вход по аккаунту


Heterogeneity of RNA protein antigens reactive with sera of patients with systemic lupus erythematosus description of a cytoplasmic nonribosomal antigen.

код для вставкиСкачать
Heterogeneity of RNA Protein Antigens Reactive with Sera
of Patients with Systemic Lupus Erythematosus
Description of a Cytoplasmic Nonribosomal Antigen
Martha Mattioli and Morris Reichlin
A soluble cytoplasmic RNA protein antigen (La) is described, which precipitates with sera of patients with systemic lupus erythematosus and a
small number of patients who lack antinuclear factors but who have
various connective tissue symptoms. Antibodies to the cytoplasmic RNA
protein are almost invariably (8 of 9 cases) accompanied by antibodies
to a non-nucleic acid cytoplasmic antigen, Ro. Description of this antigen enlarges the spectrum of nucleic acid antigens reactive with the sera
of patients with systemic lupus erythematosus.
Patients with systemic lupus erythematows (SLE) produce antibodies reactive
with an ever widening spectrum of nucleic
acid antigens. Early work in this area
pointed out the occurrence of antibodies
to both pure DNA and to DNA histone
conjugates or deoxyribonucleoprotein.
Much study has clarified the clinical and
biologic significance of such antigen-antibody reactions and has emphasized their
diagnostic specificity and their role in the
pathogenesis of certain clinical manifestations. Evidence for the major participation
From the Department of Medicine and Biochemistry, Veterans Administration Hospital, State University of New York at Buffalo School of Medicine,
3495 Bailey Avenue, Buffalo, New York.
Supported by Designated research funds from the
Veterans Administration Hospital and a research
grant from the US Public Health Service, AM10428.
A preliminary account of these findings was presented at the American Rheumatism Association
meeting held in Pittsburgh, Pennsylvania, December
8. 1972.
MARTHA MATTIOLI, MD: Research Instructor, Department of Medicine; MORRIS REICHLIN, MD: Professor of Medicine and Biochemistry.
Address reprint requests to Dr. Reichlin.
Submitted for publication September 4, 1973; accepted November 29, 1973.
of DNA-anti-DNA complexes for example
in the development of nephritis in SLE is
particularly strong (1,Z).
I n recent years, antibodies to various
RNA-containing antigens have also been
demonstrated in the sera of SLE patients,
and such antigens comprise a list which
includes ribosomes, double stranded RNA
species such as polyinosinicytidylic acid
(poly IC), single stranded RNA, nucleolar
RNA, and nuclear RNA protein. There is
compelling evidence that these are all s e p
arate and distinct antigen-antibody reactions. The significance of some of these
reactions is also emerging. Several reports
indicate that patients with antibodies to
nuclear RNA protein experience a much
more favorable course and a lower incidence of nephritis than the usual group of
patients with SLE (3,4).
This report describes a previously unrecognized cytoplasmic RNA protein antigen
to which antibodies are found not only in
the sera of classic SLE patients but also in
a small group of patients who lack detectable antinuclear factors but who have a
clinical picture consistent with SLE.
Arthritis and Rheumatism, Vol. 17, No. 4 (July-August 1974)
Double diffusion in agar gel, preparation of calf
thymus extract (CTE), sucrose density gradient
ultracentrifugation, immunuelectrophoresis. digestion with nucleolytic and proteolytic enzymes, and
quantitative complement fixation reactions were
performed as described in previous publications
Calf liver RNA, yeast RNA, and polyinosinicytidylic acid were purchased from Sigma Chemical
Co., St Louis, Missouri.
Preparation of ribosomes and extracts of ribosomes were carried out according to Sturgill and
Carpenter (7) and also according to Shur el a1 (8).
Nuclear and cytoplasmic fractions of calf thymus
tissue were prepared by a modification of the method of Wang and Pate1 as described previously (9).
All patients designated SLE possessed at least
four of the preliminary criteria of the American
Rheumatism Association as published in the Bulletin of Rheumatic Diseases, 21:643-648, 1971.
Immunologic and Physicochemical
Characterization of the Antigen
The first aim in this study was to establish the distinction of this RNA protein
antigen from RNA containing antigens
described previously. Several approaches
were used to establish the uniqueness of
this antigen-antibody reaction and their description follows. The availability of one
serum (La) that formed a single precipitin
line with CTE was central to these efforts.
Among the relevant serologic properties of
this serum was that it stained mouse liver
nuclei very weakly in indirect immunofluorescent tests. This serum stained the cytoplasm of thymocytes strongly, with a notable absence of nuclear or nucleolar staining. I n addition to the fluorescence studies,
which suggested cytoplasmic localization of
the antigen, a cell fractionation in sucrose
solutions was carried out with isolation of
the nuclear and cytoplasmic fractions of
calf thymus tissue. La antigen was demonstrable only in the cytoplasmic fraction and
no La activity was found in the nuclear
fraction. This serum formed a single precipitin line in Ouchterlony experiments
when diffused against CTE, which was
completely sensitive to both RNase and
trypsin. Not only did the precipitin reaction disappear after RNase or trypsin treatment but complement fixing activity of
CTE with the La serum was also completely abolished by such treatment. Incubation of the extracts with RNase required
3 hours for complete destruction of the La
Antigenic uniqueness of the RNA protein system La was established relative to
the ribosomal antigen and nuclear RNA
protein antigen (Mo) by diffusing sera
monospecific for each antigen against whole
CTE which contained all three antigens.
In Figure 1 is seen CTE in the center well
and serum La in well 1. Well 2 contains
antibody specific for ribosomes* and well 3
contains antibody specific for nuclear RNA
protein. Note that 1 is nonidentical to both
2 and 3, which are completely independent
of each other. Thus, the cytoplasmic RNA
protein, nuclear RNA protein, and ribosomal antigens are all immunologically
distinct from one another, sharing no immunologic specificity by the criterion of
precipitation in agar gel.
The following defined RNA preparations
neither precipitated nor fixed complement
with serum La nor did they block the reaction of serum La with its antigen in
CTE: soluble calf liver RNA, total soluble
yeast RNA, and poly IC.
To provide physical as well as immunologic evidence for the distinction of the
cytoplasmic RNA protein from other known
RNA-containing antigens in CTE, a sedimentation study of these antigens was per'Kindly provided by Dr. Peter H. Schur, of
Boston, Massachusetts.
Arthritis and Rheumatism, Vol. 17, No. 4 (July-August 1974)
with Normal
Control Mechanism
Acquisition of New Antigen
Derepression t o Fetal Antigen
Formation of AntiLymphocyte Antibody
Decrease i n
T-cell Function
Altered Lymphs
. .
Survival of
Abnormal Cells
Increase Incidence
of Cancer
Release of
Fig 1. Independence of the three RNA protein immune systems by Ouchterlony analysis.
In the center well is calf thymus extract (CTE) and in wells 1 (anti-La), 2 (antiribosomes),
and 3 [antinuclear RNA protein (Mo)] are the human sera with designated specificities.
formed utilizing the sucrose density ultracentrifugation technique. I n Figure 2 is
depicted the sedimentation characteristics
of nuclear RNA protein antigen, a soluble
form of the ribosomal antigen, and the cytoplasmic RNA protein antigen. On the
ordinate is plotted the strength of precipitation on a 1+-4+ scale of two-drop fractions of a 10% to 40% sucrose density gradient. Markers used to calibrate the gradient were a purified monoclonal IgM, purified normal IgG, human hemoglobin and
human cytochrome c. The distribution of
each protein was ascertained by precipitation with monospecific serum for each
marker. It is seen that the cytoplasmic
RNA protein is considerably smaller than
the other two RNA containing antigens.
Size estimations of the cytoplasmic RNA
protein antigen from such studies leads to
an approximate value for a molecular
weight of 80,000 and a sedimentation coefficient of 3.0s.
These data taken together provide strong
evidence for the antigenic and physical independence of the cytoplasmic RNA pro-
Arthritis and Rheumatism, Vol. 17, No. 4 (July-August 1974)
Fig 2. Sucrose density ultracentrifugation of calf thymus extract (CTE) and antigenic reactivity with antinuclear RNA protein ,).(
with antiribosomes (m), and anti-La (A). Two
drop fractions were analyzed and the distribution of the marker proteins are indicated.
tein antigen from the other recognized
RNA containing antigens.
Stability of the La Antigen
and Demonstration in Human Tissue
As previously reported for the nuclear
RNA protein antigen (Mo) (5). La antigen
activity was not easily demonstrable in extracts of human organs obtained from autopsy. Moreover, it was noted that the
strong activity noted in fresh extracts of
calf thymus tissue decayed measurably at
4OC and quickly lost activity at moderate
temperatures. With extracts made from
surgically removed human thymus tissue or
extracts of chronic lymphocytic leukemia
cells, cytoplasmic RNA protein antigen was
easily demonstrable. This activity could be
stabilized indefinitely by lyophilization of
active extracts which then could be reconstituted with full retention of antigenicity.
These findings mimicked what had previously been found with the labile nuclear
RNA protein antigen (5). As in that case,
failure to recognize the lability of this antigen has been no doubt in part responsible
for the previous failure to recognize this
system while working with SLE sera and
extracts of human tissues taken at autopsy.
Mutual Occurrenceof Anti-Ro
in ANA-Negative Patients
In a previous report, the occurrence of
antibodies to Ro, a cytoplasmic nonnucleic
acid antigen had been described as being
restricted to patients with SLE and Sjog-
Arthritis and Rheumatism, Vol. 17, No. 4 (July-August 1974)
Fig 3. Independence of the La and Ro immune systems by Ouchterlony analysis. In the upper well is human spleen extract (HSE) and opposite to it is a human serum containing
anti-Ro. In the right large well is calf thymus extract (CTE) and opposite to it is a human
serum containing anti-La.
ren’s syndrome (9). It was also noted that
the prototype serum lacked detectable antinuclear factors and was from a patient (Ro)
with the clinical picture of SLE and that
two other such patients had been found
with clinical pictures consistent with SLE
but lacking detectable antinuclear factors.
Ro is a relatively stable antigen easily
recoverable from extracts of human tissues
taken at autopsy and is totally RNase resistant. Human sera containing only antiRo react well in Ouchterlony analysis with
extracts of human spleen or liver but react
weakly with the same antigen in CTE
where it exists in low concentration. Human sera containing both anti-Ro and antiLa give one RNase resistant line in human
spleen or liver extracts and give two independent lines with CTE; one is the RNase
resistant system identical to that in extracts
of human spleen or liver, the other is RNase sensitive and is the La-anti-La reaction.
Finally, human sera containing only antiLa give a weak RNase sensitive line with
human spleen but a single strong RNase
sensitive precipitin line with CTE. Confident identification of the Ro precipitin line
was achieved by comparison of an unknown
serum with a serum containing only antiRo in which measurable RNase sensitivity
of the antigen was excluded by quantitative complement fixation studies.
Complete antigenic independence of the
Ro and La immune systems is illustrated
by the Ouchterlony plate in Figure 3. Serum CH precipitates with HSE (Ro antiRo) and this precipitate is crossed by that
formed between CTE and serum La (Laanti-La). Note that in this preparation of
HSE there is demonstrable La activity.
Much to our surprise, 8 of the 9 sera
containing anti-La also contained anti-Ro.
Thus, there is an extremely high concordance of anti-Ro in human sera containing
antibodies to the cytoplasmic RNA protein.
Equally interesting is the absence of antibodies to some other defined antigens in
these sera containing anti-La. Table 1 tabulates these findings and illustrates that none
of these sera contain anti-Mo (nuclear RNA
Arthritis and Rheumatism, Vol. 17, No. 4 (July-August 1974)
Table 1. Positive and Negative Associations
of Anti-La and other Antibodies
Human Sera with
anti-La and:
antiribosomal antibody
protein), anti-Sm (lo), or antibodies against
ribosomes (8) demonstrable as precipitins.
While failure to find other antibodies in
these sera containing anti-La may reflect
the small number of anti-La sera examined
thus far, the positive association of anti-Ro
in anti-La sera is clearly due to some nonrandom process.
The serologic crossover exhibited by the
ANA negative Ro positive patients previously reported (9) is shared and further
extended by the description of the cytoplasmic RNA protein antigen-antibody system.
The diagnoses and relevant serologic data
for the La+ patients are listed in Table 2.
There are 5 patients with the clinical diagnosis of SLE and 3 whose sera lack detectable ANA but have connective tissue syndromes consistent with but not diagnostic
of SLE (except for Ro), according to ARA
preliminary criteria. A ninth patient had
lesions of discoid LE without sufficient
Table 2. Anti-La Positive Patients
Diagnosis ANF
a Ro
criteria for the diagnosis of SLE. The patients with SLE all had significant ANA
titers on mouse liver nuclei and 3 of the 5
had demonstrable antibodies to single
stranded DNA by the complement fixation
I n Table 3 are listed some clinical data
on the 3 ANA-negative patients whose sera
contain antibodies to the cytoplasmic KNA
protein. I n the first column are those clinical findings which would be considered significant for the diagnosis of SLE, according
to ARA preliminary criteria. I n the second
column are some ancillary findings, which
in the first 2 patients are not otherwise
explained and could be related to their
connective tissue disease.
Certain comments are in order with regard to these findings. All of these patients
had polyarthritis but none had erosive destructive disease with deformity. Patient
PR had sufficient findings for the clinical
diagnosis of SLE while the other two cases
do not. In the case of patient AP, his proteinuria never exceeded 1.5 g/24 hr and his
urinary sediment never contained formed
elements. His creatinine clearance was normal. A renal biopsy showed only mild cellular proliferation in the glomeruli and
some capsular thickening.
Patient RG was a 54-year-old woman
with documented cirrhosis. Her connective
tissue system symptoms and signs were minor and antibodies to the cytoplasmic RNA
protein were discovered in a survey.
Lastly it is worth mentioning a fourth
ANA-negative patient who lacked anti-La
but who possesses anti-Ro. This lady* was
seen in 1963 for an episode of acute synovitis of the left hip accompanied by a positive LE prep and a positive ANA test. No
other clinical findings were present and
*Patient of Dr. Ralph Jacox of Rochester, New
York,who provided the serum and clinical summary.
Arthritis and Rheumatism, Vol. 17, No. 4 (July-August 1974)
Table 3. Data on 3 ANA-Negative Patients whore Sera Contained Antibodies
to Cytoplasmic RNA Proteln
Findings Significant for
Diagnosis of SLE
Ancillary Findings
Butterfly rash
Raynaud's phenomenon
Positive Latex fixation 3.2 g/100 mi;
Proteinuria (0.6-1.5 9/24 hr);
positive Latex fixation reaction,
hypertension: 2.1 g/100 ml
Coombs positive
hemolytic anemia
Positive Latex fixation;
6.1 g/100 ml y-globulin;
Lannec's cirrhosis
subsequently the synovitis and serologic
findings disappeared. Presently she has no
symptoms referrable to LE and she is ANAnegative but is positive for anti-Ro. This
case illustrates the well known serologic
point that the ANA titer may fluctuate and
raises the possibility that its absence in our
other 3 cases may be a temporal phenomena.
It is pertinent to raise a number of
points for discussion about the description
of yet another nucleic acid containing antigen reactive with SLE sera. These are: 1)
to establish that such a system is not one
already reported; 2) to question the clinical
significance of such a reaction in SLE patients, and 3) to question the significance of
the mutual occurrence of antibodies to the
cytoplasmic antigen Ro and antibodies to
the cytoplasmic RNA protein antigen, La.
The evidence that the cytoplasmic RNA
protein antigen is distinct from the other
known nuclear RNA containing antigens
is several fold. The nuclear RNA protein
(Mo) and nucleolar antigens are distinguishable from La firstly by a different cellular localization and secondly in the nonidentical precipitin patterns of the nuclear
and cytoplasmic RNA protein antigens with
their specific antibodies. Thirdly, the cytoplasmic RNA protein antigen is considerably smaller than either the nuclear RNA
protein or the nucleolar antigens.
The differentiation between the cytoplasmic RNA protein and either double or
single stranded RNA is established by the
failure of poly IC, yeast, or calf liver RNA
to react with sera possessing antibodies to
cytoplasmic RNA protein antigen.
Differentiation of the cytoplasmic RNA
protein antigen (La) from the ribosomal
antigen can be made on the basis of size,
antigenic specificity, and the failure of
either thymic or liver ribosomes (or extracts
of these) to react with sera possessing antibodies to the La antigen. These data taken
Arthritis and Rheumatism, Vol. 17, No. 4 (July-August 1974)
together provide strong evidence for the
chemical uniqueness of the cytoplasmic
RNA protein antigen from other known
RNA-containing antigens.
There are no obvious correlations between the presence of antibodies to the cytoplasmic RNA protein antigen and the
clinical status of the patients. In this small
group of patients, those with complement
fixing antibodies to single-stranded DNA
have had nephritis and have done poorly.
Indeed, 2 of the 5 patients with SLE in this
group died, 1 with advanced renal disease.
Of interest is the subgroup of patients with
anti-La who lack detectable antinuclear factors on the mouse liver substrate. These
are patients who have clinical features suggestive of SLE but do not in every case
have sufficient criteria for the diagnosis of
SLE as set forth by the ARA committee in
their statistical study of preliminary criteria for the diagnosis of SLE.
As has been repeatedly reported (11,12),
active untreated SLE is accompanied by the
invariable presence of antinuclear factors
detectable by immunofluorescence on mouse
liver sections. These patients are related to
other patients with SLE not only by their
clinical picture but also by the fact that
the only groups of patients who possess
anticytoplasmic RNA protein and anti-Ro
are patients with classic SLE or Sjogrens
Lastly, it is relevant to ask about the
significance of the nearly invariable presence of anti-Ro in the sera of patients with
anticytoplasmic RNA protein. The situation may be analogous to the concurrence
of antinuclear RNA protein (Mo) in the
serum of patients with SLE possessing antiSm. In that situation 17 of 19 patients with
anti-Sm in their sera also possessed anti-Mo.
It was shown in that study that there was
complex formation between Sm and Mo
and the suggestion was made that complex
formation might be associated with concurrent immunization of the two antigens and
thus account for the higher than expected
mutual Occurrence of antibodies to these
two antigens (6). Experiments are in progress to evaluate the possibility of complex
formation between La and Ro.
It is also of interest that there seems to
be some dissociation of occurrence of these
various antibodies. For example, we have
not observed a single example of the mutual occurrence of antibodies to the two
RNA protein antigens, Mo (nuclear) and
La (cytoplasmic). As the frequency of antiMo in SLE sera in our material is 0.30 and
that of La is 0.07, we would expect a mutual frequency of 0.02 or 3 instances of
both antibodies occurring in 150 SLE sera.
Yet, no instance of both antibodies in a
single serum has been observed thus far.
Another correlative point is that this mutual occurrence of antibodies to La and Ro
on the one hand and Mo and Sm on the
other is that these pairs are also segregated
according to cellular localization. La and
Ro are both cytoplasmic antigens while Mo
and Sm are both nuclear antigens. Thus,
certain SLE patients exhibit a major immune response to a pair of cytoplasmic
antigens while others (a larger number)
emphasize an immune response to a specific
pair of nuclear antigem with little overlap
between the two groups. Whether such a
serologic partition reflects different expressions of the same basic disease process or in
fact reflects the expression of two different
pathogenetic processes perhaps originated
by different etiologic agents is of course unknown. Such questions may perhaps be
experimentally approached when information about etiologic agents in SLE and related connective tissue diseases become
Arthritis and Rheumatlsm, Vol. 17, No. 4 (July-August 1974)
1. T a n EM, Schur PH, Carr RI, et al: Deoxyribonucleic acid (DNA) and antibodies to
DNA in the serum of patients with systemic
lupus erythematosus. J Clin Invest 45: 17321740, 1966
2. Koffler D, Schur PH, Kunkel HG: Immunological studies concerning the nephritis
of systemic lupus erythematosus. J Exp Med
126:607-623, 1967
3. Sharp GC, Irvin WS, LaRoque RL, et al:
Association of autoantibodies to different
nuclear antigens with clinical patterns of
rheumatic disease and responsiveness to
therapy. J Clin Invest 50:35&359, 1971
4. Reichlin M, Mattioli M: Correlation of a
precipitin reaction to an RNA protein antigen and a low prevalence of nephritis in
patients with systemic lupus erythematosus.
N Engl J Med 286:908-911, 1972
5. Mattioli M, Reichlin M: Characterization
of a soluble nuclear ribonucleoprotein antigen reactive with SLE sera. J Immunol 107:
1281-1290, 1971
6. Mattioli M, Reichlin M: Physical association of two nuclear antigens and mutual
occurrence of their antibodies: The rela-
tionship of the Sm and RNA protein (Mo)
systems in SLE sera. J Immunol 110:13181324, 1973
Sturgill BC, Carpenter RR: Antibody to
ribosomes in systemic lupus erythematosus.
Arthritis Rheum 8:213-218, 1965
Schur PH, Moroz LA, Kunkel HG: Predpitating antibodies to ribosomes in the serum of patients with systemic lupus erythematosus. Immunochemistry 447-453, 1967
Clark G, Reichlin M, Thomas TB: Characterization of a soluble cytoplasmic antigen
reactive with sera from patients with systemic lupus erythematosus. J Immunol 102: 117122, 1969
T a n EM, Kunkel HG: Characteristics of a
soluble nuclear antigen precipitating with
sera of patients with systemic lupus erythematosus. J Immunol 96:464471, 1966
Gonzales EN, Rothfield NF: Immunoglobulin class and pattern of nuclear fluorescence in systemic lupus erythematosus. N
Engl J Med 274:1333-1338, 1966
Seligmann M, Cannat A, Hammard M:
Studies on antinuclear antibodies. Ann NY
Acad Sci 124:816-832, 1965
Arthritis and Rheumatism, Vol. 17, No. 4 (July-Augusf 79741
Без категории
Размер файла
550 Кб
sera, patients, rna, systemic, antigen, erythematosus, reactive, cytoplasmic, lupus, description, heterogeneity, protein, nonribosomal
Пожаловаться на содержимое документа