вход по аккаунту


Serologic subsets in systemic lupus erythematosus.

код для вставкиСкачать
An Examination of Autoantibodies in Relationship to
Clinical Features of Disease and HLA Antigens
A search for anti-Ro or anti-nRNP antibodies by
precipitin analysis among a population of 64 patients
with systemic lupus erythematosus (SLE) or mixed connective tissue disease was undertaken. The 25%of SLE
patients with anti-Ro did not have any distinct clinical
features, compared to patients without anti-Ro or to
anti-nRNP patients with SLE or mixed connective tissue disease. However, these patients did have a significantly greater frequency of rheumatoid factor (80%).
Most importantly however, patients with anti-Ro had a
significantly increased frequency of HLA-B8 (81%) and
HLA-DRw3 (100%) compared to patients with antinRNP (B8-29%; DRw3-33%) or SLE patients without such antibodies (B8-41%; DRw3-25%). These
data suggest the existence of a specific immune response gene for Ro close to the D region of the major
histocompatibility complex.
Recent studies have called attention to the possible existence of various subtypes of systemic lupus
erythematosus (SLE). Most of these investigations have
From the Department of Medicine (Rheumatic Diseases
Unit), University of Western Ontario, and the Department of Medicine (Immunology Unit), Veterans Administration Hospital, State
University of New York, Buffalo, New York.
Presented in part at the Northeastern Regional Meeting of
the American Rheumatism Association, Toronto, November, 1979.
Supported by a grant from the Canadian Arthritis and Rheumatism Society.
David A. Bell, MB: University Hospital, London, Ontario;
Peter J. Maddison, MD: Royal National Hospital for Rheumatic Diseases, Upper Borough Walls, Bath BAI IRL, Avon, England.
Address reprint requests to Dr. D. A. Bell, Department of
Medicine, Rheumatic Diseases Unit, University Hospital, Room
40F4, London, Ontario, Canada N6A 5A5.
Submitted for publication December 28, 1979; accepted in
revised form May 16, 1980.
Arthritis and Rheumatism, Vol. 23, No. 11 (November 1980)
attempted to segregate these subtypes on the basis of
various disease manifestations, such as the predominance of skin involvement (discoid SLE) or the
presence or absence of variable degrees of renal disease
(diffuse glomerulonephritis). Furthermore, considerable
attention has been given to the identification of serologic markers which identify clinical subgroups with
different patterns of disease expression and prognosis.
Thus, one particular syndrome sharing many of
the clinical and serologic features of SLE, often referred
to as mixed connective tissue disease, has been identified by the presence of high titers of antibodies to a nuclear ribonucleoprotein (nRNP) (1). On the other hand,
autoantibodies to a cytoplasmic glycoprotein, termed
Ro (2), which are often accompanied by antibodies to
another soluble tissue protein, termed La (3), have been
reported to occur in lupus patients with a predilection to
severe photosensitive dermatitis (4), some of whom are
persistently ANA-negative (5). Some of these lupus patients have Sjogren’s syndrome. Anti-Ro and anti-La
are also a frequent finding in sicca syndrome and have
been shown to be immunologically identical to SS-A
(6,7) and SS-B (6) [Ha (3,7)] respectively, and share
physicochemical properties similar to SjT (8,9).
In addition to attempting to divide this disorder
into distinct clinical or serologic subgroups, recent studies have examined the presence or absence of genetically homogeneous subgroups based on the presence of
certain HLA antigens. In one such study, it was observed that among different families with several affected members who shared common HLA haplotypes,
the clinical expression of disease was often quite variable (1). In previous reports of unrelated patients with
SLE with heterogeneous clinical manifestations, a consistently increased frequency of any particular HLA-A
Table 1. Clinical features in SLE patients with or without precipitating antibodies to nRNP and Ro
Clinical features
> 4 ARA criteria
Anti-Ro, nRNP (-)
Anti-Ro and nRNP
Skin rash
Renal disease
Total patients
* P < 0.05 111 versus I.
t P < 0.05 I1 versus I or 111
or B locus antigen was not observed more often than in
the general population (11). Recent studies in which
HLA-D locus antigens were identified among SLE patients revealed a significant increase in the D related B
cell alloantigens DRw2 and DRw3 (12,13).
We considered the possibility that among unrelated SLE patients sharing the same HLA antigens or
among family members with SLE sharing the same
HLA haplotype, antibodies with identical serologic
specificity might be identified. Moreover, based on the
observation that patients with SLE whose sera contain
either anti-Ro or anti-nRNP usually form distinct
groups (14), the prediction was made that the basis for
this different antibody response might be the existence
of different immune response genes for these antigens
and hence, different HLA-D locus antigens.
Fifty patients with SLE were selected on the basis of
readily available clinical information and willingness to undergo HLA typing. Forty-nine of these patients fulfilled at
least 4 of the preliminary American Rheumatism Association
(ARA) criteria for SLE (15). Seventeen additional patients selected from the same center whose serum was known to contain anti-nRNP antibodies were also analyzed. Nine had more
than 4 ARA criteria for SLE and the remaining 7 had at least
2 criteria with other serologic abnormalities consistent with
SLE. In addition, 7 families were studied in whom the propositus had either anti-Ro or anti-nRNP and the other members had some clinical expression of a connective tissue disease. All patients studied were white.
The laboratory tests performed on these patients included quantitative immunoglobulin measurements by radioimmunodiffusion, serum complement measurements (CH50),
rheumatoid factor by the latex-fixation technique, and screening for antinuclear antibodies by an indirect immunofluorescent technique employing human leukocyte buffy coat
preparations as tissue substrate (16). Most sera were analyzed
for antibodies to native DNA by a radioimmunoassay technique (17). In addition, antibodies to Ro, La, and nRNP were
identified by precipitin analysis as previously described (2,18).
HLA typing using a standard microcytotoxicity assay was performed with commercially prepared reagents for the identification of HLA-A, B, and C antigens. In addition, D related B
ce!l alloantigens were identified using Terasaki reagents by a
cytotoxicity technique employing peripheral blood lymphocytes enriched for B cells by rosette (ERFC) separation (19). B
cell typing was performed using UCLA research trays TE-B
5-7 which detect DRw antigens 1 through 7. A standard microcytotoxicity test was used with cells incubated with antisera for 1 hour at 37°C and with complement for 2 hours at
room temperature.
Studies of unrelated SLE patients. The 64 patients with probable (less than 4 ARA criteria) or definite SLE (4 or more ARA criteria) were subdivided into
4 groups on the basis of the presence or absence of precipitating antibodies to Ro and nRNP. (See Table 1.)
Sixteen of 64 (25%) had anti-Ro antibodies only, 14 of
64 (21%) had anti-nRNP antibodies, 3 1 of 64 (48%) had
neither antigen antibody system, and 3 of 64 patients
had both anti-Ro and anti-nRNP antibodies. The frequency of certain clinical features in each of these 4 separate categories is compared in Table 1. Many of the
clinical features were present with similar frequency in
all 4 serologic categories, with the exception of serositis
and renal disease which occurred with significantly
lower frequency in patients with anti-nRNP, Raynaud’s
phenomenon which occurred less often among patients
whose serum lacked anti-Ro or anti-nRNP, and skin
rash which occurred slightly more frequently among
anti-Ro positive patients. The diagnosis of Sjogren’s
syndrome was based on the presence of obvious features
of this disorder (sicca symptoms and positive results of
Shirmer’s test). Since special diagnostic studies to uncover evidence of sicca were not undertaken, the true
frequency of this syndrome among these patients may
be significantly underestimated. Among the 7 patients
with renal disease in the anti-Ro positive group, 4 had
Table 2. Laboratory and serologic features in SLE patients with or without precipitating antibodies to
nRNP or Ro
- .
. -I
Serologic features
Anti-Ro. nRNP (-)
Anti-Ro, n R N P
Elevated Ig
Anti-DNA antibody
Rheumatoid factor
Total patients
- ...
. ... .
.- . ..
-.- . .- .. -
I versus 11.
diffuse involvement (2 membranous and 2 proliferative
Some serologic and laboratory findings were
compared among these subgroups (Table 2). Hypergammaglobulinemia was seen frequently and was often
striking in patients with anti-Ro. Anti-DNA antibodies
detectable by a radioimmunoassay method were present
in abnormal quantities among the majority of patients
in each category but were slightly more frequent among
patients who lacked either of the precipitin systems.
Rheumatoid factor, on the other hand, was strikingly
more frequent among anti-Ro positive patients (8Wo)
than among those with anti-nRNP (38%) or among
those who lacked either precipitin system (18%). Among
the patients with anti-Ro antibodies, 2 also had anti-La.
The 3 patients with both anti-Ro and anti-nRNP had
clinical and serologic features similar to those patients
with anti-Ro or anti-nRNP only.
HLA analysis. Thirteen of 16 anti-Ro positive
patients (81%) had the HLA-B locus antigen B8. This
HLA antigen was also present in 4 of 14 (29%) patients
with anti-nRNP antibody only and 13 of 31 (41%) SLE
patients lacking either serum precipitating system.
HLA-B35 was present in 2 of 16 (13%) anti-Ro positive
patients, 4 of 3 1 (12%) patients lacking either precipitin
system, and 4 of 14 (29%) patients with anti-nRNP antibody only. Two of 3 patients with both precipitin systems had HLA-B8 whereas the third patient had HLAB35.
HLA-D typing was performed in 44 of 64 patients. The increased frequency of HLA-B8 among
anti-Ro positive patients is attributed to the increased
frequency of HLA-DRw3 present in 10 of 10 anti-Ro
positive patients in whom this test was performed
(100%) and 1 of the 2 patients tested whose serum contained both anti-Ro and anti-nRNP. The frequency of
HLA---DRw3in the remaining patients was not significantly different than in controls, nor was there any significant increased frequency of other HLA-D related B
cell alloantigens.
Family studies. Several families with multiple occurrence of a connective tissue disorder consistent with
SLE or similar to it, were examined to determine
whether affected individuals sharing the same HLA
haplotype also shared autoantibodies of identical or different specificity. Two of these families were the subject
of a previous publication (10). In 3 of these families, the
propositus had anti-nRNP antibody, and at least 1 other
family member had clinical or serologic expression of
disease. In the remaining 3 families, the propositus had
anti-Ro antibodies.
Anti-nRNP (Figure I). In Family I the proband
[7], a 3 1-year-old woman with polyarthritis, severe Raynaud’s phenomenon with digital gangrene, dysphagia,
positive ANA, and hypergammaglobulinemia had antinRNP antibody. Her mother [ 11, with whom she shared
the haplotype 1,38, also had polyarthritis, less severe
Raynaud’s phenomenon, mild dysphagia, hypergammaglobulinemia, and was ANA positive but lacked
anti-nRNP or anti-Ro antibodies by precipitin analysis
or by hemagglutination. Anti-Ro and anti-nRNP were
not identified among asymptomatic family siblings of
the proband, some of whom were ANA positive.
In Family I1 the proband [3], a 33-year-old
woman with polyarthritis, Raynaud’s phenomenon, diffuse facial skin rash, and digital swelling, had antinRNP antibody. Her maternal haplotype was 3,12
DRw4 and the inferred paternal haplotype 24,8 DRw3.
The mother [ I ] had Raynaud’s phenomenon and a positive ANA, but the deceased father [2] had no history of
a connective tissue disorder. Her younger brother [ 5 ]
also had the maternal haplotype 3,12 DRw4, but did
not have anti-nRNP antibodies. Thus, in this family
HLA--B8 DRw3 occurred only in the patient with antinRNP antibody. The pattern of inheritance suggested
that if HLA related genetic factors were involved in the
response to this antigen in this case, closer linkage exists
with the maternal haplotype 3,12 DRw4.
In Family 111 the propositus and his father [ I ]
shared the same serologic features of SLE, but only the
Figure 1. 1 I :> = Normal male or female.
0 = Male or female
with fully clinically expressed MCTD (Family I, 11) or SLE (Family
111). fl (3 = Male or female with partially expressed MCTD (Family
1.11) or SLE (Family 111).
son (31 had clinical symptoms and signs of this disorder.
Although he and his father both had anti-DNA antibodies and anti-red cell antibody, only the son had detectable anti-nRNP antibody. His 2 siblings and mother
were clinically and serologically normal.
Anti-Ro (Figure 2). In Family I the propositus (31
had Raynaud's phenomenon, polyarthritis, serositis,
and anti-Ro antibodies. He and his youngest sibling [ 5 )
with rheumatoid arthritis shared the maternal HLA
haplotype 2,8 DRw3. The mother 121, who had transient
polyarthritis 20 years previously and is now clinically
normal, had rheumatoid factor and anti-Ro antibodies.
Anti-Ro was not detected in either her daughter with
seropositive rheumatoid arthritis or her middle son [4]
who shared the same HLA haplotype 2,8 DRw3.
In Family I1 the propositus [6], a 23-year-old
man with polyarthritis, skin rash, hypergammaglobulinemia, and positive ANA, had anti-Ro and antiLa antibodies as well as an antibody to an unidentified
antigen detectable with calf thymus extract. His mother
[3], with polyarthritis, Raynaud's phenomenon, and hypergammaglobulinemia, had antibodies to the same unidentified antigen (detectable by a precipitin line of
complete identity with her son's serum against calf
thymus extract). The other members of the family, including the proband's 2 siblings [5,7] as well as a mater-
nal aunt [ I ] and uncle (41 with hypergammaglobulinemia. did not have detectable anti-Ro or other precipitating systems.
In Family 111 numerous siblings have been
afflicted with clinical and serologic features of SLE.
Two members (6,8] had anti-Ro antibodies and shared
the same haplotype 2,8 DRw3. The clinically normal
sibling [ 101 also had anti-Ro and shared the same haplotype while several other affected and unaffected siblings with this haplotype were anti-Ro negative [2,5,12].
In contrast, a younger sibling (131 with SLE did not
share this haplotype but had anti-nRNP only.
Occurrence of anti-Ro and anti-nRNP in rheumatoid arthritis. The sera from 19 patients with classic
or definite rheumatoid arthritis who had the HLA antigens B8 DRw3 were examined for the presence of antibody to Ro and nRNP by precipitin analysis. Eighteen
of these 19 patients were ANA positive and 13 of 19
rheumatoid factor positive. Neither anti-Ro nor antinRNP was detected among any of these individuals.
These studies reveal that among patients with
SLE, a very strong association exists between HLA-B8
Figure 2. r l
Q ?.'<:*,,.ELIS
(2 = Normal male or female. W 0 = Male or female
with fully clinically expressed disease. fl
partially expressed disease.
(3 =
Male or female with
Table 3. HLA antigens in SLE patients with or without precipitating antibodies to Ro or nRNP
Anti-Ro (+)
Total patients
No. 9%
Anti-Ro, nRNP (-)
No. %
Anti-Ro, nRNP (+)
No. %
No. %
Controls (260) IiLA-B8- -21% HLA-DRw3-26%.
t P < 0.025 I versus 11; P < 0.01 I versus 111.
+ P < 0.001 I versus 11; P < 0.05 1 versus I l l .
and DRw3 and the presence of anti-Ro antibodies. Although HLA-B8 has been previously reported as
slightly increased among unselected, unrelated SLE patients, (20,2 l), recent reports have suggested a much
stronger association with the D locus related alloantigen
DRw3 (12,13). Studies of 2 different rheumatic disease
populations (4,22) have demonstrated that anti-Ro is
present in approximately 25% of unselected SLE patients, a proportion similar to that found in this study;
therefore this may account for the increased frequency
of HLA-DRw3 in this disorder. Antibodies to cellular
protein antigens identical to Ro have been detected
among most patients with Sjogren’s syndrome (23)
alone or in association with SLE (22), and isolated Sjogren’s syndrome has been shown to be strongly associated with HLA-DRw3 (24). One report has suggested
that through the use of a number of special diagnostic
procedures, virtually all patients with SLE (25) and with
mixed connective tissue disease (26) have Sjogren’s syndrome. Although we did not specifically undertake to
demonstrate occult evidence of Sjogren’s syndrome
among these patients, only 2 among the anti-Ro Fatients had overt clinical manifestations of this disorder.
We do not believe, therefore, that we have unwittingly
selected patients with Sjogren’s syndrome among this
SLE population through the measurement of anti-Ro
Although Raynaud’s phenomenon occurred significantly less frequently among patients lacking antiRo or anti-nRNP and severe renal disease was less
frequent among patients with anti-nRNP, no other remarkable clinical differences served to distinguish patients with these antibodies. The only remarkable serologic abnormality in the anti-Ro positive subgroup was
the very high frequency of rheumatoid factor, although
the occurrence of hypergammaglobulinemia in some
patients was often striking.
Since anti-Ro antibodies were not found among
all SLE patients with DRw3 or among rheumatoid arthritis patients with this HLA antigen as well as rheumatoid factor, HLA-DRw3 can be considered a necessary but not sufficient prerequisite factor for the immune
response to this antigenic molecule. The family studies
also support this contention, since anti-Ro antibodies
were frequently undetected among closely related relatives sharing the same HLA-D antigens or sharing the
same haplotype as the propositus with anti-Ro antibody.
The relationship between this antibody and the
disease is not clearcut. For example, in Family 111 (Figure 2), 1 clinically normal member who shared the haplotype 1,8 DRw3 with 2 clinically affected anti-Ro positive patients also had anti-Ro. In contrast to this in
Families I and 11 (Figure 2), clinically affected individuals sharing the same HLA haplotype 1,8 DRw3 did
not have detectable anti-Ro antibody.
The most straightforward interpretation of the
increased association between anti-Ro and HLA-DRw3
is that a genetically determined locus controlling the immune response to this antigen is closely linked to HLADRw3. However, since many SLE patients with anti-Ro
antibody had striking hypergammaglobulinemia and a
high frequency of rheumatoid factor, and since similar
serologic phenomena have been noted in some patients
with Sjogren’s syndrome (27), HLA-DRw3 may not
necessarily reflect a specific immune response gene but
rather a more general phenomenon involving polyclonal B cell activation. HLA-B8 and Dw3 have also
been seen among patients with other immunologic disorders such as juvenile rheumatoid arthritis (28),
chronic active hepatitis (29), and celiac disease (30). In
all of these disorders the common denominator may be
the existence of a disturbance in the regulation of the
immune response closely associated with HLA-DRw3.
The expression of this defect obviously varies among
patients with different diseases. The defects in immune
regulation that have been proposed as a mechanism to
explain SLE in general may be most readily expressed
among patients with anti-Ro antibodies.
Finally, these studies have confirmed the impression that patients with anti-Ro and anti-nRNP usually
form distinct subgroups with overlapping clinical manifestations. The failure to identify a strong association
between anti-nRNP and any particular HLA-B or D
locus antigen may simply reflect the relative heterogeneity of the molecules that constitute this antigen.
Thus, if genetic factors related to HLA are important in
the immune response to this materiaI, multipIe HLA
specificities would be expected to be present.
The authors wish to thank Ms. Olga Few for her help
in the preparation of the manuscript.
1. Sharp GC, Irwin WS, Tan EM: Mixed connective tissue
disease: an apparently distinct rheumatic disease syndrome associated with antibody to an extractable nuclear
antigen (ENA). Am J Med 52:148-159, 1972
2. Clark G, Reichlin M, Tomasi TD: Characterization of a
soluble cytoplasmic antigen reactive with sera from patients with systemic lupus erythematosus. J Immunol
102:117-122, 1968
3. Mattioli M, Reichlin M: Heterogeneity of RNA protein
antigens reactive with sera of patients with systemic lupus
erythematosus. Arthritis Rheum 17:421429, 1974
4. Maddison PJ, Mogavero H, Provost TT, et a1 The clinical
significance of autoantibodies to a soluble cytoplasmic antigen in systemic lupus erythematosus and other connective tissue diseases. J Rheumatol 6:189-195, 1979
5. Provost TT, Ahmed R, Maddison PJ, et al: Antibodies to
cytoplasmic antigens in cutaneous lupus: serological
markers for systemic diseases. Arthritis Rheum 20: 14571460, 1977
6. Alspaugh MA, Tan EM: Antibodies to cellular antigens in
Sjogren’s syndrome. J Clin Invest 55:1067-1073, 1975
7. Alspaugh MA, Maddison P: Resolution of the identity of
certain antigen-antibody systems in systemic lupus erythematosus and Sjogren’s syndrome: an interlaboratory
collaboration. Arthritis Rheum 22:796-798, 1979
8. Tan EM, Christian C, Holman HR, et a1 Anti-tissue antibodies in rheumatic disease: standardization and nomenclature. Arthritis Rheum 2 0 1419-1420, 1977
9. Anderson JR, Gray KG, Beck JS, Buchanan WW, et a 1
Precipitating autoantibodies in the connective tissue diseases. Ann Rheum Dis 21:360-369, 1962
10. Cleland LG, Bell DA, Willans M: Familial lupus: family
studies of HLA and serologic findings. Arthritis Rheum
2 1: 183-19 1, 1978
11. Russell AS, Barraclough D R E HLA studies in other
rheumatic diseases: systemic lupus erythematosus. Clin
Rheum Dis 3:365-375, 1977
12. Reinertsen JL, Klippel JH, Johnson AH, et al: B-lymphocyte alloantigens associated with systemic lupus erythematosus. N Engl J Med 299515-518, 1978
13. Gibofsky A, Winchester RJ, Patarroyo M, et al: Disease
associations of the Ia-like human alloantigens: contrasting
patterns in rheumatoid arthritis and systemic lupus erythematosus. J Exp Med 148:1728-1732, 1978
14. Reichlin M, Mattioli M: Antigens and antibodies characteristic of SLE. Bull Rheum Dis 24:756-758, 1974
15. Cohen AS, Reynolds WE, Franklin EC, et a 1 Preliminary
criteria for the classification of systemic lupus erythematosus. Bull Rheum Dis 21:643-648, 1971
16. Condemi JJ, Barnett EV, Atwater EC, et al: The significance of antinuclear factors in rheumatoid arthritis. Arthritis Rheum 8:1080-1093, 1965
17. Pincus T: Immunochemical conditions affecting the measurement of DNA antibodies using ammonium sulfate
precipitation. Arthritis Rheum 14:623-630, 1971
18. Mattioli M, Reichlin M: Characterization of a soluble nuclear ribonucleoprotein antigen reactive with SLE sera. J
Immunol 197:1281-1290, 1971
19. Terasaki PI, Bernoco D, Park MS, et al: Microdroplet
testing for HLA A-B-C-D antigens. Am J Clin Pathol
69:103-120, 1978
20. Ivanyi D, Dostal C, Mrklas L: HLA-B8 in systemic lupus
erythematosus. Tissue Antigens 8:91-93, 1976
21. Grumet FC, Coukell A, Bodmer J, et al: Histocompatibility (HL-A) associated with SLE. N Engl J Med 285193196, 1971
22. Scopelitis E, Biundo JJ, Alspaugh MA: Anti-SSA antibody and other antinuclear antibodies in systemic lupus
erythematosus (abstract). Arthritis Rheum 21590, 1978
23. Alspaugh MA, Tala1 N, Tan EM: Differentiation and
characterization of autoantibodies and their antigens in
Sjogren’s syndrome, Arthritis Rheum 19:216-222, 1976
24. Chused TM, Kassan SS, Opelz G, et a1 Sjogren’s syndrome associated with HLA DW3. N Engl J Med
296~895-897, 1977
25. Alarcon-Segovia D, Ibanez G, Velazquez-Forero F:
Sjogren’s syndrome in systemic lupus erythematosus: clinical and subclinical manifestations. Ann Intern Med
81577-583, 1974
26, Alarcon-Segovia D: Symptomatic Sjogren’s syndrome in
mixed connective tissue disease. J Rheumatol 3: 191-195,
27. Martinex-Lavin M, Vaughan JH, Tan EM: Autoantibodies and the spectrum of Sjogren’s syndrome. Ann Intern Med 91:185-190, 1979
28. Gershwin ME, Opelz G, Terasaki PI, et a 1 Elevated frequency of HLA-DW3 in juvenile rheumatoid arthritis.
Tissue Antigens 10:330-336, 1977
29. Opelz G, Vogten AJM, Summerskil WHJ, et al: HLA determinants in chronic active liver disease: relation of
HLA-DW3 to prognosis. Tissue Antigens 96:3640, 1977
30. Solheim DG, Ek J, Thune PO, et al: HLA antigens in dermatitis herpetiformis and coeliac disease. Tissue Antigens
7157-60, 1976
Без категории
Размер файла
543 Кб
subsets, lupus, serological, systemic, erythematosus
Пожаловаться на содержимое документа