In this paper we examine the effects of ethnicity on the gene flow between two groups living in Limón, Costa Rica. Our main interest is to determine if ethnicity has acted as a barrier to the exchange of genes, and if the groups have remained distinct genetically. We report the admixture estimates, F(st) values, and inbreeding coefficients of the two samples. The data consist of blood samples and surnames obtained from 375 individuals. The subjects' two surnames were analyzed to determine the ethnicity of their parents (individuals carry their father's and mother's first surnames). We used the formula of Crow and Mange ([1965] Eugen Q 12:199-203) to compute F(t), F(n), and F(r) with the surnames. Admixture estimates were computed for both groups using the computer program ADMIX.PAS kindly provided by Jeffrey Long. The estimates for the Hispanic-Limonense group are M1 = 0.5866 European, M2 = 0.3383 Amerindian, and M3 = 0.0751 African ancestry. For the Afro-Limonense group, the admixture estimates indicate M1 = 0.1047 European, M2 = 0.1357 Amerindian, and M3 = 0.7595 African ancestry. The F(st) values are F(st) = 0.00558 for the Hispanic group and F(st) = 0.05137 for the Afro-Limonense group. These F(st) values indicate that the Afro-Limonense group has experienced more genetic drift than has the other group, possibly as a result of its long history of isolation in Costa Rica. Indeed, when plotted along a scaled eigenvector R matrix of Caribbean gene frequencies, the two Limonense groups did not cluster with each other. Thus we conclude that the two ethnic groups have remained distinct breeding populations.

No abstract

Individuals belonging to six different Amerindian tribes and two African groups of Costa Rica were genotyped for factor V Leiden (FV), factor V haplotype HR2 (FV HR2), Factor II 20210G>A (FII), the methylenetetrahydrofolate reductase (MTHFR), factor VII polymorphisms (FVII IVS7, FVII R353Q), factor XIII (FXIII V34L), and the insertion/deletion (I/D) polymorphism of the gene of angiotensin converting enzyme (ACE). Clear differences in the prevalence were found and are first reported. The prevalence of some of the established genetic risk factors was low in Amerindians of Costa Rica (ACE) or even absent (FVL, FII), and others (MTHFR, FVHR2) had an extremely high prevalence. People of African origin carried very rare FVL or FII polymorphisms, but the DD genotype of ACE is the highest reported. Concerning the protective factors, the QQ genotype of FVII R353Q was absent in Amerindians, but the protective 7/7 genotype of FVII IVS7 frequently found. Novel alleles of FVII IVS7 (4, 8, and 9 monomers) were found. Intertribal heterogeneity was observed that may reflect the evolutionary history of these tribal groups and their admixture with other populations.

Variation in the frequency of twinning among human populations has been presumed to reflect genetic differences. It has been commonly reported that populations of African ancestry have the highest, those of Asian ancestry the lowest, and those of European and Middle-Eastern ancestry intermediate frequencies of twinning. Populations from the Americas have been reported to have intermediate twinning frequencies, presumably reflecting their admixture. In this context, Madrigal (1994. Am J Hum Biol 6:215-218) reported virtually identical (and high) twinning frequencies in two Costa Rican ethnic groups, one of African, the other of Euro-Amerindian ancestry. These frequencies were interpreted in light of frequent inter-ethnic unions, and it was predicted that the two groups would not differ substantially in gene frequencies of several blood enzyme systems. This paper reports the gene frequencies of both groups for such systems. The samples differ significantly for systems that have clearly different frequencies in African and European populations. Given that the groups are actually different in gene frequencies and not homogenous as predicted earlier, the conclusion that twinning frequencies are similar as a result of a similar genetic make up can be questioned. The results challenge the assumption that if populations have similar twinning frequencies it is because they are genetically similar and argue for a stronger environmental component for twinning frequencies.

No abstract