Population

Blood samples were collected from 213 unrelated individuals from Buenos Aires city (106 women and 107 men) with informed consent. Samples were stripped of any personal identifiers and made anonymous prior to testing at our lab. The protocols used in this study were approved by the Institutional Review Board of Sam Houston State University.

DNA extraction and quantification

Genomic DNA (gDNA) was extracted from blood samples using a modified Chelex^® method [9] suspending the resin in TE buffer (Tris-HCl 10mM, EDTA 0.1 mM, pH 8.0). DNA samples were quantified by real time PCR in a StepOne™ Real Time System (Life technologies, Carlsbard, CA) using 2 μL extracted gDNA, 400nM D21S11 primers (GenBank Accession number AP000433) (Integrated DNA Technologies, Coralville, IA), and 1X SYBR^® Green PCR Master Mix (Life technologies). The following real time PCR cycling parameters were used: 10 min at 95°C and 40 cycles of 15 s at 95°C and 1 min at 60°C.

PCR amplification and genotyping

Three X-STRs were amplified in a multiplex amplification reaction including DXS10079, DXS10074 and DXS10075 according to previous studies [4]. Amplified PCR products were detected by capillary electrophoresis in either an ABI PRISM™ 310 or a 3500 Genetic Analyzer (Life Technologies). Allele calling was performed using the software GeneMapper v. 4.1 (Life Technologies) and compared with allelic ladders as recommended by the ISFG [10]. A custom designed bin set was implemented to allow automation of genotyping. A ladder containing all the observed alleles in the population was prepared according to Sajantila et al. [11]. Allelic ladders were designed for each X STR marker using a mixture of individual samples that represent each allele observed in the population data set. The samples were amplified separately for each locus accordingly to the previously described protocol. Each of the amplified products were diluted 1:1,000 with H₂O, mixed together in a total volume of 1 mL and re-amplified under the same conditions but using 20 cycles instead of 30. The ladder was calibrated with DNA control samples 9947A, 9948, and K562 (Promega, Madison, WI) s [12]. Microvariants 14.3 (DXS10074), 17.2, 18.2, and 19.2 (DXS10075) were sequenced. DXS10074 and DXS10075 primers were used in a direct Taq-cycle sequencing procedure using the BigDye^® Direct Cycle Sequencing Kit and the 3500 Genetic Analyzer with Sequencing Analysis software version 4.1 (Life Technologies). Sequences were assembled and proof-read using the Geneious Pro software v. 5.5.7 (Biomatters Ltd., Auckland, New Zealand). Sequenced alleles were used to calibrate the allelic ladder. Allele frequencies and parameters of forensic interest were generated with PowerStats v. 1.2 software [13]. Hardy-Weinberg equilibrium (HWE) in females, observed heterozygosity and gene diversity were carried out with the Genetic Data Analysis software [14]. Homogeneity of allele frequencies of men and women and overall linkage disequilibrium (LD) in the male sample was assessed using GENEPOP version 4.1 software package [15]. Inter-allelic linkage disequilibrium (ID) in males was estimated with the MIDAS software [16]. Power of exclusion calculations were estimated according to Desmarais et al. [17] Population comparisons were carried out by estimating genetic distances based on Fstusing the Arlequin v. 3.5 software [18]. Bonferroni correction for the number of loci analyzed in HWE and LD evaluation and the correction for the number of inter-population comparisons at adjusted significance level were performed [19].

Quality control

The laboratory of the first author successfully participated in proficiency testing of the Latin-American Society of Forensic Genetics 2007 (http://www.slagf.org.ar).

---

Results and Discussion

Combined allele frequencies for men and women and parameters of forensic interest are depicted in (Table 1). Fisher’s exact test did not reveal allele distribution differences between men and women, therefore their frequencies can be combined. No significant deviations from expectations of HWE were detected in any of the markers analyzed in women (p > 0.015; significant level after Bonferroni correction). Overall LD and ID are shown in (Table 2). Analysis of the three X-STR markers in males showed no detectable evidence of LD except for pair DXS10074-DXS10075 (p < 0.015; significant level after Bonferroni correction). These results are consistent with the close genetic distance of these two markers (21 kb). Moreover, significant inter-allelic associations were detected among these three markers when the disequilibrium coefficient (D’) was evaluated (D’ > 0.8) (Table 2). Most studies in forensic X-STRs use pvalues of a Fisher’s exact test as a measured variable to infer the value of gametic disequilibrium (GD); which is not a test to evaluate GD. In linkage studies, it is important not only to measure overall linkage disequilibrium between pairs of loci but also to detect specific interallelic associations between markers, which can be evaluated using D’ [20]. A value of D’ > 0.8 indicates the presence of strong linkage disequilibrium between two markers [21]. Haplotype frequencies formales are displayed in (Table 3). Microvariants observed in loci DXS10074 and DXS10075 were sequenced and submitted to Genbank (Accession numbers JX645356,JX645357, JX645358 and JX645359). Interestingly, the repeat structure of allele 14.3 in DXS10074 reflects the motif of “short” alleles (< 10) instead of “long” alleles (> 13), since it is composed of contiguous repetitions of the (AAGA)n unit, with a AGA downstream the core repeat unit [4]. The repeat structure of microvariants 17.2, 18.2,and 19.2 of marker DXS10075 is consistent with the repeat pattern of intermediate X.2 alleles described by Hering et al. [4]. Population sample comparisons by pairwise genetic distance analysis, based on Fst were performed with previously published data that contained the cluster DXS10079-DXS10074-DXS10075, when available. These included Brazil [22], China [23], Japan [24],and Germany [4] (Table 4). Results showed statistically significant differences between all population pairs except for Alagoas-Rio, Buenos Aires-Germany and when the comparison involved the Argentinian and Brazilian population samples. This is not unexpected because the three South American populations (Buenos Aires, Alagoas and Rio) were subjected to a similar Amerindian and European (Portuguese for Brazil and Spaniard/Italian for Argentina) complex admixture process. Although X STR comparative data between Argentina and Germany were not found in the literature, two previous reports showed similarities in allele frequencies between Buenos Aires and other European sample sets when other X STR markers were used (i.e., Spain, Portugal and Italy) [25,26]. The genetic differences observed between Japanese and Chinese populationswere previously reported when autosomal single nucleotide polymorphisms were evaluated [27]. In conclusion, a population database for the cluster DXS10079-DXS10074-DXS10075 was developed for forensic and anthropological purposes. The use of these linked markers in conjunction with other X-STR markers should be considered especially in deficiency cases when a daughter is involved and complex kinship scenarios.