вход по аккаунту


Demonstration and characterization of a transient arthritis in rats following sensitization of synovial mast cells with antigen-specific ige and parenteral challenge with specific antigen.

код для вставкиСкачать
We have developed a model of IgE-dependent,
mast cell-mediated arthritis in rats. One knee joint (test
joint) of a Sprague-Dawley rat was injected with 1 pg of
a monoclonal IgE specific for dinitrophenol, and the
contralateral (control) joint was injected with the same
amount of an irrelevant monoclonal IgE in phosphate
buffered saline or with phosphate buffered saline alone.
Within 5 minutes of intravenous injection of antigen, an
acute, transient arthritis occurred in the test joints only,
with swelling and extravasation of intravascular blue
dye and '251-labeled albumin, decreased numbers of
stainable mast cells, and decreased histamine content of
the joint synovium. Pretreatment of experimental animals with H1 and H2 antihistamines did not completely
block the reaction. These data show that IgE-dependent
synovial mast cell degranulation causes a transient,
nondestructive arthritis, reminiscent of lupus arthritis
and intermittent hydrarthrosis
Several investigators have reported that the
number of synovial mast cells increases in various
types of arthritis, including rheumatoid arthritis (RA)
(1-3), but precisely how mast cells contribute to
inflammation in synovitis is poorly understood. Indeed, several types of human arthritis, such as lupus
arthritis, intermittent hydrarthrosis, and palindromic
rheumatism, are characterized clinically by their paroxysmal onset, relatively short duration (hours to
From the Mast Cell Physiology Section, Laboratory of
Clinical Investigation, National Institute of Allergy and Infectious
Diseases, National Institutes of Health, Bethesda, Maryland.
Daniel G. Malone, MD; Dean D. Metcalfe, MD.
Address reprint requests to Daniel G. Malone, MD, Department of Medicine, University of Wisconsin, CSC H6/367, 600
Highland Avenue, Madison, W1 53792.
Submitted for publication November 3, 1987; accepted in
revised form February 23, 1988.
Arthritis and Rheumatism, Vol. 31, No. 8 (August 1988)
days), and lack of destructive changes in bone and
cartilage. Such a clinical course is consistent with
synovial mast cell degranulation and the resulting
release and generation of mediators such as histamine
and leukotrienes.
We have begun studies of mast cell involvement
in human arthritis; however, because an IgE-dependent
mast cell-mediated arthritis would likely be paroxysmal
and transient in nature, its effect on more chronic arthritides (such as RA and the ankylosing disorders) would be
difficult to assess in humans. We have, therefore, developed an animal model in rats. The model demonstrates
that an acute, nondestructive arthritis of short duration,
with features typical of an immediate-type hypersensitivity reaction, can be induced by triggering synovial
mast cell degranulation in vivo.
Materials and methods. Outbred male SpragueDawley rats weighing approximately 500 gm were used
in all experiments. Rats were ether-anesthetized, and 1
knee (test-sensitized joint) of each animal was injected
with 50 p1 of phosphate buffered saline (PBS), pH 7.4,
containing 1 pg of purified mouse monoclonal IgE with
antigen specificity for dinitrophenol (DNP) (provided
by Dr. Henry Metzger, NIAMS, NIH, Bethesda,
MD). In all preliminary experiments and in some
subsequent experiments, the contralateral (control)
knee was injected with 50 pl of PBS that contained
either 1 pg of a human IgE myeloma protein directed
against an unknown antigen (provided by Dr. Teruko
Ishizaka, The Johns Hopkins University, Baltimore,
MD) or 1 pg of a mouse monoclonal IgE directed
against ovalbumin (provided by Dr. Reubin Siraganian, NIH).
Neither PBS alone nor PBS with either of the
control myeloma IgEs caused arthritis or extravasation
of Evans blue dye after intravenous (IV) DNP chal-
lenge. In most subsequent experiments, therefore, PBS
alone was used as the control. Acute arthritis was
induced by IV injection of PBS (0.5 cc) containing 0.5%
(weight/volume) Evans blue dye and 500 pg of bovine
serum albumin (BSA) conjugated to DNP (molar ratio
1:5), with or without 1 . 1 1 x 10' Bequerels (3 pCi) of
'251-labeledalbumin (NEN, Boston, MA).
At several timepoints after antigen injection,
selected rats were killed by COz asphyxiation. The
patella and entire infrapatellar tendon (containing the
infrapatellar fat pad) were then removed en bloc.
Immediately thereafter, the rest of the knee joint
(hereafter referred to as the whole knee joint) was
removed by cutting the femur just superior to the
epicondyles, and cutting the tibia just distal to the
tibia1 plateau.
The patella, infrapatellar tendon, and infrapatellar fat pad, all of which are covered by wellcharacterized synovium (4),were used in this study to
demonstrate changes in synovial histology (mast cell
number) and vascular permeability. All tissues were
weighed immediately after removal. In experiments in
which radiolabeled albumin was injected, tissues were
put into vials and counted in a gamma counter
(Gamma 310 system; Beckman, Fullerton, CA), and
were then lyophilized and reweighed. Results are
expressed as counts per minute/mg dry tissue weight,
to correct for weight differences among samples and
the added water weight of edematous tissues. The
percentage of tissue weight that was water weight was
calculated by dividing the difference between the wet
weight and the dry weight by the wet weight.
Tissue histamine content was determined in
infrapatellar tendon structures. Tissues were removed, weighed, immediately boiled in distilled water
for 10-20 minutes, and the supernatants were frozen
and stored at -20°C until assayed. The volume of
supernatant was brought to 1 ml, and the histamine
content was measured using a modification of a sensitive radioenzymatic assay (5). Ten microliters of the
supernatant was mixed with a NaKPO, buffer containing 3H-S-adenosylmethionineand histamine N-methyltransferase, pH 7.9, and incubated at 37°C for 90
The reaction was terminated by adding 2.5M
perchloric acid. The precipitate was removed by centrifugation, and the supernatant was made alkaline and
extracted twice with chloroform. The aqueous phase
was removed, and 100 pl of the remaining organic
phase was counted in a scintillation counter (Model LS
9000; Beckman). Each specimen was assayed in triplicate, and the histamine content of the specimen was
determined by comparing the mean disintegrations per
minute of the 3 samples against a standard curve
constructed from results on specimens of distilled
water containing known amounts of histamine. Results
are expressed as ng of histamine/mg dry tissue weight.
Infrapatellar fat pad synovial mast cells were
enumerated in Carnoy's fluid-fixed tissue sections.
Two sections of each specimen were stained: one with
acidic toluidine blue, and the other with hematoxylin
and eosin (H & E). The number of mast cells per 160 x
field was easily counted on the acidic toluidine
blue-stained sections because of the deep purple metachromatic staining of mast cell granules. Data are
expressed as the percentage of change in mast cell
number/l60 x field, calculated as lOO(Q - I), where
No. of mast cells/l60 x field in test-sensitized knee
No. of mast cells/l60 x field in control knee
To assess the role of histamine in the reaction,
some animals were administered HI and H2 receptor
antagonist drugs before induction of arthritis. Knees of
6 animals were sensitized as described above, except
that 1 pg of monoclonal IgE directed against ovalbumin in 50 pl of PBS was injected into the control knee.
One group of 3 rats was treated with subcutaneous injections of diphenhydramine HCI in PBS (20
mg/kg body weight twice a day for 3 doses) and
intraperitoneal injections of cimetidine (Smith Kline &
French, Philadelphia, PA) in PBS (200 mgikg twice a
day for 3 doses). A second group of 3 rats received
subcutaneous and intraperitoneal injections of equal
volumes of PBS, and served as controls for the antihistamine-treated animals.
Several hours after the last dose of antihistamines, animals were injected with antigen as described
above, and were killed 15 minutes later. Tissues were
processed using the same methods described above.
Differences in means were compared using Student's
1-tailed t-test.
Results. Within 5 minutes of IV injection of
antigen and Evans blue dye (a molecule bound by
albumin and therefore ordinarily confined to the intravascular space), obvious swelling was observed. In
addition, upon removal of the infrapatellar tendon,
swelling and blue coloration of the synovium covering
the deep surface of the tendon could be seen, along
with a small amount of blue synovial fluid. No swelling, color change, or effusion occurred in any of the
experiments in any of the joints injected with either of
the control myeloma IgEs or with PBS. Moreover,
there was no swelling, color change, or effusion in any
test joints when BSA without DNP was injected
intravenously. (In all figures, time 0 is the timepoint
immediately after injection of BSA without DNP.)
To confirm vascular leakage, the water content
of tissues from test-sensitized and control (PBSinjected) joints was measured before and after antigen
challenge. Figure 1 shows that the mean ? SEM water
content in the infrapatellar structures of test-sensitized
joints was greater than that in the control joints in all
rats at 15 minutes after antigen injection (67.6 & 3.1%
in test joints versus 56.9 & 1.3% in control joints; P =
0.016) and at 1 hour after antigen injection (68.1 ?
6.8% in test joints versus 52.9 ? 1.2% in control
joints; P = 0.0025), correlating well with the appearance of clinically observable joint swelling. There was
no measurable difference at 8 hours after antigen
injection or thereafter.
To more precisely quantify the increased vascular
permeability of intraarticular structures, radiolabeled
albumin was intravenously injected at the same time as
the IV antigen and dye. Compared with control joints,
the leakage of radiolabled albumin from the vascular to
the interstitial space in infrapatellar fat pad synovium of
test-sensitized joints was statistically significantly increased for up to 8 hours after antigen challenge.
The mean iz SEM c p d d r y tissue weight of the
test-sensitized joints and that of the control joints,
respectively, and the probability that the difference
between the 2 values was due to chance alone, were as
follows: at 15 minutes, 171.3 ? 54.5 versus 8.5 ? 0.3,
P = 0.0015; at 1 hour, 111.1 ? 35.1 versus 16.8 ?
2.7, P = 0.014; at 8 hours, 32.1 ? 8.1 versus 13.6
iz 0.9, P = 0.012 (Figure 2). The exposure of synovium
to IgE did not by itself account for this leakage, since
there was no difference between test-sensitized and
control joints at time 0, i.e., after injection of dye,
labeled albumin, and BSA without antigen.
Although the time course was similar, when
measuring leakage into whole knee joints, the differences between test-sensitized and control joints were
not as great (at 1 hour, 48.4 ? 7.3 cpm/mg dry tissue
weight versus 29.0 t 3.3, P = 0.025; at 8 hours, 18.8
? 4.1 versus 13.5 2 2.6, P = 0.16), although the ratio
at 8 hours was still > I . The smaller differences resulted from the large background contribution to measured cpm of labeled albumin in tissues such as bone,
bone marrow, and cartilage, which did not participate
in the IgE-dependent, mast cell-mediated reaction.
To demonstrate that mast cell degranulation
was a component of this antigen-specific, IgE-dependent arthritis, we enumerated mast cells in the infrapatellar fat pad synovium. Animals in which joints had
Figure 1. Comparison of water content of infrapatellar fat pad
tissues in test-sensitized and control joints of rats. Water content of
tissues was determined as described in Materials and Methods.
I3 .,N-%o-.
Figure 2. '251-albumin in infrapatellar fat pad tissues after induction
of arthritis with intravenous (IV) antigen injection. At various
timepoints after IV antigen injection, selected animals were killed,
and tissues were counted in a gamma counter. Values are the mean
and SEM cpm/mg dry tissue weight of all animals killed (at least 3
animals, except at 48 hours, where n = 1) at the timepoint shown.
= control-sensitized joints; @....O = test-sensitized
joints. Inset, Mean ratio of c p d m g dry tissue weight in IgEsensitized joints:control joints, plotted according to time after
antigen injection. The ratio in each animal was determined by
dividing the c p d m g dry tissue weight in the test-sensitized knee by
the cpm/mg dry tissue weight in the contralateral control knee.
been injected as previously described, but which received BSA without DNP, showed no significant differences in mast cell counts between test-sensitized
and control joints (mean ? SEM increase 22.2 t
19.3%). A decrease in stainable mast cells is indicative
of degranulation, since it is the granules that stain
metachromatically with acidic toluidine blue. Compared with control joints, there was a decrease in
stainable synovial mast cells in test-sensitized joints
after antigen challenge, which persisted for up to 24
hours. Change was as follows: at 1 hour, -66.6 t
11.3%, P = 0.008 versus control joints; at 8 hours,
-58.6 t 9.9%, P = 0.0029; at 24 hours, -84.3 t
2.0%, P = 0.0027.
In addition, there were readily apparent morphologic differences between the mast cells in the
tissues of test-sensitized joints and those in the control
joints after antigen challenge. Mast cells in control
joints were large cells of various shapes, with numerous tightly packed, purple-staining granules in the
cytoplasm. In contrast, although the granules of mast
cells in test-sensitized joints were the same size as
those in the control tissue mast cells, they were far less
numerous and less densely packed in the cell cytoplasm. We interpreted this as further evidence that
mast cell degranulation had occurred in the testsensitized joints after antigen injection.
To confirm the participation of mast cells in the
process, the histamine content of the infrapatellar
tendon structures was measured. Since histamine is
quickly eliminated in tissues after release from mast
cell granules, histamine content decreases in tissues in
which mast cell degranulation has occurred. As shown
in Figure 3, there was a marked reduction in tissue
histamine content in test-sensitized joints after antigen
injection, while that of control (PBS-injected) joints
was unaffected.
The mean ? SEM ng histamine/gm tissue of the
test-sensitized joints and that of the control joints,
respectively, and the probability that the difference
between the 2 values was due to chance alone, were as
follows: at 0 hour (injection of BSA without antigen),
3,374 ir 520 versus 2,721 2 663, P = 0.24; at 1
hour, 3,884 ? 605 versus 172 k 99, P = 0.002; at
8 hours, 2,519 ? 109 versus 0 t 0, P = 0.000021; at
24 hours, 2,241 ? 487 versus 573 t 572, P = 0.045.
H & E-stained infrapatellar fat pad synovium
was examined by light microscopy to determine
whether there are any observable histologic changes.
None were documented, except for the presence of
interstitial edema and decreases in the number of mast
Y 2000
men Y R I D Q
Figure 3. Histamine content, in ng/gm wet tissue weight, of infrapatellar fat pad tissues at various times after intravenous antigen
challenge. Values are the mean k SEM histamine content of
test-sensitized or control-sensitized tissues; n = 3 at each timepoint.
cell granules and stainable mast cells in test-sensitized
To assess the relative importance of histaminemediated changes in this process, drugs were used to
block both types of histamine receptors before injection of antigen. Treatment with diphenhydramine HC1
and cimetidine (H1 and H2 histamine receptor antagonists, respectively) statistically significantly decreased the infrapatellar fat pad synovial cpm/mg dry
tissue weight at 15 minutes post-antigen injection
(mean t SEM 35.49 k 4.46 versus 171.3 & 54.48,
drug-treated versus controls treated with subcutaneous and intraperitoneal PBS without drug, P = 0.042),
but did not completely abrogate the reaction (mean t
SEM of control-sensitized knees 8.5 ? 0.33). Similarly, the antihistamine treatment resulted in a smaller
increase in measurable tissue water than was seen in
animals not given the antihistamine drugs (8.2 t 2.4%
versus 18.7 t 4.1%, P = 0.05).
Discussion. In this report, we present and characterize for the first time an IgE-dependent, mast cellmediated arthritis in rats. Our investigation was
prompted by increasing evidence that these mechanisms contribute to inflammation in various forms of
human arthritis. In our model, parenteral injection of
specific antigen induced extravasation of intravascular
blue dye in the test-sensitized joints, with accompanying swelling and effusion. Neither control joints nor
periarticular tissues showed any of these abnormalities
in any of the experiments, which confirms that the
reaction involves cells in synovial tissue, and indicates
antigen and antibody specificity.
Measurement of tissue water content revealed a
significant increase in test-sensitized joints by 15 minutes after antigen challenge, and this was still measurable 1 hour later (Figure 1). This is consistent with the
clinically observable joint swelling and blue coloration. We quantified this vascular leakage by demonstrating an increase in interstitial accumulation of
radiolabled albumin, in test-sensitized whole knee
joints, and in the synovium that covers the structures
of the infrapatellar tendon. The increase was detectable for as long as 8 hours after antigen injection
(Figure 2).
An essential aspect of immediate-type hypersensitivity reactions is mast cell degranulation. Test
joint tissues were depleted of stainable mast cells for
as long as 24 hours after antigen challenge, and we
measured a marked decrease in test joint tissue histamine content after antigen challenge (Figure 3).
Blockade of histamine HI and H2 receptors
with antihistamine drugs resulted in an incomplete
reduction of ‘251-albuminleakage into infrapatellar fat
pad tissues, showing that the IgE-dependent mast cellmediated arthritis involves pathophysiologic mechanisms in addition to those attributable to histamine.
This is not surprising, since clinical and experimental
observations have shown that antihistamine drugs
cannot completely inhibit or prevent anaphylactic reactions and that other mediators are involved in the
inflammatory process.
Rapidly expanding experimental and clinical
evidence (6-15) suggests a role for mast cells and mast
cell mediators in arthritis, and has made it clear that
studies of mast cell participation in arthritis are relevant to human disease. Thus, it will be valuable to
study the participation of mast cells in synovitis using
the animal model described in this report. Investigation into the effect of repeated mast cell-mediated
arthritis attacks on the development of chronic, destructive disease is particularly important.
Herein we have provided the first description
and characterization of a reproducible, in vivo model of
synovial mast cell degranulation, which permits study
of the consequences of this event and of how these mast
cell-mediated reactions affect joint tissues in the presence or absence of other models of inducible arthritis in
animals. From such studies, it should be possible to
gain insights into the role of mast cells and mast cell
mediators in the development of acute and of chronic
destructive human arthritides, and into drug or immunologic therapies that might interrupt both processes.
1. Godfrey HP, Ilardi C, Engber W, Graziano FM: Quantitation of human synovial mast cells in rheumatoid
arthritis and other rheumatic diseases. Arthritis Rheum
27~852-856, 1984
2. Malone DG, Irani A-M, Schwartz LB, Barrett KE,
Metcalfe DD: Mast cell numbers and histamine levels in
synovial fluids from patients with diverse arthritides.
Arthritis Rheum 29:956-963, 1986
3. Malone DG, Wilder RL, Saavedra-Delgado AM, Metcalfe DD: Mast cell numbers in rheumatoid synovial
tissues: correlations with quantitative measures of lymphocytic infiltration and modulation by antiinflammatory
therapy. Arthritis Rheum 30: 130-137, 1987
4. Caulfield JP, Hein A, Dynesius-Trentham R, Trentham
DE: Morphologic demonstration of two stages in the
development of type I1 collagen-induced arthritis. Lab
Invest 46:321-343, 1982
5 . Dyer J, Warren K , Merlin S, Metcalfe DD, Kaliner M:
Measurement of plasma histamine: description of an
improved method and normal values. J Allergy Clin
Immunol 70:82-87, 1982
6. Gryfe A, Sanders PM, Gardner DL: The mast cell in
early adjuvant arthritis. Ann Rheum Dis 30:24-30, 1971
7. Bromley M, Fraser WD, Woolley DE: Mast cells at sites
of cartilage erosion in the rheumatoid joint. Ann Rheum
Dis 43:76-79, 1984
8. Marone G, Columbo M, Galeone D, Guidi G, KageySobotka A, Lichtenstein LM: Modulation of the release
of histamine and arachidonic acid metabolites from
human basophils and mast cells by auranofin. Agents
Actions 18: 100-102, 1986
9. Taylor DJ, Yoffe JR, Woolley DE: Mast cell products
and heparin stimulate the production of mononuclear
cell factor by cultured human monocyteslmacrophages.
Biochem J 230:83-88, 1985
10. Grennan DM, Rooney PJ, St. Onge RA, Brooks PM,
Zeitlin IJ, Dick WC: Histamine receptors in synovial
microcirculation. Eur J Clin Invest 5:75-82, 1975
11. Taylor DJ, Yoffe JR, Brown DM, Woolley DE: Histamine stimulates prostaglandin E production by rheumatoid synovial cells and human articular chondrocytes in
culture. Arthritis Rheum 29: 160-165, 1986
12. Marquardt JL, Snyderman R, Oppenheim JJ: Depression of lymphocyte transformation and exacerbation of
Behset’s syndrome by ingestion of English walnuts. Cell
Immunol 9:263-272, 1973
13. Panush RS, Stroud RM, Webster EM: Food-induced
(allergic) arthritis: inflammatory arthritis exacerbated by
milk. Arthritis Rheum 29:220-226, 1986
14. Klofkorn RW, Lehman TJA: Eosinophilic synovial effusions complicating chronic urticaria and angioedema.
Arthritis Rheum 25:708-709, 1982
15. Wasserman SI: The mast cell and synovial inflammation: or, what’s a nice cell like you doing in a joint like
this? Arthritis Rheum 27:841-844, 1984
Без категории
Размер файла
522 Кб
mastr, following, antigen, parenteral, cells, challenge, transiente, ige, specific, demonstration, arthritis, characterization, sensitization, synovial, rats
Пожаловаться на содержимое документа