Hum. Hered. 25: 370-377 (1975)
Immunoglobulin Allotypes of European Populations I. Gm and Km(Fnv) Allotypic Markers in Hungarians1
M.S. Schanfield , J .G ergely and H. H. F udenberg Section of Immunology, Department of Medicine, University of California School of Medicine, San Francisco, Calif., and the Central Institute of Hematology and Blood Transfusion, Budapest
Key Words. Gm • Km(lnv) • Hungarian population studies Abstract. The distribution of G 1m(f, z, a, and x), G3m(b0, b 1, b3, c3, c5, g, s, and t) and Km(l) (formerly lnv[l]) allotypic markers have been examined in 184 Hungarians. The results indicate that the frequency of the immunoglobulin haplotypes Gm13'-s, Gm“ *¡8, Gm('-b and Km1 is similar to the frequencies observed in surrounding populations. In ad dition, Hungarians were found to be polymorphic for the Oriental haplotype GmZ3'-bsx, and had low- frequencies of other uncommon haplotypes. Our data indicate that about 5% of the Hungarian genome is of Oriental origin.
Uncom m on Gm haplotypes have been observed in individuals o f dif ferent Caucasian ethnic origins in the United States. The occurrence o f some o f these haplotypes can only be explained by admixture with non-Caucasian populations o f either Oriental or African extraction. This is particularly true o f G/MzaJb°>3«5>s>t>v (hereafter referred to as Gmza:bst), an Oriental haplotype whose frequency is highest in Northern Orientals. In the United States, this haplotypc has been observed in individuals o f Italian, German, and Czechoslovakian extraction [Schanfield , unpubl.], and in Ashkanazic Jews [Schanfield , unpubl.; Steinberg, 1973]. Similarly, the haplotypc (jw za;bO,I,c3,5,u (hereafter referred to as Gm7-a;bc3,5) ¡s exclusively o f African
1 Supported in part by USPHS Grants (AI-09145 and HD-05894), and NIH Post doctoral Research Fellowship (HD-52597-1).
Downloaded by: Univ. of California Santa Cruz 128.114.34.22 - 2/14/2018 11:28:37 PM
origin. Unfortunately, other haplotypes are o f more ambiguous origin. The haplotype G/nza:btM>3.4.5,u,v (hereafter referred to as Gmza'-b) can be either o f African origin, or due to a recombination between two common Cauca-
Schanfield/G ergely/F udenberg
371
sian haplotypes Gm7Av (hereafter referred to as Gm7-a'b haplotype. If the reagents for the linked IgA2 determinants A2m(l and 2) are available, the haplotype can be identified more accurately, since the African haplotype is much more frequently A2m(2)-positivc than A2m(l) positive. A2m(2) is a very infrequent allotypic determinant in Caucasians with only Caucasian phenotypes [S c h a n field and F u d e n be rg , 1975], It is therefore of interest to map the frequency of non-Caucasian haplotypes in European populations not previously tested for them. Therefore, extensive immunoglobulin allotyping was carried out in a sample of unsclected Hungarians. Previous studies of Hungarians were limited to only a few specificities, namely Glm(a and x), G3m(bl) and Km(l) [W alter and N emeskeri, 1967], or Glm(f, a, and x), G3m(bl and b5) and Km(l and 2) [R o pa r t z et al., 1968]. This paper reports for the first time the distribution of the allotypic markers Glm(z), G3m(b0, b3, c3, c5, g, s, and t) and A2m(l and 2) in a sample of Hungarians. The use o f these allotypic determinants permits the detection of the majority o f the defined Caucasian and non-Caucasian haplotypes known [S ch a n field and F u d e n b e r g , 1975].
Throughout this report the alphanumeric nomenclature will be used [WHO, 1975]. The numeric notation preferred by some investigators can be found in table I. Scrum specimens from Hungarian blood donors collected by the Central Institute of Hematology and Blood Transfusion in Budapest were shipped to San Francisco for allotyping. A total of 184 sera were tested for Glm(a, x, z, and f), G3m(b0, bl, b3, c3, c5, g, s, and t) and Km(l). In addition, selected specimens were tested for the IgA2 allotypic markers A2m(l and 2). The reagents used arc presented in table 1. All sera were tested at a dilution of 1/20 using a standard test system on microtiter plates [Schanfield, 1971 ; Schanfield and F udenberg, 1975]. Any sera showing direct agglutinating activity were heatin-activated at 65 C and retested. Any sera which could not be heat-inactivated were excluded. Haplotype frequencies for the Gm system were estimated by direct gene counting including all of the observed phenotypes. The only assumption regarding genotypes in volved individuals of the phenotype Gm(z, a, x; g). For purposes of gene counting they were assumed to be G»izaxts/Gmzats. The net effect of this is to produce a slightly lower estimated frequency of Gin1**-;b and a slightly higher estimated frequency of G/nza:s than that produced by maximum likelihood computer estimation. However, the differences are usually trivial. The frequency estimates for all other haplotypes are the maximum likelihood estimates. The frequency of Km1 was estimated using the square root of the K m (l—) frequency. All calculations were done on a Wang 600 programmable calculator.
Downloaded by: Univ. of California Santa Cruz 128.114.34.22 - 2/14/2018 11:28:37 PM
Materials and Methods
372
S chanfield/G ergely/F udenberg
Tabic I. Reagents used for immunoglobulin allotyping Specificity1
G lm
numeric
a
i 2 17 3 11 5 13 6 24 21 21 21 15 16 1 2 1 1
X
G3m
A2m Km (Inv)
Agglulinalor
alphanumeric1
z f bO bl b3 c3 c5 g g g s i i 2 1 1
Hel Gey Pon Sta Har Blc F841 And Hod B7553 Gha Lch Gai Ros Far* Tay4 Cia Sim
Coat
Spr Har Har Dan Hun Hun Hun Bor2 Bor2 Sul Sul Sul 3068= 3068Hcr For Abr Abr
1 Nomenclature recommended by the WHO Workshop on Human Immunoglobulin Allotypes, 16-19 July, 1974, Rouen (France). 2,3 Reagents generously provided by E . van Loghem and S. Litwin , respectively. 4 Only selected specimens were tested for these markers.
The distribution of Gm and Km phenotypes in the sera studied, and the estimated haplotype frequencies are presented in table II. To simplify further discussion, only positive results will be reported. Further, any samplepositivc for one or more G3m(b) reagents will be termed G3m(b)-positive. 10 sera had phenotypes which could not be explained by the 3 haplo types considered to be polymorphic in Caucasian populations, e.g. Gm23¡8, (7 ,?jzax;g, anci ç Mjf;b These included 4 Gm(f, z, a; bO, 1, 3, s, t), 2 Gm(f, a; bO, 1,3) and 1 each Gm(z, a ; bO, 3, g, s, t), Gm(f, z, a ; bO, 1, 3), Gm(f, z, a ; g) and Gm(f, z, a, x; g). Without family data it is not possible to define with certainty all o f the haplotypes involved; however, several of the phenotypes can be explained by the presence o f a common Caucasian haplotype and
Downloaded by: Univ. of California Santa Cruz 128.114.34.22 - 2/14/2018 11:28:37 PM
Results
373
Gm and Km in Hungary
Table II. Distribution of immunoglobulin phenotypes and haplotypes in Hungarians Phenotypes G lm z,a; z, a, x: f, z, a ; f, z, a, x; f; f, z, a ; z ,a ; f,z ,a ; f,a ; f, z ,a ; f, z, a, x; Total Km(l + )
Haplotypes G3m g g b, g b, g b b, s, t b,g, s, t b b g g 26/194
2 5
G/nz»:s G » iza*:e
43
12 112 4 1 1 2 1 1 184
Gnpa;bst GWa;b or Gnfo G /;p a ;b or Gm G nf '-z or Gm**-
Km'
0.147 0.049 0.777 0.014 0.005 0.003 0.005
0.073
either a haplotype known to exist in another racial group or a rare Caucasian haplotype. The simultaneous presence of Gm(z, a; b, s, t) in 5 individuals is almost certainly due to the Oriental haplotype C/«zaibst. This haplotype is commonly A2m(2)-positive; 4 of the 5 individuals with this haplotype were A2m(2)-positive. The 2 individuals of the phenotypes Gm(f, a ; b) are probably heterozygous for the Oriental G/wfa:b and the Caucasian Gwl;b haplotypes. but this haplotype could result from unequal crossover; since this is extremely unlikely, the Oriental admixture appears to be a far more likely explanation. The Cmfaib haplotype is also normally A2m(2)-positive in Oriental popula tions; however, in the 2 sera studied, neither were A2m(2)-positive. Of the 3 remaining uncommon phenotypes, two can be explained by either the presence of the rare Gw/iS haplotype or the rarer Gnfi'~ haplotype, while the last phenotype can be explained by either the presence of a
Immunoglobulin Allotypes of European Populations I. Gm and Km(Fnv) Allotypic Markers in Hungarians1
M.S. Schanfield , J .G ergely and H. H. F udenberg Section of Immunology, Department of Medicine, University of California School of Medicine, San Francisco, Calif., and the Central Institute of Hematology and Blood Transfusion, Budapest
Key Words. Gm • Km(lnv) • Hungarian population studies Abstract. The distribution of G 1m(f, z, a, and x), G3m(b0, b 1, b3, c3, c5, g, s, and t) and Km(l) (formerly lnv[l]) allotypic markers have been examined in 184 Hungarians. The results indicate that the frequency of the immunoglobulin haplotypes Gm13'-s, Gm“ *¡8, Gm('-b and Km1 is similar to the frequencies observed in surrounding populations. In ad dition, Hungarians were found to be polymorphic for the Oriental haplotype GmZ3'-bsx, and had low- frequencies of other uncommon haplotypes. Our data indicate that about 5% of the Hungarian genome is of Oriental origin.
Uncom m on Gm haplotypes have been observed in individuals o f dif ferent Caucasian ethnic origins in the United States. The occurrence o f some o f these haplotypes can only be explained by admixture with non-Caucasian populations o f either Oriental or African extraction. This is particularly true o f G/MzaJb°>3«5>s>t>v (hereafter referred to as Gmza:bst), an Oriental haplotype whose frequency is highest in Northern Orientals. In the United States, this haplotypc has been observed in individuals o f Italian, German, and Czechoslovakian extraction [Schanfield , unpubl.], and in Ashkanazic Jews [Schanfield , unpubl.; Steinberg, 1973]. Similarly, the haplotypc (jw za;bO,I,c3,5,u (hereafter referred to as Gm7-a;bc3,5) ¡s exclusively o f African
1 Supported in part by USPHS Grants (AI-09145 and HD-05894), and NIH Post doctoral Research Fellowship (HD-52597-1).
Downloaded by: Univ. of California Santa Cruz 128.114.34.22 - 2/14/2018 11:28:37 PM
origin. Unfortunately, other haplotypes are o f more ambiguous origin. The haplotype G/nza:btM>3.4.5,u,v (hereafter referred to as Gmza'-b) can be either o f African origin, or due to a recombination between two common Cauca-
Schanfield/G ergely/F udenberg
371
sian haplotypes Gm7Av (hereafter referred to as Gm7-a'b haplotype. If the reagents for the linked IgA2 determinants A2m(l and 2) are available, the haplotype can be identified more accurately, since the African haplotype is much more frequently A2m(2)-positivc than A2m(l) positive. A2m(2) is a very infrequent allotypic determinant in Caucasians with only Caucasian phenotypes [S c h a n field and F u d e n be rg , 1975], It is therefore of interest to map the frequency of non-Caucasian haplotypes in European populations not previously tested for them. Therefore, extensive immunoglobulin allotyping was carried out in a sample of unsclected Hungarians. Previous studies of Hungarians were limited to only a few specificities, namely Glm(a and x), G3m(bl) and Km(l) [W alter and N emeskeri, 1967], or Glm(f, a, and x), G3m(bl and b5) and Km(l and 2) [R o pa r t z et al., 1968]. This paper reports for the first time the distribution of the allotypic markers Glm(z), G3m(b0, b3, c3, c5, g, s, and t) and A2m(l and 2) in a sample of Hungarians. The use o f these allotypic determinants permits the detection of the majority o f the defined Caucasian and non-Caucasian haplotypes known [S ch a n field and F u d e n b e r g , 1975].
Throughout this report the alphanumeric nomenclature will be used [WHO, 1975]. The numeric notation preferred by some investigators can be found in table I. Scrum specimens from Hungarian blood donors collected by the Central Institute of Hematology and Blood Transfusion in Budapest were shipped to San Francisco for allotyping. A total of 184 sera were tested for Glm(a, x, z, and f), G3m(b0, bl, b3, c3, c5, g, s, and t) and Km(l). In addition, selected specimens were tested for the IgA2 allotypic markers A2m(l and 2). The reagents used arc presented in table 1. All sera were tested at a dilution of 1/20 using a standard test system on microtiter plates [Schanfield, 1971 ; Schanfield and F udenberg, 1975]. Any sera showing direct agglutinating activity were heatin-activated at 65 C and retested. Any sera which could not be heat-inactivated were excluded. Haplotype frequencies for the Gm system were estimated by direct gene counting including all of the observed phenotypes. The only assumption regarding genotypes in volved individuals of the phenotype Gm(z, a, x; g). For purposes of gene counting they were assumed to be G»izaxts/Gmzats. The net effect of this is to produce a slightly lower estimated frequency of Gin1**-;b and a slightly higher estimated frequency of G/nza:s than that produced by maximum likelihood computer estimation. However, the differences are usually trivial. The frequency estimates for all other haplotypes are the maximum likelihood estimates. The frequency of Km1 was estimated using the square root of the K m (l—) frequency. All calculations were done on a Wang 600 programmable calculator.
Downloaded by: Univ. of California Santa Cruz 128.114.34.22 - 2/14/2018 11:28:37 PM
Materials and Methods
372
S chanfield/G ergely/F udenberg
Tabic I. Reagents used for immunoglobulin allotyping Specificity1
G lm
numeric
a
i 2 17 3 11 5 13 6 24 21 21 21 15 16 1 2 1 1
X
G3m
A2m Km (Inv)
Agglulinalor
alphanumeric1
z f bO bl b3 c3 c5 g g g s i i 2 1 1
Hel Gey Pon Sta Har Blc F841 And Hod B7553 Gha Lch Gai Ros Far* Tay4 Cia Sim
Coat
Spr Har Har Dan Hun Hun Hun Bor2 Bor2 Sul Sul Sul 3068= 3068Hcr For Abr Abr
1 Nomenclature recommended by the WHO Workshop on Human Immunoglobulin Allotypes, 16-19 July, 1974, Rouen (France). 2,3 Reagents generously provided by E . van Loghem and S. Litwin , respectively. 4 Only selected specimens were tested for these markers.
The distribution of Gm and Km phenotypes in the sera studied, and the estimated haplotype frequencies are presented in table II. To simplify further discussion, only positive results will be reported. Further, any samplepositivc for one or more G3m(b) reagents will be termed G3m(b)-positive. 10 sera had phenotypes which could not be explained by the 3 haplo types considered to be polymorphic in Caucasian populations, e.g. Gm23¡8, (7 ,?jzax;g, anci ç Mjf;b These included 4 Gm(f, z, a; bO, 1, 3, s, t), 2 Gm(f, a; bO, 1,3) and 1 each Gm(z, a ; bO, 3, g, s, t), Gm(f, z, a ; bO, 1, 3), Gm(f, z, a ; g) and Gm(f, z, a, x; g). Without family data it is not possible to define with certainty all o f the haplotypes involved; however, several of the phenotypes can be explained by the presence o f a common Caucasian haplotype and
Downloaded by: Univ. of California Santa Cruz 128.114.34.22 - 2/14/2018 11:28:37 PM
Results
373
Gm and Km in Hungary
Table II. Distribution of immunoglobulin phenotypes and haplotypes in Hungarians Phenotypes G lm z,a; z, a, x: f, z, a ; f, z, a, x; f; f, z, a ; z ,a ; f,z ,a ; f,a ; f, z ,a ; f, z, a, x; Total Km(l + )
Haplotypes G3m g g b, g b, g b b, s, t b,g, s, t b b g g 26/194
2 5
G/nz»:s G » iza*:e
43
12 112 4 1 1 2 1 1 184
Gnpa;bst GWa;b or Gnfo G /;p a ;b or Gm G nf '-z or Gm**-
Km'
0.147 0.049 0.777 0.014 0.005 0.003 0.005
0.073
either a haplotype known to exist in another racial group or a rare Caucasian haplotype. The simultaneous presence of Gm(z, a; b, s, t) in 5 individuals is almost certainly due to the Oriental haplotype C/«zaibst. This haplotype is commonly A2m(2)-positive; 4 of the 5 individuals with this haplotype were A2m(2)-positive. The 2 individuals of the phenotypes Gm(f, a ; b) are probably heterozygous for the Oriental G/wfa:b and the Caucasian Gwl;b haplotypes. but this haplotype could result from unequal crossover; since this is extremely unlikely, the Oriental admixture appears to be a far more likely explanation. The Cmfaib haplotype is also normally A2m(2)-positive in Oriental popula tions; however, in the 2 sera studied, neither were A2m(2)-positive. Of the 3 remaining uncommon phenotypes, two can be explained by either the presence of the rare Gw/iS haplotype or the rarer Gnfi'~ haplotype, while the last phenotype can be explained by either the presence of a
