GENETIC MARKERS FOR SKIN COLOR FORENSIC DNA PHENOTYPING IN THE BRAZILIAN ADMIXED POPULATION

CERQUEIRA, CCS; RAMALLO V; HÜNEMEIER, T; BARBOSA, AAL; SCHULER-FACCINI, L; SALZANO, FM; BORTOLINI, MC

Congreso; 59 Congresso Brasileiro de Genética; 2013

e human pigmentation pathway can be explained by a complex network controlled by many genes, as well as by environmental, mechanical, and epigenetic factors; thus, to understand it is a challenging task. Although dozens of genes have previously been associated with human skin color, our knowledge about this trait is still incomplete. Particularly limited is the number of studies with populations o the Europe-North American axis, and seldom until now admixed populations were considered.  e present study was planned to help  ll this gap.  e objective was to verify the possible association of 18 SNPs located within ten genes/pseudogene regions (ADAM17rs1524668, AFG3L1rs4785763, ASIPrs6058017, HERC2rs1129038, MC1Rrs1805007, MC1Rrs1805008, MC1Rrs1805009, OCA2rs1800401, OCA2rs1800407, OCA2rs1800414, SLC24A5rs1426654, SLC45A2rs6867641, SLC45A2rs26722, SLC45A2rs16891982, TPCN2rs3750965, TPCN2rs3829241, TYRrs1042602, and TYRrs1126809), with the MI (Melanin Index) in two admixed Brazilian populations (Gaucho, n=354; Baiano, n=149) with di erent histories of geographic and ethnic colonization. Some of these polymorphisms have never been studied in the context of human skin color normal variation. DNA extractions were done using the salting out method from total blood, while genotyping procedures were performed by Taqman?. MI measurements were taken on the proximal medial portion of both arms to generate a single average value (range in humans from 20, associated to the fairest skin, to 100, to darkest skin).  e SPSS software, version 17.0, was used to calculate ANOVA or the non-parametric Kruskal-Wallis test, comparing MI versus genotypes. We identi ed 4 markers associated with MI in the Gaucho sample (HERC2rs1129038, MC1Rrs1805009, SLC24A5rs1426654, and SLC45A2rs16891982), and 6 in the Baiano (HERC2rs1129038, OCA2rs1800407, SLC24A5rs1426654, SLC45A2rs16891982, TYRrs1042602, and TYRrs1126809), but only 3 were signi cantly associated in the two samples (HERC2rs1129038, SLC24A5rs1426654, SLC45A2rs16891982).  erefore, only these 3 should be preliminarily chosen as being of forensic signi cance, since they consistently show the association independently of the population considered. Furthermore, to consider possible population di erences is also important to choose an appropriate set of SNPs as phenotype predictors in forensic practice.