вход по аккаунту


Release of Collagenase Neutral Protease and Prostaglandins From Cultured Mammalian Synovial Cells by Hydroxyapatite and Calcium Pyrophosphate Dihydrate Crystals.

код для вставкиСкачать
Hydroxyapatite (HA) and calcium pyrophosphate dihydrate (CPPD) crystals were phagocytosed
when added to cultured human rheumatoid or normal
canine synovial cells. Collagenase and neutral protease
secretion into the culture medium was increased 5- to 8fold over control values in the presence of HA and
increased 3-fold in the presence of CPPD crystals. HA
but not CPPD crystals induced a 300-fold increase in
human rheumatoid synovial cell culture fluid prostaglandin (PG) E2 levels and an 8-fold increase in PGFa
levels. This mechanism may be important in the pathogenesis of the destructive arthropathies associated with
HA and CPPD crystals.
Recently we described a syndrome affecting the
shoulder joints of 4 patients (“Milwaukee shoulder
syndrome”). Bilateral rotator cuff defects, glenohumeral osteoarthritis, and joint stiffness or instability
were accompanied in each case by the finding of
hydroxyapatite (HA) crystal clumps (microspheroids),
activated collagenase, neutral protease, and particulate collagen types I, 11, and 111 in a nearly acellular
From the Rheumatology Section. Department of Medicine,
Medical College of Wisconsin, Milwaukee, WI 53226.
Supported by USPHS Grants AM26062, AM05621. and
AM18074 and by agrant from the Wisconsin Chapter of the Arthritis
Herman S. Cheung, PhD: Assistant Profesor of Medicine,
Biochemistry, and Orthopedic Surgery; Paul B . Halverson, MD:
Assistant Professor of Medicine; Daniel J. McCarty, MD: Professor
and Chairman, Department of Medicine.
Address reprint requests to Herman S. Cheung. PhD,
Department of Medicine, Medical College of Wisconsin. Milwaukee
County General Hospital, 8700 West Wisconsin Avenue, Milwaukee, WI 53226.
Submitted for publication March 6, 1981; accepted in
revised form May 27, I98 1 .
Arthritis and Rheumatism, Vol. 24, No. 11 (November 1981)
synovial fluid (1-4). Synovium obtained from 1 joint
showed typical chondromatosis (4). Electron microscopy showed crystal clumps shedding into the joint
space through areas denuded of synovial cells. Some
crystal masses were found within synoviocytes (4).
In 1975 Werb and Reynolds reported that endocytosis of latex particles by cultured rabbit synovial
cells stimulated the secretion of collagenase and neutral protease into the surrounding medium ( 5 ) . We
postulated that the HA crystals shed from synovial
depots into the joint fluid might be phagocytosed and
stimulate synovial cell secretion of the observed enzymes. Enzymatic tissue digestion might then release
more crystals, completing a pathogenetic cycle.
Endocytosis of both HA and calcium pyrophosphate dihydrate (CPPD) crystals by cultured human or
canine synoviocytes is associated with the release of
collagenase, neutral protease, and prostaglandins
(PGs) into the surrounding medium.
Tissue culture. Two human rheumatoid synovial cell
strains were established using the technique of Biwas and
Dayer (6). Confluent secondary cell cultures were used in
these experiments. Cells were grown in 1Wmm plastic
Falconware in 10 ml of Dulbeccos modified Eagle’s medium
(DMEM), supplemented with 10% fetal calf serum and 1%
penicillin and streptomycin (all from Gibco, Grand Island,
NY). The number of cells per plate varied from 1.2 to 1.5 x
Normal canine synovial cell cultures were established using a modification of the method described by
Clarris et al (7). Confluent secondary cell cultures from 4
dogs were used experimentally. These cells were grown as
outlined above except that 10% horse serum was used. The
number of cells per plate varied from 1.7 to 2 x lo6.
Cells were harvested using 0.25% trypsin at the end
of each experiment and were counted by hemocytometer.
Crystal preparation and characterization. Clumps of
crystals (Sigma County, St. Louis, MO.) were sieved to
<20p in greatest diameter (Endicott test sieves, London,
England). HA crystals showed a remarkable tendency to
clump-the individual crystals were < Ip. These crystals
showed only the interplanar spacings of hydroxyapatite by
x-ray diffraction. Their molar calcium/phosphorus (CdP)
was 1.6 by x-ray energy dispersive analysis. Triclinic CPPD
crystals were synthesized, as recorded earlier (8). These
crystals showed only the interplanar spacings of CPPD by xray diffraction and a molar C d P of 1.0 by both direct
chemical analysis and x-ray energy dispersive analysis. They
were crushed and sieved to <50p in greatest diameter before
Enzyme analyses. Collagenase activity was determined using 'H glycine and 'H proline, labeled guinea pig
skin collagen, an adaptation of the method of Evanson,
Jeffrey, and Krane (9). One unit of collagenase activity was
equal to solubilization of 1 pg of reconstituted collagen
fibrils/minute at 37°C. One hundred microliters of tissue
culture fluid were incubated with the collagen gels for 6 to 12
hours. All assays were performed in duplicate and included a
control gel incubated with trypsin (Sigma Co., St. Louis,
MO) 100 pg/ml (w/v). Results were discarded if release of
radioactivity by trypsin was 15% above control gels incubated with buffer alone. A trypsin activation standard curve was
determined using I . 10, 20, and 200 pg trypsin/ml. These
samples were incubated at room temperature for 30 minutes
and the trypsin was then inactivated by addition of 4-fold
excess of soybean trypsin inhibitor (Sigma Co., St. Louis,
MO). Trypsin activation collagenase was performed as outlined by Dayer et al(10). Collagenase reaction products were
examined by SDS polyacrylamide gel electrophoresis ( I I ) .
Neutral protease activity was estimated in duplicate
in each conditioned medium as outlined previously (3),
except that '"1 human gamma globulin (specific activity =
1.5 x 10' counts per minute/mg) was substituted for "I
bovine serum albumin. One unit of activity was equal to I pg
of protein digested per minute at 37°C.
Assay for lactic dehydrogenase was performed as
described by others (12,13). Total enzyme contents of cells
were determined by addition of Triton X-100. 0.2% final
concentration and sonication. Kesults were expressed as the
mean 2 the standard error of the mean (SEM) of the
percentage of total enzyme released.
Prostaglandins (PG) E2 and F2u were measured by
radioimmunoassay as outlined by Dunn et al ( 14). From 0.5
to 3 ml of conditioned medium were extracted for each
determination to give the desired sensitivity in the PG
assays. The largest values required dilution of the extracted
samples to yield results that fell on the standard curve.
Experiments. Crystals were sterilized by heating
(180°C x 2 hours) and added to DMEM with fetal calf or
horse serum, depending on the species of cells used. HA or
CPPD crystals were used in final concentrations of 2 and 20
mg/lO ml for the experiments with human cells; 0.5 and 1 mg
HA or CPPD crystals/lO ml were used in the canine cell
experiments. After 24 hours the medium was replaced by
0.2% lactoalbumin in DMEM, as outlined by Werb et al (15).
Medium was then replaced daily for 7 days. Collagenase
activity was determined in each conditioned medium after
dialysis against 50 mM Tris HCI buffer, pH 7.5, containing
200 mM sodium chloride and 3 mM calcium chloride. Lactic
dehydrogenase was estimated in conditioned medium from
the canine cell line and expressed as a percent of total
intracellular lactic dehydrogenase found a1 the end of the
experiment, i.e., 7 days after crystals had been added.
Prostaglandins were measured in an aliquot of fluid from
human cell line DE obtained 4 to 5 days after crystals had
been added. At the end of the initial 24-hour incubation with
crystals, all cell cultures were examined by phase contrast
microscopy. Scanning and transmission electron microscopy
were performed at this time and again at the end of experiment, i.e., 7 days after the crystals were added. The cultures
were fixed in situ in Karnovsky's solution. The botlorn of the
plastic culture dish was scored and broken. Scanning electron microscopy was performed as outlined (3). The tissues
were postfixed in 1%) osmium tetroxide and embedded in
Spurr's low-viscosity epoxy resin for transmission electron
HA a n d CPPD crystals were each phagocytosed by both rheumatoid human and normal canine
synovial cells as judged by phase contrast microscopy,
scanning electron microscopy (Figures 1 a n d 2), and
transmission electron microscopy (Figure 3).
Ten micrograms of trypsin per milliliter were
the optimal concentration for activation of latent collagenase. Concentration of 200 &ml decreased collagenase activity by about 30% (data not shown). All
subsequent collagenase activation w a s done with 10
I*.g t r y p s i d m l before assay.
Both types of crystals increased collagenase
a n d neutral protease release from human synovial cells
as compared to control cells incubated without crystals. The data from cell line DE a r e shown (Figure 4).
Almost identical results were found when the second
human cell line (JD) w a s used. H A crystals induced a
5-fold increase in collagenase and an 8-fold increase in
neutral protease. There w a s relatively little difference
in the effectiveness of the 2-mg and 20-mg crystal
doses (Figure 4A).
CPPI) crystals induced similar increases in enz y m e secretion. Collagenase and neutral protease secretion increased approximately 3-fold over control
values (Figure 4B). Again, there was very little difference between the effects of the 2 doses ( 2 and 20 mg)
There w a s a striking difference in the prostaglandin levels observed after addition of HA o r CPPD
crystals t o t h e cultured human cell line DE (Table 1).
HA crystals induced a marked increase in both PGEP
Figure 1. Cultured canine cells exposed to crystals 7 days earlier. Scanning electron micrograph showing a
tnclinic calcium pyrophosphate dihydrate crystal within a synoviocyte ( ~ 7 5 0 ) .
Figure 2. Scanning electron micrograph showing several masses of hydroxyapatite crystals within a synoviocyte
Cell exposed to crystals 7 days earlier ( x 1,400).
Figure 3. Transmission electron micrograph of a synoviocyte which had phagocytosed a large m a s of hydroxyapatite crystals. Phagolysosomal membrane can be seen but these are breaks in the membrane (arrows). Cells
exposed to crystals 2 hours earlier ( ~ 2 6 , 6 0 0 ) .
and PGF2a levels in the extracellular medium. PGE2
rose approximately 300-fold and PGF2a, 8-fold over
control values. As noted with the enzyme levels, there
was no difference between 2-mg and 20-mg doses,
suggesting that even the lower dose had induced
maximal stimulation. CPPD crystals induced almost
negligible prostaglandin release. PGE2 levels were not
greater than control values in culture fluid from cells
given the 2-mg dose, but were increased 12-fold in
fluid from a culture given 20 rng of CPPD. Neither
dose induced a rise in PGF2a.
Addition of HA and CPPD crystals to monolayer cultures of normal canine cells also augmented
secretion of both enzymes; 0.5-mg and 1.0-mg HA
crystals induced approximately a 6-fold increase in
collagenase and a 4-fold increase in neutral protease
(Figure 5). CPPD crystals in the same doses increased
secretion of both collagenase and neutral protease
approximately 3-fold (Figure 6). The effects of the two
crystal doses were similar, suggesting maximal stiniulation of the cells even by the lower dose. Cell lysis
could not account for enzyme release, since levels of
lactic dehydrogenase in conditioned medium were
nearly identical to control values. About 8% of the
total lactic dehydrogenase had been released into the
medium by the seventh day (Figure 7). Similar lactic
dehydrogenase data were obtained with CPPD crystals
(data not shown).
These data support the hypothesis outlined in
the introduction with respect to the pathogenesis of
the destructive arthropathy in the “Milwaukee shoulder syndrome.” HA and CPPD crystals were each
phagocytosed by both human rheumatoid and normal
canine synovial cells. Crystal ingestion was accompanied by augmentation of secretion of collagenase and
neutral protease. Such secretion is a nonspecific sequel of endocytosis of particulate matter by synovial
cells ( 5 ) .
The basal level of enzyme secretion by our
cultures of human rheumatoid synovial cells was within the range of values reported by Biwas and Dayer
(6). Similarly, the basal level of enzyme secretion by
cultured canine cells was very close to that found by
TIME (Doys)
TIME (Days)
Figure 4. A , The cumulative release of collagenase and neutral protease from cultured human rheumatoid synovial cells is shown; A := control
= 2 mg hydroxyapatitc. See text for details. B. Parallel experiments using triclinic calcium
(no crystals); 0 = 20 mg hydroxyapatite:
pyrophosphatc dihydrate crystals; 0 = 20 mg; = 2 mg; control is shown in Figure 4A. See text for details.
Werb and Reynolds for lapine synovial cells ( 5 ) . The
increment in enzyme release induced by latex particles
(5,16) by soluble collagens (6) and by crystals are
similar-all cause a 3- to 8-fold increase.
There was little difference in total enzyme secretion when amounts of HA or CPPD of 2 or 20 mg
were used (Figure 4A and B). HA crystals appeared to
be more potent than CPPD as inducers of enzyme
release during the I-week period of incubation. However, Werb and Reynolds showed that enzyme secretion continued for at least 2 weeks after phagocytosis
of nonbiodegradable latex particles but not after ingestion of biodegradable particles such as dextran sulfate
( 5 ) . The CPPD crystals are much larger, so that both
the surface area and the number of particles per
milligram are orders of magnitude smaller.
A dose-response curve for each particle is needed and will require the use of uniformly isotopically
labeled crystals and a technique to allow removal of
unphagocytosed crystals. The conditions used in these
experiments do not satisfy these requirements. The
clearance of synthetic triclinic CPPD crystals from
joints has been shown to depend on synovial cell
endocytosis (17). Clearance of one-half of a CPPD
Table 1. Prostaglandin levels in tissue culture fluids*
2 me
20 me
2 me
20 me
PGE2 (pglml)
PGFa ( d m l )
* Values in
10 ml of conditioned medium obtained 1 to 5 days after
crystals added to human rheumatoid synovial cells.
t Calcium pyrophosphate dihydrate crystals.
TIME (Days)
Figure 5. The cumulative release of collagenase and neutral protease from cultured normal canine synovial cells, A = control (no
crystals); 0 = 1 mg hydroxyapatite,
= 0.5 mg hydroxyapatite.
Each point represents the mean -+ SEM (n=4). See text for detailb.
dose injected into arthritic human joints has been
estimated to require 30 to 90 days and 7 to 84 days
from normal rabbit joints, depending on the mean
crystal size (8). No comparative data with HA crystals
are available. Estimates of the relative stimulating
potency of HA and CPPD crystals will require not only
a dose-response curve for each crystal but also more
information on the intracellular degradation rate and a
longer period of in vitro study. The concentrations of
HA crystals added to normal canine synovial cell
cultures were 100 and 50 kg/ml, close to the amount
found in the shoulder joint fluids from our patients
(10-50 pg/ml) ( 3 ) . The concentrations of CPPD crystals used here were greater than those found in synovia1 fluid from patients with CPPD-related acute arthritis
(-3 to 12 pg/ml) (18).
Prostaglandin release has been shown to accompany release of collagenase and neutral protease
from cultured human rheumatoid synovial cell lines
(10). We speculate that HA but not CPPD crystals
release arachidonic acid from membrane phospholip-
a n
a .-2
I3- -5
0 2
Days in Culture
Figure 7. The cumulative release of lactic dehydrogenase from
cultured normal canine cells after exposure to hydroxyapatite
crystals. Each point represents the mean ? SEM (n=4).
ids, perhaps by activation of phospholipase A?. The
disruption of phagosomal membrane shown in Figure 3
appears to support this idea. The mechanism of prostaglandin generation from synovial cells by crystals of
various chemical composition needs further study.
HA crystals have been found in synovial fluid
from severely degenerated knee joints, often in association with CPPD crystals (19,20). Dieppe has shown
that such fluids do not have a higher total leukocyte
concentration than fluids from osteoarthritic joints
that do not contain crystals (21). Erosive arthritis of
finger joints has also been associated with HA crystal
deposition (22). Some cases of CPPD crystal deposition disease develop a destructive arthropathy, sometimes even resembling a neuropathic (Charcot) joint
(reviewed in reference 23). The mechanism proposed
here may also apply to the pathogenesis of the destructive arthropathy in these cases.
TIME (Days)
Figure 6. The cumulative release of collagenase and neutral protease from cultured normal canine synovial cells; A = control (no
crystals); 0 = I mg CPPD crystals; W = 0.5 mg CPPD crystals.
Each point represents the mean ? SEM (n=4).
We are grateful to Dr. Neil S. Mandel for the x-ray
crystallographic analyses of the crystal preparations used
here, to Dr. John Garancis for the transmission electron
microscopy, to Dr. Harold Itskovitz for the prostaglandin
.assays, and to Mrs. Janine Struve and Mr. T. Nichols for
their excellent technical help.
1. Halverson PB, Cheung HS, Carrera GF, Kozin F,
Garancis J, McCarty DJ: Rotator cuff tears associated
with synovial fluid hydroxyapatite crystals and activated
collagenase (abstract). Arthritis Rheum 23:687. 1980.
2. McCarty DJ, Halverson PB, Carrera GF. Brewer BJ,
Kozin F: “Milwaukee shoulder”: association of microspheroids containing hydroxyapatite crystals, active collagenase and neutral protease with rotator cuff defects.
1. Clinical aspects. Arthritis Rheum 24:464-473, 1981
3. Halverson PB. Cheung HS, McCarty DJ, Garancis JC,
Mandel N: “Milwaukee shoulder”: association of microspheroids containing hydroxyapatite crystals. active
collagenase and neutral protease with rotator cuff defects. 11. Synovial fluid studies. Arthritis Rheum 24:474483. 1981
4. Garancis JC, Cheung HS, Halverson PB, McCarty DJ:
“Milwaukee shoulder”: association of microspheroids
containing hydroxyapatite, active collagenase and neutral protease with rotator cuff defects. 111. Morphologic
and biochemical studies of an excised synovium showing chondromatosis. Arthritis Rheum 24:484-491, 1981
5. Werb Z, Reynolds JJ: Stimulation of endocytosis of the
secretion of collagenase and neutral proteinase from
rabbit synovial fibroblasts. J Exp Med 140: 1482-1497.
6 . Biwas C, Dayer JM: Stimulation of collagenase production by collagen in mammalian cell cultures. Cell
18:1035-1041, 1979
7. Clarris BJ, Fraser JRE, Moran CJ, Muirden KD: Rheumatoid synovial cells from intact joints. Ann Kheum Dis
36~293-301, 1977
8. McCarty DJ, Palmer DW, Halverson PB: Clearance of
calcium pyrophosphate dihydrate (CPPD) crystals in
vivo. I. Studies using 16”Yb labeled triclinic crystals.
Arthritis Rheum 22:718-727, 1979
9. Evanson M, Jeffrey JJ, Krane SM: Studies on collagenase from rheumatoid synovium in tissue culture. J Clin
Invest 47:2639-2651, 1968
10. Dayer JM, Krane SM, Russell RGG. Robinson DR:
Production of collagenase and prostaglandins by isolated
adherent rheumatoid synovial cells. Proc Natl Acad Sci
USA 73:945-949, 1976
11. Cheung HS, Harvey W, Benya PB, Nimni ME: New
collagen markers of “derepression” synthesized by rabbit articular cartilage. Biochem Biophys Res Commun
68:1371-1378, 1976
12. Skosey JL, Kozin F, Ginsberg MH: Protein absorption
to monosodium urate crystals: differential responses of
human peripheral blood neutrophils. J Clin Chem Clin
Biochem 14:303-306, 1976
13. Spilberg 1, Gallagher A, Mendell B: Studies on crystalinduced chemotactic factors. 11. Kole of phagocytosis. J
Lab Clin Med 85:631-636. 1975
14. Dunn MJ, Laird JF, Dray F: Basal and stimulated rates
of renal secretion and excretion of prostaglandins F2,
Fz, and 13.14 dihydro I5 beta F, in the dog. Kidney Int
13:136-143, 1978
15. Werb Z, Mainardi CL, Vater CA. Harris ED Jr: Endogenous activation of collagenase by rheumatoid synovial
cells: evidence for a role of plasminogen activator. N
Engl J Med 296:1017-1023, 1977
16. Berman M. Leary R, Gage J: Collagenase from corneal
cell cultures and its modulation by phagocytosis. Invest
Ophthal Visual Sci 18588-601, 1979
17. McCarty DJ, Palmer DW, James C: Clearance of calcium pyrophosphate dihydrate (CPPD) crystals in vivo. 11.
Studies using triclinic crystals doubly labeled with 4’Ca
and ”Sr. Arthritis Rheum 22: 1122-1 131, 1979
18. McCarty DJ: Calcium pyrophosphate dihydrate crystal
deposition disease-1975, Proceedings of the Conference on Pseudogout and Pyrophosphate Metabolism.
Edited by DJ McCarty. Arthritis Rheum (suppl) 19:275285, 1976
19. Dieppe PA, Doyle DV, Huskisson EC, Willoughby DA,
Crocker PR: Mixed crystal deposition disease and osteoarthritis. Br Med J I:150, 1978
20. Halverson PB, McCarty DJ: Identification of hydroxyapatite crystals in synovial fluid. Arthritis Rheum
22:389-395, 1979
21. Dieppe PA, Crocker PR, Corke CG, Doyle DV, Huskisson EC, Willoughby DA: Q J Med 48:533-553, 1979
22. Schumacher HR, Miller JL, Ludivico C, Jessar RA:
Erosive arthritis associated with apatite crystal deposition. Arthritis Kheum 24:31-37, 1981
23. Menkes CJ, Simon F, Delrieu F, Forest M, Delbarre F:
Destructive arthropathy in chondrocalcinosis articularis. Arthritis Rheum (suppl) 19:329-348, 1976
Без категории
Размер файла
615 Кб
crystals, prostaglandin, cells, neutral, collagenase, dihydrate, mammalia, release, pyrophosphate, hydroxyapatite, culture, calcium, synovial, protease
Пожаловаться на содержимое документа