Circum-Mediterranean influence in the Y-chromosome lineages associated with prostate cancer in Mexican men: A Converso heritage founder effect?

Abstract

Prostate cancer is the second most common neoplasia amongst men worldwide. Hereditary susceptibility and ancestral heritage are well-established risk factors that explain the disparity trends across different ethnicities, populations, and regions even within the same country. The Y-chromosome has been considered a prototype biomarker for male health. African, European, Middle Eastern, and Hispanic ancestries exhibit the highest incidences of such neoplasia; Asians have the lowest rates. Nonetheless, the contribution of ancestry patterns has been scarcely explored among Latino males. The Mexican population has an extremely diverse genetic architecture where all the aforementioned ancestral backgrounds converge. Trans-ethnic research could illuminate the aetiology of prostate cancer, involving the migratory patterns, founder effects, and the ethnic contributions to its disparate incidence rates. The contribution of the ancestral heritage to prostate cancer risk were explored through a case-control study (152 cases and 372 controls) study in Mexican Mestizo males. Seventeen microsatellites were used to trace back the ancestral heritage using two Bayesian predictor methods. The lineage R1a seems to contribute to prostate cancer (ORadjusted:8.04, 95%CI:1.41-45.80) development, whereas E1b1a/E1b1b and GHIJ contributed to well-differentiated (Gleason ≤ 7), and late-onset prostate cancer. Meta-analyses reinforced our findings. The mentioned lineages exhibited a connection with the Middle Eastern and North African populations that enriched the patrilineal diversity to the southeast region of the Iberian Peninsula. This ancestral legacy arrived at the New World with the Spanish and Sephardim migrations. Our findings reinforced the contribution of family history and ethnic background to prostate cancer risk, although should be confirmed using a large sample size. Nonetheless, given its complex aetiology, in addition to the genetic component, the lifestyle and xenobiotic exposition could also influence the obtained results.

Fig 1. Map of the Mexican Republic depicting the frequency distributions of the haplogroups found in the present study distributed by birthplace and its comparison with previous studies.

Note: The Central-East region comprising the states of Hidalgo, Estado de México, Morelos, Puebla, Querétaro, and Tlaxcala; the Central-West including the states of Aguascalientes, Colima, Guanajuato, Jalisco, and Michoacán; and the Central Valley of Mexico (CVM) was represented by Mexico City. The East region included the states of Tabasco and Veracruz. Regarding the North region it involved the states of Chihuahua, Coahuila, Nayarit, San Luis Potosí, Sinaloa, and Zacatecas whereas the South one included the states of Chiapas, Guerrero, Oaxaca, and Yucatán. The frequencies of population data were obtained from published reports: Lopez-Ramirez et al (2020); Martinez-Cortes et al (2012), and Santana et al (2014). Nonetheless, not all the states included by region were reported in these articles.

Fig 2. Mean influence of the ancestral burden of the Y-chromosome lineages that contributed to prostate cancer obtained from linear discriminate analyses.

Note: The proportions included regions within the Mexican Republic, the Iberian Peninsula and those ancestral populations that contributed to their genetic architecture such as ANG, Anatolian Greeks; ARAG, Arabian Gulf, CHU, Chuetas; COR, Corsica; CVM, Central Valley of Mexico; CYP, Cyprus, FRA, France; IBP, Iberian Peninsula; ITA, Italy; LBN, Lebanon; LBY, Libya; MEX, Mexico; NL, Nuevo Leon; PRT, Portugal; T&C, Turkish and Cypriots; UGA, Uganda.

Fig 3. Mean influence of the ancestral burden of the Y-chromosome lineages that contributed to prostate cancer obtained from linear discriminate analyses in samples from the Nuevo Leon region.

Note: The proportions included the Iberian Peninsula and those ancestral populations that contributed to its genetic architecture such as CYP, Cyprus, FRA, France; IBP, Iberian Peninsula; GRC, Greece; IBP, Iberian Peninsula; ITA, Italy; MEA, Middle East; PRT, Portugal; UGA, Uganda.

Fig 4

Median-joining network (A), and multidimensional scale plot of R_ST values (B) of the E1b1a lineage with 14 Y-STRs using ethnicity as a criterion. Note: ARAG, Arabian Gulf; CVMs (Central Valley of Mexico, data from Santana et al); CYP, Cyprus; ITA, Italy; LBN, Lebanon; MEX, Mexico; NL, Nuevo Leon; PRT, Portugal; UGA, Uganda; YUC, Yucatan. Numbers in red represent the number of differences between one haplotype and other(s). All p values were adjusted with the method of false discovery rates. The dotted lines indicate that no significant differences were found among the populations. The circle size indicates the approximate number of individuals with shared haplotypes; the smallest circles represent one individual.

Fig 5

Median-joining network (A), multidimensional scale plot of R_ST values (B) of the E1b1b lineage with 14 Y-STR using ethnicity as a criterion. Note: BAR, Barcelona; CAT, Cataluña; CVMs (Central Valley of Mexico, data from Santana et al); CYP, Cyprus; FRA, France; GIR, Girona; IBP, Iberian Peninsula; ITA, Italy; LBN, Lebanon; LBY, Libya; LLE, Lleida; MEX, Mexico; NL, Nuevo Leon; TAR, Tarragona. Numbers in red represent the number of differences between one haplotype and other(s). All p values were adjusted with the method of false discovery rates. The dotted lines indicate that no significant differences were found among the populations. The circle size indicates the approximate number of individuals with shared haplotypes; the smallest circles represent one individual.

Fig 6

Median-joining network (A), multidimensional scale plot of R_ST values (B) of the G lineage with 14 Y-STRs using ethnicity as a criterion. Note: AND, Andalusia; ANG, Anatolian Greeks; ARA, Aragon; CAT, Cataluña; CHU; Chuetas; COR, Corsica; CVA; Valencian Community; CVMg (Central Valley of Mexico, data from Gomez et al); CVMs (Central Valley of Mexico, data from Santana et al); CYL, Castilla and Leon; GAL, Galicia; IBP, Iberian Peninsula; MEX, Mexico; NAV, Navarra; NL, Nuevo Leon; PVA, Basque Country; SEPH-PRT, Sephardic Jews from Portugal; T&C, Turkish and Cypriots. Numbers in red represent the number of differences between one haplotype and other(s). All p values were adjusted with the method of false discovery rates. The dotted lines indicate that no significant differences were found among the populations. The circle size indicates the approximate number of individuals with shared haplotypes; the smallest circles represent one individual.

Fig 7

Median-joining network (A), multidimensional scale plot of R_ST values (B) of the I lineage with 14 Y-STRs using ethnicity as a criterion. Note: AND, Andalusia; ANG, Anatolian Greeks; ARA, Aragon; AST, Asturian Community; CAT, Cataluña; CHU; Chuetas; CLM, Castilla la Mancha; CVA; Valencian Community; CVMg (Central Valley of Mexico, data from Gomez et al); CVMs (Central Valley of Mexico, data from Santana et al); CYL, Castilla and Leon; EXT, Extremadura; FRA, France; GAL, Galicia; GTO, Guanajuato; IBP, Iberian Peninsula; ITA, Italy; MAD, Madrid; MEX, Mexico; MUR, Murcia; NAV, Navarra; NL, Nuevo Leon; PVA, Basque Country; RIO, La Rioja; SEPH-PRT, Sephardic Jews from Portugal; T&C, Turkish and Cypriots. Numbers in red represent the number of differences between one haplotype and other(s). All p values were adjusted with the method of false discovery rates. The dotted lines indicate that no significant differences were found among the populations. The circle size indicates the approximate number of individuals with shared haplotypes; the smallest circles represent one individual.

Fig 8

Median-joining network (A), multidimensional scale plot of R_ST values (B) of the J lineage with 14 Y-STRs using ethnicity as a criterion. Note: ARA, Aragon; CANT, Cantabria; CAT, Cataluña; CHU; Chuetas; CLM, Castilla la Mancha; COR; Corsica; CVA; Valencian Community; CVMg (Central Valley of Mexico, data from Gomez et al); CYL, Castilla and Leon; EXT, Extremadura; FRA, France; GTO, Guanajuato; IBP, Iberian Peninsula; ITA, Italy; MAD, Madrid; MAJ, Majorca; MEX, Mexico; MUR, Murcia; NL, Nuevo Leon; PVA, Basque Country; RIO, La Rioja. Numbers in red represent the number of differences between one haplotype and other(s). All p values were adjusted with the method of false discovery rates. The dotted lines indicate that no significant differences were found among the populations. The circle size indicates the approximate number of individuals with shared haplotypes; the smallest circles represent one individual.

Fig 9

Median-joining network (A), multidimensional scale plot of R_ST values (B) of the R1a lineage with 14 Y-STRs using ethnicity as a criterion. Note: AND, Andalusia; ARA, Aragon; ASH, Ashkenazi Jews; CAT, Cataluña; CHU; Chuetas; COHEN, Cohen Jews; CVA; Valencian Community; CVMg (Central Valley of Mexico, data from Gomez et al); CVMs (Central Valley of Mexico, data from Santana et al); CYL, Castilla and Leon; EXT, Extremadura; GAL, Galicia; IBP, Iberian Peninsula; ISR, Jews from Israel; MAD, Madrid; MAJ, Majorca; MEX, Mexico; NL, Nuevo Leon; SEPH-IBP, Sephardic Jews from the Iberian Peninsula; SEPH-PRT, Sephardic Jews from Portugal; T&C, Turkish and Cypriots. Numbers in red represent the number of differences between one haplotype and other(s). All p values were adjusted with the method of false discovery rates. The dotted lines indicate that no significant differences were found among the populations. The circle size indicates the approximate number of individuals with shared haplotypes; the smallest circles represent one individual.