Avaliação de um painel de microssatélites para identificação animal/averiguação de paternidade e estrutura genética da população de cavalos da raça Mangalarga Marchador

Abstract

Introduction: Mangalarga Marchador is the most numerous horse breed in Brazil, with more than 400 thousand animals registered in the studbook. It was started by the ending of the XIX century in the south of Minas Gerais State from crosses between Alter, Andaluz, Arab, Creole, and Quarter Horse animals. The breed is defined by a particularly smoth gait and by its rusticity. The present work was developed in order to evaluate the performance of the parentage microsatellite panel in Mangalarga Marchador breed and to ascertain its population genetic structure. Material and Methods: Data from 6,092 born in 2008 and 2009 were included in order to avoid generation overlapping. Results from genotyping of 14 microsatellite markers were analyzed (AHT4, AHT5, ASB2, ASB17, ASB23, CA425, HMS1, HMS2, HMS3, HMS6, HMS7, HTG4, HTG7, and VHL20). Analyses included presence of null alleles, large allele dropout, and allele misassignment due to stuttering, allelic and genotypic frequencies, observed and expected heterozygosity, polymorphic information contend (PIC), mean PIC, power of exclusion (PE), paternity index (PI), combined paternity index (CPI), Hardy-Weinberg Equilibrium tests (EHW), AMOVA, and F statistics (FST, FIS e FIT). P values were Bonferroni corrected for an alpha = 0.05. Softwares used included GENEPOP 4.0, Arlequim 3.11, Powerstats 1.2, MicroChecker 2.2.3. Results and discussion: a total of 129 alleles were found, with a mean of 9.2 alleles per marker (range: 5-20). Mean observed heterozygosity was 0.702 (0.536 to 0.813), PE global: 0.9998, a mean PIC: 0.68 (0.5 a 0.8), and a combined PI: 99.9687%. Lower values of number of alleles, observed and expected heterozygosity, PIC, PE, and PI per locus were observed for HTG7 and the higher ones for ABS17. Null alleles were identified for some markers in some populations: MG (AHT5, ASB2, HMS2, HMS3, HMS6, HTG4, and CA425), BA (HMS3 and HTG7), and PE (HTG7 and VHL20). No significative evidences were found for allele dropout or allele misassignment due to stuttering. For genetic structure analyses tests were underwent in the complete sample and in the sample divided according the state of birth of the animals. In the complete sample, significative deviations from HWE expectations were found for markers ASB2, ASB23, CA425, HMS2, HMS6, and VHL20. Some markers showed HWE deviations in the subsamples of some states: MG, five markers; MS and PA, one marker each. In these subsamples, HWE deviations were detected most of the times for HMS2. AMOVA results locate most of the variation within the populations (99.4), with a small fraction of it between them (0.6). FST for the complete population over all the loci was 0.006, pointing to a slight differentiation between the populations. FIS values were close to null and non-significant for most of the markers tested. Higher pairwise FST values were found for the populations of the following pairs of states: MS-AL, MS-PA, and MS-RN. Lower pairwise FST was found for SP-MG. However, most of the pairwise FST values were very low and highly significant, suggesting that these populations are very similar. Conclusions: results obtained for mean PIC, PE and CPI indicated that the analysed panel is highly efficient for this breed. No strong evidence for population structure was found, meaning that, if there actually is any, it is not geopolitically based.