- Email: [email protected]
Haplospecific polymorphism between HLA B and tumor necrosis factor
Haplospecific Polymorphism Between HLA B and Tumor Necrosis Factor Xiongwen Wu, Wen Jie Zhang, Campbell S. Witt, Lawrence J. Abraham, Frank T. Christiansen, and Roger L. Dawkins
ABSTRACT: Polymorphisms were sought between HLA B and tumor necrosis factor (TNF) using three genomic probes. Extensive polymorphism was detected within a panel of 50 cell lines including 37 homozygotes representing 21 different ancestral haplotypes (AH). Following Taq I digestion of genomic DNA, we observed three allelic patterns with probe X (RI7A) and four with probe V (RgA). Seven different allelic patterns were found with probe Y (M20A) after Taq I + Rsa I digestion. Family studies showed that the Y, X, and V alleles were inherited and segregated with HLA haplotypes. A striking feature of the allelic patterns detected by these probes was that cells with the same AH had identical Y, X, and V alleles (i.e., the alleles were haplotypic). Of 15 different Y-X-V haplotypes observed, 11 were found to be specific for a particular AH (i.e., were haplospecific).
ABBREVIATIONS AH ancestral haplotype MHC major histocompatibility complex RFLP restriction fragment length polymorphism
Four were shared by more than one AH, but in these instances there were extensive similarities in other regions within the major histocompatibility complex (MHC), for example, the Japanese 46.2 (HLA Bw46DRw8) and the Chinese 46.1 (Bw46-DR9) share all alleles between HLA C and C4 and differ only in class II, suggesting their relatively recent divergence by recombination between C4 and DR. Surprisingly, two insulindependent diabetes mellitus (IDDM)-resistant but racespecific AHs 52.1 (Bw52-DRB 1" 1502, Japanese) and 7.1 (B7-DRB 1"1501, Caucasoid) carry the same Y-X-V haplotype, suggesting the possibility of localizing gene(s) relevant to IDDM. The present study confirms that MHC AHs have been conserved en bloc, including the region between HLA B and TNF. Human Immunology 33, 8997, (1992)
SSO TNF
sequence-specific oligonucleotide tumor necrosis factor
INTRODUCTION The m a j o r histocompatibility complex (MHC) is characterized by polymorphism. All polymorphisms, whether within coding or noncoding sequences at M H C classes I, II, and c o m p l e m e n t loci are haplotypic and often haPublication number 9053 of the Department of Clinical Immunology. Royal Perth Hospital, Sir Charles Gairdner Hospital and the University of Western Australia, Perth, Western Australia. From the Departments of Clinical Immunology, Royal Perth Hospital Sir Charles Gairdner Hospital, and the University of Western Australia, Perth, Western Australia. AddreJs reprint requests to R. L. Dawkins, Department of Clinical Immunology, Royal Perth Hospital, GPO Box X2213, Perth, Western Australia 6001. Received August 26, 1991; accepted October 30, 199•. Human Immunology 33, 89-97 (1992) © American Society for Histocompatibility and Immunogenetics, 1992
plospecific, as shown by the collaborative studies on a cell panel o f well-defined M H C haplotypes within the 10th International Histocompatibility Workshop [1]. The fact that polymorphisms are invariably haplotypic indicates that M H C haplotypes have been conserved en bloc. Thus, apparently unrelated individuals often share the same M H C haplotypes. Accordingly, we use the term ancestral haplotype (AH) to refer to conserved M H C haplotypes that have been maintained through many generations of an ancestral family [ 2 - 4 ] . During recent years a n u m b e r o f novel genes have been identified and m a p p e d to the region between H L A B and C2 [ 5 - 7 ] . W e predicted that this region would 89
0198-8859/92/$5.00
90
X. Wu et al.
contain polymorphic sequences that would provide haplotypic and/or haplospecific markers for different AHs. This was shown to be the case for tumor necrosis factor (TNF) [8] and BAT3 [4]. In this report we have characterized the 250-kb region between the HLA B and TNF by using four genomic probes Y (M20A), X (R17A), V (R9A), and W (RSA) [5, 6] in conjunction with a panel of wellcharacterized cell lines. MATERIALS A N D M E T H O D S
Cell lines. Fifty Epstein-Barr virus-transformed human cell lines, including 24 studied during the 10th International Histocompatibility Workshop, were selected. The local cell lines were genotyped for HLA classes I, II, and complement loci by serology and/or restriction fragment length polymorphism (RFLP) typing. The cell panel covered 21 different AHs (Table 1). AHs have been named according to the HLA B allele characterizing that AH [4]. Genomic DNA was extracted from the cell lines according to the standard phenol/chloroform/ isoamyl alcohol method. Disease associations with some of these AHs have been reported previously [2, 9-12].
D N A probes. Genomic DNA probes Y, X, V, and W were approximately 200-300 bp BamH I + EcoR I fragments derived from the clones M20A, R17A, R9A, and R5A, respectively (kindly provided by Dr. T. Spies,
TABLE 1
Boston MA). The locations of the probes are centromeric to the HLA B, approximately 36 kb for probe Y, 73 kb for probe X, 100 kb for probe W, and 128 kb for probe V [5] (Fig. 1). The probes were labeled with c~[~2p]-dCTP by random priming [13].
Electrophoresis and Southern blot analysis. Genomic DNA (10 ttg) was digested to completion with the Taq I (Promega, Madison, WI) according to the manufacturer's instruction. In some blots double digestions of Taq I + Rsa I (Promega) were used since additional fragments and improved patterns were obtained with the Y probe. The digests were electrophoresed on 1.1% agarose gels at 25 V for 62 hours (or on 0.8% agarose gels at 40 V for 18 hours) in 1 x TBE buffer at room temperature (RT) and transferred to nylon membranes (GeneScreenPlus, Du Pont, USA) by the method of Reed and Mann [14], using 0.4M NaOH as a transfer medium. Prehybridization and hybridization of the membranes were performed in hybridization bottles (Hybaid, Middlesex, UK) at 42°C. The membranes were washed twice with 2 x SSPE for 10 minutes at RT, once in 2 × SSPE + 0.5% SDS at 65°C for 10 minutes, and once in 0.5 x SSPE at 65°C for 5-10 minutes. Kodak X-OMAT films (Rochester, NY) were exposed to the membranes at -70°C for 1-10 days according to standard methods. The sizes of the restriction fragments were determined by reference to a 1-kb DNA ladder (BRL, Gaithersburg, MD), Probes were stripped off the
Ancestral haplotypes used in this study
AH
HLA A
Cw
B
C2
Bf
C4A
C4B
DR
DQw
Race
n
7.1 7.2 8.1 13.1 18.1 18.2 35.1 42.1 44.1 44.2 44.3 46.1 46.2 47.1 52.1 54.1 57.1 60.3 62.1 62.2 65.1
3 24 1 30 25 30 ? ? 2 29 29 2 2 3 24 ~ 1 2 2 ~ >
7 7 7 6 ~ 5 4 2 5 4 ? 11 11 6 ~ 1 6 3 3 3 8
7 7 8 13 18 18 35 42 44 44 44 46 46 47 52 54 57 60 62 62 65
C C C C QO C ? c C c ~ C C C C C C C C B C
S S S S S F1 S F S F S S S F S S S S S S S
3 3 + 3 Qo 3 4 3 3 1 3 3 Qo 4 4 91 3 + 2 3 6 QO 3 4 2
1 1 1 I 2 Qo 1 Qo Qo 1 1 2 2 Q0 QO 5 1 2 3 2 1 + 2
15 1 3 7 15 3 11 3 4 7 7 9 8 7 15 4 7 13 4 4 1
6 5 2 2 6 2 7 4 7 2 2 9 6 2 6 4 9 6 8 8 5
C M c C/M c c C N c C M M M C M M C/M/N C C C C
8 6 11 3 2 6 4 2 5 3 2 8 4 2 6 3 6 4 6 2 6
Abbreviations: n - number of haplotypes tested; ? = not defined; Race: C ~ Caucasoid, M = Mongoloid, N = Negroid.
Genomic Conservation and Ancestral Haplotypes
HLA Class II
Central non-HLA
91
tocol similar to the 32p protocol. Signal detection was achieved by dipping membranes into a chemiluminescent substrate for H RP (RPN 2105, Amersham, UK) and subsequent exposure of the membrane to a photographic film (Hyperfilm-ECL, Amersham, UK).
HLA Class I
COMPLEMENT OP iim
DO DR ii BiI
& CYP21GENES
......
TNF A8 ==
wu
BAT1 m m
I
I
0
50
TNF
B
::
C
A
1~1
V
W
X
Y
n u
n u
n u
. u
I
I
100
150
HLA B m m
RESULTS
I
J
200
250kb
FIGURE 1 The approximate locations of the genomic probes Y, X, W, and V [5] are indicated by open boxes. The solid boxes refer to some of the known genes of the region.
membranes by treatment with 0.4M NaOH. Stripping was carefully checked by exposure to fresh films and the membranes were then used for reprobing [15, 16].
Sequence-specific oligonucleotide (SSO) typing for DRB I alleles. DRB1 alleles were determined by SSO typing based on 11th International Histocompatibility Workshop protocols, with modifications. Briefly, horseradish-peroxidase-labeled SSOs (HRP-SSOs) were synthesized by incorporating a C6-thiol modified phosphoramidite (RPN 2112, Amersham, UK) as the last nucleotide and subsequent coupling of HRP via the thiol group. The synthesized HRP-SSOs were found to be stable and have been used over a period of 4 months without noticeable loss of activity. DRB1 typing of polymerase chain reaction amplified genomic DNA bound to nylon membranes could be accomplished using a pro-
FIGURE 2 Three different fragments (see Table 2) may be seen after Taq I digestion and probing for X. Each homozygous cell line yields a single fragment and it is possible to assign a fragment to each haplotype. The heterozygous cell 60.3/57.1 has two fragments (S and L) in keeping with the fact that homozygous 57.1 and 60.3 have S and L fragments, respectively. The results are tabulated in Table 3.
kb
5.0
4.0
3.0
P a t t e m L L L L L L L M M M L L L L L L S S L L L L L L L L L L
Region Between HLA B and T N F Loci is H i gh ly Pol ym orphi c and the Y-X-V Haplotypes segregate with H L A Haplotypes Digestion ofgenomic DNA with Taq I or Taq I + Rsa I and probing for X, V, and Y revealed extensive polymorphisms, and a number of RFLP patterns (allelic patterns) were assigned to the region between the HLA B and TNF. Among 37 homozygous cell lines three allelic patterns for probe X, four for probe V, and seven for probe Y were observed (Figs. 2-4). As shown in Table 2, a single enzyme, Taq I, could detect all the patterns for X and V as well as most patterns for Y. The addition of Rsa I for Y improved the assignment of the patterns of heterozygotes (Fig. 4). The family inheritance of the Y, X, and V patterns could be clearly demonstrated. As shown in Fig. 5, the Y, X, and V allelic patterns were inherited within the family and the Y-X-V haplotypes segregated with HLA haplotypes. It is important to note that at least three AHs could be defined in the family, and the segregation of Y-X-V haplotypes with these AHs was in agreement with that seen in unrelated homozygous individuals with the same AHs (Table 3). Similar results were also obtained from another informative family. A striking finding illustrated by Figs. 2 - 4 was that a particular AH represented by multiple unrelated homozygous individuals often gave identical Y, X, and V allelic patterns. Only two exceptions were observed (AHs 7.2 and 35.1), and possible explanations are given in Table 4. To date, no polymorphism has been observed for probe W (R5A) on Taq I digests. However, with Bst EII, polymorphism of this genomic region has been reported recently [ 19]. Homozygous Cells for AHs Provide Allelic Pattern Predictions for Heterozygous Cells By examining homozygous cell lines we were able to assign particular allelic patterns to each AH. Cells that were heterozygous for AHs gave the Y patterns predicted from the homozygous cells. As shown in Fig. 4, cells that were homozygous for the AHs 8.1, 57.1, and 60.3 gave patterns B, E, and A, respectively. In the case of heterozygotes with these AHs it was apparent that the patterns B + E and A + E could be assigned to cells carrying both AHs 8.1 + 57.1 and 60.3 +57.1, respec-
92
X. Wu et al.
AH kb
I
D
3.0
tively. This was also the case for X and V p a t t e r n s (Figs. 2 and 3). B a s e d o n family studies and allelic p a t t e r n s o b t a i n e d f r o m h o m o z y g o t e s , it was p o s s i b l e to d e r i v e g e n o t y p e s for b o t h h o m o z y g o u s and h e t e r o z y g o u s individuals (Table 3). Y-X-V Haplotypes Are Haplotypic and Often Haplospecific T o date, at least 15 d i f f e r e n t Y - X - V p a t t e r n c o m b i n a tions ( Y - X - V h a p l o t y p e s ) have b e e n identified within a
1.6
O
F I G U R E 5 Y, X, and V allelic patterns are inherited within family D A R members and segregate with HLA haplotypes. HLA haplotypes (as shown at top of the probe V) for the family are as follows: a = HLA A2, Cw3, B62, BfS, C4A3, C4B3, DR4 (AH 62.1); b = HLA A3, Cw-, B7, BfS, C4A3, C4B1, DR2 (AH 7.1); c = HLA A2, Cw3, B62, B f S C4A3, C4B1, D R X (where X is probably DR4); d = HLA A29, Cw-, B44, Bf'F, C4A3, C4B1, DR7 (AH 44.2). Genomic D N A of the family individuals was digested with Taq I (probes V and X) or Taq I + Rsa I (probe Y). The resulting patterns were obtained for each probing as described in Table 2 and the Y-X-V haplotypes were assigned as: a = A, L, B; b = A, L, A; c = A, L, B; d = D, L, C. The Y-X-V haplotypes segregate with HLA haplotypes and the genotypes are in agreement with those derived from unrelated homozygous subjects carrying the same AHs (Table 3). F = father, M = mother, S = son, and D = daughter.
1.0 probe V
F
M
S
D
ab
CO ac
ad
F M S D kb 3.1
~
probe
,
kb
20
Pattern
A A B B B A B B C C C 1.8
F I G U R E 3 After Taq l digestion and probing for V six, fragments were seen but two appear to be nonpolymorphic. Each homozygous cell line carries two of the four polymorphic fragments. The results are tabulated in Table 3.
2.2 1.6 1.9
F I G U R E 4 Double digestion with Taq I - Rsa I and probing for Y results in seven patterns (see Table 2). Each homozygous cell line is associated with two fragments and a single pattern. For example, 8.1 carries the B pattern, 57.1 the E pattern, and the two heterozygous cells carry both the B and E patterns. The exception is 35.1 since one cell is homozygous for C and the other is homozygous for G (see also Table 4).
G ~
patlern
A
AD
A
AD
F M S D probe
X
i i !i¸i¸iiii !ii ii i:ii!i iii¸¸!i 5.2
kb
%%LII
2.0
1.00 O~i
EE
E
OO
o
B
IIIIIIIi ii ii i i i
33
1.6
Pattern A A B B B B E A A A A A A A A A C C C D D E B G F F
45
I~L:IIT!/ i!~ ~ ii~
1.05
18
iii
~
~
iJ~iii~~i~ 0.88
~7 ~
A
pattern
AB BC
B
BC
pattern
L
L
L
L
Genomic Conservation and Ancestral Haplotypes
TABLE 2
Probe
93
Designation of allelic patterns detected by genomic probes X, V, and Y Enzyme
Pattern 3.3
X (R17A)
V (R9A)
Taq I
Taq I
Taq I + Rsa I
4.5
5.2
M
-
+
-
S
+
-
-
0.88
1.00
1.05
1.90
2.2
3.1
A
+
+
--
+
--
+
B
+
-
+
+
+
-
C
+
+
-
+
+
-
D
Y (M20A)
(kb)
Fragments
+
-
+
+
-
+
1.69
1.73
1.77
1.80
1.85
1.88
A
+
.
B
-
+
-
-
-
+
C
+
-
-
+
-
-
D
-
-
+
-
-
+
E
-
-
+
-
+
-
F
+
+
.
G
+
-
-
+
-
.
.
.
+
.
.
.
-
Allelic patterns (RFLP patterns) were designated to various combinations o f fragments detected by probes X, V, and Y. H u m a n g e n o m i c D N A samples w e r e digested with Taq I for X and V probing and with Taq l + Rsa 1 for Y probing. As shown, three different patterns (L, M, and S) w e r e o b s e r v e d with probe X, four distinct patterns (designated A, B, C, and D) with probe V, and seven patterns with probe Y (designated A - G ) . T h e Y patterns for Taq I are not shown as the addition o f R s a ] improves separation o f f r a g m e n t s and therefore the assignment o f patterns o f heterozygous cells. Abbreviations: + = f r a g m e n t present; - = f r a g m e n t absent.
panel o f 50 cell lines representing 21 different ancestral haplotypes. For most AHs all examples of that A H have the same Y-X-V haplotypes, for example, AHs 7.1, 8.1, 18.2, 46.1, 46.2, 52.1, 62.1, and 65.1 (Table 3, Fig. 6), and therefore the Y-X-V is haplotypic. It is important to note that the Y-X-V haplotypes can also be haplospecific, that is, a particular Y-X-V haplotype is carried by a specific A H but not by any of the other AHs tested. Fig. 6 shows 11 Y-X-V haplotypes found to be specific for particular AHs, for example, ASB is only found in A H 18.1 while CMA is exclusive to A H 18.2, etc. Thus, the HLA B to T N F segment o f the A H 18.1 can be distinguished from that of the 18.2.
FIGURE 6 The composite Y, X, and V pattern carried by each haplotype is indicated by a point. Thus, for example, BLB is carried by all 10 haplotypes bearing 8.1 but by no other haplotypes. Similarly, ASB and CMA are entirely haplotypic (two of two examples for 18.1 and, six of six examples for 18.2, respectively) and also entirely haplospecific (carried by no other haplotype tested). However, the two cells that have been designated 35. I differ, although each is homozygous (see Table 4). The four patterns that do not appear to be haplospecific (ALA, ALB, ALC, DLC) are not shown here (see Table 5). ¥-x-v Qo
ASB BLB
Y, X, and V Allelic P a t t e r n s M a y Reveal D i f f e r e n c e s B e t w e e n A p p a r e n t l y Similar A H s
8LC
As shown in Table 4, two cells that were homozygous for the 35.1 A H were shown to differ at Y and V. Interestingly, SSO typing demonstrated different splits of DR5 carried by these two cell lines. Thus, although both cells were homozygous, it appears that the 35.1 A H may in fact include two different AHs that are serologically identical but distinguishable at Y-X-V and DRB 1. The cell line R6/12286 was thought to be homozygous for the 7.2 A H defined serologically. However, the V probe pattern o f this cell line showed the C pat-
CLA
BLD oo
CMA
oeo coo
DLB EMC ESC Ft~
gee coo
000 000
GLB
8.1
18.1
18.2
35.1
42.1
Ancestral
44.3
47.1
H aplotypes
54.1
57.1
65.1
TABLE 3
Allelic patterns defined by Y, X, and V are haplotypic and often haplospecific and the genotypes of cell lines can be predicted based on homozygous cell lines Derived genotype
Local I D
10WS I D
06/3975 R6/12367 R6/12368
9082 9083
7.1 7.1 7.1
7.1 7.1 7.1
A A A
A A A
L L L
L L L
A A A
A A A
R7/4708 R7/12583 R6/12286"
9001
7.2 7.2 7.2
7,2 7.2 7.2
A A A
A A A
L L L
L L L
A A A
A A C
R5/1518 R5/843 R6/12307 R6/12308
9086 9022 9023
8.1 8.1 8.1 8.1
B B B B
B 8 B B
L L L L
L L
[3 B B B
B E] B B
E
B
k k
C
B
E
E
S
C
C
A H assigned
Q6/8187 R7/17217 R5/9759 R6/12337
9052
i57,1
8.1 8.1 8.1 8.1 8.1 8.1 8.1 57.1
R6/12333
9048
13.1
13.1
57.11 I
Y
X
V
1 3 . 1 146.1 -
R9/52517 o R6/12293
9008
18.1
181
A
A
S
S
B
B
R5/5054 R6/12303 R6/12370
9018 9085
182 18.2 18.2
18.2 18.2 182
C C C
C C C
M M M
M M M
A A A
A A A
Q5/8086 R6/12327
9042
35.1 35.1
351 35.1
C G
C G
L L
L L
A B
A B
R6/12306
9021
42.1
421
B
B
L
L
D
D
44.1 44.1
44 11 71 44 ×
A
A
C
C
D
0
R0/7379F Q5/2711 R6/12312
9027
144.11
R7/12504
442
R6/12335
9050
443
R6/12361"
9076
46.1 46 1 46.1 46 1 46 1 46 1 46.1 ~
R9/52519" R9/52523" R9/52518" R6/15375"
44 3
54.1
54 1
L
L
B
B
L L
L L
C C
C C
L
M
C
C
A A
L L
L L
C C
C C
R7/4714 R6/12351
9066
46.2 46.2
46,2 46,2
R6/12332
9047
47.1
47.1
B
B
L
L
C
C
52.1 52.1 52.1
52.1 52.1 521
A A A
A A A
L L L
L L L
A A A
A A A
R7/4709 R7/12579" R7/12580" R6/12382* R8/5618 R5/13141
9097
R6/12316 R6/12317
9031 9032
60.3
603
160.31
44 2
~
571 62.1 62.1
A A
62.1 621
R5/10168 Q5/7952
16211
71
~
44 2
R0/26468
622
62.2
A
A
L
L
B
B
65.1 65.1 65.1
65.1 65.1 65.1
F F F
F F F
L L L
L L L
A A A
A A A
R6/12364" R6/2553" R6/12363"
9079 9078
Fifty human cell lines examined are shown. AHs were asstgned to each of the cell lines based on serological data for HLA class l, class II, and complement allotyping (Table 1). For the individuals marked by asterisks, the C4 and Cyp21 were determined by Taq I RFLP [12, 17, 18]. Boxes are used to show derived genotypes for heterozygous cells. Abbreviations: 10WS ID = 10th International Histocompatibility Testing Workshop identity number; X - uncertain ancestral haplotype.
Genomic Conservation and Ancestral Haplotypes
T A B LE 4
95
Y, X, and V patterns may reveal differences between apparently similar haplotypes HLA and complement typing
Local [D
R6/12327 R7/4708 R7/12583 R6/12286
10WS [D
A
9042
24
9001
24 24 24
C
B
Bf
C4A
4
35
S
7 7 7
7 7 7
S S NT
Derived genotype
C4B
DR
DQw
AH assigned
3
1
11
7
35 1
351
3+3 3+3 NT
1 1 NT
1 1 1
1 1 1
72 72 (72)
72 72 {72)
Y
X
V
DRBI
L L A A A A A A
L L L L L L
A~ A~C A
NT 0101 0101
Although the first two cells appeared to be similar in terms of HLA and complement typing the Y, X, and V patterns were different. Each of the two ceils appears to be homozygous, suggesting that it may be possible to split 35. l A H into two. The last three cells appear to be similar, although complement typing was not available for the third cell. The Y, X, and V patterns suggest that the third cell is in fact heterozygous at V. Abbreviations: N T = not tested; ( ) = putative assignment.
tern in addition to the A pattern found in the other two cells that were homozygous for the 7.2 AH. Furthermore, Taq I RFLP typing of the C4 loci of this cell line showed an extra 5.4-kb fragment besides the 7.0- and 6.0-kb fragments seen in the other two homozygous cells for the 7.2 AH [18]. It is likely, therefore, that this cell line is not homozygous for the 7.2 AH.
region. For AHs sharing ALC the similarities can be seen from HLA A to C4. In contrast, the AHs in the two lower panels seem to share similarities in HLA class II and complement regions in addition to the Y-X-V region.
AHs Sharing the Same Y-X-V Haplotypes Also Show Extensive Similarities Elsewhere Within the M H C
In this study we have shown that there are extensive polymorphisms for the sequences between the HLA B and TNF by using three genomic probes Y, X, and V [5, 6]. The polymorphisms of Y, X, and V were shown to be haplospecific, that is, specific to a particular AH (Fig. 6). The heterogeneity of the region should provide very useful markers for assigning AHs, for example, the 18.2 AH can be distinguished from 18.1 AH although
As shown in Table 5, the AHs sharing Y-X-V haplotypes have extensive similarities in other regions (boxed) within the MHC. In the first two panels the patterns AL A and ALC are shared by different AHs. AHs sharing A L A are also similar from Bf to the class I TABLE 5
AH
DISCUSSION
Extensive similarities among the AHs sharing Y-X-V haplotypes
Race
A
Cw
B
Y
X
V
52.1
M
2,,
7.1
C
7.2
M
3 2,
17,A
46.1
M
2
11 46 A
L
C
C
S
46.2
M
2
11 46 A
L
C
C
S
? s21A < A
--
C2 Bf C4A C4B
o sl3,2o
!
DQB1
n
,502 060,
o
,
, o,o,o2 0,0,050,
8 5
4
2
0901 0303
8
4
2
0803b 0601
4
0
0,,01 0301
os13 OSl3
i
DRB1
,
44.1
C
2
5 ,,41,,
,. c
c
s i3
62.1
C
2
3 62 A
L
B
C
S
3
3
10401 0302
6
62.2
C
?
3 62 A
L
B
B
S
4
2
10401 0302
2
3 601A
L
B
C
S
0
2
1302 0604
4
II
60.3
C
2
13.1
M/C
30
6
13[D
L
CI
C
SI 3
1
07
0201
3
44.2
O
29
4 44 D
L
C
C
F
1
07
0201
3
3
HLA A, Cw, B, and complement alleles were determined by serology. DRB 1 and DQB 1 alleles were based on SSO typing. The boxed areas encompass regions of similarity between AHs. Abbreviations: n = number of haplotypes tested; Race: C = Caucasoid, M = Mongoloid, ? = not defined.
96
they both appear to carry the same HLA B allele. Furthermore, three HLA B44 bearing AHs can be differentiated as they carry different Y-X-V haplotypes ALC, DLC, and DLB, respectively (Fig. 6, Table 5). The polymorphisms between the HLA B and T N F are haplotypic, providing further evidence for the retention of the sequences in unrelated individuals with the same AH. By employing two probes used here (Y and W) and a different enzyme, Egea et al. [19] also showed polymorphisms that appeared to be haplotypic. Thus, not only has there been conservation of the classes I, II, and complement regions but, in addition, the region between HLA B and T N F also appears to be both polymorphic and conserved, suggesting the possibility of important functions for genes in this region. Although generally haplotypic, there were two exceptions to the rule that the Y, X, and V haplotypes would be predicted by the serological assignment of HLA classes I and II and by complotyping. As shown in Table 4, Y, X, and V patterns may reveal differences between apparently similar haplotypes. Two homozygous cells assigned as A H 35.1 on the basis of available markers were found to differ at Y, V, and DRB1, suggesting that AHs currently assigned may be further split by the Y, X, and V as well as DRB1 alleles. It is conceivable that these two cells represent two different AHs that are nevertheless related to one another in an evolutionary sense. The utility of Y-X-V haplotyping was illustrated further in the second case. A cell that was thought to be homozygous for the 7.2 A H was found to be heterozygous at V and further studies of its C4 genes also revealed heterozygosity. In this instance the original classification of this cell as a homozygous 7.2 was incorrect. There were four Y-X-V haplotypes, ALA, ALC, ALB, and DLC, shared by more than one AH. Interestingly, the AHs sharing the same Y-X-V haplotypes also show extensive similarities in other regions regardless of their ethnic origins. The AHs 46.1 (mainly in Chinese) and 46.2 (mainly in Japanese) [20, 21 ] share most markers, including Y, X, and V, but differ in the class II region, suggesting their relatively recent divergence by recombination between C4 and DR. Surprisingly, two insulin-dependent diabetes mellitus (IDDM)-resistant but race-specific AHs 52.1 (Japanese) and 7.1 (Caucasoid) carry the same Y-X-V haplotype, and two IDDMsusceptible AHs 62.1 and 62.2 (4, 11) also share the same Y-X-V haplotype and other similarities along the MHC. The Y-X-V haplotypes may be very useful for delineating the evolutionary relationship between AHs. The remarkable conservation of the M H C haplotypes suggests a strategy for mapping disease susceptibility genes. As the same AHs in unrelated individuals appear identical at all loci, it should be possible to use
X. Wu et al.
human AHs in the same way that recombinant mouse strains are used to map disease genes. The fact that these polymorphic sequences between HLA B and class II have been conserved en bloc may have implications for transplantation. In the mouse the polymorphic Hh-1 genes that determine hybrid resistance to bone marrow transplantation map to this region [22]. Therefore it may not be sufficient to match donors and recipients on the basis of class I and class II genes. On the other hand matching based on AHs should guarantee matching not only for classes I and II genes but also for Hh-1 like genes and any other polymorphic genes affecting transplantation. Taking the current observations together with those found in the classes I, II, and complement regions, the present study confirms that, in populations, M H C AHs have been conserved en bloc, possibly as a result of extreme polymorphism. The rate of sequencedependent recombination is low where sequence diversity is high [ 2 3 - 2 5 ] . By contrast there may be preferential recombination within homologous (homozygous) regions leading to reformation of entire AHs. These AHs can be recognized by haplospecific allele combinations [2, 4, 26] and have a specific allele at all loci tested. Furthermore, Each A H contains a particular number of genes at all loci tested [16] and has a defined genomic length between HLA B and D Q determined by pulsed field gel electrophoresis [3, 4]. Therefore, the AHs appear to be very similar, if not identical, in unrelated subjects. We propose that entire AHs are favored as a result of functional interactions between M H C alleles involved in the regulation of the immune response. We predict that the newly described polymorphisms between the HLA B and T N F will enrich the usefulness of AHs for different purposes, such as mapping of disease susceptibility genes, haplotype matching for transplantation, and for evolutionary studies of the MHC.
ACKNOWLEDGMENTS
We are grateful to Dr. T. Spies (USA) for providing genomic probes, to Dr. T. Juji (Japan), and to Dr. S. H. Chan (Singapore) for providing some Japanese and Chinese cell lines, respectively, to Dr. G. Grimsley for helpful advice and discussion, and to Dr. Y. Tian, H. Tabarias, T. Causerano, and R. Darovic for technical assistance. This work was supported by the National Health and Medical Research Council, Western Australian Arthritis and Rheumatism Foundation, and the Immunogenetics Research Foundation.
REFERENCES 1. Dupont B: Immunobiology of HLA. Vol 1. New York, Springer-Verlag, 1989. 2. Dawkins RL, Christiansen FT, Kay PH, Garlepp MJ,
Genomic Conservation and Ancestral Haplotypes
McCluskey J, Hollingsworth PN, Zilko PJ: Disease associations with complotypes, supratypes and haplotypes. Immunol Rev 70:5, 1983. 3. Dawkins RL, Kay PH, Martin E, Christiansen FT: The genomic structure of ancestral haplotypes revealed by pulsed field gel electrophoresis (PFGE). In Dupont B (eds): Immunobiology of HLA, Vol 1. New York, Springer-Verlag, 1989, p. 893. 4. Dawkins RL, Zhang WJ, Degli-Esposti MA, Abraham L, McCann V, Christiansen FT: Studies of haplotypes by pulsed field gel electrophoresis. In Bailliere Clinical Endocrinology and Metabolism, Vol 5, No 2. London, Bailliere Tindall Limited, 1991, p. 285. 5. Spies T, Blanck G, Bresnahan M, Sands J, StromingerJL: A new cluster of genes within the human major histocompatibility complex. Science 243:214, 1989. 6. Spies T, Bresnahan M, Strominger JL: Human major histocompatibility complex contains a minimum of 19 genes between the complement cluster and HLA-B. Proc Natl Acad Sci USA 86:8955, 1989. 7. Sargent CA, Dunham I, Campbell RD: Identification of multiple HTF-island associated genes in the human major histocompatibility complex class III region. Eur Mol Biol Organ J 8:2305, 1989. 8. Dawkins RL, Leaver A, Cameron PU, Martin E, Kay PH, Christiansen FT. Some disease-associated ancestral haplotypes carry a polymorphism of TNF. Hum Immunol 26:91, 1989. 9. Dawkins RL, Uko G, Pestell R, McCann VJ: Association of MHC supratypes and complotypes with endocrine disorders. In Farid NR (eds): Immunogenetics of Endocrine Disorders. New York, Alan R Liss, 1988, p. 433. 10. Wilton AN, Cobain TJ, Dawkins, RL: Family studies of IgA deficiency. Immunogenetics 21:333, 1985. 11. Kelly H, McCann VJ, Kay PH, Dawkins RL: Susceptibility to IDDM is marked by MHC supratypes rather than individual alleles. Immunogenetics 22:643, 1985.
97
in different MHC supratypes detected by pulsed field gel electrophoresis. J Exp Med 168:933, 1988. 16. Zhang WJ, Degli-Esposti MA, Cobain TJ, Cameron PU, Christiansen FT, Dawkins RL: Differences in gene copy number carried by different MHC ancestral haplotypes: quantitation after physical separation of haplotypes by pulsed field gel electrophoresis. J Exp Med 171:2101, 1990. 17. Schneider PM, Carroll MC, Alper CA, Ritmer C, Whitehead AS, Colten HR: 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, 1986. 18. Tokunaga K, Zhang WJ, Christiansen FT, Dawkins RL: The genomic structure of two ancestral haplotypes carrying C4 duplications. Immunogenetics 34:247, 1991. 19. Egea GE, Yunis I, Spies T, Strominger J, Awdeh ZL, Alper CA, Yunis EJ: Association ofpolymorphisms in the HLA-B region with extended haplotypes. Immunogenetics 33:4, 1991. 20. Tokunaga K, Omoto K, Akaza T, Akiyama N, Amemiya H, Naito HS, Sasazuki T, Satoh H, Juji T: Haplotype study on C4 polymorphism in Japanese. Association with MHC-alleles, complotypes and HLA-complement haplotypes. Immunogenetics 22:359, 1985 21. Chan SH: Antigen Report: Bw46. In Aizawa M (ed.): HLA in Asia-Oceania. Sapporo, Japan, Hokkaido University Press, 1986, p. 106. 22. Rembecki RM, Kumar V, David CS, Bennett M: Hemopoietic histocompatibility (Hh-1) regulatory and structure genes of the f haplotype map to H-2. Transplantations 49:633, 1990. 23. Walsh JB: Sequence-dependent gene conversion: Can duplicated genes diverge fast enough to escape conversion? Genetics 117:543, 1987. 24. Koop BF, Siemieniak D, Slightom JL, Goodman M, Dunbar J, Wright PC, Simons EL: Tarsius delta- and betaglobulin genes: Conversions, evolution, and systematic implications. J Biol Chem 264:68, 1989.
12. Garlepp MJ, Wilton AN, Dawkins RL, White PC: Rearrangement of 21-hydroxylase genes in disease associated MHC supratypes. Immunogenetics 23:100, 1986. 13. Feinberg AP, Vogelstein B: A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6, 1983.
25. Hughes AL: Testing for interlocus genetic exchange in the MHC: a reply to Andersson and co-workers. Immunogenetics 33:243, 1991.
14. Reed KC, Mann DA: Rapid transfer of DNA from agarose gels to Nylon membranes. Nucleic Acid Res 13:7207, 1985. 15. Tokunaga K, Saueracker G, Kay PH, Christiansen FT, Anand R, Dawkins RL: Extensive deletions and insertions
26. Kay PH, Dawkins RL, Williamson J, Tokunaga K, Christiansen FT, Charoenwong P: Coexistence of an MHC chromosomal segment marked by HLA B 17, BfS, C4A6, B1, DR7, and DQw9 in different ethnic groups. Hum Immunol 23:27, 1988.
ABSTRACT: Polymorphisms were sought between HLA B and tumor necrosis factor (TNF) using three genomic probes. Extensive polymorphism was detected within a panel of 50 cell lines including 37 homozygotes representing 21 different ancestral haplotypes (AH). Following Taq I digestion of genomic DNA, we observed three allelic patterns with probe X (RI7A) and four with probe V (RgA). Seven different allelic patterns were found with probe Y (M20A) after Taq I + Rsa I digestion. Family studies showed that the Y, X, and V alleles were inherited and segregated with HLA haplotypes. A striking feature of the allelic patterns detected by these probes was that cells with the same AH had identical Y, X, and V alleles (i.e., the alleles were haplotypic). Of 15 different Y-X-V haplotypes observed, 11 were found to be specific for a particular AH (i.e., were haplospecific).
ABBREVIATIONS AH ancestral haplotype MHC major histocompatibility complex RFLP restriction fragment length polymorphism
Four were shared by more than one AH, but in these instances there were extensive similarities in other regions within the major histocompatibility complex (MHC), for example, the Japanese 46.2 (HLA Bw46DRw8) and the Chinese 46.1 (Bw46-DR9) share all alleles between HLA C and C4 and differ only in class II, suggesting their relatively recent divergence by recombination between C4 and DR. Surprisingly, two insulindependent diabetes mellitus (IDDM)-resistant but racespecific AHs 52.1 (Bw52-DRB 1" 1502, Japanese) and 7.1 (B7-DRB 1"1501, Caucasoid) carry the same Y-X-V haplotype, suggesting the possibility of localizing gene(s) relevant to IDDM. The present study confirms that MHC AHs have been conserved en bloc, including the region between HLA B and TNF. Human Immunology 33, 8997, (1992)
SSO TNF
sequence-specific oligonucleotide tumor necrosis factor
INTRODUCTION The m a j o r histocompatibility complex (MHC) is characterized by polymorphism. All polymorphisms, whether within coding or noncoding sequences at M H C classes I, II, and c o m p l e m e n t loci are haplotypic and often haPublication number 9053 of the Department of Clinical Immunology. Royal Perth Hospital, Sir Charles Gairdner Hospital and the University of Western Australia, Perth, Western Australia. From the Departments of Clinical Immunology, Royal Perth Hospital Sir Charles Gairdner Hospital, and the University of Western Australia, Perth, Western Australia. AddreJs reprint requests to R. L. Dawkins, Department of Clinical Immunology, Royal Perth Hospital, GPO Box X2213, Perth, Western Australia 6001. Received August 26, 1991; accepted October 30, 199•. Human Immunology 33, 89-97 (1992) © American Society for Histocompatibility and Immunogenetics, 1992
plospecific, as shown by the collaborative studies on a cell panel o f well-defined M H C haplotypes within the 10th International Histocompatibility Workshop [1]. The fact that polymorphisms are invariably haplotypic indicates that M H C haplotypes have been conserved en bloc. Thus, apparently unrelated individuals often share the same M H C haplotypes. Accordingly, we use the term ancestral haplotype (AH) to refer to conserved M H C haplotypes that have been maintained through many generations of an ancestral family [ 2 - 4 ] . During recent years a n u m b e r o f novel genes have been identified and m a p p e d to the region between H L A B and C2 [ 5 - 7 ] . W e predicted that this region would 89
0198-8859/92/$5.00
90
X. Wu et al.
contain polymorphic sequences that would provide haplotypic and/or haplospecific markers for different AHs. This was shown to be the case for tumor necrosis factor (TNF) [8] and BAT3 [4]. In this report we have characterized the 250-kb region between the HLA B and TNF by using four genomic probes Y (M20A), X (R17A), V (R9A), and W (RSA) [5, 6] in conjunction with a panel of wellcharacterized cell lines. MATERIALS A N D M E T H O D S
Cell lines. Fifty Epstein-Barr virus-transformed human cell lines, including 24 studied during the 10th International Histocompatibility Workshop, were selected. The local cell lines were genotyped for HLA classes I, II, and complement loci by serology and/or restriction fragment length polymorphism (RFLP) typing. The cell panel covered 21 different AHs (Table 1). AHs have been named according to the HLA B allele characterizing that AH [4]. Genomic DNA was extracted from the cell lines according to the standard phenol/chloroform/ isoamyl alcohol method. Disease associations with some of these AHs have been reported previously [2, 9-12].
D N A probes. Genomic DNA probes Y, X, V, and W were approximately 200-300 bp BamH I + EcoR I fragments derived from the clones M20A, R17A, R9A, and R5A, respectively (kindly provided by Dr. T. Spies,
TABLE 1
Boston MA). The locations of the probes are centromeric to the HLA B, approximately 36 kb for probe Y, 73 kb for probe X, 100 kb for probe W, and 128 kb for probe V [5] (Fig. 1). The probes were labeled with c~[~2p]-dCTP by random priming [13].
Electrophoresis and Southern blot analysis. Genomic DNA (10 ttg) was digested to completion with the Taq I (Promega, Madison, WI) according to the manufacturer's instruction. In some blots double digestions of Taq I + Rsa I (Promega) were used since additional fragments and improved patterns were obtained with the Y probe. The digests were electrophoresed on 1.1% agarose gels at 25 V for 62 hours (or on 0.8% agarose gels at 40 V for 18 hours) in 1 x TBE buffer at room temperature (RT) and transferred to nylon membranes (GeneScreenPlus, Du Pont, USA) by the method of Reed and Mann [14], using 0.4M NaOH as a transfer medium. Prehybridization and hybridization of the membranes were performed in hybridization bottles (Hybaid, Middlesex, UK) at 42°C. The membranes were washed twice with 2 x SSPE for 10 minutes at RT, once in 2 × SSPE + 0.5% SDS at 65°C for 10 minutes, and once in 0.5 x SSPE at 65°C for 5-10 minutes. Kodak X-OMAT films (Rochester, NY) were exposed to the membranes at -70°C for 1-10 days according to standard methods. The sizes of the restriction fragments were determined by reference to a 1-kb DNA ladder (BRL, Gaithersburg, MD), Probes were stripped off the
Ancestral haplotypes used in this study
AH
HLA A
Cw
B
C2
Bf
C4A
C4B
DR
DQw
Race
n
7.1 7.2 8.1 13.1 18.1 18.2 35.1 42.1 44.1 44.2 44.3 46.1 46.2 47.1 52.1 54.1 57.1 60.3 62.1 62.2 65.1
3 24 1 30 25 30 ? ? 2 29 29 2 2 3 24 ~ 1 2 2 ~ >
7 7 7 6 ~ 5 4 2 5 4 ? 11 11 6 ~ 1 6 3 3 3 8
7 7 8 13 18 18 35 42 44 44 44 46 46 47 52 54 57 60 62 62 65
C C C C QO C ? c C c ~ C C C C C C C C B C
S S S S S F1 S F S F S S S F S S S S S S S
3 3 + 3 Qo 3 4 3 3 1 3 3 Qo 4 4 91 3 + 2 3 6 QO 3 4 2
1 1 1 I 2 Qo 1 Qo Qo 1 1 2 2 Q0 QO 5 1 2 3 2 1 + 2
15 1 3 7 15 3 11 3 4 7 7 9 8 7 15 4 7 13 4 4 1
6 5 2 2 6 2 7 4 7 2 2 9 6 2 6 4 9 6 8 8 5
C M c C/M c c C N c C M M M C M M C/M/N C C C C
8 6 11 3 2 6 4 2 5 3 2 8 4 2 6 3 6 4 6 2 6
Abbreviations: n - number of haplotypes tested; ? = not defined; Race: C ~ Caucasoid, M = Mongoloid, N = Negroid.
Genomic Conservation and Ancestral Haplotypes
HLA Class II
Central non-HLA
91
tocol similar to the 32p protocol. Signal detection was achieved by dipping membranes into a chemiluminescent substrate for H RP (RPN 2105, Amersham, UK) and subsequent exposure of the membrane to a photographic film (Hyperfilm-ECL, Amersham, UK).
HLA Class I
COMPLEMENT OP iim
DO DR ii BiI
& CYP21GENES
......
TNF A8 ==
wu
BAT1 m m
I
I
0
50
TNF
B
::
C
A
1~1
V
W
X
Y
n u
n u
n u
. u
I
I
100
150
HLA B m m
RESULTS
I
J
200
250kb
FIGURE 1 The approximate locations of the genomic probes Y, X, W, and V [5] are indicated by open boxes. The solid boxes refer to some of the known genes of the region.
membranes by treatment with 0.4M NaOH. Stripping was carefully checked by exposure to fresh films and the membranes were then used for reprobing [15, 16].
Sequence-specific oligonucleotide (SSO) typing for DRB I alleles. DRB1 alleles were determined by SSO typing based on 11th International Histocompatibility Workshop protocols, with modifications. Briefly, horseradish-peroxidase-labeled SSOs (HRP-SSOs) were synthesized by incorporating a C6-thiol modified phosphoramidite (RPN 2112, Amersham, UK) as the last nucleotide and subsequent coupling of HRP via the thiol group. The synthesized HRP-SSOs were found to be stable and have been used over a period of 4 months without noticeable loss of activity. DRB1 typing of polymerase chain reaction amplified genomic DNA bound to nylon membranes could be accomplished using a pro-
FIGURE 2 Three different fragments (see Table 2) may be seen after Taq I digestion and probing for X. Each homozygous cell line yields a single fragment and it is possible to assign a fragment to each haplotype. The heterozygous cell 60.3/57.1 has two fragments (S and L) in keeping with the fact that homozygous 57.1 and 60.3 have S and L fragments, respectively. The results are tabulated in Table 3.
kb
5.0
4.0
3.0
P a t t e m L L L L L L L M M M L L L L L L S S L L L L L L L L L L
Region Between HLA B and T N F Loci is H i gh ly Pol ym orphi c and the Y-X-V Haplotypes segregate with H L A Haplotypes Digestion ofgenomic DNA with Taq I or Taq I + Rsa I and probing for X, V, and Y revealed extensive polymorphisms, and a number of RFLP patterns (allelic patterns) were assigned to the region between the HLA B and TNF. Among 37 homozygous cell lines three allelic patterns for probe X, four for probe V, and seven for probe Y were observed (Figs. 2-4). As shown in Table 2, a single enzyme, Taq I, could detect all the patterns for X and V as well as most patterns for Y. The addition of Rsa I for Y improved the assignment of the patterns of heterozygotes (Fig. 4). The family inheritance of the Y, X, and V patterns could be clearly demonstrated. As shown in Fig. 5, the Y, X, and V allelic patterns were inherited within the family and the Y-X-V haplotypes segregated with HLA haplotypes. It is important to note that at least three AHs could be defined in the family, and the segregation of Y-X-V haplotypes with these AHs was in agreement with that seen in unrelated homozygous individuals with the same AHs (Table 3). Similar results were also obtained from another informative family. A striking finding illustrated by Figs. 2 - 4 was that a particular AH represented by multiple unrelated homozygous individuals often gave identical Y, X, and V allelic patterns. Only two exceptions were observed (AHs 7.2 and 35.1), and possible explanations are given in Table 4. To date, no polymorphism has been observed for probe W (R5A) on Taq I digests. However, with Bst EII, polymorphism of this genomic region has been reported recently [ 19]. Homozygous Cells for AHs Provide Allelic Pattern Predictions for Heterozygous Cells By examining homozygous cell lines we were able to assign particular allelic patterns to each AH. Cells that were heterozygous for AHs gave the Y patterns predicted from the homozygous cells. As shown in Fig. 4, cells that were homozygous for the AHs 8.1, 57.1, and 60.3 gave patterns B, E, and A, respectively. In the case of heterozygotes with these AHs it was apparent that the patterns B + E and A + E could be assigned to cells carrying both AHs 8.1 + 57.1 and 60.3 +57.1, respec-
92
X. Wu et al.
AH kb
I
D
3.0
tively. This was also the case for X and V p a t t e r n s (Figs. 2 and 3). B a s e d o n family studies and allelic p a t t e r n s o b t a i n e d f r o m h o m o z y g o t e s , it was p o s s i b l e to d e r i v e g e n o t y p e s for b o t h h o m o z y g o u s and h e t e r o z y g o u s individuals (Table 3). Y-X-V Haplotypes Are Haplotypic and Often Haplospecific T o date, at least 15 d i f f e r e n t Y - X - V p a t t e r n c o m b i n a tions ( Y - X - V h a p l o t y p e s ) have b e e n identified within a
1.6
O
F I G U R E 5 Y, X, and V allelic patterns are inherited within family D A R members and segregate with HLA haplotypes. HLA haplotypes (as shown at top of the probe V) for the family are as follows: a = HLA A2, Cw3, B62, BfS, C4A3, C4B3, DR4 (AH 62.1); b = HLA A3, Cw-, B7, BfS, C4A3, C4B1, DR2 (AH 7.1); c = HLA A2, Cw3, B62, B f S C4A3, C4B1, D R X (where X is probably DR4); d = HLA A29, Cw-, B44, Bf'F, C4A3, C4B1, DR7 (AH 44.2). Genomic D N A of the family individuals was digested with Taq I (probes V and X) or Taq I + Rsa I (probe Y). The resulting patterns were obtained for each probing as described in Table 2 and the Y-X-V haplotypes were assigned as: a = A, L, B; b = A, L, A; c = A, L, B; d = D, L, C. The Y-X-V haplotypes segregate with HLA haplotypes and the genotypes are in agreement with those derived from unrelated homozygous subjects carrying the same AHs (Table 3). F = father, M = mother, S = son, and D = daughter.
1.0 probe V
F
M
S
D
ab
CO ac
ad
F M S D kb 3.1
~
probe
,
kb
20
Pattern
A A B B B A B B C C C 1.8
F I G U R E 3 After Taq l digestion and probing for V six, fragments were seen but two appear to be nonpolymorphic. Each homozygous cell line carries two of the four polymorphic fragments. The results are tabulated in Table 3.
2.2 1.6 1.9
F I G U R E 4 Double digestion with Taq I - Rsa I and probing for Y results in seven patterns (see Table 2). Each homozygous cell line is associated with two fragments and a single pattern. For example, 8.1 carries the B pattern, 57.1 the E pattern, and the two heterozygous cells carry both the B and E patterns. The exception is 35.1 since one cell is homozygous for C and the other is homozygous for G (see also Table 4).
G ~
patlern
A
AD
A
AD
F M S D probe
X
i i !i¸i¸iiii !ii ii i:ii!i iii¸¸!i 5.2
kb
%%LII
2.0
1.00 O~i
EE
E
OO
o
B
IIIIIIIi ii ii i i i
33
1.6
Pattern A A B B B B E A A A A A A A A A C C C D D E B G F F
45
I~L:IIT!/ i!~ ~ ii~
1.05
18
iii
~
~
iJ~iii~~i~ 0.88
~7 ~
A
pattern
AB BC
B
BC
pattern
L
L
L
L
Genomic Conservation and Ancestral Haplotypes
TABLE 2
Probe
93
Designation of allelic patterns detected by genomic probes X, V, and Y Enzyme
Pattern 3.3
X (R17A)
V (R9A)
Taq I
Taq I
Taq I + Rsa I
4.5
5.2
M
-
+
-
S
+
-
-
0.88
1.00
1.05
1.90
2.2
3.1
A
+
+
--
+
--
+
B
+
-
+
+
+
-
C
+
+
-
+
+
-
D
Y (M20A)
(kb)
Fragments
+
-
+
+
-
+
1.69
1.73
1.77
1.80
1.85
1.88
A
+
.
B
-
+
-
-
-
+
C
+
-
-
+
-
-
D
-
-
+
-
-
+
E
-
-
+
-
+
-
F
+
+
.
G
+
-
-
+
-
.
.
.
+
.
.
.
-
Allelic patterns (RFLP patterns) were designated to various combinations o f fragments detected by probes X, V, and Y. H u m a n g e n o m i c D N A samples w e r e digested with Taq I for X and V probing and with Taq l + Rsa 1 for Y probing. As shown, three different patterns (L, M, and S) w e r e o b s e r v e d with probe X, four distinct patterns (designated A, B, C, and D) with probe V, and seven patterns with probe Y (designated A - G ) . T h e Y patterns for Taq I are not shown as the addition o f R s a ] improves separation o f f r a g m e n t s and therefore the assignment o f patterns o f heterozygous cells. Abbreviations: + = f r a g m e n t present; - = f r a g m e n t absent.
panel o f 50 cell lines representing 21 different ancestral haplotypes. For most AHs all examples of that A H have the same Y-X-V haplotypes, for example, AHs 7.1, 8.1, 18.2, 46.1, 46.2, 52.1, 62.1, and 65.1 (Table 3, Fig. 6), and therefore the Y-X-V is haplotypic. It is important to note that the Y-X-V haplotypes can also be haplospecific, that is, a particular Y-X-V haplotype is carried by a specific A H but not by any of the other AHs tested. Fig. 6 shows 11 Y-X-V haplotypes found to be specific for particular AHs, for example, ASB is only found in A H 18.1 while CMA is exclusive to A H 18.2, etc. Thus, the HLA B to T N F segment o f the A H 18.1 can be distinguished from that of the 18.2.
FIGURE 6 The composite Y, X, and V pattern carried by each haplotype is indicated by a point. Thus, for example, BLB is carried by all 10 haplotypes bearing 8.1 but by no other haplotypes. Similarly, ASB and CMA are entirely haplotypic (two of two examples for 18.1 and, six of six examples for 18.2, respectively) and also entirely haplospecific (carried by no other haplotype tested). However, the two cells that have been designated 35. I differ, although each is homozygous (see Table 4). The four patterns that do not appear to be haplospecific (ALA, ALB, ALC, DLC) are not shown here (see Table 5). ¥-x-v Qo
ASB BLB
Y, X, and V Allelic P a t t e r n s M a y Reveal D i f f e r e n c e s B e t w e e n A p p a r e n t l y Similar A H s
8LC
As shown in Table 4, two cells that were homozygous for the 35.1 A H were shown to differ at Y and V. Interestingly, SSO typing demonstrated different splits of DR5 carried by these two cell lines. Thus, although both cells were homozygous, it appears that the 35.1 A H may in fact include two different AHs that are serologically identical but distinguishable at Y-X-V and DRB 1. The cell line R6/12286 was thought to be homozygous for the 7.2 A H defined serologically. However, the V probe pattern o f this cell line showed the C pat-
CLA
BLD oo
CMA
oeo coo
DLB EMC ESC Ft~
gee coo
000 000
GLB
8.1
18.1
18.2
35.1
42.1
Ancestral
44.3
47.1
H aplotypes
54.1
57.1
65.1
TABLE 3
Allelic patterns defined by Y, X, and V are haplotypic and often haplospecific and the genotypes of cell lines can be predicted based on homozygous cell lines Derived genotype
Local I D
10WS I D
06/3975 R6/12367 R6/12368
9082 9083
7.1 7.1 7.1
7.1 7.1 7.1
A A A
A A A
L L L
L L L
A A A
A A A
R7/4708 R7/12583 R6/12286"
9001
7.2 7.2 7.2
7,2 7.2 7.2
A A A
A A A
L L L
L L L
A A A
A A C
R5/1518 R5/843 R6/12307 R6/12308
9086 9022 9023
8.1 8.1 8.1 8.1
B B B B
B 8 B B
L L L L
L L
[3 B B B
B E] B B
E
B
k k
C
B
E
E
S
C
C
A H assigned
Q6/8187 R7/17217 R5/9759 R6/12337
9052
i57,1
8.1 8.1 8.1 8.1 8.1 8.1 8.1 57.1
R6/12333
9048
13.1
13.1
57.11 I
Y
X
V
1 3 . 1 146.1 -
R9/52517 o R6/12293
9008
18.1
181
A
A
S
S
B
B
R5/5054 R6/12303 R6/12370
9018 9085
182 18.2 18.2
18.2 18.2 182
C C C
C C C
M M M
M M M
A A A
A A A
Q5/8086 R6/12327
9042
35.1 35.1
351 35.1
C G
C G
L L
L L
A B
A B
R6/12306
9021
42.1
421
B
B
L
L
D
D
44.1 44.1
44 11 71 44 ×
A
A
C
C
D
0
R0/7379F Q5/2711 R6/12312
9027
144.11
R7/12504
442
R6/12335
9050
443
R6/12361"
9076
46.1 46 1 46.1 46 1 46 1 46 1 46.1 ~
R9/52519" R9/52523" R9/52518" R6/15375"
44 3
54.1
54 1
L
L
B
B
L L
L L
C C
C C
L
M
C
C
A A
L L
L L
C C
C C
R7/4714 R6/12351
9066
46.2 46.2
46,2 46,2
R6/12332
9047
47.1
47.1
B
B
L
L
C
C
52.1 52.1 52.1
52.1 52.1 521
A A A
A A A
L L L
L L L
A A A
A A A
R7/4709 R7/12579" R7/12580" R6/12382* R8/5618 R5/13141
9097
R6/12316 R6/12317
9031 9032
60.3
603
160.31
44 2
~
571 62.1 62.1
A A
62.1 621
R5/10168 Q5/7952
16211
71
~
44 2
R0/26468
622
62.2
A
A
L
L
B
B
65.1 65.1 65.1
65.1 65.1 65.1
F F F
F F F
L L L
L L L
A A A
A A A
R6/12364" R6/2553" R6/12363"
9079 9078
Fifty human cell lines examined are shown. AHs were asstgned to each of the cell lines based on serological data for HLA class l, class II, and complement allotyping (Table 1). For the individuals marked by asterisks, the C4 and Cyp21 were determined by Taq I RFLP [12, 17, 18]. Boxes are used to show derived genotypes for heterozygous cells. Abbreviations: 10WS ID = 10th International Histocompatibility Testing Workshop identity number; X - uncertain ancestral haplotype.
Genomic Conservation and Ancestral Haplotypes
T A B LE 4
95
Y, X, and V patterns may reveal differences between apparently similar haplotypes HLA and complement typing
Local [D
R6/12327 R7/4708 R7/12583 R6/12286
10WS [D
A
9042
24
9001
24 24 24
C
B
Bf
C4A
4
35
S
7 7 7
7 7 7
S S NT
Derived genotype
C4B
DR
DQw
AH assigned
3
1
11
7
35 1
351
3+3 3+3 NT
1 1 NT
1 1 1
1 1 1
72 72 (72)
72 72 {72)
Y
X
V
DRBI
L L A A A A A A
L L L L L L
A~ A~C A
NT 0101 0101
Although the first two cells appeared to be similar in terms of HLA and complement typing the Y, X, and V patterns were different. Each of the two ceils appears to be homozygous, suggesting that it may be possible to split 35. l A H into two. The last three cells appear to be similar, although complement typing was not available for the third cell. The Y, X, and V patterns suggest that the third cell is in fact heterozygous at V. Abbreviations: N T = not tested; ( ) = putative assignment.
tern in addition to the A pattern found in the other two cells that were homozygous for the 7.2 AH. Furthermore, Taq I RFLP typing of the C4 loci of this cell line showed an extra 5.4-kb fragment besides the 7.0- and 6.0-kb fragments seen in the other two homozygous cells for the 7.2 AH [18]. It is likely, therefore, that this cell line is not homozygous for the 7.2 AH.
region. For AHs sharing ALC the similarities can be seen from HLA A to C4. In contrast, the AHs in the two lower panels seem to share similarities in HLA class II and complement regions in addition to the Y-X-V region.
AHs Sharing the Same Y-X-V Haplotypes Also Show Extensive Similarities Elsewhere Within the M H C
In this study we have shown that there are extensive polymorphisms for the sequences between the HLA B and TNF by using three genomic probes Y, X, and V [5, 6]. The polymorphisms of Y, X, and V were shown to be haplospecific, that is, specific to a particular AH (Fig. 6). The heterogeneity of the region should provide very useful markers for assigning AHs, for example, the 18.2 AH can be distinguished from 18.1 AH although
As shown in Table 5, the AHs sharing Y-X-V haplotypes have extensive similarities in other regions (boxed) within the MHC. In the first two panels the patterns AL A and ALC are shared by different AHs. AHs sharing A L A are also similar from Bf to the class I TABLE 5
AH
DISCUSSION
Extensive similarities among the AHs sharing Y-X-V haplotypes
Race
A
Cw
B
Y
X
V
52.1
M
2,,
7.1
C
7.2
M
3 2,
17,A
46.1
M
2
11 46 A
L
C
C
S
46.2
M
2
11 46 A
L
C
C
S
? s21A < A
--
C2 Bf C4A C4B
o sl3,2o
!
DQB1
n
,502 060,
o
,
, o,o,o2 0,0,050,
8 5
4
2
0901 0303
8
4
2
0803b 0601
4
0
0,,01 0301
os13 OSl3
i
DRB1
,
44.1
C
2
5 ,,41,,
,. c
c
s i3
62.1
C
2
3 62 A
L
B
C
S
3
3
10401 0302
6
62.2
C
?
3 62 A
L
B
B
S
4
2
10401 0302
2
3 601A
L
B
C
S
0
2
1302 0604
4
II
60.3
C
2
13.1
M/C
30
6
13[D
L
CI
C
SI 3
1
07
0201
3
44.2
O
29
4 44 D
L
C
C
F
1
07
0201
3
3
HLA A, Cw, B, and complement alleles were determined by serology. DRB 1 and DQB 1 alleles were based on SSO typing. The boxed areas encompass regions of similarity between AHs. Abbreviations: n = number of haplotypes tested; Race: C = Caucasoid, M = Mongoloid, ? = not defined.
96
they both appear to carry the same HLA B allele. Furthermore, three HLA B44 bearing AHs can be differentiated as they carry different Y-X-V haplotypes ALC, DLC, and DLB, respectively (Fig. 6, Table 5). The polymorphisms between the HLA B and T N F are haplotypic, providing further evidence for the retention of the sequences in unrelated individuals with the same AH. By employing two probes used here (Y and W) and a different enzyme, Egea et al. [19] also showed polymorphisms that appeared to be haplotypic. Thus, not only has there been conservation of the classes I, II, and complement regions but, in addition, the region between HLA B and T N F also appears to be both polymorphic and conserved, suggesting the possibility of important functions for genes in this region. Although generally haplotypic, there were two exceptions to the rule that the Y, X, and V haplotypes would be predicted by the serological assignment of HLA classes I and II and by complotyping. As shown in Table 4, Y, X, and V patterns may reveal differences between apparently similar haplotypes. Two homozygous cells assigned as A H 35.1 on the basis of available markers were found to differ at Y, V, and DRB1, suggesting that AHs currently assigned may be further split by the Y, X, and V as well as DRB1 alleles. It is conceivable that these two cells represent two different AHs that are nevertheless related to one another in an evolutionary sense. The utility of Y-X-V haplotyping was illustrated further in the second case. A cell that was thought to be homozygous for the 7.2 A H was found to be heterozygous at V and further studies of its C4 genes also revealed heterozygosity. In this instance the original classification of this cell as a homozygous 7.2 was incorrect. There were four Y-X-V haplotypes, ALA, ALC, ALB, and DLC, shared by more than one AH. Interestingly, the AHs sharing the same Y-X-V haplotypes also show extensive similarities in other regions regardless of their ethnic origins. The AHs 46.1 (mainly in Chinese) and 46.2 (mainly in Japanese) [20, 21 ] share most markers, including Y, X, and V, but differ in the class II region, suggesting their relatively recent divergence by recombination between C4 and DR. Surprisingly, two insulin-dependent diabetes mellitus (IDDM)-resistant but race-specific AHs 52.1 (Japanese) and 7.1 (Caucasoid) carry the same Y-X-V haplotype, and two IDDMsusceptible AHs 62.1 and 62.2 (4, 11) also share the same Y-X-V haplotype and other similarities along the MHC. The Y-X-V haplotypes may be very useful for delineating the evolutionary relationship between AHs. The remarkable conservation of the M H C haplotypes suggests a strategy for mapping disease susceptibility genes. As the same AHs in unrelated individuals appear identical at all loci, it should be possible to use
X. Wu et al.
human AHs in the same way that recombinant mouse strains are used to map disease genes. The fact that these polymorphic sequences between HLA B and class II have been conserved en bloc may have implications for transplantation. In the mouse the polymorphic Hh-1 genes that determine hybrid resistance to bone marrow transplantation map to this region [22]. Therefore it may not be sufficient to match donors and recipients on the basis of class I and class II genes. On the other hand matching based on AHs should guarantee matching not only for classes I and II genes but also for Hh-1 like genes and any other polymorphic genes affecting transplantation. Taking the current observations together with those found in the classes I, II, and complement regions, the present study confirms that, in populations, M H C AHs have been conserved en bloc, possibly as a result of extreme polymorphism. The rate of sequencedependent recombination is low where sequence diversity is high [ 2 3 - 2 5 ] . By contrast there may be preferential recombination within homologous (homozygous) regions leading to reformation of entire AHs. These AHs can be recognized by haplospecific allele combinations [2, 4, 26] and have a specific allele at all loci tested. Furthermore, Each A H contains a particular number of genes at all loci tested [16] and has a defined genomic length between HLA B and D Q determined by pulsed field gel electrophoresis [3, 4]. Therefore, the AHs appear to be very similar, if not identical, in unrelated subjects. We propose that entire AHs are favored as a result of functional interactions between M H C alleles involved in the regulation of the immune response. We predict that the newly described polymorphisms between the HLA B and T N F will enrich the usefulness of AHs for different purposes, such as mapping of disease susceptibility genes, haplotype matching for transplantation, and for evolutionary studies of the MHC.
ACKNOWLEDGMENTS
We are grateful to Dr. T. Spies (USA) for providing genomic probes, to Dr. T. Juji (Japan), and to Dr. S. H. Chan (Singapore) for providing some Japanese and Chinese cell lines, respectively, to Dr. G. Grimsley for helpful advice and discussion, and to Dr. Y. Tian, H. Tabarias, T. Causerano, and R. Darovic for technical assistance. This work was supported by the National Health and Medical Research Council, Western Australian Arthritis and Rheumatism Foundation, and the Immunogenetics Research Foundation.
REFERENCES 1. Dupont B: Immunobiology of HLA. Vol 1. New York, Springer-Verlag, 1989. 2. Dawkins RL, Christiansen FT, Kay PH, Garlepp MJ,
Genomic Conservation and Ancestral Haplotypes
McCluskey J, Hollingsworth PN, Zilko PJ: Disease associations with complotypes, supratypes and haplotypes. Immunol Rev 70:5, 1983. 3. Dawkins RL, Kay PH, Martin E, Christiansen FT: The genomic structure of ancestral haplotypes revealed by pulsed field gel electrophoresis (PFGE). In Dupont B (eds): Immunobiology of HLA, Vol 1. New York, Springer-Verlag, 1989, p. 893. 4. Dawkins RL, Zhang WJ, Degli-Esposti MA, Abraham L, McCann V, Christiansen FT: Studies of haplotypes by pulsed field gel electrophoresis. In Bailliere Clinical Endocrinology and Metabolism, Vol 5, No 2. London, Bailliere Tindall Limited, 1991, p. 285. 5. Spies T, Blanck G, Bresnahan M, Sands J, StromingerJL: A new cluster of genes within the human major histocompatibility complex. Science 243:214, 1989. 6. Spies T, Bresnahan M, Strominger JL: Human major histocompatibility complex contains a minimum of 19 genes between the complement cluster and HLA-B. Proc Natl Acad Sci USA 86:8955, 1989. 7. Sargent CA, Dunham I, Campbell RD: Identification of multiple HTF-island associated genes in the human major histocompatibility complex class III region. Eur Mol Biol Organ J 8:2305, 1989. 8. Dawkins RL, Leaver A, Cameron PU, Martin E, Kay PH, Christiansen FT. Some disease-associated ancestral haplotypes carry a polymorphism of TNF. Hum Immunol 26:91, 1989. 9. Dawkins RL, Uko G, Pestell R, McCann VJ: Association of MHC supratypes and complotypes with endocrine disorders. In Farid NR (eds): Immunogenetics of Endocrine Disorders. New York, Alan R Liss, 1988, p. 433. 10. Wilton AN, Cobain TJ, Dawkins, RL: Family studies of IgA deficiency. Immunogenetics 21:333, 1985. 11. Kelly H, McCann VJ, Kay PH, Dawkins RL: Susceptibility to IDDM is marked by MHC supratypes rather than individual alleles. Immunogenetics 22:643, 1985.
97
in different MHC supratypes detected by pulsed field gel electrophoresis. J Exp Med 168:933, 1988. 16. Zhang WJ, Degli-Esposti MA, Cobain TJ, Cameron PU, Christiansen FT, Dawkins RL: Differences in gene copy number carried by different MHC ancestral haplotypes: quantitation after physical separation of haplotypes by pulsed field gel electrophoresis. J Exp Med 171:2101, 1990. 17. Schneider PM, Carroll MC, Alper CA, Ritmer C, Whitehead AS, Colten HR: 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, 1986. 18. Tokunaga K, Zhang WJ, Christiansen FT, Dawkins RL: The genomic structure of two ancestral haplotypes carrying C4 duplications. Immunogenetics 34:247, 1991. 19. Egea GE, Yunis I, Spies T, Strominger J, Awdeh ZL, Alper CA, Yunis EJ: Association ofpolymorphisms in the HLA-B region with extended haplotypes. Immunogenetics 33:4, 1991. 20. Tokunaga K, Omoto K, Akaza T, Akiyama N, Amemiya H, Naito HS, Sasazuki T, Satoh H, Juji T: Haplotype study on C4 polymorphism in Japanese. Association with MHC-alleles, complotypes and HLA-complement haplotypes. Immunogenetics 22:359, 1985 21. Chan SH: Antigen Report: Bw46. In Aizawa M (ed.): HLA in Asia-Oceania. Sapporo, Japan, Hokkaido University Press, 1986, p. 106. 22. Rembecki RM, Kumar V, David CS, Bennett M: Hemopoietic histocompatibility (Hh-1) regulatory and structure genes of the f haplotype map to H-2. Transplantations 49:633, 1990. 23. Walsh JB: Sequence-dependent gene conversion: Can duplicated genes diverge fast enough to escape conversion? Genetics 117:543, 1987. 24. Koop BF, Siemieniak D, Slightom JL, Goodman M, Dunbar J, Wright PC, Simons EL: Tarsius delta- and betaglobulin genes: Conversions, evolution, and systematic implications. J Biol Chem 264:68, 1989.
12. Garlepp MJ, Wilton AN, Dawkins RL, White PC: Rearrangement of 21-hydroxylase genes in disease associated MHC supratypes. Immunogenetics 23:100, 1986. 13. Feinberg AP, Vogelstein B: A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6, 1983.
25. Hughes AL: Testing for interlocus genetic exchange in the MHC: a reply to Andersson and co-workers. Immunogenetics 33:243, 1991.
14. Reed KC, Mann DA: Rapid transfer of DNA from agarose gels to Nylon membranes. Nucleic Acid Res 13:7207, 1985. 15. Tokunaga K, Saueracker G, Kay PH, Christiansen FT, Anand R, Dawkins RL: Extensive deletions and insertions
26. Kay PH, Dawkins RL, Williamson J, Tokunaga K, Christiansen FT, Charoenwong P: Coexistence of an MHC chromosomal segment marked by HLA B 17, BfS, C4A6, B1, DR7, and DQw9 in different ethnic groups. Hum Immunol 23:27, 1988.