Download HLA-DRB1*04 Subtypes are Associated with Increased

British Journal of Rheumatology 1997;36:941±944
HLA-DRB1*04 SUBTYPES ARE ASSOCIATED WITH INCREASED
INFLAMMATORY ACTIVITY IN EARLY RHEUMATOID ARTHRITIS
C. SEIDL, U. KOCH,* T. BUHLEIER, R. FRANK,* B. MOÈLLER,* E. MARKERT,$
G. KOLLER-WAGNER,$ E. SEIFRIED AND J. P. KALTWASSER*
Institute for Transfusion Medicine and Immunohematology, Red Cross Blood Donor Service Hessen,
Frankfurt/Main, *Department of Rheumatology, Internal Medicine III, J-W Goethe University, Frankfurt/Main
and $Sandoz AG, NuÈrnberg, Germany
SUMMARY
The sequence polymorphism of HLA-DRB1 molecules in 84 rheumatoid arthritis (RA) patients with early RA has been analysed to evaluate whether particular HLA-DR alleles in¯uence disease progression in the early stage of the disease. Clinical
data were analysed by grouping the patients according to disease-associated haplotype combinations (DRB1*04,04/
DRB1*04,01/DRB1*04,X/DRB1*01,X) in comparison to patients who did not carry these haplotypes (DRB1*X,X). Our
results indicate that patients with early RA who are homozygous for DRB1*04 exhibit an elevated in¯ammatory activity and
an increase of joint a€ections. In addition, the amino acid polymorphism (QR/KRAA) at position 70±74 seems to a€ect the
production of rheumatoid factors. These results support the role of HLA-DRB1 alleles in the pathogenesis of RA and indicate that patients with particular HLA-DRB1*04 haplotype combinations may require intensi®ed therapeutic interventions in
the early stage of the disease to prevent disease progression.
KEY WORDS: HLA, Human leucocyte antigen, Early rheumatoid arthritis, Genetics, Sequence analysis, Disease severity, HLA
and disease association, Rheumatoid factor.
(ACR) for in¯ammatory active RA as revised in
1987. The disease duration was between 6 and 36
months, median age at disease onset was 51.5 yr
(range 25±65 yr). Fifty-two patients were seropositive
and 32 were seronegative for rheumatoid factors
(RFs).
RHEUMATOID arthritis (RA) is a major representative
of a large group of rheumatic diseases with autoimmune components [1]. The occurrence of RA is
strongly associated with the expression of particular
HLA-DRB1 alleles [2]. Despite di€erences in the type
of HLA-DRB1 alleles that are associated with RA in
some ethnic groups, sequence studies have shown
that the disease-associated HLA molecules share a
common amino acid (AA) sequence in the third
hypervariable region of the DRB chain (AA 67±74:
LLEQRRAA or LLEQKRAA) [3]. These shared AA
epitopes are the main structural part of a peptide
binding pocket in the DR heterodimer [4, 5]. Recent
studies revealed that patients who are homozygous
for disease-associated HLA-DR haplotypes exhibit a
more progressive form of the disease [6±10]. These
studies were conducted on patients in a late and
progressive stage of the disease. In the present study,
we have analysed the HLA-DRB1 allele distribution
among RA patients with a disease duration of 3 yr to
investigate whether particular alleles already in¯uence
disease progression in the early stage of the disease.
Clinical and laboratory parameters
Clinical and laboratory data of patients who were
recruited by a multicentre study protocol (SIMERA,
OL2 404-E-00, Sandoz AG) were recorded. In the
present study, only those parameters re¯ecting the
actual clinical status of the patient before entering
the therapeutic protocol of the SIMERA study were
used. All laboratory parameters were determined
according to standard protocols in a central
laboratory facility. From each patient radiographs of
both hands (including ®nger and wrist joints), both
forefeet and the chest were taken. Radiographs were
evaluated using three X-ray scores: number of juxtaarticular erosions, number of eroded joints (possible
score 0±32) and joint damage score (possible score
0±200), according to Larsen and Dale [11, 12]. In
addition to the X-ray scores, joint a€ection was
assessed by determining the Ritchie index [13].
Physical disability was determined by the Health
Assessment Questionnaire (HAQ) [14].
PATIENTS AND METHODS
RA patients and controls
Eighty-four German RA patients (25 male/59
female) and 277 healthy blood donors were included
in our study. All patients ful®lled the diagnostic
criteria of the American College of Rheumatology
Polymerase chain reaction (PCR) SSP typing of
HLA-DRB1 alleles
Genomic DNA was extracted from frozen (±208C)
whole-blood samples anticoagulated with EDTA
following the salting-out procedure [15]. HLA typing
of patients for DRB1*01±DRB1*16 was conducted
by PCR ampli®cation of genomic DNA using nested
Submitted 22 July 1996; revised version accepted 2 April 1997.
Correspondence to: J. P. Kaltwasser, Medizinische Klinik III,
Bereich Rheumatologie, J-W Goethe UniversitaÈt, Theodor-Stern
Kai 7, D-60596 Frankfurt, Germany.
# 1997 British Society for Rheumatology
941
942
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 9
heterozygote position. With this strategy, we were
able to de®ne whether individuals were homozygous
or heterozygous for DRB1 alleles.
primer combinations. The ®rst round of ampli®cation
consisted of two primer sets: one speci®c for exon 2
of all HLA-DRB genes and the second speci®c for
the DRw52 allele group. Subsequently, a total of 19
PCR reactions was performed to determine the
DRB1*01±DRB1*16 speci®cities. Primer sequences
and PCR reaction conditions were chosen as
described elsewhere [16].
Statistical methods
The numbers of patients and controls positive for
an allele were compared by the w2 test or Fisher's
exact test. The level of signi®cance was set to 0.05
and P values were corrected (Pc) for the number of
comparisons. The relative risk (RR) was calculated as
described by Woolf [17]. Clinical data were analysed
using the Kruskal±Wallis test (Statistical Software
Program SPSS, release 6.01, SPSS Inc., Chicago, IL,
USA).
Sequence-based analysis of HLA-DRB1 alleles
Individuals carrying RA-associated DRB1 alleles
(DRB1*01, DRB1*04, DRB1*10 and DRB1*14)
were analysed by sequence-based typing. Sequencebased typing was performed by PCR ampli®cation
using the same strategy as for HLA-SSP typing, but
including one biotinylated primer in the nested PCR
reactions. Subsequently, single-stranded DNA was
generated by magnetic separation with streptavidincoated magnetic beads (Dynabeads M-280, Dynal
A.S., Norway) that bound to the biotin-labelled PCR
product. Sequencing reactions were performed including ¯uorescence-labelled dideoxynucleotides according to standard protocols (PRISM T7 Sequencing
kit, Applied Biosystem Division, Perkin-Elmer,
Foster City, CA, USA). Sequence analysis was conducted on a 373A automated DNA sequencer
(Applied Biosystem Division, Perkin-Elmer) using the
Sequence EDITOR2 and NAVIGATOR2 software
(Applied
Biosystem
Division,
Perkin-Elmer).
Sequence ambiguities of heterozygote samples were
de®ned by comparing the ¯uorescence intensities
of the two di€erent dideoxynucleotides at each
RESULTS
Sequence analysis revealed the characteristic
restriction of particular DR4 subtypes in Caucasian
RA patients to the subtypes DRB1*0401/04/05/08
(Table I). Only HLA-DRB1*0401 was signi®cantly
increased in the patient group (P < 10 ±8) and there
was no signi®cant increase in HLA-DR4 homozygotes given the increased frequency of DRB1*04
alleles in the patient group overall.
In order to evaluate the in¯uence of HLA-DRB1
allele combinations on disease progression, patients
were grouped according to the HLA-DRB1 alleles
associated with RA. Patients who did not carry either
of these alleles were grouped separately and are
referred to as DRB1*X,X. The majority of RA
patients expressed either the DR1 or the DR4 allele
TABLE I
Frequencies of HLA-DRB1 alleles in RA patients and controls (CO)
DRB1*
01
0101/02
0103
15
16
03
04
0401
0402
0403
0404
0405
0406
0407
0408
0409
11
12
13
14
07
08
09
10
RA (N = 84)
CO (N = 277)
Antigen gene frequency
N (%)
N (%)
Antigen gene frequency
N (%)
N (%)
28
27
1
20
4
18
49
42
3
2
5
7
1
9
4
9
4
3
4
(33.3)
(33)
(1.2)
(23.8)
(4.8)
(21.4)
(58.3)
(50)
0
0
(3.6)
(2.4)
0
0
(6.0)
0
(8.3)
(1.2)
(10.7)
(4.8)$
(10.7)
(4.8)
(3.6)
(4.8)
28
27
1
21
4
18
56
46
±, P > 0.05.
$All RA patients were typed as DRB1*1401.
3
2
5
7
1
9
4
9
4
3
4
(16.7)
(16)
(0.6)
(12.5)
(2.4)
(10.7)
(33.3)
(27.4)
0
0
(1.8)
(1.2)
0
0
(0.3)
0
(4.2)
(0.6)
(5.4)
(2.4)$
(5.4)
(2.4)
(1.8)
(2.4)
62
61
1
69
16
59
57
38
1
3
4
1
2
6
2
2
72
10
58
9
69
14
5
12
(22.4)
(22)
(0.4)
(24.9)
(5.8)
(21.3)
(20.6)
(13.7)
(0.4)
(1.1)
(1.4)
(0.4)
(0.7)
(2.2)
(0.7)
(0.7)
(26.0)
(3.6)
(20.9)
(3.2)
(24.9)
(5.1)
(1.8)
(4.3)
66
65
1
75
16
64
62
41
1
3
4
1
2
6
2
2
76
11
65
10
76
15
5
13
(11.9)
(11.2)
(0.2)
(13.5)
(5.8)
(11.6)
(11.2)
(7.4)
(0.2)
(0.5)
(0.7)
(0.2)
(0.4)
(1.1)
(0.4)
(0.4)
(13.7)
(2.0)
(11.7)
(1.8)
(13.7)
(2.7)
(0.9)
(2.3)
Pc
RR
±
±
±
±
±
±
<10±8
<10±8
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
1.48
1.41
±
0.91
0.82
0.92
3.97
4.72
±
±
2.5
±
±
0.0
8.47
±
0.27
0.30
0.43
1.33
0.36
0.88
2.0
1.02
943
SEIDL ET AL.: HLA-DRB1 AND INFLAMMATORY ACTIVITY IN RA
TABLE II
Clinical and laboratory characteristics of RA patients according to HLA-DRB1 alleles
DRB1*
Patients
ESR
CRP
RF value
Larsen score
Ritchie index
HAQ index
04,04
n=7
04,01
n = 10
04,X
n = 32
01,X
n = 18
X,X
n = 17
60
(36±80)
71
(36±106)
446
(226±787)
28
(4±44)
22
(20±29)
1.75
(1.0±2.0)
33
(24±46)
14
(7±42)
80
(25±165)
24
(1.5±33.5)
17
(13±22)
1.13
(0.94±1.53)
30
(16±48)
8
(5±23)
113
(25±268)
26
(13.5±41.3)
14
(10±24)
1.0
(0.53±1.38)
36
(17±56)
7
(3±53)
34
(20±193)
17
(10.8±37.8)
15
(12±20)
0.75
(0.59±1.03)
45
(23±63)
16
(6±42)
20
(20±78)
17
(6.5±42)
17
(12±24)
1.13
(0.5±1.19)
P
0.045
0.02
0.015
n.s.
<0.1
0.07
Values are given as median with 0.25/0.75 quartiles.
(79.7%). The analysis of the clinical data from our
patients according to these allele groups did exhibit a
signi®cant increase of RF in patients carrying allele
combinations with the QR/KRAA motif on both
haplotypes (16.6% vs 3.6%, P < 0.01) or on one
haplotype (25% vs 13.1%, P < 0.05) in comparison
to patients without the QR/KRAA motif (7.2% vs
13.1%). Similar results were obtained when
C-reactive protein (CRP) values and erythrocyte
sedimentation rate (ESR) were compared between the
di€erent DRB1 allele combinations (Table II).
DRB1*04 homozygous patients had signi®cantly
increased CRP or ESR values when compared to
DRB1*04 heterozygous or DRB1*04/01 negative
patients. These data indicate that DRB1 alleles
in¯uence the level of disease activity in RA. In this
respect, we were also interested to study whether the
genetic heterogeneity might be re¯ected in an
increased incidence of erosions or extra-articular
manifestations. Erosions were observed in the
majority of early RA patients (78.6% vs 21.4%), but
patients who carry the DRB1*04 or DRB1*01 alleles
did not exhibit an increase in the incidence of
erosions or in the severity of erosive destruction as
determined by the evaluation of X-ray scores
according to Larsen (Table II). Since severe erosive
destructions are unlikely to occur in the early stage
of the disease, we also evaluated two indexes (Ritchie
and HAQ). In comparison to radiographic data,
these indexes are questionnaire measures for joint
a€ection and physical disability, and may provide
more sensitive information in the early stage of the
disease regarding the long-term natural history of
RA. The results of this analysis demonstrated a
tendential (P < 0.1) increase of both indexes in the
group of DRB1*04 homozygous RA patients
(Table II). This tendential increase is slightly more
prominent for the HAQ index, which may be due to
the fact that in comparison with the Ritchie index the
HAQ index uses a questionnaire to de®ne the overall
functionality of the patient, whereas the Ritchie index
only includes the number of painful/tender joints. In
contrast to the evaluation of disease progression, the
analysis of extra-articular disease manifestations, as
de®ned by the occurrence of s.c. nodules, vasculitis
or secondary SjoÈgren's syndrome, did not reveal any
association for particular DRB1 allele combinations.
Extra-articular manifestations were observed only in
11 (13.0%) of these early RA patients.
DISCUSSION
In the present study, we have analysed the HLADRB1 allele distribution among patients with early
manifestation of the disease. The HLA-DRB1*04
allele frequency was clearly increased in patients with
elevated levels of CRP and ESR. Furthermore, the
majority of patients carrying the disease-associated
HLA alleles DRB1*04 and DRB1*01 produced RFs.
The variation of these laboratory parameters may
re¯ect an increased in¯ammatory and autoimmune
activity among DRB1*04- and DRB1*01-positive
RA patients. In this respect, our observation supports
the pathogenetic function of particular AA residues
in the third hypervariable region for RA. In contrast
to previous studies, we did not observe a signi®cant
increase in the incidence or severity of erosions
among patients carrying arthritogenic alleles
[10, 18, 19]. In particular, patients who express
shared
AA
epitopes
on
both
haplotypes
(DRB1*04,04 or DRB1*04,01) did not exhibit
an increased risk of developing erosions. This
discrepancy is most likely due to the very early phase
of the disease course (<3 yr) in our patient cohort.
Thus, not all patients in our study might have
developed erosions at this early phase of the disease
course that could be veri®ed by X-ray examination.
The elevated indexes for joint a€ection (Ritchie) and
physical disability (HAQ) among DR4 homozygous
RA patients may, however, re¯ect the genetic
predisposition to mount an elevated autoimmune
response. These indexes assess the joint a€ection by
either evaluating the tenderness of joints or by using
a questionnaire to evaluate individual disabilities.
Thus, in contrast to the evaluation of the radiological
joint damage as de®ned by the Larsen score, the
944
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 9
Ritchie and HAQ indexes also assess in¯ammatory
reactions that a€ect the joints and have not yet led to
erosive alterations [11±14, 20]. In this respect, the
increase in these indexes among RA patients who
express alleles carrying shared AA on both
haplotypes supports the pathogenetic concept that an
increased cell surface expression of particular alleles
results in a higher susceptibility for severe autoimmune reaction. The increased in¯ammatory
reactivity that we observed in early RA patients may
re¯ect the stronger response of genetically predisposed RA patients to putative autoantigens. In
this respect, our data support the role of HLADRB1*04 as the most potent allele in the determination of disease severity among RA patients. Patients
who carry this allele are already in the early stage of
the disease, characterized by an increased in¯ammatory reactivity that may indicate an elevated risk of
developing a more progressive type of the disease.
ACKNOWLEDGEMENTS
The authors would like to thank Dr Prestele
(Sandoz AG, Basel, Switzerland) for preparing the
computerized clinical data set and Dr Laasonen for
performing the radiological evaluation. The authors
are also indebted to the colleagues of the rheumatology departments who collected clinical data on RA
patients for this study: Abt. Rheumatologie,
Medizinische
Hochschule
Hannover
(Zeidler);
Rheumaklinik, Klinikum Berlin-Buch (GromnicaIhle); Rheumaklinik Rappenau (Bolten), Institut
fuÈr Immunologie, UniversitaÈtsklinikum Erlangen
(Manger),
Abt.
Rheumatologie,
Seehospital
Cuxhaven (KaÈstner), Schloûparkklinik Berlin (Alten);
Abt. Rheumatologie, Klinikum der Alberts-LudwigsUniversitaÈt
Freiburg
(Peter);
Klinik
fuÈr
Rheumakranke Bad Kreuznach (Dreher); Klinik
Niedersachsen Bad Nenndorf (Becker-Capeller);
Rheinisches Rheumazentrum St Elisabeth-Hospital
Meerbusch-Lank
(Langer);
Marburg
(MuÈllerBrodmann); Abt. Innere medizin Rotes-KreuzKrankenhaus
Bremen
(Botzenhardt);
Abt.
Nephrologie/Rheumatologie UniversitaÈt GoÈttingen
(Weber) and Medizinische Poliklinik UniversitaÈtsKlinik Leipzig (HaÈntzschel).
REFERENCES
1. Snowden N, Kay RA. Immunology of systemic rheumatoid disease. In: Saklatvala J, Walport MJ, eds.
Immunology of rheumatoid disease. Br Med Bull
1995;51:437±48.
2. Winchester R. The molecular basis of susceptibility to
rheumatoid arthritis. Adv Immunol 1994;56:389±466.
3. Gregersen PK, Silver J, Winchester RJ. The shared
epitope hypothesis: an approach to understand the
molecular genetics of susceptibility to rheumatoid
arthritis. Arthritis Rheum 1987;30:1205±13.
4. Hiraiwa A, Yamanaka K, Kwok WW, Mickelson EM,
Masewicz S, Hansen JA et al. Structural requirements
for recognition of the HLA-Dw14 class II epitope: a
key HLA determinant associated with rheumatoid
arthritis. Proc Natl Acad Sci USA 1990;87:8051±5.
5. Weyand CM, Goronzy JJ. Inherited and noninherited
risk factors in rheumatoid arthritis. Curr Opin
Rheumatol 1995;7:206±13.
6. Weyand CM, Goronzy JJ. Disease mechanisms in
rheumatoid arthritis: Gene dosage e€ect of HLA-DR
haplotypes. J Lab Clin Med 1994;124:335±8.
7. McMahon MJ, Hillarby MC, Clarkson RW, Hollis S,
Grennan DM. Major histocompatibility complex
variants and articular disease severity in rheumatoid
arthritis. Br J Rheumatol 1993;32:899±902.
8. Deighton CM, Cavanagh G, Rigby AS, Lloyd HL,
Walker DJ. Both inherited HLA-haplotypes are
important in the predisposition to rheumatoid arthritis.
Br J Rheumatol 1993;32:893±8.
9. Weyand CM, McCarthy TG, Goronzy JJ. Correlation
between disease phenotype and genetic heterogeneity in
rheumatoid arthritis. J Clin Invest 1995;95:2120±6.
10. Gough A, Faint J, Salmon M, Hassell A, Wordsworth
P, Pilling D et al. Genetic typing of patients with
in¯ammatory arthritis at presentation can be used to
predict outcome. Arthritis Rheum 1994;37:1166±70.
11. Larsen A, Dale K, Eek M. Radiographic evaluation of
rheumatoid arthritis and related conditions by standard
reference ®lm. Acta Radiol Diagn 1977;18:481±91.
12. Larsen A, Thoen J. Hand radiography of 200 patients
with rheumatoid arthritis after an interval of one year.
Scand J Rheumatol 1987;16:395±401.
13. Ritchie D et al. Clinical studies with an articular index
for the assessment of joint tenderness in patients with
rheumatoid arthritis. Q J Med (NS) 1968;37:393±406.
14. Pincus T, Callhan LF, Vaughn WK. Questionnaire,
walking time and button test measures of functional
capacity as predictive markers for mortality in RA. J
Rheumatol 1987;14:240±51.
15. Miller SA, Dykes DD, Polesky HF. A simple salting
out procedure for extracting DNA from human
nucleated cells. Nucleic Acids Res 1988;16:1215.
16. Bein G, GlaÈser R, Kirchner H. Rapid HLA-DRB1
genotyping by nested PCR ampli®cation. Tissue
Antigens 1992;39:68±73.
17. Tiwari JL, Terasaki PI. HLA and disease association.
Heidelberg: Springer-Verlag, 1985:20.
18. Emery P, Salmon M, Bradley H et al. Genetically
determined factors as a predictor of radiological
changes in patients with early symmetrical arthritis. Br
Med J 1992;305:1387±9.
19. Toda Y, Minamikawa Y, Akagi S, Sugano H, Mori Y,
Nishimura H et al. Rheumatoid-susceptible alleles of
HLA-DRB1 are genetically recessive to non-susceptible
alleles in the progression of bone destruction in the
wrists and ®ngers of patients with RA. Ann Rheum
Dis 1994;53:587±92.
20. Pincus T. Clinical observations often provide more
accurate information concerning the longterm
natural history and results of treatment for rheumatoid
arthritis than structural research studies. Rheumatol
Eur 1995;24(suppl. 2):179±84.