close

Вход

Забыли?

вход по аккаунту

?

Localization of C-reactive protein in synovium of patients with rheumatoid arthritis.

код для вставкиСкачать
1491
LOCALIZATION OF C-REACTIVE
PROTEIN IN SYNOVIUM OF
PATIENTS WITH RHEUMATOID
ARTHRITIS
JONATHAN D. GITLIN, JOAN 1. GITLIN, and DAVID GITLIN
Synovial biopsies from 8 patients with rheumatoid
arthritis, 2 patients with degenerative osteoarthritis, and 4
patients with nonarthritic disease were studied for localization of C-reactive protein (CRP) using immunofluorescence microscopy. The nuclei of many synoviocytes
and histiocytes in rheumatoid synovial membrane were
found to bind CRP. Cultures of rheumatoid synovium in
14C-labeledamino acids produced radioactive IgG, IgM,
IgA, and C3, but not CRP, indicating the synovial-bound
CRP was not of local origin. A few CRP-binding nuclei
were present in one osteoarthritic synovium, but none was
found in the other and none in synovium from the 4 nonarthritic patients. The nature of the nuclear CRP ligand in
rheumatoid synovium was not determined.
C-reactive protein (CRP) is a normal constituent
of human plasma (1). It is synthesized by all normal
individuals, and its average concentration ranges from
From the Department of Pediatrics, University of Pittsburgh
School of Medicine and the Children's Hospital of Pittsburgh, Pittsburgh, Pennsylvania.
Presented in part before the American Pediatric Society, San
Francisco, California, April 28, 1977.
Supported by USPHS Grants HD-01031 and RR-05507.
Jonathan D. Gitlin, B.S., Joan 1. Gitlin, M.S.H.: medical
students at the University of Pittsburgh School of Medicine and
recipients o f summer research fellowships from the Medical Alumni
Association of the University of Pittsburgh. David Gitlin, M.D.:
Professor of Pediatrics, University of Pittsburgh School of Medicine.
Address reprint requests to Jonathan D, Gitlin,Children.s
Hospital of Pittsburgh, 125 DeSoto Street, Pittsburgh, PA 15213.
Submitted for publication April 8, 1977; accepted July 2,1977.
Arthritis and Rheumatism, Vol. 20, No. 8 (November-December 1977)
70 ng per ml in neonates to 580 ng per ml in adults ( I ) .
The protein has a molecular weight of approximately
120,000 daltons and is composed of identical subunits
having a molecular weight of 24,000 daltons (2). It can
bind to a number of different phosphate monoesters
(3,4) including such diverse substances as pneumococcal
C-polysaccharide (5-7), lecithin, and sphingornyelin (8),
to a variety of polyanions such as DNA, hyaluronic
acid, and chondroitin sulfate (9), and to polycations
including protamine and histone (10). Although the
function of C R P in vivo is unknown, it has been demonstrated in vitro that the protein is endowed with certain
biological attributes: 1) in binding to a ligand, C R P can
activate complement via the classic pathway (8-13); 2) it
can inhibit the response of T-lymphocytes to specific
antigens and to allogeneic cells in mixed lymphocyte
cultures (14-16); 3) it can increase phagocytosis of bacteria by polymorphonuclear leukocytes ( I 7); and 4) it
can inhibit platelet aggregation induced by aggregated
IgG or thrombin (18).
Increases in plasma C R P concentrations usually
occur in response to tissue damage or inflammation and
thus may be observed in such different disorders as
infection, collagen disease, malignancy, and myocardial
infarction. In patients with rheumatoid arthritis, increased plasma C R P levels are usually indicative of active disease (19). Not all patients with clincially apparent rheumatoid arthritis manifest elevated C R P levels,
although almost all adults with the disorder do (19);
only a third of all children with the juvenile form of the
GITLIN ET AL
I492
disease have serum C R P concentrations that are detectable by the usual clinical laboratory procedures (20,21 ).
Current concepts concerning the possible etiology of rheumatoid arthritis are numerous, but may
be grouped, for brevity, into 2 major categories: 1 ) those
that hold that the inflammation is caused by an infectious agent, such as a virus or mycoplasma, and 2)
those that maintain that the disease is the result of an
anomalous immunologic response to a hypothetic antigen, either an infectious agent or a tissue derivative.
Immunologic theories of the disease process can be subdivided into two broad groups: a ) those implicating the
humoral system, involving such proteins as rheumatoid
factor, IgG, and complement either collectively or separately (22,23), and b) those that state that the T-lymphocyte or delayed hypersensitivity is the principal inciter (24-26). It is to be emphasized that these theories
are not necessarily mutually exclusive.
Since CRP when bound to an appropriate ligand
can activate complement and since activation of complement has been implicated in the development of rheumatoid arthritis, the possibility that C R P might bind to
the synovium in this disorder was investigated in the
present study using immunofluorescence and tissue culture.
MATERIALS AND METHODS
Biopsies of knee joint synovium were obtained from
8 patients with rheumatoid arthritis and from 2 patients with
Table 1. C R P Levels in Serum and Joint Fluids and Serum LatexFixation Titers at the Time of Biopsy
C R P level, rg/ml
Age
Patient
(yrs)
Serum
Joint
Fluid
Serum
LatexFixation
Titer
~~
Rheumatoid arthritis
TQ
KR
RM
DS
JB
DD
AF
MM
Postmortem biopsies
DM
MK
PR
FP
17
19
28
6
< I*
92
-
-
4 ‘h
5 Y2
29
49
54
58
60
62
-
15
<I
47
49
42
* Limit of sensitivity of method
3
18
<I
-
II
-
2
6
52
=
1 pg C R P per ml.
I:20
1:640
I : I280
I :320
1:640
1:30
degenerative osteoarthritis undergoing synovectomy for joint
reconstruction. The 8 rheumatoid arthritis patients were female, and each fulfilled the American Rheumatism Association’s criteria for definite or classical rheumatoid arthritis (27).
Two of the 8 patients were 4%and 5 % years of age, respectively, and presented with polyarticular disease; the remaining
6 patients were from 29 to 6 2 years of age (Table I). Four of
the 6 adults with rheumatoid arthritis had “antinuclear antibody” (ANA) titers from 1:30 to 1:lOO using sections of rat
liver as the test antigen in the indirect immunofluorescence
method. Patients JY and AS with degenerative osteoarthritis
of the knee were males, ages 57 and 73 years respectively; the
diagnoses were supported by clinical history, physical findings,
and x-ray changes.
Knee joint synovium was also obtained at autopsy from
4 patients who died of disorders unrelated to arthritis and who
revealed no gross or microscopic evidence of arthritis on postmortem examination: 1 patient died of peritonitis and sepsis
following a pancreatic abscess (FP, Table l ) , another (DM)
died of primary pulmonary hypertension, the third (M K ) succumbed to intracranial hemorrhage, and the remaining patient
( P R ) died of gastrointestinal hemorrhage secondary to hepatic
cirrhosis.
Aliquots of synovial tissue were cut into slices 3 to 5 mm
thick immediately upon procurement, and the slices were frozen
on the sides of dry glass vials using a mixture of 95% alcohol
and solid CO, at approximately -70°C as an external coolant.
Aliquots of fresh synovium were fixed in Bouin’s solution for
histologic examination by light microscopy, and fresh tissue
from 2 of the patients with rheumatoid arthritis was placed in
tissue culture as described below. Renal biopsies, obtained from
2 children ages 4% and 12 years, respectively, who had anaphylactoid purpura, and from 2 adults, ages 39 and 71 years, who
had sclerosing and proliferative glomerulonephritis, respectively, were frozen in the same fashion as the synovial tissues.
Additional control tissues included lymph nodes obtained at
postmortem from 2 patients, ages 2 and 4 years, who died of
congenital heart disease.
Sera and joint fluids were obtained from 5 of the 8
rheumatoid arthritis patients during surgery and from 3 of the
4 patients a t postmortem; serum but not joint fluid was obtained from the fourth (Table I). The sera and joint fluids
were assayed for C R P using radial immunodiffusion in agar
(28); the limit of sensitivity of the method using goat antiserum
specific for human C R P (Miles Laboratories, Elkhart, Indiana) was 1 p g per ml. The C R P standard used for the assay
was a preparation of purified C R P isolated from the pleural
fluid of patients with neoplastic disease by the method of
Hokama and his colleagues (29).
Plasma proteins other than C R P were not detected in
this standard, when it was examined by means of immunoelectrophoresis (30) and double diffusion in agar (31). The
preparation gave a precipitin line with the goat anti-CRP, but
gave no precipitin lines with any of the antisera used in this
study against other plasma proteins including the rabbit antiserum against pooled normal adult plasma ( 3 2 ) . When
immunoelectrophoresis and double diffusion in agar were
used, the goat anti-CRP was shown to be specific for C R P the
antiserum formed a single precipitin line when it was reacted
with adult human plasma containing detectable amounts of
CRP, and this line fused with a precipitin line formed in the
C-REACTIVE PROTEIN
reaction between the antiserum and the purified C R P standard.
Rabbit antisera specific for human IgG, IgA, IgM, C3,
ceruloplasmin, and fibrinogen were prepared as described elsewhere (32). The antiserum against fibrinogen reacted with
fibrin as well as fibrinogen; the other antisera were specific for
their homologous protein antigens as determined by testing
against adult plasma and purified plasma proteins using
immunoelectrophoresis. A rabbit antiserum was also prepared
using pooled adult plasma as the antigen as previously described (32); the antiserum contained precipitating antibodies
against at least 17 different plasma proteins, including albu-
min, a,-antitrypsin, IgM, and IgG among others.
Aliquots of the rabbit antisera specific for human IgG,
IgA, IgM, and fibrinogen and- aliquots of goat anti-human
C R P were fluoresceinated by conjugation with fluorescein isothiocyanate as described by Rinderknecht (33). Briefly, antibody-containing globulins were precipitated from the antisera
at 4°C by mixing the antisera with equal volumes of a saturated solution of ammonium sulfate. The precipitated globulins were collected by centrifugation, redissolved in minimal
amounts of distilled water, and dialyzed against repeated
changes of 0.15 M NaCl until the dialysates were free of
detectable NH,+. Conjugation of the globulin fractions was
performed a t 4°C in an acetone, saline, bicarbonate buffer
mixture at p H 9.0 using 0.05 mg of fluorescein isothiocyanate
per mg of protein in the globulin fraction. To remove fluorescein products not bound to protein, the conjugated antisera
thus prepared were dialyzed at 4°C against several changes of
0.15 M NaCl containing 0.001 M phosphate buffer, pH 8.5,
over a 3-day period, and were then passed through columns of
Sephadex G-25 (Pharmacia, Uppsala, Sweden).
Frozen synovial tissues were cut into 6 p sections by
means of a cryostat and placed on glass slides without fixative.
Sections were immersed in 95% alcohol for 15 to 30 seconds to
adhere the sections to the slides, and the slides were then
washed in 0.1 M NaCl containing 0.05 M phosphate buffer at
pH 7.4. Duplicate sections were not treated with 95% alcohol
before washing, and although the results of immunofluorescence were exactly the same with or without alcohol,
there was a disconcerting loss of nontreated sections from the
slides during subsequent staining with hematoxylin and eosin.
The brief treatment in 95% alcohol resulted in better adherence
of the sections to the slides and did not inhibit the removal of
free or unbound C R P and other plasma proteins from the
sections by washing.
Although the binding of C R P to a t least some of its
ligands is C a + + dependent, the results of immunofluorescence
using either phosphate buffered saline or tris (hydroxymethyl)
aminomethane buffer at pH 7.4 as the washing solution were
the same with or without added 0.01 M CaCI,. After washing
in 3 or more changes of the phosphate buffered saline over a
20-minute period, the slides were drained and the sections were
flooded with specific fluoresceinated antiserum. After incubation at room temperature for 20 minutes, antiserum was
washed from the sections using at least 3 changes of phosphate
buffered saline over a period of 20 to 30 minutes. The sections
were then mounted in glycerol buffered with one part of 0.1 M
phosphate buffer, pH 8, to 9 parts of glycerol.
Examination for immunofluorescence was performed
using a tungsten lamp in conjunction with interference filters
1493
(FITC filters, Optisk Laboratorium, Lyngby, Denmark) as
described by Rygaard and Olsen (34,35). Selected microscopic
fields were photographed using Tri-X film (Eastman Kodak,
Rochester, New York), and the fields were located on the
slides using a vernier mechanical stage. The sections were then
fixed in 95% alcohol containing I% HCI and stained with
hematoxylin and eosin. The fields previously photographed in
the ultraviolet microscope were relocated using the vernier
stage and rephotographed using light microscopy. To ascertain the specificity of observed fluorescence, aliquots of the
fluoresceinated antisera were absorbed with putified homologous antigens in the zone of antigen excess, thus removing or
binding specific fluoresceinated antibodies. These absorbed
sera were then used to stain duplicate synovial sections.
As a further control for the specificity of immunofluorescence, sections from each synovium were flooded with
unlabeled antiserum and washed in buffered saline before
staining with unabsorbed homologous fluoresceinated antiserum. Other tissue sections, which were not pretreated in 95%
alcohol, were incubated with C R P for 20 minutes at room
temperature by flooding them with either purified C R P or
rheumatoid serum containing C R P (patient R M ) before washing; the sections were then washed and stained with fluoresceinated an ti-CRP .
To examine the role of RNA and D N A on the localization of CRP, frozen sections of synovium were placed in
95% ethanol for 30 seconds. These sections and duplicate
sections not pretreated with alcohol were washed in 0.1 M
NaCl for 20 minutes and then treated with a 1% solution of
either crystalline bovine pancreas ribonuclease (Worthington
Biochemical, Freehold, New Jersey) in 0.1 M acetate buffer,
pH 5.0, or purified porcine spleen deoxyribonuclease (Worthington Biochemical) in 0.1 M acetate buffer, pH 4.6, for a
period of 10 minutes to I hour (36,37). The sections were
washed in several changes of phosphate buffered saline over a
period of 25 minutes and were then incubated with serum from
patient RM at room temperature for 20 minutes. After washing i n phosphate buffered saline for 25 minutes, the sections
were stained with fluoresceinated anti-CRP. After examination for im munofluorescence the sections were stained for
RNA and D N A using standard methyl green-pyronin and
Feulgen-Schiff methods, respectively.
Aliquots of synovium from 2 of the rheumatoid arthritis patients (RM and A F ) were minced with scissors immediately after acquisition, and approximately 250 mg of tissue
were incubated in roller tubes for 3 days at 37°C (32). The
tubes contained 2 ml of minimum essential medium (Grand
Island Biological, Grand Island, New Y o r k ) with 4 pCi of
either L-leucine uniformly labeled with “C or a mixture of
amino acids obtained from “C-labeled algal hydrolysates. Six
roller tube cultures were prepared from each synovium using
different areas of the biopsy for each culture. At the end of
incubation, the cultures were frozen and thawed twice,
dialyzed against distilled water for 4 days, lyophilized, and
reconstituted to 0.1 ml with 0.1 M borate buffer, pH 8.6.
Serum from a patient with rheumatoid arthritis who had an
elevated C R P level was added to aliquots of each culture as a
protein carrier, and immunoelectrophoresis was performed in
agar. All electrophoreses were performed in duplicate. The
slides were developed with antiserum specific for CRP, IgG,
IgA, C3, or ceruloplasmin or with antiserum against adult
GITLIN ET AL
I494
plasma for 3 days, washed in daily changes of 0. I h4 NaCl for
5 days, and dried. They were then inverted on Tri-X film Cor 7
weeks for autoradiography.
RESULTS
The pathology observed by light microscopy in
the synovial tissues that had been fixed in Bouin's solution was compatible with the clinical diagnoses. The
synovium appeared to be normal in each of the biopsies
obtained at postmortem from the 4 nonarthritic patients. I n the 2 patients with degenerative arthritis, the
villi were hyperplastic, because of a n increase in synoviocytes and connective tissue, but some follicles of lyrnphocytes were present in the synovial membrane. I n the
rheumatoid arthritis patients the synovium was greatly
thickened and often disorganized, the synoviocyteswere
hyperplastic and covered in areas by fibrin, edema was
apparent, the synovial membrane was markedly infiltrated with lymphocytes and plasma cells, and polyrnorphonuclear leukocytes were evident.
FiE 1. Synovid .ttNiUnSfrOitl rheumatoid patiem. /irtirtuno~uore.Pcenrp
shuninz fihrin at jrer synouial surface and in synocioryie l a w ofputieni
DD 1 A 1, lgG, IgA , and CR P in .rynocial nienrbrane of pmirnr M M I B.
C.and D. rexpccriL'pIL'J. Arruw in C indirarm IgA in u typical fimnia
crll. 1 X 84. /
1 mmunofluorescence revealed fibrin deposition
on the articular surface of and within the synoviocyte
layer at various sites in the synovium in each of the
arthritic patients studied, including the 2 patients with
degenerative osteoarthritis (Figure I A ) . This was not
observed in synovium from the 4 nonarthritic patients.
I n the arthritic patients IgG, IgA, and IgM were Found
in the cytoplasm of many plasma cells (Figure I); of the
three immunoglobulins, the IgG-containing cells were
greatest in number and the IgM-containing cells were
the least in number. In the rheumatoid synovium the
distribution of IgG was extensive (Figure 131, as has
been described by Munthe and Natvig (38); 1gG was
Found in some intercellular spaces even after prolonged
washing prior to staining.
On t h e other hand, fluorescence attributable to
bound CRP was seen in the synovium o f each rheumatoid patient in discrete bodies distributed in the synoviocyte layer and in the subsynoviocyteconnective tissue
Of the 'ynovial membrane (Figures I D and 2 ) + That Ibis
Ruorcscence did indeed represent localized CRP was
C-REACTIVE PROTEIN
indicated by the finding that it was not seen in those
sections stained with fluoresceinated anti-CRP from
which antibodies against C R P had been removed by
absorption with purified CRP. In addition, when sections were treated with unlabeled antiserum against
CRP before staining with fluoresceinated anti-CRP, the
fluorescence was either markedly inhibited or abolished.
CRP was not detected in synovium obtained at
postmortem from the 4 nonarthritic patients, nor was it
found in any of the renal or lymph node biopsies examined. No C R P binding could be found in sections of
synovium from 1 of the osteoarthritic patients despite
repeated search (Figure 2). However, in the other osteoarthritic patient, one small area in a single section of
four synovial sections examined contained approximately half a dozen fluorescent bodies that were specific
for CRP; the number of such bodies in this synovium
was relatively insignificant compared to the number seen
in any of the patients with rheumatoid arthritis.
When frozen synovial, renal, and lymph node
sections were incubated with either C R P or rheumatoid
1495
serum containing CRP, no detectable change in CRP
binding was found in the tissues. Those tissues in which
CRP binding was not observed before the in vitro exposure to CRP were found not to bind detectable
amounts of C R P after the in vitro exposure to CRP.
Incubating sections with either deoxyribonuclease or
ribonuclease for 1 hour resulted in the virtually complete loss of DNA or RNA, respectively, as indicated by
the Feulgen-Schiff and methyl green-pyronin reactions,
but CRP binding remained unaffected. The enzymatic
digestions neither increased nor decreased the number
of CRP binding bodies, whether the sections had been
pretreated with 95% alcohol or not.
On examination of the frozen sections with light
microscopy after restaining with hematoxylin and eosin,
it was found that the fluorescence specific for CRP appeared to have emanated from cell nuclei (Figure 3).
Most of the nuclei were obviously vesicular; a few were
somewhat more compact. The morphology of the cells
in these sections was considerably altered by the procedures used, since tissue fixatives were not used until
Fig 3. Synovial section from rheumatoid patient K R , showing localization of CR P. A. Immunofluorescence; arrows indicate 3 typical
fluorescent bodies. B. Light microscopy of same area as in A after restaining with hematoxylin and eosin: arrows point to the nuclei from which
fluorescence emanated, the upper two are within the synoviocyte layer. ( X 210.)
I496
Fig 4. Ittiniunoelectrophoresis of synovial cultures developed with antiCRP 1 A). anri-\t,hole plastna (B), anti-lgG ( C ) ,anti-lgA (D).anti-C3
(El. and anti-ceruloplastnin (Fi. The lower halJ of each figure is the
autoradiograph of the ittitnunoelectrophoretic partern. Arrows in A indicate the trough in which anti-CRP had been placed; comparison ofthe
upper half' of Jigure with the lower half' reveals absence of radioactive
CRP. In B.arrows Alb. A t . M . and G indicate albutnin. a,-antitrypsin,
IgM, and IgG, respective1.v: the precipirin line for IgM. hidden in the
irnniunoelectrophoretic pattern. is easily seen as radioactive in the autoradiograph. Arrows in E and F indicate radioactive C3 and ceruloplasmiti. respecticelj,.
after immunofluorescence microscopy. Therefore, the
identity of the cells that bound C R P could not be established with any degree of certainty. From the localization of many of the CRP-binding nuclei in the synoviocyte layer (Figure 3) and the resemblance of these
nuclei to the predominant type of nucleus in this layer, it
seemed that most of these CRP-binding nuclei represented synoviocytes. Of the CRP-binding nuclei in the
subsynoviocyte tissue, most were larger and more lightly
staining than the nuclei of obvious lymphocytes and
appeared to represent histiocytes, macrophages, and
possibly fibroblasts. Lymphocyte and plasma cell nuclei
did not appear to bind CRP, but this observation could
not be definitely established. Although many CRP-binding nuclei were found in each of the rheumatoid synovia,
the number of such nuclei were relatively few compared
to the total number of nuclei of similar cell types in the
tissue.
The numbers of CRP-binding nuclei in the sections of rheumatoid synovia studied did not correlate
with the serum latex-fixation titer or the CRP level in
either serum or joint fluid. The greatest numbers of
CRP-binding nuclei were seen in patients TQ, KR, and
MM, with somewhat lesser numbers in patients DS, JB,
and AF, and still less in patients RM and DD. Although
patients FP and PR (Table 1 ) had CRP levels comparable t o those found in the rheumatoid patients, no
GITLIN ET AL
CRP binding sites were detected in the synovium of
either individual. Incubating sections of synovium with
either purified CRP or serum from patient RM that
contained 28 pg of C R P per ml did not result in any
synovial localization of CRP in the nonarthritic patients
or in the osteorarthritic patient who did not previously
manifest CRP binding. In the other osteoarthritic patient and in the rheumatoid patients such treatment with
CRP did not increase C R P localization beyond that seen
without pretreatment. Thus absence of synovial CRP
localization in any of the patients studied did not appear
to be simply the result of an absence of C R P in the
tissue.
The cultures of synovium in 14C-labeled amino
acids revealed the production of radioactive IgG, IgM,
IgA, C3, and surprisingly, ceruloplasmin (Figure 4), the
production of these proteins being a manifestation of
the viability of the tissue in the culture. On the other
hand detectable amounts of radioactive CRP, albumin,
or a,-antitrypsin were not found (Figure 4,A and B).
DISCUSSION
When C R P binds to an appropriate ligand in
vitro, it can activate the complement system (8-13). In
the present study CRP was found bound to certain
synovial nuclei in patients with rheumatoid arthritis. It
does not necessarily follow, however, that these nuclei
did actually bind C R P in vivo to activate complement:
first, it is not known whether the cells to which these
nuclei belong were alive at the time of the biopsy, and
second, it is not known whether CRP can penetrate a
living cell to reach the nucleus. It is to be noted that
plasma proteins with the molecular size of CRP are
present in virtually all interstitial fluids as well as blood
vessels (39). Thus, when frozen tissue sections thaw,
even momentarily, the cut cells are exposed to plasma
proteins, including CRP.
This study does not reveal whether CRP actually
interacted with a nuclear ligand in vivo, but it does
reveal that certain synovial nuclei in patients with rheumatoid arthritis do contain a substance that is capable
of binding CRP. Although both DNA and histone can
bind CRP (9, lo), if either of these macromolecules was
the ligand, then it must have been structurally different in some way from those in the other nuclei of the
same cell type that did not bind CRP. If it is argued that
the CRP ligand was either partially depolymerized
DNA or a histone that had dissociated from the DNA,
it should be noted that many cells present were in various states of degeneration and none of these appeared to
C-REACTIVE PROTEIN
bind CRP detectable under the conditions of study. In
addition, depolymerization of DNA with deoxyribonuclease did not result in an increase in the number of
CRP-binding nuclei, nor did it decrease the number that
did bind CRP. Treatment with ribonuclease also did not
affect CRP binding. Although dissociation of DNA
from histone might occur temporarily during cell replication, the CRP-binding cells did not appear to be in
any active stage of mitosis.
Whatever the CRP-binding ligand was, it was
diffusely distributed within the nucleus. It is almost unnecessary to state that it may be an anomalous product
and could be either endogenous or exogenous in origin.
Four of the adult rheumatoid patients in this study,
RM, JB, AF, and MM, had ANA titers from 1:30 to
1: 100. These antibodies, although directed against nuclear material, were relatively nonspecific regarding the
cell types with whose nuclei they would react. In fact the
antibody titers given were obtained by using the nuclei
in rat liver sections as the antigen in the standard indirect immunofluorescence test. It would seem unlikely,
therefore, that CRP-binding by specific nuclei was the
result of interaction between the nuclei and these antinuclear antibodies. Even incubation of the sections with
rheumatoid serum from patient RM which had an ANA
titer of 1:30 did not alter the number of nuclei that
bound CRP. Serum antibodies specific for the nucleoprotein of polymorphonuclear leukocytes have been
reported in patients with rheumatoid arthritis (40), but
the cells that bound C R P were not polymorphonuclear
leukocytes. The possibility that a specific antinucleoprotein antibody might, in binding to its antigen,
also result in C R P binding cannot be ruled out at this
point, but by the same token, one must also consider the
converse, that the nuclear alteration that permits CRP
binding might be a stimulus for antinuclear antibody
production.
The nuclear ligand for C R P was present in each
of the rheumatoid synovia, and in terms of numbers of
nuclei involved, to a much lesser degree in one of the
two osteoarthritic synovia studied. Bound C R P was not
observed in synovium from the other osteoarthritic patient or from any of the 4 nonarthritic patients; flooding
the synovial sections with C R P did not result in detectable C R P binding. Similarly, CRP did not bind to kidney or lymph node from nonarthritic patients when
sections of these tissues were treated with CRP. The
relative numbers of CRP-binding nuclei in rheumatoid
synovium did not appear to change when sections were
exposed to C R P before staining with fluoresceinated
CRP.
1497
It is clear from the tissue culture studies that the
synovium does not synthesize CRP, and hence the CRP
that was bound to nuclei came from the plasma. This is
in accord with the observation that the liver appears to
be the source of C R P in humans (41). The synovial
cultures did produce IgG, IgM, IgA, and C3 as has been
found by others (23,42), indicating the viability of the
tissues in culture.
The CRP-binding nuclei appeared to be the nuclei of synoviocytes, histiocytes, and possibly fibroblasts. Few, if any, lymphocytes or plasma cells seemed
to bind CRP. Although C R P binding by the cell surface
of lymphocytes can be demonstrated by immunofluorescence using preparations of intact peripheral lymphocytes (1 4), lymphocytic binding, surface or otherwise, was not detected in the sections studied here. This
was not unexpected, since surface immunoglobulins on
lymphocytes are not readily detected in tissue sections
either. In rabbits given typhoid vaccine intramuscularly,
Kushner and Kaplan (43) found Cx-reactive proteir
(CxRP) localized to the cytoplasm of necrotic myo
fibers in the inflammatory area, but lymphocytes, macrohpages, and polymorphonuclear leukocytes did not
contain the protein. In rabbits with myocardial infarction produced by coronary artery occlusion, CxRP
localized only in necrotic cardiac myofibers (44). It is
apparent that the localization of CxRP in inflammatory
or necrotic sites in rabbits was different from the nuclear
localization of CRP in the synovial inflammation of
rheumatoid arthritis.
ACKNOWLEDGMENTS
The authors are grateful for this opportunity to express their indebtedness to Dr. William T. Green, Jr. of the
Children’s Hospital of Pittsburgh for his interest and encouragement and for his cooperation in obtaining the synovial
biopsies, and to the orthopedic residents at St. Margaret’s
Hospital in Pittsburgh for their patience with our requests for
blood, joint fluid, and synvoial biopsies.
REFERENCES
I . Claus DR, Osmand AP, Gewurz H: Radioimmunoassay
of human C-reactive protein and levels in normal sera. J
Lab Clin Med 87:120-128, 1976
2. Gotschlich EC, Edelman G M : C-reactive protein: a molecule composed of subunits. Proc Natl Acad Sci
54~558-566, 1965
3. Hornung MO, Berenson GS: Some biochemical and serologic properties of the pneumococcal C polysaccharide.
Proc SOC Exp Biol Med 114:31-37, 1963
4. Gotschlich EC, Edelman G M : Binding properties and
GITLIN ET AL
specificity of C-reactive protein. Proc Natl Acad Sci
57:706-712, 1967
5. Tillet WS, Francis T Jr: Serological reactions in pneumonia with a non-protein somatic fraction of pneumococcus. J Exp Med 52561-571, 1930
6. Abernathy TJ, Avery OT: The occurrence during acute
infections of a protein not normally present in the blood. J
Exp Med 73173-182, 1941
7. Volanakis JE, Kaplan MH: Specificity of C-reactive protein for choline phosphate residues of pneumococcal polysaccharide. Proc SOCExp Biol Med 136:612-614, 1971
8. Kaplan MH, Volanakis JE: Interaction of C-reactive protein complexes with the complement system. I. Consumption of human complement associated with the reaction of
C-reactive protein with pneumococcal C-polysaccharide
and with the choline phosphatides, lecithin and sphingomyelin. J Immunol 112:2135-2147, 1974
9. Claus DR, Siegel J, Petras K, et al: Complement activation by interaction of polyanions and polycations. I I I .
Complement activation by interaction of multiple polyanions and polycations in presence of C-reactive protein. J
lmmunol I 1 8:83-87, 1977
10. Siegel J, Osmand AP, Wilson MF, Gewurz H: Interactions of C-reactive protein with the complement system.
11. C-reactive protein-mediated consumption of complement by poly-L-lysine polymers and other polycations.
J Exp Med 142:709-721, 1975
11. Volanakis JE, Kaplan MH: Interaction of C-reactive protein complexes with the complement system. 11. Consumption of guinea pig complement by C R P complexes:
requirement for human Clq. J Immunol 113:9-17, 1974
12. Siegel J, Rent R, Gewurz H: Interactions of C-reactive
protein with the complement system. I. Protamine-induced consumption of complement in acute phase sera. J
Exp Med 140:631-646, 1974
13. Osmand AP, Mortensen R F , Siegel J, Gewurz H: Interactions of C-reactive protein with the complement system.
111. Complement-dependent passive hemolysis initiated by
CRP. J Exp Med 142:1065-1076, 1975
14. Mortensen RF, Osmand AP, Gewurz H: Effects of Creactive protein on the lymphoid system. I. Binding to
thymus-dependent lymphocytes and alteration of their
functions. J Exp Med 141:821-839, 1975
15. Mortensen, RF, Gewurz H: Effects of C-reactive protein
on the lymphoid system. 11. Inhibition of mixed lymphocyte reactivity and generation of cytotoxic lymphocytes. J Immunol 116:1244-1250, 1976
16. Mortensen RF, Braun D, Gewurz H: Effects of C-reactive
protein on lymphocyte functions. I l l . Inhibition of antigen-induced lymphocyte stimulation and lymphokine
production. Cell Immunol 2859-68, 1977
17. Kindmark C - 0 : Stimulating effect of C-reactive protein
on phagocytosis of various species of pathogenic bacteria.
t l i n Exp lmmunol 8:941-948, 1971
18. Fiedel BA, Gewurz H: Effects of C-reactive protein on
platelet function. I. Inhibition of platelet aggregation and
release functions. J Immunol I16:1289-1294, 1976
19. McEwen C, Ziff M: Basic sciences in relation to rheumatic
diseases. Med Clin North Am 39:765-782, 1955
20. Griffin PP, Tachdjian OM, Green WT: Pauciarticular arthritis in children. JAMA 184:23-28, 1963
21. Brewer EJ Jr: Laboratory data, Juvenile Rheumatoid Arthritis, Edited by EJ Brewer Jr, Philadelphia, WB Saunders, 1970, pp 68-79
22. Zvaifler NJ: Rheumatoid synovitis: an extravascular immune complex disease. Arthritis Rheum I7:297-3O4, 1974
23. Ruddy S, Colten H R: Rheumatoid arthritis: biosynthesis
of complement proteins by synovial tissues. N Engl J Med
290:1284-1288, 1974
24. Janeway CA, Gitlin D, Craig J M , Grice DS: “Collagen
disease” in patients with agammaglobulinemia. Trans Assoc Am Physicians 69:93-97, 1956
25. Gitlin D, Janeway CA, Apt L, Craig JM: Agammaglobulinemia, Cellular and Humoral Aspects of Hypersensitive
States. Edited by HS Lawrence, New York. Hoeber-Harper, 1959, pp 375-441
26. van Boxel JA, Paget SA: Predominantly T-cell infiltrate in
rheumatoid synovial membranes. N Engl J Med
293:s 17-520, 1975
27. Ropes MW, Bennett GA, Cobb S, et al: 1958 Revision of
diagnostic criteria for rheumatoid arthritis. Arthritis
Rheum 2:16-18, 1959
28. Mancini G, Carbonara AO, Heremans JF: Immunochemical quantitation of antigens by single radial
immunodiffusion. Immunochemistry 1:235-254, 1965
29. Hokama Y, Tam R, Hirano W, Kimura L: Significance of
C-reactive protein binding by lecithin: a simplified procedure for C R P isolation. Clin Chim Acta 5053-62, 1974
30. Scheidegger JJ: Une micro-mbthode de I’immunoelectrophorese. Int Arch Allergy Appl Irnmunol 7: 103110, 1955
31. Ouchterlony 0:Antigen antibody reactions in gels. IV.
Types of reactions in coordinated systems of diffusion.
Acta Path Microbiol Scand 32:231-240, 1953
32. Gitlin D, Biasucci A: Development of yG, yA, y M ,
PLC/PIA,C’I esterase inhibitor, ceruloplasmin, transferrin, hemopexin, haptoglobin, fibrinogen, plasminogen, a,antitrypsin, orosomucoid, &lipoprotein, a,-macroglobulin, and prealbumin in the human conceptus. J Clin Invest
48: 1433-1446, 1969
33. Rinderknecht H: Ultra-rapid fluorescent labelling of protein. Nature 193:167-168, 1962
34. Rygaard J, Olsen W: Interference filters for improved
immunofluorescence microscopy. Acta Path Microbiol
Scand 76:146-148, 1969
35. Rygaard J, Olsen W: Determination of characteristics of
interference filters. Ann NY Acad Sci 177:430-433, 1971
36. Kalnitsky G, Hummel JP, Dierks C: Some factors which
affect the enzymatic digestion of ribonucleic acid. J Biol
Chem 234:1512-1519, 1959
37. K unitz M: Crystalline desoxyribonuclease. 11. Digestion
of thymus nucleic acid (desoxyribonucleic acid): the kinetics of the reaction. J Gen Physiol 33:363-377, 1950
38. Munthe E, Natvig JB: Gammaglobulin complexes and
C-REACTIVE PROTEIN
anti-y-globulin antibodies in rheumatoid tissue and tissue
eluates. Proceedings of the International Symposium on
Immune Complex Diseases. Edited by L Bonomo, JL
Turk, Milan, Carlo Erba Fdn, 1970, pp 86-97
39. Gitlin D, Gitlin JD: Fetal and neonatal development of
human plasma proteins, The Plasma Proteins. Edited by
FW Putnam, New York, Academic Press, 1975, pp
32 1-374
40. Elling, P: Antinuclear factors in rheumatoid arthritis: increasing organ-non-specificity with increasing reactivity of
rheumatoid factor with heterologous gamma globulin.
Ann Rheum Dis 27:406-413, 1968
4 I . Hurlimann J, Thorbecke GJ, Hochwald GM: The liver as
1499
the site o f C-reactive protein formation. J Exp Med
123:365-378, 1966
42. Ford DK, Smiley JD: Continuous culture of a B-immunocyte from rheumatoid synovium. Arthritis Rheum
16:341-344, 1973
43. Kushner I , Kaplan MH: Studies on acute phase protein. I.
An immunohistochemcal method for the localization of
Cx-reactive protein in rabbits: association with necrosis in
local inflammatory lesions. J Exp Med I14:961-979, 1961.
44. Kushner I , Rakita L, Kaplan MH: Studies of acute-phase
protein. 11. Localization of Cx-reactive protein in heart in
induced myocardial infarction in rabbits. J Clin Invest 42:
286-292, I963
Документ
Категория
Без категории
Просмотров
2
Размер файла
1 955 Кб
Теги
synovium, patients, protein, arthritis, localization, reactive, rheumatoid
1/--страниц
Пожаловаться на содержимое документа