Anthropological survey on red cell glutathione peroxidase (GPX1) polymorphism in Central Western Africa A tentative hypothesis on the interaction between GPX1.13307902 and Hb.pdfS allelic productsкод для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 79217-224 (1989) Anthropological Survey on Red Cell Glutathione Peroxidase (GPXI) Polymorphism in Central Western Africa: a Tentative Hypothesis on the Interaction between GPXI*2 and Hbp*S Allelic Products GIOVANNI DESTRO-BISOL AND GABRIELLA SPEDINI Departments of Human and Animal Biology and Anthropology, University of Rome “La Sapienza,” 1-00185Rome (G.S.)and Institute of Forensic Medicine, Catholic University, 1-00168,Rome (G.D.-B.),Italy KEY WORDS Mbugu, Sango, Goun, Bamileke ABSTRACT Phenotype and allele frequencies for erythrocyte glutathione peroxidase (GPX1) polymorphism are reported in the Mbugu and Sango (Central African Republic), Goun (Benin), and Bamileke (Cameroon) ethnic groups. The GPX1*2 allele frequencies (from 0.012 in the Sango to 0.058 in the Bamileke) fit into the range of the data already known for the Subsaharan populations. The value of GPXl*2 for study of the genetic admixture between Negro and Pygmy populations is suggested. Three different unusual GPXl electrotypes are described. Finally, we hypothesize a n interaction between GPX1*2 and Hb beta*S allelic products occurring in the sickle cells infected by Plasmodium falciparum. For almost 15 years the staff of the Laboratory of Anthropology at the University of Rome “La Sapienza” has been actively engaged in the anthropogenetic study of the Subsaharan populations. In particular, we have concentrated on the biological interrelations of the Subsaharan populations and the possible interactions between the strong selective forces to which these groups are exposed and some of their biological characteristics (Spedini et al., 1980a,b, 1981, 1982, 1983, 1985-86, 1986). Erythrocyte glutathione peroxidase (GPX1: EC 22.214.171.124) (Mills, 1957) is a selenoenzyme (Flohe et al., 1973; Rotruck et al., 1973) that catalyzes the oxidation of reduced glutathione (GSH) by peroxides, and is thus vital in the defence against oxidative damage. The importance of this enzyme within the biochemical pathway of red blood cells is shown by the haemolytic anaemia caused by GPXl deficiency (Necheles et al., 1970), along with the noticeable variations of GPXl activity observed in various haemolytic diseases: HbS (Chiu and Lubin, 1979; Das and Nair, 1980; Ponzetto-Zimmermann and Natta, 1981; Beretta et al., 1983); GGPD deficiency (Beutler, 1977; Swarup-Mitra, 1977; Gerli et al., 1982; 0 1989 ALAN R. LISS, INC. Mavelli et al., 1984); alpha-thalassaemia (Beutler et al., 1977); and beta-thalassemia (Gerli et al., 1980). The electrophoretic polymorphism of GPXl, first reported in US blacks (Beutler and West, 1974), is determined by the two codominant alleles GPX1*1 and GPX1*2 (following the nomenclature introduced by Meera Khan et al., 1984a).The previous GPXl studies (Beutler and West, 1974; Beutler et al., 1974; Nurse and Jenkins, 1976; Board, 1983; Meera Khan et al., 1984a, 1986; Destro-Bisol et al., 1986; Meera Khan, 1986) have pointed out that GPX1*2 is present only among Subsaharans and in populations with a probable African genetic heritage, thus supporting the theory of “African marker” value of GPX1*2 (Meera Khan et al., 1986). Some rare variants have been observed: GPXl “Musi” (Beutler et al., 1974); GPXl “Lebanese” homozygous and heterozygous variants (Board, 1983); GPXl “Djuka” or GPXl 3-1 and GPXl 4-1 (Meera Khan et al., 1984b).However, there have been no reported family studies demonstrating the genetic Received March 20, 1987; accepted July 20, 1988. 218 G. DESTRO-BISOL AND G. SPEDINI inheritance of any of these variants. Recently Prof. P. Meera Khan informed us that he had encountered three further GPXl variants, including a silent allele in some Jewish populations. Interestingly, the catalytic activity of the GPX1*2 allele product was found to be almost twice that of GPX1*1 (Meera Khan et al., 1986). Given the importance of the GPXl physiological role and the considerably higher activity of the GPX1*2 allele product, it would be interesting to investigate a possible selective role of the latter polymorphism. We have previously reported the GPXl frequencies obtained from the Beti, Bateke, and Babenga populations of the Congo (DestroBisol et al., 1986).In this paper we present the GPXl data relative to the Goun (Benin), Mbugu and Sango [Central African Republic (CAR)], and Bamileke (Cameroon) populations. All the groups examined have been settled in Central-Western Africa for some time, where they and other populations have interacted sometimes peacefully, sometimes violently, and where there has been a great spreading of malarial parasites (Fig. 1). POPULATIONS According to the linguistic classification of Greenberg (1980), the Goun, Sango, and Mbugu populations belong to the non-Bantu or West African family of the Niger-Congo Fig. 1. Geographic location of the populations examined in this survey on GPXl polymorphism in CentralWestern Africa: 1,Mbugu;2, Sango;3, Bamileke;4, Goun;5, cluster, while the Bamileke of Cameroon fit into the Bantu “lato-sensu” group of the Benue Congo family. The Goun live in the southern region of Benin and the Mbugu and Sango in the Basse Kotto district on the banks of the Ubangui river in the CAR. Details of the history and distribution of some other genetic markers in the examined populations are reported in Spedini et al., 1980a and Destro-Bisol et al., 1987 for the Goun; Spedini et al., 1983 and Destro-Bisol et al., 1987 for the Mbugu and Sango; and Spedini et al., 1988, in preparation for the Bamileke. MATERIALS AND METHODS The material included blood samples from 45 Mbugu and 47 Sango of the Basse Kotto district (CAR) collected in 1979 and from 18 Goun residing in Port0 Novo (Benin) collected in 1980. The methods of blood collection, shipment, and storage are reported in Spedini et al. (1983). In addition, during a n expedition led by G.S. in 1985 in SouthWestern Cameroon, 452 blood samples were collected from a s many Bamileke residing in three different parts of the Department of Moungo: 259 from Manengole village, 116 from Kongsamba Town (including 11related individuals), and 77 from Ndoungue Village, of which 54 were from pregnant women. The blood samples, obtained by venipuncture, South Bateke; 6 , Babenga Pygmies; 7, North Bateke; 8, Beti. 219 GLUTATHIONE PEROXIDASE IN AFRICA were collected in K3-EDTA tubes a s in our previous GPXl survey (Destro-Bisol et al., 1986). The samples were separated, salinewashed, and then stored separately a s serum and washed packed red cells a t -20°C within 12 hours of collection in the laboratory of the Centres des Grandes Endemies a t Kongsamba. The samples were then flown on ice to Rome within 12days and transported directly to the Laboratory of Anthropology at the University of Rome “La Sapienza,” where they were stored at -30°C before electrophoresis. Preparation of samples and electrophoretic procedures were performed in accordance with Meera Khan et al. (1984a), except that 0.05 M dithiothreitol (DTT) was used as lysis solution instead of 0.1 M 2-mercaptoethanol (2-ME).Samples exhibiting unusual patterns were treated by 0.05 or 0.1 M DTT, 1M or 2 M 2-ME, or 0.05 M GSH. For gene counting only unrelated individuals were taken into account. RESULTS Of the 562 samples analyzed, 492 were easily typed a s one of the previously reported GPXl 1,2-1, and 2 phenotypes; the remaining 70 samples showed unusual electrotypes (Fig. 2). I n order to avoid any mistyping caused by protein denaturation, the samples exhibiting unusual electrotypes were treated,with 2-ME and GSH solutions (Beutler and West, 1974), but no change in the electrophoretic pattern was observed in any of these samples. Of the Bamileke samples, 53 showed a single GPXl band slightly faster than that of the GPXl 1 type. This unusual electrotype was named “Bamileke.” It was possible to perform a family analysis for one carrier of “Bamileke” variant of the Nkongsamba group, who was excluded from gene counting. The paternity was confirmed by the determination of 12 other gene markers, including red cell enzymes and serum proteins; both parents of the propositus showed the GPXl 1 phenotype. This sample and the other identical variants were therefore considered a s GPXl 1 types. In addition, GPXl 1 and “Bamileke” electrotypes were found to be undistinguishable by ultrathin-layer polyacrylamide gel isoelectric focusing in the range of pH 4-6 (Destro-Bisoland Briziobello, 1987). Thirteen Mbugu and four Sango showed two different unusual electrotypes, which we named “A” and “B.” For these last two variants, family data were not available; as a precautionary measure, they were excluded from gene counting. The results are summarized in Table 1. + 0 I PHENOTYPE I 2-2 ’BAM n X 2-1 “B’ 1-1 Fig. 2. Diagrammatic representation and zymograms of some GPXl phenotypes. Variants “A” and “B’ were 2.2 ‘BAM’ ‘A’ 2-1 ‘6’ drawn following the interpretation of Prof. Meera Khan (see text). 220 G. DESTRO-BISOLAND G. SPEDINI TABLE 1. Distribution of GPXI phenotypes and allele frequencies in the examined populations1 Pouulation 1 CAR Mbugu2 0bserved Expected Sang03 Observed Expected Benin Goun Observed Expected Cameroon Barnileke (Manengole) Observed Expected (Kongsambayl Observed Exuected (Ndo~ngue)~ Observed Expected Total6 Observed Expected GPXl phenotype 2-1 2 Total Allele frequency GPX1*1 GPX1*2 X:: 29 29.1 3 2.8 0 0.1 32 32.0 0.953 f 0.026 0.047 - 42 42.0 1 1.0 0 0.0 43 43.0 0.988 j, 0.010 0.012 - 17 17.0 1 1.0 0 0.0 18 18.0 0.972 zk 0.027 0.028 - 228 227.9 30 30.1 1 259 259.0 0.938 f 0.011 0.062 0.001 1.0 91 90.6 13 13.9 1 0.5 105 105.0 0.929 k 0.018 0.071 0.562 73 73.1 4 3.9 0 0.1 77 77.0 0.974 j, 0.01 3 0.026 - 392 391.3 47 48.2 2 1.5 441 441.0 0.942 f 0.008 0.058 0.188 lThese data were communicated at the Fifth Congress of the European Anthropological Association, Lisboa, September 28 to October 4, 1986. 2Ten “A”and three “B’ variants were excluded. 3Three “ A and one “B’ variants were excluded. 4Includes eight “Bamileke” variants; 11 related individuals (including one GPXl “Bamileke”, one GPXl 2, and nine GPX1) were excluded. SIncludes 44 “Bamileke” variants. 6No statistically significant difference was found among the three Barnileke subgroups by a classical chi-square test for comparison. DISCUSSION Unusual variants In terms of electrophoretic mobility and the single banded pattern, the variant phenotypes encountered among the Bamileke resemble the GPXl “Musi” variant, previously found in two related individuals affected by refractory anaemia and in several cord blood samples (Beutler et al., 1974). It is interesting to note that 33 of the 53 “Bamileke” variants were found in the group of 54 pregnant women included in the Ndongue sample. Unfortunately, the “Musi” sample was not available for a comparison (Prof. E. Beutler, Scripps Foundation, La Jolla, CA, personal communication). However, the fact that for both the “Bamileke” and the “Musi” variants the patterns of inheritance did not confirm the genetic origin of the changed electrophoretic mobility supports the identity of these two variants. It is worth mentioning that a remarkable decrease in the selenium content was found in both the blood of pregnant women and in cord blood (Rudolph and Wong, 1978; Behne and Wolters, 1979; Butler et al., 1982;Pleban et al., 1982).These findings led us to suggest tentatively that a scarcity of blood selenium could be related to the changed electrophoretic mobility of the “Musi” variants found in the cord blood samples and of the “Bamileke” variants found among the pregnant women of Ndoungue. It is well known that lack of a n inorganic element of the prosthetic group of proteins can cause a change in their conformation and, consequently, a different electrophoretic mobility. Obviously our hypothesis await further experimental confirmations. Two different unusual electrophoretic variants, which we have called “A’ and “B,” were found in the CAR populations. They were apparently heterozygotes for the common GPXl allele and two further GPXl alleles. “A” and “B” variants showed a n electrophoretic mobility very similar to those of the “Lebanese variants” (Board, 1983).However, owing to the low activity of the “Lebanese variants” a t our disposal, we could not reach a definite conclusion. Prof. P. Meera 221 GLUTATHIONE PEROXIDASE IN AFRICA Khan, after examining the photograph shown in Figure 2, suggested that the “ A ’ and “B” variants could be respectively the GPXl 3-1 (found among the Djuka of Surinam: Meera Khan et al., 1986) and GPXl 4-1 (previously found in a Negro: Meera Khan et al., 1984b). The Djuka of Surinam are descendants of captives from Central-Western Africa and probably from a wider region of West Africa (Meera Khan et al., 1986). The geographic location of their forefathers is therefore in accordance with the suggested identity between the GPXl “Djuka” and “A” variants. GPXl*2 allele as anthropological marker All the populations examined were characterized by polymorphic frequencies of GPX1*2 (from 0.012 in the Sango to 0.058 in the Bamileke), which fit into the range of values reported for the Subsaharan populations (Table 2). The GPX1*2 allele was not found in any population of non-African origin, with the exception of Ashkenazy Jews of the United States and Punjabis of India; this finding could be the result of their ancient genetic admixture with African populations (Mourant et al., 1976; Meera Khan et al., 1984a).The data obtained by us strongly support the theory of the “African marker” value of GPX1*2 (Meera Khan et al., 1984a) and enable us to outline more exhaustively the GPXl distribution in Subsaharan Africa. When we group the Sudanic population on one side and the Bantu on the other, the GPX1*2 frequencies do not show a noteworthy variation. These results are in accordance with previous findings on other genetic polymorphisms in Subsaharan Africa. As to the Pygmies, data obtained from the Babenga of the Congo are in line with the low frequencies (Babinga of the CAR, Twa of Rwanda) or absence (Mbuti of Zaire and Western Pygmies from Cameroon) of the GPX1*2 noticed in the other Pygmy groups. TABLE 2. GPXl x2 allele freouencies reDorted in Drevious studies Population Negroids Bantus Bamileke Beti North Bateke South Bateke Hutu Tutsi Denasena Zulu West Africans Yoruba Goun Mbugu Sango Lissongo and Bagandu San Glaokx’ate Pygmies Babin ga Mbuti Twa “Western” Babenga Blacks Afro-Jamaicans Afro-Americans Djuka Caucasians’ Ashkenazi Jews Punjabi Indians Country Sample size GPX1*2 Cameroon Congo Congo Congo Rwanda Rwanda Botswana SAR 441 94 79 57 122 16 n.r. 303 0.058 0.058 0.025 0.079 0.045 0.156 0.077 0.058 This study Destro-Bisol et al. (1987) Destro-Bisol et al. (1987) Destro-Bisol et al. (1987) Meera Khan et al. (1986) Meera Khan et al. (1986) Nurse and Jenkins (1977) Board (1983) Nigeria Benin CAR CAR CAR 65 18 32 43 43 0.000 0.028 0.047 0.012 0.093 Ojikutu et al. (1977) This study This study This study Meera Khan et al. (1986) Botswana 33 0.000 Nurse and Jenkins (1977) CAR Zaire Rwanda Cameroon Congo 927 122 188 18 48 0.005 0.000 0.023 0.000 0.011 Meera Khan et al. (1986) Meera Khan et al. (1986) Meera Khan et al. (1986) Meera Khan et al. (1986) Destro-Bisol et al. (1987) Jamaica USA Surinam 72 392 715 0.007 0.032 0.054 Meera Khan et al. (1984a) Beutler et al. (1974) Meera Khan et al. (1986) USA 90 0.011 India 116 0.013 Estimated bv Meera Khan et al. (1984) from Beutler et al. (1974) Meera Khan et al. (1984a) Reference lThe GPXb2 allele was nut found in 110 Norwegians,398 Dutch, and 76 Sardinians(Meera Khan et al., 1984a),300 non-Jewish whites of the U S A . (estimated by Meera Khan et al., 1984a from Beutler et al., 1974) and 47 Marathi and 198 Telugu of India (Meera Khan et al., 1984a). As for Amerindians, G P X M was not found in 149 Quechua, 312 Wajana, and 504 Trio Amerindians (Meera Khan et al., 1984a). 222 G. DESTRO-BISOL AND G. SPEDINI GPXl*2 allele in the HbA and HbAS subsamples of each population (Table 3): the GPX1*2 was found to be absent in some HbAS subsamples (Mbugu, Sango, Goun, Bamileke of Kongsamba and of Ndoungue, North Bateke, and Babenga), or its frequency was noticeably lower in others (Bamileke of Manengole and Beti). Only in the South Bateke was the GPX1*2 frequency higher among the HbAS than among the HbA carriers. This group was found to be significantly different from the Babenga and the Sango with respect to GPXl phenotype distribution (GPX1 2-1 and GPXl 2 2-2 pooled: GPXl*2 and malaria South Bateke vs. Babenga xf= 4.547,0.05 > P Meera Khan et al. (1986) suggested that the > 0.02; South Bateke vs. Sango X ? = 4.104, GPXl*2 allele product could be less favour- 0.05 > P > 0.02), while no significant differable than the GPX1*1 to the survival of Plas- ences were found among the other population modium falciparum because of its ability to samples. Interestingly, when the indepenoxidise a higher rate of GSH, a compound dence of GPX1*2 and HbP*S allele distribution essential for growth and development of the was tested in a 2 X 2 contingency table P. falciparum parasite. This is a very attrac- (GPX1 2-1 and GPX 2-2 pooled-the South tive hypothesis. However, the considerably Bateke were excluded owing to their differhigher peroxidasic activity of the GPX1*2 ence in GPXl distribution), a significant allelic product could have remarkable conse- value was obtained (x?= 7.543, 0.01 > P > quences on other steps of the biochemical 0.001). These results seem therefore to suppattern of erythrocytes. In the case of the port our hypothesis. However, their value Subsaharan populations, it would be of par- must be considered with great caution for the ticular interest to study the possible interac- following reasons: first, this study of interactions between the GPX1*2 allele and the tion between the GPX1*2 allele product and genetic factors involved in the protection HbAS was performed on too small a scale for any firm conclusion to be drawn; and second, against malaria. Friedman (1981) suggested that the prema- a more faithful evaluation of the occurrence ture destruction of the red cells infected by of malaria selection on a n association bemalaria parasites in which HbS, or other tween genetic traits requires the simultaneabnormal hemoglobins, are present could be ous investigation of other biochemical polymainly due to reactive species generated by morphisms (e.g., red cell glucose-6-phosphate the interaction between membrane-bound dehydrogenase) a s well as environmental abnormal hemoglobins and hydrogen perox- factors (e.g., dietary selenium intake) that ide excreted by the Plasmodium. It is known could affect the phenotypic expression of the that GPX1, along with catalase, plays a genetic traits under study and/or the indiprimary role in the elimination of H202 in the vidual resistance to malaria. In conclusion, our findings support the red cells (Cohen and Hochstein, 1963), and a remarkable reduction of GPXl activity could value of GPXl a s a n admixture marker of be involved in the sickling events (Das and Pygmy with neighbouring Negro populations, Nair, 1980). Hence, basing our hypothesis on and stimulate us toward further investigathese findings, it seems possible to conclude tions on the existence of an interaction that the Hbp*S protection against malaria between GPXl*2 and Hb beta*S allelic might be counteracted by the higher peroxi- products. We have further GPXl studies in dase activity of the GPXl*2 allelic product. progress, aimed at checking the suggested The subsequent increase in oxidized gluta- negative association between GPX1*2 and thione (GSSG) might be counterbalanced by Hbp*S alleles. a n increased activity of the NADPH/GSSG reductase system (Wendel, 1980) and/or by ACKNOWLEDGMENTS GSSG efflux from erythrocytes (Srivastava This work was supported by the Minister0 and Beutler, 1969; Sies et al., 1972). In order to check this speculative hypothe- della Pubblica Istruzione. We are indebted to sis, we have compared the frequencies of the Prof P. Meera Khan (University of Leiden, The Babenga have established a gene flow with the neighboring South Bateke (Spedini et al., 1985-86), who are characterized by rather high frequencies of GPX1*2 (0.079; Destro-Bisol et al., 1986). By supporting the notion that the presence of GPX1*2 in the Pygmy group is presumably due to genetic admixture with neighbouring Negro populations (Meera Khan et al., 1986), our data allow us to suggest that GPX1*2 could be a valuable marker in the study of the genetic admixture between Negro and Pygmy populations. 223 GLUTATHIONE PEROXIDASE IN AFRICA TABLE 3. Association of GPXl polymorphism with Hbp types Population CAR Mbugu Sango Benin Goun Cameroon Bamileke (Manengole) (Kongsamba) (Ndoungue) Congo Beti South Bateke Babenga Pooled sample2 Total GPXl*2 frequency 0 0 0 0 28 4 34 9 0.054 0.000 0.015 0.000 0 0 16 2 0.031 0.000 28 1 1 0 235 21 94 13 70 7 0.064 0.024 0.080 0.000 0.029 0.000 1 63 31 66 13 46 0 0 0 3 0 43 4 650 102 0.079 0.016 0.030 0.000 0.076 0.091 0.012 0.000 0.099 0.001 1-1 A AS A AS 25 4 33 9 3 0 1 0 A AS 15 2 1 0 A AS A 206 20 80 11 66 7 53 30 62 13 40 9 42 4 982 100 10 1 AS A AS A North Bateke GPXl phenotype 2-1 2-2 Hbp type’ AS A AS A AS A AS A AS 13 0 4 0 4 0 5 2 1 0 65 2 1 0 0 0 0 0 0 0 11 ‘The data relative to the Hbp types are reported in Spedini et al. (1983)for the Mbugu and Sango;Spedini et al. (198Ob)for the Goun; Spedini et al. (1988) for the Bamileke;and Spedini et al. (1986)for the Beti, Bateke, and Babenga. 2The South Bateke were excluded (see text). The Netherlands) for his help in the interpretation of the unusual GPXl electrotypes, to Prof. P.G. Board (University of Canberra, Australia) for providing us with the “Lebanese” variants, and to Dr. A. Briziobello for his cooperation in GPXl typing. LITERATURE CITED Behne D, and Wolters W (1979) Selenium content and glutathione peroxidase activity in the plasma and erythrocytes of nonpregnant and pregnant women. J . Clin. Chem. Clin. Biochem. 17:133-135. Beretta L, Gerli GC, Ferraresi R, Agostoni A, Gualandri V, and Orsini GB (1983) Antioxidant system in sickle red cells, Acta Haematol. 70:194-197. Beutler E (1977) Glucose-6-phosphate dehydrogenase deficiency and red cell glutathione peroxidase. Blood 49:467-469. Beutler E, and West C (1974) Red cell glutathione peroxidase polymorphism in Afro-Americans. Am. J . Hum. Genet. 26:255-258. Beutler E, West C, and Beutler B (1974) Electrophoretic polymorphism of glutathione peroxidase. Ann. Hum. Genet. 381163-169. Beutler E, Matsumoto F, Powars D, and Warner J (1977) Increased glutathione peroxidase activity in alphathalassemia. Blood 50:647-655. Board PG (1983) Further electrophoretic studies of erythrocyte glutathione peroxidase. Am. J. Hum. Genet. 35:914-918. Butler JA, Whanger PD, and Tripp M J (1982) Blood selenium and glutathione peroxidase activity in pregnant women: Comparative assays in primates and other animals. Am. J. Clin. Nutr. 36:15-23. Chiu D, and Lubin B (1979) Abnormal vitamin E and glutathione peroxidase levels in sickle cell anemia. J . Lab. Clin. Med. 94:542-548. Cohen G, and Hochstein P (1963) Glutathione peroxidase: The primary agent for the elimination of hydrogen peroxide in erythrocytes. Biochemistry 21420-1428. Das SK, and Nair RC (1980) Superoxide dismutase, glutathione peroxidase, catalase and lipid peroxidation of normal and sickled erythrocytes. Br. J . Haematol. 4437-97. Destro-Bisol G, Briziobello A (1987) Adoption of isoelectric focusing (IEF) for the study of red cell glutathione peroxidase (GPX1) polymorphisms: Preliminary results and anthropological perspectives. Proceedings of the Fifth Congress of the European Anthropological Association, Lisboa, September 28 to October 4, 1986, in press. Destro-Bisol G, Briziobello A, Adriani A, and Spedini G (1986) Frequencies of the GPXI*T (or GPX1*2) and CAII*2 alleles in some Congo populations. Hum. Hered. 36:58-61. Destro-Bisol G, Adriani A, Briziobello A, and Spedini G (1987) Further data on the distribution of PGMl (by IEF) and CAI1 polymorphism among the Subsaharan populations (Central African Republic and Benin). Anthropol. Anz. 45:145-152. Friedman MJ (1981) Hemoglobin and the red cell membrane: Increased binding of polymorphic hemoglobins and measurement of free radicals in the membrane. In 224 G. DESTRO-BISOLAND G. SPEDINI GJ Brewer (ed.): The Red Cell. 5th Ann Arbor Conference. New York: Alan R. Liss, pp. 519-531. Flohe L, Gunzler WA, and Schock HH (1973)Glutathione peroxidase: A selenoenzyme. FEBS Lett. 32132-134. Gerli GC, Beretta L, Bianchi M, Pellegatta A, and Agostoni A (1980) Erythrocyte superoxide dismutase, catalase and glutathione peroxidase activities in betathalassaemia (major and minor). Scand. J. Haematol. 2537-92. Gerli GC, Beretta L, Bianchi M, Agostoni A, Gualandri V, and Orsini GB (1982) Erythrocyte superoxide dismutase, catalase and glutathione peroxidase in glucose6 phosphate dehydrogenase deficiency. Scand. J. Haematol. 29:135-140. Greenberg J (1980) Classification des langues d‘Afrique. In: Histoire GQnAraled’Afrique. Vol. I. Paris: Jeune Afrique/UNESCO, pp. 321-346. Mavelli I, Ciriolo MR, Rossi L, Meloni T, Forteleoni G, De Flora A, Benatti U, Morelli A, and Rotilio G (1984) Favism: A hemolytic disease associated with increased superoxide dismutase and decreased glutathione peroxidase activities in red blood cells. Eur. J. Biochem. I39t13-18. Meera Khan P (1986) Red cell GGPD, GPGD, GPX1, GLOI, LDH, PGP, and DIA2 polymorphisms. In LL CavalliSforza (ed.): African Pygmies. Orlando, FL: Academic Press. Meera Khan P, Verma C, Wijnen LMM, and Jairaj S (1984a) Red cell glutathione peroxidase (GPXI) variation in AfroJamaican, Asiatic Indian, and Dutch populations. Is the GPX1*2 allele of “Thomas” variant an African marker? Hum. Genet. 66:352-355. Meera Khan P, Verma C, Wijnen LMM, and Zurcher C (1984b) Formal genetics and ethnic distribution of the electrophoretic variants of red cell glutathione peroxidase (GPX1). Is GPX1*2 an African allele? Antropol. Contemp. 7:120-121 (Abstract). Meera Khan P, Verma C, Wijnen LMM, Wijnen JT, Prins HK, and Nijenhuis LE (1986) Electrotypes and formal genetics of red cell glutathione peroxidase (GPXI) in the Djuka of Surinam. Am. J. Hum. Genet. 38:712-723. Mills GC (1957) Hemoglobin catabolism: I. Glutathione peroxidase, an erythrocyte enzyme which protects hemoglobin from oxidative breakdown. 3 . Biol. Chem. 229: 189-1097. Mourant AE, Kopec AC, and Domaniewska-Sobczak K (1976) The Distribution of Human Blood Groups and Other Polymorphisms. Ed. 2. Oxford: Oxford University Press. Necheles TF, SteinbergMH, and Cameron D (1970)Erythrocyte glutathione-peroxidase deficiency. Br. J. Haematol. 19:605-612. Nurse GT, and Jenkins T (1976) Health and the hunter gatherer: Biomedical studies of the hunting and gathering populations of Southern Africa. In S Beckman and L Hauge (eds.):Monographs in Human Genetics. Basel: S. Karger, pp. 65-66. Ola Ojikutu R, Nurse GT, and Jenkins T (1977) Red cell enzyme polymorphisms in the Yoruba. Hum. Hered. 27:444-453. Pleban PA, Munyani A, and Beachum J (1982) Determination of selenium concentration and glutathione perox- idase activity in plasma and erythrocytes. Clin. Chem. 28:311-316. Ponzetto Zimmerman C and Natta C (1981) Glutathione peroxidase activity in whole blood of patients with sickle cell anaemia. Scand. J . Haematol. 26:177-181. Rotruck JT, Pope AL, Ganther HE, Swanson AB, Hafeman DG, and Hoekstra WG (1973) Selenium: Biochemical role a s a component of glutathione peroxidase. Science 179:588-590. Rudolph N and Wong SL (1978) Selenium and glutathione peroxidase activity in maternal and cord plasma and red cells. Pediatr. Res. 12:789-792. Sies H, Gerstenecker C, Menzel H, and Flohe L (1972) Oxidation of the NADPH system and releaseof GSSG from hemoglobin free perfused rat liver during peroxidatic oxidation of glutathione by peroxides. FEBS Lett. 27:171-175. Spedini G, Capucci E, Fuciarelli M, and Rickards 0 (1980a) The AcP polymorphism frequencies in the Mbugu and Sango of Central Africa (correlations between the Per allele and some climatic factors in Africa). Ann. Hum. Biol. 7:125-128. Spedini G, Fuciarelli M, and Rickards 0 (1980b) Blood polymorphism frequencies in the Tofinu, the “Water Men” of Southern Benin. Anthropol. Anz. 38:121-130. Spedini G, Capucci E, Crosti N, Danubio ME, and Romagnoli S (1982)Erythrocyte glyoxalaseI (GLO) and superoxide dismutase (SOD A) polymorphisms in the Mbugu and some other populations of the Central African Republic. Hum. Hered. 32:253-258. Spedini G, Walter H, Capucci E, Fuciarelli M, Rickards 0, Aebischer ML, and Crosti N (1983) An anthropobiological study in Basse Kotto (Central Africa). I. Erythrocyte and serogenetic markers: An analysis of the genetic differentiation. Am. J. Phys. Anthropol. 60:39-47. Spedini G, Menchicchi F, and Destro-Bisol G (1985-1986) Recherche biologiques, nutritionelles et sanitaires sur des populations de la RQpubliquePopulaire du Congo et probkmes lies au developpement rural: V. Les polymorphismes gAn6tiques Arithrocitayres et s6riques chez le Bbtis, Les Battkts et les Babingas du Congo: Analyse de l’hBtBrogen6it6 gBn6tique intra-et inter-groupes. Ri. Antropol. LXVIII:76-92. Spedini G, Menchicchi F, Destro-Bisol G, and Schanfield M (1986) Migration and genetic polymorphism in some Congo peoples. In DF Roberts and GF De Stefan0 (eds.): Genetic Variation and its Maintenance. Cambridge: Cambridge University Press, pp. 191-197. Spedini G, Bodioli L, Destro-BisolG, BattaggiaC, Bailly C (1988) Indagine antropobiologica nel Cameroun occidentale: dati preliminari sulla variabilita genetica dei Bemileke. Antropol. Comtemp. in press. Srivastava SK and Beutler E (1969) The transport of oxidized glutathione from human erythrocytes. J. Biol. Chem. 244~9-16. Swarup-Mitra S (1977) Activity of glutathione peroxidase and glutathione reductase in G-6-PD deficient subjects. Indian J. Med. Res. 66:253-259. Wendel A (1980) Glutathione peroxidase. In WB Jakoby (ed.): Enzymatic Basis of Detoxication. New York: Academic Press, pp. 334-336.