Introduction

The current method used to sex determine ancient human remains is anthropometric measurements of the bones, such as skull, pelvis, vertebrae, teeth, long bones, and the pars petrosa ossis temporalis (Ovchinnikov et al. 1998; Graw et al. 2005; Veroni et al. 2010; Harris and Case 2012; Krishan et al. 2016; Ubelaker and DeGaglia 2017; Hora and Sládek 2018; García-Campos et al. 2018a, b). Sex determination using this method is more accurate in adult whole-body skeletons, than in the subadult skeletons, and very old or fragmented bone remains (Álvarez-Sandoval et al. 2014) because the anatomical differences between men and women are more clearly distinguished after the onset of puberty due to adolescent hormonal changes. These female characteristics begin to develop at different ages and ageing process can be affected by nutritional status and related socioeconomic factors. In many cases, these anthropometric methodologies do not yield conclusive results in subadult individuals. On the other hand, in modern humans, secondary sexual traits usually emerge in the skull or postcranial skeleton during adolescence. Female characteristics begin to develop at different ages with the ageing process also being affected by nutritional status and related socioeconomic factors (Moore 2013; Candelas-González et al. 2017).

Similar procedures are used in the sex identification of newborns, and subadults, although the accuracy of sex estimation is lower in subadult than adult skeletons (Fazekas and Kósa 1978; Schutkowski 1987; Luna et al. 2017). Sexing can be also performed in 66% of the specimens with skulls that have been largely destroyed, using the usually well-preserved pars petrosa ossis temporalis of skulls (Graw et al. 2005). Furthermore, sex was correctly assigned in male maxillary canines of subadults that contained greater amounts of dentine (García-Campos et al. 2018a, b). Still, these methods are useless in very fragmented adult, and subadult bones, or when the adequate bone remains are not available (Álvarez-Sandoval et al. 2014).

There are several molecular methods to determine sex, such as the amelogenin test based on the amplification of the AMEL gene in the X and Y chromosomes (AMELX/AMELY) (Stone et al. 1996; Tschentscher et al. 2008; Gibbon et al. 2009; Álvarez-Sandoval et al. 2014). The AMEL locus has 2 homologous genes: AMELX, which is located on the distal short arm of X chromosome (Chr-X) (p22.1-p22.3), and AMELY, which is located near the centromere of the Y chromosome (Chr-Y) (p11.2). These genes have 89% homology, with there being a 6-bp deletion in the third intron of AMELX which is absent in AMELY (Stone et al. 1996; Butler and Li 2014; Álvarez-Sandoval et al. 2014; Dutta et al. 2017). The amplification of other genes, such as DXYS156, SOX3, STS, and TSPYL2 (Bashamboo et al. 2003; Butler and Li 2014), are also used for the sex determination of ancient remains. However, the limitation of this and other molecular methods is DNA degradation, which depends on the archaeological context and bone sample (Quincey et al. 2013; Álvarez-Sandoval et al. 2014; Gaudio et al. 2019). Sex identification of ancient human remains was recently performed using the quantification of sequences that are aligned with Chr-Y and Chr-X by considering that males have half of the amount of DNA that corresponds to Chr-X compared to females (Green et al. 2010; Skoglund et al. 2012, 2013). Sex determination with this methodology was obtained by massive sequencing and the exclusion of the homologous regions in both chromosomes (Green et al. 2010; Skoglund et al. 2013). However, the read depth must be reasonably high to avoid any confounding effects due to sequencing errors.

Chr-Y typically determines male sex in humans. It is the third smallest chromosome comprising ~ 2–3% of the haploid genome (Quintana-Murci and Fellous 2001; Halder et al. 2017). It is composed of a pseudoautosomal portion (PAR) that is divided into the two regions: PAR1, located in the terminal region of the short arm (Yp11.32), and PAR2, located in the long arm (Yq12). The pseudoautosomal regions can recombine with the pseudoautosomal region of Chr-X (Xp22.3 and Xq28) during meiosis (Quintana-Murci and Fellous 2001; Bashamboo et al. 2003; Halder et al. 2017). PAR1 and PAR2 represent ~ 5% of the whole chromosome, with the remaining ~ 95% corresponding to the non-recombinant region of human Chr-Y (NRY). Genes in the NRY region are classified into two categories. The first region is comprised of genes that are ubiquitously expressed, show homology with Chr-X, and exhibit housekeeping cellular functions. The second region includes genes with specialized functions that are expressed in the testes (Halder et al. 2017).

The development of new and improved methodologies for sex estimation and the revaluation of existing methods is very important for both ancient and forensic researchers to achieve more accurate results. The objective of this study was to develop an alternative technique. For this, we selected four genes from fifteen Chr-Y-specific genes in contemporary male samples in silico. These male-specific DNA markers were experimentally tested in contemporary and ancient samples from individuals found in Tacotalpa, Tabasco, Mexico. The results confirmed that these regions were specific to the male sex.

Materials and methods

Contemporary samples

The present study contained ten contemporary samples (five males and five females) from peripheral blood obtained from the “Servicio de Medicina Interna” of the Regional Hospital by Dr. Carlos MacGregor Sánchez Navarro from the Instituto Mexicano del Seguro Social (IMSS). The samples were registered to project number 2017-785-071, and the National Ethics Commission (IMSS) ethics committee approved the study with the number and R-2017-785-071. The tests were performed by following the rules established in the Helsinki Declaration. All data were kept strictly confidential, as indicated by national and international regulations. The participants were informed about the procedure prior to providing their written consent.

Archaeological material

The study contained nineteen samples discovered in three different archaeological sites in the municipality of Tacotalpa in Southern Tabasco (Supplementary Table 1). Ten of these samples were found at the Puyil cave located at San Felipe Mountain in Puxcatan town at the coordinates 17° 27′ 38.04″ N and 92° 39′ 28.5″ W; seven samples were located at the Abrigo Rocoso Fidencio López, Ejido Lázaro Cárdenas, at the coordinates 17° 31′ 21.35″ N and 92° 47′ 54.21″ W; and two samples were located at the Sima Cuesta Chica, Ejido la Pila, at the coordinates 17° 26′ 33.59″ N and 92° 44′ 31.53″ W. Figure 1 shows the geographic locations of all ancient samples. The archaeological samples recovered from the Puyil cave were found by Luis Alberto Martos-López in 2007 and Abrigo Rocoso Fidencio López and Sima Cuesta Chica by Eladio Terreros-Espinosa in 2016.

Fig. 1
figure 1

Location of pre-Hispanic samples from Tacotalpa, Tabasco, that belonged to individuals from the Mesoamerican region. The samples were found in the ARLF (Abrigo Rocoso Fidencio López), SCC (Sima Cuesta Chica), and the PC (Puyil cave) at the coordinates 17° 31′ 21.35″ N and 92° 47′ 54.21″ W; 17° 26′ 33.59″ N and 92° 44′ 31.53″ W; and 17° 27′ 38.04″ N and 92° 39′ 28.5″ W, respectively

Full size image

The sex was determined in fifteen of the samples using anthropometric measurements as described previously and according to conservation of the bone (Bass 1987; White and Folkens 2005; Blau and Ubelaker 2009; Ubelaker 2014).

Pretreatment of ancient samples

To eliminate exogenous DNA and surface contaminants, each sample was washed by rinsing in water, and the samples then incubated for 5 min with 10% full-strength Clorox bleach. The supernatant was discarded, the sample was incubated in distilled water for 5 min, and the water was then discarded. After the 3rd rinse with distilled water, the sample was dried by incubation at 37 °C overnight.

All experiments were performed in a clean room, according to the guidelines of working with ancient remains (Adler et al. 2011; Campos et al. 2012). The clean room was routinely cleaned with bleach and equipped with a UV light. All containers were wiped with bleach before and after being placed in the laboratory.

DNA extraction

Contemporary DNA extraction was performed in duplicate using magnetic beads (PerkinElmer, Waltham, MA, USA) with the Prepito DNA Blood250 Kit (PerkinElmer, Waltham, MA, USA) and the Prepito-D instrument (PerkinElmer, Waltham, MA, USA) following the manufacturer’s instructions. The extracted DNA was maintained at − 70 °C. One negative control without DNA was used during the entire process.

Ancient DNA (Supplementary Table 1) was extracted from 20 mg of bone powder by incubation in 200 μl of lysis buffer and 6 μl proteinase K under gentle rotation at 56 °C until lysis was completed, following the manufacturer’s instructions (PerkinElmer, Waltham, MA, USA). Aliquots of the extracted DNA were maintained at − 70 °C until use. Ancient DNA was also extracted in duplicate from 50 mg of bone powder via incubation in 500 μl ethylenediaminetetraacetic acid (EDTA; pH 8, 0.5 M) and 20 μl proteinase K (0.25 mg/ml) under gentle rotation at 37 °C overnight. The supernatant obtained following centrifugation was used for DNA extraction with silica (silicon dioxide; Sigma) as previously described by Rohland and Hofreiter (2007). Purified DNA was eluted in 50 μl TE buffer, and 20 μl aliquots were stored at − 70 °C until use. This method was also performed in the Insectary at CINVESTAV-IPN, where human DNA is not handled. One negative control without DNA was used during the entire procedure. The ancient DNA extracted, amplified, and then sequenced were used for haplogroup identification and replication in four samples from Puyil cave in the Department of Anthropology, University of Kansas, Lawrence, KS. This laboratory is specific to work with ancient DNA. Results were the same from both laboratories.

Analysis of the Y chromosome in silico

The non-homologous regions of the X and Y chromosomes of the reference sequence CRs_human_hg38/Chr Y (NC_000024.10) were analysed using IGV software (Thorvaldsdóttir et al. 2013; Integrative Genomics Viewer 2018) to identify the regions specific to Chr-Y. Table 1 shows the sequences identified in the following fifteen genes: TTTY7, TSPY3, TTTY2, TTTY22, USP9Y, UTY, NLGN4Y, TTTY14, FAM41AY1, AC007359.1, TTTY4, PBY2B, DAZ1, DAZ3, and DAZ4. The sequences were classified into the following 3 groups: (1) the sequence was present only in Chr-Y; (2) the sequence showed similarity with Chr-X; and (3) the sequence was also identified in autosomal chromosomes (Fig. 2).

Table 1 Description of 15 genes of Chr-Y analysed to select specific sequences that allow the identification of the male sex
Full size table
Fig. 2
figure 2

Schematic representation of Chr-Y shown in the PAR1, PAR2, and NRY regions, and the location of the 15 genes used to find specific sequences to identify the male sex in ancient samples

Full size image

Male identification using specific Chr-Y sequences

To demonstrate that the region identified in this study was specific to male individuals, a pair of primers (F_TTTY7-R_TTTY7) specific for the TTTY7 gene were designed to amplify a 140-bp fragment from nucleotides 6453443–6453582. A specific pair of primers (F_TSPY3-R_TSPY3) for TSPY3 amplified a fragment of 121 bp from nucleotides 9400901–9401021. A specific pair of primers (F_TTTY22-R_TTTY22) for TTTY22 amplified a fragment of 124 bp from nucleotides 9811189–9811312, and a specific pair of primers (F_TTTY2-R_TTTY2) for TTTY2 amplified a fragment of 133 bp from nucleotides 9738313–9738445 (Table 2). All primer sets were designed using the program PrimerX-Bioinformatics.org. The contemporary DNA samples were tested with the primer set designed to target the Chr-Y-specific regions to confirm the effectiveness of male sex identification. The following PCR conditions using the enzyme Phusion Hot Start II Fidelity DNA Polymerase (Thermo Fisher Scientific, Waltham, MA, USA) were used: 98 °C for 30 s; 35 cycles at 98 °C for 10 s, 60.5 °C or 61.5 °C (according to the primer pair used; Table 2) for 30 s, and 72 °C for 20 s; and 72 °C for 10 min. After optimization and confirmation of this procedure in the contemporary samples, this assay was tested on the DNA of ancient bone samples at least in triplicate.

Table 2 Primers used to identify the male sex in contemporary and ancient human samples
Full size table

Sanger sequencing was performed using the forward and reverse primers (Table 2) following the manufacturer’s instructions for the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA, USA).

Female and male sex identification using the AMEL gene (AMELX/AMELY)

Sex was determined via amplification of the AMEL gene (AMELX/AMELY) in contemporary human samples using the specific pairs of primers shown in Table 2. Amplicons of 130 bp and 170 bp indicated female and male sex, respectively. After sex identification of the contemporary samples, the test was performed on the ancient bone samples. PCR amplification was performed using the enzyme Phusion Hot Start II Fidelity DNA Polymerase (Thermo Fisher Scientific, Waltham, MA, USA) with the following conditions: 98 °C for 30 s; 35 cycles at 98 °C for 10 s, 52 °C for 30 s (Table 2), and 72 °C for 20 s; and 72 °C for 10 min (Lin et al. 1995).

Results

Analyses of X and Y chromosomes for male sex identification

The TTTY7, TSPY3, TTTY2, TTTY22, USP9Y, UTY, NLGN4Y, TTTY14, FAM41AY1, AC007359.1, TTTY4, PBY2B, DAZ1, DAZ3, and DAZ4 genes specific for Chr-Y (Stelzer et al. 2016; GeneCards-Human Genes n.d.) were analysed using IGV software (Thorvaldsdóttir et al. 2013; Integrative Genomics Viewer 2018) to identify the regions not homologous with Chr-X and autosomal chromosomes (Table 1). The regions specific to Chr-Y, selected to identify the male sex, were in the following genes: TTTY7 from nucleotide sites 6453472–6453552 and 6457811–6457861; TSPY3 from nucleotide sites 9712721–9712755, 9340012–9340052, and 9338261–9338286; TTTY2 from nucleotide sites 9738350–9738414 and 9740557–9740591; and TTTY22 from nucleotide sites 9811090–9811129 and 9811192–9811267 (Table 1).

Amplification of TTTY7, TSPY3, TTTY2, and TTTY22 gene-specific regions for Chr-Y in contemporary human samples using PCR

To verify the specificity of the sequences selected for each gene in identifying males, amplicons of 140, 121, 124, and 133 bp were obtained for TTTY7 (from nucleotides 6453443–6453582), TSPY3 (from nucleotides 9400901–9401021), TTTY22 (from nucleotides 9811189–9811312), and TTTY2 (from nucleotides 9738313–9738445), respectively. Female and male contemporary samples were used to optimize the test. The expected molecular band size for each gene in the male samples and the absence of the band in the female samples were observed, as shown in Fig. 3. This test was also repeated in ten contemporary DNA samples from five males and five females to confirm our results. The minimal amount of DNA needed to obtain a reliable determination of males with contemporary DNA was 3 ng/μl. The PCR products were sequenced by Sanger, which confirmed the specific amplification of the TTTY7, TSPY3, TTTY2, and TTTY22 gene regions (GenBank accession numbers: MEXCHRYTTTY22, MN385559; MEXCHRYTTTY7, MN385560; MEXCHRYTSPY3, MN385561; MEXCHRYTTTY2, MN385562).

Fig. 3
figure 3

Amplification of specific regions in the TTTY22, TTTY7, TTTY2, and TSPY3 genes to identify male sex in contemporary human DNA samples. The DNA was amplified using the specific primers to amplify each region of Chr-Y in contemporary human samples. The specific amplicon size is indicated by the arrows on the right side, and the molecular marker is displayed on the left side. (FC) Female DNA sample (negative control). The specific amplifications for TTTY22, TTTY7, TTTY2, and TSPY3 genes with sizes of 124, 140, 133, and 121 bp, respectively, are shown as indicated in top of the figure

Full size image

PCR amplification of TTTY7, TSPY3, TTTY2, and TTTY22 sequences from ancient pre-Hispanic samples

The bone remains of this study were discovered in 2007 and 2016 at the following sites: Puyil cave, Abrigo Rocoso Fidencio López, and Sima Cuesta Chica in Tacotalpa, Tabasco, Mexico. Due to the archaeological context, cave environmental conditions, and body-skeleton integrity, the sex identification was determined in all samples using the new molecular approach developed in this study, the AMEL test, and anthropometric measurements.

The sex of the bone samples from Tabasco was determined via amplification of the TTTY7, TSPY3, TTTY2, and TTTY22 genes. The minimal amount of DNA needed to amplify the specific region in Chr-Y in human ancient samples was 15 ng/μl. None of the ancient samples showed amplification of the TTTY22 gene. From a total of nineteen ancient samples, fourteen showed amplification of TTTY2 (Fig. 4a), eleven showed amplification of TTTY7 (Fig. 4b), and eight showed amplification of TSPY3 (Fig. 4c), as shown in Table 3. These results indicated that all samples from Puyil belonged to males (Fig. 4). The samples from Abrigo Rocoso Fidencio López showed amplification of the specific amplicon for Chr-Y in four of the samples, and one sample from Sima Cuesta Chica was a male positive (Fig. 4), which confirms that at least these samples belonged to males. Four of the samples did not show amplification of any of these genes.

Fig. 4
figure 4

Amplification of the specific regions in TTTY2, TTTY7, and TSPY3 genes to identify male sex in ancient human DNA samples. The pre-Hispanic samples from the geographic areas Puyil cave, Abrigo Rocoso Fidencio López, and Sima Cuesta Chica were tested to determine male sex via amplification of the specific regions in TTTY2, TTTY7, and TSPY3 genes. a Amplification of the specific region in TTTY2 gene displaying the specific band of 133 bp. b Amplification of the specific region in TTTY7 gene displaying the specific band of 140 bp. c Amplification of the specific region in TSPY3 gene displaying the specific band of 121 bp. Female ancient DNA was used as a negative control (FC), displayed an absence of any specific band, and male ancient DNA was used as a positive control showing the amplification of the specific band in each gene (MC), and the negative control with no DNA shows the absence of any amplification band (−). The size of the specific amplicon is indicated on the right side of each panel and the molecular weight markers on the left side. The name of each sample is indicated on the top of each line

Full size image
Table 3 Sex identification using different approaches in ancient samples. Sex identification in pre-Hispanic samples from Tacotalpa, Tabasco, was performed via amplification of the specific regions in the TTTY2, TSPY3, TTTY7, and AMELX/AMELY genes and the anthropometric measurement method as indicated in each column
Full size table

Sex identification using the AMELX/AMELY test

To determine the sex of contemporary and ancient samples using molecular techniques, specific regions of the X and Y chromosomes were amplified according to previous publications using the AMELX/AMELY test as described in the “Materials and methods” section (Lin et al. 1995; Ovchinnikov et al. 1998; Tschentscher et al. 2008; Gibbon et al. 2009; Dutta et al. 2017). This procedure produced amplicons of 130 bp for female and 130 and 170 bp for male contemporary DNA samples. The amplicons for Chr-X and Chr-Y in all contemporary samples are shown in Fig. 5a. All ancient samples from the Puyil cave were positive for Chr-X, and four of the samples were positive for Chr-Y (PUXTABMEX001–PUXTABMEX004) (Fig. 5b). Five ancient samples from Abrigo Rocoso Fidencio López were positive for Chr-X (ARFLTABMEX001, ARFLTABMEX003, ARFLTABMEX004, ARFLTABMEX006, and ARFLTABMEX007), and three of these samples were positive for Chr-Y (ARFLTABMEX001, ARFLTABMEX006, and ARFLTABMEX007) (Fig. 5c). The two ancient samples from Sima Cuesta Chica were positive for Chr-X and Chr-Y (SCCTABMEX001 and SCCTABMEX002) (Fig. 5c). These results suggest that at least nine of the nineteen samples corresponded to the males using the AMELX/AMELY test (Fig. 5). The Chr-X amplicon was observed in eight samples, but this result does not indicate that these samples corresponded to the females because DNA that matches the Chr-Y region may be degraded.

Fig. 5
figure 5

Sex identification using the AMELX/AMELY test. To determine the sex of contemporary and ancient samples, specific regions of the X and Y chromosomes of female and male genomes were amplified according to previous publications using the AMELX/AMELY test described in the “Materials and methods” section. This procedure showed amplicons of 130 bp for females and 130 and 170 bp for males in contemporary DNA samples. a The amplicons for X and Y chromosomes in contemporary samples using the AMELX/AMELY test are shown; the arrows show amplicons of 130 bp for females (Chr-X) or 130 (Chr-X) and 170 bp (Chr-Y) for males. The amplicons in ancient samples from the archaeological sites are shown in b the Puyil cave, c Abrigo Rocoso Fidencio López, and Sima Cuesta Chica. The amplicon sizes are shown by the arrows on the right side, and the molecular weighs are shown on the left side

Full size image

Sex identification by anthropometric measurements

The bone remains used in this study for each sample and the results of the sex identification of fifteen of the samples that were subjected to anthropometric measurements are displayed in Table 3 and Supplementary Table 1. The sex was determined in fifteen of the samples using anthropometric measurements as described in the “Materials and methods” section and according to conservation of the bone remains. Twelve of the fifteen bone samples characterized using this approach were found to be from males. The sex identification of five of these ancient samples was confirmed using the amelogenin test, and ten of the samples were confirmed using the anthropometric procedure.

Discussion

To correctly identify males in ancient samples, we developed a new approach to identify males in ancient samples using the amplification of four specific regions in the TTTY7, TSPY3, TTTY2, or TTTY22 genes of Chr-Y that are absent in the Chr-X and autosomes. The analysis of these four genes using IGV software (Thorvaldsdóttir et al. 2013; Integrative Genomics Viewer 2018) indicated that they are exclusively found in males. The results of this study confirmed the specificity of this region for the male sex in samples of contemporary human DNA (Fig. 3). Confirmation of the specificity for male was demonstrated as well as the absence of the amplification of these genes in contemporary DNA samples from females.

The TTTY7 gene region encodes a non-coding RNA (Y-linked 7) (Stelzer et al. 2016; GeneCards-Human Genes n.d.) and was amplified from contemporary and ancient human DNA samples from Mexico (Fig. 4b), confirming the male sex in eleven of the ancient samples, nine from the Puyil cave, and two from Abrigo Rocoso Fidencio López.

The TSPY3 gene, which encodes the male-specific protein Y-linked 3 gene (Stelzer et al. 2016; GeneCards-Human Genes n.d.) showed amplification of the specific region in eight of the ancient samples, six from the Puyil cave, and two from Abrigo Rocoso Fidencio López (Fig. 4c), indicating that they were males.

The TTTY2 gene, which is transcribed into a non-coding RNA (Y-linked 2) (Stelzer et al. 2016; GeneCards-Human Genes n.d.), was amplified in nine samples from the Puyil cave, four samples from Abrigo Rocoso Fidencio López, and one sample from Sima Cuesta Chica (Fig. 4a).

The TTTY22 gene, which is transcribed into a testis-specific non-coding RNA (Y-linked 22) (Stelzer et al. 2016; GeneCards-Human Genes n.d.), did not amplify in any of the ancient samples, which may be related to allelic dropout (Takayama et al. 2009; Kim et al. 2013; Stevens et al. 2017) or DNA degradation in ancient DNA samples. Ancient DNA degradation depends on the environmental conditions and the microbial activity in the decaying tissues (Lindahl 1993). Environmental conditions that favour ancient DNA survival from tissues are low temperatures and dryness, which would limit hydrolytic and oxidative processes (Lindahl 1993; Dabney et al. 2013). Overrepresented purines (depurination) at positions adjacent to the breaks in the ancient DNA may also contribute to its degradation (Willerslev and Cooper 2005; Briggs et al. 2007). Consequently, this DNA region is not a good candidate to determine the male sex in ancient samples.

AMELX/AMELY was also tested to specifically identify sex (Stone et al. 1996; Esteve-Codina et al. 2008; Butler and Li 2014; Dutta et al. 2017). The results from using the AMELX/AMELY method indicated that nine of the samples were males, and eight samples may be females.

Anthropometric measurements of bones (Stone et al. 1996; Veroni et al. 2010; Harris and Case 2012; Krishan et al. 2016; Ubelaker and DeGaglia 2017; Hora and Sládek 2018; García-Campos et al. 2018a, b) have been largely used to identify sex. Fifteen of the nineteen samples were tested for sex determination using the anthropometric measurements and results showed twelve males and three females. It is worth emphasizing that the method of sex identification using anthropometric measurements of full adult skeletons, a coxal femur bone, a right iliac bone, and three out of four skulls correctly identified males. These results are in agreement with previous studies that successfully identified the sex of adult full skeletons and coxal femur bones (Ovchinnikov et al. 1998; Veroni et al. 2010; Harris and Case 2012; Krishan et al. 2016; Ubelaker and DeGaglia 2017; Hora and Sládek 2018; García-Campos et al. 2018a, b). Similar approaches have been also used in the sex determination of newborns and subadults. However, they are more laborious than the sex identification of adult skeletons and their accuracy of sex identification estimation is lower in subadult than in adult skeletons (Fazekas and Kósa 1978; Schutkowski 1987; Luna et al. 2017). Therefore, the molecular procedures should be preferred when skeletons or useful bone remains are not available for sex determination.

Amplification of the specific regions in the TTTY7, TSPY3, or TTTY2 genes from ancient bone samples was positive in fifteen of the nineteen samples (Table 3). Four of these fifteen samples were also positive using the AMELX/AMELY procedure, and nine samples were positive using the anthropometric measurement method. Four samples that were negative for all the male-specific DNA markers proposed to identify males in the ancient samples were positive for the AMELX/AMELY test, and two of these samples were also positive using the anthropometric measurement method.

The molecular procedures developed in this study used along with the AMELX/AMELY and anthropometric measurements are very useful in confirming the male sex from the samples. From a total of nineteen samples in the present study that were identified as males, seven samples showed the specific amplicons for the three genes of this study; four samples showed two of the specific gene regions; and four samples showed one of the specific gene regions, which gives a total of fifteen samples positive for male identification. Four samples of a total of nineteen which were not amplified any of the three new gene regions proposed in this study tested positive for Chr-Y using the AMELX/AMELY method. Although seventeen samples showed the amplicon for Chr-X, only nine of these samples were identified as males using the AMELX/AMELY method (Table 3). Therefore, the method developed here is complementary and accurate in identifying males from ancient samples and/or useful for forensic analysis. These results also indicate that the AMELX/AMELY method is not always accurate in identifying males. This may be due to deletion of a large fragment and mutations of the amelogenin gene in the Y-homologue, which results in the process known as allelic dropout that leads to gender typing errors (Esteve-Codina et al. 2008; Takayama et al. 2009; Kim et al. 2013; Stevens et al. 2017). This depends on the type of population, possibly specific to the lineage, and inherited deletions (Kashyap et al. 2006). Similarly, failure of the amplification of the TTTY22 gene may be due to allelic dropout or DNA degradation of ancient DNA samples.

Molecular procedures are very important to identify sex in ancient and forensic samples using the specific TTTY7, TSPY3, or TTTY2 genes and the AMELX/AMELY test (Table 3) in addition to anthropometric measurements when a correct bone sample is available. Therefore, we recommend sex identification by the simultaneous use of molecular procedures for the TTTY7, TSPY3, or TTTY2 gene male-specific DNA markers identified in this study and the AMELX/AMELY test.

The molecular identification method has certain advantages over morphometric methods, because it is not affected by individual variation in the size and the morphology of skeletal materials. It can be used to identify sex from foetal and subadult remains and is not restricted by physical fragmentation, requiring only one complete bone or tooth for sex determination. However, molecular methods have several constraints: (a) molecular contamination, which can usually be avoided by using strict laboratory conditions (Llamas et al. 2017); (b) molecular preservation of the specimen, which is the greatest challenge for molecular methods, because of the difficulty in predicting the level of preservation (Lindahl 1993); (c) environmental factors, which can induce molecular degradation that severely impairs the process of obtaining DNA for forensic analysis (Dabney et al. 2013); (d) molecular methods can be costly and thus their use is often restricted to forensic material where other methods are not useful; and (e) molecular methods are destructive in obtaining DNA from an ancient bone remain. However, it is possible to extract DNA from bones with minimal damage to the ancient bone because ancient DNA extraction needs minimal quantities (25–700 mg). Of importance is that bone destruction is justifiable in cases when the results are likely to inform on debates to test hypotheses, and/or when their destruction is not detrimental to other research areas (Kaestle and Horsburgh 2002).

The anthropometric methods are usually based on the assessment of more than one anatomical feature from the skeleton, and their effectiveness depends, to a great extent, on the state of conservation of the anatomical structures, which in many cases are of poor quality, with these methods not yielding conclusive results from subadult skeletons.

Conclusions

The new approach developed to identify male sex in ancient samples via amplification of the male-specific regions of the TTTY7, TSPY3, or TTTY2 genes is a useful tool to confirm male sex in ancient human remains and/or forensic studies when sex cannot be determined using anthropometric measurements or the AMELX/AMELY test. This approach is especially important in ancient samples and forensic research when the age of the individual is unknown, the availability of the bone under study is limited, or the DNA is highly damaged. Furthermore, these genes may also be searched in silico to identify sex if massive sequencing is developed. Since male sex identification was not assessed in all used samples, we would like to stress the value of using complementary approaches in sex identification of human remains, especially when methods applied in one study may not be applicable to others with the same level of confidence.