close

Вход

Забыли?

вход по аккаунту

?

Hla-dr alleles with naturally occurring amino acid substitutions and risk for development of rheumatoid arthritis.

код для вставкиСкачать
939
HLA-DR ALLELES WITH NATURALLY OCCURRING
AMINO ACID SUBSTITUTIONS AND RISK FOR
DEVELOPMENT OF RHEUMATOID ARTHRITIS
XIAOJIANG GAO, NANCY J. OLSEN, THEODORE PINCUS, and PETER STASTNY
To determine the HLA-DR4 subtypes associated
with rheumatoid arthritis (RA), we performed amplification of DR4 DRBl genes by the polymerase chain
reaction and dot-blots with oligonucleotide probes. In 52
HLA-DR4+ RA patients, Dw4 was the predominant
subtype. This subtype was found in 45 of 52 patients
(86.596) compared with 33 of 59 DR4+ controls (55.9%;
P C 0.001). In the whole population, Dw4 also gave the
highest relative risk for RA (RR = 5.31). Relative risk
was also associated with DR1.1, the common white DRI
(Dwl) type, which has a third hypervariable region
amino acid sequence similar to some forms of DR4 and
has glycine at position 86. Variants of DR1 (DR1.2) or
DR4 (Dw13.1, Dw14.1) with valine at position 86 a p
peared less able to confer risk for RA. Substitution of
residues in the third hypervariable region of the first
domain of DRBl appeared to correlate with relative risk
for RA. Among subjects having 0-1 amino acid substitutions, RA developed in 53%, whereas in subjects with
From the Department of Internal Medicine, The University
of Texas Southwestern Medical Center, Dallas, and the Division of
Rheumatology and Immunology. Vanderbilt University School of
Medicine, Nashville, Tennessee.
Supported in part by N I H grants PO-AI-23271. RON-21278. and P50-AR-39169, and the Texas Higher Education
Coordinating Board Advanced Technology Program.
Xiaojiang Gao. MD: Fellow in Internal Medicine. The
University of Texas Southwestern Medical Center; Nancy J. Olsen,
MD: Assistant Professor of Medicine, Division of Rheumatology
and Immunology. Vanderbilt University School of Medicine; Theodore Pincus, MD: Chief, Division of Rheumatology and Immunology, Vanderbilt University School of Medicine; Peter Stastny, MD:
Professor of Internal Medicine, The University of Texas Southwestern Medical Center.
Address reprint requests to Peter Stastny. MD. Department
of Internal Medicine, The University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75235.
Submitted for publication December 6, 1989; accepted in
revised form January 25, 1990.
Arthritis and Rheumatism, Vol. 33, No. 7 (July 1990)
2-4 amino acid changes, RA was present in only 17.4%
(P < 0.ooOOl). DQw7 (formerly DQw3.1) was slightly
increased in DR4+ RA patients compared with controls,
but a striking excess of Dw4,DQw7 homozygous patients
was observed. The results suggest that DQw7 may have
an additional effect, possibly with a recessive mechanism, since it was observed only in DR4 homozygous
patients.
HLA-DR4 is associated with risk for development of rheumatoid arthritis (RA) (1-7). Subsets of
HLA-DR4 were recognized initially using mixed lymphocyte cultures (8) and, more recently, using oligonucleotide probes (9,lO). The HLA-DR region is known
to contain 1 DRA (previously DRa) gene and 1-3 DRB
(previously DRP) genes (11). In HLA-DR4 haplotypes, there are usually 3 DRB genes, of which DRBl
encodes the molecules that carry the determinants that
characterize DR4 and its subtypes, DRB2 is a
pseudogene that is not expressed, and DRB4 encodes
a molecule that has the serologic determinants known
as DRw53 (12). The HLA-DR4 subsets differ by only
1-3 amino acids in the third hypervariable region of the
first domain of the DRBl molecule (12,13). These
differences are not detectable serologically.
An association of HLA-DRl with risk for development of RA has also been found, by Bardin et al
(14) and by our group (6). DRI was reported to have
shared epitopes with DR4 recognized by B cells (15)
and by T cell clones in cells from normal subjects and
patients with RA (16,17). A variant of HLA-DRl
(Dw1) was recognized by mixed lymphocyte culture as
not being Dwl (18) and more recently, has been
identified as Dw20 ( 1 1). DNA sequencing results have
shown that the variant of DRI from an American black
940
GAO ET AL
donor was different from DRl (Dwl) only at codons 85
and 86 (18).
Naturally occurring variants, including 3 new
variants (Dw13.2, Dw14.2, and DRCnew) that have
been described in more detail elsewhere (19) and that
contain substitutions in various amino acids from
positions 67-86, were available to evaluate their effect
on the risk of developing RA. Determination of these
sequences in a population of RA patients and controls
was performed by the polymerase chain reaction
(PCR), followed by hybridization with oligonucleotide
probes (19). Risk for development Of RA was found to
be associated with DR4 (Dw4) and with 3 Other HLADR alleles that have only 1 amino acid substitution in
the region from residues 67-86 of the first domain of
DRB 1. Other alleles with 2 4 residue differences from
the DR4 (Dw4) sequence appeared to lack the
to confer risk for development of RA.
PATIENTS AND METHODS
Patients. Fifty-two unrelated, white, HLA-DR4+
patients with RA were selected for this study from a group of
82 patients. The patients were classified as having definite or
classic RA by the American Rheumatism Association criteria (20). Fifty-nine HLA-DR4+ white individuals who were
randomly selected from 177 members of the tissue typing
laboratory reference panel served as healthy controls.
H L A typing. All subjects were typed for HLA-A. B,
C, DR, and DQ antigens using standard methods as previously described (21). Some of the typing sera were from our
laboratory collection, and others were obtained through exchange with other investigators. The definition of all of the
antigens was based on panel typings with reagents and criteria
of the Tenth International Histocompatibility Workshop.
Preparation of genomic DNA. DNA was prepared
from peripheral blood leukocytes using a standard procedure
as previously described (22). The cells were washed with
Tris-EDTA buffer, pH 8.0, lysed with cold lysis buffer (320
mM sucrose, 10 mM Tris HCI, pH 7.4, 1% Triton X-100.5
mM MgC13, treated with proteinase K (0.1 mg/ml) at 37°C
overnight, and then extracted with phenol, phenol/
chloroform, and chloroform, dialyzed, and precipitated with
ethanol.
PCR primers. In order to perform allele-specific
amplification, we have designed a primer homologous to
codons 6 1 3 of the first hypervariable region of DRBl of
DR4 (12.23). This primer, called PSP-25, consists of the
sequence S’-GTTTCTTGGAGCAGGTTAAAC-3’.
It is mismatched by 3-7 nucleotides with all other DRB genes,
including at least 2 of the last 3 nucleotides at the 3’ end of
the primer. The other primer used was GAMP DRB I , which
is complementary to the DNA sequence at the 3’ end of the
first domain of all DRBl genes including codons 87-94 (23);
its sequence is S’-GCCGCTGCACTGTGAAGCTCTC-3’.
For determination of the variants of HLA-DRI. we have
used a DRIB1-specific primer, PSP-43, which has the sequence 5’-TTGTGGCAGCTI’AAGTTTGAA-3’ and is homologous to the first hypervariable region of DRI B1.
Oligonucleotide probes. The probes used for the
variants of DR4 in these experiments consisted of oligomers
of 19 nucleotides covering the region between codons 68 and
75 (12). In this region, there are differences that distinguish
between Dw4, DwlO. Dw13, and Dw14. The probes used
have been described in detail (19). Briefly, PS-4 (Dw4) was
complementary to codons 68-74 and consists of the sequence S’-CGCGGCCCGCTTCTGCTCC-3’,
PS-5 (Dw 10)
was complementary to codons 68-74 and consists of the
sequence 5’-CGCGGCCCGCTCGTCTTcc-3’ ps-15
(Dw13) was complementary to codons 69-75 and has the
sequence ~,-CACCTCGGCCCGCCTCTGC-~,,and ps-6
( ~ ~ 1 Dw15)
4 , hybridized with codons 6P-75 and includes
the sequence S’-CACCGCGGCCCGCCTCTGC-3’.
In addition, we have used PS-7. which is composed of the sequence
5’-GTACTCGGCGCTAGGCCGC-3‘ and is homolo~ousto
the region between codons 54 and 60 of DR4BI(Dw15) (12),
PS-79, which hybridizes with codons 31-39 of DR4 (KT2)
and includes the sequence 5’-CACCAAGAGGAGTCCGTGC-3’ (24). PS-34. which hvbridizes with codons 83-89
of the recently’described varian; of DRI (Dw20) and has the
sequence 5’-CGGGGCTGTGGAGAGCTTC-3’,
PS-4 1,
which has the structure 5’-CTACGGGGTTGGTGAGAGC3’ between codons 82 and 88 and contains a glycine codon at
position 86, and PS-36. which is homologous to the sequence
between codons 83 and 89, has a valine codon at position 86,
and has the structure 5’-CGGGGTTGTGGAGAGCTC-3’.
Prior to use in hybridization experiments, the probes
were end-labeled with 32P.Briefly, 44 ng of probe, 0.33 mCi
of y‘2P-labeled adenosine 5’-triphosphate (7,000 Ci/mmole;
New England Nuclear, Boston, MA), 10 units of T, polynucleotide kinase (Fisher Scientific, Pittsburgh, PA), and 50
mM Tns. pH 7.6, 10 mM MgCI,. 5 mM dithiothrietol, 0.1
mM EDTA were mixed and incubated at 37°C for 1 hour. A
full-length DRB complementary DNA probe was used as a
control for hybridization to DRB.
PCR amplification. Genomic DNA ( I pg), PCK buffer
(50 mM KCI, 1.5 mM MgCI,, 10 mM Tris HCI, pH 8.3, 0.1
mg/ml of gelatin). 100 mM of dATP. dCTP, dGTP, and
d l T P , 0.5 mM of each primer, and double-distilled water (to
give a total volume of 100 pl) were mixed in a W p l
microfuge tube. The mixture was overlaid with 70 pI of
mineral oil to prevent evaporation. The DNA was denatured
by heating for 10 minutes at 94°C. and 2.5 units of Taq I
polymerase (Perkin-EImerKetus, Norwalk, CT) was added.
The amplification was accomplished by 30 cycles of incubation at 94°C for I minute, 59°C for 2 minutes, and 72°C for 3
minutes, using a thermal cycling machine designed for this
purpose (Perkin-ElmerKetus).
Dot-blot hybridization. Amplified DNA (20 ng/
sample) was boiled for 10 minutes and blotted onto nylon
membranes (Biotrace; Gelman Sciences, Ann Arbor. MI)
with the aid of a dot-blot manifold (Bethesda Research
Laboratories, Gaithersburg, MD). The filters were prehybridized at 55°C for 1 hour in 20 ml of a solution containing
1M NaCI, 1% sodium dodecyl sulfate (SDS). 10% dextran
sulfate, 25% formamide. and 0.2 mg/ml of boiled salmon
sperm DNA. ”P-labeled probe was added to the solution,
3
94 1
HLA-DR4 SUBTYPES IN RA
and hybridization was allowed to continue for 2 hours at
55°C. The filters were then washed in 5 x SSC (1 x SSC =
0.15M NaCI. O.Ol5M sodium citrate) and 1% SDS at 65°C for
30 minutes. Dots were visualized after exposure to x-ray film
(XAR-5; Kodak, Rochester, N Y ) with Kodak intensifier
screens for 1-8 hours at -70°C.
Determination of DQw3 subtypes. Genomic DNA was
digested with the restriction endonuclease Bgl 11. electrophoresed on a 0.8% agarose gel, blotted onto a nylon
membrane, and hybridized with a full-length DQB probe.
DQw7 was scored by the presence of a 4.2-kilobase band; a
6.0-kb band defined the presence of DQw4, DQw8. or
Pfl6BE
PS-4
. PS-5
PS-ls
PS-6
DQfl.
Statistical analysis. The significance of differences in
antigen frequencies was evaluated by chi-square test with
Yates' correction. Relative risks were calculated by the
method of Woolf (a x d/b x c), except for cases in which I
of the elements was 0, where the formula of Haldane [(2a
+1)(2d + 1)/(2b + 1)(2c + l ) ] was used (25).
RESULTS
Specific dot-blot hybridization. DNA extracted
from B cell lines defined at the Tenth International
Histocompatibility Workshop was used to determine
the specificity of typing with oligonucleotide probes.
The donors of these B cell lines were known to have
the 5 major subtypes of DR4, as determined by typing
with homozygous typing cells by mixed lymphocyte
culture. Five of these samples were homozygous for
both DR4 and a given subtype, and the other 3 were
homozygous for DR4 and heterozygous for a given
subtype. These latter samples had the genotypes
Dw4.Dwl4, DwlO.Dwl4. and Dw13.Dw14. respectively (Figure 1). The first 5 samples from homozygous
donors hybridized only to 1 oligomer having the
matched sequence. The other 3 samples from heterozygous donors were positive with both matched oligonucleotide probes. Probe PS-6 hybridized with both
Dw14+ and Dwl5+ samples because the sequences in
the third hypervariable regions are identical (12).
DNA from Dwl5 cell lines hybridized specifically with probe PS-7, which hybridizes to codons
54-60, where there is a sequence present in Dw15 that
is not found in any of the other DR4 subtypes. This
sequence is, however, also found in DRw8.1,, and
cross-hybridization was detected when generic DRB
DNA amplification was performed. HLA-DRl also
has an identical sequence to that of the probe for
Dw 14,Dw 15 in the third hypervariable region and
would also hybridize with probe PS-6. In addition,
reported sequences for some DR5 and DRw6 DRBl
genes are homologous to the DwlO probe (PS-5)
(12,23,26.27). To avoid these cross-hybridizations, we
F w r e 1. DNA typing for HLA-DK4 subsets by dot-blot hybridization with oligonucleotide probes. Samples 1-5 contain DNA from
donors who were homozygous for both HLA-DR4 and a given
subtype. and samples 6-8 are from donors who were homozygous
for HLA-DR4 and heterozygous for a given subtype. Genomic
DNA was extracted from well-characterized B cell lines and amplified with HLA-DR4. DRBI-specific primers. The sequences of the
probes used and the hybridization conditions are described in
Patients and Methods. These conditions permitted recognition of
differences of a single nucleotide.
have used primers that selectively amplify only the
DR4B1 genes when determining the subsets of DR4
and DRlBl for the analysis of the DRl variants. The
technical aspects of the procedure are discussed more
extensively in another report (19). Dw13.1 and Dw13.2
as well as Dw14.I and Dw14.2 represent polymorphisms at position 86. Dw13.2 and Dw14.2 are the
variants that have glycine at residue 86, instead of the
more common valine.
Two subjects (1 patient and 1 control) in this
study were found to have a DR4 subtype that has not
been previously described in white individuals. This
subtype of DR4, which we have called "new", hybridized with PS-6 (Dwl4,DwlS) and with PS-7 (Dwl5).
Serologically, DR4 was accompanied by DQw3
(DQw8) in both cases. A DK4 (Dw15) molecule associated with DQw3 was also observed recently in a
white individual by Celis and coworkers (28). Another
DR4 subtype, which was found in a Japanese population and named KT2. has 1 amino acid mismatch at
position 37 of the first domain with Dw13.1 (24). This
subtype of DR4 has never been reported in white
individuals and was not observed in the present
study.
HLA-DR4 subsets in RA patients. Fifty-two
patients with RA and 59 controls, all known to have
HLA-DR4 by serologic typing, were studied (Table 1).
Dw4 was enriched in the RA group, in which 86.5% of
942
GAO ET AL
Table 1. Phenotypic frequency of HLA-DR4 subtypes detected
by dot-blot hybridization with oligonucleotide probes in DR4 positive controls and rheumatoid arthritis patients’
DR4 allele
Controls
(n = 59)
Patients
(n = 52)
RR
P
Dw4
Dw 10
Dw13
Dwl3.l
Dw13.2
Dw14
Dw14. I
Dw14.2
New
33 (55.9)
5 (8.5)
10 (17.0)
6 (10.2)
4 (6.8)
17 (28.8)
17 (28.8)
0 (0)
l(1.7)
45 (86.5)
0 (0)
I(1.9)
I (1.9)
0 (0)
8 (15.4)
7 (13.5)
I(1.9)
I (1.9)
5.07
0.09
0.33
0.35
0.23
0.45
0.38
3.5
1.14
<O.OOI
NS
<0.05
NS
NS
<0.05
C0.05
NS
NS
PCO,,
<0.007
NS
NS
NS
NS
NS
NS
NS
NS
* Values are the number (%) of subjects. Relative risk (RR) was
obtained by the method of Woolf (25). except for DwlO, Dw13.2.
and Dw14.2. where the formula of Haldane (25) was used. P,,, is
the P value corrected for the number of antigens tested: NS = not
significant.
the DR4+ patients were found to have this subtype,
compared with 55.9% of the controls. This gave a
relative risk (RR) of 5.07. which was statistically
significant ( P < 0.001) even when P was multiplied by
7 to correct for the number of antigens tested (P,,, <
0.007).
All of the other subtypes were decreased in RA
patients, except for Dw14.2. None of the RA patients
in our study had DwlO. a specificity known to be
frequent among Jews, and which was present at a very
low frequency in our control group. Dw13 was observed in 17% of the control group and in only 1.9% of
the RA patients. Four control subjects and none of the
patients had Dw13.2. The frequency of Dw14.1 was
somewhat lower in the RA group than in controls (RR
= 0.38). while Dw14.2 was observed in only 1 patient
and in none of the normal controls. Thus, the Dw4
subtype was the only predominant DR4 type seen in
our group of RA patients. None of the other differences observed were statistically significant after correction of P values for the number of antigens tested
(Table 1).
When the antigen frequencies were calculated
for the population of RA patients and controls as a
whole (Table 2), Dw4 gave the highest relative risk for
RA (RR = 5.31). The association of RA with DR4 was
also highly significant ( P < 0.00001). while the differences in the frequencies for the other DR4 subsets
were not significant.
Typing of patients homozygous for HLA-DR4.
Among the RA patients, there were 12 who typed
positive only for DR4 and who appeared to be homozygous for DR4 by restriction fragment length
polymorphism analysis with a full-length DRB probe
(data not shown). Of these 12 patients, 10 showed a
positive hybridization reaction only with the Dw4specific oligomer (PS-4). One patient typed as
Dw4.Dw14 by a positive hybridization reaction with
probes PS-6 and PS-41, indicating that the subtype was
Dw14.2.
HLA-DR1 and the D R l variants. A small increase in the frequency of DRl was observed in R A
patients compared with controls (Table 2). DRI. 1,
which is the common form in white individuals and
which has glycine at position 86, was observed in
27.3% of RA patients compared with 15.5% of normal
controls (RR = 2.04. P < 0.05). DRl.2, the allele with
valine at position 86, was infrequent in both groups
and appeared to be more infrequent in the RA patients
compared with controls (RR = 0.32, P not significant).
DR1.l and DR1.2 together accounted for all the samples that were typed serologically as DR1. The antigen
DR “Br” (29). or DRl-Cetus (30). was not detected,
since it does not react with DRl sera. The independence of DRI from DR4 was investigated by analyzing
the frequency of DRI in DR4- RA patients and
controls (Table 2). Results of this analysis gave a
relative risk of 2.73 ( P < 0.01).
Possible effect of glycine at position 86. The
frequency of the Dw14 (Dw14.1) allele with valine at
position 86 was decreased in RA patients (RR = 0.741,
while Dw14.2, which has glycine at position 86, was
Table 2. Frequency (%) of HLA-DRI and DR4 subtypes in the
group of rheumatoid arthritis patients and normal controls studied as
a whole’
Allele
DR I
DR1.I
DRI.2
DRI (in
DR4-)
DR4
Dw4
DwlO
Dw13
Dw13.1
Dw13.2
Dw 14
Dw14.1
Dw14.2
New
Controls
(n = 177)t
Patients
(n = 82)$
19.2
15.5
3.5
18.9
28.3
27.3
33.3
18.6
2.8
5.7
3.4
2.3
9.6
9.6
0
0.56
RR
P
<0.05
<0.05
38.9
1.66
2.04
0.32
2.73
63.4
54.9
0
1.2
1.2
0
8.5
7.3
1.2
1.2
3.47
5.31
0. I9
0.21
0.35
0.23
0.88
0.74
4.36
2. I7
<o.m1
<0.00001
1.1
NS
<0.01
NS
NS
NS
NS
NS
NS
NS
NS
* See Table 1 for explanations and abbreviations.
t For DRI, n = 214; for DRI.1, n = 116; for DRI.2, n = 116; for
DRI (DR4-L n = 143.
t F o r D R 1 . n = 19l;forDRl.l,n=88;forDR1.2,n
=88:forDRI
(DR4-), n = 72.
943
HLA-DR4 SUBTYPES IN RA
Table 3. Percentages of DQw7 and DQw8 among DR4 haplotypes
in controls and rheumatoid arthritis patients"
Controls
Allele
DQw7
Patients
DQw8
DQw7
DQw8
DR4
Dw4
DwlO
Dw13.1
Dw13.2
Dw14.1
Dw14.2
New
* Values in parentheses are the number of haplotypes of each
subtype obtained by direct counting, where homozygous individuals
= 2 and heterozygous individuals = I .
found in 1 RA patient and in none of the controls. The
variant of DRI (DR1.2) with valine at position 86 was
also decreased in RA patients (RR = 0.32). An analysis of glycine at position 86 versus valine at position 86
in DR1 Bl and DR4 B1 showed a strong association of
RA with glycine at position 86 (RR = 4.0, P <
0.ooool).
Subsets of DQw3. HLA-DR4 in white haplotypes is virtually always associated with DQw3. DQw3
has been subdivided by biochemical and serologic
methods into DQw7, DQw8, and DQw9 (1 1). Only the
first 2 subsets are associated with DR4. In these
experiments, we have determined the DQw3 subtypes
by Southern blotting after digestion of genomic DNA
with Bgl 11 and hybridization with a full-length DQP
probe. The relationship between DQ and the DR4
subtypes was analyzed, and the results are shown in
Table 3. In RA patients, DQw7 (DR4) was associated
only with Dw4. The distribution of DQw7 and DQw8
among Dw4 haplotypes of patients was almost the
same as that of controls (Table 3). However, it was
quite different in DR4 homozygotes compared with
DR4 heterozygotes. In the 24 DR4 haplotypes from 12
DR4 homozygotes. 17 (70.8%) had DQw7, with only 1
patient being homozygous for DQw8. In contrast, only
14 of 40 Dw4 haplotypes (35%) from DR4 heterozygotes were associated with DQw7. The results also
showed that there was an excess of DQw7/DQw7
homozygotes (expected 2.7, observed 6) among the
patients who were homozygous for Dw4, which accounted for most of the DQw7 increase in the patients.
Overall, the increase of DQw7 among DR4+ R A
patients compared with DR4+ controls in our samples
was small (Table 3) and less than that reported by
other investigators (28.29).
DISCUSSION
In these experiments, when typing for DR4
subsets, we amplified the genes from genomic DNA
using primers designed to hybridize with a sequence in
the first hypervariable region of the first domain of
DRB 1 of the DR4 alleles. By performing allele-specific
amplification, one avoids the possibility of crosshybridization of the probes with other genes that share
sequences with the DR4 subtypes. For example, the
Dw14 probe (PS-6) also hybridizes to DRBl of DRI.
However, using the DR4-specific primers, only DR4
genes are amplified, and even in DRl,DR4 heterozygotes, hybridization to DRl is not detected. Similarly,
for the determination of polymorphism among DR1
genes, selective amplificationof DRI BI was performed.
Because most of the variations among the DR4
subsets occur in the third hypervariable region (amino
acid positions 67-74) (12), we have designed oligonucleotide probes to recognize substitutions in this region. Sequences that characterize Dw4, DwlO, Dw13,
and Dw14/Dwl5. respectively, were distinguished
with these probes. The method was found to be
specific and reproducible. The Dw15 variant, which is
the most common subtype of DR4 among Japanese
individuals (7). is identical to Dw 14 in the third hypervariable region. However, this variant has a substitution at residue 57 (serine instead of aspartic acid), and
this polymorphism was detectable with our probe
PS-7. In the present study, 2 subjects who were
initially typed as having Dwl4,DwlS were found to
have a serine substitution in position 57. This subtype
has been designated as "new" pending further studies.
In both subjects, the DR4 variant was associated with
DQw3.DQw8 and not with DQw4 as it was in the
Japanese individuals.
In this group of KA patients. Dw4 was the main
risk factor associated with susceptibility. In contrast.
the other DR4 subtypes were somewhat decreased
among R A patients, but none of the other differences
were statistically! significant when P values were corrected for the number of antigens tested.
Recently, a variant of DRl, which involves
amino acids at positions 85 and 86 and appears to
determine a T cell recognition determinant, has been
reported (18). In addition, polymorphism at position 86
occurs among some DR4 variants (19). We have
therefore examined DRI and DR4 alleles with probes
that recognize nukleotide sequences that encode either
glycine or valine iat residue 86. Interestingly, a statistically significant correlation was found between gly-
GAO ET AI,
944
Table 4. Correlation of amino acid substitutions in HLA-DR alleles with risk for rheumatoid arthritis (RA)
Amino acid position
Allele
Dw4
DRI.1
DwlS
Dw14.2
New7
Dw 14.1
Dw13.2
DRI.2
Dw13.1
DwlO
67
L
-
-
I
68
L
-
69
1:
-
70
71
72
73
74
85
86
Q
K
R
R
R
R
R
R
R
R
E
R
A
-
V
A
-
G
-
A
E
E
-
D
-
-
-
-
-
V
V
V
V
Number of
differences
KR for
I
I
I
I
2
2
3
3
5.3t
4
RA*
2.0$
3.25
4.4
2.2
0.7
0.2
0.3
0.4
0.2
Relative risk (RR) was calculated as in Table I . except for DwlS. where the data of Ohta et al (7) were used. Amino acid residues are
designated by the I-letter code; residues identical to the original sequence are indicated by dashes.
t P < o.oooo1.
t P < 0.05.
8 P < 0.0004 (ref. 7).
P The "new" allele was found to have a sequence identical to DwlS in the second hypervariable region (see Discussion).
cine at position 86 and the development of R A , suggesting that this residue contributes to the specificity
that confers the risk for RA.
We also examined the distribution of the DQw3
alleles DQw7 and DQw8 among DR4 RA patients and
controls because of reports suggesting that DQw7 was
a risk factor for R A (31,32). Various studies have
shown different results with respect to DQw7 in RA.
One group of investigators found DQw7 to be increased mainly in patients with Felty's syndrome (29),
and others observed an essentially normal distribution
of this allele (33). Our results do show a small increase
in DQw7 in DR4 haplotypes of RA patients compared
with controls (Table 3). The increase appears to be
explained by an excess of IIw4.DQw7 homozygotes.
A striking increase in DQw7 was found in the patients
who were homozygous for DR4, while in heterozygous
patients, the distribution of DQw7 and DQw8 was as
expected. The role of DQw7 homozygosity uncovered
by this analysis is not fully understood at present. The
distortion observed could indicate that DQw7 might play
an ancillary and recessive role in susceptibility for the
development of RA. In the whole patient group, DQw7
was only slightly increased, and the difference from that
found in controls was not statistically significant.
Our best interpretation is that DQw7 has an
independent recessive effect that is not required for,
but may contribute to, the DR effect on susceptibility
for the development of RA. Determinants that differentiate DQw7 from DQw8 are recognized by human T
cells (34), and thus it is likely that DQ-specific T cells
may also be involved. One possible explanation for the
increased frequency of DQw7 homozygotes could be
an inhibitory effect associated with the presence of
DQw8. Such an inhibitory effect could arise, for
example, if DQw8 had the capacity to stimulate suppressor cells capable of inhibiting the development of
RA. Functional studies would be required to investigate this possibility, but they are difficult to conduct
presently, since the antigen that elicits the immune
response that causes R A is presently not known.
The correlations shown in Table 4 suggest that
the maximal risk of developing R A is associated with
the amino acid sequence LLEQKRAA for positions
67-74 and VG for residues 85 and 86. Substitution of
lysine ( K ) by arginine (R) at position 71, which occurs
in DRI.1, Dwl5, D14.2, and "new," produced significant, but somewhat lower risk in the first 2 subtypes
and similar risks in the others, but the samples were
too small to achieve statistical significance in the
latter. One additional substitution of alanine (A) by
glutamic acid (E) at position 74 in Dw13.2. or of
glycine (G) by valine (V) at position 86 as seen in
Dw 14. I , markedly reduced the risk of developing RA.
Alleles with 3 substitutions (DR1.2 and Dw13.1) and 4
substitutions (DwlO) further decreased the risk of R A
development. When the groups were combined. subjects with alleles having 0 or I substitutions had a
much higher frequency of R A (53%) than those with
alleles having 2-4 substitutions (17.4%). and the difference between these groups was highly statistically
significant ( P < 0.00001).
The role of the LLEQKRAA-VG sequence in
determining susceptibility to R A is presently not understood. The region in which these substitutions
945
HLA-DR4 SUBTYPES IN RA
occur is known to form epitopes which are recognized
by T cells (8,12). A change of 2 residues at positions 85
and 86 is sufficient for T cells to differentiate between
Dwl and Dw20 (18). Lang et al (35) recently reported
experiments with a T cell clone that recognized a DR4
(Dw13) epitope including valine at position 86. It was
concluded that the substitution of glycine by valine at
residue 86 directly influenced T cell recognition
events. Our data suggest that positions 85 and 86 may
contribute to the effects of the broader region encompassing residues 67-74. More work will be required to
study the effect of these epitopes on T cells that
recognize peptides and T cells involved in immune
regulation, as well as their role in the development of
autoimmunity and RA. Because many of these alleles
are rare in the North American white population,
studies in other populations, in which these alleles are
more frequent, may provide useful additional information. Such studies are currently in progress in our
laboratory.
Several investigators have previously speculated about the role of the third hypervariable region of
DRBl in susceptibility to RA (36.37). on the basis of
sequence information derived from prototype cells.
The role of the shared sequences between Dw14 and
DRI has also been discussed (37.38). Perhaps in some
populations, Dw14.2 is more common than in the
population groups that we have studied. The contribution of the polymorphism at position 86 was not
previously recognized. Our results suggest that this
residue is important together with the LLEQKRAA
sequence, since the Dw 14. I allele did not confer risk
for RA (Table 4).
The correlation between risk and amino acid
sequences determined with oligonucleotide probes in
population studies is of considerable interest. Our
analysis shows that the substitution of 1 amino acid
from the prototype sequence found in Dw4 had little
effect. However, substitution of 2 or more of these
residues resulted in complete loss of the ability to
produce a susceptible phenotype for the development
of RA.
In conclusion, our results suggest that a dominant effect that determines susceptibility to RA is
associated with an epitope formed by amino acids.
including residue 86, of the third hypervariable region
of DRBI. in the DR alleles Dw4, DR1.l, Dw15.
Dw14.2, and "new" (Dw 15-like). An independent
recessive effect of DQw7 was suggested by the excess
number of DQw7 homozygotes observed. One explanation for the latter might be the possibility of circum-
venting a protective effect of DQw8. The relevant DR
and DQ epitopes are known to be 'recognized by T
cells, making it likely that they exert their effects
through an immune response that is critical to the
development of RA.
ACKNOWLEDGMENT
The authors gratefully acknowledge the assistance of
Chris Danielson in the preparation of the manuscript.
REFERENCES
I . Stastny P: Association of the Bcell alloantigen DRw4
with rheumatoid arthritis. N Engl J Med 298:869-871.
1978
2. Stastny P: Rheumatoid arthritis, Histocompatibility
Testing 1980. Edited by PI Terasaki. Los Angeles,
UCLA Tissue Typing Laboratory, 1980
3. Young A, Jaraquemada D. Awad J. Festenstein H.
Corbett M , Hay FC. Roitt IM: Association of HLADR4/Dw4 and DR2/Dw2 with radiologic changes in a
prospective study of patients with rheumatoid arthritis:
preferential relationship with HLA-Dw rather than
HLA-DR specificities. Arthritis Rheum 27:20-25, 1984
4. Zoschke D, Segall M: Dw subtypes of DR4 in rheumatoid arthritis: evidence for a preferential association with
Dw4. Hum Immunol IS:lI8-124, 1986
5. Thomsen M, Morling N. Snorrason E, Svejgaard A,
Sorensen SF: HLA-Dw4 and rhematoid arthritis. Tissue
Antigens 13:5$-60. 1979
6. Stastny P, Ball El. Khan MA. Olsen NJ, Pincus T. Gao
X: HLA-DR4 and other genetic markers in rheumatoid
arthritis. Br J Rheumatol 27 (suppl II):132-138, 1988
7. Ohta N , Nishimura YK, Tanimoto K , Horiuchi Y, Abe
C, Shiokawa Y, Abe T. Katagiri M, Yoshiki T, Sasazuki
T: Association between HLA and Japanese patients
with rheumatoid arthritis. Hum Immunol 5 : 123-132.
I982
8. Reinsmoen NL, Bach FH: Five HLA-D clusters associated with HLA-DR4. Hum lmmunol 5:123-132, 1982
9. Freeman SM. Noreen HJ, Dahl CA. Nelson PJ, Reinsmoen NL. Bach FH: Determination of DRB alleles of
DR41Dw subtypes by oligonucleotide probing. Hum
Immunol 20: 1-1 1, 1987
10. Nepom GT, Seyfried CE. Holbeck SL. Wiske K R ,
Nepom BS: Identification of HLA-Dw14 genes in DR4'
rheumatoid arthritis. Lancet 11:1002-1004, 1986
I I . Nomenclature Committee on Leukocyte Antigens: Nomenclature for factors of the HLA system. Tissue Antigens 32: 177-187, 1987
12. Gregersen PK, Shen M, Song Q-L, Merryman P, Degar
S. Seki T. Maccari I, Goldberg D, Murphy H, Schwenzer J, Wang CY. Winchester RJ. Nepom GT. Silver J:
946
Molecular diversity of HLA-DR4 haplotypes. Proc Natl
Acad Sci USA 83:2642-2646, 1986
13. Cairns JS, Curtsinger JM, Dahl CA, Freeman S, Alter
BJ, Bach FH: Sequence polymorphism of HLA DRBl
alleles relating to T-cell-recognized determinants. Nature 317:166-168, 1985
14. Bardin T. Legrand L, Naveau B, Marcelli-Barge M,
Debeyre N. Lathrop GM, Poirier JC, Schmid M, Ryckewaert A, Dryll A: HLA antigens and seronegative
rheumatoid arthritis. Ann Rheum Dis 4450-53, 1985
15. Duquesnoy RJ, Marrais M, Hockbarth S, Zeevi A:
Serological and cellular definition of a new HLA-DR
associated determinant, MC I , and its association with
rheumatoid arthritis. Hum Immunol 10: 165-176, 1984
16. Goronzy J, Weyand CM, Fathman CG: Shared T cell
recognition sites on human histocompatibility leukocyte
antigen class I1 molecules of patients with seropositive
rheumatoid arthritis. J Clin Invest 77: 1842-1849. 1986
17. Weyand CM. Goronzy J: Seropositive rheumatoid arthritis is associated with T-cell epitopes encoded within
the third hypervariable region of the HLA-DRI and
HLA-DR4 haplotypes, Immunobiology of HLA. Vol. 11.
New York. Springer-Verlag, 1989
18.. Hurley CK, Ziff BL, Silver J. Gregersen PK, Hartzman
R, Johnson AH: Polymorphism of the HLA-DRl haplotype in American blacks: identification of a DRIpchain
determinant recognized in the mixed lymphocyte reaction. 1 Immunol 140:40194023, 1988
19. Gao X, Fernandez M, Shumway W. Stastny P: DNA
typing for class I1 HLA antigens with allele-specific or
group-specific amplification. I. Typing for subsets of
HLA-DR4. Hum Immunol 27:40-50. 1990
20. Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar RA:
1958 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9:175-176. 1958
21. Nunez G, Moore SE, Ball GV, Hurd ER, Stastny P:
Study of HLA antigens in ten multiple-case rheumatoid
arthritis families. J Rheumatol I1:129-135, 1984
22. Ball EJ, Dombrausky L, Hoover M. Capra JD, Stastny
P: Restriction fragment length polymorphism of the
human T cell receptor alpha gene. I. Two polymorphic
restriction sites localized to different regions of the gene.
Immunogenetics 26:48-55, I987
23. Bell JI, Denney D Jr, Foster L, Belt T , Todd JA,
McDevitt HO: Allelic variation in the DR subregion of
the human major histocompatibility complex. Proc Natl
Acad Sci USA 84:6234-6238, 1987
24. Gregersen PK, Goyert SM. Song Q,Silver J: Microheterogeneity of HLA-DR4 haplotypes: DNA sequence
analysis of LD "KT2" and LD "TAS" haplotypes.
Hum Immunol 19:287-292, 1987
25. Svejgaard A, Jersild C, Nielsen LS, Bodmer WF: HL-A
antigens and aisease statistical and genetical considerations. Tissue Antigens 4:95-105, 1974
GAO ET AL
26. Steimle V, Hinkkanen A. Schlesier M, Epplen JT: A
novel HLA-DR-beta-I sequence from the DRwl 1 haplotype. Immunogenetics 28:208-210, 1988
27. Gorski J, Mach B: Polymorphism of human Ia antigens:
gene conversion between two DR-beta loci resulted in a
new HLA-D/DR specificity. Nature 322:67-70, 1986
28. Celis E, Karr RW, Gregerson PK. Reinsmoen NL:
MHC-restriction analysis of antigen-specific helper Tcells derived from a Caucasian individual with a
DR4(Dwl5); DQw3 haplotype (abstract). Hum Immunol
26:13, 1989
29. Bidwell JL, Bidwell EA, Sansom DM, Klouda PT,
Bradley BA: The origin of HLA-DR "Br": exon 2
nucleotide sequence implicates possible gene conversion of DR1 by DR4-Dw10, DR5 or DRw6-Dwl8. Hum
Immunol 26: 191-197, 1989
30. Gregersen PK, Todd JA, Erlich HA, Long E, Servenius
B, Choi E, Kao HT, Lee JS: First domain sequence
diversity of DR and DQ subregion alleles, Immunobiology of HLA. Vol. 1. Edited by B Dupont. New York.
Springer-Verlag, 1989
31. Sansom DM. Bidwell JL, Maddison PJ, Campion G,
Klouda IT,Bradley BA: HLA-DQ-alpha and DQ-beta
restriction fragment length polymorphisms associated
with Felty's syndrome and DR4-positive rheumatoid
arthritis. Hum Immunol 19:269-278. 1987
32. Singal DP, D'Souza M, Reid B, Bensen WG, Kassam
YB, Adachi JD: HLA-DQ beta-chain polymorphism in
HLA-DR4 haplotypes associated with rheumatoid arthritis. Lancet 11: 1 118-1 119, 1987
33. Nepom GT. Byers P, Seyfried C, Healey LA, Wilske
KR. Stage D, Nepom BS: HLA genes associated with
rheumatoid arthritis: identification of susceptibility alleles using specific oligonucleotide probes. Arthritis
Rheum 32: 15-2 1, 1989
34. Ball EJ. Stastny P: Antigen-specific HLA-restricted
human T-cell lines. 11. A GAT-specific T-cell line restricted by a determinant carried by an HLA-DQ molecule. Immunogenetics 20547-564, 1984
35. Lang B, Winchester R. Nepom GT. Silver J. Gregersen
PK: Further DNA sequence microheterogeneity of the
HLA-DR4Dwl3 haplotype group: importance of amino
acid position 86 of the DR-beta-1 chain for T cell
recognition (abstract). Human Irnmunol 26:35. 1989
36. Nepom GT, Hansen JA. Nepom BS: The molecular
basis for HLA class I1 associations with rheumatoid
arthritis. J Clin Immunol 7:l-7, 1987
37. Gregersen PK, Silver J, Winchester RJ: The shared
epitope hypothesis: an approach to understanding the
molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum 30:1205-1213, 1987
38. Todd JA, Acha-Orbea H. Bell JI, Chao N, Fronek Z,
Jacob CO, McDermott M, Sinha AA. Timmerman L.
Steinman L, McDevitt HO: A molecular basis for MHC
class 11-associated autoimmunity. Science 240: 10031009, 1988
Документ
Категория
Без категории
Просмотров
3
Размер файла
765 Кб
Теги
development, acid, substitution, allele, hla, occurring, naturally, amin, arthritis, risk, rheumatoid
1/--страниц
Пожаловаться на содержимое документа