HLA-DR,DQ sequence polymorphisms in Polynesians, Micronesians, and Javanese

HLA-DR,DQ Sequence Polymorphisms in Polynesians, Micronesians, and Javanese Xiaojiang Gao, Paul Zimmet, and Susan W. Serjeantson

ABSTRACT: The origins of the Polynesians remain an enigma. Linguistic reconstructions of proto-Austronesian languages suggest a shared origin for Polynesians, Micronesians, and Javanese with dispersal from northern Borneo and Sulawesi. Analysis of 810 chromosomes for nucleotide sequence polymorphism at HLA-DRB1, DRB3, DRB5, DQA1, and DQB 1 loci in Polynesian (Rarotonga, Western Samoa, and Niue), Micronesian (Nauru and Kiribati), and Javanese populations showed virtually no overlap of HLA class II haplotypes between contemporary Polynesians and Javanese. Further, there were marked differences in population distributions of some

HLA-DRB1 alleles that could not be distinguished in earlier serologic or restriction fragment length polymorphism (RFLP) studies, e.g., for DRI2, DRBI*1201 had a frequency of 15%-30% in Polynesians (1% in Micronesians and Javanese), whereas DRBI*I202 had a frequency of 28%-38% in Micronesians and 51% in Javanese (1% in Polynesians). A novel DR6-related allele, DRBI*1408, was found in all three Polynesian study populations. The Polynesian HLA class II genetic repertoire is not readily derived from the island Southeast Asian gene pool. Human Immunology 34, 153-161 (1992)

ABBREVIATIONS

MHC MLC PCR

major histocompatibility complex mixed lymphocyte culture polymerase chain reaction

RFLP SSO

restriction fragment length polymorphism sequence-specific oligonucleotides

INTRODUCTION Early serologic studies of H L A - D R antigen distributions in Oceania, in Micronesians [1], Polynesians [2, 3], and Balinese [4], suggested comparatively limited genetic diversity at the H L A - D R locus, but mixed lymphocyte culture (MLC) typing has not been done in these populations. Restriction length fragment polymorphism (RFLP) typing confirmed that the supertypic antigens were essentially confined to DR2, DR4, DR5, DR6, and D R 8 in Micronesians, with one additional antigen, DR9, in Polynesians [5]. However, RFLP typing found diverse and, in some instances, novel D R , D Q linkage disequilibrium relationships in Pacific Islanders, so that the allelic subtypes associated with the serologic

From the Human GeneticsGroup (X.G.; S.W.S.),John Curtin School of Medical Research, The Australian National University, Canberra, ACT," and the International Diabetes Institute (P.Z.), CaulfieldGeneral Medical Centre, Caulfield, VIC, Australia. Address reprint requests to X. Gao, Human Genetics Group, John Curtin Schoolof Medical Research, The Australian National University, GPO Box 334, CanberraACT 2601, Australia. ReceivedDecember27, 1991; acceptedApril 27, 1992.

HumanImmunology34, 153-161 (1992) © AmericanSocietyfor Histocompatibilityand Immunogenetics,1992

antigens could not be discriminated with any confidence. For instance, the allelic subtypes of HLA-DR15, H L A - D w 2 , and D w l 2 , can be readily assigned in Caucasoids on the basis of D R , D Q linkage arrangements [6], but DRi5-positive Micronesians had DQA1 RFLPs characteristic of D w 2 ( D R B I * 1 5 0 1 ) and DQB1 RFLPs characteristic of D w I 2 (DRBI*1502) [5]. Similarly, D R 6 subtypes, Dw9, D w l 8 , and D w l 9 , can be discriminated in Caucasoids by particular D R , D Q RFLP patterns [7~], but the common D R 6 haplotype in Polynesians was a combination of D R 1 8 (or D w l 9 ) RFLPs at DRB and D w 9 RFLPs at DQA1 and DQB1 [8]. For HLA-DR5-related alleles, there was an unusual, highfrequency DRB pattern in Micronesians, designated " D R 5 * N A U R U " [9], which occurred in Polynesians also, but with different DQA1 linkage disequilibrium relationships [8]. Finally, D R 8 RFLPs were associated with D Q I RFLPs rather than with DQ3 or DQ4 as in Caucasoids [5]. The unanswered question arising from these RFLP studies, and of importance in terms of population affini153 0198-8859/92/$5.00

X. Gao et al.

154

ties as well as molecular evolution of the major histocompatibility complex (MHC), is whether the novel DR,DQ arrangements in Pacific Island populations represent ancestral recombination between DQA1, DQB 1, and DRB1 loci, or whether unusual DR,DQ linkage arrangements are markers for novel DR or DQ alleles. A related question, but one of longer standing, concerns the origins of the Polynesians who, according to linguistic reconstructions, have evolved out of island Southeast Asia [10]. For insular Southeast Asia, there is a dearth of genetic information and this is particularly true for HLA class II genes. With the development of polymerase chain reaction (PCR)-based HLA class II typing protocols with sequence-specific oligonucleotides (SSOs) [11-13], it is possible to address these questions and to define more precisely the sequence polymorphisms of the HLA-DR and DQ genes in Oceanic populations. In this study, we have examined nucleotide sequence polymorphisms at the DRB1, DRB3, DRB5, DQA1, and DQB1 loci in one Javanese, three Polynesian, and two Micronesian populations, by SSO typing of PCR products. Linkage disequilibrium relationships between the DRB1 locus and the four other class II loci have been determined for the three ethnic groups. MATERIALS A N D M E T H O D S

Study populations. Polynesians from Rarotonga (n = 78) or Niue (n = 61) in the Cook Islands were randomly sampled from among participants in a broadly based medical survey that sought complete ascertainment of adults aged 20 years and older, as described elsewhere [14]. These two series have not been examined previously for HLA distributions. Further DNA specimens from Niue (n = 9) and from the central Polynesian populations of Western Samoa and Tokelau (n -- 51) were from cord-blood samples from newborns in Auckland, New Zealand, as reported previously in HLA class II RFLP analysis of a mixed Polynesian series [8]. These populations represent three of the four Polynesian linguistic subgroups [15]: Tongic (Niue), Samoic (Western Samoa, and Tokelau) and Central-Eastern (Rarotonga). The fourth unrepresented subgroup is Easter Island. Micronesians from the Republic of Nauru (n =67) and Kiribati (n = 62) were randomly sampled from among adult participants in comprehensive surveys of diabetes, as described for Nauru [16] and Kiribati [17], previously known as the Gilbert Islands. These populations speak languages classified as "nuclear Micronesian," as distinct from the Western Malayo-Polynesian languages spoken in Western Micronesia (Palau and the Marianas). Nuclear Micronesian and Polynesian languages belong to the same subgroup of Austronesian,

known as "Oceania" [18]. Although a Nauruan series has been examined for HLA-DR serology [1] and for DR,DQ RFLP distributions [5], the earlier series do not correspond with samples studied here. Austronesian speakers from Java were sampled in Yogyakarta.

DNA preparation. Genomic DNA was prepared from hemolysates or from buffy coats using either the routine method of phenol-chloroform extraction [ 19] or automated nucleic acid extraction, using an Applied biosystems Inc. (ABI) model 340A.

Oligonucleotide primers and PCR amph'fication. The second exons ofDRB1, DRB3, DRB5, DQA1, and DQB1 genes were amplified by PCR for all samples. All of the oligonucleotide primers used in the present study have been described previously. DRB generic amplification was performed with primers GAMPDRB and GLPDRB [11], which amplify all the DRB genes. Amplifications of DR1, DR2, DR4, and DR52-related DRB1 genes were achieved using GAMPDRB at the 3' end, together with group-specific primers SRIP (an equivalent of PSP-43) [13], SRP-5 [20], SRP-4 [21], and SRP-6 [21], respectively, at the 5' end. DRB3 and DRB5 genes were amplified by locus-specific primers SRP-3 and SR5P [20, 21] at the 5' end and GAMPDRB at the 3' end. DQA1 and DQB1 genes were amplified using primers described by Scharf et al. [22] and by Morel et al. [23], respectively. The conditions of the PCR amplifications were as given in the original references.

Oligonucleotide probes and

hybridization. Oligonucleotide probes, all 19 mers, were locally constructed on an ABI model 380B synthesizer and then labeled with 32p before hybridizations. The sequences and specificities of 12 probes used for DRB generic typing are shown in Table 1. The details of the other probes, together with the hybridization conditons, have been described elsewhere including eight for DRI [13], 12 for DR2-related DRB1 and DRB5 [24], 11 for DR4 [12], 31 for DR52-related DRB1 and DRB3 [25], 11 for DQA1, and 14 for DQB1 [20] typing. Statistical analyses. Allele frequencies were determined by gene counting. Two-locus haplotype frequencies, linkage disequilibrium delta values (D) and chi-squared estimates of the statistical significance of linkage disequilibrium, were calculated as given by Baur and Danilovs [26]. RESULTS

HLA-DRBI allele distributions. Oligonucleotide hybridizations of PCR products detected 21 of the 47

HLA Class II Polymorphisms in Pacific Islanders

155

TABLE 1 Oligonucleotide probes for DRB generic typing Name

Codon

Sequence (5'-3')

Specificity

SR-1 SR-2 SR-32 SR-4 SR-11 SR-12 SR-6

27-33 66-72 73-79 28-34 54-60 24-30 32-38

G GAA AC ATC C CGG C AGA G CGG G CGG T AAC

AGATGC CTG GAG GTG GAC TAC "ITC CCT GAT TTA CTG CAG GAG

ATC CAG AAC TAT GAG GAG GAG

TAT GCG TAC CAC GAG AGA AAC

AAC CG TGC CAA TAC CAC GTG

SR-14 SR-57 SR-14.8 SR-7 SR-8 SR-9 SR-10 SR-45 SRB5

55-61 55-61 54-60 25-31 69-75 25-31 28-34 54-60 24-30

G CCT G CCT G CGG CAGTTC AA GAC CGGTAT GAA AGA GTA CTC GTG CGG

GCT GCG GAC GCT CCT GAT CTG GAA AGGCGG CTG CAC CGC GTC GGC GCT TTC CTG

GAG CAC GAGTAC GAG GAG AAG ACT GCC CTG AGA GGC CAT AAC AGG CCG CAC AGA

TGG TGG CAC CT GT A C Ca G

DRB1*0101,0102,0103 DRB1*1501,1502,1503, DRB5*02 DRBl*0301,0302 DRBl*0401-0411 DRB 1" 1101-1104 DRB1*1201,1202 DRB 1"0301,0302,1301,1302,1305, 1402,1403 DRB1*1401,1404 DRB1*1405 DRBl*1408 DRB1*0701,0702 DRB 1*0801-0804 DRB 1"0901 DRBI*1001 DRB 1"0405,1303,1304,0801,0803 DRB5*0101,0102,0201,0202

" Complementary to coding sequence.

officially named [27] DRB1 alleles. In addition, the newly defined DRBI*1408 allele, described previously in Australian Aborigines [28], was detected in the present study. Table 2 gives the gene frequencies of DRB alleles in the six populations. Two DR2-related alleles, DRBl*1501 and 1502, were detected in all populations, but DRBl*1502 showed markedly higher gene frequencies in Javanese and Micronesian populations (11.3%-23.9%) than in the three Polynesian groups (1.4%-3.9%). DRBI*0403 (Dw13) dominated the DR4 group in Pacific Islanders with allele frequencies ranging from 11.2% to 22.9%, but there was a near absence of DR4-related alleles in Javanese. The DR11 group was mainly represented by DRBI*IlO1 in all of the populations, with the highest gene frequency found in Rarotongan Polynesians (20.5%). Two DRl2-related alleles showed markedly different distributions among the study populations; DRBI*1201 was detected with relatively high gene frequencies in all three Polynesian populations (14.7%-30.0%), but was virtually absent from the Javanese and Micronesian samples, whereas DRBI*1202 was detected with gene frequencies of 28.4% in Nauru, 37.9% in Kiribati, and 50.7% in Javanese; only three Polynesian donors were found to have this allele. DRBI*I401 and the closely related allele DRBI*I408 were the major DR6-related alleles detected in this study, although DRBl*1401 was found in only one Kiribati donor, and DRBI*I408 was absent from the Nauruan sample; HLA-DR6 alleles were virtually absent from Javanese. DRBI*0803 was the most common DRw8 subtype in these populations, but DRBI*0802 was detected in some Polynesians. Several

other alleles also showed different distributions in the study populations; DRBI*0901 was well represented in Polynesians, but absent from Micronesians, whereas Javanese had an appreciable frequency of DRBI*07. Some alleles suggest foreign admixture: the Caucasoid DR1 (DRBI*OIO1) and DR3 (DRBI*0301) alleles in Polynesians and the African DR3 (DRB 1"0302) allele in Nauruans.

HLA-DRB3 and HLA-DRB5 allele distributions. Three of the four known DRB3 alleles were detected in all populations and showed strong linkage with DR52related DRB1 alleles. The associations between DRB3 locus alleles and DRB1 alleles were DRB3*OIO1 with DRBI*0301, 0302, 1201, and 1408; DRB3*0202 with DRBI*llOI, 1102, 1301, 1401, 1404, and 1405; and DRB3*0301 with DRBI*1202 and 1302. DRB3*020I was not observed. DRB5*OIOI was associated with both DRBI*150I and 1502. In Javanese, DRBI*I502 was more usually associated with DRB5*OI02.

HLA-DQA1 and HLA-DQBI allele distributions. The frequencies of DQA1 and DQB1 alleles are given in Table 3. The distributions of DQ alleles in these populations corresponded with relevant DRB1 alleles because of strong DR,DQ linkage disequilibrium, as shown in Table 4. Among the DR,DQ haplotypes showing population-specific distributions, the Polynesian allele DRBI*1201 was linked with DRB3*OIOI, DQAI*0501, and DQBI*0301, whereas the Micronesian allele DRBI*I202 was associated with DRB3*0301, DQAI*0601, and DQBI*0301. In Poly-

156

x. Gao et al.

TABLE 2

HLA-DRB allele frequencies of Polynesians, Micronesians, and Javanese

RARO ~ n = 156 b

NIUE n = 140

WSMA n = 102

NARU n = 134

KIRI n = 124

JAVA n = 154

Allele

n

%

n

%

n

%

n

%

n

%

n

%

DRBI* 0101 0103 1501 1502 0301 0302 0401 0402 0403 0404 0405 0406 0410 1101 1102 1201 1202 1301 1302 1401 1404 1405 1408 07 0801 0802 0803 0901 1001

1 1 2 3 1 0 4 1 28 1 8 0 1 32 1 23 1 0 3 8 l 0 8 l 0 4 9 14 0

0.6 0,6 1.3 1.9 0.6 0.0 2.6 0.6 18.0 0.6 5.1 0.0 0.6 20.5 0.6 14.7 0.6 0.0 1,9 5,1 0.6 0,0 5.1 0.6 0.0 2.6 5.8 9.0 0.0

0 0 3 2 0 0 0 0 32 0 0 0 0 I1 0 42 2 0 0 7 0 0 6 2 1 6 14 12 0

0.0 0.0 2.1 1.4 0.0 0.0 0.0 0.0 22.9 0.0 0,0 0.0 0.0 7.9 0.0 30.0 1.4 0.0 0.0 4.3 0.0 0.0 5.0 1.4 0.7 4.3 10.0 8.6 0.0

1 0 4 4 2 0 l 0 16 0 8 0 0 11 0 19 0 0 2 3 0 0 7 2 0 1 10 11 0

1,0 0,0 3.9 3,9 2,0 0.0 1.0 0.0 15.7 0.0 7.8 0.0 0.0 10.8 0.0 18.6 0.0 0.0 2.0 2.9 0.0 0.0 6.9 2.0 0.0 1.0 9.8 10.8 0.0

0 0 11 32 0 3 5 0 15 0 1 0 0 9 0 1 38 0 2 7 0 1 0 0 2 0 7 0 0

0.0 0.0 8.2 23.9 0.0 2.2 3.7 0.0 11.2 0.0 0.8 0.0 0.0 6.7 0.0 0.8 28.4 0.0 1.5 5.2 0.0 0.8 0.0 0,0 1.5 0.0 5.2 0.0 0.0

0 0 4 14 0 0 0 0 25 0 0 0 0 2 0 1 47 0 0 2 0 2 13 0 0 0 14 0 0

0.0 0.0 3.2 11.3 0.0 0.0 0.0 0.0 20.2 0.0 0.0 0.0 0.0 1.6 0.0 0.8 37.9 0.0 0.0 1.6 0.0 1.6 10.5 0.0 0.0 0.0 11.3 0.0 0.0

1 0 5 32 1 0 0 0 0 1 1 1 0 1 0 1 78 1 1 1 0 0 0 17 0 0 2 4 4

0.7 0.0 3-3 20.8 0.7 0.0 0.0 0.0 0.0 0.7 0.7 0.7 0.0 0.7 0.0 0.7 50.7 0.7 0.7 0.7 0.0 0.0 0.0 11.0 0.0 0.0 1.3 2.6 2.6

DRB3* 0101 0202 0301

32 42 4

20.5 26.9 2.6

49 17 2

35.0 12.1 1.4

27 15 2

26.5 14.7 2.0

4 17 40

3.0 12.7 29.9

14 6 47

11.3 4.8 37.9

1 7 76

0.7 4.6 49.4

DRB5* 0101 0102

4 1

2.6 0.6

5 0

3.6 0.0

8 0

7.8 0.0

42 1

31.3 0.8

16 2

12.1 1.6

19 20

12.3 13.0

° RARO, Rarotonga; NIUE, Niue; WSMA, West Samoa; NARU, Nauru; and KIRI, Kiribati b Number of chromosomes examined,

nesians, DRBI*0405 was associated with DQBI*0402 rather than DQBI*0401 as observed more frequently in other ethnic groups. DRBI*0302, found in three Nauruans from families with known African ancestry, was associated with DRB3*OIO1, DQAI*0401, and DQBI*0402, as in Africans. DRBI*1401, detected in Polynesians and Nauruan Micronesians, was more frequently related to DQBI*0502 than DQBI*0503, whereas the major DR4 subset in both groups, DRBI*0403, was invariably linked with DQAI*0301

and DQBI*0302. Hardy-Weinberg equilibrium at the HLA-DRB1, DQA1, and DQB1 loci was examined in the six populations separately. The allelic distributions at all the loci examined were in equilibrium (data not shown). The D R , D Q haplotypes for DRB1 alleles found commonly in Pacific Islanders can be compared with those reported for Caucasoids [29]. For DR2-related haplotypes, the linkage disequilibrium between DRBI*1501, DRB5*OIO1 and DRBI*I502,

HLA Class II Polymorphisms in Pacific Islanders

TABLE 3

157

H L A - D Q allele frequencies in Polynesians, Micronesians, and Javanese

RARO" n = 156 b

NIUE n = 140

WSMA n = 102

NARU n = 134

KIRI n = 124

JAVA n = 154

Allele

n

%

n

%

n

%

n

%

n

%

n

DQAI* 0101 0102 0103 0201 0301 0401 0501 0601

18 7 10 1 57 4 58 1

11.5 4.5 6.4 0.6 36.5 2.6 37.2 0.6

13 5 14 2 44 7 53 2

9.3 3.6 10.0 1.4 31.4 5.0 37.8 2.1

11 10 10 2 36 1 32 0

10.8 9.8 9.8 2.0 35-3 1.0 31.4 0.0

9 44 7 0 21 5 10 38

6.7 32.8 5.2 0.0 15.7 3.0 8.2 28.4

19 16 14 0 25 0 3 47

15.3 12.9 11.3 0.0 20,2 0.0 2.4 37.9

29 20 4 17 7 0 3 74

18.8 13.0 2.6 11.0 4.6 0.0 2.0 48.1

DQBI* 0501 0601 0602 0603 0604 0605 0502 0503 0201 0301 0302 0303 0401 0402

1 9 2 0 2 1 5 15 2 60 32 14 1 12

0.6 5.8 1.3 0.0 1.3 0.6 3.2 9.6 1.3 38.5 20.5 9.0 0.6 7.7

0 16 2 0 0 0 6 8 2 55 32 12 0 7

0.0 11.4 1.4 0.0 0.0 0.0 4.3 5.7 1.4 39.3 22.9 8.6 0.0 5.0

1 14 4 0 2 0 1 9 4 31 16 11 0 9

1.0 13.7 3.9 0.0 2.0 0.0 1.0 8.8 3.9 30.4 15.7 10.8 0.0 8.8

1 38 11 0 2 0 7 1 0 49 19 0 1 5

0.8 28.4 8.2 0.0 1.5 0.0 5.2 0.8 0.0 36.6 14.2 0.0 0.8 3.7

2 28 2 0 0 0 0 18 0 49 25 0 0 0

1.6 22.6 1.6 0.0 0,0 0.0 0.0 14.5 0.0 39-5 20.2 0.0 0.0 0.0

26 10 1 1 0 1 13 1 14 76 2 8 1 0

16.9 6.5 0.7 0,7 0.0 0.7 8.5 0.7 9.1 49.4 1.3 5.2 0.7 0.0

a RARO, Rarotonga; NIUE, Niue; WSMA, West Samoa; NARU, Nauru; and KIRI, Kiribati. h Number of chromosomes examined.

DRB5*O102 in Caucasoids is near absolute. In Micronesians, in whom DR15 was relatively common, DRBI*1502 was invariably associated with DRB5*OIOI rather than with DRB5*OI02. For DR4-related haplotypes in Pacific Islanders, the DRB 1"0403, DQB 1"0302 association was absolute, as in Caucasoids. For DR5related alleles, DRBI*I 1O1 in Pacific Islanders and Caucasoids shared D R , D Q linkage arrangements, but DRBI*1201 was associated with a different DRB3 allele in Polynesians. The DR6 allele DRBI*1401 was associated with different DQB 1 alleles in Pacific Islanders (DQBI*0502) and Caucasoids (DQBI*0503), although examples of the DRBI*14OI,DRBI*0503 haplotype were found in Polynesians; it was on this haplotye, presumably, that the point mutation giving rise to DRBI*I408 occurred. The previously reported DR6-related D R B 1 , D R B 3 , D Q RFLP haplotype in Polynesians [8] not found in Caucasoids, carried the novel DRB1 allele, DRBI*I408. The common DR8 haplotype in Pacific Islanders carried the DRBI*0803 allele, rather than DRBI*0801 more usually seen in Caucasoids.

DISCUSSION This study, using SSO hybridizations of PCR-amplified exon-2 segments of HLA class II genes, has characterized HLA-D region alleles that were difficult or impossible to detect in Pacific Islanders by the more traditional procedures of serologic and RFLP analyses. For instance, in previous studies a DR5-related DRB RFLP pattern, designated "DR5*NAURU," was detected both in Micronesians [9] and in Polynesians [8], in whom it was associated with different HLA-B antigens (B35 and B39, respectively) [5], and with different DQA1 RFLPs in the two groups. The DRI2 allele could be confused with DR6 when HLA-DR serology was in its infancy [30], possibly accounting for an earlier report of a high frequency of a B35-1inked DR6 allele in Balinese [4]. We can now report that DR6-related alleles are virtually absent from Javanese and that the different DR,DQ haplotypes associated with "DR5*NAURU" indicate different DRB1 alleles: DRBI*I201, found almost exclusively in Polynesians in this study, and DRBI*1202, essentially confined here to Micronesians and Javanese.

158

TABLE 4

X. Gao et al.

H L A - D R , D Q two-locus haplotypes with significant linkage disequilibrium in Polynesians, Micronesians, and Javanese Polynesian n = 199 ~

Haplotype DRB1-DQA1 1501-0102 1502-0101 1502-0102 1602-0102 0401-0301 0403-0301 0405-0301 1101-0501 1201-0501 1202-0601 1302-0102 1401-0101 1405-0101 1408-0101 07 -0201 0802-0401 0803-0103 0901-0301 1001-0101 DRB1-DQB1 1501-0502 1501-0602 1502-0501 1502-0601 1602-0502 0301-0402 0403-0302 0405-0402 1101-0301 1201-0301 1202-0301 1401-0502 1408-0503 07 -0201 07 -0303 0802-0402 0803-0601 0901-0303 1001-0501

HF

D

Micronesian n = 129 HF

D

2.3

2.2

5.6

4.4

2.0

1.9

16.9

12.9

1.3 17.9 4.1 12.3 20.4

0.8 11.9

2.0 16.0

1.6 13.1

2.7 8.0 13.1

4.0

3.7

31.2

21.5

1.0 4.1

1.0 3.7

3.2

2.8

5.4 1.3 2.8 8.4 8.9

4.8

2.0

2.0

2.3

2.1

1.2 2.7 7.7 5.9

1.2 4.0

8.1

5.2

1.1 3-6

HF

3.3

4.9

1.2 17.3

1.0 12.5

]..2 16.0 4.0

1.1 13.2 2.5

14.5 3.5 7.9 13.2

30.6

19.0

3.1 5.4 1.3

2.9

2.8

2.7

5.0

4.0

3.7

2.8

13.7

10.2

]..3

1.1

49.0

23.7

11.7

10.4

2.6 2.6

2.5 2.2

2.8 7.3 8.9

2.6 6.4 8.1

2.6

2.3

12.2 5.1

9.2 3.8

1.3

1.2

49.0

23.0

8.8

7.7

2.6

2.5

2.6

2.2

2.4

8.1

D

7.4

17.9 3.8 12.3 20.7

1.2

Javanese n = 154

1.9

6.0

Number of haplotypes examined. HF, haplotype frequency ( × 100) (HF < I% are not listed); and D, deIta value (× 100).

The two DR12 alleles were associated not only with different D Q A 1 alleles, but also with different DRB3 alleles. Among the three Polynesian populations, HLA class II allele distributions were very similar; the particular features were high frequencies of DRBI*0403,

DRBI*llO1, and DRBI*I20I, and the near absence of DR2-related alleles. Unique to Polynesians among Oceanic populations was the presence of DRBI*0802, described previously only in Amerindians [31]; other Polynesian features were the relatively high frequencies of DRBI*0901, a common allele in East Asians [32], and the novel allele DRBI*1408. The Javanese and Micronesians populations shared extraordinarily high frequencies of DRBI*1202, which, together with high frequencies of DRBI*I502, clearly distinguished them from Polynesians. Kiribati, geographically close to the Polynesian nation of Tuvalu, showed little Polynesian influence except that DRBI*I408, not found in Nauru, w a s prevalent. The origins of the Polynesians remain a matter for debate. Some conclude Polynesians evolved out of Melanesia (e.g., Allen and White [33]; Houghton [34]), mainly because the Bismarck Archipelago region near Rabaul has been suggested as a source of Lapita or Polynesian culture [35]; others suggest a substantial Melanesian contribution on the basis of the high frequency (15%) of a Melanesian-specific a-globin deletion in Eastern Polynesia [36]. HLA class I analyses have shown that the Polynesian repertoire can be derived from the East-Asian gene pool, except that Polynesians and Amerindians share uniquely high frequencies of HLA-B48 [5] that may reflect common Mongoloid ancestry rather than direct contact. However, some South American connection, possibly by seagoing Polynesians, is indicated by the prehistoric introduction of the Andean sweet potato to Eastern Polynesia [37]. Linguistic reconstructions of Austronesian languages suggest that proto-Austronesian expansion from Taiwan and Southern China some 6000 years ago led to coastal settlements in the Philippines, northern Borneo, and Sulawesi, with later dispersal to Java and Oceania [10]. By 3 0 0 0 years ago, pre-Polynesians had carried Lapita-style pottery as far east as Samoa, via island Melanesia [38]. For Micronesia, linguistic and archeologic evidence suggest two separate waves of migration about 3500 years ago, one from island Southeast Asia (represented by the Western Malayo-Polynesian speakers of Palauan and Chamorro languages) and another from island Melanesia, giving rise to nuclear Micronesian languages that include those spoken in Nauru and Kiribafi [18]. Our analysis of H L A - D R , D Q haplotype distributions shows a common ancestry of Micronesian and Javanese populations that is compatible with linguistic and archeologic evidence that they derived from the same ancestral gene pool, because the DRBI*1202 RFLP occurs commonly in Southern Chinese (Hong Kong and Singapore) [30]. Further, the presence of DRBI*0403 in Micronesians, but not in Javanese, is compatible with linguistic evidence for dual settlement

HLA Class II Polymorphisms in Pacific Islanders

of Micronesia, because DRBI*0403 is common in coastal Melanesians [21]. This supports findings based on HLA class I [1] and globin gene polymorphisms [39] that there is a substantial Melanesian contribution to the Micronesian gene pool. The presence of DR7 in Javanese, but not in Micronesians, may reflect the influence of later immigrants from China, because this allele is common in Sino-Tibetan speakers, but is rare in Japanese [32]. The H L A - D R , D Q distributions in Polynesians are not so readily reconciled with prehistoric linguistic and archeologic reconstructions. There is virtually no overlap between the HLA class II profiles in contemporary Polynesian and Javanese populations. Mongoloid elements DRBI*0405, 1201, 0802, and 0901, particularly DRBl*0802, could suggest an origin further north than insular Southeast Asia, which is not inconsistent with findings that Polynesians cluster with Ainu and Japanese on craniometric analyses [40] rather than with other peoples of Oceania. Similarly,/3-globin haplotypes cluster Micronesian and Javanese distinctively from Polynesians [41], but the high frequency of DRBI*0403 does suggest a Melanesian contribution to the Polynesian gene pool. N o n - H L A genetic markers indicate population bottlenecks or small population sizes have had a profound influence on the evolution of the Polynesians, as evidenced in the restricted V N T R (variable number of tandem repeats) repertoire [42] and by the near loss of wild-type mitochondrial D N A [43]. The apparently unique and restricted HLA class II profile in Polynesians also suggests rapid evolution, exemplified by the isolated population of Niue, which had the highest frequencies of DRBI*0403 and 1201 and apparent loss of

DRB 1*0405. The unusual DR5-, DR6-, and DR8-related DRB1, DQA1 RFLP haplotypes previously reported in Pacific Islanders were shown in this study to be associated with unusual DRB1 alleles. The DR5 alleles DRBI*I201 and DRBI*I202 differ by a single base pair in codon 67 (ATC v TTC) [44], but are in linkage disequilibrium with different DQA1 and DRB3 alleles. The DR6 alleles DRBI*1401 and DRBI*1408 differ by a single base pair in codon 57 (GCT v GAT) [28], but are in linkage disequilibrium with different DRB3, and D Q B 1 alleles in Pacific Islanders. Similarly, the DR8 alleles DRBI*0801 and DRBI*0803 differ only at codon 67 (TTC v ATC) [45], but have different D Q A 1 , D Q B 1 linkage arrangements in Pacific Islanders compared with Caucasoids [29]. This suggests that the rate of recombination between H L A - D R B 1 and HLA-DQA1 loci is not much greater than the mutation rate at HLA-DRB 1; only for DR2 alleles were there novel D R B 1 , D Q A 1 haplotypes that did not correlate with detectable nucleotide substitutions at the HLA-DRB1 locus.

159

The Polynesian HLA class II genetic repertoire is not readily derived from the island Southeast Asian gene pool, contrary to linguistic and archeologic reconstructions. The Javanese, themselves a derivative population, may not be ideal for comparison with Polynesians, but serologic studies in Malays [46], Balinese [4], and Micronesians [1] show that high frequencies of B35 and DRI2 are characteristic of mainland and island Southeast Asia; HLA-B35 [5] and DRBI*1202 have frequencies of < 1% in Polynesians. Perhaps then the HLA class II system is not ideal in searching for population affinities, with a large number of alleles at relatively low frequencies making it particularly vulnerable to founder effects and genetic drift, and its role in immune response making it potentially subject to natural selection [47]. Although Mongoloid ancestry is clearly evident in Polynesians, the HLA class II profile may well have changed beyond recognition of any immediate ancestral source.

ACKNOWLEDGMENT

These studies have been supported in part by grant DK25446 from the National Institute of Diabetes and Digestive and Kidney Disease.

REFERENCES 1. Serjeantson SW, Kirk RL, Ranford P, Beauchamp M: HLA antigens and non-HLA chromosome 6 markers in Micronesians from Nauru. Tissue Antigens 22:49, 1983. 2. Dagger J, Faed J, Fong R, Milligan L, Roberts M, Thomas H, Webster A, Woodfield DG, Whyte J: HLA antigens in New Zealand Maoris. In Aizawa A (ed): HLA in AsiaOceania 1986. Sapporo, Hokkaido University Press, 1986, p 328. 3. Paik YK, Yoshida M, Wong L, Park MS: HLA in native Hawaiians. In Aizawa M (ed): HLA in Asia-Oceania 1986. Sapporo, Hokkaido University Press, 1986, p 308. 4. Breguet G, Wolnizer CM, Doran T, Bashir H, Jeannet M, Benz6nana G, Ney R, Adiputra N, Scherz R, Blake NM: HLA antigens in Bali (Indonesia) with a special reference to an isolated community. Tissue Antigens 20:229, 1982. 5. Serjeantson SW: HLA genes and antigens. In Hill AVS, Serjeantson SW (eds): The Colonization of the Pacific: A Genetic Trail. Oxford, Oxford University Press, 1989, p 126. 6. Cohen N, Brauther C, Font M-P, Dausset J, Cohen D: HLA-DR2-associated Dw subtypes correlate with RFLP clusters: most DR2 IDDM patients belong to one of these clusters. Immunogenetics 23:84, 1986. 7. Trowsdale J, Young JAT, Kelly AP, Austin PJ, Carson S, Meunier H, So A, Erlich HA, Spielman RS, Bodmer J,

X. Gao et al.

160

Bodmer WF: Structure, sequence and polymorphism in the HLA-D region. Immunol Rev 85:5, 1985. 8. Serjeantson SW, White BS, Jazwinska EC, Yenchitsomanus PT, Mickleson KNP, Trent RJ: HLA-DR and -DQ polymorphisms: new linkage relationships established by RFLP genomic typing in Polynesians and Melanesians. Hum Immunol 20:145, 1987. 9. Kohonen-Corish MR, Serjeantson SW: RFLP analysis of HLA-DR and -DQ genes and their linkage relationships in the Pacific. A m J Hum Genet 39:751, 1986. 10. Blust RA: The Austronesian homeland: a linguistic perspective. Asian Perspect 26:45, 1988. 11. Todd JA, Bell J1, McDevitt HO: HLA-DQ3 gene contributes to susceptibility and resistance to insulin-dependent diabetes mellitus. Nature 329:599, 1987. 12. Gao X, Fernandez-Vifia M, Shumway W, Stasmy P: DNA typing for class II HLA antigens with allele-specific or group-specific amplification: typing for subsets of HLA-DR4. Hum Immunol 27:40, 1990. 13. Gao X, Moraes JR, Miller S, Stasmy P: DNA typing for class II HLA antigens with allele-specific or group-specific amplificauon. V. Typing for subsets of HLA-DR1 and HLA-DR'Br'. Hum Immunol 30:147, 1991. 14. King H, Taylor R, Koteka G, Nemaia H, Zimmet P, Bennett PH, Raper RL: Glucose tolerance in Polynesia. Med J Aust 145:505, 1986. 15. Clark R: Language. In Jennings JD (ed): The Prehistory of Polynesia. Cambridge, MA, Harvard University Press, 1979, p 249. 16. Zimmet P, King H, Taylor R, Raper LR, Balkau B, Borger J, Heriot W, Thoma K: The high prevalence of diabetes mellitus, impaired glucose tolerance and diabetic ~etinopathy in Nauru: the 1982 survey. Diabetes Res 1:13, 1984. 17. King H, Taylor R, Zimmet P, Pargeter K, Raper RL, Beriki T, Tekanene J: Non-insulin-dependent diabetes (NIDDM) in a newly independent Pacific Nation: the Republic of Kiribati. Diabetes Care 7:409, 1984. 18. Bellwood PS: The colonization of the Pacific: some current hypotheses. In Hill AVS, Serjeantson SW (eds): The Colonization of the Pacific: A Genetic Trail. Oxford, Oxford University Press, 1989, p 1. 19. Maniatis T, Fritsch EF, Sambrook J: Molecular Cloning. A Laboratory Manual. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory, 1982. 20. Gao X, Serjeantson SW: Heterogeneity in HLA-DR2related DR,DQ haplotypes in eight populations of AsiaOceania. Immunogenetics 34:401, 1991. 21. Gao X, Serjeantson SW: Diversity in HLA-DR4-related DR,DQ haplotypes in Australia, Oceania and China. Hum Immunol 32:269, 1991. 22. Scharf SJ, Horn GT, Erlich HA: Direct cloning and sequence analysis of enzymatically amplified genomic sequences. Science 233:1076, 1986.

23. Morel PA, Dormans JS, Todd JA, McDevitt HO, Trucco M: Aspartic acid at position 57 of the HLA-DQ3 chain protects against type I diabetes: a family study. Proc Natl Acad Sci USA 85:8111, 1988. 24. Moraes ME, Fernandez-Vifia M, Stasmy P: DNA typing for class II HLA antigens with allele-specific or groupspecific amplification. IV. Typing for alleles of the HLADR2 group. Hum Immunol 3l:139, 1991. 25. Fernandez-Vifia M, Shumway W, Stastny P: DNA typing for class II HLA antigens with allele-specific or groupspecific amplification. IV. Typing for alleles of the DRw52-associated group. Hum Immunol 27:5 I, 1990. 26. Baur MP, Danilovs JA: Population analysis of HLAA,B,C,DR, and other genetic markers. In Terasaki PI (ed): Histocompatibility 1980. Los Angeles, UCLA Tissue Typing Laboratory, 1980, p 955. 27. Bodmer JG, Marsh SGE, Albert ED, Bodmer WF, Dupont B, Erlich HA, Mach B, Mayr WR, Parham P, Sasazuki T, Schreuder GMTh, Strominger JL, Svejgaard A, Terasaki PI: Nomenclature for factors of the HLA system, 1990. Hum Immunol 31:186, 1991. 28. Gao X, Veale A, Serjeantson SW: HLA class II diversity in Australian Aborigines: unusual HLA-DRB1 alleles. Immunogenetics 1992 (in press). 29. Fernandez-Vifia M, Gao X, Moraes MEM, Stasmy P: Alleles at four HLA class II loci and their associations determined by oligonucleotide hybridization in five different ethnic groups. Immunogenetics 34:299, 1991. 30. Kohonen-Corish MR, Dunckley H, Serjeantson SW: HLA-DR and -DQ DNA genotyping in seven populations of Asia-Oceania and Australia. Tissue Antigens 32:32, 1988. 31. Gorski J: The HLA-DRw8 lineage was generated by a deletion in the DRB region followed by first domain diversification. J Immunol 142:4041, 1989. 32. Aizawa M, Natori T, Wakisaka A, Konoeda Y: Antigen and gene frequencies of ethnic groups. In Aizawa M (ed): HLA in Asia-Oceania 1986. Sapporo, Hokkaido University Press, 1986, p 1080. 33. Allen J, White PJ: The Lapita homeland: some data and an interpretation. J Polynesian Soc 98:129, 1989. 34. Houghton P: The early human biology of the Pacific: some considerations. J Polynesian Soc 100:167, 1991. 35. Spriggs M: The lapita cultural complex: origins, distribution, contemporaries and successors. In Kirk RL, Szathmary E (eds): Out of Asia: Peopling the Americas and the Pacific. Canberra, J Pacific Hist 1985, p 185. 36. Hill AVS, Bowden DK, Trent RJ, Higgs DR, Oppenheimer SJ, Thein SL, Mickelson KNP, Weatherall DJ, Clegg JB: Melanesians and Polynesians share a unique thalassaemia mutation. Am J Hum Genet 37:571, 1985. 37. Yen DE: The Sweet Potato and Oceania. Honolulu, Bishop Museum Bull 236, 1974. 38. Spriggs M: The dating of the Island Southeast Asian Neo-

HLA Class II Polymorphisms in Pacific Islanders

lithic: an attempt at chronometric hygiene and linguistic correlations. Antiquity 63:587, 1989. 39. O'Shaughnessey DF, Hill AVS, Bowden DK, Weatherall DJH, Clegg JB: Globin genes in Micronesia: origins and affinities of Pacific Island peoples. Am J Hum Genet 46:144, 1990. 40. Brace CL: Micronesians, Asians, Thais and relations: a craniofacial and odontometric perspective. Micronesia Suppl 2:323, 1990. 41. Chen LZ, Easteal S, Board PG, Kirk RL: Evolution of/3globin haplotypes in human populations. Mol Biol Evol 7:423, 1990. 42. Flint J, Boyce AJ, Martinson JJ, Clegg JB: Population bottlenecks in Polynesia revealed by minisatellites. Hum Genet 83:257, 1989. 43. Hertzberg M, Mickleson KNP, Serjeantson SW, Prior JF, Trent RJ: An Asian-specific 9 base pair deletion of

161

mitochondrial DNA is frequently found in Polynesians. Am J Hum Genet 44:504, 1989. 44. Abe A, Ito I, Ohkubo M, Kaneko T, Ito K, Kato H, Kashiwagi N, Obata F: Two distinct subtypes of the HLA-DRwl2 haplotypes in the Japanese population detected by nucleotide sequence analysis and oligonucleotide genotyping. Immunogenetics 30:422, 1989. 45. Marsh S, Bodmer J: HLA class II nucleotide sequences, 1991. Immunogenetics 33:321, 1991. 46. Sukumaran KD, Joo OK, Har DS, Thavemalar, Lopez CG: HLA antigen, gene and haplotype frequencies of Malays. In Aizawa M (ed): HLA in Asia-Oceania 1986. Sapporo, Hokkaido University Press, 1986, p 266. 47. Hughes AL, Nei M: Nucleotide substitution at major histocompatibility complex class II loci: evidence for overdominant selection. Proc Natl Acad Sci USA 86:958, 1989.