Mitochondrial DNA diversity of the founder populations of the Asturcón pony

A 380 bp fragment of the horse D-loop region was analysed in 42 founder mare samples of the bay-coated Asturcón pony obtained in three different and reproductively isolated mountainous areas of Western Asturias: range of “La Bobia” (20), range of “Carondio” (13) and range of “El Aguión” (9). These sequences were compared with the information provided by 37 founder matrilines of black-coated Asturcón assigned to the range of “Sueve” (26) and the “out-of-Sueve” (11) founder populations. The aim of this research was to ascertain the differences in founder mtDNA diversity between the two strains of the Asturcón pony and if such differences have geographical consistency. The 79 sequences analysed gave 16 different haplotypes defined by 33 polymorphic sites. The two Asturcón strains shared eight haplotypes that gathered 76% and 81% of the samples available in bay-coated and black-coated Asturcón, respectively. Both haplotypic (0.027 ± 0.006) and nucleotide (0.021 ± 0.011) diversity were higher in the bay-coated than in the black-coated Asturcón (0.024 ± 0.005 and 0.016 ± 0.009, respectively). AMOVA analyses failed in assessing any statistically differentiation among Asturcón geographical populations or strains. Most genetic variability is due to the individuals (estimates varying from 96.34% to 99.81%). Differentiation among strains or population took low and non-significant values. Differentiation between Asturcón pony strains using mtDNA marker would not have clear support. The two strains of the Asturcón pony breed likely derive from the same ancestral mare population. Additional key words: D-loop; genetic variability; population structure; horse. * Corresponding author: fgoyache@serida.org Received: 25-02-13. Accepted: 18-07-13. This work has one Supplemenary Figure that does not appear in the printed article but that accompanies the paper online. Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA) Spanish Journal of Agricultural Research 2013 11(3): 702-707 Available online at www.inia.es/sjar ISSN: 1695-971-X http://dx.doi.org/10.5424/sjar/2013113-4127 eISSN: 2171-9292

that the main matrilineal composition of the Asturcón pony belongs to the Jansen et al.'s (2002) haplotypic family D (Royo et al., 2005a).However, the mtDNA haplotypic variability in the black-coated Asturcón is much larger than that assessed in the bay-coated Asturcón (Royo et al., 2007;Álvarez et al., 2012) suggesting that the founder population of the black-coated Asturcón has a larger influence of foreign horse breeds.
Here the mtDNA haplotypes assigned to the founder matrilines of black-coated Asturcón analysed in Álvarez et al. (2012) are compared with new sequences obtained from founder bay-coated Asturcón individuals sampled in three different and reproductively isolated areas of Western Asturias.The black-coated Asturcón pony matrilines were assigned to either a "Sueve" or to an "out-of-Sueve" area in accordance with Royo et al. (2005b).The aim of this research was to ascertain the differences in founder mtDNA diversity between the two strains of Asturcón pony and if such differences have geographical consistency.Inferences on the ancestral mtDNA composition of the Asturcón pony will also be made.
Sampling locations are illustrated in Fig. 1.Blood samples were obtained from a total of 42 bay-coated Asturcón founder matrilines sampled in three different, geographically isolated, areas.Twenty matrilines were sampled in the range of "La Bobia"; thirteen matrilines were sampled in the range of "Carondio"; and, finally, nine matrilines were sampled in the range of "El Aguión".Within Western Asturias, the "La Bobia" and the "El Aguión" areas have the most westerly and the most easterly locations, respectively, while the "Carondio" area has a central location.The 37 founder matrilines of black-coated Asturcón analysed in Álvarez et al. (2012) were assigned to a "Sueve" founder population (26) and to an "out-of-Sueve" founder population (11) according to Royo et al. (2005b).The "Sueve" population gathered those founder mares recovered in different management units within the range of Sueve.The out-of-Sueve population included those mares recovered during the early 1990s out in the lowlands of Eastern Asturias and the area of Picos de Europa.
Total DNA was isolated from blood samples following standard procedures (Sambrook et al., 1989).The laboratory methods used to amplify a 380 bp region of the horse mtDNA are described in full in Álvarez et al. (2012).Mitochondrial DNA sequences were aligned using the program CLUSTAL implemented in the Short communication.Founder mitochondrial DNA diversity in Asturcón pony 703 Figure 1.Geographical areas in which the Asturcón pony populations analysed were sampled.Abbreviations mean the following: "La Bobia", B; "Carondio", C; "El Aguión", A; "Sueve", S; and "out-of-Sueve", oS.software MEGA 5.03 (Tamura et al., 2007; available at http://www.megasoftware.net/).Analyses were restricted to 380 bp.The nucleotide positions were numbered following Acc.No. X79547 (Xu & Arnason, 1994).According to the polymorphic sites, the identified haplotypes were assigned to the haplogroups defined by Jansen et al. (2002; see Table 2 of that paper).Def inition of haplotypes has been done following Álvarez et al. (2012).The haplotypes that were not identified in such manuscript are here numbered starting from haplotype (Hap) 16.To derive the betweenhaplotypes phylogenetic relationships we constructed a median-joining network (MJ network) using the Network 4.5.1.0program (Bandelt et al., 1999), available at http://www.fluxus-engineering.com/.The software Arlequin 3.5 (Excoffier & Lischer, 2010) was used to assess: a) haplotype and nucleotide diversity at the population level (Table 1); and b) population structure via AMOVA analyses.Statistical confidence for computed values was estimated by permutation analysis using 1000 replicates.
The 79 sequences analysed gave 16 different haplotypes including those 11 previously identified in blackcoated Asturcón by Álvarez et al. (2012) (Table 1).The bay-coated Asturcón gave five haplotypes (from H_16 to H_20) which were not present in the black-coated Asturcón.In both Asturcón strains H_1 was the most frequent haplotype (50% and 30% in bay-coated and black-coated Asturcón, respectively).H_1 was identified in all the populations analysed.Three haplotypes identified in black-coated Asturcón were not identif ied in bay-coated Asturcón.The haplotypes shared between the two Asturcón pony strains gathered 76% (bay-coated) and 81% (black-coated) of the samples available in each Asturcón strain.Haplotypic and nucleotide diversity were slightly higher in bay-coated Asturcón pony (0.027 ± 0.006 and 0.021 ± 0.011, respectively) than in the bay-coated strain (0.024 ± 0.013 and 0.016 ± 0.009, respectively).The out-of-Sueve black-coated Asturcón population had relatively high haplotype diversity (0.091 ± 0.039) but particularly high nucleotide diversity (0.024 ± 0.013).Fig. 2 illustrates the high haplotypic diversity assessed.No clear phylogenies could be ascertained.Most sequences belonged to the haplotypic superfamily D (62% and 42% in bay-coated and black-coated Asturcón, respectively).In any case, both Asturcón strains shared sequences assigned to 10 of the haplogroups reported by Jansen et al. (2002) (see Table 1).AMOVA analyses did not give evidence of genetic structure.Regardless the model fitted, most genetic variability was assigned to the individuals, varying from 96.34% to 99.81%.When the analysis was fitted in two hierarchical levels, considering the population sampled or the strain, the Φ ST values estimated were low and statistically non significant.The three hierarchical levels analysis, considering both the Asturcón strains and the populations within strain, gave a significant Φ CT (relative divergence between strains) of 0.051, explaining 5.10% of the genetic variability.However, this may be an artifact due to the fact that between-populations differentiation within strains took a negative value.Further analyses (not shown) informed that Φ ST distance computed between each pair of geographical populations were low and statistically non significant.
The current research points out the existence of noticeable haplotypic diversity in the sampled Asturcón pony populations.Despite the intense genetic bottleneck underwent by the Asturian native pony populations, a considerable number of matrilines remained.Domestic horses originated from multiple matrilines (Vilà et al., 2001;Jansen et al., 2002).However, it has also been reported that genetic bottlenecks, such as that well known occurred in the Exmoor pony, do not always affect haplotype diversity (Jansen et al., 2002).In any case, the analysed sequences were obtained considering founder matrilines only.This is clearly biasing upwards the overall diversity of the populations studied which is probably lower in the whole present population (Álvarez et al., 2012).
The analyses carried out also failed in assessing any statistically differentiation among Asturcón geographical populations or strains.Since the Asturcón populations analysed have been reproductively isolated for decades, one could expect some genetic differentiation due to genetic drift.Transmission of mtDNA is subject to major stochastic processes, such as practically inexistent paternal transmission, which can be amplified by local bottlenecks (White et al., 2008).Many different local processes of drift and fixation could have affected the maternal genetic scenario in the Asturcón pony.However, if happened, they could not be traced using the current data.
Furthermore, mitochondrial lineages have been widely shown as not useful to identify horse breeds.From a technical point of view, it is difficult to establish a pattern for horse mtDNA haplotypes neither at a regional nor breed level: the ability of the species for move and the use of the horse as the primary means of long distance transportation leaded to extensive mixing of formerly distinct geographic populations before and after domestication (Vilà et al., 2001;Jansen et al., 2002).Recently, Achilli et al. (2012), using 83 horse mitochondrial genomes, suggested that the proposed specificity of certain mtDNA clusters to some horse breeds, such as the British pony breeds (Jansen et al., 2002), was only due to the traditional use of a small fragment of the control region in horse mtDNA studies.The information provided by the complete mitochondrial genome suggests that this specificity is no longer justifiable (Achilli et al., 2012).Moreover, during the last two centuries allochtonous horses have been exploited in the Asturian mountains.However, even though an influence of such breeds on the mitochondrial DNA composition of the Asturcón pony cannot be excluded, it has been reported that these breeds (namely the so called Hispano-Bretón) are more likely present-day reservoirs of ancestral mtDNA lineages of Celtic ponies of the Iberian Peninsula (Pérez-Gutiérrez et al., 2008).Therefore, to trace any mtDNA influence on the maternal composition of the founder Asturcón mare populations analysed is unlikely.
Finally, the current research suggests that the two strains of the Asturcón pony breed derive from the same ancestral mare population.Most sequences analysed could be assigned to the Jansen et al. (2002) D haplofamily which has been suggested to be the main representative of the ancestral Iberian horse population (Royo et al., 2005a).This is also truth for the out-of-Sueve population of the black-coated Asturcón pony, which was obtained in areas with not many historical references on the existence of ancient local Asturcón herds.Even though this population gathers a relatively large number of different Jansen et al.'s (2002) haplogroups, the maternal composition of this, or any other population analysed, does not allow to consider it as a Jansen et al. (2002); Hap_17 could not be assigned to any haplogroup with certainty.

Figure 2 .
Figure2.Median-joining tree constructed with the mtDNA haplotypes described in Table1.Circles are proportional to the number of samples displaying each haplotype.Slashes on the branches mean the number of mutations (higher than 1) separating nodes.Capital letters near circles identify the haploclusters reported byJansen et al. (2002) for the species.Within each circle, white portion represents the bay-coated Asturcón and black portion the black-coated Asturcón pony samples in the nodes.Missing haplotypes are denoted as mv.