close

Вход

Забыли?

вход по аккаунту

?

The effects of the phospholipase a2 inhibitor manoalide on cartilage degradation stromelysin expression and synovial fluid cell count induced by intraarticular injection of human recombinant interleukin-1╨Ю┬▒ in the rabbit.

код для вставкиСкачать
AK1?IKII'IS & KIIEUMATISM
Vol. 39, No. 8, August 1996, pp 1292-1299
0 1996, American College of Kheurnatology
1292
THE EFFECTS OF THE PHOSPHOLIPASE A2 INHIBITOR7MANOALIDE,
ON CARTILAGE DEGRADATION, STROMELYSIN EXPRESSION7AND
SYNOVIAL FLUID CELL COUNT INDUCED BY
INTRAARTICULAR INJECTION OF
HUMAN RECOMBINANT INTERLEUKIN-la IN THE RABBIT
DENIS J. SCHRIER, CRAIG M. FLORY, MARTIN FINKEL, SUSAN L. KUCHERA,
MARK E. LESCH, AND PEER B. JACOBSON
Objective. To evaluate the effects of the phospholipase A, (PLA,) inhibitor manoalide on cartilage degradation, stromelysin expression, and inflammatory
cell accumulation in rabbits treated intraarticularly
with recombinant human interleukin-la (rHuIGla).
Methods. Rabbits were given an intraarticular
injettion of rHuIL-la. At various time points over a
24-hour period, the rabbits were euthanized and the
articular space was lavaged with sterile PBS. The proteoglycan content of the lavage fluid was measured using
a dimethylmethylene blue assay. PLA, activity and
differential cell counts were also measured. The femur
was removed and cartilage proteoglycan content determined. In some experiments, levels of synovial stromelysin messenger RNA (mRNA) were assessed. Manoalide or vehicle was administered 30 minutes before the
rHuIL-la injection.
Results. The rHuIL-la-induced arthritic response is characterized by significant accumulation of
inflammatory cells, loss of proteoglycan from the condylar cartilage, and induction of mRNA for stromelysin.
PLA, activity was also elevated in synovial fluids from
rhIL-la-injected joints. Pretreatment with manoalide
(0.3 mg/joint) significantly inhibited PLAz activity in the
sflovial fluid, prevented the loss of proteoglycan from
the condylar cartilage, and reduced proteoglycan levels
in lavage fluids. However, manoalide either had no effect
on, or stimulated, cell accumulation. To assess the
Denis J. Schrier, PhD, Craig M. Flory, PhD, Martin Finkel,
L. Kuchera, BS, Mark E. Lesch, MS, Peer B. Jacobson,
PhD: Parke Davis Pharmaceutical Research, Ann Arbor, Michigan.
Address reprint requcsts to Denis Schricr, PhD, Parke Davis
Pharmaceutical Research, 2800 Plymouth Road, A n n Arbor, MI
48105.
Submitted for publication August 9, 1995; accepted in revised
form March 13. 1996.
Ids, Susan
relationship between the induction of PLA, and stromelysin, levels of stromelysin mRNA were measured in
synovial tissue from manoalide- and vehicle-treated
joints. Stromelysin message levels were significantly
suppressed in a dose-dependent manner.
Conclusion. These studies demonstrate that
manoalide is a potent inhibitor of inflammation and
cartilage catabolism, and suggest that PLA, is involved
in the pathophysiology of rHuIL-la-induced arthritis
in rabbits.
The release of proteoglycan fragments from articular cartilage is an important feature of both rheumatoid arthritis (RA) and osteoarthritis. The cartilage
proteoglycan matrix gives elasticity and resilience to
cartilage, allowing the distribution of mechanical forces
across the weight-bearing joint. The synovium releases a
variety of mediators including interleukin-1 (IL-1) that
stimulate chondrocytes to degrade extracellular proteoglycan, and extensive evidence suggests that IL-1 is
involved in the pathology of chronic inflammatory arthritis. These studies have shown that IL-1 levels are
elevated in synovial fluid from arthritis patients (1-4),
messenger RNA (mRNA) for IL-1 is expressed at
increased levels in RA synovial cells (5), IL-1 induces
chondrocyte production of matrix metalloproteinases,
including stromelysin and collagenase ( 6 - 4 , IL-1 depresses proteoglycan biosynthesis by articular chondrocytes (9,10), and intraarticular injection of IL-la or
IL-lP into rabbits induces an arthritic response that
includes cellular infiltration and proteoglycan release
from articular cartilage (11-13).
In addition to its direct effects on cartilagedegrading enzymes and connective tissue biosynthesis,
IL-1 has a number of other characteristics that contrib-
MANOALIDE EFFECTS ON IL-1 a-INDUCED ARTHRITIS
ute to its mode of action. Recent studies have shown that
IL-1 has a pronounced effect on cicosanoid biosynthesis
a n d phospholipase A, (PLA,) activity in synovial cells
and chondrocytes. In rheumatoid synovial fibroblasts,
stimulation with IL-1 leads to induction of prostaglandin
H synthase-2 and thc cytosolic form of PLA, (14,15),
which correlate with a n increase in prostaglandin E,
rclcasc. Moreover, treatment of rabbit chondrocytcs
with IL-1 leads t o a n increase in m R N A for the 14-kd
synovial fluid PLA, (1 6). These findings are analogous t o
observations that clevated levels of multiple isoforms of
PLA, are present in the synovial fluid, cartilage, a n d
serum of patients with chronic arthritis (17-23). PLA, is
also a major component of the salt-extractable pool of
matrix proteins in normal cartilagc (24). Purified o r
recombinant PLA, is known to cause a n inflammatory
response when injected into the synovial space (25). The
direct effects of PLA, on cartilage metabolism have not
been evaluated.
To examine t h e relationship between IL-1induced cartilage degradation and PLA,-related mechanisms, we evaluated the effects of manoalide, a potent
PLA, inhibitor, on recombinant human IL-la (rHulLla)-induced arthritis in rabbits. Because r H u I L - l a induces the influx of inflammatory cells and significant loss
of cartilagc proteoglycan, the effects of the compound
on both parameters was determined. In addition, we
examined the relationship between these parameters
and the generation of message for stromelysin, a metalloproteinase implicated in proteoglycan catabolism (for
review, see ref. 26). The results indicate that manoalide
has an inhibitory effect on the arthritic response t o IL-I.
These effects may b e mediated, in part, by PLA, and t h e
subscqucnt expression of stromelysin.
MATERIALS AND METHODS
Animals. Malc New Zealand white rabbits (3-4 kg)
from Hazclton Laboratories (Denvcr, PA) wcrc used in all
cxperiments. The rabbits wcrc given free access to food and
water, and wcrc housed in the Parke Davis Rescarch animal
facilities. All protocols were approved by the Parke Davis
Kesearch Animal Welfare Committee.
Induction of arthritis with rHuIL-la. Rabbits were
ancsthctizcd with a ketamine/xylazine/phosphatcbuffered saline (PBS) mixture (2 ml of ketamine to 1 ml each of xylazine
and PBS). Whilc under anesthesia, the rabbits (6 pcr cxpcrimental group, cxccpt whcre noted) were injected intraarticularly with 0.5 ml of PBS containing 25 ng of rHuIL-la
(Genzymc, Cambridge, MA) into the right stifle joint (knee).
The contralateral joint was injected with 0.5 ml of the PBS
vchiclc.
At various time points, thc animals wcrc killcd with an
intravenous injection of sodium pentobarbital. Sterile saline (1
1293
mi) was injected into each stifle joint with a 26-gauge, onehalf-inch needle. The knee was flexcd 10 times, and thc lavage
fluid was withdrawn with a 23-gaugc, thrce-quarter-inch needle. Thc volume of synovial fluid retrieved from the synovial
cavity was vcry consistcnt and stimulus-independent (mean -+
SEM 0.972 2 0.16 ml in rHuIL-la-treatcd and 0.969 2 0.14 ml
in PBS-treated rabbits). An aliquot of lavagc fluid was rcmoved, and the number of white blood cells was determined
using a cell counter (Coultcr Electronics, Hialeah, FL). Slides
were prcparcd using a cytocentrifuge, and differential counts
wcrc performed. The remaining lavage fluid was centrifuged
for 10 minutes. The supernatant was rcmovcd and frozcn at
-20°C until analyzed.
The fcmur was removed, and the cartilage was carefully disscctcd from the femoral condyle using a no. 11 scalpel
bladc. The cartilage sample was immediately weighed (wct)
and frozcn at -20°C until analyzed for proteoglycan contcnt.
Manoalidc (molecular weight 416.5; Calbiochem, La Jolla,
CA) was administered intraarticularly into both knees (0.1,0.3,
or 1 .O mg/kncc; in a 0.5-ml volume [OS ml of PBS containing
20% methanol]) of treated animals 30 minutes before the
rHuIL-la injection. Both knees of all animals not receiving
manoalide were given intraarticular injections of the dosing
vehicle (0.5 ml of PBS containing 20% methanol).
Phospholipase A2 assay. PLA, assays were performed
on synovial fluid as previously described (27). Briefly, samples
were incubated for 15 minutes at 37°C with Escherichia coli
suspension (6.3 mCi/rnmole of inorganic phosphate; New
England Nuclear, Wilrnington, DE), [ 1-I4C] arachidonate labeled (100,000 counts per minute containing -37% phosphatidylethanolaminc/phosphatidylglycerol and 45% cardiolipin)
in a total volumc of 200 pl of 250 mM Tris HCI, pH 9,
containing 10 mM Ca-' and 1 mgiml bovine serum albumin
(BSA). Substrate was also added to control tubes containing
PBS to monitor the nonenzymatic release of labclcd arachidonic acid.
The reaction was stopped by the addition of 100 ~1 of
20 mg/ml fatty acid-free BSA. After incubation at 4°C for 30
minutes, the samples wcrc centrifuged at l0,OOQg for 4 minutes, and 250 p1 of the supernatants containing the released
'"C-labclcd arachidonic acid was analyzed by scintillation
counting.
Proteoglycan assay. A dimcthylmcthylcnc blue assay
was used to analyze the proteoglycan content of the cartilagc
and synovial lavage fluid (28). Thc cartilage was thawed and
added to 960 pl of 0.1M PBS, pH 7.0, with EDTA ( 2 mM) and
1.-acctyl-cysteine (2 mM). Forty microliters of papain (25
pg/ml) was addcd, and the mixture was placed in a watcr bath
for 7 hours at 65°C. The lavage fluid was digcsted by adding 12
pI of the papain solution to 180 p1 of sample for 1 hour at
65°C. A dimethylene blue (DMB) solution was prepared by
dissolving 16 mg of DMB, 2 gm of sodium formatc, and 2.0 ml
of formic acid in 5 ml of ETOH. The solution was brought to
a final volume of 1 liter with distilled water. Thc DMB solution
was protected from light.
To perform the assay, each well of a 96-well platc
received 270 ~1 of dimethylmethylene blue dye and 30 pl of
sample, blank, or dilutions of the standard (chondroitin sulfate
C from shark cartilage), and absorbance was immcdiatcly
measurcd at 540 nm. A linear rcgrcssion analysis was pcrformed on thc absorbance values from thc various concentra-
SCHRIER ET AL
60
500
*
50
400 A
40
c
'Po
30
([I
0
300 2
0
X
0
-
Q)
v)
20
w
s
2
n
200
s
10
0
0
0
5
10
15
20
100
25
Time (hours)
Figure 1. EUects of recombinant human intcrlcukin-la (rHuIL-la) treatment o n phospholipase
A, (PLA,), cell accumulation, and proteoglycan content in synovial lavage fluid. Rabbits wcrc
injected intraarticularly in 1 stifle joint with rIIuIL-la (solid symbols) and the contralateral joint
with phosphate buffered saline (PBS) (open symbols). At various times after injection, rabbits we!%
killcd, joint cavities were lavagcd with sterile PBS, and cell counts, proteoglycan content, and PLA,
activity were determined. Valucs arc the mean and SEM. * = P < 0.05 by one-way analysis of
variance, with Ihnnctt's multiple comparisons test. Each rlIuIL-la-treated group was compared
with its respective vehicle-treated control; n = 6 for each experimental group. A4 = arachidonic acid.
tions of the chondroitin sulfate. The proteoglycan content of
the samples was calculated using the standard curve.
Stromelysin mRNA expression. Manoalide and
rHuIL-la were administered to anesthetized rabbits as described above. The synovium was harvested 6 hours after
infusion of rHulL-la, snap-frozen in liquid nitrogen, and
stored at -80°C. Kinetic data have shown that stromelysin
expression in synovium is maximal at 6 hours post-lL-1
infusion (data not shown).
Total RNA was extracted with TRIzol reagent (Gibco
BRL, Gaithersburg, MD) and analyzed by Northern blot.
Briefly, 5 mg of RNA was separated on a 1.0% agarose/
formaldchydc gel and transferred to Zeta-Probe G T nylon
membranes (Bio-Kad, Hercules, CA) by overnight capillary
transfer. The blots were baked for 2 hours and then prehybridized for 2 hours at 65°C in 10 ml of Church hybridization buffer
(7% sodium dodccyl sulfate [SDS], 500 mM phosphate, 1 mM
EUI'A, 1% BSA). The blots wcrc hybridized overnight at 65°C
with 10' counts per minutc/ml of full-length 72P-labclcd complementary RNA (cRNA) probes to rabbit strornelysin (added
to prehybridization buffer along with 100 ml of 10 mg/rnl
salmon sperm DNA), rinsed once for 10 minutes at room
temperature with 5 % SIX, 25 mM phosphate, 1 mM EIITA,
and 0.5% BSA, twice for 10 minutes at 65°C in the same bulfer,
and twice for 10 minutes at 65°C in 1% SDS, 25 mM phosphate, and 1 mM EDTA, then autoradiographed.
The blots were then stripped and reprobed with a
100-nucleotide cKNA probe to 18s ribosomal RNA (rRNA) to
control for equal loading (18s rRNA and complementary
DNA from Ambion, Austin, TX). Following autoradiography,
the stromelysin and 18s rRNA expression were quantified on
a scanning densitometcr, and the stromelysin data were normalized to the 18s rKNA values.
Statistical analysis. Statistical power was determined
by either a one-way or two-way analysis of variance, and a
Dunnett's test for multiple comparisons.
RESULTS
Recombinant human IL-la-induced synovial
fluid PLA, activity. Rabbits were injected intraarticularly with rHuIL-la in 1knee joint and the same volume
of sterile PBS in the contralateral knee. The animals
were killed at various time points and the knee joints
were lavaged with PBS. PLA, activity was significantly
elevated at 8 hours, reached maximal levels at 12 hours,
and remained elevated at 24 hours (Figurc 1). The kinetics
MANOALIDE EFFECTS ON IL-la-INDUCED ARTHRITIS
..-
.v
I
I
I
I
Vehicle Knee
IL-1 Knee
*
I
T
E0
--
Veh
various doses, and lavage fluid and cartilage proteoglycan measurements were made 24 hours after rHuIL-la
injection. At 0.3 mg/knee, manoalide significantly reduced the proteoglycan content of the lavage fluid, as
measured by a dimethylmethylene blue assay (Figure 3).
The inhibitory effects of the compound were evident at
8, 12, and 24 hours after manoalide treatment (Figure
4). A slight increase in proteoglycan fragment concentration was observed in the rabbits treated with manoalide in the absence of rHuIL-la. However, the effect
was transient, and returned to baseline by 24 hours.
Manoalide also reversed the rHuIL-la-induced
release of proteoglycan from cartilage removed from the
femoral condyles at 24 hours (Figure 5). A single time
point was used in these studies because significant losses
of cartilage proteoglycan in response to rHuIL-la were
not observed earlier in the model (data not shown).
Manoalide inhibited proteoglycan loss by 78% ( P <
0.3
Manoalide (mglknee)
400
h
Figure 2. Effects of manoalide on PLA, activity in the synovial fluid of
rabbits treated with rHuIL-la. Rabbits were injected intraarticularly in
1 stifle joint with rHuIL-la and the contralateral joint with PBS;
manoalide or vehicle (Veh) was administered intraarticularly 30
minutes before rHuIL-la injection. At 12 hours after injection, rabbits
werc killed, joint cavities were lavaged with sterile PBS, and PLA,
activity was determined. Values are the mean and SEM. * = P < 0.05
by two-way analysis of variance, with Dunnett’s multiple comparisons
test. Comparisons with data from rHuIL-la-treated joints of manoalide vehicle-treated rabbits are shown; n = 6 for each experimental
group. See Figure 1 for other definitions.
1295
0
1 350
LL
a,
-
Vehicle Knee
IL-1 Knee
-
I
0)
2
300
([I
-I
-I
E
\
T
2
Y
200
c
C
of this response was similar to the appearance of proteoglycan fragments and the influx of inflammatory cells
(>%I% ncutrophils) in the synovial fluid. In each instance,
maximal leveLs werc obtained by 12 hours and remained
elevated at 24 hours. Injection of vehicle also induced a
slight, but significant (P < 0.05), increase in PLA, activity
at 8 hours, which subsided by 12 hours. None of the other
parametcrs was affected by vehicle treatment.
To ascertain whether the PLA, inhibitor, manoalide, blocked PLA, activity in the lavage fluid, rabbits
were predosed intraarticularly with the compound (0.3
mgiknee), and PLA, activity was measured 12 hours
after rHuIL-la injection (Figure 2). Manoalide reduced
synovial PLA, activity by 74% ( P < O.OOl), to baseline
levels.
Effects of the PLA, inhibitor manoalide on
rHuIL-la-induced cartilage catabolism. To assess the
relationship between PLA, inhibition and cartilage proteoglycan degradation, manoalide was administered at
s
150
C
la
0
x
a, 100
0
al
c
$
1
*
250
0)
Ea,
!
0.1
50
Veh
0.3
Manoalide (mg/knee)
Figure 3. Dose-response showing the effect of manoalide on lavage
fluid proteoglycan content of rabbits treated with rHuIL-la. Rabbits
were injected intraarticularly in 1 stifle joint with rHulL-la and the
contralateral joint with PBS; various doses of manoalide or the
rnanoalide vehicle (Veh) were administered intraarticularly 30 minutes
before rHuIL-la injection. At 24 hours after injection, rabbits were
killed, joint cavities were lavaged with sterile PBS, and proteoglycan
content was determined. Values are the mean and SEM. * = P < 0.05
by two-way analysis of variance, with Dunnett’s multiple comparisons
test. Comparisons with data from rHull>-la-treated joints of manoalidc vchicle-treated rabbits are shown; n = 6 for each experimental
group. See Figure 1 for other definitions.
SCHRIER ET AL
1296
0.05) at 0.3 mg/knec. The 0.1-mg dose was ineffective.
Thc effect of 1 mg of manoalide on rHuIL-la-induced
cartilagc protcoglycan was not studied because compound treatment alone caused significant ccllular influx
and proteoglycan release from cartilage (data not
shown). Thcsc rcsults indicate that manoalide significantly reduces cartilage protcoglycan loss induced by
rHuIL-la at 0.3 mg/knee.
Effects of the PLA, inhibitor manoalide on
rHuIL-10-induced synovial cell count. Intraarticular
injection of rHuIL-1 a induced a significant influx of
inflammatory cells into the articular space (Figure 6).
Neutrophils wcrc the predominant cell type (>90%) in
the exudate. At 0.3 mg/knce, manoalide slightly increased the neutrophil count in vehicle-treated knees.
However, the compound had no effect on rHuIL-lainduced neutrophil accumulation at any time point
tested. These results suggest that the beneficial effects of
-
0Vehicle Knee
IL-1 Knee *
*
-
*
I
1-
L
-
i
0.3
Manoalide (mglknee)
Figure 5. Dose-response showing the cffcct of manoalide on cartilage
proteoglycan content of rabbits treated with rHuIL-la. Rabbits were
injected intraarticularly in 1 stifle joint with rHuIL-la and the contralateral joint with PBS;various doses of manoalide or the manoalide
vehicle (Veh) wcrc administered intraarticularly 30 minutes before
rHuIL-la injection. At 24 hours after injection, rabbits were killed,
cartilage was removed from thc femoral condyles, and proteoglycan
content was dctcrmincd. Values are the mean and SEM. * = P < 0.05
by two-way analysis of variance, with Dunnett’s multiplc comparisons
test. Comparisons with data from rHuIL-1 a-trcatcd joints of manoalide vehicle-treated rabbits arc shown; n = 6 for each experimental
group. See Figure 1 for othcr definitions.
600
A
7-
500
a
IL-1
IL-1 + Manoalide
Manoalide
U
a,
0)
2m
400
-I
-1
€
\
C
m
0
x
-
300
m
0
2!
0
200
a‘
[5)
3-
100
r0
4
i
I
a
12
~
16
7
20
-
T
24
Time (Hours)
Figure 4. Timc-rcsponse showing the effccts of manoalide on lavage
fluid protcoglycan content of rabbits treated with rHuI1,-la. Rabbits
were injected intraarticularly in 1 stifle joint with rHulL-la and the
contralateral joint with PBS; manoalide or vehicle was administered
intraarticularly 30 minutes before rHulL-1 a injection. At various time
points after injcction, rabbits were killcd, joint cavities wcrc lavaged
with sterile PBS, and proteoglycan content was determined. Values arc
thc mean and SEM. * = P < 0.05 by two-way analysis of variancc, with
Dunnctt’s multiple comparisons test. Comparisons with data from
rHu11,-1 a-treated joints of manoalide vehicle-trcatcd rabbits are
shown; n - 6 for cach cxpcrimental group. SCCFigurc 1 for definitions.
manoalide on cartilage mctabolism are not related lo its
effects on cell counts.
Inhibition of stromelysin mRNA by manoalide.
The levels of mRNA for the metalloprotcinasc stromelysin were measured by Northern blot analysis in synovial tissue from rHuIL-1 a-treated and manoalidetreated animals. Previous studies have shown that
increased levels of mcssage are present as early as 1hour
after intraarticular injection of rHuIL-la, with the levels
peaking 6 hours after infusion, and returning to bascline
by 24 hours (data not shown). At 0.3 mg/kncc, manoalide caused a significant reduction (I’ < 0.05) in stromelysin levels in synovium harvested 6 hours after
rHuTL-la infusion (Figurc 7). The reduction observed
with the 0.1-mgknee dose was not statistically significant. Although a significant amount of underlying tissuc
was harvested along with the synovium, in situ hybridization studies have confirmcd that the synovial lining is
the source of the mcssage analyzed by Northern blot.
1297
MANOALIDE EFFECTS ON IL-la-INDUCED ARTHRITIS
*
IL-I
IL-1 + Manoalide
A
vation of the enzyme with an alkylating agent or with
heat blocks the inflammatory reaction (25). It is unclear
whether other isoforms are directly proinflammatory,
because most of them arc cytosolic. The results from this
study indicate that a correlation exists between PLA,
activity and arthritis in rabbits injected with rHulL-la. A
temporal relationship exists between the appearance of
PLA, activity, proteoglycan fragments, and cellular infiltration in the synovial fluid. A causal relationship is
suggested by the observation that manoalide, a PLA,
inhibitor, blocks connective tissue metabolism, as assessed by measuring the proteoglycan content in the
synovial fluid and cartilage.
The demonstration that PLA, is involved in d
0
0
4
a
12
16
20
24
Time (Hours)
Figure 6. Time-response showing the effects of manoalide on lavage
fluid cell count of rabbits treated with rHuIL-la. Rabbits were injected
intraarticularly in 1 stifle joint with rHuIL-la and the contralateral
joint with vehicle; manoalide or vehicle was administered intraarticularly 30 minutes before rHulL-la injection. At various time points
after injection, rabbits were killed, joint cavities were lavagcd with
sterile PBS, and cell counts were performed. Values are the mean and
SEM. * = P < 0.05 by two-way analysis of variance, with Dunnett's
multiple comparisons test. Comparisons with cell counts from PBStreatcd joints of manoalide vehiclc-trcatcd rabbits are shown; 11 = 6
for cach experimental group. SCCFigure 1 for definitions.
5
-
0Vehicle Knee
T
2
2000
2
IL-I Knee
7
0
1500
7
Q)
.N
m
5
.-E
I000
UJ
C
0"
500
s
.-
-RI
c
Because stromelysin is proposed to be a key enzyme
involved in proteoglycan degradation, these results suggest a potential mechanism by which manoalide inhibits
cartilage damage.
2
0
Veh
0.1
0.3
Manoalide (mglknee)
DISCUSSION
PLA, is implicated in a variety of inflammatory
diseases including arthritis, inflammatory bowel disease,
reperfusion damage, septic shock, and arthritis (for
review, see ref. 29). In RA,elevated levels of PLA, arc
observed in serum, synovial fluid, and articular cartilage
(17-23). The cartilage is a potentially important source
of the 14-kd isoform and has been purified, cloned, and
expressed. The proinflammatory characteristics of PLA,
have been demonstrated in a number of models. In
rabbits, the subcutaneous injection of PLA, induces an
inflammatory response (30). Proliferation of the synovial
lining, a hallmark of RA, is also observed after intraarticular injection of human recombinant PLA,. Inacti-
- + - +- +
Control 1
0.1
-+-+-+
1
0.3
rlL-la
Manoalide (mglKnee)
Figure 7. Effects of manoalide on stromelysin expression in the
synovium of rabbits treated with rIIuIL-la. Six hours after rHuIL-la
infusion, rabbits were killed, synovium was removed, snap-frozen in
liquid nitrogen, and stored at -80"C, until processing for Northern
blot analysis. Manoalide or vehicle was administered intraarticularly 30
minutes before rIIuIL-la injection. Values arc the mean and SEM
(n =: 2 per experimental group) of a representative experiment. * =
P < 0.05 by two-way analysis of variance, with Dunnett's multiple
comparisons test. Comparisons with data from rHuIL-1 a-treated
joints of manoalide vehicle-trcatcd rabbits are shown. See Figure 1 for
dcfinitions.
1298
variety of cellular responses apparently unrelated to a
direct effect on arachidonic acid metabolism emphasizes
its potential involvement in signal transduction. In addition to its role as a source of proinflammatory mediators,
PLA, is involved in a variety of other cellular functions,
including degradation (31), phagocytosis (32), adhesion and adhesion molecule expression (33), exocytosis,
and cell spreading (34,35). In several instances, the
observed effects are not attributed solely to direct regulation by metabolites of arachidonic acid (31,36). The
present studies suggest that the inhibitory effects of
marloalide on proteoglycan degradation are not due to
an indirect effect on inflammatory cells, since the compound has no effect on cellular migration. Moreover,
previous studies have demonstrated that inhibition of
neutrophil accumulation has no effect on the release of
proteoglycan fragments into the synovial fluid (37).
A potential source of proteoglycan catabolic activity is the metalloproteinase stromelysin, which is
released by synovial fibroblasts. Indeed, a relationship
between PLA, and metalloproteinase release by synovial
fibroblasts (38) and matrix catabolism by chondrocytes
(39) has been reported. Increased levels of stromelysin
message and protein are correlated with the appearance
of proteoglycan fragments in the IL-1 arthritis model
(37,401. The present studies examined the effects of
manoalide on the levels of stromelysin mRNA in the
synovihm of rHuIL-lu-treated rabbits. Thc results indicate that manoalide blocks expression of the enzyme at
a dose that also blocks proteoglycan release, which
suggests a potential cxplanation for the observed effects
of the compound on connective tissue metabolism.
Further studies are planned to explore the potential link
between PLA, activity and strornelysin expression, including the use of blocking antibodies.
While it is conceivable that the inhibitory effects
of manoalide might be attributable to an alternative
biochemical mechanism, manoalide appeared to have
similar effects on proteoglycan degradation and PLA,
activity at an equivalent dose, suggesting a common
mechanism. Moreover, many in vitro studies have shown
that manoalide is an effective inhibitor of PLA, in a
variety of systems, including those involving synovial
enzyme (41,42). It is probable that cartilage-derived
PLA, and stromelysin are also important for proteoglycan degradation, and future studies will address this
issue.
This is among the first reported studies to evaluate a pharmacologic agent in the rabbit IL-1-induced
arthritis model. Although it is routine to administer
antiinflammatory compounds by the oral route, the
SCHRIER ET AL
intraarticular route remains somewhat unconventional.
Because manoalide was administered intraarticularly, it
is more likely that the compound might have nonspecific
effects on the joint tissue. For instance, intraarticular
injection of steroids in humans causes a transient arthropathy (43). At a dose of 0.3 mgljoint, manoalide caused
a slight increase in the synovial fluid neutrophil count.
This cffcct was minimal compared with the change
induced by rHuIL-la, and no augmentation of the cell
count was observed in the animals that received both
treatments. At the highest dose of manoalide (1 mg/
knee), an increase in the release of proteoglycan fragments and a drop in cartilage proteoglycan contents
were observcd. No evidence of direct cartilage degradation was evident at the lower doses of manoalide. These
results suggest that a balance exists between the protective and potentially injurious effects of a compound
administered by the intraarticular route.
In conclusion, these results indicate that PLA,
activity is elevated in the synovial fluid of rabbits treated
intraarticularly with rHuIL-1 a. This activity correlates
with several indicators of the arthritis response. Treatment with the PLA, inhibitor manoalide blocks PLA,
activity in the synovial fluid and prevents the degradation of proteoglycan induced by rHuIL-la. The latter
effect may be partly explained by an inhibitory effect on
the expression of mRNA for stromelysin. These results
suggest that PLA, may play a broader role in the
pathogenesis of rHuIL-la-induced arthritis than has
previously been appreciated.
ACKNOWLEDGMENTS
T h e authors would like to acknowledge the technical
assistance of Godwin Okonkwo and administrative assistance
of Doreen Gasparella.
REFERENCES
Natsuyama M, Inoue K,
1. Miyasaka N, Sato K, Goto M, Sasano IM,
Nishioka K Augmented interleukin-1 production and IILA-DR
expression in thc synovium o f rheumatoid arthritis patients: possiblc involvement in joint destruction. Arthritis Rheum 31:480486, 1988
2. Bcndtzen K, Petcrsen J, I Ialkjaer-Kristcnscn J, Ingemann-Hanscn
T Intcrlcukin-1-like activities in synovial fluids of patients with
rhcumatoid arthritis and traumatic arthritis. Rheumatol Int 5:79 82, 1985
3. Nouri AME, Panayi Cis, Goodman SM: Cytokincs and the chronic
inflammation of inflammatory disease. I. The presence of interleukin 1 in synovial fluids. CXn Exp Immunol 55:295-302, 1984
4. Wood DD, Ihrie EJ, Dinarcllo CA, Cohen PL: Isolation of an
interleukin-1-like factor from human joint effusions. Arthritis
Rheum 26975-983, 1983
5. MacNaul KL, Hutchinson NI, Parsons JN, Rayne EK, Tocci MJ:
Analysis of IL-1 and TNF-a gene expression in human rheumatoid
MANOALIDE EFFECTS ON IL-la-INDUCED ARTHRITIS
synoviocytes and normal monocytcs by in situ hybridization.
J Immunol 145:4154-4166, 1990
6 MacNaul KL, Chartrain N, Lark M, Tocci MJ, Hutchinson NI:
Discoordinatc expression of fibroblasts: synergistic elfects of
interleukin-1 and tumor necrosis factor-n on stromclysin expression. J Hiol Chcm 26917238-17245. 1990
7 Kandel RA, Petclycky M, Dinarello <:A, Minden M, Ptitzker KPH,
Cruz TF: Comparison of the eflect of interleukin 6 and interleukin
1 on collagenase and proteoglycan production by chondrocytes. J
Rheumatol 17:953-957, 1990
8 Lefebre V, Peelers-Joris C. Vaes G: Modulation by interleukin 1
and tumor necrosis factor LY of production of collagenase, tissue
inhibitor of metalloproteinases and collagen types if diffcrcntiated
and dedirerentiated articular chondrocytes. Hiochim Biophys Acta
1052:366-378, 1990
9. Arncr EC, Pratta MA: Independent effects of interleukin-1 on
proteoglycan breakdown, proteoglycan synthesis, and prostaglandin El release from cartilage in organ culturc. Arthritis Rheum
32:288-297, 1989
10 Dingle JT: The effect of synovial catabolin on cartilage synthetic
activity. Connect Tissue Res 12:277-286, 1984
I 1 Arner EC, DiMeo TM, Rhuhl DM, Pratta MA: In vivo studies on
the effects of human rccombinant interleukin-lp on articular
cartilage. Agents Actions 27:254-257, 1989
12 Pcttipher ER, IIiggs GA, llenderson B: Interleukin 1 induces
leukocyte infiltration and cartilage proteoglycan degradation in
the synovial joint. Proc Natl Acad Sci U S A 83:8749-8753, 1986
13. Pcttipher ER, Henderson B, Hardingham T, Ratcliffe A: Cartilage
proteoglycan depletion in acute and chronic antigen-induced
arthritis. Arthritis Rheum 32:601-607, I989
14. Angel J, Colard 0, Chevy F, Fournier C: Interleukin-1-mediated
phospholipid breakdown and arachidonic acid release in human
synovial cells. Arthritis Rheum 36:158-167, 1993
15. Hulkowcr KI, Wertheimer SJ, Levin W, Coffey JW, Anderson CM,
Chcn T, DeWitt DL, Crow1 RM, Hope WC, Morgan D W
Interleukin-lp induces cytosolic phospholipasc A, and prostaglandin H synthase in rheumatoid synovial fibroblasts: evidence for
their role in the production of prostaglandin E,. Arthritis Rheum
37:653-661, 1994
16. Lyons-Giordano B, Galbraith W, Pratta MA, Arner EC: Interleukin Ip stimulates phospholipase A, mRNA synthesis in rabbit
articular chondrocytes. Biochem Biophys Res Commun 164:488495, 1989
17. Sakamoto K, Arakawa H, Yamashita S, Mizuta H, Takkaki K,
Ogawa M: Membrane-associatcd phospholipase A, detected by a
radioimmunoassay is a sensitive marker of inflammation in rheumatoid arthritis. Rcs Commun Mol Pathol Pharmacol76:279-285,
1992
18. Kortekangas P, Aro HT, Ncvalaincn TJ: Group 11 phospholipase
A, in synovial fluid and serum in acute arthritis. Scand J Rheurnatol 23:68-72, 1994
19. Pruzanski W, Bogoch E, Wloch M, Vadas P: The role of phospholipasc A, in the pathophysiology of osteoarthritis. J Rheumatol
Suppl -37:117-119, 1991
20. Pruzanski W, Bogoch E, Stefanski E, Wloch M, Vadas P: Enzymatic activity and distribution of phospholipase A, in human
cartilage. Life Sci 482457-2462, 1991
21. Scilhamcr JJ, Plant S, PruLanski W, Schilling J. Stefanski E, Vadas
P, Johnson LK: Multiple forms of phospholipase A3 in arthritic
synovial fluid. J Biochem 106:38-42, 1989
22. Lai C-Y, Wada K: Phospholipasc A, from human synovial fluid:
purification and structural homology to the placental engnie.
Hiochcm Biophys Res Commun 157:488-493, 1988
23. Vadas P, Pruzanski W Role of extracellular phospholipase A, in
inflammation. Adv Inflamm Rcs 752-59: 1984
24. Recklies AD, White C: Phospholipase A, is a major component of
the salt-extractable pool of matrix proteins in adult human articular cartilage. Arthritis Rheum 34:1106-1115, 1991
1299
25. Bomalaski JS, Iawton I’, Browning JL: Iluman extracellular
rccombinant phospholipase A, induces an inflammatory response
in rabbit joints. J Immunol 146:3904-3910, 1991
26. Wocssncr JF: Matrix metalloproteinases and their inhibitors in
connective tissue remodeling. FASEB J 5:2145-2154, 1991
27. Kramer RM, IIession C, Johanson B, Hayes G , McGray P,
Pingchang-Chow E, Tizard R, Pcpinsky RB: Structure and properties of a human non-pancreatic phospholipasc A2 J Bid Chcm
3945768-5778, 1989
28. Baragi VM, Jordan H, Renkicwicz RR: A protocol for rapid
screening of proteoglycan-degrading metalloproteinasc inhibitors.
J Pharmacol Toxicol Methods 27:101-105, 1992
29. Kaiser E, Chiba P, Zaky K Phospholipases in biology and
medicine. Clin Biochem 23:349-370, 1990
30. Pruzanski W, Vadas P, Fourasier V: Inflammatory effect of
intrademal administration of soluble phospholipase A, in rabbits.
J Invest Ilcrmatol 86:380-383, 1986
31. Smolcn JE. Weissmann G : ElTects of indomethacin. 5.8.11.14,
,
cicosatctraynoic acid and p-bromphenacylbromide on lysosomal
enzyme release and superoxide anion gcncration by human polymorphonuclear leukocytes. Biochem Pharmacol 29533-538, 1980
32. Schwacha MG, Gudcwicz PW, Snyder JA, Loegering DJ: Depression of macrophagc respiratory burst capacity and arachidonic acid
release after Fc receptor-mediated phagocytosis. J Immunol 150:
236-245, 1993
33 Jacobson PB, Schrier SJ: Regulation of CDI Ib/CD18 expression
in human neutrophils by phospholipase A,. J Immunol 15156395652, 1993
34. Meers PK, Hong K, Papahadjopoulos D: Free fatty acid enhanccment of cation-induced fusion of liposomes: synergism with synexin and other promoters of vesicle aggregation. Biochemistry
27~6784-6794, 1988
35. Creutz CE:Cis-unsaturated fatty acids induce the fusion of chromaffin granules aggregated by synexin. J Cell Biol 91:247-256, 1981
36. Perretti M, Flower RJ: Modulation of IL-1-induced neutrophil
migration by dexamethasone and lipocortin 1. J Immunol150:992999, 1993
37. McDonnell J, Hoerrner LA, Lark MW, Harper C, Dey T, h b n e r
J, Eiermann G, Kazazis D, Singer 11, Moore V L Recombinant
human interleukin-lkinduced increase in levels of proteoglycans,
stromelysin, and leukocytes in rabbit synovial fluid. Arthritis
Rheum 35799-805, 1992
38. Mehindate K, Al-Daccak R, Daycr J-M, Kennedy BP, Kris C,
Borgeat P, Poubelle PE, Mourad W Superantigen-induced collagenasc gene-expression in human IFN-y-treated fibroblast-like
synoviocytes involves prostaglandin E,: evidence for a role of
cyclooxygcnasc-2 and cytosolic phospholipase A,. J Immunol
155~3570-3577, 1995
39. Richard M, Vignon MJ, Peschard MJ, Broquet P, Carret JP,
Louiso P: Seritonin-stimulated phospholipasc A2 and collagenase
activation in chondrocytes from human osteoarthritis articular
cartilage. FERS I.ett 278:38-40, 1991
40. Hutchinson N1,Lark MW, MacNaul KI., Harpcr C, Hocrrncr LA,
McDonnell J, Donatelli S, Moore V, Bayne EK: In vivo expression
of stromclysin in synovium and cartilagc of rabbits injected
intraarticularly with interleukin-lp. Arthritis Rhcum 3512271233, 1992
41. Jacobson PH, Marshall LA, Sung A, Jacobs KS: Inactivation of
human synovial fluid phospholipase A2 by the marine natural
product, manoalide. Hiochem Pharmacol 39:1557-1564, 1990
42. Glascr KH, Jacobs KS: Molecular pharmacology of manoalide:
inactivation of bee venom phospholipase A,. Biochem Pharmacol
35:449-453. 1986
43. Gray RG, Gottlieb N L Intra-articular corticosteroids: an updated
assessment. Clin Orthop 177:235-263, 1983
Документ
Категория
Без категории
Просмотров
3
Размер файла
767 Кб
Теги
expressions, stromelysin, degradation, induced, recombinant, count, human, fluid, cells, intraarticular, effect, injections, inhibitors, rabbits, phospholipase, interleukin, synovial, cartilage, manoalide
1/--страниц
Пожаловаться на содержимое документа