1HIlllllIlO-

Immunogenegcs 34: 247-251, 1991

genetics

© Springer-Verlag 1991

The genomic structure of two ancestral haplotypes carrying duplications

C4A

Katsushi Tokunaga*, Wen Jie Zhang, Frank T. Christiansen, and Roger L. Dawkins Departments of Clinical Immunology,RoyalPerth Hospital, Sir Charles Gairdner Hospital and the Universityof Western Australia, Perth, Western Australia Received January 28, 1991; revised version received April 5, 1991

Abstract. Two major histocompatibility complex (MHC) ancestral haplotypes (AH) HLA A24, Bw52, C2C, BCS, C4A3 + 2, C4BQO , DRw15 , DQw6 (52.1) and HLA A24, CwT, B7, C2C, BfS, C4A3+ 3, C4B1, DR1, DQw5 (7.2), which occur with the haplotype frequencies of approximately 10 % and 4 % respectively in the Japanese population, carry duplicated C4A alleles by C4 allotyping. Southern blot analysis with Taq I indicated that the 52.1 A H has two C4 genes defined by 7.0 kilobase (kb) and 6.0 kb C4 hybridizing fragments but both encode C4A allotypes, being C4A3 and C4A2 respectively. The 7.2 AHcarries two C4A3 and one C4B1 alleles and restriction length polymorphism (RFLP) analysis with Taq I showed that 6.0 kb and 7.0 kb fragments are in the proportion of 2:1. By pulsed field gel electrophoresis (PFGE) analysis, the lengths of the Pvul fragments carrying C4 and Cyp21 genes were approximately 390 kb for 52.1 and 440 kb to 7.2. The results indicate that the RFLP markers do not correlate with C4 isotype (A or B) or allotype and that the C4 gene copy number is a function of the number of genomic blocks containing C4 and Cyp21.

Introduction The human major histocompatibility complex (MHC) includes at least several dozen genes which can be subgrouped into HLA Class I, HLA Class II and the central non-HLA genes. At least most are polymorphic and many are duplicated. The term "ancestral haplotype" (AH) has been used to describe conserved MHC haplotypes which appear to be identical between apparently unrelated subjects * Presentaddress: Departmentof BloodTransfusion, TokyoUniversity Hospital, Tokyo, Japan. Address correspondence and offprint requests to: R.L. Dawkins, Department of Clinical Immunology,Royal Perth Hospital, GPO Box X2213, Perth, Western Australia 6001.

(Dawkins et al. 1983; Kay et al. 1988). These AHs have a specific content of alleles at all MHC loci and have a particular genomic length as determined by pulsed field gel electrophoresis (PFGE) (Tokunaga et al. 1988). More recently, it has been shown that AHs have a particular gene copy number at all loci tested so far (Zhang et al. 1990). In the case of the DRB genes it can be shown that a given A H carries 1, 2, 3, or even more copies (Zhang et al. 1990) which are classified DRB1-5 as if there were alleles at quite separate loci (WHO Nomenclature Committee 1990). By contrast, although each AHhas 1, 2, or 3 C4 genes it is usual to classify these as either C4A or C4B and to assign alleles on the basis of a two locus system. By this convention there are haplotypes that carry 1 × A , 0 × B ; 0 x A , 1 ×B; 1 × A , 2 × B , etc., (Table 1). The assignment of C4 allotypes to A or B is based on electrophoretic mobility (Awdeh and Alper 1980) in the first instance but difficulties do arise. Other characteristics such as hemolytic activity, molecular mass, Chido/ Rodgers antigens, and RFLP have also been used, but the results are often conflicting and therefore confusing (Table 2). On this occasion we describe two ancestral haplotypes with two C4 genes both of which express C4A products by electrophoresis. By RFLP, one of these haplotypes would be classified as having one copy of C4A and one copy of C4B (Schneider et al. 1986). The second haplotype has an 'insertion' of approximately 50 kilobases (kb) resulting in three copies of C4 (and Cyp21). By RFLP it could be concluded that there are two copies of C4B and one copy of C4A, whereas electrophoresis of the products suggests 'homoduplication' of C4A with one copy of C4B.

Materials and methods Blood samples. Peripheral blood lymphocytes(PBL) and ethylenediaminetetraacetate(EDTA)-plasmaswere obtained from homozygousor heterozygous individuals with the AHs HLA .424, Bw52, C2C, BfS,

248

K. Tokunaga et al.: Ancestral haplotypes carrying C4A duplications

Table 1. Ancestral haplotypes used in this study.

techniques using the defined antisera from international and regional workshops.

AH HLA A

Cw

B

7.1 7.2 8.1 18.2 35.2 42.1 44.1 44.3 44.4 46.2 52.1 61.1 62.4 65.1

7 7 7 5 4 2 5 11 8 4 8

7 7 8 18 35 42 44 44 44 w46 w52 w61 w62 w65

3 24 1 w19 3 2 29 w33 2 24 26 11 -

C2 Bf

C4A C4B DR DQw Pvul/C4 n

C C C C C C C C C C C C C

3 3+3 QO 3 3+2 1 3+3 QO 3 4 3+2 3 3 2

S S S F1 F F S S F S S S S S

1 1 1 QO QO QO QO 1 1 2 QO 1 1 1+2

w15 1 3 3 1 3 4 7 w13 8 w15 9 4 1

6 5 2 2 5 4 7 2 6 6 6 9 8 5

390 440 340 350 NT NT 390 NT 390 390 390 390 390 400

4 15 10 10 0 0 0 0 15 12 21 3 2 6

- Not defined. NT= Not tested. n = Number of unrelated haplotypes used in this study. Some haplotypes are given for the purpose of comparison only. Pvul/C4 = fragment lengths (kb) by PFGE after Pvu I digestion and probing for C4 (see Results section, Fig. 4, and Tokunaga et al. 1988, 1989).

Conventional southern blot analysis. Genomic DNA was prepared from buffy coat cells using standard phenol/chloroform/isoamyl alcohol extraction. Digestion by restriction endonuclease Taq I was carried out according to the manufacturer's instructions. Fully digested DNA was electrophoresed in 0.8% agarose at 20 V for 40 h and transferred to nylon membrane (Biotrace, Gelman Sciences, Ann Arbor, MI) by the method of Southern (1975), using 0.4 M NaOH as a transfer medium. The resulting blots were prehybridized and hybridized with a C4 (pATA; Belt et al. 1984) or a Cyp21 (pC21/3c; White et al. 1986) cDNA probe labeled by the random priming method as described previously (Garlepp et al. 1986). PFGE analysis. Preparation of high molecular DNA in agarose blocks, digestion with a rare cutting restriction endonuctease Pvul, separation of digests by pulsed field gel electrophoresis (PFGE) and Southern hybridization were performed as previously described (Anand 1986; Toknnaga et al. 1988). Gene copy number. The use of densitometry has been described previously (Zhang et al. 1990).

Results Table 2. Comparison between C4A and C4B. Nature

C4A

Location Telomeric Proximity to Cyp21 A Rodgers + Chido Hemolytic activity Low Electrophoretic Acidic mobility Taq I RFLP 7.0 kb c~ chain Isotypic residues 1101, 1102, 1105, 1106

96 kd P-C-L-D

C4B

Apparent/possible exceptions

Centromeric B + High Basic

AHs 42.1, 44.1, 52.1 AHs 35. 2, 52.1, 65.1 A1 B5 A6 (inactive) A1, A91; B3, B4, B5

6.4, 6.0, or 5.4 kb 94 kd L-S-I-H

AHs 7.2, 44.1, 44,3, 52.1 ? ?

+ positive negative. ? unknown. References: Dupont 1989; Giles 1987, 1988b; Mauffet al. 1983; O'Neill et al. 1980; Schneider et al. 1986; Roos et al. 1982; White et al. 1986; Yu et al. 1986. -

C4A3+2, C4BQO, DRw15, DQw6 (52.1) and HLA A24, Cw7, B7, C2C, Bj~S, C4A3 + 3, C4B1, DR1, DQw5 (7. 2). Several individuals with other AHs were also examined for comparison. All the individuals were fully genotyped by family study. AHs used in this study are given in Table 1. Complement and HLA allotyping. Typing of C4 was performed on carboxypeptidase and neuraminidase treated EDTA-plasma using agarose gel electrophoresis followed by immunofixation or hemolytic overlay (Mauff et al. 1983; Sim and Cross 1986; Zhang et al. 1988). C2 allotypes were determined by polyacrylamide gel isoelectric focusing and hemolytic overlay (Alper 1976). Bf allotypes were determined by immunofixation electrophoresis (Alper et al. 1972). HLA A, B, C, DR, and DQ specificities were determined by standard microcytotoxicity

A H 52.1 carries C 4 A 3 - C 4 A 2 . T h e 52.1 A H is the m o s t c o m m o n haplotype in the Japanese population with the f r e q u e n c y o f a p p r o x i m a t e l y 10 % (Tokunaga et al. 1985). By C4 allotyping h o m o z y g o t e s for this A H s h o w e d two distinct bands c o r r e s p o n d i n g to C 4 A 3 and C 4 A 2 respectively (Fig. 1, lane 1 and lane 5). T h e two bands w e r e o f similar optical density indicating that there w e r e similar concentrations o f each product and p r e s u m a b l y one copy o f each gene. N o C4B allotype could be o b s e r v e d so that the c o m p l o t y p e was assigned as C 4 A 3 + 2, C 4 B Q O . This conclusion is supported by the p r e v i o u s demonstration that the 5 2 . 1 A H is C h i d o 4 negative and the C 4 A 3 and C 4 A 2 allotypes carry different R o d g e r s specificities (Giles et al. 1988a). Interestingly, C4 probing after Taq I digestion revealed 7.0 kb and 6.0 kb fragments o f equal intensity (Fig. 2, lane 1). The results indicate that one of the C 4 A allotypes is m a r k e d by a 6.0 kb rather than a 7.0 kb fragment. Probing for Cyp21 y i e l d e d 3.7 kb and 3.2 kb fragments o f equal intensity (Fig. 3, lane 4). These observations suggest that there are two C4 and Cyp21 genes on this haplotype. F u t h e r m o r e , after probing for C4 and Cyp21, P F G E analysis o f Pvul-digested D N A f r o m 5 2 . 1 , showed a 390 kb f r a g m e n t which was also o b s e r v e d in other A H s (Fig. 4).

The 7.2 AH carries C4A3-C4A3-C4B1. The frequency of the AH is 4-6% in Japanese (Tokunaga et al. 1985). The homozygous 7. 2 showed both C4A3 and C4B1 bands by C4 allotyping but the intensity for C4A3 was about t w o - f o l d stronger than that for C4B1 as indicated by the C 4 A / B ratio o f 1.9 (Fig. 1, lane 2). M o r e o v e r , in the

K. Tokunaga et al. : Ancestral haplotypes carrying C4A duplications

249

Fig. 3. Cyp21A gene duplication in the 7.2 AH as revealed by comparison of fragment intensities. Genomic DNA from four homozygous individuals (7.1 in lane 1; 7.2 in lane 2; 65.1 in lane 3, and 52.1 in lane 4) were digested as in Figure 2 but probed with Cyp21. In lane 1 and lane 4, no difference in intensity can be observed between 3.2 kb and 3.7 kb fragments, i. e., two copies of Cyp21 are present on the 7.1 and 52.1 AHs. Approximately two-fold ratios (3.2 kb/3.7 kb) are obvious in lane 2 and lane 3 suggesting that there are three copies of Cyp21 in the 7.2 and 65.1 AHs. Fig. 1. C4 allotyping reveals duplicated C4A alleles carried by Japanese ancestral haplotypes 52.1 and 7.2. Five plasma samples with AHs homozygous or heterozygous for 52.1 or 7.2 were treated, separated and visualized as described in the text. The AHs for the samples are as follows: lane 1 and lane 5, homozygous 52.1; lane 2, homozygous 7.2; lane 3, heterozygous 52.1 + Z 2; lane 4, heterozygous 7. 2 + 44. 4. As seen in lane 1 and lane 5, the 52.1 AH shows two distinct C4A alleles, C4A3 and A2 with equal intensities and no C4B band ("homozygous C4B null"). In lane 2, a ratio of 1.9 implies twice as much C4A3 product as C4B1. The complotype for this AH therefore can be assigned as C4A3+3, C4B1. Furthermore, the assignments for the 52.1 and 7.2 AHs were confirmed by testing heterozygotes as shown in lane 3. An A/B ratio of approximately 4 suggests that both AHs are duplicated at the C4A locus. Lane 4 confirms a C4A duplication on the 7. 2 AH when compared to the 44. 4 AH.

Fig. 4. Duplicated Z 2 AH is 50 kb larger than other AHs by PFGE. Five DNA samples were digested with Pvul and subjected to PFGE. After probing with Cyp21 two distinct fragments of 440 kb and 390 kb were observed. In lane 1, the homozygous 7.2 shows a 440 kb fragment while three ceils in lane 3 (44.4+62.4), lane 4 (52.1) and lane 5 (46.2 + 61.1) have fragments of 390 kb. The fact that the 7. 2 AHis larger by approximately 50 kb was confirmed when a heterozygote 7.2 + 44.4 was tested (lane 2), It is known that 390 kb and 440 kb fragments carry both C4 and Cyp21 genes (Tokunaga et al. 1989). Therefore a 50 kb difference in size between the 7.2 and other AHs provides further evidence of a tandem duplication of C4 and Cyp21 genes on this haplotype. h e t e r o z y g o t e s o f 7.2 vs 52.1 o r 7.2 vs 4 4 . 4 , the d e n s i t o m e t r i c ratios, t o g e t h e r w i t h the h o m o z y g o u s data c o n f i r m e d the a s s i g n m e n t o f c o m p l o t y p e C 4 A 3 + 3, C4B 1 for

Fig. 2. Difference in fragment intensities confirms gene duplication with a probe for C4. Genomic DNAs derived from five homozygous individuals were digested with Taq I and probed for C4. The AHs are: lane 1, 52.1; lane 2, 46.2; lane 3, 7.1; lane 4, 7.2, and lane 5, 65.1. As shown in lane 1, lane 2, and lane 3, the intensities of the 7.0 kb Taq I fragments are similar to those of the 6.0 kb or 5.4 kb fragments within each lane. It is of interest to note that in the 52.1 AH only the C4A product was detectable by protein allotyping (Fig. 1). Therefore, the 6.0 kb Taq I C4 gene on this AH actually encodes for a C4A product. A similar situation is seen in lane 4, where the 6.0 kb fragment has stronger signals than the 7.0 kb fragment indicating that there are two 6.0 kb genes and only one 7.0 kb gene on this haplotype. At protein level, however, this haplotype expresses two C4A alleles and one C4B allele suggesting that the 7.0 kb and one of the two 6.0 kb genes encode C4A allotypes. The 65.1 AH seen in lane 5 shows a stronger signal for 5,4 kb relative to 7.0 kb, in keeping with the C4B 1 +2 duplication as shown previously (Garlepp et al. 1986).

7.2. It is t h e r e f o r e c o n c l u d e d that t w o C 4 A 3 alleles and a single C4B1 allele are e x p r e s s e d o n this h a p l o t y p e . H o w e v e r , S o u t h e r n b l o t analysis o f Taq 1-digested D N A and p r o b i n g for C4, r e v e a l e d a m u c h m o r e i n t e n s e 6.0 kb f r a g m e n t as c o m p a r e d to 7.0 kb f r a g m e n t for this A H (Fig. 2, lane 4). T h e r e f o r e , t w o 6.0 k b - a s s o c i a t e d g e n e s are p o s s i b l e w h e n c o m p a r e d to the 7 . 0 kba s s o c i a t e d C4 g e n e . B e c a u s e the A H p r o d u c e s a single C4B1 allotype, it w a s s u g g e s t e d that o n e o f the 6.0 kb Taq I C 4 g e n e s o n this A H i n fact e n c o d e s a C 4 A allotype. Similarly the T a q I digests p r o b e d w i t h Cyp21 y i e l d e d a s t r o n g 3.2 kb f r a g m e n t and relatively w e a k 3.7 kb fragm e n t (Fig. 3), i n d i c a t i n g a C y p 2 1 A duplication. D i r e c t e v i d e n c e o f the d u p l i c a t e d C 4 A and Cyp21 g e n e s w a s also o b t a i n e d b y P F G E analysis as d e m o n -

250 strated in Figure 4. The homozygous 7.2 A H showed a 440 kb band which is some 50 kb larger than those of other haplotypes. This was further confirmed by testing heterozygotes of 7. 2 vs 44. 4 (Fig. 4, lane 2) and the results showed that the intensity of the 440 kb band of the 7.2 A H was stronger than that of the 390 kb band of the 44. 4 AH.

Discussion The concept of ancestral haplotypes arose from the observation that certain combinations of alleles at diverse MHC loci (supratypes) were highly conserved in that they occurred in individuals who were not known to be related and must, therefore, have been separated by many generations (Dawkins et al. 1983). The possibility that these haplotypes have been conserved en bloc was tested by examining the genomic structure and comparing polymorphisms at intervening loci. Several studies have now shown that it is possible to identify some thirty different but highly conserved haplotypes within Caucasoids. Each haplotype has a specific length between HLA B and DQ and each has a given content of alleles, deletions, and insertions (Tokunaga et al. 1988; Tokunaga et al. 1989). At several regions there are variations in gene copy number but every AH has the same gene copy number (Zhang et al. 1990). These findings would seem to imply retention/selection for large genomic segments rather than individual coding regions (Klein and Takahata 1990). Elsewhere, we have suggested that there might be some form of interaction between the products of different classes of MHC genes. An obvious possibility is that various segments of the MHC encode products which are involved in one or other aspect of immunoregulation (French and Dawkins 1990). In order to pursue the analysis we chose to study the C4 genes in more detail. As expected, we show here that two Japanese AHs have a specific genomic structure and content. The data support the concept of AHs as genomic segments which are conserved en bloc. The two haplotypes characterized here are of particular interest because both contain two C4 genes which encode C4A products. One contains two genes expressing different C4A alleles and the other contains three genes of which two encode C4A products but the third expresses a C4B protein. Much of the present confusion surrounding the terminology of C4 genes appears to have derived from the assumption that there is a simple two locus system reflected by A and B products (by electrophoresis) and by A or B genomic patterns (by RFLP). In fact, as shown in Table 2, there appear to be exceptions to any single method of classifying C4 genes and products into a two locus system. Although C4 genes marked by Taq I 7.0 kb fragments often encode C4A products and although genes marked by Taq I 6.4, 6.0, and 5.4 kb fragments

K. Tokunaga et al. : Ancestral haplotypes carrying C4A duplications often encode C4B products (Schneider et al. 1986), we show here that there are exceptions. One of the common Japanese haplotypes studied (7.2) has a C4 gene marked by a 6.0 kb fragment and expresses C4A3. Similarly, the second, even more common haplotype (52.1), has a 6.0 kb fragment and expresses C4A2. Furthermore, we have shown (unpublished data) that the 44. 3 haplotype contains a C4 gene marked by a 7.0 kb fragment but with a C4B1 product and the 44.1 A H contains a C4 gene marked by a 6.0 kb Taq I fragment but encoding C4A3. Previously Yu and Campbell (1987) have provided evidence that a 6.0 kb Taq I gene can encode C4A3. It follows that there is no strict relationship between the RFLP classification and the nature of the protein product. It is pertinent that the same RFLP patterns are seen in nonhuman primates as well as humans (Dawkins et al. 1989; Kawaguchi et al. 1990). Most chimpanzees have C4 genes marked by 6.4 and 5.4 kb fragments. The 6.4 fragment is also found in the gorilla. In the orangutan 7.0, 6.0, and 5.4 fragments are found. These observations suggest that nonhuman primates and humans may have various combinations of primordial C4 genes represented by the four different Taq I fragments. Different species have retained different frequencies of the four subtypes but apparently most, if not all, encode a functional product (Klein et al. 1990). It should be noted that few, if any, human haplotypes lack a C4 gene completely (Uring-Lambert et al. 1989) suggesting that at least one gene in the region may be necessary for survival. A remarkable feature of the C4 system is that the genomic structure suggests that there are 1, 2, 3, or even more block segments of some 40-50 kb (Carroll et al. 1985; Garlepp et al. 1986; Collier et al. 1989; Dunham et al. 1989) which contain C4, Cyp21, and probably other genes (Levi-Strauss et al. 1988; Morel et al. 1989). Thus, those haplotypes containing one C4 gene also contain one Cyp21 gene and a Pvul fragment of approximately 340-350 kb (Table 1); those containing two C4 and two Cyp21 genes have Pvul fragments of approximately 390 kb; those with three C4 and three Cyp21 genes appear to have fragments of some 400-440 kb. These findings suggest that segments are inserted or deleted en bloc rather than that single C4 genes are duplicated or deleted. Acknowledgments. We are gratefi,dto M. MacDougalland B. Harvey for preparingthe manuscript. This work was supportedby the National Healthand MedicalResearchCouncil, WesternAustralianArthritis and RheumatismFoundationand the ImmunogeneticsResearchFoundation. •PublicationNo. 9020 of the Departmentsof ClinicalImmunology,Royal Perth Hospital, Sir Charles Gairdner Hospital, and the University of Western Australia.

References Alper, C.A.: Inherited structural polymorphismin C2: evidence for genetic linkage between C2 and Bf. J Exp Med 144." 1111-1115, 1976

K. Tokunaga et al.: Ancestral haplotypes carrying C4A duplications Alper, C.A., Boenisch, T., and Watson, L.: Genetic Polymorphism in Human Glycine-rich Glycoprotein. JExp Med 135: 86-90, 1972 Anand, R.: Pulsed field gel electrophoresis: a technique for fractioning large DNA fragments. Trends Genet 23: 278, 1986 Awdeh, Z. L. and Alper, C. A.: Inherited structural polymorphism of the fourth component of human complement. Proc Natl Acad Sci USA 77." 3576-3580, 1980 Belt, K. T., Carroll, M.C., and Porter, R.R.: The structural basis of the multiple forms of human complement component C4. Cell 36: 907-914, 1984 Carroll, M. C., Palsdottir, A., Belt, K. T., and Porter, R. R.: Deletion of complement C4 and steroid 21-hydroxylase genes in the HLA class III region. EMBO J 4: 2547-2552, 1985 Collier, S., Sinnott, P.J., Dyer, P. A., Price, D. A., Harries, R., and Strachan, T.: Pulsed field gel electrophoresis identifies a high degree of variability in the number of tandem 21-hydroxylase and complement C4 gene repeats in 21-hydroxylase deficiency haplotypes. EMBO J 8: 1393-1402, 1989 Dawkins, R.L., Christiansen, F.T., Kay, P.H., Garlepp, M., McCluskey, J., Hollingsworth, P.N., and Zilko, P.J.: Disease associations with complotypes, supratypes, and haplotypes. Immunol Rev 70: 5-22, 1983 Dawkins, R. L., Kenter, M., Bontrop, R., Lawler, D., Christiansen, F. T., Saueracker, G., Martin, E., Leaver, A. L., Parham, P., Van Rood, J. J., and Giphart, M.: Interspecies MHC homology at the DNA level. In M. A. Hardy (ed.) Xenograft 25, pp. 191-198, New York Elsevier Science, 1989 Dunham, I., Sargent, C.A., Dawkins, R.L., and Campbell, R.D.: Direct observation of the gene organization of the complement C4 and 21-hydroxylase-loci by pulsed field gel electrophoresis. J Exp Med 169: 1803-1806, 1989 Dupont, B.: RFLP stadardization reports. In B. Dupont (ed.): Immunobiology of HLA, Vol I: Histocompatibility Testing 1987, pp. 575-670, Springer, New York, 1989 French, M. A. H. and Dawkins, R. L.: Central MHC genes, IgA deficiency and autoimmune disease. Immunol Today 11: 271-274, 1990 Garlepp, M. J., Wilton, A. N., Dawkins, R. L., and White, P. C.: Rearrangement of 21-hydroxylase genes in disease associated MHC supratypes. Immunogenetics 23: 100-105, 1986 Giles, C. M.: Three chido determinants detected on the B5Rg of human C4: their expression in Ch-types donors and families. Hum Immunol 18: 111-122, 1987 Giles, C. M., Tokunaga K., Zhang, W. J., Tanaka, H., Endoh, N., and Juji, T. : The antigenic determinants, Rg/Ch/WH, expressed by Japanese C4 allotypes. Immunogenetics 15: 267-275, 1988a Giles, C.M., Uring-Lambert, Goetz, J., Hauptmann, G., Fielder, A. H. L., Rittner, C., and Robson, T.: Antigenic determinants expressed by human C4 allotypes; a study of 325 families provides evidence for the structural antigenic model. Immunogenetics 27." 442-448, 1988b Kawaguchi, H., Golubic, M., Figueroa, F., and Klein, J.: Organization of the chimpanzee C4-CYP21 region: implications for the evolution of human genes. Eur J Immunol 20: 739-745, 1990 Kay, P.H., Dawkins, R.L., Williamson, J., Tokunaga, K., Christiansen, F.T., and Charoenwong, P.: Coexistence of an MHC chromosomal segment marked by HLA B17, BJS, C4A6, B1, DR7, and DQw9 in different ethnic groups. Hum Immunol 23: 27-35, 1988 Klein, J. and Takahata, N. : The major histocompatibility complex and the quest for origins. Immunol Rev 113: 5-25, 1990 Klein, J., Bontrop, R.E., Dawkins, R.L., Erlich, H.A., Gyllensten, U.B., Heise, E. R., Jones, P.P., Parham, P., Wakeland, E.K., and Watkins, D. I.: Nomenclature for the major histocompatibility complexes of different species: a proposal. Immunogenetics 31:

251 217-219, 1990 Levi-Strauss, M., Carroll, M.C., Steinmetz, M., and Meo, T.: A previously undetected MHC gene with an unusual periodic structure. Science 240: 201-204, 1988 Mauff, G., Alper, C.A., Awdeh, Z., Batchelor, J.R., Bertrams, J., Dawkins, R.L., and others: Statement on the nomenclature of human C4 aUotypes. Immunobiol 164: 184-191, 1983 Morel, Y., Bristow, J., Gitelman, S. E., and Miller, W. L.: Transcript encoded on the opposite strand of the human steroid 21-hydroxylase/complement component C4 gene locus. Proc Natl Acad Sci USA 86: 6582-6586, 1989 O'Neill, G.J., MiNter, P., Pollack, M. S., and Dupont, B.: Different HLA antigen associations for the functionally active and inactive products of the complement C4F1 allele. Hum Immunol 1: 23-30, 1980 Roos, M.H., Mollenhauer, E., Demant, P., and Rittner, C.: A molecular basis for the two locus model of human complement component C4. Nature 298: 854-856, 1982 Schneider, P. M., Carroll, M. C., Alper, C. A., Rittner, C., Whitehead, A. S., and Colten, H. R.: Polymorphism of the human complement C4 and steroid 21-hydroxylase genes. Restriction fragment length polymorphisms revealing structural deletions, homoduplications, and size variants. J Clin Invest 78: 650-657, 1986 Sim, E. and Cross, S. J.: Phenotyping of human complement component C4, a class 11"IHLA antigen. Biochem J 239: 763-767, 1986 Southern, E.M.: Detection of specific sequences among DNA fragments separated by gel electrophoresis. J Mol Bio198: 503-517, 1975 Tokunaga, K., Omoto, K., Akaza, T., Akiyama, N., Amemiya, H., Naito, S., Sasazuki, T., Satoh, H., and Juji, T.: Haplotype study on C4 polymorphism in Japanese. Association with MHC alleles, complotypes and HLA-complement haplotypes. Immunogenetics 22: 359-365, 1985 Tokunaga, K., Saueracker, G., Kay, P. H., Christiansen, F. T., Anand, R., and Dawkins, R. L.: Extensive deletions and insertions on different MHC supratypes detected by pulsed field gel electrophoresis. j Exp Med 168: 933-940, 1988 Tokunaga, K., Kay, P.H., Christiansen, F.T., Saueracker, G., and Dawkins, R.L.: Comparative mapping of the human major histocompatibility complex in different racial groups by pulsed field gel electrophoresis. Hum Immunol 26: 99-106, 1989 Uring-Lambert, B., Mascard-Lemone, F., Tongio, M. M., Goetz, J., and Hauptmann, G. : Molecular basis of complete C4 deficiency: A study of three patients. Hum lmmunol 24: 125-132, 1989 White, P. C., New, M. I., and Dupont, B.: Structure of human steroid 21-hyroxylase genes. ProcNatlAcadSci USA 83: 5111-5115, 1986 WHO Nomenclature Committee: Nomenclature for factors of the HLA system, 1989. Immunogenetics 31: 131-140, 1990 Yu, C. Y. and Campbell, R. D.: Definitive RFLPs to distinguish between the human complement C4A/C4B isotypes and the major Rodgers/Chido determinants: Application to the study of C4 null alleles, lmmunogenetics 25: 383-390, 1987 Yu, C.Y., Belt, K.T., Giles, C.M., Campbell, R.D., and Porter, R. R. : Structural basis of the polymorphism of human complement components C4A and C4B: Gene size, reactivity and antigenicity. EMBO J 5: 2873-2881, 1986 Zhang, W.J., Kay, P.H., Cobain, T.J., and Dawkins, R. L. : C4 allotyping on plasma or serum: application to routine laboratories. Hum Immunol 21: 165-171, 1988 Zhang, W.J., Degli-Esposti, M.A., Cobain, T.J., Cameron, P.U., Christiansen, F. T., and Dawkins, R. L.: Differences in gene copy number carried by different MHC ancestral hyplotypes: quantitation after physical separation of haplotypes by Pulsed Field Gel Electrophoresis. J Exp Med 171: 2101-2114, 1990