close

Вход

Забыли?

вход по аккаунту

?

Preferential utilization of a novel v╨Ю┬╗3 gene in monoclonal rheumatoid factors derived from the synovial cells of rheumatoid arthritis patients.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM Volume 37
Number 6, June 1994, pp 860468
0 1994, American College of Rheumatology
860
PREFERENTIAL UTILIZATION OF A NOVEL VA3 GENE IN
MONOCLONAL RHEUMATOID FACTORS DERIVED FROM THE
SYNOVIAL CELLS OF RHEUMATOID ARTHRITIS PATIENTS
RICHARD W. ERMEL, THOMAS P. KENNY, ALICE WONG, ALAN SOLOMON,
POJEN P. CHEN, and DICK L. ROBBINS
Objective. To further our understanding about
the molecular genetics of rheumatoid factor (RF) in
rheumatoid arthritis (RA).
Methods. The heavy and light chain variable
region (V) genes of 5 new human monoclonal IgM RFs
were cloned and sequenced using the polymerase chain
reaction and the dideoxynucleotidetermination method.
Results. The results reveal the recurrent usage in
two RA patients of a novel VA3 germline gene, designated Humlv3c93. Specifically, in 2 of 3 RFs (C93 and
D53) from one patient, the light chains in the VA
geneencoded region were identical to each other and to
the light chain of an RF (H4) from another patient.
Serologically, the light chains of these 3 RFs were
classified as members of the VA3b sub-subgroup. Each
of the RFs was encoded by a different VH gene. Both
C93 and D53 bound specifically with human and rabbit
IgG, whereas H4 was monospecific for rabbit IgG.
Conclusion. Since the lv3c93 gene is not homologous to any reported VA sequence from natural autoantibodies, it is possible that lv3c93 may represent a
disease-specificRF-related VA gene. Moreover, the amino
acid sequence CSGGSCY in the third complementaritydetermining regions of 2 of the RF heavy chains is
encoded by the DLR2 gene segment and has been found
Supported in part by NIH grants AR-39831 and CA-10056
and the Arthritis National Research Foundation. Dr. Solomon is an
American Cancer Society Clinical Research Professor.
Richard W. Ermel, DVM, MPVM, PhD: University of
California at Davis; Thomas P. Kenny, BS: University of California
at Davis; Alice Wong, BS: University of California at Davis; Alan
Solomon, MD: University of Tennessee Medical Center, Knoxville;
Pojen P. Chen, PhD: University of California at San Diego; Dick L.
Robbins, MD: University of California at Davis.
Address reprint requests to Dick L. Robbins, MD, Division
of Rheumatology , Allergy, and Clinical Immunology, University of
California School of Medicine, TB192, Davis, CA 95616.
Submitted for publication July 28,1993; accepted in revised
form December 29, 1993.
previously in 2 other RA-derived RFs, and thus may
play a significant role in antigen binding.
The pathogenic role and derivation of rheumatoid factor (RF) in rheumatoid arthritis (RA) is not well
understood, although a great deal of evidence suggests
that RF participates in sustaining inflammatory synovitis. RFs are polyclonal autoantibodies directed
against antigenic sites in the Fc portion of the IgG
molecule (1). Since the synovium is primarily involved
in the immunopathologic process of RA and in RF
synthesis (2-5), we have focused on RF produced de
novo by rheumatoid synovial cells (RSC). Our serologic studies showed that the RFs from rheumatoid
synovia have unique fine subclass specificities and
affinities that are different from those of serum RFs.
For example, RFs from rheumatoid synovia bind well
to IgG3, with higher avidity than to IgG1, 2, and 4; in
contrast, serum RFs react strongly with IgG1, 2, and
4, but poorly with IgG3 (6). Thus, the molecular
characterization of RSC RFs could be of key importance in understanding the etiology and pathogenesis
of RA.
In previous studies, we derived hybridomas
from RSC that secreted monoclonal RF (mRF) (7,8),
which were distinctly different from mRF paraproteins
found in mixed (type 11)cryoglobulinemia and lymphoproliferative diseases. Serologic analysis of the mRFs
showed that they were monospecific for human IgG
and displayed their own characteristic specificities
toward human IgG subclasses (7,8). To define the
genetic basis of RF in RA, we have cloned and
sequenced the heavy (H) and light (L) chain variable
region (V) genes of these RSC-derived mRFs. Our
studies and those from other laboratories have provided further evidence that the RSC RFs utilize a
relatively large V gene repertoire. Specifically, our re-
RF Vh3 GENE
cent studies have identified 3 new RF-related V genes:
HK102, ha3~6-correspondinggene, and Humlv3 18 (8).
Thus, currently, the IgM human RF-related V genes
identified include 10 V H genes (hv1051, hvlfl0, GL-SJ2,
1.9111, VH26, VH4.18, VH4.21, V71-2, V79, and the
ha3~6-corresponding gene), 6 V K genes (HK102, Vd,
A23, kv325, kv328, and Vg), and 5 V h genes
(Humlv 1 17, Humlv 1L 1, the mAb63-corresponding
Humlv418 gene, the HAFlO-corresponding VA8 gene,
and Humlv318) (reviewed in refs. 8-10).
To define further the molecular genetic basis for
R F in RA, we have recently generated 5 new mRFs
that were derived from the RSC of 2 unrelated RA
patients. Sequence analysis of the mRF V L genes
indicated the utilization of a heretofore undescribed
germline VA3 gene in 3 of the 4 A-type mRFs. In these
mRFs, the Vh3 gene was associated with 3 different
VH genes and expressed 3 different specificities. Our
studies also have led t o the identification of 5 new IgM
RF-related V germline genes. W e also found that the
H chains of 2 of the 5 RFs used a DLR2-encoded
heptapeptide identified previously in 2 other RAderived monospecific RFs (10).
MATERIALS AND METHODS
Generation and characterization of IgM RF-secreting
hybridomas from RSC. The methods and serologic data are
provided in detail elsewhere (7). Briefly, cells from rheumatoid synovial tissue were fused with F3B6 humardmouse
heterohybridoma cells using the plate fusion technique as
described (7,ll). RSC and F3B6 cells were mixed and
seeded in 6-well plates; the cells were then fused with 2 ml of
40% PEG. The fused cells were seeded into 96-well plates.
Growing hybrids were screened for IgM-RF production, and
the positive cells were subcloned repeatedly by limiting
dilution.
Eventually, we obtained 5 IgM mRFs from 2 unrelated RA patients; the monoclonal nature of the mRFs was
evidenced by the fact that each expressed only the p heavy
chain and a single light chain isotype (4 A and 1 K ) . RFs B42,
C93, and D53 were from patient NIK, and RFs H4 and H6
were from patient REB. Four mRFs (B42, D53, H4, and H6)
were monospecilk for IgG, since none reacted with any other
antigens tested (including histone, bovine serum albumin,
thyroglobulin, single-stranded DNA [ssDNA] and doublestranded DNA [dsDNA], collagen, and ovalbumin; data not
shown). However, one mRF (C93) was reactive with IgG,
thyroglobulin, and human ssDNA and dsDNA (data not
shown).
Oligonucleotide primer design. Oligonucleotide polymerase chain reaction (PCR) primers were designed as
described previously (7). Briefly, the 5’ primers corresponded to the consensus sequences in the leader regions of
human H and L chains. The H-chain leader sequences were
divided into 3 major groups, and a mixed primer of 27 bases
86 1
(beginning at the initiator methionine codon ATG) for the
consensus sequence in each group was synthesized; they
were designated 5‘ Eco RI Sign.HH1, 5’ Eco RI Sign.HH2,
and 5’ Eco RI Sign.HH3 (7). Additionally, we synthesized a
new 27-base 5’ primer beginning at the first framework
region of VH, 5’ Eco RI/FRl, 5’-GGGAATTCAGGTGCAGCTG(CGT)(AT)G(CG)AGTC-3’.
The primer for the A
chain, 5’ Eco RI Sign.HL, was described previously (7); the
primer for the K chain was 5’ Eco RI Sign.HK, 5’-G-
GGAATTCATGGACATG(AG)(AG)(AG)(AGT)(CT)CC(ACT)(ACG)G-3’.
The 3’ primers were constructed from conserved
sequences within the constant regions of the L or H chains.
The primers for A and p chains, 3’ Hind III/HL constant and
3‘ Hind III/HH constant, respectively, were described previously (7); the primer for the K light chain was 3’ Hind
III/HK constant, 5’-CCAAGCTTCATCAGATGGCGGGAAGAT-3’. All primers were designed with Eco RI or
Hind I11 restriction endonuclease sites on their 5’ ends; an
additional 2 bases (G or C) were added outside the restriction
endonuclease site to improve enzyme digestion.
Amplification and cloning of the VL and VH complementary DNA (cDNA). Total RNA was prepared from 106
hybridoma cells by a microadaptation of the guanidinium
thiocyanate/cesium chloride procedure (12) and was used to
synthesize the first strand of cDNA using the Boehringer
Mannheim cDNA kit (13). Ten microliters of cDNA and 5 pl
of each primer (1 pM final concentration) were added to 80
pl of the PCR mixture. The mixture was amplified in a
thermal cycler for 35-40 cycles, with each cycle consisting of
melting at 94°C for 1 minute, annealing at 45-50°C for 1
minute, and extension at 72°C for 2 minutes.
DNA sequence analysis. Gel-purified PCR products
were digested with Eco RI and Hind I11 restriction enzymes
and ligated into pBluescript SK (M13+/-) sequencing vectors (14). Appropriate clones were isolated and sequenced
by the dideoxynucleotide chain-termination method. At least
2 independent SK clones of opposite orientation were used
to validate each sequence. The computer programs of the
Genetics Computer Group (15) were used to analyze the
sequence data.
RESULTS
Molecular characterization of the B42 V h and
VH cDNA. T h e amplified D N A was cloned into the
pBluescript sequencing vector, and the suitable recombinants were identified and analyzed. T h e results
showed that the B42 RF was the product of VA1 and
VH3 genes, designated Humlalb42 and Humha3b42,
respectively.
H u m l a l b 4 2 was most homologous t o t h e
Humlv1042 germline VAl gene (16), sharing 98% homology in a 298-basepair overlapped region (Figure 1).
In this region, lalb42 differed from Humlv1042 by only
6 nucleotides, with 3 differences occurring in the
framework regions (FRs) and 3 differences occurring
ERMEL ET AL
862
18
20
CDRl
T I S S H G S S S N I G A G Y D V Q
ACCATCTCCTCGCATGGGAGCAGCTCCAACATCGGGGCAGGTTATGATGTACAA
~~
.
*C *T
35
34
*H
W Y Q Q L P A T A P K L L I Y G N S
TGGTACCAGCAGCTTCCAGCAACAGCCCCCAAACTCCTCATCTATGGTAACAGC
52
53
CDRZN R P S G V P D R F S G S K S G T S
70
AATCGGCCCTCAGGGGTCCCTGACCGATTCTCTGGCTCCAAGTCTG~ACCTCA
71
88
89
104
105
T V
ACCGTC
106
*****.
Figure 1. Nucleotide and amino acid sequences of the VA1 and Jh
gene segments of B42 monoclonal rheumatoid factor (mRF) (lalb42)
and the most homologous human Ig germline A L-chain V-region
gene (Humlv1042). Homology (dashes) is 98% in a 298-bp overlap; *
= positions where the germline amino acid sequence differs from
lalb42;/ = position of possible N region. *** = position of JA gene
segment. The complementarity-determining regions (CDRs) are
marked.
in the first complementarity-determining region
(CDR1). In addition, lalb42 used a JA2/3 gene, since
both Jh genes were identical in their coding regions
(Table 1 and Figure 1) (17,18). One nucleotide in the
CDR3 of lalb42 cannot be accounted for by either the
Vhl or the Jh2/3 genes, which suggests the existence
of either an N region (19) or a longer VA gene segment
(Figure 1).
Humha3b42 was identical to the VH26 germline
gene in a 275-bp overlap (20). The D gene-encoded
CDR3 of ha3b42 shared an identical stretch of 23 bp
with the DLR2 D gene (21) and the JH portion of
ha3b42 was the product of the JH4 gene (22). In
addition, the CDR3 of ha3b42 contained 5‘ and 3‘
N-regions that were 11 bp and 9 bp in length, respectively (19) (Figure 2A).
While there is no somatic mutation in the VHencoded region and the Jh-encoded region, la1b42
differs from the most homologous lv1042 germline
gene by 6 nucleotides. Thus lalb42 may be encoded by
an unknown VAl germline gene that is yet to be
identified. Alternatively, the observed difference in
lalb42 may represent somatic mutations.
Molecular characterization of C93 RF VA and
VH cDNA. Sequence analysis of the C93 RF showed
that it was the product of VA3 and VH3 genes,
designated Humla3c93 and Humha3c93, respectively.
Humla3c93 was most homologous to the VAIII. 1 germline gene (23), but with only 82% homology in a 288-bp
overlap. This suggests that la3c93 was encoded by
another VA3 germline gene, designated Humlv3c93
(Figure 3). In addition, la3c93 was 93% homologous to
the expressed Vh gene for the 1B8 human monoclonal
anti-HIV antibody directed against the transmembrane
segment gp41 surface antigen of the human immunodeficiency virus 1 (HIV-1) (24) (Figure 3).
Humla3c93 used a Jh2/3 gene, beginning at
position 292 (Table 1 and Figure 2B) (17,18). Assuming
the VA gene-encoded region ends at position 288, the
Table 1. Characterization of 5 IgM monoclonal rheumatoid factors (mRF) generated from 2
unrelated rheumatoid arthritis patients
Gene segment
mRF
Isotype
Specificity
VL
JL
VH
D
JH
B42
PA
Pan-IgG
JA2/3
JH4
PA
Pan-IgG*
DLR2
JH6
D53
PA
Pan-IgG
D21/9
JH2
H4
PA
IgGS
D2119
JH4
H6
PK
Pan-IgG
VH26
(VH3)
8-IB
(VH3)
VH4.41
(VH4)
VH26
(VH3)
HumhvlLlR
(VHI)
DLR2
c93
Humlv1042
(VA1)
Humlv3c93
(Vh3)t
Humlv3c93
(VWt
Humlv3c93
(Vh3)t
Humigkl8
(vK4)
DXP4
JH5
JA2/3
JAl
JA2/3
J K ~
* C93 reacted to IgG, thyroglobulin, and human single-stranded and double-stranded DNA.
t Humlv3c93 represents a newly identified VA3 germline gene encoding VA3b light chains.
$ H4 monospecific for rabbit IgG and had
no human IgG reactivity.
863
RF VA3 GENE
A
95
Humha3b42
103
designated Humla3d53 and Humha4d53, respectively.
Humla3d53 was identical to Humla3c93 and the lv3c93
--/ll///////l#######################////llll/*~*******
germline
gene (Figure 3). The remainder of la3d53
DLR2
used
a
JAl
gene (Table 1 and Figure 2B) (17,18).
W G Q G T L V T V S S
113
TGGGGCCAGGGAACCCTGGTCACCGTCTCCTCA
Humha4d53 was 99% homologous to the VH4
germline gene VH4.41 (26), deviating by only 1 doubleCDR3
107
V P R D S S G Y Y Y F D Y W G Q G T
base difference in a 274-bp overlap (Figure 5). These
GTCCCCCGCGATAGTAGTGGTTATTACTACTTTGACTACTGGGGCCAGGGAACC
- / / / / / / / / # # # # l # # # # # # # # # # # # # ~ ~
. ~ differences
. ~ ~ ~ ~ ~ .were
~ ~ ~
~ ~FR2
~ ~ region
~ ~ ~ ~and
* ~ resulted
. ~
two
in~the
in
D2119
JH 4
an amino acid substitution. The CDR3 of ha4d53 was
L V T V S S
113
CTCGTCACCGTCTCCTCA
the unmutated product of the 21/9 D gene (27) (Figure
********.*********
5). The JH region of ha4d53 was identical in sequence
to that of the JH2 gene (22) (Table 1 and Figure 5).
Two N regions were present in the 5' and 3' portions of
the CDR3 of ha4d53, totaling 6 bp and 8 bp in length,
CDR 3
Q S A D S S G T Y R V V F G G G T K
103
respectively (19) (Figure 5).
CAATCAGCAGACAGCAGTGGTACTTATCGGGTGGTATTCGGCGGAGGGACCAAG
CDR3
D M A W D C S G G S C Y D L I F D Y
102
GATATGGCGTGGGATTGTAGTGGTGGTAGCTGCTACGA~TGATATTTGACTAC
*tt**~***~*"+*.***+.*It++l*t+l**.",,,*".
95
Humha3h4
108
B
89
Humla3c93
,/l*'*'**~'***"'***+f*X*"
-
..
lx2/3
104
L T V
CTGACCGTC
106
*********
89
Humla3d53
18
CDR 3
Q S A D S S G T Y V F G T G T K V T
CAATCAGCAGACAGCAGTGGTACTTATGTCTTCGGPACTGGGACCAAGGTCACC
105
****t*******+t****t*****ttt
Jhl
106
188
108
V L G
GTCCTAGGT
***I***
89
Hmla3h4
CDR 3
Q S A D S S G T Y P Y V V F G G G T
CAATCAGCAGACAGCAGTGGTACTTATCCGTATGTGGTATTCGGCGGAGGGACC
////I/"**"***++*'**""**
19
37
38
55
56
73
102
5x213
103
K L T V L V
AAGCTGACCGTCCTAGTC
108
****+****.******
Figure 2. A, Nucleotide and amino acid sequences of the CDR3 and
the fourth framework region (FR4) of B42 mRF (ha3b42) and H4
mRF (ha3h4). / / / = position of N regions; ### = position of D
gene segment; *** = position of JH gene segment. The CDRs are
marked. B, Nucleotide and amino acid sequences of the terminal
portion of the A light-chain V regions with their associated JA gene
segments of la3c93, la3d53, and la3h4. / / / = position of possible N
regions; *** = position of J A gene segments. The CDRs are marked.
See Figure 1 for other definitions.
14
CDR3 of la3c93 would have an N region of 3 bp (19)
(Figure 2B). These 3 bp could also be explained by the
existence of a longer VA gene segment.
Among all the reported VH3 germline genes,
ha3c93 was most similar to 8-1B (25), with a homology
of 98% in a 272-bp overlap (Figure 4). The CDR3 of
ha3c93 had an identical stretch of 21 bp with the DLR2
D gene, and shared 22 of 23 bp overall (21) (Figure 4). In
addition, ha3c93 used a JH6 gene, sharing 57 of 58 bp
(Table 1 and Figure 4) (22). The data indicated that the
CDR3 of ha3c93 had a 4-bp N region (19) (Figure 4).
Molecular characterization of D53 RF VA and
VH cDNA. Genetic analysis of the D53 RF revealed
that it was the product of VA3 and VH4 genes,
T
I
S
G
V
Q
A
E
D
E
A
D
Y
Y
C
-
91
ACCATCAGTGGAGTCCAGGCAGAAGACGAGGCTGACTATTACTGTCAATCAGCA
92
97
Figure 3. Comparison of the nucleotide and amino acid sequences
of the VA3 germline gene Humlvc93 and of the VA gene segments of
C93 mRF (la3c93), D53 mRF (la3d53), H4 mRF (la3h4), and 1B8
human monoclonal anti-human immunodeficiency virus antibody
directed against the transmembrane gp41 surface antigen. Dashes
indicate homology. The CDRs are marked. See Figure 1 for other
definitions.
ERMEL ET AL
864
Molecular characterization of H4 RF VA and VH
cDNA. Molecular analysis of the H4 RF indicated that
it was the product of Vh3 and VH3 genes, designated
Humla3h4 and Humha3h4, respectively. Humla3h4
was identical to Humla3c93, Humla3d53, and the
lv3c93 germline gene (Figure 3). Humla3h4 also used a
Jh2/3 gene identical to that of la3c93 (Table 1 and
Figure 2B) (17,18). This observation is interesting
since la3h4 and la3c93 were derived from different RA
patients. Six nucleotides in the CDR3 of la3h4 cannot
be accounted for by either the Vh3 or the JA2/3 genes,
which again suggests the possible existence of an N
region (19) or a longer Vh gene segment (Figure 2B).
Humha3h4 was identical to the VH26 VH3
germline gene in a 274-bp overlap (20) as was found for
ha3b42. In contrast to the latter, the CDR3 of ha3h4
was encoded by an unmutated 21/9 D gene (27) (Figure
2A). The JH region of ha3h4 was the product of an
unmutated JH4 gene (as was found for ha3b42) (Table
B G G G L V Q P G G S L R L S C A A S
Humha3c93 GGAGGAGGCTTGGTCCAGCCTGGGGGGTCCCTGAGACTCTCCTGTGCAGCCTCT
8-1B
B G P G L V K P S G T L S L T C A V S
Humha4d53 GGCCCAGGACTGGTGAAGCCTTCGGGGACCCTGTCCCTCACCTGCGCTGTCTCT
VH4.41
25
26
CDRl42
G G S I S S S N W W S W V R Q P P G
GGTGGCTCCATCAGCAGTAGTAACTGGTGGTGGAGTTGGGTCCGCCAGCCCCCAGGG
43
60
61
P S L K S R V T I S V D K S K N Q F
78
CCGTCCCTCAAGAGTCGAGTCACCATATCAGTAGACAAGTCCAAGAACCAGTTC
79
S L K L S S V T A A D T A V Y Y C A
TCCCTGAAGCTGAGCTCTGTGACCGCCGCGGACACGGCCGTGTATTACTGTGCG
94
93
CDR3
R A P R I T M I V V V I T T I D R Y
99
AGAGCTCCACGTATTACTATGATAGTAGTGGTTATTACTACGATCGATCGGTAC
Il//II1WX###WX###~~##~~~####1X~##X~##I/Illffl~~***
D21/9
_--_
25
100
F D L W G R G T L V T V S S
TTCGATCTCTGGGGCCGTGGCACCCTGGTCACTGTCTCCTCA
113
*tt***t******t*t*t**tt**.***t**ttt**~~,"***,*~~
JH2
26
44
62
43
CDR2
G L E W V S V I Y S G G S T Y Y A D
GGGCTGGAGTGGGTCTCAGTTATTTATAGCGGTGGTAGCACATACTACGCAGAC
61
Figure 5. Nucleotide and amino acid sequences of the VH4, D, and
JH gene segments of D53 mRF (ha4d53) and the most homologous
human Ig germline H-chain V-region gene (VH4.41). Homology
(dashes) is 99% in a 274-bp overlap. * = positions where the
germline amino acid sequence differs from ha4d53; / / / = position of
N regions; ### = position of D gene segment; *** = position of JH
gene segment. The CDRs are marked. See Figure 1 for other
definitions.
79
1 and Figure 2A) (22). In addition, the CDR3 of ha3h4
contained a 5' N region that was 8 bp in length (19)
(Figure 2A).
80
94
Molecular characterization of H6 RF VKand VH
cDNA. Sequence analysis of the H6 RF showed that it
was the product of V K and
~ VH1 genes, designated
CDR3
95 D G V Y C S G G S C Y Y Y Y G M D V
102
Humka4h6
and
Humhalh6,
respectively. Humka4h6
GATGGCGTATATTGTAGTGGTGGTAGCTGCTACTACTACTACGGTATGGACGTC
- - / / / / # G # # # X # Y # # # # # X # # # Y # # X # # # * ~ ~ ~ ~ ~ i ~ ~ ~ ~ ~ ~ * ~ * ~ ~ * ~was
* * * ~ ~W O
homologous in a 305-bp overlap to the HuDLRZ
JH6
migkl8 germline V K gene
~
(Figure 6A) (28). The
3 nucleotide differences (2 in FR1 and 1 in CDR3)
103
resulted in amino acid substitutions and a replacementto-silent ratio (R:S) of infinity. The remainder of ka4h6
Figure 4. Nucleotide and amino acid sequences of the VH3, D, and
used an unmutated J Kgene
~ sharing 36 of 36 bp (Table 1
JH gene segments of C93 mRF (ha3c93) and the most homologous
human Ig germline H-chain V-region gene (8-1B). Homology (dashand Figure 6A) (29). The CDR3 of ka4h6 contained a
es) is 98% in a 272-bp overlap. * = positions where the germline
2-bp N region (19,30) (Figure 6A).
amino acid sequence differs from ha3c93; / / / = position of N
Humhalh6 was 100% homologous to the
region; ### = position of D gene segment; *** = position of JH
HumhvlLlR germline gene in a 274-bp overlap (Figgene segment. The CDRs are marked. See Figure 1 for other
ure 6B) (31). The CDR3 of halh6 had an identical
definitions.
865
RF Vh3 GENE
A 1 E L V M T Q S P D S L A V 5 L C E R
H u m k a l h l GAGCTCGTGATGACCCAGTCTCCAGACTCCCTGGCTGTGTCTCTGGGCGAGAGG
Hurnig*18
.
.
CR.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
‘U * I
1
9
A
T
I
N
C
K
S
S
Q
S
V
L
CDRl
S S
Y
N
N
18
K
30
GCCACCATCRiiCTGCAAGTCCAGCCAGAGTGTTTTATACAGCTCC~CAATAAC
3
1
N
Y
L
A
W
Y
Q
Q
K
P
G
Q
P
P
K
L
L
I
48
AACTACTTAGCTTGGTACC~==AGA~CCAGGACAGCCTCCTAAGCTGCTCATT
CDRZ-
4
9
Y
m
T
R
E
S
G
V
P
D
R
f
S
G
S
G
66
TACTGGGCATCTACCCGGGAATCCGGGGTCCCTGACCGATTCAGTGGCAGCGGG
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6
1
S
G
T
O
F
T
L
T
I
S
S
L
Q
A
E
D
V
A
84
TCTGGGACAGATTTCACTCTCACCATCAGCAGCCTGCAGGCTGAAGATGTGGCA
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
8
5
CUR3
V Y Y C Q Q Y Y C T P P V T F G P G
102
GTTTAITACTGTCAGCAATATTATTGTACTCCTCCGGTCACTTTCGGCCCTGGG
......~......~----~~~-~~~~~-------//I....//~~~~..**~,****~~,
‘5
Jk3
T K V D I K R
ACCARAGIGGITATCAAACGA
103
LO9
...................
25
Humhalh6
GGGGCTGAGGTGAAGAAGCCTGGGGCCTC~GTG~GGTCTCCTGC~GGCTTCT
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Humh”lI,lR
26
CDRl43
G
Y
1 i T E
Y M H
W
V G P A P G 0
GGATACACCTTCACCGGCTACTATATGCACTGGGTGCGACAGGCCCCTGGACAA
44
G
L
E
W
M
G
R
I
G
P
N
S
G
G
CUR2T N
60
Y
A
GGGCTTGAGTCGATGGGACGGATC~~CCCTAACAGTGGTGGCACA~C~~TGCA
61
78
~
X
F
Q
G
R
V
T
M
T
R
D
T
S
I
S
T
A
CAGAAGTTTCAGGGCAGGGTCACCATGACCAGGGACACGTCCATCAGCACAGCC
93
9
4
100
CDR3
R A P S P R l T I F G V P R N N W F
AGAGCCCCTAGCCCCCGTATTACGATTTTTGGAGTGCCCAG~AC~CTGGTTC
. .
//~
/ / / / I I~
II//OY1#IYItXl)ltl)))l(l//l///~**~*******
DXPl
101
W
G 0 G T L V T V S 5 G S U P P
118
GACCCCTGGGGCCAG~GAACCCTGGTCACCGTCTCCTCAGGGAGTGCATCCGCCC
........................................
JH5
Figure 6. A, Nucleotide and amino acid sequences of the V Kand
~
J K gene segments of H6 mRF (ka4h6) and the most homologous
human Ig germline L-chain V-region gene (Humigkll). Homology
(dashes) is 99% in a 305-bp overlap. * = positions where the
germline amino acid sequence differs from ka4h6; / / = position of N
region; *** = position of JK gene segment; The CDRs are marked.
B, Nucleotide and amino acid sequences of the VHl, D, and JH
gene segments of H6 mRF (halh6) and the most homologous human
Ig germline H-chain V-region gene (HumhvlLIR). Homology (dashes) is 99%in a 274-bp overlap. / / / = position of N regions; ### =
position of D gene segment; *** = position of JH gene segment. The
CDRs are marked. See Figure I for other definitions.
stretch of 20 bp with the DXP4 gene (21) (Figure 6B).
In addition, halh6 used a JH5 gene with an identical
stretch of 51 bp (Table 1 and Figure 6B) (22). Two N
regions were present in the 5’ and 3’ portions of the
CDR3 of halh6, totaling 12 bp and 7 bp in length,
respectively (19) (Figure 6B).
DISCUSSION
In our continuing effort to define the genetic
basis of potentially pathogenic RFs in rheumatoid
synovium, we have now characterized the H and L
chain V regions of 5 new RSC-derived monoclonal
IgM RFs from 2 unrelated RA patients (see Table 1).
Notably, the V gene segments that encoded the light
chain regions of la3c93, la3d53 and la3h4 were completely identical, but each had a different VJh sequence (Figure 4). The data suggest very strongly that
these L chains are encoded by a single germline VA3
gene; the putative germline gene is designated
Humlv3c93, and has a nucleotide sequence identical to
the Vh gene-encoded region of la3c93. It should be
noted that the L chains of these 3 RFs were classified
serologically as the Vh3b L chain (data not shown).
Although lv3c93 is employed by 3 of the 5 RFs
analyzed in these studies, it has not been found in any
of the other 5 RSC-derived RFs from other unrelated
RA patients previously analyzed in our laboratory or
in any RA-derived RFs reported by other investigators. Because the total numbers of RA patients and the
RA-derived RFs analyzed are small, the available data
should be interpreted with caution. However, the data
may suggest that lv3c93 is polymorphic (present or
absent) in humans.
While the L-chain V regions of C93, D53, and
H4 are encoded by a single Vh gene, each uses a
distinct VH gene (Table 1). On the other hand, the H
chain V regions of B42 and H4 share the same VH26
germline gene, but each employs a different L chain V
gene (Table 1). The findings are consistent with numerous examples of loose pairing of autoantibodyrelated H- and L-chain V genes (8,9) and provide
additional genetic and structural basis for the reported
“connectivity” among many autoantibodies.
The fact that all 3 lv3c93-derived RF L chains
are identical to lv3c93 suggests very strongly that
lv3c93 may possess some intrinsic binding property
toward the IgG Fc region. This contention is consistent with the lack of antigen-driven mutation in all 3
lv3c93-encoded RFs. Moreover, lv3c93 has not been
found to encode any natural (or physiologic) RFs and
autoantibodies in normal healthy subjects, suggesting
that lv3c93 may encode only the disease-specific RFs
in RA patients.
Generally, random mutation in a neutral envi-
ERMEL ET AL
866
Table 2. Human Ig V genes that encode IgM rheumatoid factors in patients with rheumatoid arthritis
VH
Germline
HumhvlOSl
HumhvlflO
HumhvLl
HumhvlLlR
GL-SJ2
1.9111
VH26
ha3c6 gene
8-1B
VH4.18
VH4.21
Vll-2
v79
VH4.41
VK
Ref.
VHl
VHl
VHl
VH3
VH3
VH3
VH3
VH3
VH4
VH4
VH4
VH4
VH4
39
7
31
*
8,lO
10
10
8
*
VA
Germline
Hk102
Vd
vK1
VKI
A23
Humkv32S
Humkv328
Vg
Humigkll
V K ~
V K ~
V K ~
v K3
vK4
Ref.
Germline
8
40
Humlvll7
HumlvlLl
VAl
Vhl
39
31,42
40
40
40
8
Humlv1042
Humlv3c93
Humlv318
Humlv418
Humla8flO
Vhl
Vh3
Vh3
VM
Vh8
*
*
*
Ref.
8
42
7
36
41,42
10
42
*
* Data are from the present report.
ronment without selection pressure would lead to an
R:S ratio of 2.9, while a ratio higher than 2.9 would
indicate positive selection (8,32). Thus, the high R:S
ratio found in ka4h6 suggests that H6 arises from an
antigen-driven response in the rheumatoid synovium,
resulting in a somatically mutated RF with higher
avidity toward antigen that has been selected and
expanded preferentially (8). It should be noted that
high R:S ratios were also observed in the FRs of
ka4h6, indicating that the mutated amino acid residues
in these FRs may contribute to higher avidity.
In addition to the many RA-derived IgM RFs,
several RA-derived monoclonal IgG RFs have been
generated and characterized recently (3 1,33-35). The
results from these studies indicate that most IgG RFs
contain more somatic mutations than do the IgM RFs,
and apparently derive from an antigen-driven response. However, several IgG RFs employ Ig V genes
similar to those of the IgM RFs. For example,
humhvlLl encodes the H chains of the L1 IgG and the
mAb111 IgM RFs (31,36); humhvlflO encodes the H
chains of the C1 IgG and the HAFlO IgM RFs (7,37);
and hvlLlR encodes the H chains of TS7 IgG and the
H6 IgM RFs (31,38). In contrast, a few V genes are
found to encode only IgM or IgG RFs. The implications of these data cannot be precisely determined
because of the smaller number of IgG RFs that have
been analyzed.
In the CDR3 of the 5 presently analyzed RF H
chains, each contained a stretch of 18-31 bp that was
identical to a known D gene (Table 1). Two of the H
chains used DLR2 and two 2119. While the 2119 gene
was used in 2 different reading frames in ha4d53 and
ha3h4, the DLR2 gene was translated in the same
reading frame in ha3b42 and ha3c93. These data suggest that the DLR2-encoded amino acid residues (CSGGSCY) may contribute to binding to IgG. This
conclusion is supported by the occurrence of the same
heptapeptide in the CDR3 of 2 RA-derived monospecific RFs, RF-SJ1, and RF-SJ2 (10). In the future, it
will be important to examine experimentally the role of
these 7 amino acid residues in IgG-binding activity of
the 4 involved RFs.
In summary, the A light chains of 3 mRFs ((293,
D53, and H4) were encoded by a novel VA3 germline
gene, designated as Humlv3c93. Humlv3c93 has not
been found to encode any natural (or physiologic) RFs
or other type of autoantibody, which indicates that
lv3c93 may be a “pathogenic” RF-related Ig V gene,
instead of a common autoantibody-related V gene that
may serve some physiologic function. Moreover,
Humlv3c93 can be distinguished by an intrinsic binding property toward the IgG Fc region and the ability
to interact with high affinity with IgG molecules. The
identification of Humlv3c93 and the 5 additional
IgM RF-related V germline genes (Humlvl042-VA1,
Humigk18-V~4, Humhv 1L1R-VH1 , 8-1B-VH3, and
VH4.41-VH4) brings the number of currently identified IgM human RF-related V genes in RA patients to
13 VH genes, 7 V K genes, and 7 VA genes (Table 2).
Finally, the extent of somatic mutation in the light
chains of 2 RSC-derived mRFs (B42 and H6) are
suggestive of an antigen-driven response in the rheumatoid synovium; presumably mutated RFs that have
a higher avidity for antigen are selected and expanded
preferentially.
RF Vh3 GENE
ACKNOWLEDGMENTS
T h e excellent secretarial assistance of Theresa Andreozzi and the continued assistance of our orthopaedic
colleagues are gratefully acknowledged.
REFERENCES
1. Carson DA: Rheumatoid factor, Textbook of Rheumatology.
Third edition. Edited by WN Kelley, ED Harris Jr, S Ruddy,
CB Sledge. Philadelphia, WB Saunders, 1989
2. Munthe E, Natvig JB: Immunoglobulin classes, subclasses and
complexes of IgG rheumatoid factor in rheumatoid plasma cells.
Clin Exp Immunol 123-70, 1972
3. Munthe E, Natvig JB: Complement fixing intracellular complexes of IgG rheumatoid factor in rheumatoid plasma cells.
Scand J Immunol 1:217-229, 1972
4. Zvaifler NJ: The immunopathology of joint inflammation in
rheumatoid arthritis. Adv Immunol 16:265-336, 1973
5. Vaughan JH, Chihara T, Moore TL, Robbins DL, Tanimoto K,
Johnson JS, McMillan R: Rheumatoid factor producing cells
detected by direct hemolytic plaque assay. J Clin Invest 58:933941, 1976
6. Robbins DL, Skilling J, Benisek WF, Wistar R Jr: Estimation of
the relative avidity of 19s IgM rheumatoid factor secreted by
rheumatoid synovial cells for human IgG subclasses. Arthritis
Rheum 29:722-729, 1986
7. Robbins DL, Kenny TP, Coloma MJ, Gavilondo-Cowley J,
Soto-Gil R, Chen PP, Lamck JW: Serologic and molecular
characterization of a human monoclonal rheumatoid factor
derived from rheumatoid synovial cells. Arthritis Rheum 33:
1188-1 195, 1990
8. Ermel RW, Kenny TP, Chen PP, Robbins DL: Molecular
analysis of rheumatoid factors derived from rheumatoid synovium suggests an antigen-driven response in inflamed joints.
Arthritis Rheum 36:380-388, 1993
9. Pascual V, Victor K, Randen I, Thompson K, Steinitz M, Forre
0, Fu SM, Natvig JB, Capra JD: Nucleotide sequence analysis
of rheumatoid factors and polyreactive antibodies derived from
patients with rheumatoid arthritis reveals diverse use of V, and
V, gene segments and extensive variability in CDR-3. Scand J
Immunol 36:349-362, 1992
10. Pascual V, Randen I, Thompson K, Sioud M, Forre 0, Natvig
J, Capra JD: The complete nucleotide sequences of the heavy
chain variable regions of six monospecific rheumatoid factors
derived from Epstein-Barr virus-transformed B cells isolated
from the synovial tissue of patients with rheumatoid arthritis. J
Clin Invest 86:1320-1328, I990
11. Robbins D, Kenny T, Lanick J, Wistar R Jr: Production of
human monoclonal rheumatoid factor secreting hybridomas
derived from rheumatoid synovial cells. Med Sci Res 17:157159, 1989
12. Rappolee DA, Brenner CA, Schultz R, Mark D, Werb Z:
Developmental expression of PDGF, TGF-a, and TGF-P genes
in preimplantation embryos. Science 241: 1823-1825, 1988
13. Larrick JW, Danielsson L, Brenner CA, Abrahamson M, Fry
KE, Borrebaeck CAK: Rapid cloning of rearranged immunoglobulin genes from human hybridoma cells using mixed primers
and the polymerase chain reaction. Biochem Biophys Res
Commun 160:1250-1256, 1989
14. Short JM, Fernandez JM, Sorge JA, Huse WD: Lambda ZAP:
a bacteriophage lambda expression vector with in vivo excision
properties. Nucleic Acids Res 16:7583-7600, 1988
15. Devereaux J, Haeberli P, Smithies 0: A comprehensive set of
sequence analysis programs for the VAX. Nucleic Acids Res
12:387-395, 1984
867
16. Daley M, Olee T, Peng H, Soto-Gil R, Chen P, Siminovitch K:
Molecular characterization of the human immunoglobulin VAI
germline gene repertoire. Mol Immunol29: 1031-1042, 1992
17. Dariavach P, LeFranc G, LeFranc M-P: Human immunoglobulin Clambda6 genes encodes the Kern+02 lambda chain and
Clambda4 and Clambda5 are pseudogenes. Proc Natl Acad Sci
U S A 84:9074-9078, 1987
18. Udey JA, Blomberg B: Human lambda light chain locus: organization and DNA sequences of three genomic J regions.
Immunogenetics 25:63-70, 1987
19. Alt FW, Blackwell TK, Yancopoulos GD: Development of the
primary antibody repertoire. Science 238: 1079-1087, 1987
20. Matthyssens G, Rabbitts TH: Structure and multiplicity of
genes for the human immunoglobulin heavy chain variable
region. Proc Natl Acad Sci U S A 77:65614565, 1980
21. Yamada M, Wasserman R, Reichard BA, Shane S, Caton AJ,
Rovera G: Preferential utilization of specific immunoglobulin
heavy chain diversity and joining segments in adult human
peripheral blood B lymphocytes. J Exp Med 173:395-407, 1991
22. Pascual V, Capra JD: Human immunoglobulin heavy-chain
variable region genes: organization, polymorphism, and expression. Adv Immunol 49: 1-74, 1991
23. Combriato G, Klobeck HG: VA and JA-CA gene segments of the
human immunoglobulin A light chain locus are separated by 14
kB and rearranged by a deletion mechanism. Eur J Immunol
21:1513-1522, 1991
24. Larrick JW, Danielsson L, Brenner CA, Wallace EF, Abrahamson M, Fry KE, Borrebaeck CAK: Polymerase chain reaction
using mixed primers: cloning of human monoclonal antibody
variable region genes from single hybridoma cells. Biotechnology 7:934-938, 1989
25. Berman JE, Mellis SJ, Pollock R, Smith CL, Suh H, Heinke B,
Kowal C, Surti U, Chess L, Cantor CR: Content and organization of the human Ig VH locus: definition of three new VH
families and linkage to the Ig CH locus. EMBO J 7:727-738,
1988
26. Weng N, Snyder J, Yu-Lee L, Marcus D: Polymorphism of
human immunoglobulin VH4 germ-line genes. Eur J Immunol
22:1075-1082, 1992
27. Baluwela L, Albertson DG, Shemngton P, Rabbitts PH, Spurr
N, Rabbitts TH: The use of chromosomal translocations to
study human immunoglobulin gene organization: mapping DH
segments within 35 kb of the C mu gene and identification of a
new DH locus. EMBO J 7:2003-2010, 1988
28. Klobeck HG, Bornkamm GW, Combriato G, Mocikat R,
Pohlenz HD, Zachau HG: Subgroup IV of human immunoglobulin K light chains is encoded by a single germline gene. Nucleic
Acids Res 13:65154529, 1985
29. Martin T, Blaison G, Levallois H, Pasquali JL: Molecular
analysis of the VKIII-JKjunctional diversity of polyclonal rheumatoid factors during rheumatoid arthritis frequently reveals
addition. Eur J Immunol22:1773-1779, 1992
30. Heller M, Owens JD, Mushinski JF, Rudikoff S: Amino acids at
the site of Vk-Jk recombination not encoded by germline
sequences. J Exp Med 166:637-646, 1987
31. Olee T, Lu E, Huang D, Soto-Gil R, Deftos M, Kozin F, Carson
D, Chen P Genetic analysis of self-associating immunoglobulin
G rheumatoid factors from two rheumatoid synovia implicates
an antigen-driven response. J Exp Med 17k831-842, 1992
32. Shlomchik MJ, Marshak-Rothstein A, Wolfowicz CB,
Rothstein TL, Weigert MG: The role of clonal selection and
somatic mutation in autoimmunity. Nature 328:805-811, 1987
33. Lu EW, Deftos M, Tighe H, Carson DA, Chen PP: Generation
and characterization of two monoclonal self-associating IgG
rheumatoid factors from a rheumatoid synovium. Arthritis
Rheum 35:lOl-105, 1992
ERMEL ET AL
34. Newkirk MM: Complete amino acid sequence of the variable
region of two highly related heavy and light chains of human IgG
(SFL) and IgM (RIV) rheumatoid factors. Scand J Immunol
35:651-658, 1992
35. Kaplan S, Hyman K, Brooks R, Wakai M, Hashimoto S, Furie
R, Chiorazzi N: Monoclonal IgM, IgG, and IgA human rheumatoid factors produced by synovial tissue derived, EBVtransformed B cell lines. Clin Immunol Immunopathol66: 18-25,
1993
36. Mantovani L, Wilder RL, Casali P: Human rheumatoid B-la
(CDS +B) cells make somatically hypermutated high affinity IgM
rheumatoid factors. J Immunol 151:473488, 1993
37. Lu EW, Deftos M, Olee T, Huang D-F, Soto-Gil RW, Carson
DA, Chen PP: Generation and molecular analyses of two
rheumatoid synovial fluid-derived IgG rheumatoid factors. Arthritis Rheum 36:927-937, 1993
38. Randen I, Pascual V, Victor K, Thompson KM, Forre 0, Capra
JD, Natvig JB: Synovial IgG rheumatoid factors show evidence
of an antigen-driven immune response and a shift in the V gene
39.
40.
41.
42.
repertoire compared to IgM rheumatoid factors. Eur J Immunol
23:1220-1225, 1993
Ezaki I, Kanda H, Sakai K, Fukui N, Shingu M, Nobunaga M,
Watanabe T: Restricted diversity of the variable region nucleotide sequences of the heavy and light chains of a human
rheumatoid factor. Arthritis Rheum 34:343-350, 1991
Victor KD, Randen I, Thompson K, Forre 0, Natvig JB, Fu
SM, Capra JC: Rheumatoid factors isolated from patients with
autoimmune disorders are derived from germline genes distinct
from those encoding the Wa, Po, and Bla cross-reactive idiotypes. J Clin Invest 87:1604-1613, 1991
Pascual V, Capra JD: V,4-21, a human V, gene segment
overrepresented in the autoimmune repertoire. Arthritis Rheum
35:Il-18, 1992
Harindranath N, Goldfarb IS, Ikematsu H, Burastero SE,
Wilder RL, Notkins AL, Casali P: Complete sequence of the
gene encoding the VH and VL regions of low- and high-affinity
monoclonal IgM and IgAl rheumatoid factors produced by
CD5+ B cells from a rheumatoid arthritis patient. Int Immunol
3:865-875, 1991
Документ
Категория
Без категории
Просмотров
3
Размер файла
807 Кб
Теги
factors, patients, utilization, monoclonal, arthritis, genes, novem, synovial, derived, rheumatoid, cells, preferential
1/--страниц
Пожаловаться на содержимое документа