Aug 2, 1991 - We have constructed the physical map of the 3' region of the human immunoglobulin heavy chain variable region (VH) genes. DNA segments ...
The EMBO Journal vol. 1 0 no. 1 2 pp.3641 - 3645, 1991
Physical map of the 3' region of the human immunoglobulin heavy chain locus: clustering of autoantibody-related variable segments in one haplotype Euy Kyun Shin1, Fumihiko Matsuda2, Hitoshi Nagaoka', Yosho Fukita1, Takashi Imai3, Kazushige Yokoyama3, Eiichi Soeda3 and Tasuku Honjo1l2 IDepartment of Medical Chemistry, Faculty of Medicine, and 2Center for Molecular Biology and Genetics, Kyoto University, Yoshida, Sakyo-ku, Kyoto 606, and 3Gene Bank, Tsukuba Life Science Center, The Institute of Physical and Chemical Research (RIKEN), 3-1-1 Koyadai, Tsukuba, Ibaraki 305, Japan Communicated by J.Tooze
We have constructed the physical map of the 3' region of the human immunoglobulin heavy chain variable region (VH) genes. DNA segments extending to 200 kb upstream of the JH segment were isolated in two YAC clones. Five VH segments were identified in this region in the 5' to 3' order, VII-5 VIV4, VI39 VI29 and VVI-1 segments which were all structurally normal and orientated in the same direction as the JH segments. From DNA of a different cell line we have isolated a cosmid contig containing the same DNA region which has extraordinary polymorphism. The YAC and cosmid DNAs were called haplotypes A and B, respectively. Haplotype B contained an additional VH-I segment (VI4.lb) between the VII5 and VIV4 segments. VI4.1b segment is almost identical to a previously published VH sequence encoding a rheumatoid factor. Another VH segment in the B haplotype (VI-3b) corresponding to the VI3 segment also showed 99.7% nucleotide sequence homology with an anti-DNA autoantibody VH sequence. However, none of the VH sequences in haplotype A showed such strong homology with autoantibody VH sequences. The results suggest that VH haplotypes may have linkage with autoantibody production.
Introduction Generation of immunoglobulin (Ig) variable region diversity relies on three major mechanisms: evolutionarily determined germline repertoire, somatic association of germline V, D and J segments to create a functional V gene and somatic mutation (Tonegawa, 1983; Honjo and Habu, 1985). Relative contribution of the three mechanisms is hard to assess, partly because we do not know the number and organization of Ig V segments. The majority of the human Ig heavy chain (H)V region segments are mapped at the distal region of chromosome 14 (q32) which is estimated to encompass 3000 kb DNA by pulsed field gel electrophoresis (PFGE) (Croce et al., 1979; Berman et al., 1988; Matsuda et al., 1988). The human VH segments are classified into six families which are intermingled with each other (Kodaira et al., 1986; Lee et al., 1987; Shen et al., 1987; Berman et al., 1988). The human VH locus is thus expected to Oxford University Press
contain many repeated regions which have been generated by extensive duplication and recombination during evolution. In addition, the human VH locus appears to have considerable polymorphisms. A highly restricted usage of VH segments was reported. Namely, preferential utilization of the 3' proximal VH segments was reported in human fetal B cells (Schroeder et al., 1987), leukemic cells (Shen et al., 1987; Logtenberg et al., 1989a), and autoantibodies (Logtenberg et al., 1989b,c). These results indicate that construction of the physical map of the human VH lOCUS is important for studying not only evolution of the multi-gene family but also association of the physical location of VH segments with their usage. Previous studies on the physical mapping of the human VH locus using cosmid vectors allowed us to isolate many contigs which altogether cover more than 2000 kb. However, we had difficulties in connecting these contigs, partly because of the presence of wide regions without any VH segments and partly because of abundant repetitive regions that made the identification of the overlapping clones obscure. To overcome these problems we introduced as a vector yeast artificial chromosome (YAC) which can clone several hundred kb DNA (Burke et al., 1987). In the present report we describe the physical map of the 3' proximal 200 kb DNA of the human VH locus and nucleotide sequences of four VH segments newly identified in this region. In adddition, we found an extraordinarily polymorphic allele that contains an additional VH segment. This haplotype contained two VH segments which are almost identical to autoantibody VH sequences, suggesting a possible linkage of autoimmunity and VH haplotypes.
Results and discussion Isolation of JH proximal 200 kb region from YAC library We have screened a YAC library of a human Epstein -Barr virus-transformed cell line CGM1 (Imai and Olson, 1990) using VHI and VHIII specific primers, and isolated 18 independent YAC clones containing VH segments. Among these YAC clones, we found a contig consisting of two YAC clones, Y103 and Y20, which hybridized with D, JH and VH-VI probes. The two YAC clones were shown to overlap 100 kb and to cover a 240 kb region by mapping with several restriction enzymes (Figure 1). We constructed a cosmid library from Y103 clone DNA, screened it by using VH probes and obtained 15 overlapping cosmid clones to cover the entire Y 103 clone. The 3' portion of the Y20 clone which does not overlap with the Y 103 clone, was identical to a cosmid clone contig isolated previously (Sato et al., 1988). We identified five VH segments in the two YAC clones. We propose to rename all the VH segments by family number and the order from the 3' end. Therefore, the order of VH segments in this region is; VII 5, VIVP4, VI3, -
3641
E.K.Shin et al.
7
YC45 YC24YC49
YC20
YC25
YC19
YC3 YC44
u2-2
YC53
Y103
Y20
A
D4 D1 D2 D3
i.51-1- -Ia
VD- 5 MV-.4Vj-3
Pi[Fi]fTTI IaII
II
Eco RI Rem HT
K 7
3-64
4-1 8' D31 S-74
II
k.
I
lI
Iif
EcoRI "L Ir 11,I I
#79 i7-11
I iIZE|IJ
3lull
1.
II
L
11
11
1
I I
VM l
nijnj
,,.,1,,
I,, 1111111
Hindm
VI-2
JH UCA_. C8
-
lI
III
1 ,1 1
III1 I I
III[ 11
lil
I
II
II II
l
l
II
II I III Bam HI 4414L4 I II I I lilt I I I iAP r i I1 LI &IfUIL -L:J- I J 1 JI LI
B
VI-4.lb
VI-3b VI-2b VIV-4b
VII-Sb ,,
24-53
\. I- - -
+200
lii ,
lo
M-21 1 3-35 3 _-79S
+150
+100
I +50
0
(kb)
-40
Fig. 1. Restriction map of the 3' proximal 200 kb of the human VH locus and comparison of the restriction maps of A and B haplotypes. Y103 and Y20 clones were isolated from a YAC library of CGM1 DNA. Cosmid clones derived from Y103 DNA are indicated by YC and cosmid clones covering between Vvl-l and CA were isolated previously (Sato et al., 1988). Cosmid clones 3-79 and 3-35 were obtained from a human placenta DNA (Kodaira et al., 1986) and the other cosmid clones in haplotype B from FLEB14-14 (Matsuda et al., 1990). The position of #79 unique fragment which is used for Southern hybridization of DNAs in Figure 3 is shown by a bar in B haplotype. Homologous regions between A and B haplotypes are indicated by squares. Restriction sites of EcoRI, BamHI and HindIII are indicated by vertical lines.
VI-2 and VVlI I from 5' to 3'. The restriction fragment sizes of the VH segments in this region agree with those estimated by two-dimensional PFGE (Walter et al., 1990). Not all human VH segments preferentially expressed during early stages of ontogeny map in the proximity of the JH segments The nucleotide sequences of VH segments within the Y103 clone indicate that none of them are structurally defective (Figure 2). Essential nucleotides of the heptamers and nonamers of recombination signals are conserved (Hesse et al., 1989). The transcriptional orientations of the five VH segments are all identical to those of the D and JH segments. We examined the nucleotide sequence homology between these VH segments and those reported previously. We found that the VI-2 segment was identical to the 20P3 VH gene, one of the rearranged VH genes found in human fetal liver (Schroeder et al., 1987). The VvIl1 segment has already been shown to be expressed in acute lymphatic leukemia and autoimmune patients (Berman et al., 1988; Logtenberg et al., 1989b) as well as in fetal liver. However, none of the other three VH segments located within this region are homologous to reported VH genes. Several investigators have reported that during the early stage of human and murine ontogeny, VH segments located in the JH proximal region are more frequently utilized than those located distantly from the JH locus (Yancopoulos et al., 1984; Schroeder et al., 1987). Schroeder et al. (1987) 3642
reported that a few of the VH-I, VH-III and VH-IV family genes are used most frequently in an 18 week old fetal liver. However, no germline VH segments corresponding to the VvlIII and VHWIV family genes were found in the present region. In fact, we found two VH segments identical to the expressed VH-m (60 P2) and VH IV (58P2) family genes at around 850-900 kb upstream from the JH segment (unpublished data). The VH segment usage in early ontogenic stage does not always appear to be closely associated with the distance from the JH segments in man. This conclusion is in agreement with a recent report that no clear distinction of human VH usage was found between 30 weeks of gestation and adult (Cuisinier et al., 1990). Isolation of a polymorphic counterpart by cosmid
contig Among cosmid contigs previously isolated from DNA of different individuals (Kodaira et al., 1986; Matsuda et al., 1990), we found a 100 kb cosmid contig that has extensive homology with, but which is considerably different from the YAC contig by restriction mapping. This homologous region is located between 180 and 80 kb from the JH segment (Figure 1). Comparison of the restriction maps of the two contigs shows that the restriction sites surrounding VH segments are almost identical, and the order of VH segments is the same except for the insertion of the VI4 lb
segment between VII-5b and VIv4b. The five VH segments in the cosmid contig were named with b after the ordered
Human VH lOCUS (a)
-5
-19
VI-2 V143 VI-3
Not
Asp
Trp
Thr
Trp Arg
Ilie
Lou
tgaqaqotoc gttectcacc ATG
GAC
TGG
ACC
TGG
ATC
CTC TTC
.a,aac,...
AGG
Ph.
Va1 GTG
Lou TTG
Ala
Ala
OCA
OCA
Ala 0CC
Thr
G
ACA
G/
ccctaqtccc
qtaaqaqqct
aqtqatqaqa aaqaqattqa
..q*q aaa...... q-.q
.T
.Tq--q**a-a...... q..q
cc,..a*a acc..t.I c*a,a -c -a.-.A
VI.'i
VI-2
qqqaqatctc
Via3 VI-a.lb
qtqttctctc cacaq/
atccacttct ....C.
*.q-c.tc
--qa-tco
a. t
t*..
*t...ct
ly
Ala
His
liar
Gin
Va1
GCC
CAC
TCC
CAG
GTG
*T
*T
Gin
Lou
Val
Gin
liar
Glly
Ala
Glu
Va1
Lys Lys
Pro
Gly
Ala
CAG
CTG
GIG
CAG
TCT
GOG
GCT
GAG
GmG
AAG
AAG
CCT
GOG
GCC
Siar Gly OCT TCT GGA
Ala
Cys
Lya
TGC
AAG
Tyr Thr Ph. TAC
TTC
ACC
Ser
Val Lys Va1
TcA
GTG
liar
GTC
AAG
TCC
I.T
T.
T
C
I-T..A.....T-.T.:...... 50
40
30
VI.2 VioS VI-lb
-a.
-
20
10
GA
./
..ct. ..t/... .-.-c. *-q...c-cac.c......
qtccaqtcca
q..c"-aq ....a.
q.-.c-qq ....a.
q...aGcca
...C.-t.q.c ....q
q.
+1
-4
... ...
Thr
Gly
Tyr Tyr
Hat Rli srpVal
Arq
Gin
Ala
Pro
Gly
Gin
Gly
Lou
Glu
Trp
ACC
GGC
TAC
ATG
GTG
CGA
CAG
0CC
CCT
OGA
CMA
GGG
CIT
GAG
TGG
TAT
TOG
CAC
Not Gly Trp Ile Asn OGA TOG ATC AAC
ATG
Pro
Asn
Sar
Gly
CCT
AAC
AGT
GOT
TGC. ~~~~~A.. AT.AC .GG
VI4.lb 60
Gly
VI.2 Vi1s VIaJ
Thr
V14.Ib
AAMC.
Gin
Lys
Ph.
Gin
Gly
Arg
Val
Thr
Not
Thr
Arq Asp
Thr
liar
AAG
TTT
CAG
GOC
AGG
GTC
ACC
ATG
ACC
AGO
ACG
TCC
G-
-C
GAC
.-------A
....T -A ..
62A
82Bi 82C
Sar
Thr
Ala
Tyr
Net
Glu
Lou
liar
Arq Lou
Arq
Siar
Asp Asp
Thr
AGC
ACA
GCC
TAC
ATG
GAG
CTG
AGC
AGO
AGA
TOT
GAC
ACG
62
CTG
GAC
*T,
G
0CG.C
G..C -T........ A CAC.C-C -A-AG GG AT C T-TT.....C -TG .-
.-C -A .-CG.. GC-C ACA--A C-T-TGT
-CG
s0
Ila ATC
70
Tyr Ala
Asn
GGC ACA AAC TAT GCA CAG ..A :::A A AM
A
.-T
*G
TC
90
Va1
Ala
VI-' VI4s
Ala
Tyr Tyr Cys
Arg
CAGjA
TOT 0CG AGA GA CCAGGGAGGAGOCAG TGAAAACCCA CATCCTGAGG GTO ZOAGU&G ..T ...........................A .......... 0CC
VI-&
GIG
TAT
TAC
. . . -. . . . . . . .A..A .TOT .~~~~~~~~~~~~~~.G..........A
. . . . ..
VI-4.lb
(b)
-5
-19
VU.. V114
ccaqotcoca ccctcctctq
gqttqaaaaa
qqtaccaqct
qeogaqoaca
.....nnnnnn.....A..........a
caqtqactcc
tqtqcaccacc
Nat
Asp
Thr
Lou
Cys
Siar
Thr
Lou
Lou
Lou
Lou
Thr
ATO
GAC
ACA
CTT
T0C
TCC
AGO
CTO
CTO
CTO
CTO
ACC
qtqaqtqctq
V11a
tqqtcaqqqa
ctccttcaoq
Pro
Sar
T
CCT
TCA
TI
.C............nnn..
caqttttctt qtttqtqgqc tteatcttct
gqtqaaacat
ccacaq/
tatgctttct
/
+1
-4
Vu.a
Ile ATC
...................
rp
Val
GO
GTO
Clu TTO
liar
Gin
Ile
Thr
Lou
Lys
Glu
liar
TCC
CAG
ATO
ACC
TTO
AAG
GAG
TCT
..... .......... ..... .......... ... ./.........
10
Vus4 V11.a
35 35A
30
20
Gly
Pro
Thr
Lou
Va1
Lys
Pro
Thr
Gin
Iixr
Lou
Thr
Lou
Thr
Cy.
Thr
Ph.
GOT
GOT
ACG
CTG
GTG
AAA
CCC
ACA
CAG
ACC
CTC
ACG
CTG
ACC
TOC
ACC
TTC
liar
Gly
Ph.
TOT
GGG
TTC
liar
Lou
liar
Thr
liar
Gly
TOA
CTC
AGC
ACT
AGT
GOA
V.1 GTG
Gly
V.1
GTG
GOT
35B
40
Gly
Trp
Iie Arq
Gin
Pro
GGC
TOG
ATC
CGOT
CAG
CCC
TO- T*** .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~AG
............
VII.
V11a
Pro
Gly
Lys
Al a
Lou
Glu
CCA
GOA
AAG
0CC
CTG
GAG
Trp L.au Ala Lou Ile Tyr Trp As n Asp Asp Lys Arq Tyr liar Pro liar Lou Lys liar TOG CIT 0CA CTC ATT TAT TOG AAT GAT GAT AAG GOC TAC AGC CCA TOT CTG AAG AGC .~~~~~~~~~~~~~~G.-
so
S2
S82A S829S2C
Gin
VlVlLuTrNat
Thr
Asn
Hot
Asp
Pro
V.1
Asp
Thr
Ala
Thr
Tyr Tyr
Cys
Ala
Hils
Arq
AAC
CAG
GIG
ACC
AAC
ATG
GAC
CCT
GTG
GAC
ACA
GCC
ACA
TAT
TOT
0CA
CAC
AGA
Vos6
AG0CTGCTT
CIT
ACA
CTCATTGOTO
ATG
CTOCCTOCCC
Arq Lou
Thr
CTC
ACC
AGG
Ila Thir Lys ATC ACC AAG
Asp
Thr
liar
GAC
ACC
TCAA
Ly s
90
Asn
VUl
GTC
;La
60
50
C
Z&"&Mg
GGGCACCTCC
ACACAGCCCA
TOT
Ae&&a&hGr
C
TOTOTCTOCT
AACAGGAAAG
ACCTCT0CAG
TAC
-19
(c) V1V-4
cacaqgaa
caqctcacat
attcaqgqtc
atttccttaa
accaccacac
qqqaaatact
ttctqaqact
catgqacctc
ctqcacaaqa
ac
Nat
Lys
His
Lou
Trp
Ph.
Ph.
Lou
Lou
Lou
ATG
AAA
CAC
CTG
TOG
TTC
TIC
CTC
CTG
CTG
-5
Va1 GTG
Ala
Ala
Pro
Arq
0CA
OCT
CCC
AGA
qtqaqtqtct
caaqqctqca
qacatqqqqa
tatqqqaqqt
in
Lou
Gin
Glu
liar
Gly
Pro
Gly
Lou
V.1
Lys
Pro
CTO
CAG
GAG
TCG
G0C
CCA
GGA
CTO
GTO
AAG
CC?
____40 Arq Gin
Pro
Ala
CGOG
CCC
0CC
CAG
Gly
Swr Lys TOC
AGA""
liarGl
TCGO
GAG
ccaqqqctca
hLoSrLuTrCy CTO TCC CTO AGO T0C
AGO
Thr
Va1
AC?
GTO
Asn
Lou
Glu
Top
Ile Gly
Tyr Thr
C?G
GAG
TOOG
ATT
TAT
Gin
Ph.
liar
CAG
TIC
TOC
liar
TC?
Arq Ila OGG CGOT ATC
ACC
G0C
liar
TGO
rp
Val
Lou
P.r
Gin
Va1
G
GO
GTC
CTG
TOC
CAG
GIG
C
Ile ATO
liar
lior
Tyr
Tyr
Trp
AGT
AGT
TAC
TAC
TOGO AGC TGO
liar
Ile
Top
ATO
70
Gly
liar
Thr
Asn
Tyr
Asn
Pro
liar
Lou
GGG
AGC
ACC
MAC
TAC
AAC
CCC
TOC
CTC
Ala
Arg
0CG
AGA
Lys
Arq Val
liar
MAG AGT CGOA
GTC
Thr
Nat
liar
Val
Asp
ACC
ATG
TOA
GTA
GAC
90
Lou
lior
liar
Val
Thr
Ala
Ala
Asp
Thr
Ala
Val
Tyr
Tyr Cys
C?G
AGC
TOT
GTG
ACC
0CC
0CG
GAC
ACG
0CC
GTO
TAT
TAC
GOACCG0C0C
GOT
liar
Lys
TOCCTOCAGO GAGGCGGAGG
Gly Gly
AGT
C?G MAG
Lou
aql
ctctqttcac
60
Gly OGA
$~~ ~~0
MAG AAC
ctgtqqqtct
a30-
50
Ltya MAG
10
Thr
ACG
qcctctqatc
20I
10
AG
+1
-4 T
T/
TOT
GA
GAGALM
AGGGGAGOTG
AGTOTOAGCC
CAG
ALC
AGOTGCT0CT CAAGACCAGC AGGGGOC0CG CGGGOCGCAC AGAGCAAGAG GCCGGGTOAG
haplotypes. VH sequences are grouped by famnilies (a) VHI1, (b) VH-1I and (c) VH-IV. VI-2b and VIV-4b previously published (Matsuda et al., 1988; Lee et al., 1987). Bases identical to the above are shown by dots. Intron sequences are shown in lower case. Amino acid sequences deduced from the top sequences are shown above. Recombination signals are underlined and deletion is shown by asterisks. Numbers according to Kabat et al. (1987) are given above amino acid residues. Primers used for sequencing are indicated by horizontal arrows. The nucleotide sequences have been subm-itted to the EMBL database (accession nos X62106-X621 12).
Fig.
2.
VH
sequences within A and B
sequences in B
VH
haplotype
They
notation.
with the Table
JH
I
the cosmid
segments in
have the
segments and
shows
comparisons
were
transcriptional orientation structurally non-defective.
same are
nucleotide
and
amino
acid
sequence
corresponding VH segments in Y 103 and contig, respectively. The corresponding VH two DNA regions are highly conserved in of
al., 1984), CESS cells (Takahashi
et al., 1984; human line), four hydatidiform moles (homozygous cells) and 15 individuals (Figure 3). BamHl digests of these DNAs were hybridized with a unique fragment # 79, which was et
B cell
isolated from the EcoRI
fragment
of the 3-79 clone
(Figure
BamHl band in 3-79 kb BamHl band in Y20
1). The probe # 79 detected the 8.0 kb
nucleotide ( >92 %) and amino acid sequences ( >88 %). The
and M-24 cosmid clones, but the 3.5
VIA4
probe cross-hybridized with the 8.5 was located in the 14q32 region but not within Mb upstream from the JH lOCUS (unpublished data). We detected the 3.5 and 8.5 kb fragments in DNAs one hydatidiform mole (lane of CGM1I, Rag/G04, E3D
segment does
b
not
have
a
homologue
in the YAC
In order to determine whether these
regions locus,
arose
we
from
polymor-phism
or
examined the presence of
various DNAs derived from
homologous DNA duplication in the VH two DNA regions in
FLEB14-14
cells (Matsuda
germiine control), Rag/G04 cells [Matsuda et al., 1990, human -mouse hybrid cell containing a single human chromosome of t(X: 14) (p22; q32)], CGMI1 cells (source of YAC library), E3D cells (Noma
et
al.,
and Y 103 clones. This kb
DNA.
1990,
human
BamHl band,
which
10,
20), and three individuals the
8.0
and
hydatidiform
8.5 kb moles
(lanes
13, 14 and 15). We detected
fragments
in
FLEB
14-14,
two
(lanes 21 and 23) and six individuals
(lanes 7, 10, 11, 12, 16 and 18). The other six individual (lanes 5, 6, 8, 9, 17 and 19) contained 3.5, 8.0 and
DNAs
3643
E.K.Shin et al. Table I. Comparison of nucleotide and amino acid sequences
V115
V,,-5b
VIV4
V,-3b VI-2b RF-TS3 21/28 (= 8E10)
V12
55/296 (25/98)
56/296 (25/98)
1/294 (1/98)
55/296 (23/98)
(13/98) (3/98)
56/294 (24/98) 58/294 (26/98)
1/296 (0/98) 31/296 (15/98)
RF-TS3
9/301 (5/100)
VI4. Ib
VIV-4b
21/28 (= 8E10)
Vi3
29/296 (14/98) 6/296 34/296 56/294 6/296
(3/98) (18/98) (26/98) (3/98)
30/296 4/296 55/294 29/296
(24/98) (13/98)
The nucleotide (or amino acid) sequences of coding regions except for the signal peptide sequence are compared. VIV 4b and VI-2b segments are identical to V79 and V35, respectively, which were described previously (Lee et al., 1987; Matsuda et al., 1988). Numbers are mismatched basepairs (residues) over total basepairs (residues) between corresponding pairs of A and B haplotype VH segments or between those of autoantibody and other
VH segments. 8.5 kb fragments. One hydatidiform mole (lane 22) and CESS (lane 24) had only the 8.5 kb fragment (see below). Since homozygous hydatidiform moles and Rag/GO4 showed clear segregation of the 3.5 and 8.0 kb fragments, these two fragments are likely due to polymorphic difference. Studies on 15 individual DNAs also indicate that the 3.5 and 8.0 kb fragments segregated. In agreement with the above conclusion, both 8.0 and 3.5 kb BamHI fragments were absent from CESS DNA which had deleted VH segments located within 300 kb upstream of the JH segment from both chromosomes, as assessed by the germline location of the rearranged VH segments on both chromosomes (data not shown). We referred to the Y103 clone and the cosmid contig DNAs as A and B haplotypes, respectively. The VI-3b sequence is identical with the rearranged VH sequence of E3D10 cell line. However, E3Dl0 DNA contained the 3.5 kb BamHI fragment hybridized with # 79 probe, the marker of the A haplotype. The result suggests that the A and B haplotypes may have recombinants and thus more haplotypes will be found among the population. One hydatidiform mole (lane 22) had only the 8.5 kb fragment, which is due to a large deletion including the VIV4 segment (data not shown). The same deletion was found in several other individuals (data not shown), suggesting the presence of another haplotype. We then studied segregation of haplotypes A and B in five families and confirmed the above conclusion. Figure 4 shows Southern blot hybridization analyses of one family which has four children including twins. The parents (I-1 and 1-2) have both 8.0 kb (the B haplotype marker) and 3.5 kb (the A haplotype marker) fragments, while three children including twins (II-2, 11-3, 1-4) have both 8.0 and 3.5 kb fragments and child II-1 has only the 3.5 kb fragment. Considering the double intensity of the 3.5 kb fragment in 11-1 child, she is likely to have homozygous A haplotype. Because identical twins (II-3, II-4) should be regarded as one zygote, the ratio of a/a to a/b is 1:2 and agrees with Mendel's law. The relative intensities of the 8.0 and 3.5 kb bands are the same in all the heterozygous individuals. Taking all these data together, haplotypes A and B are most likely allelic. Two VH segments in the B haplotype are almost identical to autoantibody VH genes When we compared the nucleotide sequences of five VH segments in the B haplotype with VH genes published previously, the VI-4 lb segment and RF-TS3 cDNA encoding a rheumatoid factor (Pascual et al., 1990) were
3644
Fig. 3. Southern hybridization of DNAs to distinguish haplotypes A and B. Origin of DNA in each lane: 1, CGM1; 2, FLEB14-14; 3, Rag/GO4; 4, E3D10; 5-19, individual blood; 20-23, complete hydatidoform moles; 24, CESS. DNAs were digested with BamHI. Probe use was shown as #79 in Figure 1.
...
,,,@,.-
Fig. 4. Mendelian segregation of the haplotypes A and B in one family. (A) Phylogenic tree of a sample family. Squares and circles indicate male and female, respectively. Letters a and b represent haplotypes A and B, respectively, as determined by Southern hybridization. (B) Southern hybridization pattern of individual DNAs of the sample family. Each member of the family tree in (A) is shown at the top of the lane.
very similar (Table I). They differ only at residue 85 in the mature protein although two additional divergences are found in the hydrophobic signal peptide. The VI-3b segment also showed a remarkable nucleotide sequence homology (99.7%) with the VH sequences of two identical anti-DNA autoantibodies, 21/28 and 8E10, which were derived from an SLE patient and a leprosy patient, respectively (Dersimonian et al., 1987; Cairns et al., 1989). Although 21/28 and 8E10
Human VH IOCUS
have different D and JH segments, their VH segments are identical and differ only in one base at residue 2 from the V1-3b sequence but six bases from the VI-3 sequence. No other VH-I genes so far sequenced are > 95 % homologous to the 21/28 VH sequence. Recently, several studies have demonstrated that unmutated germline VH genes can encode autoantibodies such as rheumatoid factors and anti-DNA antibodies (Chen et al., 1986; Radoux et al., 1986; Sanz et al., 1989). It is interesting that the VH haplotype B but not haplotype A contains two VH segments, i.e. VI3b and VI4 1b, highly homologous to autoantibody VH genes. It is important and feasible to test whether the VH haplotype B is linked with rheumatoid arthritis or other autoimmune diseases.
Materials and methods The YAC library was constructed from CGM 1 DNA partially digested with EcoRI using pYAC4 as vector (Imai et al., 1990). Screening of the YAC library by polymerase chain reaction and colony hybridization, and purification of YAC clones were performed as described (Green et al., 1990). Partial Sau3A cosmid library in Lorist2 vector was constructed from Y103 DNA (Gibson et al., 1987). Libraries were screened with 32P-labelled VH-I, VH-II and VH IV probes (Kodaira et al., 1986; Lee et al., 1987). Southern hybridization of DNA was done according to the method of Southern et al. (1975). DNAs were digested with restriction enzymes and electrophoresed in a 0.7% agarose gel and transferred to nitrocellulose membrane filters (Schleicher & Schuell). The filters were hybridized to 32P-labelled probe at 65°C for 12 h. Filters were washed three times (30 min each) in 0.1 x SSC, 0.1% SDS at 65°C and exposed to X-ray film at room temperature for 3 days. Sequences of VH segments were determined by the chain termination method (Sanger et al., 1980; Hattori and Sakaki, 1986) with the specific oligonucleotide primers indicated.
Acknowledgements We thank Drs T.Miki (Osaka University) for providing individual DNAs. C.Azuma (Osaka University) and C.Ban (National Osaka Hospital) for gifts of hydatidiform mole DNAs, and A.Shimizu (Kyoto University) for valuable comments. We are also grateful to Ms M.Wakino and T.Miwa for their technical assistance, and to M.Yamaoka and H.Kanaya for their help in preparing the manuscript. This work is supported in part by a grant from the Science and Technology Agency of Japan.
Kabat,E.A., Wu,T.T., Reid-Miller,M., Perry,H.M. and Gottesman,K.S. (1987) Sequences of Proteins of Immunological Interest. NIH publications, Washington, DC. Kodaira,M., Kinashi,T., UmemuraJ., Matsuda,F., Noma,T., Ono,Y. and Honjo,T. (1986) J. Mol. Biol., 190, 529-541. Lee,K.H., Matsuda,F., Kinashi,T., Kodaira,M. and Honjo,T. (1987) J. Mol. Biol., 195, 761-768. Logtenberg,T., Schutte,M.E., Inghirami,G., Berman,J.E., GmeligMeyling,F.H.J., Insel,R.A., Knowles,D.M. and Alt,F.W. (1989a) lit. Immunol., 1, 362- 366. Logtenberg,T., Young,F.M., van ES,J.H., Gmelig-Meyling.F.H.J. and Alt,F.W. (1989b) J. Exp. Med., 170, 1347-1355. Logtenberg,T., Young,F.M., van ES,J.H., Gmelig-Meyling,F.H.J., Berman,J.E. and Alt,F.W. (1989c) J. Autoimmunity, 2, 203-213. Matsuda,F., Lee,K.W., Nakai,S., Sato,T., Kodaira,M., Zong,S.Q., Ohno,H., Fukuhara,S. and Honjo,T. (1988) EMBO J., 7, 1047-1051. Matsuda,F., Shin,E.K., Hirabayashi,Y., Nagaoka,H., Yoshida,M.C., Zong,S.Q. and Honjo,T. (1990) EMBO J., 9, 2501-2506. Noma,Y., Yaoita,Y., Matsunami,N., Rosen,A., Klein,G. and Honjo,T. (1984) Mol. Biol. Med., 2. 337-350. Pascual,V., Randen,I., Thomson,K., Sioud,M., Forre,O., Natvig,J. and Capra,J.D. (1990) J. Clin. Invest., 86, 1320-1328. Radoux,V., Chen,P.P., Sorge,J.A. and Carson,D.A. (1986) J. Exp. Med., 164, 2119-2124. Sanger,F., Coulson,A.R., Barrel,B.G., Smith,A.J.H. and Roe,B.A. (1980) J. Mol. Biol., 143, 161-178. Sanz,I., Dang,H., Takei,M., Talal,N. and Capra,J.D. (1989) J. Immunol., 142, 883-887. Sato,T., Matsuda,F., Lee,K.H., Shin,E.K. and Honjo,T. (1988) Biochem. Biophvs. Res. Comnmun., 154, 265-271. Schroeder,H.W.,Jr, Hillson,J.L. and Perlmutter,R.M. (1987) Science, 238, 791 -793. Shen,A., Humphries,C., Tucker,P. and Blattner,F. (1987) Proc. Natl. Acad. Sci. USA, 84, 8563-8567. Southern,E.M. (1975) J. Mol. Biol., 98, 503-517. Takahashi,N., Noma,T. and Honjo,T. (1984) Proc. Natl. Acad. Sci. USA, 81, 5194-5198. Tonegawa,S. (1983) Nature, 302, 575-581. Yancopoulos,G. D., Desiderio,S.V., Paskind,M., Kearney,J F.. Baltimore,D. and Alt,F.W. (1984) Nature, 311, 727-733. Walter,M.A., Surtl,U., Hofker,M.H. and Cox,D.W. (1990) EMBO J., 9, 3303-3313.
Received on June 10, 1991; revised on August 2, 1991
References Berman,J.E., Mellis,S.J., Pollock,R., Smith,C.L., Suh,H.. Heinke.B.. Kowal.C., Surti.U., Chess,L., Cantor.C.R. and Alt,F.W. (1988) EMBO J., 7, 727-738. Burke,D.T., Carle,G.F. and Olson,M.V. (1987) Science, 236, 806-8 12. Cairns,E., Kwong,P.C., Misener,V., Ip,P., Bell,D.A. and Siminovitch,K.A. (1989) J. Immunol., 143, 685-691. Chen,P.P., Albrandt,K., Orida,N.K., Radoux.V., Chen,E.Y., Schrantz.R., Liu,F. and Carson,D.A. (1986) Proc. Natl. Acad. Sci. USA, 83. 8318-8322. Croce,C.M., Sander,M.. Martinis,J., Cicurel,L.. D'Ancona,G.G.. Dolby,T.W. and Koprowski,H. (1979) Proc. Natl. Acad. Sci. USA, 76. 3416-3419. Cuisinier,A.M., Fumoux,F., Moinier,D., Boubli,L., Guigou,V., Milii,M., Schiff,C., Fougereau,M. and Tonnelle,C. (1990) The Neu Biologist, 2, 689-699. Dersimonian,H., Schwartz,R.S.. Barrett.K.J. and Stollar.B.D. (1987) J. Immunol., 139, 2496-2501. Gibson,T.J., Coulson,A.R., SulstonJ.E. and Little,P.F.P. (1987) Gene. 53, 275 -281. Green,E.D. and Olson,M.V. (1990) Proc. Natl. Acad. Sci. USA. 87. 1213-1217. Hattori,M. and Sakaki,Y. (1986) Anal. Biochem., 152. 232-238. Hesse,J.E., Lieber,M.R., Mizuuchi,K. and Gellert,M. (1989) Genes Dev., 3, 1053-1061. Honjo,T. and Habu,S. (1985) Annu. Rev. Biochem.. 54, 803-830. Imai,T. and Olson,M.V. (1990) Genomnics. 8, 297-303.
3645
