Utility of several microsatellite markers for the genetic characterisation of three ex situ populations of threatened caprine taxa (Capra aegagrus, C. cylindricornis and C. falconeri)

Caprines belong to the most endangered group of mammals and artiodactyls suffering from many negative human impacts. Fortunately, many of them are protected and managed by national and international legislation and in situ and ex situ conservation actions. Although many microsatellite markers have been developed for wild and domestic caprines, they remain uninvestigated in respect of their utility for some taxa. We examined the utility of the International Society for Animal Genetics microsatellite set for genetic characterisations of three wild and one domestic Capra species from captive or semi-captive ex situ populations in Europe. Our data suggest the utility of this microsatellite set for detecting shared and species-specific alleles, characterising the genetic variability, and determining phylogenetic relationships and intraspecific structures in investigated taxa. We detected a depleted genetic variability in Capra falconeri and Capra cylindricornis in European ex situ populations; unrelated individuals are therefore needed for improving genetic variability parameters, as they are for the extralimital population of Capra aegagrus in the Vřísek game reserve (Czech Republic), for which we identified no genetic introgression from the domestic goat and great dissimilarity with some analysed individuals from European zoos. Current results here indicate some difficulties with the historical evidence, for example with respect to the origin and purity of particular individuals under breeding programmes. Published by Copernicus Publications on behalf of the Leibniz Institute for Farm Animal Biology. 366 E. Levý et al.: Utility of several microsatellite markers for the genetic characterisation


Introduction
Caprines (Caprini sensu, Groves and Grubb, 2011) represent a rather evolutionarily young group of bovids, but one that is certainly no less diverse in its morphological, ecological and behavioural features than other groups of ungulates (Groves and Leslie, 2011;Hassanin et al., 2012;Bibi, 2013).Caprines exhibit extraordinary adaptations to many habitats, including the harshest environments of our planet, but despite that, they are not resistant to negative human impacts (e.g.Shackleton, 1997).They suffer from habitat destruction, hunting, competition with domestic livestock, disease from domestic livestock, and also introgression from their domestic counterparts that originated in part from them (Shackleton, 1997;Groves and Leslie, 2011).In effect, Caprini belong to the most endangered group of artiodactyls, with approximate twice as many endangered and critically endangered species as other bovid groups (e.g.Bovini, Antilopini; our counting is based on Groves and Leslie, 2011).Fortunately, many of them are protected and managed by national and international legislation and specific in situ and ex situ conservation actions, and genetic markers are increasingly being used for genetic characterisation (e.g.genetic variability parameters, genetic purity) of caprine species and their particular populations (e.g.Maudet et al., 2002;Hammer et al., 2008;Gebremedhin et al., 2009).To avoid negative effects of outbreeding and inbreeding, management actions should be based upon a large number of purebred individuals; genetic variability parameters, structure and purity are relatively easily and unequivocally detectable with a good resolution by genetic markers analysed by population-genetic and phylogenetic methods (e.g.Allendorf and Luikart, 2007;Frankham et al., 2003).The study of microsatellite loci of Capra species is very important for effective scientific management from the prospective of maintaining the wild (sub)species/domestic breed purity and from the perspective of genetic variability parameters.
Microsatellite markers with a high polymorphism are one of the best tools for detecting the genetic fitness of whole populations or even of particular individuals, including paternity determination (e.g.Witzenberger and Hochkirch, 2011).Although many microsatellite markers have been developed for wild and domestic caprines (Arevalo et al., 1994;Bhebhe et al., 1994;Bishop et al., 1994;Crawford et al., 1995;Kemp et al., 1995;Kogi et al., 1995;Ma et al., 1996;Vaiman et al., 1996;Luikart et al., 1999Luikart et al., , 2006;;Maudet et al., 2001Maudet et al., , 2002Maudet et al., , 2004;;Kumar et al., 2009), they remain uninvestigated in respect of their utility for some taxa.We examined the utility of the ISAG (International Society for Animal Genetics) microsatellite set, available primarily for the domestic goat paternity specification, for genetic characterisations of three wild and one domestic Capra species from captive or semi-captive ex situ populations from EAZA (European Association of Zoos and Aquaria) or private institutions in central Europe -specifically Capra falconeri heptneri, C. cylindricornis, C. aegagrus and C. hircus.One population, of Capra aegagrus in the Vřísek game reserve (near Česká Lípa, northern Bohemia, managed by Forests of the Czech Republic, s. e.), is associated historically with a relatively well-known out-of-range population of this species in the Pálava Biosphere Reserve (southern Moravia), where several (the precise number is unknown) wild goats were released in the 1950s from Prague and Brno zoos (Ernst et al., 2011).The wild goats occupied the steppe and surrounding habitats under some game control management for 43 years, until in 1996 the remaining individuals were captured and transferred to the Vřísek game reserve due to damage to valuable steppe flora in the Pálava reserve (e.g.Heroldová, 1997;Ernst et al., 2011).
This study represents an attempt to analyse the abovementioned taxa in respect of ISAG microsatellite set utility and to assess detected genetic parameters in the conservation management programmes for these taxa.

Sample collection
We studied a total of 55 individuals of Capra aegagrus (hereinafter referred to as CA), of which n = 50 were from the Vřísek game reserve and n = 5 from the Olomouc zoo; 26 individuals of Capra falconeri (hereinafter referred to as CF), subspecies C. f. heptneri, from one breeding institution in Slovakia; 10 individuals of Capra cylindricornis (hereinafter referred to as CC) from breeding institutions in the Czech Republic; and 9 individuals of various local breeds of Capra hircus (hereinafter referred to as CH) typical for the Czech Republic in order to detect possible genetic introgression to CA.From 2008 to 2012, blood samples were collected from the jugular veins of adult animals by a sampling kit comprising an anticoagulant (150 µL 0.5 M EDTA/5 mL of blood).As for juvenile animals, hair bulbs were sampled and mouth cavity smears were acquired with the oral smear kit, to minimise stress.As for harvested or deceased individuals, the skeletal muscles were used for DNA isolation.All proce- Abbreviations: FM -fluorescence marking, PC -primer concentration (nM), T -taxon, IA -identified alleles (species-specific ones are in bold, alleles with the highest frequency are in italic), n -sample size, Ng -numbers of genotypes, Na -number of alleles, MAF -major allele frequency, He -expected heterozygosity, Ho -observed heterozygosity, PIC -polymorphic information content, F -inbreeding coefficient.
dures were carried out in accordance with the laws and ethical guidelines established in the Czech Republic and during the course of regular veterinary work with these animals.All biological samples were stored at −85 • C except for hair bulbs, which were stored in paper bags at temperatures not exceeding 24 • C.

Microsatellite genotyping and analyses
Genome DNA was isolated by the QIAamp DNA blood kit.Target DNA sequences comprising potential polymorphic loci were amplified by the polymerase chain reaction (PCR).
We tested a modified multiplex typing panel of microsatellite loci according to panels of goat markers for parentage verification tested using the 2001/2002 ISAG comparison test by the ISAG Standing Committee on "Applied Genetics in Sheep and Goats".The panel allows the co-amplification and four-colour detection of eight microsatellite markers (multi-plex1 proposed by LGS, Cremona, Italy); for more details see Table 1.A Veriti thermal cycler (Applied Biosystems) was used for the PCR 8-plex.The total volume of the amplification mixture amounted to 6.25 µL.The reagency concentration amounted to 1.2× PCR buffer (comprising 15 mM MgCl 2 ), 336 µM dNTP mixture, 0.9 U Taq Gold polymerase (Am-pliTaq Gold ™ , Applied Biosystems), 3 % DMSO, and 10-100 ng/µL DNA.Primers of selected microsatellite markers were used in concentrations ranging from 80 to 800 nM (see Table 1).Primer concentration, marking and sequence are listed under Table 1.The following cycling conditions were applied: initial denaturation 95 A total of 0.5 µL of ROX 500 size standard (Applied Biosystems) and 11.5 µL Hi-Di ™ unionised formamide (Applied Biosystems) were added to the amplification product of 0.5 µL.Samples were denaturised for 5 min at 95 • C and then laid on ice for 5 min.Polymorphism of studied loci was detected with ABI PRISM 310 ® genetic analyser (Applied Biosystems, Foster City, CA, USA).Fluorescent markers were evaluated by GeneScan ® 3.7 NT and Genotypern ® 3.7 NT (Applied Biosystems) software.

Results and discussion
Our data proved the utility of the used microsatellite set for the understanding of genetic variability and fitness of analysed populations (for details see Table 1).For example, we found polymorphism at all loci of CA, CF and CH, but at only two loci of CC (one locus was not amplified).CA exhibits the most species-specific alleles, as well as number of alleles per locus and number of genotypes, followed by CH, CF and CC.Our study enlarges the number of known alleles, for example in CA (Maudet et al., 2004;Glowatzki-Mullis et al., 2007).On the contrary, however, Pokorádi et al. (2006) detected more alleles for CF and Luikart et al. (1999) for CH, probably due the higher sample size.As CH comprise a lot of breeds (Taberlet et al., 2008), our sample of local Czech breeds cannot reflect the diversity of this one properly.Alleles with the highest frequency do not correspond in most cases with species-specific alleles, and the highest average frequency of main alleles was found in CC; it was almost identical in CF and CA, and the lowest frequency was recorded in CH.
As the heterozygosity is considered one of the main indices of viability and adaptability of animal populations, the average heterozygosity and PIC were highest in the CH (comparable with Jandurová et al., 2004, andLuikart et al., 1999, specifically in the white short-haired, brown shorthaired and Saanen goat breeds) followed by CF (higher than Pokorádi et al., 2006), CA (much higher than CA from the game park in Berne; Saitbekova et al., 1999), and the lowest was recorded in CC (lower than Maudet et al., 2004).Limited genetic variability in CF and CC is certainly associated with a limited number of founders of the captive populations (see Holma, 2007;Fainstein, 2011) and is comparable with populations of Capra species with historical bottlenecks and associated depleted genetic variability such as C. ibex (detected heterozygosity 0.13-0.4;Maudet et al., 2002;Biebach and Keller, 2009) or C. walie (0.35;Gebremedhin et al., 2009).New founders and/or their gametes using the AI procedure are of great conservation importance for their long-term ex situ management.
When assessing the population, the Hardy-Weinberg balance was calculated for CA χ 2 = ∞, df = 16, P < Highly sign.; CF χ 2 = 11.5950,df = 16, P <0.7714; CH χ 2 = 7.6063, df = 16, P < 0.9597.However, it was impossible to calculate it for CC.The probabilities of paternity exclusion, one parental genotype unavailable, and parentage exclusion were, for a panel of eight microsatellites, 99.49, 94.81 and 99.99 % respectively in all goat samples.All analysed taxa were differentiated by STRUCTURE (Fig. 1) with K = 5.Moreover, the CA set is split up in two independent sets: CA from the Vřísek game reserve and CA from the Olomouc zoo.When analysing the polymorphism of studied microsatellite loci, no influence of CH on the CA population from the Vřísek game reserve (n = 50) was established.On the contrary, CA from the Olomouc zoo (n = 5) shows a distinctive genetic link to CH, as demonstrated by K = 2, K = 3, K = 4, and K = 6, although this stock could, based on some historic evidence, have the same link to the Pálava-founding CA population.Based on our evidence, though, this link seems improbable.Our study also indicates that CA from the Pálava population at the Vřísek game reserve is genetically valuable and has no genetic introgression from the CH as was assumed by Anděra and Červený (2009).It is appropriate to point out that the material is incomplete (e.g. in relation to the study of Naderi et al., 2008) and that the whole issue is more complex than work on very different species and on wild species with no domesticated counterparts, as both species are historically related with possible mutual gene flow.Our study also detected no introgression of domestic goats into the CF stock in the European zoos (Hammer et al., 2008), which points to a high breeding-value of the analysed portion of CF.
The neighbour-joining tree (Fig. 2) shows the differentiation of two lines, namely CA together with CH and CC together with CF.All species exhibit comparable divergences from the ancestral stock, except for CC, which shows more changes unique to that group.The genus Capra is relatively young, with speciation still in progress (Hassanin et al., 2012).In effect, the phylogenetic relationships of species within the group depend on the genetic markers used (Luikart et al., 2001(Luikart et al., , 2006)), possibly due to incomplete lineage sorting and/or also former natural hybridisation events (Pidancier et al., 2006).Our results are more similar to the taxa relationships presented in Luikart et al. (2001).
In general, the information presented here indicates some difficulties with historical evidence, for example with respect to the origin and purity of particular individuals, and underlines the need to use genetic methods for valuable sciencebased conservation management of ex situ populations.

Figure 1 .
Figure 1.Results of the STRUCTURE analysis of the Capra microsatellite markers studied here.

Figure 2 .
Figure 2. Topology of neighbour-joining tree showing the genetic relationship among four Capra populations using genetic distances for eight microsatellite loci.

Table 1 .
Genetic variation analysed by eight microsatellite markers.