вход по аккаунту


Differential expression of HLA class II genes associated with disease susceptibility and progression in rheumatoid arthritis.

код для вставкиСкачать
Vol. 48, No. 10, October 2003, pp 2779–2787
DOI 10.1002/art.11251
© 2003, American College of Rheumatology
Differential Expression of HLA Class II Genes
Associated With Disease Susceptibility and
Progression in Rheumatoid Arthritis
Christian Heldt,1 Joachim Listing,1 Osman Sözeri,1 Franca Bläsing,2 Stefan Frischbutter,1
and Brigitte Müller1
ger of the 2 DRB4 promoters, independent of the DRB1
allele. Moreover, DRB1*04 alleles in RA patients
showed a reduced association with the DRB4 splice
variant, completely preventing DRB4 expression.
Conclusion. Our findings represent the first evidence of a correlation between the differential expression of HLA class II genes and both the susceptibility
and the progression of RA.
Objective. Rheumatoid arthritis (RA)–associated
HLA class II genes are assumed to promote susceptibility to and/or progression of the disease. Among the
various modes of action proposed so far is the effect of
the differential expression of HLA class II genes in
different types of antigen-presenting cells on the Th1/
Th2 balance. The aim of this study was to investigate the
differential expression of genes encoded within the
RA–associated HLA–DR4 superhaplotype and within
the neutral DR7 and DR9 superhaplotypes.
Methods. The promoters encoded within these 3
haplotypes were first analyzed for sequence polymorphisms. To test for functional consequences, we assumed that the binding of nuclear factors to the promoter elements was correlated with the transcription
activity, and we used surface plasmon resonance technology. To that end, oligonucleotides representing the
polymorphic regulatory sequences and nuclear extracts
from a monocyte cell line and a B cell line were used.
Results. While the promoters of the highly polymorphic HLA–DRB1*04, *07, and *09 alleles showed
comparable binding of nuclear factors, differential
binding was observed for the 2 promoters that drive the
relatively nonpolymorphic DRB4 alleles in linkage disequilibrium with DRB1. Interestingly, analysis of RA
patients positive for DR4, DR7, and DR9 revealed the
segregation of radiographic progression with the stron-
The association of HLA class II genes with the
development of autoimmune diseases has captivated
numerous scientists over the last decades. This association has extensively been studied in rheumatoid arthritis
(RA). The DR4, DR1, and DR10 haplotypes have been
found to be positively associated with the disease, while
others, such as DR2 and DR7, have been reported to be
neutral, if not protective (1–3). Several mechanisms have
been proposed to explain the association, among them
are epitope selection (4) and differential expression (5).
The shared epitope hypothesis centers on epitope selection and is intriguingly simple. However, evidence has
accumulated from mouse studies supporting an impact
of the differential expression of class II major histocompatibility complex (MHC) genes on the T cell cytokine
profile, and thus an association with protective/
suppressive effects (5).
In this context, differential expression of the
various I-A genes has been correlated with a single
mismatch in the X2 box of the promoter region (6).
Interestingly, the structural requirements for class II
MHC promoters (S, X1, and X2, as well as Y, CAAT,
and TATA boxes) are well conserved over long evolutionary distances but allow for considerable intraspecies
variability. It is assumed that the polymorphism in the
regulatory region confers flexibility in the immune response, both for the species and for the heterozygous
Supported by a grant from the Fritz Thyssen Stiftung and by
the Berlin Senate for Research and Cultural Affairs.
Christian Heldt, PhD, Joachim Listing, PhD, Osman Sözeri,
PhD, Stefan Frischbutter, Brigitte Müller, PhD: Deutsches
RheumaForschungs-Zentrum, Berlin, Germany; 2Franca Bläsing,
PhD: Max-Planck-Institut für Molekulare Genetik, Berlin, Germany.
Address correspondence and reprint requests to Brigitte
Müller, PhD, Deutsches RheumaForschungs-Zentrum, Schumannstrasse 21/22, Berlin 10117, Germany. E-mail:
Submitted for publication September 16, 2002; accepted in
revised form May 22, 2003.
individual (7). While the variation in the coding sequence is understood to be sustained by balancing
selection, concomitant variation in the regulatory regions is thought to result from linkage disequilibrium
(8). However, the idea has recently been put forward of
a coordinate evolution in the coding and regulatory
sequences, which ensures that class II MHC sequences
favoring Th2 differentiation associate with class II MHC
molecules that are able to present worm epitopes.
Conversely, Th1-favoring promoters are predicted to be
linked to coding sequences that are able to present
endogenous viral epitopes (9).
The enormously polymorphic and polygenic organization of the HLA complex and the lack of specific
monoclonal antibodies have so far only allowed for a
patchy analysis of the differential expression of class II
MHC genes in humans. Several hundred HLA–DRB1
alleles showing sequence variation in the coding region
(10) as well as in the regulatory region (11) can be
grouped into the 10 different superhaplotypes, DR1–
DR10, reflecting the 10 major serologically defined
HLA–DR specificities. These superhaplotypes each contain multiple haplotypes with unique DRB1 sequences
and multiple associations with the appropriate DRB3,
DRB4, or DRB5 genes plus additional nonfunctional
DRB genes. They are encoded on 5 different allele
lineages. One of these lineages encodes the DR4, DR7,
and DR9 superhaplotypes containing the DRB4 gene,
associated with an HLA–DRB1*04, *07, and *09 allele,
respectively (Figure 1). While the DR4 superhaplotype
is RA-associated, DR7 and DR9 are considered neutral.
So far, more than 40 different DRB1*04 alleles that
share the same promoter have been described. Likewise,
there is only 1 promoter reading into the single
DRB1*09 allele and 1 promoter reading into the 4
DRB1*07 alleles (10).
In contrast to the DRB1 gene, the DRB4 gene
shows relatively little polymorphism, and of the 8 functional alleles, only 3 show an amino acid substitution
within exon 2 that encodes the third hypervariable loop
of the class II molecule responsible for epitope selection
(11). However, a regulatory polymorphism is associated
with the DRB4 gene, consisting of 2 different promoters,
DRB4A and DRB4B. Furthermore, the DRB4A promoter has been reported to be in linkage disequilibrium
with a splice variant that prevents the formation of the
DRB4 gene product altogether (Figure 1) (12). If
present, both the DRB1 and the DRB4 gene products
combine with the nonpolymorphic ␣-chain to form the
heterodimeric HLA–DR molecule.
To ensure sufficient immune responses to survive
Figure 1. Genomic organization of the HLA complex on the human
chromosome 6. Top, The HLA–DR region is embedded between the
HLA–DQ and the HLA class III region. Middle, Enlargement of the
DR region, showing the 5 allele lineages that encode the 10 different
DRB1 superhaplotypes (DR1–DR10). Membership in the allele lineages is defined by the DRB1 allele. Bottom, Enlargement of the allele
lineage, showing the functional DRB1 alleles, the DRB4 and DRA
genes, as well as pseudogenes encoded by this particular allele lineage.
Promoters are shown as large arrows, the DRB4 splice variant is
indicated by a solid star, and pseudogenes are shown in gray typeface.
infections, class II MHC expression on antigenpresenting cells needs to be tightly controlled. The high
number of different class II MHC genes and alleles has
allowed for a detailed analysis of the structural requirements for this tight regulation, consisting of ubiquitous
as well as unique regulatory elements. Among the
unique elements are the X1 box, which binds the different RFX factors, which, in turn, bind the class II MHC
transcription activator (CIITA) molecule to form the
supramolecular transcription complex (13), and the S
box, which represents a duplicated X1 box. The ubiquitous regulatory elements consist of the X2 box, which
has recently been defined as a CRE box (14), the Y box,
which represents a reversed CAAT box, the CAAT box
itself, which binds the nuclear factor Y factors, and the
TATA box at position –45, which binds the TATAbinding factors (9). It has recently been proposed that
the unique elements ensure a basic transcription of the
locus, while the ubiquitous elements are responsible for
mediating the fine-tuning (15).
The differential expression of HLA class molecules has previously been described on various levels. A
ranking of transcription activity among the various
HLA–DR genes and alleles was established via chloramphenicol acetyltransferase assays performed in EpsteinBarr virus–transformed B cell lines (16). The corresponding steady-state levels of messenger RNA
(mRNA) were analyzed by reverse transcription–
polymerase chain reaction (RT-PCR) and RNase protection assays using RNA isolated from peripheral B
cells (17,18). In the context of RA, an elevated expression of HLA class II genes has been described for
peripheral B cells (19). Thus, HLA class II expression
has been studied almost exclusively in B cells. However,
the recent finding of a differential surface expression of
DRB1 and DRB4 on monocytes, but not on B cells, calls
for an in-depth analysis of HLA class II expression on all
types of antigen-presenting cells (20).
In the present study, we set out to compare the
RA-associated DR4 superhaplotype and the neutral
DR7 and DR9 superhaplotypes, performing binding
studies of nuclear extracts from monocytes and B cells to
the various promoter elements. Our particular focus was
on structural differences in the regulatory regions that
lead to functional consequences related to the autoimmune process.
Study population. The study patients met the American College of Rheumatology (formerly, the American Rheumatism Association) classification criteria for RA (21) (as
described in detail in ref. 2). A total of 73 RA patients and 28
controls expressing the DR4, DR7, and/or DR9 haplotypes
were evaluated. The mean age of the RA patients was 51 years
(range 19–74 years); 17 of them were men. The mean age of
the controls was 37 years (range 23–60 years); half of them
were men. Genomic DNA was isolated according to the
method described by Sykes (22). Typing of the polymorphism
at the HLA–DRB locus was performed using genomic DNA,
PCR, and biotinylated sequence-specific oligonucleotide
probes (23).
Cell lines and fluorescence-activated cell sorting
(FACS) analyses. The monocyte cell line THP-1 (ACC 16;
DSMZ, Braunschweig, Germany) and the B cell line BJAB
(ACC 72; DSMZ) were cultured (at 37°C in an atmosphere of
5% CO2) in complete RPMI 1640 medium containing 10%
fetal calf serum, 2 mM L-glutamine, 100 units/ml of penicillin,
100 ␮g/ml of streptomycin, and 50 ␮M 2-mercaptoethanol.
Both cell lines were typed at their HLA–DR loci as described
above. Cell surface expression of DR molecules on THP-1 and
BJAB cells was monitored via staining with the pan–reactive
monoclonal antibody L243 (clone HB55; American Type Cul-
ture Collection [ATCC], Manassas, VA), using the CD3specific monoclonal antibody OKT3 (clone CRL-8001; ATCC)
as an isotype control. Flow cytometric analyses were performed using a FACSCalibur instrument with CellQuest software (both from Becton Dickinson, Heidelberg, Germany).
Nuclear extracts. Cells were left untreated or were
stimulated with cAMP (100 ␮M), transforming growth factor
␤1 (TGF␤1; 20 ng/ml), vitamin D3 (250 ng/ml), lipopolysaccharide (LPS; 10 ␮g/ml), interleukin-4 (IL-4; 20 units/ml), TPA
(1 nM), IL-1␤ (5 ng/ml), interferon-␥ (IFN␥; 1,000 units/ml),
and anti-CD40/anti-IgM (5 ␮g/ml each), respectively. Then,
5 ⫻ 107 THP-1 and BJAB cells were washed twice in ice-cold
phosphate buffered saline and resuspended in 500 ␮l of buffer
A (10 mM HEPES, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM
dithiothreitol, pH 7.9) containing a protease inhibitor cocktail
(Roche, Mannheim, Germany). The suspension was kept on
ice for 15 minutes, and then Nonidet P40 was added to a final
concentration of 0.5%. The lysates were centrifuged at 6,500
revolutions per minute for 20 seconds, and the pellets were
resuspended in 150 ␮l of buffer C (20 mM HEPES, 1.5 mM
MgCl2, 420 mM NaCl, 0.2 mM EDTA, 25% [weight/volume]
glycerol, pH 7.9) containing the protease inhibitor cocktail.
The probes were stirred vigorously on ice for 30 minutes and
centrifuged at 13,000 rpm for 10 minutes. The nuclear extracts
containing supernatants were collected, and the protein concentrations were measured using the bichinolin-4-carbonic
acid protein assay reagent kit (Interchim, Montluçon, France).
Surface plasmon resonance. The BIAcore System 2000
(Uppsala, Sweden) was used to study the binding affinities of
nuclear extracts to promoter fragments bound to streptavidincoated chips. Experiments were performed as described previously (24). Minor variations included using TES buffer (10
mM Tris, 1 mM EDTA, 300 mM NaCl, pH 8) for pretreatment
of the chips. Before immobilization on the streptavidin chip,
the biotinylated sense strands of the promoter fragments
(Table 1) were hybridized to the antisense strands (2-fold
excess), and double-stranded oligonucleotides were diluted in
TES buffer to a final concentration of 100 nM. Another
variation included the immobilization of 400 resonance units
(RU) of DNA onto the chips. Measurements were performed
at 25°C and at a flow rate of 30 ␮l/minute in modified buffer C
(20 mM HEPES, 1.5 mM MgCl2, 420 mM NaCl, 0.2 mM
EDTA, 0.005% BIAcore Surfactant P20, pH 7.9). Two hundred fifty microliters of the nuclear extracts (100 ␮g/ml) was
injected and, at the end of the association time, was replaced
with continuous buffer flow. At the end of each cycle, the
sensor chip was regenerated by injecting two 1-minute pulses
of 0.05% sodium dodecyl sulfate in TES buffer.
The binding response was analyzed by measuring the
RU values during each cycle. Data analysis was performed with
the Misevaluation software (BIAcore). The binding affinities
were determined by subtracting the RU value before injection
from the RU value at the end of the association phase. The
specific binding affinities were then subtracted from the reference cell containing no DNA and were normalized to the
nonspecific DNA.
Electrophoretic mobility shift assay (EMSA). The
single-stranded oligonucleotides TATA box DRB4A and
TATA box DRB4B (Table 1) were end-labeled using ␥-32PATP (Amersham, Freiburg, Germany) and T4 kinase (Promega, Madison, WI). Double-stranded oligonucleotides were
Table 1. Oligonucleotides used in the present study*
To analyze DRB4 splice variants
5⬘ DRB4 (intron 1)
3⬘ DRB4 (exon 2)
To analyze DRB4 promoters
5⬘ W-Box DRB4
To analyze binding affinities
5⬘ Bio S-Box DR4
5⬘ Bio S-Box DR7, DR9
5⬘ Bio S-Box DRB4
5⬘ Bio X-Box DR4
5⬘ Bio X-Box DR7, DR9
5⬘ Bio X-Box DRB4
5⬘ Bio X-Box DRA
5⬘ Bio Y-Box DR4, DR7, DR9
5⬘ Bio Y-Box DRB4A
5⬘ Bio Y-Box DRB4B
5⬘ Bio CCAAT-Box DR4
5⬘ Bio CCAAT-Box DR4, DR7
5⬘ Bio CCAAT-Box DRB4
5⬘ Bio TATA-Box DR4, DR7
5⬘ Bio TATA-Box DR9
5⬘ Bio TATA-Box DRB4A
5⬘ Bio TATA-Box DRB4B
5⬘ Bio random
* Oligonucleotides were synthesized by Tib Molbiol (Berlin, Germany). For the oligonucleotides used to analyze binding
affinities, only the biotinylated (Bio) sense strand is shown; the antisense strand did not include biotin.
generated by annealing the 32P end-labeled sense and unlabeled antisense oligonucleotides (20-fold excess of antisense).
For competition experiments, the 50 nM double-stranded
promoter fragment was incubated for 15 minutes with unlabeled oligonucleotide competitors (2–100-fold excess), 40 ␮g
of nuclear extract, and 0.5 mg/ml of poly(dI-dC) in binding
buffer (40 mM HEPES, 10 mM MgCl2, 100 mM KCl, 30% [w/v]
glycerol, 0.02% [w/v] Triton X-100, pH 7.9) containing the
protease inhibitor cocktail (Roche) in a final volume of 20 ␮l.
Electrophoresis was carried out on a native polyacrylamide gel
(4%) using 0.5⫻ Tris–borate–EDTA as the running buffer.
After drying, the gel was exposed to the phosphorimager for
DNA amplification and sequencing. Genomic DNA
(0.5 ␮g) was amplified in a 50-␮l PCR reaction containing 1
␮M each of the 5⬘ and 3⬘ primers (Table 1), 2.5 units of Taq
polymerase (Promega), 0.3 units of Pfu polymerase (Promega),
and 250 ␮M dNTPs in PCR buffer (670 mM Tris HCl, 1.3 mM
MgCl2, 160 mM [NH4]2SO4, 0.1% [w/v] Tween 20, pH 8.8). An
initial incubation at 96°C for 10 minutes was followed by 35
cycles at 96°C for 1 minute, 60°C for 1 minute, and 72°C for 1
minute. The reaction was terminated at 72°C for 10 minutes.
The PCR products were separated on 1.4% agarose gels and
purified using a JetSorb kit (Genomed, Bad Oeynhausen,
Germany). The amplified promoter fragments and intron
1/exon 2 splice sites were then sequenced using the dideoxy
chain-termination method and the ABI Prism 310 sequencer
(Applied Biosystems, Foster City, CA). The sequencing reaction was carried out in a 10-␮l volume containing 1 ␮l of 0.5
␮M 5⬘ primer (Table 1), 2 ␮l of PCR product, and 2 ␮l of
BigDye (Applied Biosystems).
Statistical analysis. Fisher’s exact tests were performed to calculate P values and confidence intervals describing the differences between the fractions shown in Table 2 and
Figure 5.
Mediation of the IFN␥-stimulated expression of
HLA–DR via the X box. In this study, we compared the
RA-associated superhaplotype DR4 with the neutral
DR7 and DR9 superhaplotypes, looking for structural
differences that lead to functional consequences. For
analysis of the structural differences, we aligned the
promoter sequences reading into the HLA–DRB1*04,
*07, *09, and the HLA–DRB4 alleles (11), and we
identified polymorphic residues in and around the
known S, X, Y, CAAT, and TATA boxes. For analysis of
the functional consequences, we designed oligonucleotides representing the various polymorphic boxes and
tested their binding to nuclear extracts isolated from
antigen-presenting cells. We used as antigen-presenting
cells the human monocyte cell line THP-1 and the
human B cell line BJAB, which carry endogenously the
Figure 2. Differential surface expression of HLA–DR molecules on
monocytes and B cells. Histograms show the flow cytometric analyses
of unstimulated or interferon-␥ (IFN␥)–stimulated (1,000 units/ml for
48 hours) THP-1 and BJAB cells stained with the isotype control or
with the HLA–DR–specific L243 monoclonal antibody.
DR1/DR2 and DR8/DR5 (or DR6) superhaplotypes,
respectively. To confirm that both cell lines were suitable hosts for analyzing DR expression, surface
HLA–DR was measured by FACS analysis using the
monoclonal antibody L243.
Figure 2 shows that 100% of the BJAB cells were
positive for HLA–DR, and stimulation with IFN␥ only
slightly increased the amount of surface expression, as
determined by the mean fluorescence intensity. Likewise, TPA and IL-1␤ only slightly increased HLA class II
surface expression, while anti-IgM/anti-CD40, LPS,
IL-4, cAMP, vitamin D3, and TGF␤1 did not add to the
stimulatory effect induced by IFN␥ (data not shown). In
contrast, only very few of the unstimulated THP-1 cells
were DR positive. However, stimulation with 1,000
units/ml of IFN␥ for 48 hours (Figure 2), but not with
TPA or TGF␤1 (data not shown), induced DR expression on all THP-1 cells.
To determine the contribution of the individual
promoter boxes to DR expression, we analyzed the
binding affinities of THP-1 and BJAB nuclear extracts to
the different oligonucleotides representing the polymorphic promoter boxes. The surface plasmon resonance
was measured with the BIAcore apparatus, and a random oligonucleotide was used as a control. The binding
affinities obtained were then normalized to the affinity
of THP-1 or BJAB nuclear extracts to this random
sequence and are presented as percentage increase.
Figure 3A shows a comparison of the X box
sequences of the DRB1*04, *07, *09, DRB4A, and
DRB4B promoters. The DRB1*04 promoter differed
from the DRB1*07 and *09 promoters by just 1 polymorphic residue 3⬘ of the X2 box. The DRB4A and
DRB4B promoters were identical, yet differed from
DRB1*04 by a G3 A substitution within the X box and
by T3 G and T3 A substitutions immediately 3⬘ of the
X2 box. The X boxes of the DRA promoter were very
different from those of the DRB promoters, yet the
relative binding of THP-1 and BJAB nuclear proteins to
the X box DRA was highest (132% and 191%, respectively) (Figure 3B). For the X box DRB oligonucleotides, binding varied between 106% and 109% for the
THP-1 nuclear protein and between 117% and 126% for
the BJAB nuclear protein. Activation of THP-1 cells
with IFN␥ stimulated the binding of nuclear proteins to
the X box and led to a 126% (DRB1*07/*09), 139%
(DRB4), 149% (DRB1*04), and 215% (DRA) increase,
respectively. For nuclear extracts of IFN␥-stimulated
BJAB cells, the increase in transcription factor concentration was less dramatic, but was also strongest for the
X box DRA oligonucleotide (Figure 3B).
Stimulation of THP-1 or BJAB cells with cAMP,
TPA, or TGF␤1 did not lead to a significant increase in
transcription factors (data not shown). Of all the oligonucleotides tested, the X box oligo showed the strongest
increase in the binding of nuclear extracts isolated from
IFN␥-stimulated THP-1 cells, indicating that IFN␥
Figure 3. Mediation of interferon-␥ (IFN␥)–stimulated HLA–DR
expression via the X box. A, Sequence alignment of the HLA–
DRB1*04, DRB1*07, DRB1*09, DRB4A, DRB4B, and DRA promoters. Shaded areas show the core sequences of the transcription factor
target sites. Sequence variations among the alleles/genes are indicated.
B, Binding of nuclear extracts from the THP-1 and BJAB cell lines to
the X boxes of DRB1*04, DRB1*07/*09, DRB4, and DRA. Surface
plasmon resonance was measured and normalized to the random DNA
in order to calculate the percentage increase in binding. For isolation
of nuclear extracts, cells were either left untreated (solid bars) or were
treated with 1,000 units/ml of IFN␥ for 48 hours (hatched bars).
Values are the mean and SD.
stimulation of HLA–DR expression is mediated via the
X box.
Mediation of the differential expression of the
relatively nonpolymorphic HLA–DRB4 gene via the
TATA box. Alignment of the polymorphic TATA box
sequences is shown in Figure 4A. While the box itself
was completely conserved, sequence variation was found
5⬘ and 3⬘ of the TATA box. The relative binding
affinities of both THP-1 and BJAB nuclear proteins
were highest for the DRB4B (203% and 184%, respectively) and lowest for the DRB4A (126% and 106%,
respectively) promoter oligonucleotides. The binding of
the nuclear proteins to the DRB1-specific oligonucleotides ranged between 174% and 179% for THP-1 cells
and between 155% and 168% for BJAB cells, respectively. Stimulation of BJAB cells with IFN␥ increased
the concentration of transcription factors that bound the
TATA box and led to a 5% (DRB4A), 10% (DRB4B),
29% (DRB1*09), and 37% (DRB1*04/*07) increase in
the relative binding. Neither IFN␥, cAMP, nor TPA led
to an increase in the transcription factor concentration
of THP-1 nuclear extracts (data not shown).
The differential binding of nuclear proteins to the
DRB4A- and DRB4B-specific oligonucleotides was confirmed by competition assays in which 32P-labeled
TATA box DRB4A or DRB4B and cold TATA box
DRB4A or DRB4B oligonucleotides competed for binding to the nuclear proteins. The polyacrylamide gel from
the EMSA showed a more rapid decline in the signal
strength of the upper band when competed out with
unlabeled TATA box DRB4B (Figure 4C). Figure 4D
summarizes the findings of the competition analysis with
P-labeled TATA box DRB4A using a phosphorimager
for quantification of the bands. Again, the reduction in
signal strength was stronger when competed out with
unlabeled TATA box DRB4B, indicating that the
DRB4B promoter binds nuclear factors with a higher
affinity than does the DRB4A promoter and is predicted
to result in stronger transcription activities.
Association of an elevated expression of HLA–
DRB4B with radiographic progression in early RA. We
next addressed whether any of these structural and
functional promoter differences correlated with a specific disease course. We therefore evaluated a cohort of
RA patients who had previously been evaluated for risk
factors of radiographic progression (2). In these patients, a C-reactive protein (CRP) level ⱖ1.5 mg/dl at
the start of treatment with disease-modifying antirheumatic drugs was the strongest risk factor for radiographic
progression of RA and was independent of genetic
predisposition (2). We selected all DRB1*04-,
Figure 4. Mediation of the differential expression of HLA–DRB4A
and HLA–DRB4B via the TATA box. A, Sequence alignment of the
HLA–DRB1*04, DRB1*07, DRB1*09, DRB4A, and DRB4B promoters. Shaded areas show the core sequences of the transcription factor
target sites. Sequence variations among the alleles/genes are indicated.
B, Binding of nuclear extracts from the THP-1 and BJAB cell lines to
the TATA boxes of DRB1*04/*07, DRB1*09, DRB4A, and DRB4B.
Surface plasmon resonance was measured and normalized to the
random DNA in order to calculate the percentage increase in binding.
For isolation of nuclear extracts, cells were either left untreated (solid
bars) or were treated with 1,000 units/ml of interferon-␥ for 48 hours
(hatched bars). Values are the mean and SD. C and D, Competition
assays for the binding of nuclear extracts to TATA boxes DRB4B and
DRB4A were performed by electrophoretic mobility shift assay. C,
Polyacrylamide gel showing that the radiolabeled TATA box DRB4B
oligonucleotide was competed out by increasing concentrations of
unlabeled TATA box DRB4B and TATA box DRB4A. Arrowhead
indicates the band corresponding to nuclear proteins bound to the
radiolabeled TATA box DRB4B oligonucleotide. The radiolabeled
TATA box DRB4A oligonucleotide was competed out in the same
way. D, The intensity of the upper band was calculated in relation to
the intensity of the corresponding lane using the phosphorimager.
There was a reduction in signal strength resulting from the competition
between 32P-labeled DRB4A and unlabeled DRB4 oligonucleotides
for binding to nuclear proteins. Binding without competition was set at 1.
DRB1*07-, and DRB1*09-positive patients for whom
initial CRP values were available.
A total of 73 RA patients were evaluated; 35
of them had initial CRP values ⱖ1.5 mg/dl (Table 2).
These 35 patients shared a total of 2 DRB1*09, 11
DRB1*07, and 34 DRB1*04 alleles, of which 27 were
*0401; 7 alleles were *0403, *0404, and *0408, respectively. The allele distribution in the 38 patients with
initial CRP values ⬍1.5 mg/dl was comparable, with 4
DRB1*09, 12 DRB1*07, and 38 DRB1*04 alleles, of
which 32 were *0401.
All 73 RA patients were evaluated for associations with the DRB4B promoter and the splice variant.
Most interestingly, the association with the stronger
DRB4B promoter was more pronounced in the group
with rapid disease progression (22%) than in the group
with moderate disease progression (7%). The 1-sided P
value for the difference between these 2 groups was
0.035, with 95% confidence intervals ranging from 0.71
to 0.99 (relative risk) and from 0.064 to 0.098 (odds
ratio), indicating statistical significance (Table 2). In
contrast, the DRB4 splice variant preventing the formation of a functional DRB4B gene product was not
associated with either of the patient groups (data not
shown). Five patients carried the combination
DRB1*04/*07 and were therefore excluded from the
splice variant and DRB4B promoter analyses since an
exact assignment of DRB4 to the corresponding DRB1
locus is not possible. It is important to note that all of the
DRB4 genes we analyzed share the same exon 2 sequence that forms the epitope-binding groove (data not
shown). The association between regulatory polymorphisms and disease progression is therefore not perturbed by additional structural polymorphisms.
Association of RA disease susceptibility with the
differential expression of HLA–DRB4. We next investigated whether the differential expression of the DRB4
gene also affected the susceptibility to RA. To that end,
Table 2. Correlation between an elevated expression of DRB4 and
radiographic progression in early RA*
RA patient group
HLA–DR allele
CRP ⱖ1.5 mg/dl at
study entry (n ⫽ 35)
DR4, 49 (34)
DR7, 16 (11)
DR9, 3 (2)
DR4, 50 (38)
DR7, 16 (12)
DR9, 5 (4)
CRP ⬍1.5 mg/dl at
study entry (n ⫽ 38)
Association of
DRB1*04, *07,
and *09 alleles
with DRB4B†
22 (10/46)
7 (3/46)
* Values are the frequency (no. of positive alleles or no. of positive
alleles/total). RA ⫽ rheumatoid arthritis; CRP ⫽ C-reactive protein.
† The difference between the 2 groups is 15%. One-sided P ⫽ 0.035,
with confidence intervals ranging from 0.71 to 0.99 (relative risk) and
from 0.064 to 0.98 (odds ratio), indicating statistical significance.
Figure 5. Differential association of HLA–DRB1*04, DRB1*07, and
DRB1*09 alleles with the HLA–DRB4 splice variant, as determined by
amplification and sequencing of the respective gene segments in
rheumatoid arthritis (RA) patients and controls. ⴱ ⫽ P ⫽ 0.032.
28 healthy DR4-, DR7-, or DR9-positive controls were
included in the analysis, and we compared the associations with the DRB4B promoter and splice variant in the
RA patients with those in the controls. In the RA
patients, a total of 71 DRB1*04, 21 DRB1*07, and 5
DRB1*09 alleles linked to the DRB4A promoter were
analyzed for an association with the splice variant. In the
controls, we analyzed 9 DRB1*04, 14 DRB1*07, and 1
DRB1*09 alleles. As shown in Figure 5, the frequencies
of both the DRB1*07- and DRB1*09-associated splice
variants were comparable among the RA patients and
the controls. In contrast, the frequencies for the
DRB1*04-associated splice variant differed significantly
between the RA patients (1.4%) and the controls (22%)
(1-sided and 2-sided P ⫽ 0.032).
The frequencies of associations between the
DRB1*04, *07, and *09 alleles and the DRB4B promoter were also comparable between the RA patients
and controls. Again, the DRB1*07 allele showed the
highest frequencies (data not shown).
Lack of augmentation of the differential expression of HLA–DR genes by the Y, S, and CAAT boxes.
The alignment of the Y, S, and CAAT boxes of the
DRB1*04, DRB1*07, DRB1*09, and DRB4 promoters
revealed several polymorphic residues in and around the
various boxes (11), but there was only 1 G3 A substitution 5⬘ of the Y box that distinguished the DRB4A from
the DRB4B promoter. However, this substitution did
not lead to functional consequences, since the binding of
BJAB nuclear factors to the DRB4A and DRB4B Y box
oligonucleotides showed comparable increases (14%
and 15%, respectively). Likewise, there was a 22% and a
30% increase for the THP-1 nuclear factors compared
with the random oligonucleotide (data not shown).
Additional structural differences between the Y, S, and
CAAT boxes of the DRB1*04, *07, and *09 promoters
consisted of 2 single substitutions within the S box and 3⬘
of the CAAT box. These substitutions resulted in only
minor differences in the binding of nuclear factors to the
S and CAAT box oligonucleotides (14–24% and 1–9%,
respectively) (data not shown).
We investigated the differential expression of
HLA class II genes as a mechanism involved in the
susceptibility and/or progression of RA, concentrating
on the molecular analysis of the RA-associated DR4, as
well as the neutral DR7 and DR9, superhaplotypes. We
compared the various promoters encoded by these haplotypes in an effort to find sequence polymorphisms
resulting in functional consequences. Assuming that the
specific binding of nuclear factors to the various promoter sequences was related to the transcription activity, we used surface plasmon resonance technology. The
promoter sequences we tested correspond to the known
S, X, Y, CAAT, and TATA boxes (25), and the nuclear
factors were derived from the human monocyte and
lymphoma cell lines THP-1 and BJAB (26), respectively.
Our data suggest that IFN ␥ -induced upregulation of surface HLA–DR expression in the monocyte cell line is mediated via the X box, since stimulation
with IFN␥ led to an increased binding of THP-1 nuclear
factors to the various X box oligonucleotides alone
(Figure 3). Our results are therefore consistent with
previous speculations that the overall expression level, as
well as the on/off switch, is regulated via the locusspecific X1 box, which binds the unique RFX proteins
and CIITA (15). The identification of binding sites for
IFN␥-inducible transcription factors in the CIITA promoter (14) further supports our findings. Moreover, the
increased binding of THP-1 nuclear factors to the X
boxes was strongest for the DRA-specific sequence,
which supports the previously held view that DRA
expression is the rate-limiting step in HLA–DR surface
expression (16). In BJAB cells, only a minor upregulation of surface HLA–DR expression in response
to IFN␥ stimulation was observed. The small shift in the
mean fluorescence intensity (Figure 2) correlated with a
slightly increased binding of nuclear factors to the X box
of DRA (Figure 3) and to the TATA box of the DRB1
alleles (Figure 4). The observed incapacity of BJAB cells
to up-regulate surface HLA–DR expression may be
characteristic of this particular cell line, but the situation
is also reminiscent of that of peripheral blood B lymphocytes, which have recently been described to be
refractory to stimulation by LPS and phorbol myristate
acetate (20).
From our data, it can be concluded that monocytes are very important for the response to environmental stimuli since they are highly activational and capable
of affecting the Th1/Th2 balance by differentially expressing HLA class II genes. Likewise, the differential
expression of murine class II MHC genes has been
shown to be restricted to macrophages and to be associated with protective/suppressive effects via the T cell
cytokine profile (5). One approach to addressing the
impact of differential HLA–DRB4 expression on the T
cell cytokine profile in human RA will involve the
analysis of rheumatoid factor isotypes in correlation with
the respective DRB4 promoters.
Interestingly, our data suggest a differential expression of the relatively nonpolymorphic DRB4 gene,
while the highly polymorphic DRB1 alleles seem to be
expressed at comparable levels. From an evolutionary
point of view, the retention of a second functional, yet
coding, nonvariable DRB gene has for a long time
remained an enigma. However, the concept that the
DRB4 gene regulates the immune response via differential expression is intriguing. Our data thus indicate
that selection may also work on the regulatory regions,
which were previously thought to be merely flanking
sequences in linkage disequilibrium with the coding
Independent of stimulation, there was a 1.6–1.8fold increase in the binding activity of nuclear factors to
the TATA box DRB4B oligonucleotide compared with
the TATA box DRB4A oligonucleotide. Even though
this difference seems insignificant at first glance, it was
confirmed by EMSA as an alternative assay (Figure 4).
A 2-fold difference in promoter activity can well be
reconciled with previously published results by Leen and
Gorski (18), who reported 3-fold and 5-fold higher levels
for the DRB4B-driven pre-mRNA and mRNA, respectively. Measuring the surface plasmon resonance yielded
little evidence for an involvement of the S, Y, and CAAT
boxes in the differential expression of the DRB1 and
DRB4 genes we tested. However, the use of recombinant transcription factors rather than nuclear extracts
may result in higher sensitivities.
Contrary to the overall level of expression, the
fine-tuning of the DRB4A and DRB4B promoters is
regulated at a ubiquitous regulatory element, the TATA
box. We therefore assume that the differential expression of the DRB4 gene will not be restricted to the
classic antigen-presenting cell, including the dendritic
cells, but will hold true for any cell expressing HLA class
II genes. Interesting candidates include the synoviocytes,
which are known to participate in the autoimmune
process in RA.
In summary, we would like to propose a model
whereby both the susceptibility to the development of
RA and the progression of RA depend upon the differential expression of the relatively nonpolymorphic
HLA–DRB4 gene. Upon grouping the RA patients
according to their CRP levels at study entry (which, in
this cohort, was the strongest indicator of disease progression), the more potent DRB4B promoter segregated
with radiographic progression (Table 2). Interestingly,
the frequency of DRB1*0401 alleles was almost identical
in RA patients with high and low initial CRP values,
ruling out an effect of epitope selection on disease
progression. In addition, there was a concomitant reduction in the splice variant frequency associated with
DRB1*04 alleles among RA patients, indicating a protective effect of low DRB4 expression. Our speculations
about the mechanism include the enhanced signaling
between antigen-presenting cells and T cells due to
enhanced antigen presentation, favoring the development of inflammatory responses. Comparable shifts
toward Th1 responses have previously been reported to
result from increases in antigen dose and affinity, respectively (27,28).
1. Lang B, Melchers I, Urlacher A, Tanzi-Fetta RF, Kohlbrenner S,
Tongio MM, et al. HLA-DR1 and DRw6 association in DR4negative rheumatoid arthritis patients. Rheumatol Int 1990;10:
2. Listing J, Rau R, Muller B, Alten R, Gromnica-Ihle E, Hagemann
D, et al, for the Berlin Collaborating Rheumatological Study
Group. HLA-DRB1 genes, rheumatoid factor, and elevated C-reactive protein: independent risk factors of radiographic progression in early rheumatoid arthritis. J Rheumatol 2000;27:2100–9.
3. Weyand CM, McCarthy TG, Goronzy JJ. Correlation between
disease phenotype and genetic heterogeneity in rheumatoid arthritis. J Clin Invest 1995;95:2120–6.
4. Gregersen PK, Silver J, Winchester RJ. The shared epitope
hypothesis: an approach to understanding the molecular genetics
of susceptibility to rheumatoid arthritis. Arthritis Rheum 1987;30:
5. Baumgart M, Moos V, Schuhbauer D, Muller B. Differential
expression of major histocompatibility complex class II genes on
murine macrophages associated with T cell cytokine profile and
protective/suppressive effects. Proc Natl Acad Sci U S A 1998;95:
6. Janitz M, Mitchison A, Reiners-Schramm L, Lauster R. Polymorphic MHC class II promoters exhibit distinct expression patterns in
various antigen-presenting cell lines. Tissue Antigens 1997;49:
7. Cowell LG, Kepler TB, Janitz M, Lauster R, Mitchison NA. The
distribution of variation in regulatory gene segments, as present in
MHC class II promoters. Genome Res 1998;8:124–34.
8. O’hUigin C, Satta Y, Hausmann A, Dawkins RL, Klein J. The
implications of intergenic polymorphism for major histocompatibility complex evolution. Genetics 2000;156:867–77.
Mitchison NA, Roes J. Patterned variation in murine MHC
promoters. Proc Natl Acad Sci U S A 2002;99:10561–6.
Marsh SG, Bodmer JG, Albert ED, Bodmer WF, Bontrop RE,
Dupont B, et al. Nomenclature for factors of the HLA system,
2000. Hum Immunol 2001;62:419–68.
Louis P, Eliaou JF, Kerlan-Candon S, Pinet V, Vincent R, Clot J.
Polymorphism in the regulatory region of HLA-DRB genes correlating with haplotype evolution. Immunogenetics 1993;38:21–6.
Sutton VR, Kienzle BK, Knowles RW. An altered splice site is
found in the DRB4 gene that is not expressed in HLA-DR7,Dw11
individuals. Immunogenetics 1989;29:317–22.
Fontes JD, Jiang B, Peterlin BM. The class II trans-activator
CIITA interacts with the TBP-associated factor TAFII32. Nucleic
Acids Res 1997;25:2522–8.
Moreno CS, Beresford GW, Louis-Plence P, Morris AC, Boss JM.
CREB regulates MHC class II expression in a CIITA-dependent
manner. Immunity 1999;10:143–51.
Mitchison NA. Genetic variation in the MHC class II promoter:
lessons for regulation and for comparative genomics. Folia Biol
Louis P, Vincent R, Cavadore P, Clot J, Eliaou JF. Differential
transcriptional activities of HLA-DR genes in the various haplotypes. J Immunol 1994;153:5059–67.
Vincent R, Louis-Plence P, Gaillard F, Clot J, Eliaou JF. Qualitative and quantitative analysis of HLA-DRB gene expression.
J Rheumatol 1997;24:225–6.
Leen MP, Gorski J. DRB4 promoter polymorphism in DR7
individuals: correlation with DRB4 pre-mRNA and mRNA levels.
Immunogenetics 1997;45:371–8.
Kerlan-Candon S, Louis-Plence P, Wiedemann A, Combe B, Clot
J, Eliaou JF, et al. Specific overexpression of rheumatoid
arthritis–associated HLA–DR alleles and presentation of lowaffinity peptides. Arthritis Rheum 2001;44:1281–92.
Czerwony G, Alten R, Gromnica-Ihle E, Hagemann D, Reuter U,
Sorensen H, et al. Differential surface expression of HLA-DRB1
and HLA-DRB4 among peripheral blood cells of DR4 positive
individuals. Hum Immunol 1999;60:1–9.
Arnett FC, Edworthy SM, Block DA, McShane DJ, Fries JF,
Cooper NS, et al. The American Rheumatism Association 1987
revised criteria for the classification of rheumatoid arthritis.
Arthritis Rheum 1988;31:315–24.
Sykes BC. DNA in heritable disease [letter]. Lancet 1983;2:787–8.
Scharf SJ, Griffith RL, Erlich HA. Rapid typing of DNA sequence
polymorphism at the HLA-DRB1 locus using the polymerase
chain reaction and nonradioactive oligonucleotide probes. Hum
Immunol 1991;30:190–201.
Blaesing F, Weigel C, Welzeck M, Messer W. Analysis of the
DNA-binding domain of Escherichia coli DnaA protein. Mol
Microbiol 2000;36:557–69.
Benoist C, Mathis D. Regulation of major histocompatibility
complex class-II genes: X, Y and other letters of the alphabet.
Annu Rev Immunol 1990;8:681–715.
Clements GB, Klein G, Povey S. Production by EBV infection of
an EBNA-positive subline from an EBNA-negative human lymphoma cell line without detectable EBV DNA. Int J Cancer
Hemmer B, Stefanova I, Vergelli M, Germain RN, Martin R.
Relationships among TCR ligand potency, thresholds for effector
function elicitation, and the quality of early signaling events in
human T cells. J Immunol 1998;160:5807–14.
Ise W, Totsuka M, Sogawa Y, Ametani A, Hachimura S, Sato T,
et al. Naive CD4⫹ T cells exhibit distinct expression patterns of
cytokines and cell surface molecules on their primary responses to
varying doses of antigen. J Immunol 2002;168:3242–50.
Без категории
Размер файла
145 Кб
class, expressions, progressive, associates, hla, differential, arthritis, disease, genes, susceptibility, rheumatoid
Пожаловаться на содержимое документа