close

Вход

Забыли?

вход по аккаунту

?

Patterns of antihistone antibody specificity in systemic rheumatic disease. I. Systemic lupus erythematosus mixed connective tissue disease primary sicca syndrome and rheumatoid arthritis with vasculitis

код для вставкиСкачать
285
PATTERNS OF ANTIHISTONE ANTIBODY
SPECIFICITY IN SYSTEMIC RHEUMATIC DISEASE
I. Systemic Lupus Erythematosus, Mixed Connective Tissue Disease,
Primary Sicca Syndrome, and Rheumatoid Arthritis with Vasculitis
ROBERT M. BERNSTEIN, RODERICK N. HOBBS, DONALD J. LEA,
DONALD J. WARD, and GRAHAM R. V. HUGHES
A new fluorimetric assay was used to measure the
relative amounts of antibodies to individual nuclear
histones in sera from 102 patients with systemic lupus
erythematosus (SLE), mixed connective tissue disease,
primary sicca syndrome, and rheumatoid arthritis with
vasculitis. In SLE sera, the predominant responses were
to histones H-1, H-2B, and H-3, with marked elevations
of binding to H-1 and H-2B in one-third of the patients,
and to H-3 in one-fourth; antibodies of both the IgG and
IgM classes were also detected. In a few SLE sera, the
pattern of histone response differed or was restricted to
1 immunoglobulin class. In mixed connective tissue
disease, only 2 of 9 sera showed elevated histone binding
activity, the response being predominantly to H-3 in 1
patient and to H-1 and H-2B in the other. Binding to H2B was also prominent in 2 of 3 patients with primary
sicca syndrome. The highest antihistone reactivity and
the most heterogeneous response patterns were observed
in patients who had rheumatoid arthritis with vasculitis;
6 of 8 of those sera had elevated histone reactivity. In
SLE, the highest histone binding results were found
among patients with a history of photosensitivity. His-
From the Rheumatology Unit, Hammersmith Hospital,
London, and the Rheumatology Department, Robert Jones and
Agnes Hunt Orthopaedic Hospital, Oswestry, Shropshire, England.
Robert M. Bernstein, MA, MD, MRCP: Hammersmith
Hospital; Roderick N. Hobbs, PhD: Robert Jones and Agnes Hunt
Orthopaedic Hospital; Donald J. Lea, PhD: Robert Jones and Agnes
Hunt Orthopaedic Hospital; Donald J. Ward MB, FRCP: Robert Jones and Agnes Hunt Orthopaedic Hospital; Graham R. V.
Hughes, MD, FRCP: Hammersmith Hospital.
Address reprint requests to Dr. R. M. Bernstein, Rheumatism Research Unit, Manchester Royal Infirmary, Oxford Road,
Manchester MI3 9WL, England.
Submitted for publication September 14, 1983; accepted in
revised form September 13, 1984.
Arthritis and Rheumatism, Vol. 28, No. 3 (March 1985)
tones are closely associated with DNA in the nucleosome, and we speculate that antihistone antibodies could
arise as a result of damage to DNA, induced by drugs or
irradiation.
Autoantibodies reactive with cellular components are a characteristic finding in systemic rheumatic disease. Among the antigens best characterized are
DNA and the histones. The histones are a group of
similar, small, highly conserved nuclear proteins (1)
that bind to DNA by virtue of their many basic
residues. They can be prepared in high purity as 5
distinct molecular fractions: H-1 , H-2A, H-2B, H-3,
and H-4.
The presence of antihistone antibodies has been
reported in idiopathic and drug-induced systemic lupus erythematosus (SLE) (2-5), rheumatoid arthritis
(RA) (5,6), and other conditions (7), and they are
thought to be responsible for the lupus erythematosus
cell phenomenon. We have developed a solid-phase
fluorimetric assay for the measurement of antihistone
antibodies, using highly-purified histone fractions (8).
In this paper we report the diverse patterns of antihistone reactivity seen in different connective tissue
diseases, emphasizing the predominant response to
histones H-1, H-2B, and H-3 in SLE.
PATIENTS AND METHODS
Patients. We studied 82 patients with SLE fulfilling
the American Rheumatism Association classification criteria
(9), 9 patients with mixed connective tissue disease (MCTD)
defined by the presence of anti-RNP antibody and high-titer
speckled antinuclear antibody (ANA) in the absence of
antibodies to Sm and with normal DNA binding (lo), and 8
patients who had RA with vasculitis, presenting as cutan-
286
eous purpura and ulceration in 6 and mononeuritis multiplex
in 5. Also studied were 3 patients who had primary sicca
syndrome with keratoconjunctivitis sicca and salivary gland
lymphoid follicles observed on lip biopsy; all 3 had rheumatoid factor and precipitating antibodies to Ro, and 2 had antiLa. Eight healthy controls were studied. Sera were stored at
-20°C and examined with the investigator having no knowledge of the diagnosis.
Antihistone antibody assay. Purified individual histones were prepared using standard methods (11) and further
purified by Biogel P-100. Histones H-1 and H-3 were from
pig thymus; H-2A, H-2B, and H-4 were from calf thymus
(Figure 1). In addition, whole calf thymus histones (Sigma,
St. Louis, MO) were used in a screening study comparing
the overall antihistone response with clinical features and
HLA antigens. The assay buffer contained 3.02 g d i t e r boric
acid, 1.87 g d i t e r disodium tetraborate (Na2B4O7-IOHzO),
and 100 mM sodium chloride. Tween 20 (0.1%) was added to
make up Tween buffer, a solution that minimizes nonspecific
protein binding (12). The fluorescein-labeled sheep antisera
to whole human IgG and IgM (Wellcome Reagents Ltd.) had
fluorescein/protein molar ratios of about 3.5.
The assay was performed as described in detail
previously (8). Polystyrene enzyme immunoassay cuvettes
(Gilford, Oberlin, OH) were coated with tyrosineglutamic
acid copolymer with an average molecular weight of 66,000
(TGA; Miles-Yeda Ltd., Rehovot, Israel) by adding 100 pl of
a 100-pg/ml solution in borate buffered saline to the cuvette
bottom and tapping gently to ensure complete coverage.
Cuvettes were sealed with Parafilm M (American Can Co.,
Greenwich, CT), left overnight at 20"C, and washed thoroughly with distilled water.
Individual histones were coated onto the prepared
surfaces by overnight incubation of 100-pl volumes of histone (50 pg/ml) in borate buffered saline. Unbound protein
was washed away quickly with buffer. Duplicate samples of
5% serum in Tween buffer were added to the cuvettes and
left overnight at 20°C. Cuvettes were washed 3 times with
Figure 1. Purified histones used in the fluorimetric assay, displayed
on a 6M urea/l5% polyacrylamide gel. Left to right: H-1, H-2A, H2 8 , H-3 (with a trace of dimer), and H-4 (showing the mono-, di-,
tri- , and tetra-acetylated forms).
BERNSTEIN ET AL
Tween buffer, and a 100-pl solution containing a mixture of 1
volume fluorescein-labeled antiserum, 3 volumes normal
sheep serum, and 56 volumes Tween buffer was added.
After incubation for 5 hours and then washing 3 times
with Tween buffer, bound conjugate was released from the
cuvette surface by addition of 750 pl of 0.2% sodium dodecyl
sulfate (SDS) in 0.1M NaOH. The fluorescence of this
solution was measured in a Locarte fluorimeter (13), using
an SDS-NaOH solution blank. The fluorimeter was adjusted
to read 1.000 with a 200-fold dilution of the mixture containing labeled antiserum in the SDS-NaOH solution. All binding values were the mean of estimations made in duplicate.
An upper limit of normal was set arbitrarily at 2 standard
deviations above the mean of the values for healthy controls.
Antibodies to DNA and soluble cellular antigens. Antibody to double-stranded DNA was measured by the Farr
assay (14). Antibodies to acidic soluble cellular antigens
were detected by counterimmunoelectrophoresis using extracts of rabbit thymus and human spleen (15).
HLA typing and clinical correlation in SLE. HLA-A,
B, C, and DR antigens were determined in 64 SLE patients,
and in all 12 patients with MCTD or primary sicca syndrome,
by a microcytotoxicity assay (16) using antisera standardized
against sera of the 8th Histocompatibility Workshop (17).
Medical records of 64 SLE patients were reviewed retrospectively, and the following individual features were recorded: discoid lupus, facial rash, photosensitivity (lightinduced rash or flare of lupus activity), alopecia, mouth
ulcers, livedo reticularis, Raynaud's phenomenon, digital
ischemia or vasculitis, arthritis, myositis, tendinitis, pleurisy, pericarditis, dyspnea, abnormal pulmonary function test
results, proteinuria (>0.5 gm in 24 hours), urine granular or
red cell casts, renal biopsy findings, hypertension, noninfective fever, sicca syndrome (dryness on Schirmer's test, with
wetting <5 mm in 5 minutes, and/or salivary lymphoid
follicles on lip biopsy), seizures, psychosis, depression,
thrombosis, stroke, abortion, splenomegaly, lymphadenopathy, raised IgG, low IgA, low C3, low C4, anemia (<lo
gm/dl), leukopenia (<4.0 x 109/liter), lymphopenia, thrombocytopenia (<lo0 x lO9/liter on 2 occasions unrelated to
therapy), steroid therapy, cytotoxic therapy, antimalarial
therapy, Coombs' test result, rheumatoid factor, lupus anticoagulant, biologic false-positive test result for syphilis, and
antibodies to nuclei DNA, Sm, RNP, Ro, La, SL, proliferating cell nuclear antigen, ribosomal RNP, and PL-4 (18).
For the purposes of this analysis, whole histone
binding results were defined as normal (<0.1 fluorimeter
units), moderately elevated (0.1-0.2), or high (>0.2), and the
results were compared with the presence or absence of
clinical features and HLA antigens by chi-square tests with 2
degrees of freedom (do. The 1 statistically significant correlation obtained was then examined further by Student's ttest using the binding results for each of the purified histones.
RESULTS
Controls. In the absence of serum, very little
fluorescent conjugate was bound to the histone-TGA-
HISTONE ANTIBODIES IN RHEUMATIC DISEASE
287
Table 1. Fluorimeter readings of 8 healthy control sera and comparison with 13 systemic lupus
erythematosus (SLE) sera selected for results just above the arbitrary normal range
Purified histones
Normal sera
1
2
3
4
5
6
7
8
No serum
Mean of 8 normal sera
SD
Mean + 2 SD
(arbitrary upper end
of normal range)
Mean of 13 SLE sera
SD
H- 1
H-2A
H-2B
H-3
H-4
0.111
0.105
0.094
0.082
0.091
0.108
0.085
0.092
0.012
0.096
0.010
0.116
0.096
0.078
0.091
0.068
0.087
0.100
0.058
0.060
0.009
0.080
0.015
0.110
0.078
0.069
0.073
0.053
0.062
0.075
0.065
0.086
0.016
0.070
0.010
0.090
0.098
0.085
0.099
0.106
0.086
0.111
0.069
0.107
0.015
0.095
0.013
0.121
0.083
0.084
0.094
0.063
0.080
0.096
0.062
0.076
0.008
0.080
0.012
0.104
0.131
0.031
0.095
0.029
0.134
0.031
0.123
0.035
0.114
0.028
coated cuvettes, and fluorimeter readings were about
0.01. Readings for 8 normal sera averaged approximately 0.08 for individual histone fractions (Table l ) ,
suggesting that low levels of antihistone antibodies are
a normal finding, although the possibility of some
nonspecific Fc binding of IgG is not excluded.
SLE patients. Figure 2 shows SLE sera binding
values >2 standard deviations above the mean values
0.7
-
0
I
I
I
0.6-
1
1
I
0.5m
.-c
V
E 0.4-
L
L
QJ
c
a
l
.-E
5
-
0.3-
LL
0.2-
.....
0.1-
0 '
Figure 2. Binding values to the 5 individual histones in sera from 82
patients with systemic lupus erythematosus. Elevated values were
defined as >2 standard deviations above the mean values of 8
normal sera.
I
I
I
'0
I
i
H1
H2A
H2B
H3
H4
Figure 3. Typical bindisg patterns to the 5 individual histones in 6
systemic lupus erythematosus sera, showing greatest reactivity with
histones H-1, H-2B, and H-3. One serum (0- -0) is unusual in
that it shows mainly H-3 reactivity.
BERNSTEIN ET AL
of the normal sera, for each of the individual histones.
Binding to H-1 was elevated in 51 (62%), to H-2A in 35
(43%), to H-2B in 68 (83%), to H-3 in 47 (57%), and to
H-4 in 44 (54%) of the 82 SLE sera tested. Although
many SLE sera gave results only slightly above the
normal range, a more marked elevation (>0.2 fluorimeter units) of binding to H-1 and H-2B was seen in
about 32% of sera, to H-3 in 24% of sera, and to H-2A
and H-4 in 10%.
Typical patterns of binding to the 5 individual
histones are shown in Figure 3. In almost all cases,
binding to histones H-1, H-2B, and H-3 was greater
than binding to H-2A or H-4. As shown in Table 1, this
pattern of reactivity was apparent even in SLE sera
showing only modest binding (0.1-0.2 units), in contrast with the normal sera, where H-2B binding was
lowest. A few SLE sera showed a more restricted
pattern of binding to just 1 or 2 histones, such as to H-1
and H-2B or only H-3. Histone H-4 appeared to be
least immunogenic, and antibodies to H-4 were predominant in only 1 patient.
MCTD patients. In keeping with the restricted
nature of the antinuclear antibody response implicit in
the serologic definition of MCTD (lo), histone binding
was normal in 7 of 9 MCTD sera tested. However,
binding was elevated in 2 patients (Figure 4). The
predominant response was to H-3 in 1 patient with
photosensitive rash, alopecia, and Raynaud's phenomenon, and to H-1 and H-2B in a patient with Raynaud's
phenomenon, depression, telangiectasia, and active
ulcerative colitis. The latter patient was receiving
therapy with salazopyrine, a drug capLble of inducing
ANA. All patients with normal histone binding activity
had severe Raynaud's phenomenon, 2 had myositis
and sicca syndrome, 1 had erosive arthritis and telangiectasias, and 2 had recurrent pleuritic pain controlled by low-dose prednisone.
Primary sicca syndrome patients. Histone binding was elevated in 2 of the 3 sera from patients with
primary sicca syndrome. In 1 serum the response was
to H-1 and H-2B; in the other it was to all histones
except H-2A, with strikingly high binding to H-2B
(Figure 4).
RA patients with vasculitis. Histone binding
values were greatly elevated in 6 of the 8 sera from
patients who had RA with vasculitis. The binding
pattern was much more heterogeneous than in SLE,
and very high binding to H-2A and H-4 was observed
in this group (Figure 5 ) . Titers of IgM rheumatoid
factor were high in all cases, ranging from 160-1,280,
1.57
L
01
%
0 . 6i
-
.-E
0
I
-3
Y
0.4
-
0.2
-
0-
I
I
I
I
1
H1
H2A
H2B
H3
H4
+ 1' sicca syndrome
MCTD
Figure 4. Histone reactivity patterns of sera from 2 patients with
mixed connective tissue disease (MCTD) and 2 with primary sicca
syndrome. Seven additional MCTD sera and 1 serum from a patient
with orimarv sicca svndrome gave normal results.
c
CT,
'
e 0.8m
L
01
01
L
c
b
0.6-
3
-
I&
0.4-
0.2-
...%
.:'
-f..
/--7-*'
.....
.....
.....
H1
H2A
H2B
H3
H4
Figure 5. Heterogeneous histone binding patterns of sera from 6
patients with rheumatoid vasculitis. Two additional sera showed
normal binding.
HISTONE ANTIBODIES IN RHEUMATIC DISEASE
289
Table 2. Histone binding values in 60 systemic
erythematosus patients, by history of photosensitivity
History of
photosensitivity
Histone
~
but there was no correlation with the level, pattern, or
immunoglobulin class of histone binding.
IgG and IgM antihistone responses. Eleven sera
with high binding values were selected for separate
examination of the IgG and IgM responses. In SLE
(Figure 6), histone-binding antibodies were predominantly of the IgM class in 4 of 6 sera tested, and
exclusively IgM in the 1 serum showing mainly antiH-3 specificity. Only the sixth serum with predominantly anti-H-2B and anti-H-4 reactivity contained
mainly IgG antibodies. In rheumatoid vasculitis, a
greater proportion of the antibodies were of the JgG
class. Only 1 of these sera showed clearly different
P
(Student,s
Mean
SD
Mean
SD
t-test)
0.241
0.135
0.289
0.161
0.143
0.189
0.138
0.088
0.137
0.128
0.101
0.068
0.033
0.073
0.072
0.035
<0.005
c0.005
<0.005
<0.05
<0.05
~~
H- 1
H-2A
H-2B
H-3
H-4
Figure 6. IgG and IgM binding patterns of sera from 6 systemic lupus
erythematosus patients. 0- -0 is the same H-3-reactive serum
shown in Figure 3. IgM reactivity predominates in 5 sera, and IgG
reactivity in 1 (0-0).
No history of
photosensitivity
lupus
0.084
0.279
0.078
0.111
patterns of histone reactivity when the IgG and IgM
responses were compared (Figure 7).
Clinical-serologic correlation in SLE. In SLE
sera, elevated binding to total histones showed no
correlation with clinical features, HLA antigens, abnormal laboratory findings, or other autoantibodies,
1.3-
.
1.21.1-
1.0-
c3
0.90.8-
4
0'70'6W
I-
-:O m s -
5 0'4-
.
.
:
3
0*3-
.*
.
.
0.2 0-1HI
I
I
I
,
I
H2A
H25
H3
H4
H1
1
I
I
I
H2A
H2E
H3
H4
Figure 7. IgG and IgM binding patterns of sera from 5 rheumatoid
vasculitis patients. The pattern of binding differed in I serum
(A-A).
IgG binding predominated in the highest binding serum,
and in one other (0-0).
I
1
I
A
B
Figure 8. Histone H-2B binding values in systemic lupus erythemat o w s patients, according to the presence (A) or absence (B) of a
history of photosensitivity.
BERNSTEIN ET AL
290
except for a possible association with a history of
photosensitivity (x2 = 9.2, 2 df, P = 0.01), but even
this was not significant when corrected for the number
of chi-square tests carried out. Nevertheless, the relationship between photosensitivity and elevated binding values for the individual histones was apparent on
analysis by Student’s t-test (Table 2). This relation is
illustrated in Figure 8, which shows H-2B binding; all 6
patients with highest histone binding had histories of
photosensitivity. Among the patients included in the
clinical review, these 6 showed the highest binding to
all histones, and an additional 2 patients had particularly high H-1 binding. All 8 patients were among the
30 with a history of photosensitivity, although this was
not particularly severe or recent at the time blood was
drawn for the study.
DISCUSSION
The pattern or relative magnitude of the immune response to the 5 histone proteins described in
this study shows differences between SLE and the
other rheumatic conditions examined. Further patterns are emerging among the antihistone antibodies
induced by drugs (ref. 19 and Hobbs RN, Lea DJ,
Bernstein RM: manuscript in preparation).
In SLE, the predominant responses were usually to H-1, H-2B, and H-3, but not to H-2A or H-4.
Only 1 of the 82 SLE sera showed markedly elevated
binding to a single histone (H-3), and this patient was
unusual in that features of SLE were newly developing
after a long prodrome of idiopathic thrombocytopenic
purpura. In a serial study of patients with SLE (8), we
found that the pattern of histone reactivity remains
stable, although absolute levels may fluctuate with
disease activity. Using the immunoblot technique,
Hardin and Thomas (20) found antihistone antibodies
in 11 of 24 SLE sera, with anti-H-1 and anti-H-2B
predominant in most cases. We found above-normal
levels of binding to 1 or more histones in 83% of the
SLE patients tested, and marked elevations in 32% (as
is readily apparent in Figure 2). This is in keeping with
previous findings on antihistone antibodies in 30-60%
of SLE sera (2,5,20,21).
Although overlap syndromes abound, the concept of MCTD remains controversial. The condition
was originally defined by the presence of anti-RNP
antibodies in the absence of other types of ANA
(10,22), yet 2 of our 9 MCTD patients had elevated
histone binding, and 2 others had antibody to the
cellular antigen Ro (SS-A). Thus, it would appear
unwise to adhere too rigidly to the serologic definition
of MCTD, both because additional anticellular antibody specificities may be defined in these sera and
because anti-RNP-negative overlap syndromes can be
so clinically similar (22,23).
Although the highest binding values were found
in serum from patients with rheumatoid vasculitis,
these measurements are complicated by the possibility
of rheumatoid factors binding either to the specific IgG
antihistone antibody or to histones directly. A subset
of rheumatoid factors does cross-react with histones
(6). However, using the present assay, we demonstrated cross-reactivity only when rheumatoid factors were
isolated from serum (24), and ANA-negative, rheumatoid factor-positive RA sera showed normal antihistone reactivity (8).
In normal sera, the low level of immunoglobulin
binding measured in our assay may represent true
antihistone antibody or nonspecific binding. The presence of autoantibodies in normal serum is well-documented for ANA (25), rheumatoid factor (26), antimitochondria1 antibodies (27), and immune complexes
(28). In SLE, many results were clustered just above
the normal range, which we defined as 2 standard
deviations above the mean of healthy control values.
These slightly elevated values could reflect the elevated immunoglobulin levels often found in SLE, but as
evidence against that possibility, histone binding was
mostly normal in MCTD, where serum IgG is also
often elevated. That the higher binding values measured in the patients’ sera represent specific antibodies
is supported first by the finding of various patterns of
histone reactivity, and second by the restriction of
antibody class in some cases.
Several assays for the detection of antihistone
antibodies in SLE have been described, including (a)
immunofluorescence after reconstitution of acid-extracted tissue sections with purified histones (2-4), (b)
solid-phase radioimmunoassays (5,19,29), (c) enzymelinked immunosorbent assays (ELISA) using a solid
phase of polystyrene (21,30) or glass (31), and (d)
immunoblot techniques (20). These methods have given conflicting results. For example, by an ELISA
method (31), antibodies were found to be restricted to
H-1 and H-2B in some cases, whereas by radioimmunoassay (3,little antibody to these histones was
detected.
The fluorimetric assay developed for our studies is simple, specific, and sensitive, and gives results
that are in substantial agreement with the findings of
the 2 most recent reports (20,21). Specificity is ensured by the use of well-defined, highly-purified histones, while sensitivity is enhanced by the increased
HISTONE ANTIBODIES IN RHEUMATIC DISEASE
amount of histone antigen bound to the solid phase via
the TGA copolymer (8). Although antibody binds
nonspecifically to naked TGA, this binding is greatly
diminished when the TGA is saturated with histones;
indeed, normal sera show lower reactivity with histone-TGA than with naked TGA (8). In the absence of
TGA, the sensitivity of the assay is lower, but the
patterns of histone reactivity are unchanged.
There is one interesting exception. As shown in
Figure 9, reactivity of antibody with histone H-4 was
sometimes undetectable when H-4 was adsorbed directly onto the polystyrene. This observation suggests
that antigenic sites on H-4 are restricted to the hydrophobic C-terminal region that is probably hidden when
the protein binds directly to plastic. H-4 is known to
undergo major conformational changes on binding to
DNA or other histones (32,33), and TGA may have a
similar effect on the conformation of H-4, which could
be important to its recognition by antibody. To analyze in detail the antigenic site on this histone, we have
predicted likely epitopes from the amino acid sequence and synthesized the relevant polypeptides to
test with antibody.
The diverse patterns of antihistone reactivity
described in this report may reflect differences in the
underlying ability of each patient to mount an immune
3.2
-
2.8
-
2.4
-
2.0
-
1.6
-
1.2
-
0.8
-
0.4
-
0,
.-
p
?!
L
2
.-E
L
J
LL
29 1
response to the histones. However, other factors such
as differences in clearance and degradation cannot be
ignored. A similar differential responsiveness to the
closely-related Sm and RNP antigens appears to have
a racial basis, in that the anti-Sm specificity is nearly
10 times more common in black and Chinese patients
than in white patients (18). There are now several
examples of autoantibody specificities involving more
than 1 component of a nucleoprotein particle: antibodies are found simultaneously to DNA and several
histones in the nucleosome, to high molecular weight
RNA and 3 proteins in the ribosome (18), to the Ro and
La antigens (which, together, form complexes with
hyRNAs), and to several proteins in the soluble (U)
RNP particles (34). It seems possible that, in each
case, the immune response develops against the particle as a whole. In patients with SLE, MCTD, and
primary sicca syndrome, antibodies to the different
types of particle occur separately more often than
together. The antinuclear response in all these related
conditions is not merely polyclonal, but highly selective, and quite possibly, antigen-driven (18).
The highest antihistone reactivity in SLE was
found among patients with photosensitivity. Light
therapy with psoralens and ultraviolet A irradiation
damages DNA and may lead to ANA (but not antinative DNA) production (35). Therapy with hydralazine (a drug that binds to DNA) induces antibodies
directed against histones (16,36) and poly(ADP-ribose)
(37), but rarely against DNA. In SLE, the most
common antihistone response is to H-1 and H-2B
(20,21, and this report). Both these histones lie in
exposed positions on the nucleosome, again suggesting that the whole particle is the immunogen. In
experimental studies the immunogenicity of histones is
greatly enhanced by the addition of nucleic acid (38),
and the soluble cellular Jo-1 and La antigens can
interact with viral RNA (39). We would speculate that
DNA acts as an immunologic carrier to which T
lymphocytes are normally tolerant; damage to DNA
by drugs or ultraviolet irradiation (40) could lead to a
breakdown of tolerance, enabling anti-hapten responses to develop against DNA-associated molecules such
as histones.
ACKNOWLEDGMENTS
0 1 2 A 2 0
3
Histone alone
4
1 2 A 2 0 3
4
Histone-TCA
Figure 9. Immunoglobulin binding to tubes coated with purified
histones in the absence (left) or presence (right) of the tyrosineglutarnic acid (TGA) copolymer. Two examples are shown.
We are indebted to Dr. G. H. Goodwin for the kind
gift of highly-purified histones. We thank Dr. J. R. Batchelor
and A. Dodi for the HLA determinations, C. C. Bunn for the
DNA binding measurements, and Dr. Evelyn Hess for
helpful discussions.
BERNSTEIN ET AL
292
REFERENCES
1. Isenberg I: Histones. Annu Rev Biochem 48:159-191,
1979
2. Tan EM, Robinson J, Robitaille P: Studies of antibodies
to histones by immunofluorescence. Scand J Immunol
55311-818, 1976
3. Fritzler MJ, Tan EM: Antibodies to histones in druginduced and idiopathic lupus erythematosus. J Clin
Invest 62560-567, 1978
4. Grossman L , Barland P: Histone reactivity of druginduced antinuclear antibodies: a comparison of symptomatic and asymptomatic patients. Arthritis Rheum
24:927-931, 1981
5. Rubin RL, J o s h FG, Tan EM: A solid-phase radioimmunoassay for anti-histone antibodies in human sera:
comparison with an immunofluorescence assay. Scand J
Immunol 15:63-70, 1982
6. Aitcheson CT, Peebles C, J o s h F, Tan EM: Characteristics of antinuclear antibodies in rheumatoid arthritis:
reactivity of rheumatoid factor with a histone-dependent
nuclear antigen. Arthritis Rheum 23528-538, 1980
7. Kubokawa T, Ishikawa H: Antibody to histone commonly observed in pemphigus and connective tissue
disease. Arch Dermatol Res 273: 149-152, 1982
8. Hobbs RN, Lea DJ, Ward DJ: fluorimetric assay for
human antibodies to all the histones. J Immunol Methods 65:235-243, 1983
9. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ,
Rothfield NF, Schaller JG, Talal N, Winchester RJ: The
1982 revised criteria for the classification of systemic
lupus erythematosus. Arthritis Rheum 25: 1271-1277,
1982
10. Tan EM: Autoantibodies to nuclear antigens: their immunobiology and medicine. Adv Immunol 33: 167-240,
1982
11. Johns EW: The isolation and purification of histones,
Methods in Cell Biology. Edited by G Stein, J Stein, LJ
Kleinsmith. London, Academic Press, 1977, pp 183-203
12. Smith JA, Hurrell JGR, Leach SJ: Elimination of nonspecific adsorption of serum proteins by Sepharosebound antigens. Anal Biochem 87:299-305, 1978
13. Lea DJ, Ward DJ: Estimation of IgM rheumatoid factors
by fluorimetry. Ann Rheum Dis 37:247-251, 1978
14. Pincus T, Schur PH, Rose JA, Decker JL, Talal N:
Measurement of serum DNA binding activity in SLE. N
Engl J Med 281:701-705, 1969
15. Bernstein RM, Bunn CC, Hughes GRV: Identification of
antibodies to acidic antigens by counterimmunoelectrophoresis. Ann Rheum Dis 4134-555, 1982
16. Compston DHS, Batchelor JR, McDonald WI: B-lymphocyte alloantigens associated with multiple sclerosis.
Lancet 2:1261-1265, 1976
17. Terasaki P, editor: Histocompatibility Testing 1980. Los
Angeles, University of California Press, 1980
18. Bernstein RM, Bunn CC, Hughes GRV, Francoeur AM,
Mathews MB: Cellular protein and RNA antigens in
autoimmune disease. Mol Biol Med (in press)
19. Portanova JP, Rubin RL, J o s h FG, Agnello VD, Tan
EM: Reactivity of anti-histone antibodies induced by
procainamide and hydralazine. Clin Immunol Immunopathol 25:67-79, 1982
20. Hardin JA, Thomas JO: Antibodies to histones in systemic lupus erythematosus: localization of prominent
autoantigens on histones HI and H2B. Proc Natl Acad
Sci USA 80:7410-7414, 1983
21. Costa 0, Monier JC: Detection of antibodies to histones
in human systemic lupus erythematosus and in murine
lupus-like syndromes using micro-ELISA. Ann Immunol (Paris) 134:365-376, 1983
22. Alarcon-Segovia D: Mixed connective tissue disease: a
decade of growing pains. J Rheumatol 8535-540, 1981
23. Bernstein RM, Morgan SH, Chapman J, Bunn CC,
Mathews MB, Turner-Warwick M, Hughes GRV: AntiJo-l antibody: a marker for myositis with interstitial lung
disease. Br Med J 289:151-152, 1984
24. Hobbs RN, Lea DJ, Phua KK, Johnson PM: Binding of
isolated rheumatoid factors to histone proteins and basic
polycations. Ann Rheum Dis 42:435-438, 1983
25. Lowenstein MB, Rothfield NF: Family study of systemic lupus erythematosus: analysis of the clinical history,
skin immunofluorescence, and serologic parameters.
Arthritis Rheum 20: 1293-1303, 1977
26. Bartfield H: Distribution of rheumatoid factor activity in
non-rheumatoid states. Ann NY Acad Sci 168:30-38,
1969
27. Baum H , Palmer C: Significance of anti-mitochondria1
antibodies. Lancet 2: 1411-1412, 1981
28. Contreras CE, Orozco A, Sanchez P, Ortega G, Bianco
NE: Physiological aspects of circulating immune complexes in the normal population. Clin Exp Immunol
48~693-699, 1982
29. Blankstein LA, Stollar BD, Levy SB: Immunochemical
distinction among histones and their variants in a solidphase radioimmunoassay. Anal Biochem 104: 168-172,
1980
30. Romac J , Bouley JP, Van Regenmortel MHV: Enzymelinked immunosorbent assay in the study of histone
antigens and nucleosome structure. Anal Biochem 113:
366-371, 1981
31. Aitkaci A, Monier JC, Mamelle N: Enzyme-linked immunosorbent assay for antihistone antibodies and their
presence in systemic lupus erythematosus sera. J Immunol Methods 44:311-322, 1981
32. Lewis PN: Fluorescently labelled histones as probes of
nucleosome structure: preparation and general properties of methionine-labelled histone H4. Eur J Biochem
99:315-322, 1979
33. Burch JBE, Martinson HG: Iodination of nucleosomes
at low ionic strength: conformational changes in H4 and
HISTONE ANTIBODIES IN RHEUMATIC DISEASE
stabilisation by H 1 . Nucleic Acids Res 9:4367-4385,
1981
34. Petterson I, Hinterberger M, Mimori T, Gottlieb E,
Steitz JA: The structure of mammalian small nuclear
ribonucleoproteins: identification of multiple protein
components reactive with anti-(Ul) RNP and anti-Sm
autoantibodies. J Biol Chem 259:5907-5914, 1984
35. Kubba R, Steck WD, Clough JD: Antinuclear antibodies
and PUVA photochemotherapy. Arch Dermatol
117:474-477, 1981
36. Portanova JP, Rubin RL, Joslin FG, Bernstein RM, Tan
EM, Agnello VD: Antihistone antibodies induced by
pronestyl and hydralazine (abstract). Arthritis Rheum
(suppl) 24:S110, 1981
293
37. Clayton AL, Bernstein RM, Tavassoli M , Shall S, Bunn
CC, Hughes GRV, Chantler SM: Measurement of antibody to poly(adenosine diphosphate-ribose): its diagnostic value in systemic lupus erythematosus. Clin Exp
Immunol 56:263-271, 1984
38. Stollar BD, Ward M: Rabbit antibodies to histone fractions as specific reagents for preparative and comparative studies. J Biol Chem 245:1261-1266, 1970
39. Mathews MB, Bernstein RM: Myositis autoantibody
inhibits histidyl-tRNA synthetase: a model for autoimmunity. Nature 304:177-179, 1983
40. Emerit I , Michelson AM: Mechanism of photosensitivity in systemic lupus erythematosus patients. Proc Natl
Acad Sci USA 78:2537-2540, 1981
Документ
Категория
Без категории
Просмотров
2
Размер файла
775 Кб
Теги
antihistone, syndrome, systemic, erythematosus, primary, disease, tissue, antibody, mixed, sicca, lupus, connection, patterns, rheumatic, arthritis, vasculitis, specificity, rheumatoid
1/--страниц
Пожаловаться на содержимое документа