/or natural selection against this disease manifested mostly Genomic analysis of resistance/susceptibility to melanoma in Old in older horses. In agreement with this situation, only non- Kladruber greying horses. Tissue Antigens 79:247–248 Glowatzki-Mullis ML, Muntwyler J, Pfister W, Marti E, Rieder S, significant changes in TLR4/a allelic frequencies were ob- Poncet PA, Gaillard C (2006) Genetic diversity among horse served. In terms of conservation genetics, disease-associated populations with a special focus on the Franches-Montagnes SNPs could be useful for monitoring endangered populations. breed. Anim Genet 37:33–39 However, despite an intensive phenotypic selection Helyar SJ, Hemmer-Hansen J, Bekkevold D, Taylor MI, Ogden R, Limborg MT, Cariani A, Maes GE, Diopere E, Carvalho GR, programme against IBH launched 4 years ago, allelic frequen- Nielsen EE (2011) Application of SNPs for population genetics cies and HWE in two IBH-associated loci, STAT-JAK2 and of nonmodel organisms: new opportunities and challenges. Mol IVL (Vychodilova et al. 2013), did not change and both loci Ecol Resour 11(Suppl 1):123–136 were in HWE. Although between 1997 and 2010 only 9 % of Horín P, Cothran EG, Trtková K, Marti E, Glasnák V,Henney P, Vyskocil grey mares remained in the herd, it is likely that a longer time M, Lazary S (1998) Polymorphism of Old Kladruber horses, a surviving but endangered baroque breed. Eur J Immunogenet period is needed before consequences of selective breeding 25:357–363 can be detected, even at the molecular level. Jakubec V, Rejfková M, Volenec J, Majzlík I, Vostrý L (2007) Analysis In general, it seems that regular continuous monitoring of of linear description of type traits in the varieties and studs of the – molecular diversity with many markers is not necessary in Old Kladrub horse. Czech J Anim Sci 52:299 307 Jakubec V, Vostrý L, Schlote W, Majzlík I, Mach K (2009) Selection in this population. Nevertheless, long-term monitoring of spe- the genetic resource: genetic variation of the linear described type cific markers, especially those associated with disease, traits in the Old Kladrub horse. Arch Tierzucht 52:343–355

94 292 J Appl Genetics (2013) 54:285–292

Lazarus R, Vercelli D, Palmer LJ, Klimecki WJ, Silverman EK, Richter Rieder S, Taourit S, Mariat D, Langlois B, Guérin G (2001) Mutations B, Riva A, Ramoni M, Martinez FD, Weiss ST, Kwiatkowski DJ in the agouti (ASIP), the extension (MC1R), and the brown (2002) Single nucleotide polymorphisms in innate immunity (TYRP1) loci and their association to coat color phenotypes in genes: abundant variation and potential role in complex human horses (Equus caballus). Mamm Genome 12:450–455 disease. Immunol Rev 190:9–25 Rosengren Pielberg G, Golovko A, Sundström E, Curik I, Lennartsson Leroy G, Callède L, Verrier E, Mériaux JC, Ricard A, Danchin-Burge J, Seltenhammer MH, Druml T, Binns M, Fitzsimmons C, C, Rognon X (2009) Genetic diversity of a large set of horse Lindgren G, Sandberg K, Baumung R, Vetterlein M, Strömberg breeds raised in France assessed by microsatellite polymorphism. S, Grabherr M, Wade C, Lindblad-Toh K, Pontén F, Heldin CH, Genet Sel Evol 41:5 Sölkner J, Andersson L (2008) A cis-acting regulatory mutation Luís C, Cothran EG, Oom MM, Bailey E (2005) Major histocompat- causes premature hair graying and susceptibility to melanoma in ibility complex locus DRA polymorphism in the endangered the horse. Nat Genet 40:1004–1009 Sorraia horse and related breeds. J Anim Breed Genet 122:69–72 Solis A, Jugo BM, Mériaux JC, Iriondo M, Mazón LI, Aguirre AI, Luís C, Juras R, Oom MM, Cothran EG (2007) Genetic diversity and Vicario A, Estomba A (2005) Genetic diversity within and among relationships of Portuguese and other horse breeds based on four south European native horse breeds based on microsatellite protein and microsatellite loci variation. Anim Genet 38:20–27 DNA analysis: implications for conservation. J Hered 96:670–678 Marklund L, Moller MJ, Sandberg K, Andersson L (1996) A missense Thomson RC, Wang IJ, Johnson JR (2010) Genome-enabled develop- mutation in the gene for melanocyte-stimulating hormone recep- ment of DNA markers for ecology, evolution and conservation. tor (MC1R) is associated with the chestnut coat color in horses. Mol Ecol 19:2184–2195 Mamm Genome 7:895–899 Volenec J, Jakubec V, Jelínek J, Přibyl J, Záliš N (1995) Analysis of Nei M, Maruyama T, Chakraborty R (1975) The bottleneck effect and inbreeding of Old Kladrub horses. Sci Agric Bohem 26:279–286 genetic variability in populations. Evolution 29:1–10 Vostrý L, Kracíková O, Hofmanová B, Czerneková V, Kott T, Přibyl J O’Brien SJ (1994) Genetic and phylogenetic analyses of endangered (2011) Intra-line and inter-line genetic diversity in sire lines of the species. Annu Rev Genet 28:467–489 Old Kladruber horse based on microsatellite analysis of DNA. Petit RJ, El Mousadik A, Pons O (1998) Identifying populations for Czech J Anim Sci 56:163–175 conservation on the basis of genetic markers. Conserv Biol Vychodilova L, Matiasovic J, Bobrova O, Futas J, Klumplerova M, 12:844–855 Stejskalova K, Cvanova M, Janova E, Osickova J, Vyskocil M, Porter V (2002) Mason’s world dictionary of livestock breeds, types and Sedlinska M, Dusek L, Marti E, Horin P (2013) Immunogenomic varieties. Revised by Valeria Porter, 5th edn. CABI Publishing, analysis of insect bite hypersensitivity in a model horse popula- Oxon tion. Vet Immunol Immunopathol 152:260–268 Rendo F, Iriondo M, Manzano C, Estonba A (2012) Effects of a 10- Vychodilova-Krenkova L, Matiasovic J, Horin P (2005) Single nucle- year conservation programme on the genetic diversity of the otide polymorphisms in four functionally related immune re- Pottoka pony—new clues regarding their origin. J Anim Breed sponse genes in the horse: CD14, TLR4, Cepsilon, and Genet 129:234–243 Fcepsilon R1 alpha. Int J Immunogenet 32:277–283

95 MS IV Immunogenomic analysis of insect bite hypersensitivity in a model horse population.

Veterinary Immunology and Immunopathology 152 (2013) 260–268

Contents lists available at SciVerse ScienceDirect

Veterinary Immunology and Immunopathology

j ournal homepage: www.elsevier.com/locate/vetimm

Research paper

Immunogenomic analysis of insect bite hypersensitivity in a model

horse population

a b a a,f a,f

Leona Vychodilova , Jan Matiasovic , Olga Bobrova , Jan Futas , Marie Klumplerova ,

a c a a a

Karla Stejskalova , Michaela Cvanova , Eva Janova , Jarmila Osickova , Mirko Vyskocil ,

d c e a,f,∗

Marketa Sedlinska , Ladislav Dusek , Eliane Marti , Petr Horin

Institute of Animal Genetics, Faculty of Veterinary Medicine, University of Veterinary and Pharmaceutical Sciences, Palackého 1/3, 61242 Brno, Czech Republic

Department of Immunology, Veterinary Research Institute, Hudcova 70, 62100 Brno, Czech Republic

Institute of Biostatistics and Analyses, Faculty of Medicine and Faculty of Science, Masaryk University, Kamenice 126/3, 62500 Brno, Czech Republic

Equine Clinic, Faculty of Veterinary Medicine University of Veterinary and Pharmaceutical Sciences, Palackého 1/3, 61242 Brno, Czech Republic

Department of Clinical Research-VPH, Vetsuisse Faculty, University of Berne, Länggassstrasse 124, 3001 Berne, Switzerland

Ceitec VFU, University of Veterinary and Pharmaceutical Sciences, Palackého 1/3, 61242 Brno, Czech Republic

a r t i c l e i n f o a b s t r a c t

Article history:

Equine insect bite hypersensitivity (IBH) is a seasonal IgE-mediated dermatosis caused by

Received 4 June 2012

bites of insects of the genus Culicoides. A familial predisposition for the disease has been

Received in revised form

shown but, except for the MHC, the genes involved have not been identiﬁed so far. An

20 September 2012

immunogenomic analysis of IBH was performed in a model population of Old Kladruby

Accepted 27 December 2012

horses, all living in the same environment. Clinical signs of IBH were used as phenotypic

manifestation of IBH. Furthermore, total serum IgE levels were determined in the sera

Keywords:

of these horses and used as an independent phenotypic marker for the immunogenetic

Insect bite hypersensitivity

Horse analysis. Single nucleotide polymorphisms (SNPs) in candidate immunity-related genes

Single nucleotide polymorphism were used for association analyses. Genotypes composed of two to ﬁve genes encoding

Immunity-related genes interferon gamma – IFNG, transforming growth factor beta 1 – TGFB1, Janus kinase 2 –

JAK2, thymic stromal lymphopoietin – TSLP, and involucrin – IVL were associated with IBH,

indicating a role of the genes in the pathogenesis of IBH. These ﬁndings were supported by

analysis of gene expression in skin biopsies of 15 affected and 15 unaffected horses. Two

markers associated with IBH, IFNG and TGFB1, showed differences in mRNA expression in

skin biopsies from IBH-affected and non-affected horses (p < 0.05). Expression of the gene

coding for the CD14 receptor molecule – CD14 was different in skin biopsies at p < 0.06.

When total IgE levels were treated as binary traits, genotypes of IGHE, ELA-DRA, and IL10/b

were associated with this trait. When treated as a continuous trait, total IgE levels were

associated with genes IGHE, FCER1A, IL4, IL4R, IL10, IL1RA, and JAK2. This ﬁrst report on non-

MHC genes associated with IBH in horses is thus supported by differences in expression of

1. Introduction

Insect bite hypersensitivity (IBH) is an important dis-

∗

Corresponding author at: Institute of Animal Genetics, Faculty of Vet-

ease of the horse (Broström et al., 1987; Björnsdóttir

erinary Medicine, University of Veterinary and Pharmaceutical Sciences,

et al., 2006; Cunningham and Dunkel, 2008; Marti et al.,

Palackého 1/3, 61242 Brno, Czech Republic. Tel.: +420 541 562 292;

2008; Schaffartzik et al., 2012). It is mainly caused by

fax: +420 549 248 841.

E-mail address: [email protected] (P. Horin). IgE-mediated reactions to allergens present in the saliva

L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268 261

of biting insects (Hellberg et al., 2006). The mechanisms of selected candidate gene SNPs with total serum IgE levels

resulting to susceptibility or resistance to IBH have not and insect bite hypersensitivity and their expression in the

been completely clariﬁed yet. IBH is a complex disease with population of Old Kladruby horses.

interplay of environmental and genetic factors. Midges

(genus Culicoides) and possibly black ﬂies (family Simuli- 2. Materials and methods

idae) are the most important groups of insects involved,

although reactions to antigens from black ﬂies might be 2.1. The population and the disease

due to cross-reactivity (Schaffartzik et al., 2012). Clinical

manifestation of IBH is due to an allergic reaction sharing The nucleus population of the gray variety of the Old

some similarities with human atopic dermatitis (AD). IBH Kladruby horse kept always in a single location under the

is also called summer eczema or summer dermatitis. These same conditions allowed a long-term follow-up, includ-

names reﬂect clinical aspects as well as seasonality of the ing clinical diagnostics, skin biopsies and analysis of gene

disease related to the occurrence of the biting insects. expression. Sixty-one breeding mares, 4–22 years old, were

Genetic studies showed that horses from some families available for the entire period of the study (10 years). Clin-

and breeds are more susceptible than others (Lazary et al., ical manifestation of IBH – summer dermatitis – and total

1994). The heritability for severity of summer eczema was IgE levels were selected as phenotypic traits. Occurrence

estimated at 0.30 in Icelandic horses (Eriksson et al., 2008) of clinical manifestations of IBH was recorded by the local

indicating that speciﬁc genes can contribute to disease sus- veterinarian during at least 4 and maximum 10 seasons

ceptibility. In humans, chromosome regions and speciﬁc for individual horses. The intensity of clinical symptoms

genes were found to be associated with allergies in gen- and the year of their onset varied among individual horses.

eral as well as with atopic dermatitis (Hoffjan and Epplen, The mean age of onset was between 2 and 3 years (2.7

2005). years). The regions affected were, in order of appearance,

Several approaches may be used for detecting genes mane, tail, linea alba, udder, head, chest and hind legs. In

contributing to resistance/susceptibility to diseases. these regions, small itching areas and crusts extending pro-

Genome-wide association studies (GWAS) using large gressively were observed. Due to extensive scratching, the

numbers of anonymous SNPs distributed over the genome areas affected were mutilated followed by appearance of

are often used for this purpose. As they do not always allow bleeding wounds. Recurrence of the clinical signs in sum-

identiﬁcation of speciﬁc genes, candidate gene studies can mer rather than quantitative expression was considered to

be a useful tool. As compared to humans, domestic animal be the attribute of susceptibility to IBH. Horses considered

populations bred for a long time are more homogenous as resistant never showed even minor clinical signs of IBH

and more informative, especially for clinically oriented during their life.

association studies with limited numbers of cases avail- Entomological analysis showed occurrence of a vari-

able. Well-characterized small isolated model populations ety of parasitic insect species with Culicoides obsoletus and

can even be valuable models for studies on allergic diseases Odagmia ornata representing species likely to cause IBH in

(Laitinen, 2002). The Old Kladruby horse breed is a more this population (Sarkova, 2005). Altogether, 22 mares were

than 400-year-old population, with well-documented identiﬁed as susceptible to IBH and 39 mares were consid-

breeding records. This population is small, isolated and ered to be resistant to IBH, as they had never had clinical

partially inbred. Its molecular genetic diversity in various signs of the disease during their lifespan. Nineteen stallions

types of markers has been characterized previously (Horin were used as sires of one to nine offspring during the period

et al., 1998). All these horses have always been kept at the analyzed.

same place and always exposed to the same environmental

factors, including biting insects. The occurrence of insect 2.2. Immunoblots on salivary gland extracts of C.

bite hypersensitivity in this population allowed using the nubeculosus

population as an informative model.

Studies in humans have shown that the total serum Immunoblots on salivary gland extracts (SGE) from

IgE levels are inﬂuenced by genetic factors (Meyers et al., C. nubeculosus were performed as described in Hellberg

1991; Weidinger et al., 2010). Some of these genes were et al. (2006) with sera of 14 randomly selected IBH

found to be independent of the speciﬁc IgE response or horses, because of the limited availability of sali-

of the susceptibility to atopy in humans as well as in ani- vary gland extract. Brieﬂy: SDS-PAGE was performed

mals (Delgado et al., 2010; Dizier et al., 1995; Meyers et al., using the discountinuous system according to Laemmli

1991; Weidinger et al., 2010). Various studies have identi- (1970) on preparative 10–20% Tris HCl gels (BioRad,

ﬁed genes associated with total serum IgE levels (reviewed www.bio-rad.com) under denaturing reducing conditions

in Weidinger et al., 2010). In horses the genetic basis of using standard protocols. After electrophoresis, the pro-

total IgE levels has not been investigated so far, but a study teins were electro-transferred to polyvinylidene diﬂuoride

in Lipizzan horses had estimated the heritability of mold (PVDF) membranes. The membranes were dried and cut

speciﬁc IgE at 0.30 (Eder et al., 2001). into strips. Horse sera were diluted 1:6 in washing buffer

Based on the hypothesis that total IgE levels might be (20 mM Tris, 025 M NaCl, 0.125% Tween-20, pH 7.5) and the

◦

inﬂuenced by genetic factors in horses independently of strips incubated at 4 C overnight on a shaker. The following

the IBH status and that some of the candidate genes for day the strips were washed and incubated with an anti-

atopy and the IgE response in general might be the same equine IgE mAb (0.5 ␮g/ml, 3H10, Wilson et al., 2006) for

genes, the aim of this work was to investigate associations 3 h at RT on a shaker. After an additional washing step, the

262 L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268

strips were incubated for 2 h with horseradish peroxidase or newly developed by resequencing of pooled DNA

(HRP) conjugated goat anti-mouse-IgG ab (Sigma–Aldrich, samples from horses of the population under study

www.sigmaaldrich.com) diluted 1:10,000 in washing (usually from ﬁve to ten horses). Horse-speciﬁc

buffer. Upon a ﬁnal washing, the strips were devel- primers were designed using Primer3 software

oped with ECL Western blotting detection reagents (GE (http://fokker.wi.mit.edu/primer3/) on annotated genes

Healthcare Life Sciences, http://www.gelifesciences.com) from the NCBI horse genome sequence assembly EcuCab2.0

according to the manufacturer’s protocol. (http://www.ncbi.nlm.nih.gov/mapview/map search.cgi?

taxid=9796) or genes predicted by BLAST search of

2.3. Determination of total IgE levels in horse sera horse genome for deﬁned human/animal counterparts.

Sequences with double or multiple peaks were aligned

Total serum IgE levels were determined by ELISA and analyzed by the BioEdit software (http://www.

(enzyme linked immunoassay) as described in Wilson mbio.ncsu.edu/BioEdit/bioedit.html) and putative restric-

et al. (2006) and Scharrenberg et al. (2010). Brieﬂy: tion enzyme cleavage sites were assigned by NEBcutter

Immunolon 2HB 96 well plates (Thermo Fisher Scientiﬁc, V2.0 (http://tools.neb.com/NEBcutter2/index.php).

www.thermoﬁsher.com) were coated with monoclonal

anti-IgE 1C12 in 0.05 M carbonate buffer pH 9.6. Follow- 2.4.2. Genotyping

ing blocking with PBS Tween 0.5% BSA, sera (diluted 1:40 In general, the total PCR reaction volume was 12.5 ␮l

in blocking buffer) and standard samples were added with 1.25 ␮l Hot Start Taq 10× buffer, 2.5 mM dNTPs,

to the plate, twofold serially diluted and incubated for 5 pmol primers, 0.5 U Hot Star Taq polymerase (QIAGEN,

1.5 hours (h) at RT. A biotinylated anti-equine IgE mAb Hilden, Germany), 100 ng DNA. The PCR protocol consisted

◦

was added and incubated for 2 h, followed by incubation usually of an initial denaturation at 95 C for 15 min fol-

◦

with extra-avidin alkaline phosphatase (Sigma–Aldrich; lowed by 30–40 cycles of 30 s at 94 C, 30 s at annealing

◦ ◦

http://www.sigmaaldrich.com). The ELISA was developed temperature (mostly 60 C), and extension at 72 C for

◦

with p-nitrophenyl phosphate. Plates were read when 1 min per kilobase, followed by a ﬁnal extension at 68–72 C

the OD 450 nm of the highest standard reached 1.00. As for 10–60 min. The primer sequences and SNP characteris-

standard, duplicate samples of six doubling dilutions of tics are given in Supplementary Table 1. The majority of

a horse serum with a known IgE concentration ranging SNP markers were genotyped by PCR–RFLP. The fragments

from 128 to 4 ng/ml were used. The OD of the test samples were run for 90 min at 300 V in 6% PAGE and silver stained.

was interpolated from the standard curve to give their IgE Enzymes (New England Biolabs, Ipswich, MA, USA) used for

concentration in ng/ml. The concentration of the original PCR–RFLP markers are in Supplementary Table 2. If neces-

test samples was calculated as the average of the values sary, primers amplifying a PCR product of 200–400 bp were

from those dilutions where the concentration fell within designed and SNPs were analyzed by PCR–SSCP.

the standard curve.

2.4.3. MHC genes

2.4. Single nucleotide polymorphisms within candidate Eqca-DRA exon 2 alleles. A RFLP-based genotyping sys-

genes tem of Eqca-DRA alleles was developed based on the

GenBank sequences available. Ampliﬁcation of the 307 bp

2.4.1. SNP markers long product was carried out with standard primers Be3

Forty-six SNP markers located in 29 candidate genes and Be4 (Albright-Fraser et al., 1996). Aliquots of PCR

coding for CD14 receptor (CD14), Collagen, type VI, alpha products digested separately with restriction enzymes

5 (COL6A5), MHC class II DQA (ELA DQA), MHC class II Cac8I, Hpy166II, BsaJI and produced speciﬁc fragments

DRA (ELA DRA), Fc fragment of the IgE receptor, alpha allowing identiﬁcation of alleles Eqca-DRA*0201 (M60100),

chain (FcER1A), interferon-induced guanylate-binding Eqca-DRA*0301 (L47172), Eqca-DRA*0401 (AJ575295) and

protein 1-like (GBP1), histamine receptor H4 (HRH4), Eqca-DRA*0501 (FJ716134). The remaining allele Eqca-

interferon gamma (IFNG), immunoglobulin heavy chain DRA*0101 (L47174) could be identiﬁed by manual editing,

epsilon-like (IGHE), interleukin 1 beta (IL1B), interleukin 1 subtracting sequences of the other alleles previously

receptor antagonist (IL1RN), interleukin 4 (IL4), interleukin identiﬁed by RFLP. The nomenclature suggested by us pre-

4 receptor (IL4R), interleukin 10 (IL10), interleukin 12A viously (Janova et al., 2009) was used for designating Eqca

(IL12A), interleukin 12B (IL12B), interleukin-12 recep- alleles.

tor subunit beta-2 (IL12RB2), interleukin-13 (IL-13), Eqca-DQA exon 2 alleles. Eqca-DQA genotypes were

interleukin-23 subunit p19 (IL23A), involucrin (IVL), Janus determined by PCR–SSCP according to the original method

kinase 1 (JAK1), tyrosine-protein kinase JAK2-like (JAK2), described by Fraser and Bailey (1998).

transforming growth factor beta 1 (TGFB1), transfor-

ming growth factor beta 3 (TGFB3), transforming growth 2.4.4. Data analysis

factor beta receptor I (TGFBR1), transforming growth Clinical disease. 61 mares were analyzed. The analysis

factor, beta receptor II (TGFBR2), tumor necrosis factor was performed in two steps. After an initial scan allowing

alpha (TNFA), toll-like receptor 4 (TLR4), thymic stromal identiﬁcation of individual markers potentially associated

lymphopoietin (TSLP) were used for analyzing the Old with IBH, complex genotypes integrating effects observed

Kladruber population (Supplementary Table 1). Markers in previous analyses were identiﬁed and tested. A standard

were identiﬁed by searching the Broad institute SNP two-sided Fisher’s exact test and Fisher’s exact test based

database (http://www.broad.mit.edu/mammals/horse/) on Monte Carlo estimates (100,000 samples) of exact

L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268 263

signiﬁcance were used to assess associations between indi- Mini Kit columns (Qiagen, Carlsbad, CA, USA) to elimi-

vidual SNP alleles and/or genotypes and manifestation of nate phenol and other inhibitors of reverse transcription.

the disease. Bonferroni corrections for multiple compar- Although this procedure itself avoided the possibility of

isons (numbers of markers analyzed) were made in all tests. genomic DNA being co-ampliﬁed, intron-spanning primers

Odds ratios with 95% conﬁdence intervals were computed were designed, if possible. Only CD14 and Involucrin gene

on the basis of logistic regression analyses. sequences did not allow designing functional intron-

Total IgE levels. 58 mares were available for this analy- spanning RT-PCR primers. RNA was then eluted to 20 ␮l

␮

sis. Based on the sample distribution pattern (one-sample and reverse transcribed by using 10 l RNA oligoT a M-MLV

␮ ␮

t-test), horses with total IgE levels higher than 32 g/ml (Invitrogen) in the ﬁnal 20 l volume. Following incubation

◦

were identiﬁed as outliers (p < 0.01) and excluded. This at 37 C for 1 h 30, cDNA was diluted to the ﬁnal volume

␮

selected group (N = 50) was then analyzed similarly to the 40 l.

complete group including outliers. Total IgE levels were Skin biopsies. Biopsies stored in RNALater were dried

treated in two ways. When the values of total IgE lev- and 10 mg of skin tissue were placed into 1 ml TRIReagent-

els were treated as a continuous trait, effects of different RT (MRC), homogenized with 10 Zirconia/Silica Cat. No.

alleles/genotypes on total IgE levels were tested by the 11079125z 2.3 mm (BioSpec Products, Inc, USA) beads by

␮

standard non-parametric Mann–Whitney U test. For ana- using the MagnaLyzer (Roche, Switzerland). 100 l bro-

lyzing them as a binary trait, cut-off values, which would manisol were added, and the samples were then treated

allow binary coding of total IgE values were sought by an a as described above for PBL.

priori performed association analysis, when signiﬁcantly

associated (p ≤ 0.1) individual markers were selected in

2.4.9. Real-time PCR

univariate logistic regression and used as input variables

Fifteen horses affected with IBH and 15 horses with

in the multivariate logistic regression model. Statistically

no history or clinical signs of IBH at the time of sampling

conclusive associations were detected within the range

were analyzed. Gene expression was measured in tripli-

17–22 ␮g/ml. IBM SPSS For association analyses; IBM SPSS

Advanced Statistics 19 for Windows (Release 19.0.1, IBM

3 pmol of each primer in the ﬁnal volume 3 ␮l was used

Corporation 2010) was used.

for real-time PCR on the LightCycler480 instrument (Roche,

Switzerland). The PCR protocol was as follows: 15 min ini-

2.4.5. Analysis of gene expression for selected markers ◦ ◦ ◦

tial denaturation at 95 C, 45 cycles at 95 C for 15 s, 58 C

Genes associated with IBH at the population level (IFNG, ◦

for 30 s and 72 C for 30 s, followed by PCR product melt-

TGFB1, IVL, TSLP, CD14) were analyzed for expression in

ing temperature analysis (using continuous heating from

peripheral blood leukocytes (PBL) and skin biopsies by real- ◦ ◦

60 C to 95 C). cDNA obtained from stimulated leukocytes

time qPCR in 30 Kladruber horses.

as described above was diluted 10×, 100× and 1000×, and

used for measuring ampliﬁcation efﬁcacy for all primer

2.4.6. Skin biopsies

pairs, the ﬁrst three genes serving as reference genes. The

Skin biopsies were collected by standard procedures

primers for reference genes used for normalization were as

from areas heavily affected with dermatitis in IBH-affected

follows:

horses. Peripheral areas with signs of active dermatitis

EqACTBF GGACCTGACGGACTACCTC

process but not damaged by scratching were biopsied. In

EqACTBR CACGCACGATTTCCCTCTC

non-affected horses, biopsies of intact skin were collected

EqSDHAF GAGGAATGGTCTGGAATACTG

from the same areas. All biopsies were taken in summer. EqSDHAR GCCTCTGCTCCATAAATCG

The samples were placed into RNALater solution (Qiagen, EqHPRTe6F GCTTGCTGGTGAAAAGGACCCCTCG ◦

EqHPRTe9R CACTAATGACACAAACGTGATTCAAATCCCTG

Carlsbad, CA) and stored at 4 C for 24 h in order to allow

the RNALater solution to penetrate into the tissue. After

◦ Analysis of variation in expression of candidate refer-

removing hair, the biopsies were stored at −21 C.

ence genes SDHA, ACTb and HPRT in skin and PBL samples

was done by the geNORM software (Vandesompele et al.,

2.4.7. Stimulation of equine peripheral blood leukocytes

2002). The least variation in expression was observed for

Due to low levels of expression of some genes in

the gene HPRT, which was therefore used as a reference

skin samples, leukocytes stimulated by phorbol and iono-

gene for analysis. Real-time PCR was run as described

mycine were used for assessment of ampliﬁcation efﬁcacy.

above. The primers used for real-time PCR were as follows:

Peripheral blood leukocytes were separated by hemoly-

sis. Leukocyte suspension in the RPMI medium with 10% EqIFNge3F GCCCAAAGCTAACCTGAGGAAGCGG

of fetal calf serum and antibiotics was incubated with EqIFNge4R ATTGCAACGCTCTCCGGCCTCG

◦ EqTGFb1e6F TCCACGAGCCCAAGGGCTACCAC

0.004 mg ionomycine and 0.0002 mg PMA at 37 C for 4 h.

EqTGFb1e6e7R GGGCCAGGACCTTGCTGTACTGC

After centrifugation, leukocytes were re-suspended in 1 ml

EqCD14F TTGCGCAAGCTCACTCACTCGC

TRI-RT solution (MRC, Cincinnati, OH, USA). EqCD14R CGCTCGCCCAGTCCAGGATTGTC

EqTSLPe2e3F2 AGACCGGCCGAGTTGTCTCACCG

EqTSLPe3R2 GGCAATGGCTGTTGAGCGTAGCG)

2.4.8. RNA extraction

EqINVOFV3 GAAGGAAGGGTCTGCAGAGCAGC

Peripheral blood leukocytes. Total RNA from peripheral

EqINVORV3 CTGGCCGGGAGCTGGGACAAATG

blood leukocytes (PBL) was isolated using TRIReagent-

EqJAK2lF GTGGCCTCAGATGTTTGGAGCTTTGG

RT (MRC, UK), which eliminated all genomic DNA from EqJAK2lR TGCCAATCATACGCATAAATTCCGCTGG

RNA extracted, and subsequently puriﬁed on RNEasy

100

264 L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268

Table 1

Composed genotypes associated with clinical summer dermatitis (R resistant, S susceptible) at p < 0.05 (in bold).

Genotype associated Horses Population Frequency Frequency p value p value Odds ratio (CI

tested (N) frequency in R in S corrected 95%)

IFNG/a (36 CG)/TGFB1 61 0.246 0.359 (14/39) 0.045 (1/22) 0.006 0.042 0.085 (0.010–0.701)

(139 AA)

IFNG/a (36 CC or 58 0.172 0.053 (2/38) 0.400 (8/20) 0.002 0.013 12.000 (2.233–64.489)

GG)/IVL/a (261 TC)

IFNG/a (36 CG)/JAK2/b 61 0.525 0.667 (26/39) 0.273 (6/22) 0.004 0.027 0.188 (0.059–0.592)

(463 CT or TT)

IFNG/a (36 CG)/TSLP/b 61 0.295 0.436 (17/39) 0.045 (1/22) 0.001 0.008 0.062 (0.008–0.505)

(130 GC or GG)

TGFB1 (139 AA)/IVL/c 57 0.298 0.421 (16/38) 0.053 (1/19) 0.005 0.034 0.076 (0.009–0.633)

(873/873 or 840/840)

TGFB1 (139 CC or 60 0.550 0.395 (15/38) 0.818 (18/22) 0.003 0.019 6.900 (1.950–24.415)

CA/TSLP/a (217 CC or

CG)

TGFB1 (139 CC or 61 0.328 0.179 (7/39) 0.591 (13/22) 0.002 0.012 6.603 (2.030–21.478)

CA)/TSLP/b (130 CC)

IVL/a (261 CC or 58 0.517 0.658 (25/38) 0.250 (5/20) 0.005 0.037 0.173 (0.051–0.584)

TT)/JAK2/b (463 CT or

TT)/

IVL/c (873/873 or 57 0.632 0.789 (30/38) 0.316 (6/19) 0.001 0.007 0.123 (0.036–0.426)

840/840)/JAK2/b (463

CT or TT)

The results obtained were expressed as ratio between RNA expression. Two markers associated with IBH, IFNG

expression of the gene of interest (goi) and expression of and TGFB1 showed signiﬁcant differences in mRNA expres-

the reference gene (ref, here HPRT) weighed by ampliﬁ- sion in skin biopsies from affected and non-affected horses

cation efﬁcacy (1/POWER(E goi;Ct goi))/(1/POWER(E ref;Ct (Table 4) Non-signiﬁcant (p < 0.060) but yet 1.6 increase in

ref)), (Zelnickova et al., 2008). Differences in gene expres- CD14 mRNA expression was observed in IBH biopsies. No

sion between IBH-affected and non-affected groups or differences in mRNA expression of JAK2 and IVL were found

groups of different genotypes were examined by the in skin biopsies from affected and non-affected horses.

Mann–Whitney test. Gene expression in peripheral blood leukocytes was similar

between IBH-affected and non-affected horses in all genes

3. Results analyzed.

Clinical disease: summer dermatitis. After all corrections, 4. Discussion

signiﬁcant (p < 0.05) associations of composed genotypes

with IBH were found (Table 1). The genes whose SNPs In association analyses, phenotyping of the trait ana-

were associated with summer dermatitis were interferon lyzed is one of the crucial points. Here, horses were

gamma (IFNG), transforming growth factor beta 1 (TGFB1), classiﬁed as resistant or susceptible to clinical IBH based

involucrin (IVL), Janus kinase 2 (JAK2) and thymic stromal on presence or absence of clinical manifestation of IBH

lymphopoietin (TSLP) coding genes. over 4–10 summer seasons. The presence and quantitative

Total IgE levels. No difference in total IgE mean/median expression of IBH can be inﬂuenced by many non-genetic

values between IBH-affected and non-affected horses was factors, such as the degree of exposure to insect bites,

observed (19,361/21,309 vs.17,430/12,510, respectively, seasonal variations, climatic variations between years.

p = 0.633 in the Mann–Whitney U test). Genes associated Therefore, recurrence over seasons was selected as the

with total IgE levels in the multivariate regression analy- main qualitative criterion of susceptibility/resistance to

sis are shown in Table 2. Associations with total IgE levels IBH.

treated as a continuous trait are described in Table 3. Mark- Total IgE levels were determined as an additional phe-

ers within IGHE and IL10 genes were detected as associated notypic marker. In humans, total IgE levels are highly

by both methods used. heritable (h up to 0.8), less inﬂuenced by environmen-

tal factors and correlate with clinical atopic dermatitis

3.1. Immunoblots on salivary gland extracts of C. (Potaczek and Kabesch, 2012). It was shown previously that

nubeculosus total IgE levels are most probably strongly inﬂuenced by the

degree of infection with endoparasites in horses (Hamza

The immunoblots showed that overall, the sera of the et al., 2010). Nonetheless, surprisingly, in one of previous

Kladruby horses with IBH displayed IgE-binding to seven studies in horses of the Icelandic breed, horses affected

different protein bands of the salivary gland extract of with IBH had signiﬁcantly higher total serum IgE levels

similar molecular weights as those described previously than non-affected control horses (Wilson et al., 2006).

(Hellberg et al., 2006), suggesting that they were sensi- This may be due to the fact that IBH-affected Icelandic

tized to similar Culicoides salivary glands allergens as the horses often have very strong Th2-type, IgE-mediated reac-

Icelandic horses living in Switzerland. tions (reviewed in Schaffartzik et al., 2012). We therefore

101

L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268 265

Table 2

IgE level in association analyses – multivariate logistic regression model.

Marker IgE cutoff (␮g/ml) Without outliers (Ntotal = 50) With outliers (Ntotal = 58)

p value OR (95% CI) p value OR (95% CI)

IGHE/b 17.0 0.043 26.955 (1.111; 654.046)

IGHE/a 18.0 0.037 15.793 (1.180; 211.309) 0.030 13.626 (1.283; 144.715)

FCER1A/b 21.5 0.044 4.385 (1.041; 18.465)

22.0 0.014 10.766 (1.609; 72.012)

IL4 18.0 0.056 0.290 (0.081; 1.031)

IL4R 18.0 0.033 0.157 (0.029; 0.861)

19.0 0.025 0.139 (0.025; 0.778)

20.0 0.045 0.190 (0.038; 0.963)

IL4R 20.0 0.008 0.239 (0.082; 0.693)

20.5 0.044 0.378 (0.147; 0.972)

IL10/b 22.0 0.043 0.232 (0.056; 0.958)

IL1RA 17.0 0.037 0.221 (0.054; 0.910)

19.0 0.075 0.230 (0.046; 1.163) 0.039 0.265 (0.075; 0.933)

IL1RA 20.5 0.064 0.263 (0.064; 1.082) 0.047 0.278 (0.079; 0.982)

JAK2/b 17.0 0.040 5.240 (1.077; 25.497) 0.105 2.450 (0.829; 7.242)

18.0 0.048 3.098 (1.011; 9.489) 0.011 4.454 (1.417; 14.002)

JAK2/b 19.0 0.071 7.607 (0.841; 68.816) 0.023 9.814 (1.366; 70.530)

Multiple values for the same marker refer to different alleles/genotypes associated.

Table 3

Markers and genotypes associated with total IgE levels.

Marker Genotype N Total IgE level (ng/ml): median p value

IGHE/b Allele A 11 24,428 0.020

Allele G 105 12,510

IGHE/a TT 51 11,757 0.018

CT 7 24,498

Allele C 7 24,498 0.021

Allele T 109 13,116

ELA-DRA DRA*0201/− 55 16,859 0.013

Others 3 4513

IL10/b Het CG 25 10,400 0.035

CC + GG 33 19,763

COL29a1 Het CT 33 19,763 0.045

CC + TT 25 10,400

Mann–Whitney test; het heterozygote.

Table 4

mRNA expression of selected genes in skin biopsies (S) and peripheral blood leukocytes (B).

Gene p value Fold icrease mean IBH Mean Median SE SD Variance Tissue

TGFB1 0.0121 1.2736 + 0.3985 0.3855 0.0283 0.1097 0.0120 S

− 0.3129 0.3007 0.0153 0.0595 0.0035 S

0.4186 + 0.3985 1.6195 0.0283 0.1097 0.0120 B

− 1.8290 1.3777 0.4090 1.5839 2.5089 B

CD14 0.0591 1.582 + 1.1432 1.0059 0.1570 0.6081 0.3698 S

− 0.7227 0.7121 0.0622 0.2410 0.0580 S

0.4186 + 1.6038 1.6650 0.1837 0.7117 0.5065 B

− 1.4381 1.1989 0.1926 0.7459 0.5564 B

IFNG 0.0020 0.0589 + 0.0582 0.0561 0.0047 0.0183 0.0003 S

− 0.0988 0.0785 0.0115 0.0446 0.0020 S

0.8846 + 0.0690 0.0545 0.0093 0.0361 0.0013 B

− 0.0688 0.0632 0.0091 0.0352 0.0012 B

JAK2 0.5476 1.0441 + 0.8820 0.9049 0.0502 0.1945 0.0378 S

−

0.8447 0.8090 0.0529 0.2049 0.0420 S

0.3952 + 0.6963 0.5875 0.1033 0.4002 0.1602 B

− 0.6100 0.4379 0.1148 0.4444 0.1975 B

IVL 0.0929 2.4470 + 1.5032 0.8825 0.4962 1.9216 3.6927 S

− 0.6143 0.6000 0.0568 0.2200 0.0484 S

TSLP 0.093 0.841 + 0.6726 0.5374 0.0703 0.2724 0.0742 S

− 0.7996 0.7708 0.0593 0.2300 0.0528 S

NT NT + 0.2646 0.2348 0.0424 0.1641 0.0269 B

− 0.2577 0.2090 0.0330 0.1277 0.0163 B

Signiﬁcant p values in bold.

Expression in peripheral blood leukocytes not anticipated.

102

266 L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268

investigated in this study whether total serum IgE levels keratinocytes. TSLP was shown to have a wide range of

could be associated with IBH in horses of another breed. effects on cells of the immune system and on allergic

The Kladruby horses were used for such a study because inﬂammation (Roan et al., 2012). Associations between

they live in the same stud, i.e. under similar environmen- the TSLP-receptor encoding gene and atopic dermatitis

tal conditions. No differences in total IgE levels between were identiﬁed in several dog breeds (Wood et al., 2010).

IBH-affected and non-affected horses were observed. The Associations of atopic disease with CD14 polymorphisms

association study showed that the IgE response and atopy (Litonjua et al., 2005), supported by functional studies

behave as two different phenotypes, underlain by two dif- (Sümegi et al., 2007) were found in humans. Besides IR

ferent gene sets in the Kladruber horses. markers, genes related to skin structure and function were

All candidate genes were selected based on their role also associated with human AD (Boguniewicz and Leung,

in immune responses and allergic reactions. Small, isolated 2011). Involucrin (IVL) is a specialized cytoplasmatic pro-

and genetically deﬁned populations were used successfully tein of the keratinocytes, major cells responsible for the

for analysis of human allergic disease. Major advantages structure of epidermis (Eckert and Green, 1986).

of using such speciﬁc populations are a lower genetic RNA expression analysis showed down-regulation of

variation, a higher predictability and more information interferon gamma production and up-regulation of TGFB1,

on disease phenotype and environmental factors available IVL and CD14 gene activity in the affected skin, simi-

(Laitinen, 2002). Candidate gene studies of canine atopic larly to human atopic dermatitis. The differences observed

dermatitis revealed a need of breed and location speciﬁc for IVL and CD14 were not signiﬁcant, but the p values

analyses to increase the likelihood of ﬁnding associations ranged between 0.05 and 0.10. No signiﬁcant differences

with the disease (Wood et al., 2010). The Old Kladruber in TSLP expression were found in biopsies from affected

population is a small but deﬁnite and genetically homoge- horses compared to controls, while it was shown to be up-

nous group (Horin et al., 1998). Its small size and weak regulated in AD lesion of human patients. Although not

effects of individual genes did not allow a robust statistical signiﬁcant (p < 0.1), expression of TSLP was lower in skin

identiﬁcation of single marker associations. After Bonfer- biopsies from IBH-affected compared to unaffected horses.

roni corrections, no individual marker was associated with The reason for this difference veriﬁed in two repeated

IBH at p < 0.05 (data not shown). However, combinations tests is not clear – perhaps it could result from hetero-

of genes associated with resistance/susceptibility to IBH geneity in TSLP expression in different parts of atopic skin

in composed genotypes led to increased signiﬁcance and regions or may be explained by the stage of the disease

strength of associations between composed genotypes and when biopsies were taken: biopsies were possibly taken

the disease, which may be interpreted in terms of a synergic too long after insect bites had occurred and TSLP was no

(additive) action of the genes involved. Complex genotypes longer increased. The timing of biopsy collection is prob-

thus allowed identiﬁcation of genes with weak individ- ably a major limitation of this kind of studies. However,

ual effects. Sometimes, different SNPs located within the signiﬁcant differences in gene expression between IBH skin

same genes were involved in composed genotypes associ- lesion and skin from control horses were identiﬁed in a

ated with IBH. This probably reﬂects variation in numbers similar study in horses (Heimann et al., 2011). Transcrip-

of corresponding genotypes present in the population, low tion of TGFB1 and TGFB3 was shown to be enhanced in

numbers compromising statistical signiﬁcance. the regenerating epidermis and dermis in humans and

The Kladruby population is stratiﬁed due to inbreed- pigs (Cox, 1995). Decreased expression of IL-10 and TGF-

␤

ing. In all cases, only results withstanding conservative 1 molecules was found to be involved in IBH in horses

Bonferroni corrections reﬂecting the numbers of markers (Hamza et al., 2008). Here, we found no direct evidence for

studied were considered to minimize the risk of false pos- involvement of TGFB3 in IBH. However, a strong association

itive results. No effect of age was observed. Similar results of IBH with the microsatellite marker AHT004 closely linked

with lower p values due probably to smaller numbers to TGFB3 on the horse chromosome 24 was observed in this

were obtained by analyzing age-matched groups (data not group (data not shown). Moreover, TGFB3 was associated

shown). The data obtained are supported by similar obser- with IBH in our study of Icelandic horses (Vychodilova et al.,

vations made by us in a genetically distinct population submitted).

of Icelandic horses where composed genotypes involv- Other genes were associated with total IgE levels,

ing the same TSLP and CD14 SNPs were associated with reﬂecting complex relationships between total IgE levels

IBH (Klumplerova et al., 2012). Moreover, markers biolog- and clinical disease (Raap et al., 2012). In theory, genes

ically related to IBH were identiﬁed, while different, but associated with IgE levels may affect total antibody levels,

again biologically plausible markers showed associations total IgE levels as well as levels of speciﬁc IgE. Associations

with other phenotypes in the same group of mares with of IgE heavy chain (IGHE) SNP markers with total IgE levels

the same marker panel – total IgE levels (this study), and were identiﬁed by both statistical approaches used. SNPs in

melanoma (Futas et al., 2012). Results of mRNA expres- this gene seem to inﬂuence IgE levels, but with no associa-

sion analysis and data from human and mouse studies are tion with clinical IBH. A variety of other immunity-related

in agreement with the assumption of biological plausibil- genes may affect IgE antibody responses with or without

ity of the associations observed. Immunity-related genes associations with clinical disease (Choi et al., 2012). Asso-

associated with IBH in this study were reported to be associations of total IgE levels with the FcER1A gene correspond

ciated with atopic dermatitis in humans (Arkwright et al., to recent ﬁndings in humans (Potaczek and Kabesch, 2012).

2001; Nomura et al., 2003; Litonjua et al., 2005). Similarly Curik et al. (2003) found associations between ELA and lev-

to interferon gamma, JAK kinases are also expressed in els of IgE antibodies against mold antigens.

103

L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268 267

Although they cannot reveal novel genes involved, Albright-Fraser, D.G., Reid, R., Gerber, V., Bailey, E., 1996. Polymorphism

of DRA among equids. Immunogenetics 43, 315–317.

studies of individual candidate loci can be useful when

Arkwright, P.D., Chase, M.J., Babbage, S., Pravica, D., David, T.J., Hutchin-

genome-wide studies produce non-informative results or

son, I.V., 2001. Atopic dermatitis is associated with a low producer

for a more precise deﬁnition of mechanisms of the disease transforming growth factor beta (1) cytokine genotype. J. Allergy Clin.

Immunol. 108, 281–284.

studied (Akey, 2009). This was also conﬁrmed for equine

Björnsdóttir, S., Sigvaldadóttir, J., Broström, H., Langvad, B., Sigurdsson,

IBH by Andersson et al. (2012). They observed associations

A., 2006. Summer eczema in exported Icelandic horses: inﬂuence of

between MHC class II region markers and IBH. Here, we environmental and genetic factors. Acta Vet. Scand. 26, 48–53.

Boguniewicz, M., Leung, D.Y., 2011. Atopic dermatitis: a disease of altered

found associations only with total IgE levels due probably

skin barrier and immune dysregulation. Immunol. Rev. 242, 233–246.

to limited polymorphism of the single DRA marker we used.

Broström, H., Larsson, A., Troedsson, M., 1987. Allergic dermatitis (sweet

Taken together, our results identiﬁed genes associ- itch) of Icelandic horses in Sweden: an epidemiological study. Equine

ated with the equine IBH and IgE levels similar to those Vet. J. 19, 229–236.

Choi, W.A., Kang, M.J., Kim, Y.J., Seo, J.H., Kim, H.Y., Kwon, J.W., Yu, J., Park,

reported in human atopic dermatitis. We showed differ-

S.J., Lee, Y.C., Hong, S.J., 2012. Gene-gene interactions between can-

ences in expression of some of the IBH-associated genes in

didate gene polymorphisms are associated with total IgE levels in

the affected and non-affected skin which suggests poten- Korean children with asthma. J. Asthma 49, 243–252.

Cox, D.A., 1995. Transforming growth factor-beta3. Cell Biol. Int. 19,

tial functional importance of the statistical associations 357–371.

observed.

Cunningham, F.M., Dunkel, B., 2008. Equine recurrent airway obstruc-

In general, the results of statistical analyses conﬁrmed tion and insect bite hypersensitivity: understanding the diseases and

uncovering possible new therapeutic approaches. Vet. J. 177, 334–344.

by gene expression data and by similar observations in a

Curik, I., Fraser, D., Eder, C., Achmann, R., Swinburne, J., Crameri, R., Brem,

genetically distinct population suggest that the role of the

G., Sölkner, J., Marti, E., 2003. Association between the MHC gene

genes involved merit to be further investigated. In speciﬁc region and variation of serum IgE levels against speciﬁc mould aller-

gens in the horse. Genet. Sel. Evol. 35 (S1), 177–190.

terms, the results valid for this deﬁnite endangered pop-

Delgado, C., Lee-Fowler, T.M., DeClue, A.E., Reinero, C.R., 2010. Feline-

ulation can be used for designing a conservation program

speciﬁc serum total IgE quantitation in normal, asthmatic and

aiming to reduce the incidence of IBH and to maintain at parasitized cats. J. Feline Med. Surg. 12, 991–994.

Dizier, M.H., Hill, M., James, A., Faux, J., Ryan, G., le Souef, P., Lathrop, M.,

the same time its diversity and speciﬁcity.

Musk, A.W., Demenais, F., Cookson, W., 1995. Detection of a reces-

sive major gene for high IgE levels acting independently of speciﬁc

Conﬂict of interest statement response to allergens. Genet. Epidemiol. 12, 93–105.

Eckert, R.L., Green, H., 1986. Structure and evolution of the human involu-

crin gene. Cell 46, 583–589.

We disclose any ﬁnancial and personal relationships

Eder, C., Curik, I., Brem, G., Crameri, R., Bodo, I., Habe, F., Lazary, S., Sölkner,

with other people or organizations that could inappropri-

J., Marti, E., 2001. Inﬂuence of environmental and genetic factors

ately inﬂuence their work. on allergen-speciﬁc immunoglobulin-E levels in sera from Lipizzan

horses. Equine Vet. J. 33, 714–720.

Eriksson, S., Grandinson, K., Fikse, W.F., Lindberg, L., Mikko, S., Broström,

Acknowledgments

H., Frey, R., Sundquist, M., Lindgren, G., 2008. Genetic analysis of insect

bite hypersensitivity (summer eczema) in Icelandic horses. Animal 2,

360–365.

The authors wish to thank Iveta Krizova, DVM from the

Fraser, D.G., Bailey, E., 1998. Polymorphism and multiple loci for the horse

Kladruby National Stud for providing long-term records

DQA gene. Immunogenetics 47, 487–490.

on clinical manifestations of IBH. The work was sup- Futas, J., Vychodilova, L., Hofmanova, B., Vranova, M., Putnova, L., Muzik, J.,

Vyskocil, M., Vrtkova, I., Dusek, L., Majzlik, I., Horin, P., 2012. Genomic

ported by the Grant Agency of the Czech Republic

analysis of resistance/susceptibility to melanoma in Old Kladruber

projects 523/06/1402 and 524/09/1939, by IGA VFU project

greying horses. Tissue Antigens 79, 247–248.

22/05/FVL, and by the Swiss National Science Founda- Hamza, E., Wagner, B., Jungi, T.W., Mirkovitch, J., Marti, E., 2008. Reduced

incidence of insect-bite hypersensitivity in Icelandic horses is associ-

tion grant No. 310030 129837/1. JM was supported by the

ated with a down-regulation of interleukin-4 by interleukin-10 and

projects MZE0002716202 of the Czech Ministry of Agri-

transforming growth factor-␤1. Vet. Immunol. Immunopathol. 122,

culture and AdmireVet project CZ.1.05/2.1.00/01.0006 – 65–75.

ED0006/01/01 from the Czech Ministry of Education. The Hamza, E., Torsteinsdottir, S., Eydal, M., Frey, C., Mirkovitch, J., Brcic, M.,

Wagner, B., Wilson, A.D., Jungi, T.W., Marti, E., 2010. Increased IL-4

authors would also like to thank Dr. V. Gerber and J.

and decreased regulatory cytokines production following relocation

Klukowska-Rötzler for sharing their IL4R primers and PCR

of Icelandic horses from a high to low endoparasite environment. Vet.

protocol, and J. Janda for the equine TSLP sequence and SNP Immunol. Immunopathol. 133, 40–50.

information. Heimann, M., Janda, J., Sigurdardottir, O.G., Svansson, V., Klukowska, J., von

Tscharner, C., Doherr, M., Broström, H., Andersson, L.S., Einarsson, S.,

Marti, E., Torsteinsdottir, S., 2011. Skin-inﬁltrating T cells and cytokine

Appendix A. Supplementary data expression in Icelandic horses affected with insect bite hypersensitiv-

ity: a possible role for regulatory T cells. Vet. Immunol. Immunopathol.

140, 63–74.

Supplementary data associated with this article can be

Hellberg, W., Wilson, A.D., Mellor, P., Doherr, M.G., Torsteinsdottir, S.,

found, in the online version, at http://dx.doi.org/10.1016/j. Zurbriggen, A., Jungi, T., Marti, E., 2006. Equine insect bite hyper-

vetimm.2012.12.013. sensitivity: immunoblot analysis of IgE and IgG subclass responses

to Culicoides nubeculosus salivary gland extract. Vet. Immunol.

Immunopathol. 113, 99–112.

References Hoffjan, S., Epplen, J.T., 2005. The genetics of atopic dermatitis: recent

ﬁndings and future options. J. Mol. Med. 83, 682–692.

Horin, P., Cothran, E.G., Trtkova, K., Marti, E., Glasnak, V., Henney, P.,

Akey, J., 2009. Constructing genomic maps of positive selection in humans:

Vyskocil, M., Lazary, S., 1998. Polymorphism of Old Kladruby horses,

where do we go from here? Genome Res. 19, 711–722.

a surviving but endangered baroque breed. Eur. J. Immunogenet. 25,

Andersson, L.S., Swinburne, J.E., Meadows, J.R., Broström, H., Eriksson, S.,

357–363.

Fikse, W.F., Frey, R., Sundquist, M., Tseng, C.T., Mikko, S., Lindgren, G.,

Janova, E., Matiasovic, J., Vahala, J., Vodicka, R., Van Dyk, E., Horin, P.,

2012. The same ELA class II risk factors confer equine insect bite hyper-

2009. Polymorphism and selection in the major histocompatibility

sensitivity in two distinct populations. Immunogenetics 64, 201–208.

104

268 L. Vychodilova et al. / Veterinary Immunology and Immunopathology 152 (2013) 260–268

complex DRA and DQA genes in the family Equidae. Immunogenetics Roan, F., Bell, B.D., Stoklasek, T.A., Kitajima, M., Han, H., Ziegler,

61, 513–527. S.F., 2012. The multiple facets of thymic stromal lymphopoietin

Klumplerova, M., Vychodilova, L., Bobrova, O., Cvanova, M., Futas, J., (TSLP) during allergic inﬂammation and beyond. J. Leukoc. Biol.,

Janova, E., Vyskocil, M., Vrtkova, I., Putnova, L., Dusek, L., Marti, http://dx.doi.org/10.1189/jlb.1211622.

E., Horin, P., 2012. Major histocompatibility complex and other Sarkova, H., 2005. Ektopraziti koni v Kladrubech n. L. vyvolavajici pruritus.

allergy-related candidate genes associated with insect bite hyper- (Ectoparasites provoking pruritus in the horses in Kladruby n. Labem).

sensitivity in Icelandic horses. Mol. Biol. Rep., http://dx.doi.org/10. Thesis. Czech Agricultural University Prague.

1007/s11033-012-2408-z. Schaffartzik, A., Hamza, E., Janda, J., Crameri, R., Marti, E., Rhyner, C., 2012.

Laitinen, T., 2002. The value of isolated populations in genetic studies of Equine insect bite hypersensitivity: what do we know? Vet. Immunol.

allergic diseases. Curr. Opin. Allergy Clin. Immunol. 2, 379–382. Immunopathol. 147, 113–126.

Lazary, S., Marti, E., Szalai, G., Gaillard, C., Gerber, H., 1994. Stud- Scharrenberg, A., Gerber, V., Swinburne, J.E., Wilson, A.D., Klukowska-

ies on the frequency and associations of equine leucocyte Rotzler, J., Laumen, E., Marti, E., 2010. IgE, IgGa, IgGb and

antigens in sarcoid and summer dermatitis. Anim. Genet. 25, IgG(T) serum antibody levels in offspring of two sires affected

75–80. with equine recurrent airway obstruction. Anim. Genet. 41 (S2)

Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly , 131–137.

of the head of bacteriophage T4. Nature 227, 680–685. Sümegi, A., Szegedi, A., Gal, M., Hunyadi, J., Szegedi, G., Antal-Szalmas, P.,

Litonjua, A.A., Belanger, K., Celedon, J.C., Milton, D.K., Bracken, M.B., Kraft, 2007. Analysis of components of the CD14/TLR system on leukocytes

P., Triche, E.W., Sredl, D.L., Weiss, S.T., Leaderer, B.P., Gold, D.R., of patients with atopic dermatitis. Int. Arch. Allergy Immunol. 143,

2005. Polymorphisms in the 5’ region of the CD14 gene are associ- 177–184.

ated with eczema in young children. J. Allergy Clin. Immunol 115, Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy,

1056–1062. N., De Paepe, A., Speleman, F., 2002. Accurate normalization of

Marti, E., Gerber, V., Wilson, A.D., Lavoie, J.P., Horohov, D., Crameri, R., real-time quantitative RT-PCR data by geometric averaging of mul-

Lunn, D.P., Antczak, D., Björnsdottir, S., Björnsdottir, T.S., Cunningham, tiple internal control genes. Genome Biol. 3, http://dx.doi.org/

F., Derer, M., Frey, R., Hamza, E., Horin, P., Heimann, M., Kolm-Stark, 10.1186/gb-2002-3-7-research0034 (article number 0034.1).

G., Olafsdottir, G., Ramery, E., Russell, C., Schaffartzik, A., Svansson, Weidinger, S., Baurecht, H., Naumann, A., Novak, N., 2010. Genome-

V., Torsteinsdottir, S., Wagner, B., 2008. Report of the 3rd Havemeyer wide association studies on IgE regulation: are genetics of IgE also

Workshop on Allergic Diseases of the Horse, Holar, Iceland, June 2007. genetics of atopic disease? Curr. Opin. Allergy Clin. Immunol. 10,

Vet. Immunol. Immunopathol. 126, 351–361. 408–417.

Meyers, D.A., Beaty, T.H., Colyer, C.R., Marsh, D.G., 1991. Genetics of total Wilson, A.D., Harwood, L., Torsteinsdottir, S., Marti, E., 2006. Production

serum IgE levels: a regressive model approach to segregation analysis. of monoclonal antibodies speciﬁc for native equine IgE and their

Genet. Epidemiol. 8, 351–359. application to monitor total serum IgE responses in Icelandic and

Nomura, I., Goleva, E., Howell, M.D., Hamid, Q.A., Ong, P.Y., Hall, C.F., non-Icelandic horses with insect bite dermal hypersensitivity. Vet.

Darst, M.A., Gao, B., Boguniewicz, M., Travers, J.B., Leung, D.Y., 2003. Immunol. Immunopathol. 112, 156–170.

Cytokine milieu of atopic dermatitis, as compared to psoriasis, skin Wood, S.H., Ollier, W.E., Nuttall, T., McEwan, N.A., Carter, S.D., 2010.

prevents induction of innate immune response genes. J. Immunol. 171, Despite identifying some shared gene associations with human atopic

3262–3269. dermatitis the use of multiple dog breeds from various locations lim-

Potaczek, D.P., Kabesch, M., 2012. Current concepts of IgE regulation and its detection of gene associations in canine atopic dermatitis. Vet.

impact of genetic determinants. Clin. Exp. Allergy 42, 852–871. Immunol. Immunopathol. 138, 193–197.

Raap, U., Weißmantel, S., Gehring, M., Eisenberg, A.M., Kapp, A., Fölster- Zelnickova, P., Matiasovic, J., Pavlova, B., Kudlackova, H., Kovaru, F., Fal-

Holst, R., 2012. IL-31 signiﬁcantly correlates with disease activity and dyna, M., 2008. Quantitative nitric oxide production by rat, bovine

Th2 cytokine levels in children with atopic dermatitis. Pediatr. Allergy and porcine macrophages. Nitric Oxide 19, 36–41.

Immunol. 23, 285–288.

105 MS V Major histocompatibility complex and other allergy-related candidate genes associated with insect bite hypersensitivity in Icelandic horses.

106 Mol Biol Rep (2013) 40:3333–3340 DOI 10.1007/s11033-012-2408-z

Major histocompatibility complex and other allergy-related candidate genes associated with insect bite hypersensitivity in Icelandic horses

Marie Klumplerova • Leona Vychodilova • Olga Bobrova • Michaela Cvanova • Jan Futas • Eva Janova • Mirko Vyskocil • Irena Vrtkova • Lenka Putnova • Ladislav Dusek • Eliane Marti • Petr Horin

Received: 6 June 2012 / Accepted: 18 December 2012 / Published online: 30 December 2012 Ó Springer Science+Business Media Dordrecht 2012

Abstract Insect bite hypersensitivity (IBH) is an allergic and exon 2 polymorphisms of the class II Eqca-DRA gene. dermatitis of horses caused by bites of insects. IBH is a Associations with Eqca-DRA and COR113 were identified multifactorial disease with contribution of genetic and (p \ 0.05). In addition, a panel of 20 single nucleotide environmental factors. Candidate gene association analysis polymorphisms (SNPs) in 17 candidate allergy-related genes of IBH was performed in a group of 89 Icelandic horses all was tested. During the initial screen, no marker from the born in Iceland and imported to Europe. Horses were clas- panel was significantly (p \ 0.05) associated with IBH. Five sified in IBH-affected and non-affected based on clinical SNPs associated with IBH at p \ 0.10 were therefore used signs and history of recurrent dermatitis, and on the results for analysis of combined genotypes. Out of them, SNPs of an in vitro sulfidoleukotriene (sLT)-release assay with located in the genes coding for the CD14 receptor (CD14), Culicoides nubeculosus and Simulium vittatum extract. interleukin 23 receptor (IL23R), thymic stromal lympho- Different genetic markers were tested for association with poietin (TSLP) and transforming growth factor beta 3 IBH by the Fisher’s exact test. The effect of the major his- (TGFB3) molecules were associated with IBH as parts of tocompatibility complex (MHC) gene region was studied by complex genotypes. These results are supported by similar genotyping five microsatellites spanning the MHC region associations and by expression data from different horse (COR112, COR113, COR114, UM011 and UMN-JH34-2), populations and from human studies.

Keywords Horse Insect bite hypersensitivity Major M. Klumplerova L. Vychodilova O. Bobrova J. Futas E. Janova M. Vyskocil P. Horin histocompatibility complex Association analysis Institute of Animal Genetics, Faculty of Veterinary Medicine, University of Veterinary and Pharmaceutical Sciences, Palackeho tr. 1/3, 61242 Brno, Czech Republic Introduction M. Klumplerova P. Horin (&) Ceitec VFU, University of Veterinary and Pharmaceutical Insect bite hypersensitivity (IBH) is an important disease of Sciences, Palackeho tr. 1/3, 61242 Brno, Czech Republic the horse [1–3], caused by IgE-mediated reactions against e-mail: [email protected] salivary proteins from midges (genus Culicoides) and M. Cvanova L. Dusek sometimes black ﬂies (Simulium spp.). The mechanisms Institute of Biostatistics and Analyses, Masaryk University, resulting in susceptibility or resistance to IBH still remain Brno, Czech Republic unclear. IBH is the result of a complex interplay of environmental and genetic factors, as illustrated by an inter- I. Vrtkova L. Putnova Laboratory of Agrigenomics, Mendel University, Brno, esting phenomenon. Due to the absence of Culicoides spp., Czech Republic IBH does not occur in Iceland. However, when Icelandic horses are imported to continental Europe and get exposed E. Marti to Culicoides, they develop IBH with a much higher Department of Clinical Research-VPH, Vetsuisse Faculty, University of Berne, La¨nggass-strasse 124, 3001 Berne, prevalence (50 %) than horses of this breed born in the Switzerland same environment (3–10 %). IBH shares some similarities 123

107 3334 Mol Biol Rep (2013) 40:3333–3340 with human atopic dermatitis [3]. In vitro degranulation of The non-affected horses were selected randomly from the basophils stimulated with Culicoides allergens and deter- same stables where IBH-affected horses were maintained. mination of released sulfidoleukotrienes (sLT) can be used Clinical manifestation of IBH was assessed as described as an in vitro diagnostic test for IBH [4]. previously [4], based on clinical signs and history of The fact that many horses exposed to Culicoides remain recurrent dermatitis during summer. Furthermore, an healthy under the same environmental conditions suggests in vitro sLT release test was used to confirm the diagnosis of that genetic factors can influence susceptibility to IBH. In IBH and to exclude horses that were sensitized to Culicoides recent studies, the heritability for IBH was estimated to allergens but had not (yet) shown clinical signs from the range from 0.07 to 0.30 [5, 6]. Earlier studies have shown control group. All horses were sampled and tested during the that horses from some families are more susceptible than summer season. The in vitro sLT release test was performed others indicating that specific genes can contribute to dis- as described by Baselgia et al. [4]. Briefly: peripheral blood ease susceptibility [7]. Major histocompatibility complex leukocytes were isolated and incubated with Concanavalin (MHC) class II genes were reported to be associated with A as positive control, Culicoides nubeculosus and Simulium IBH in different breeds [8, 9]. In humans, further chro- vittatum extracts and with buffer alone to determine the mosome regions and specific genes were found to be spontaneous sLT release. After 40 min incubation at 37 °C, associated with allergies in general as well as with atopic supernatants were collected and released sLTs determined dermatitis [10]. In a group of horses belonging to another by ELISA in the cellular allergen stimulation test (CAST, breed and living in a different environment, we have Bu¨hlmann laboratories, http://www.buhlmannlabs.ch/) fol- identified genes coding for interferon gamma (IFNc), trans- lowing the manufacturers instructions. The cut-off descri- forming growth factor beta 1 (TGFb1), involucrin (Ivl), Janus bed in Baselgia et al. [4] was used to classify the results as kinase 2 (JAK2), thymic stromal lymphopoietin (TSLP) and positive or negative test result. Only horses with a negative CD14 lymphocyte receptor (CD14) either associated with test result with C. nubeculosus (‘‘CAST-Cul’’) were selected IBH or differentially expressed in the skin, or both [Vycho- as control horses (n = 43) and all horses included in the dilova et al., Vet Immunol Immunopathol accepted]. IBH-affected group (n = 46) had a positive test result with Several approaches may be used for detecting genes this allergen. We know from previous studies that the CAST contributing to resistance/susceptibility to diseases. Gen- with S. vittatum has a lower sensitivity and specificity for ome-wide association studies (GWAS) using large numbers diagnosis of IBH. These results (‘‘CAST-Sim’’) were thus of anonymous SNPs distributed over the genome are often not used to confirm the diagnosis of IBH but were used as an used for this purpose [11]. As they do not always allow additional phenotype. 42 out of the 46 IBH-affected horses identification of specific genes, candidate gene studies can be gave a positive ‘‘CAST-Sim‘‘result and seven out of the 43 a useful complementary tool. As compared to humans, non affected horses were positive with this extract. There domestic animal populations bred for a long time are more were thus in total 49 horses with a positive and 40 horses homogenous and more informative, especially for clinically with a negative ‘‘CAST-Sim’’ test results, respectively. oriented association studies, but with limited numbers of cases available. Icelandic horses exported to Europe represent an interesting model of IBH [2]. Major histocompatibility complex markers The aim of this work was to investigate associations of the MHC region and to identify genotypes in candidate One ELA class II gene and five MHC-linked microsatellite gene SNPs associated with susceptibility to IBH in Ice- markers were used for analyzing associations with the MHC landic horses imported from Iceland. (Fig. 1). A PCR-RFLP genotyping system of Eqca-DRA exon 2 alleles was developed based on the GenBank sequences available. Amplification of the 307 bp long product was car- Materials and methods ried out with standard primers Be3 and Be4 [13]. Aliquots of PCRproductsweredigestedseparatelywithrestriction The population and the disease enzymes Cac8I, Hpy166II, BsaJI and produced specific fragments allowing identification of alleles Eqca-DRA*0201 Eighty-nine horses of the Icelandic breed born in Iceland (M60100), Eqca-DRA*0301 (L47172), Eqca-DRA*0401 with no common parents and imported to Switzerland were (AJ575295) and Eqca-DRA*0501 (FJ716134). The remain- used for the study. The horses had been living in Swit- ing allele Eqca-DRA*0101 (L47174) could be identified by zerland for at least 4 years to exclude that IBH susceptible manual editing, subtracting sequences of the other alleles horses that had not yet developed clinical signs of IBH previously identified by PCR-RFLP. The nomenclature sug- would be included in the non-affected group. Most of the gested by us previously [14] was used for designating Eqca horses developed IBH within 2 years after import [12]. alleles. 123

108 Mol Biol Rep (2013) 40:3333–3340 3335

Fig. 1 MHC markers used for association analysis (in bold)

Length variations of five microsatellite markers located Data analysis in the class I and class II regions of the horse MHC (COR112, COR113, COR114, UM011 and UMN-JH34-2) Association analysis was performed in IBM SPSS were determined using an ABI PRISM 310 automated Advanced Statistics 19 for Windows (Release 19.0.1, Ó sequencer (Applied Biosystems, Foster City, CA, USA) IBM Corporation 2010). Identification of host genotypes based on primers and procedures published [15]. associated with the results of the CAST test was performed in two steps. After an initial screen allowing identification of individual markers potentially associated with IBH at

Single nucleotide polymorphism markers praw \ 0.10, a pairwise analysis of interactions of the in allergy-related genes markers identified was performed. All possible genotypes composed of individual marker variants were tested by SNPs in genes encoding the CD14 receptor (CD14), binary logistic regression analysis to assess individual Fc fragment of the IgE receptor alpha chain (FcER1A), contributions of markers and of pairwise statistical inter- Interferon gamma (IFNG), Immunoglobulin heavy chain actions. The resulting odds ratios and confidence intervals epsilon-like (IGHE), Interleukin 4 (IL4), Interleukin 4 were calculated based on a standard two-sided Fisher’s receptor (IL4R), Interleukin 10 (IL10), Interleukin 13 (IL13), exact test and Fisher’s exact test based on Monte Carlo Interleukin 17 (IL17A), Interleukin 17 receptor (IL17AR), estimates (100,000 samples) of exact significance [16, 17]. Interleukin 23 receptor (IL23R), Involucrin (IVL), Tyrosine- Microsatellite alleles with frequencies \0.1 were pooled protein kinase JAK2-like (JAK2), Thymic stromal lympho- for this analysis. Bonferroni corrections for multiple poietin (TSLP), Toll-like receptor 4 (TLR4), Transforming comparisons were used based on [18], separately for dif- growth factor beta 1 (TGFB1), Transforming growth factor ferent types of markers. MHC-linked multi-allelic micro- beta 3 (TGFB3) were studied (Table 1). Markers were satellite markers, the MHC Eqca-DRA expressed gene with identified by searching the Broad institute SNP database multiple-SNP haplotypes, and independently segregating (http://www.broad.mit.edu/mammals/horse/)ornewlydevel- bi-allelic SNP markers are genetically different types of oped by resequencing of pooled DNA samples from horses of loci. Therefore, their associations with IBH were studied the population under study (usually from 5 to 10 horses). separately. Bonferroni corrections for multiple compari- Horse-specific primers were designed using Primer3 software sons thus were made separately for five MHC microsatel- (http://fokker.wi.mit.edu/primer3/) on annotated genes from lites, three exon two alleles of Eqca-DRA and 20 candidate the NCBI horse genome sequence assembly EcuCab2.0 gene SNP markers. In specific genotypes, Bonferroni cor- (http://www.ncbi.nlm.nih.gov/mapview/map_search.cgi?taxid rections were made for the numbers of genes involved in =9796) or genes predicted by BLAST search of horse genome combined genotypes. for defined human/animal counterparts. Sequences with double or multiple peaks were aligned and analysed by the BioEdit software (http://www.mbio.ncsu.edu/BioEdit/bioedit. Results html) and putative restriction enzyme cleavage sites were assigned by NEBcutter V2.0 (http://tools.neb.com/NEB MHC markers cutter2/index.php). SNP markers were then genotyped by PCR-RFLP by using restriction enzymes (New England Eqca-DRA was associated with CAST-Cul status (praw = Biolabs, Ipswich, MA, USA) corresponding to polymorphic 0.012, pcorr = 0.036). Genotypes with Eqca-DRA*0501 sites. The fragments generated were run for 90 min at 300 V were enriched in CAST-negative horses (f = 0.390) in 6 % PAGE and silver stained. compared to cases (f = 0.136). For CAST-Sim, praw for 123

109 3336 Mol Biol Rep (2013) 40:3333–3340

Table 1 Non-MHC candidate gene SNP markers used for analysis of associations with insect bite hypersensitivity Candidate gene Gene SNP genome position SNP genome Type of Primers 50–30 forward/ symbol (marker name used) position marker reverse (chromosome)

Fc fragment of IgE FCER1A NW_001867419 G 1417759 A Eca5 38124382 G[A Intronic gtgccgtggctggaaggat/ receptor, alpha chain (FcER1A) gccaggaagaaattgctgttgc Ig heavy epsilon chain IGHE NW_001876796 T 754996 C (IGHE) Eca24 754996 T[C Exonic gtctccaagcaagccccatta/ tttaccagggtctttggacacctc Interferon gamma IFNG NW_001867424 G 24587636 C Eca6 83346753 C[G Intronic tactctggaactcagtcaattgctgaga/ (IFNG/a) G 24585624 C (IFNG/b) 83344741 C[G intronic gaaatggattctgactcctcttc Interleukin 10 IL10 NW_001867416 A 2988098 G (IL10) Eca5 2988098 A[G Intronic tctgccctgtgaaaataagagc/ tgtcaaactcactcatggcttt Interleukin 13 IL13 NW_00186737 A 42920818 G (IL13) Eca14 42920818 Intronic cctgacccctctagagacctg/ A[G acaggctgaggtccaagcta Interleukin 17 receptor IL17RA NW_001867423.1 T 921254 C Eca6 27987884 T[C30UTR gcaggcacacacctaaacct/ (IL17RA) ggggacagaagatgaccaga Interleukin 17A IL17A NW_001867389.1 T 49864780 C Eca2049864780 T[C Intronic ctctctccttgcccattcag/ (IL17A) ggctgtcctgtgtcctatca Interleukin 23 receptor IL23R NW_001867420.1 C 44757430 T Eca5 93803835 C[T Promotor ttgaaaaggcagaacagaattt/ (IL23R) cctccatgacaccaactgaa Interleukin 4 IL4 NW_00186737 G 42902615 A (IL4) Eca14 42902615 Intronic ccttgatcaaagaatgcctga/ G[A tccaaaggccctgtgtaatc Interleukin 4 receptor IL4R NW_001867375 G 2398183 A (IL4R) Eca13 20794503 Exonic cttcttccccctttaggaagtg/ G[A gagttctgagggctgtgaggt Involucrin IVL NW_001867419 C 44674251 Eca5 44674251 C[T Intronic cagcacattctgccagtgac/ T(IVL/a) taatgctgctgctgctgttt Janus kinase 2 JAK2 NW_001867420 T 26565472 C Eca23 26565472 Intronic ggggttaagaacaaggtgga/ (JAK2) T[C tgtggaacccataaagctctg LPS receptor molecule CD14 NW_001867377 A 36268808 G Eca14 36268808 5‘UTR gagcctgagtcatcaggacattgc/ (CD14/a) C 36269096 T (CD14/b) A[G 36269096 intronic tggcttccaggctccacaca C[T cgcagctctttccagagtcca/ cggaagttctcatcgtccacct Thymic stromal TSLP NW_001867377 C 66624113 G Eca14 66624113 Intronic gctggatgagaccgcagtccc/ lymphopoietin (TSLP) C[G gctgctcctcgtcagcatttgc Toll-like receptor 4 TLR4 NW_001867396 A 21975271 G Eca25 21975271 Exonic ggcctcaaccatctctccacct/ (TLR4/a) G 21975144 A (TLR43/b) A[G 21975144 exonic ccacggtttaccatccagcaag G[A Transforming growth TGFB1 NW_001867363 C 11894258 A Eca10 11894258 Intronic ttgactttcgcaaggatctg/ factor beta1 (TGFB1) C[A ggttgtgctggttgtacagg Transforming growth TGFB3 NW_001867395 G 21504080 A Eca24 21504080 Exonic ggaaaaagtgtggctttcca/ factor beta3 (TGFBR3) G[A tgatccaagattccccaaaa

Eqca-DRA*0501 was 0.024, corresponding to pcorr 0.096. between cases and controls at p = 0.0006, while p values No microsatellite genotype was signiﬁcantly associated ranging from 0.045 to 0.006 were found for selected with CAST-Cul/Sim after Bonferroni corrections and dif- genotypes in COR112, COR113, COR114 and UMN-JH34-2. ferences in microsatellite allelic frequencies ranged Interactions of different markers in composed genotypes between 0.1 and 0.05. The COR113 allele 257 was asso- could not be analyzed due to low numbers of animals with ciated with negative results of the CAST test (Table 2). relevant combinations of alleles. Only the genotype Eqca- When only genotypes with selected resistance and/or sus- DRA*0501/COR113 257 carrying two ‘‘resistant’’ alleles ceptibility associated alleles were compared against each in both loci was informative. No carrier of this genotype other, signiﬁcant differences were observed for all loci. was found among 46 CAST-Sim cases, while seven carriers

The strongest effect was observed for UM011. In 57 horses, were identiﬁed in the control group (praw = 0.014, cor- the frequencies of genotypes with alleles 166 or 170 rected for two markers p = 0.028). No associations with compared to genotypes carrying alleles 168 or 180 differed heterozygosity in the MHC-linked markers were found.

123

110 Mol Biol Rep (2013) 40:3333–3340 3337

Table 2 Associations between MHC-ELA markers and IBH in Icelandic horses at praw \ 0.100 (N = 85)

Marker Position DA Alleles Cul/ f(DAC) f(DAC) praw pcorr OR (CI) p value (raw/corr) for tested Sim CAST CAST differences in f(DA) positive negative

DRA 32690939 DRA*0501 3 Cul 0.136 0.390 0.012 0.036 0.247 0.017/0.051 (0.085–0.716) DRA 32690939 DRA*0501 3 Sim 0.152 0.385 0.024 0.072 0.287 0.027/0.081 (0.102–0.805) COR113 33480825 257 11 Cul 0.068 0.220 0.023 0.115 0.260 NS (0.065–1.040) COR113 33480825 257 11 Sim 0.043 0.256 0.020 0.100 0.152 0.004/0.020 (0.031–0.751) COR113 33480825 253 11 Sim 0.283 0.103 0.039 0.195 3.447 0.021/0.105 (1.020–11.645) UM011 33510120 170 10 Cul 0.023 0.195 0.013 0.065 0.960 0.015/0.075 (0.011–0.805) UM011 33510120 170 10 Sim 0.022 0.205 0.013 0.065 0.074 NS (0.009–0.615) N number of horses, DA disease-associated allele, Cul/Sim Culicoides/Simulium, f(DAC) frequency of DA carriers, OR(CI) odds ratios (confidence interval), f(DA) frequency of DA, NS not significant before correction Significant p values in bold

Allergy-related candidate gene SNPs Five statistically signiﬁcant SNPs at p = 0.1 were then used for analysis of composed genotypes: TSLP, IVL, None of the SNP markers was signiﬁcantly associated CD14, IL23R and TGFB3 (Table 3). Out of all possible with CAST status at p = 0.05 (Fisher’s exact test). combinations tested, associations of genotypes composed

Table 3 Markers involved in Signiﬁcant SNP Genotypes/ p value for p value for Associated with composed genotypes alleles associated ‘‘CAST-CUL’’ ‘‘CAST-SIM’’ (praw \ 0.100) TSLP GC NT 0.057 Susceptibility CC Resistance IVL TT 0.171 (allele T: 0.087) – Susceptibility CC?TC Resistance CD14/b TC 0.086 0.073 Resistance CC Susceptibility Genotypes/alleles associations IL23R TT 0.191 (allele T: 0.099) – Resistance tested both for genotypes and/or CC?CT Susceptibility alleles, NT not tested due to TGFB3 AA 0.110 (allele: 0.053) 0.090 Resistance p [ 0.05 for all individual GG?AG Susceptibility markers

Table 4 Signiﬁcant (pcorrected \ 0.05) associations of composed genotypes with CAST status Genotypes associateda p values for p values for Odds ratios Associated with CAST-Cul (praw/pcorr) CAST-Sim (praw/pcorr) (CI 95 % OR)

TSLP CC–TGFB3AA NT 0.018/0.036 NC Resistance CD14/b CC–IL23R non-TT 0.021/0.042 NT 3.766 (1.218; 11.640) Susceptibility CD14/b CC–TGFB3 non-AA 0.021/0.042 0.009/0.018 5.161 (1.339; 19.895)—Cul Susceptibility 6.143 (1.590; 23.736)—Sim CI 95 % OR confidence interval of odds ratios, NT not tested due to p [ 0.1 for all individual markers, NC not calculable due to null values a No composed genotype with IVL was significantly associated with IBH Significant p values in bold 123

111 3338 Mol Biol Rep (2013) 40:3333–3340

Table 5 Values of population linkage disequilibrium—LD (coefficient of correlation/p) among MHC-linked markers in Icelandic horses UMN-JH34-2 xxxxx DRA 0.089/0.218 xxxxx COR112 0.098/0.0001 0.146/0.0001 xxxxx COR113 0.090/0.001 0.138/0.0003 0.134/0.0001 xxxxx UM011 0.094/0.001 0.131/0.0002 0.150/0.0001 0.182/0.0001 xxxxx COR114 0.108/0.0001 0.155/0.0001 0.162/0.0001 0.198/0.0001 0.208/0.0001 xxxxx UMN-JH34-2 DRA COR112 COR113 UM011 COR114 Significant p values in bold of CD14/TGFB3, CD14/IL23R and TSLP/TGFB3 withstood imported to continental Europe [2, 12]. This is probably corrections for multiple testing. No combined genotype due to the fact that these horses are exposed to new aller- containing IVL reached statistical significance (Table 4). gens as adults [3]. However, the fact that about 50 % of these horses remain free of IBH even though subjected to the same environmental changes suggests that genetic Discussion factors also influence susceptibility to IBH in this group of horses. Classification of the phenotypes observed is crucial for Horses born in Iceland and imported to Switzerland interpretation of the results of association analyses. We were included in our study. Environmental influences could classified horses as resistant or susceptible to IBH based on thus dilute the effect of genetic variation and consequently the results of a cellular allergy test additionally to the clinical the power of the association analysis. However, it was our diagnosis. The sLT-release assay with C. nubeculosus extract aim to investigate whether besides the environmental had been shown previously to have a relatively good sen- influences genetic factors contributing to susceptibility for sitivity (78 %) and very high specificity (97 %) for diag- IBH could be identified also in this group. Significant nosis of IBH [4]. This was important for the selection of the associations withstanding all over-conservative Bonferroni non-affected horses, as we wanted to exclude horses that corrections confirmed this hypothesis. Since rather weak were sensitized to Culicoides but had not (yet) developed genetic effects could be anticipated, there is no reason to clinical signs of IBH. On the other hand, the use of a expect false positive results due to this approach. Similar positive sLT-release test results for selection of the IBH associations for the MHC-DRA gene and for the COR113 cases allowed including only horses with IBH caused by marker in Icelandic horses born and living in Sweden [9] IgE-mediated reactions. In some cases, IBH may be support this conclusion. caused by cell-mediated type IV hypersensitivity reactions However, false negative results must be considered. [3]. Only 89 horses living in a similar environment in terms of Even when GWAS are available, studies of individual similar exposure to Culicoides, diagnosed with the same candidate loci can be useful for formulating more precise diagnostic method during the same season, belonging to the hypotheses on mechanisms of the disease studied [19, 20]. same breed but with no common parents, were available. For equine IBH, this was confirmed by Andersson et al. [9]. This is a limitation of this study, like in many of clinical Here, candidate genes were selected according to their role field studies on large domestic mammals. It is likely that in immune responses and in allergic reactions as reported the group analyzed did not allow detecting effects of fur- in horses and other species. The MHC region was shown to ther markers on IBH and negative results of this study be associated with IBH in horses [8, 9]. The set of cyto- cannot be interpreted as lack of association. kines and of genes expressed in the skin (IVL and TSLP) We confirmed effects of ELA class II markers observed was composed based on literary data and on results of our previously in two distinct horse populations [9]. Due to study on Old Kladruby horses where associations and/or within-breed variation, it is not surprising that the markers differences in gene expression were found [Vychodilova and alleles associated were not the same like in the popu- et al., Vet Immunol Immunopathol accepted]. lations analyzed. The LD values among markers (Table 5) As compared to humans, domestic animal populations are in agreement with their distances on ECA20 (Fig. 1). bred for a long time are more homogenous and more Our data thus support the view that the ELA class II region informative, especially for clinically oriented association is associated with clinical and cellular processes related to studies with limited numbers of cases available. Icelandic equine IBH across breeds and populations. It is not clear ponies are an old breed, highly susceptible to IBH when whether differences observed between CAST-Cul and

123

112 Mol Biol Rep (2013) 40:3333–3340 3339

CAST-Sim reflect low numbers of cases analyzed or the Expression of IL-10 and TGFb1 molecules was found to be specificity of allergic reactions. involved in IBH in horses [35]. Although Icelandic horses represent a model population Taken together, our results confirmed associations of for IBH investigations, no study analyzed non-MHC MHC class II, CD14 and TSLP markers with equine IBH immunity-related candidate gene SNP markers associated observed in a different population. In addition, associations in this context. Analysis of pairwise interactions in com- of IBH with two other immunity-related genes, IL23R and bined genotypes showed increased odds ratios and p values TGFB3 were found. Genes associated with IBH correspond suggesting possible cumulative effects of selected geno- to those reported in human atopic dermatitis association types/alleles. Only single markers associated at praw \ 0.05 studies. The results of statistical analyses supported by were used to increase the probability of finding p values similar observations in genetically distinct populations and significant after all corrections. Complex genotypes thus by gene expression data from another study suggest that the allowed identification of synergic interactions of genes role of these genes in equine IBH merits to be further with weak individual effects. This approach was used investigated. recently e.g. for analyzing interactions between markers underlying complex mechanisms of susceptibility to Acknowledgments The work was supported by the Grant Agency human Crohn’s disease [21]. of the Czech Republic projects 523/06/1402 and 524/09/1939, by IGA VFU project 22/05/FVL, and by the Swiss National Science Foun- Support for biological plausibility of the results of the dation Grant No. 310030_129837/1. candidate gene study comes from two sides. First, the same TSLP marker (although through different alleles) was associated with clinical IBH in a genetically distinct population of Old Kladruber horses. [Vychodilova et al., Vet References Immunol Immunopathol accepted]. As the SNP analyzed is TSLP an intronic marker, it is not supposed to have a direct 1. Cunningham FM, Dunkel B (2008) Equine recurrent airway effect on the phenotype. Second, in the same study, dif- obstruction and insect bite hypersensitivity: understanding the ferences in gene expression in skin biopsies from affected diseases and uncovering possible new therapeutic approaches. and unaffected horses were observed for CD14. Statistical Vet J 177:334–344 2. Marti E, Gerber V, Wilson AD, Lavoie JP, Horohov D, Crameri R, evidence for its association with IBH in Icelandic horses Lunn DP, Antczak D, Bjo¨rnsdo´ttir S, Bjo¨rnsdo´ttir TS, Cunningham thus represents complementary information supporting the F, De´rer M, Frey R, Hamza E, Horin P, Heimann M, Kolm-Stark G, results obtained in Kladruber horses. Olafsdo´ttir G, Ramery E, Russell C, Schaffartzik A, Svansson V, Data from human and animal studies also are in agreement Torsteinsdo´ttir S, Wagner B (2008) Report of the 3rd Have- meyer workshop on allergic diseases of the Horse, Ho´lar, Ice- with the assumption of biological plausibility of the associ- land, June 2007. Vet Immunol Immunopathol 126:351–361 ations observed. Immunity-related genes associated with 3. Schaffartzik A, Hamza E, Janda J, Crameri R, Marti E, Rhyner C IBH in this study were reported to be associated with atopic (2012) Equine insect bite hypersensitivity: what do we know? Vet dermatitis in humans [22–24]. The TSLP-receptor gene was Immunol Immunopathol 147:13–126 4. Baselgia S, Doherr MG, Mellor P, Torsteinsdottir S, Jermann T, associated with atopic dermatitis in different dog breeds [25]. Zurbriggen A, Jungi T, Marti E (2006) Evaluation of an in vitro TSLP was shown to have a wide range of effects on cells of sulphidoleukotriene release test for diagnosis of insect bite the immune system and on allergic inflammation [26]. hypersensitivity in horses. Equine Vet J 38:40–46 Associations of atopic disease with CD14 polymorphisms 5. Eriksson S, Grandinson K, Fikse WF, Lindberg L, Mikko S, Brostro¨m H, Frey R, Sundquist M, Lindgren G (2008) Genetic [24], supported by functional studies [27], were reported in analysis of insect bite hypersensitivity (summer eczema) in Ice- humans. Besides immunity-related markers, genes related to landic horses. Animal 2:360–365 skin structure and function were also associated with human 6. Schurink A, Ducro BJ, Heuven HC, van Arendonk JA (2011) AD [28]. Involucrin is a specialized cytoplasmatic protein of Genetic parameters of insect bite hypersensitivity in Dutch Friesian broodmares. J Anim Sci 89:1286–1293 the keratinocytes, cells responsible for the structure of epi- 7. Marti E, Gerber H, Lazary S (1992) On the genetic basis of dermis [29]. SNPs in the IL23R gene were associated with equine allergic diseases: II. Insect bite dermal hypersensitivity. human psoriasis. It is known that genome regions associated Equine Vet J 24:113–117 with susceptibility to psoriasis, Crohn’s disease and atopic 8. Lazary S, Marti E, Szalai G, Gaillard C, Gerber H (1994) Studies on the frequency and associations of equine leucocyte antigens in dermatitis overlap in humans [30]. Futhermore, IL-23 sig- sarcoid and summer dermatitis. Anim Genet 25:75–80 naling enhances Th2 polarisation [31] and Th2 cytokine 9. Andersson LS, Swinburne JE, Meadows JR, Brostro¨mH, production is increased in horses with IBH [32, 33]. Up- Eriksson S, Fikse WF, Frey R, Sundguist M, Tseng CT, Mikko S, regulation of the CD14 gene activity was observed in Lindgren G (2012) The same ELA class II risk factors confer equine insect bite hypersensitivity in two distinct populations. affected skin from human atopic dermatitis patients [27]. Immunogenetics 64:201–208 Transcription of TGFB1 and TGFB3 was enhanced in the 10. Hoffjan S, Epplen JT (2005) The genetics of atopic dermatitis: regenerating epidermis and dermis in humans and pigs [34]. recent findings and future options. J Mol Med 83:682–692 123

113 3340 Mol Biol Rep (2013) 40:3333–3340

11. Schurink A, Ducro BJ, Bastiaansen JW, Frankena K, van (2005) Polymorphisms in the 50 region of the CD14 gene are Arendonk JA (2012) Genome-wide association study of insect associated with eczema in young children. J Allergy Clin bite hypersensitivity in Dutch Shetland pony mares. Anim Genet. Immunol 115:1056–1062 doi:10.1111/j.1365-2052.2012.02368.x 25. Wood SH, Ollier WE, Nuttall T, McEwan NA, Carter SD (2010) 12. Bjo¨rnsdo´ttir S, Sigvaldado´ttir J, Brostro¨m H, Langvad B, Sigurdsson A Despite identifying some shared gene associations with human (2006) Summer eczema in exported Icelandic horses: influence of atopic dermatitis the use of multiple dog breeds from various environmental and genetic factors. Acta Vet Scand 48:3 locations limits detection of gene associations in canine atopic 13. Albright-Fraser DG, Reid R, Gerber V, Bailey E (1996) Poly- dermatitis. Vet Immunol Immunopathol 138:193–197 morphism of DRA among equids. Immunogenetics 43:315–317 26. Roan F, Bell BD, Stoklasek TA, Kitajima M, Han H, Ziegler SF 14. Janova E, Matiasovic J, Vahala J, Vodicka R, Van Dyk E, Horin P (2012) The multiple facets of thymic stromal lymphopoietin (2009) Polymorphism and selection in the major histocompatibility (TSLP) during allergic inflammation and beyond. J Leukoc Biol complex DRA and DQA genes in the family Equidae. Immuno- 91:877–886 genetics 61:513–527 27. Su¨megi A, Szegedi A, Ga´l M, Hunyadi J, Szegedi G, Antal- 15. Tseng CT, Miller D, Cassano J, Bailey E, Antczak DF (2010) Szalma´s P (2007) Analysis of components of the CD14/TLR Identification of equine major histocompatibility complex hap- system on leukocytes of patients with atopic dermatitis. Int Arch lotypes using polymorphic microsatellites. Anim Genet 41(Suppl. Allergy Immunol 143:177–184 2):150–153 28. Boguniewicz M, Leung DY (2011) Atopic dermatitis: a disease 16. Agresti A (2002) Categorical data analysis, 2nd edn. Wiley, of altered skin barrier and immune dysregulation. Immunol Rev Hoboken, p 710 242:233–246 17. Li W (2007) Three lectures on case-control genetic association 29. Eckert RL, Green H (1986) Structure and evolution of the human analysis. Brief Bioinform 9:1–13 involucrin gene. Cell 46:583–589 18. Zar JH (1999) Biostatical analysis, 4th edn. Prentice Hall, Upper 30. Capon F, Di Meglio P, Szaub J, Prescott NJ, Dunster C, Baumber L, Saddle River, p 662 Timms K, Gutin A, Abkevic V, Burden AD, Lanchbury J, Barker JN, 19. Akey J (2009) Constructing genomic maps of positive selection Trembath RC, Nestle FO (2007) Sequence variants in the genes for the in humans: where do we go from here? Genome Res 19:711–722 interleukin-23 receptor (IL23R) and its ligand (IL12B) confer pro- 20. Michel S, Liang LM, Depner M, Klopp N, Ruether A, Kumar A, tection against psoriasis. Hum Genet 122:201–206 Schedel M, Vogelberg C, von Mutius E, von Berg A, Bufe A, 31. Peng J, Yang XO, Chang SH, Yang J, Dong C (2010) IL-23 Rietschel E, Heinzmann A, Laub O, Simma B, Frischer T, signaling enhances Th2 polarization and regulates allergic airway Genuneit J, Gut IG, Schreiber S, Lathrop M, Illig T, Kabesch M inflammation. Cell Res 20:62–71 (2010) Unifying candidate gene and GWAS approaches in 32. Hamza E, Doherr MG, Bertoni G, Jungi TW, Marti E (2007) asthma. PLoS ONE. doi:10.1371/journal.pone.0013894 Modulation of allergy incidence in Icelandic horses is associated 21. Cso¨ngei V, Ja´romi L, Sa´frańy E, Sipeky C, Magyari L, Farago´ B, with a change in IL-4-producing T cells. Int Arch Allergy Bene J, Polga´r N, Lakner L, Sarlo´s P, Varga M, Melegh B (2010) Immunol 144:325–337 Interaction of the major inflammatory bowel disease suscepti- 33. Heimann M, Janda J, Sigurdardottir OG, Svansson V, Klukowska bility alleles in Crohn’s disease patients. World J Gastroenterol J, von Tscharner C, Doherr M, Brostro¨m H, Andersson LS, 16:176–183 Einarsson S, Marti E, Torsteinsdottir S (2011) Skin-infiltrating T 22. Arkwright PD, Chase MJ, Babbage S, Pravica D, David TJ, cells and cytokine expression in Icelandic horses affected with Hutchinson IV (2001) Atopic dermatitis is associated with a low insect bite hypersensitivity: a possible role for regulatory T cells. producer transforming growth factor beta (1) cytokine genotype. Vet Immunol Immunopathol 140:63–74 J Allergy Clin Immunol 108:281–284 34. Cox DA (1995) Transforming growth factor-beta3. Cell Biol Int 23. Nomura I, Goleva E, Howell MD, Hamid QA, Ong PY, Hall CF, 19:357–371 Darst MA, Gao B, Boguniewicz M, Travers JB, Leung DY 35. Hamza E, Wagner B, Jungi TW, Mirkovitch J, Marti E (2008) (2003) Cytokine milieu of atopic dermatitis, as compared to Reduced incidence of insect-bite hypersensitivity in Icelandic psoriasis, skin prevents induction of innate immune response horses is associated with a down-regulation of interleukin-4 by genes. J Immunol 171:3262–3269 interleukin-10 and transforming growth factor-b1. Vet Immunol 24. Litonjua AA, Belanger K, Celedoń JC, Milton DK, Bracken MB, Immunopathol 122:65–75 Kraft P, Triche EW, Sredl DL, Weiss ST, Leaderer BP, Gold DR

123

114 MS VI Genomic analysis of resistance/susceptibility to melanoma in Old Kladruber greying horses.

115 Tissue Antigens ISSN 0001-2815

LETTER TO THE EDITOR Genomic analysis of resistance/susceptibility to melanoma in Old Kladruber greying horses J. Futas1,L.Vychodilova1, B. Hofmanova2, M. Vranova1,3, L. Putnova4,J.Muzik5, M. Vyskocil1, I. Vrtkova4, L. Dusek5,I.Majzlik2 &P.Horin1,3

1 Institute of Animal Genetics, University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic 2 Department of Animal Science and Ethology, University of Life Sciences, Prague, Czech Republic 3 CEITEC-VFU, University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic 4 Laboratory of Agrigenomics, Mendel University Brno, Brno, Czech Republic 5 Institute of Biostatistics and Analyses, Masaryk University, Brno, Czech Republic

Key words: association analysis; greying horse; melanoma; microsatellites; single nucleotide polymorphisms

Melanoma is a common disease occurring in many horse Breeding mares were examined for the presence of breeds. As in all tumors its nature is complex, and specific melanoma by visual inspection according to Seltenham- genes were shown to be involved in the mechanisms of dis- mer et al. (4) and classified as negative (n = 26, mean ease in humans as well as in various animal species (1). age 11 years, range 7–23 years) or positive (n = 24, mean Grey horses, regardless of the breed, show high incidence age 16 years, range 7–18 years). Fifty microsatellite mark- of dermal melanomas. A mutation within the syntaxin 17 ers located on 29 out of 31 pairs of horse chromo- gene responsible for greying in horses was shown to be somes (ECA) and 36 polymorphic markers located in associated with occurrence of melanoma in grey horses but 26 expressed candidate genes (CD14, CTNNAL1, ELA- the association observed did not explain completely the Eqca-DRA, ELA-Eqca-DQA, FcER1A, GBP1, HRH4, IGHE, variation observed (2). Other factors influencing the forma- IFNG, IL1B, IL1RN, IL4, IL4R, IL10, IL13, IL12A, IL12B, tion of melanoma thus must be involved. In this context, IL12RB2, IL23A, MMP16, TGFB1, TGFB3, TLR4, MC1R, immunity-related genes can be considered as functional can- ASIP and STX17) were genotyped. Corrections for age were didate genes. Old Kladruber grey horses were selected as a made based on the cutoff point 12 years, set-up by using model for analyzing associations between microsatellite and the receiver operating characteristics (ROC) web calcula- candidate gene polymorphic markers and the presence of tor (http://www.jrocfit.org.). Fisher’s exact test with standard melanoma. Genetic diversity of this old, small, isolated and Bonferroni corrections for multiple comparisons was used for partially inbred population was characterized previously (3). association analyses.

Table 1 Signiﬁcant (Pcorr < 0.05) associations between multiple-gene (composed) genotypes and presence of melanomas in Old Kladruber greying horses

Associated Frequency of Frequency of Odds ratio a 2 Genotype associated with carriers in N carriers in PPuncorr/Pcorr (conﬁdence interval)

TKY312 non99/ MMP16 1471G P 0.10 (1/10) 0.85 (17/20) 0.00013/0.00214 45.516 (2.997–695.951) COR058 non220/IL12A 242GG P 0.20 (2/10) 0.85 (17/20) 0.00097/0.01559 18.291 (1.802–185.706) COR058 non220/MMP16 1471G P 0.00 (0/10) 0.80 (16/20) 0.00003/0.00053 Not calculable COR058 non220/UM011 167 P 0.11 (1/9) 0.85 (17/20) 0.0003/0.00483 71.114 (2.016–2 508.149) homozygous COR018 250/ELA-Eqca-DRA non*0401 N 0.89 (8/9) 0.20 (4/20) 0.00086/0.01380 0.005 (0.000–0.318) COR018 250/TLR4 I heterozygous N 0.78 (7/9) 0.00 (0/20) 0.00002/0.00037 Not calculable COR018 250/TLR4 II heterozygous N 0.78 (7/9) 0.00 (0/20) 0.00002/0.00037 Not calculable (IL12A 242GG/MMP16 1471G/ STX17 P 0.00 (0/9) 0.80 (16/20) 0.00007/0.00043 Not calculable G homozygous)/ (TKY312 non99/COR058 non220/ UM011 167 homozygous) non (IL12A 242GG/ MMP16 1471G/ N 0.80 (8/10) 0.05 (1/20) 0.00006/0.00032 0.017 (0.001–0.224) STX17 G homozygous)/ non (TLR4 I homozygous/ TLR4 II homozygous) aCarriers of at least one allele involved in single-locus associations in each locus were included.

After all corrections, three chromosome regions located by the Czech Ministry of Agriculture MZe (NAZV) project on ECA 25, 9 and 20 were found to be significantly QH92277. associated with the presence of melanoma (0.0004 < Pcorr < 0.02). The presence and strong association of the STX17 Correspondence duplication on ECA25 with melanoma reported (2) were Petr Horin confirmed in this population. In addition, single nucleotide Institute of Animal Genetics polymorphisms (SNPs) within the TLR4 gene linked to STX17 University of Veterinary and Pharmaceutical Sciences Brno on ECA25 were also associated with melanoma. Strong and Brno 612 42 significant linkage disequilibrium between STX17 and TLR4 Czech Republic was observed, despite their physical distance (>15 Mb). Two Tel: +420 541 562 292 + linked markers on ECA9, the microsatellite HTG004 and Fax: 420 549 248 841 e-mail: [email protected] a SNP within the matrix metalloproteinase 16 (MMP16 ), represent a second group of associated markers. Analysis of doi: 10.1111/j.1399-0039.2011.01827.x genotypes composed of multiple markers not only confirmed single gene associations, but also suggested effects of other Conflicts of interest immunity-related genes in the presence of melanoma, IL12A, coding for the interleukin-12 p35 subunit and specific major The authors have declared no conflicting interests. histocompatibility complex Eqca haplotypes (Table 1). The associations identified are in agreement with biological roles References of the genes involved (5–7). 1. Smith SH, Goldschmidt MH, McManus PM. A comparative Although some risks of false positive associations were review of melanocytic neoplasm. Vet Pathol 2002: 39: 651–78. reduced by corrections for age and multiple comparisons 2. Rosengren Pielberg G, Golovko A, Sundstrom¨ E et al. A with high P values, the results obtained provide only pre- cis-acting regulatory mutation causes premature hair graying and liminary evidence that immunity-related genes might be susceptibility to melanoma in the horse. Nat Genet 2008: 40: involved in mechanisms of the horse melanoma. On the 1004–9. other hand, the STX17 marker, if considered as a kind 3. Horin P, Cothran EG, Trtkova K et al. Polymorphism of Old of ‘positive control’ showed that even in this small pop- Kladruby horses, a surviving but endangered baroque breed. Eur ulation biologically plausible associations can be detected. J Immunogenet 1998: 25: 57–63. 4. Seltenhammer MH, Simhofer H, Scherzer S et al. Equine However, they need to be confirmed in higher numbers of melanoma in a population of 296 grey Lipizzaner horses. horses and in other populations. Within the model endan- Equine Vet J 2003: 35: 153–7. gered population analyzed, melanoma represents a serious 5. Ohnishi Y, Tajima S, Ishibashi A. Coordinate expression of threat. The markers identified can be investigated as tools membrane type-matrix metalloproteinases-2 and 3 (MT2-MMP for designing a conservation program aiming to reduce inci- and MT3-MMP) and matrix metalloproteinase-2 (MMP-2) in dence of melanoma without reducing population diversity and primary and metastatic melanoma cells. Eur J Dermatol 2001: specificity. 11: 420–3. 6. Muller¨ J, Feige K, Wunderlin P et al. Double-blind placebo-controlled study with interleukin-18 and Acknowledgments interleukin-12-encoding plasmid DANN shows antitumor effect in metastatic melanoma in gray horses. J Immunother 2011: 34: This study was supported by the Czech National Sci- 58–64. ence Foundation, project GA CR 523/09/1972, by the 7. Bogunovic D, Manches O, Godefroy E et al. TLR4 engagement project ‘CEITEC – Central European Institute of Technology’ during TLR3-induced pro-inflammatory signaling in dendritic (CZ.1.05/1.1.00/02.0068) from the European Regional Devel- cells promotes IL-10-mediated suppression of anti-tumor opment Fund, by IGA VFU – project 150/2008/FVL and the immunity. Cancer Res 2011: 71: 5467–76.