- Email: [email protected]
Expression of IL-1 genes in human and bovine chondrocytes: A mechanism for autocrine control of cartilage matrix degradation
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
December 30, 1986
Pages 904-911
Expression of IL-I Genes in Human and Bovine Chondrocytes: A Mechanism for Autocrine Control of Cartilage Matrix Degradation I
Felicia Ollivierre*, Ueli Gublert, Christine A. Towle*, Cato Laurencin*, and Benjamin V. Treadwel1.2
*Orthopaedic Research Laboratories, Massachusetts General Hospital and, Harvard Medical School, Boston, MA 02114 tDepartment of Molecular Genetics, Hoffmann-LaRoche, Inc., Nutley, NJ 07110 Received November 3, 1986
In this report we describe the presence of interleukin-i activity in medium conditioned by bovine articular cartilage. Preparations partially purified by Sephacryl $200 chromatography (Mr 18000-25000) stimulate murine thymocyte proliferation in the lymphocyte activation factor assay. Furthermore, the factor(s) activate cartilage tissue to secrete a protease which is essential for the activity of purified synovial collagenase. We also demonstrate the presence of mRNA coding for IL-I~ and ~ in human articular chondrocytes and conclude that the human monocytic and chondrocytic mRNAs are identical. Our results demonstrating cartilage expression of IL-I genes suggest the possibility of an autocrine mechanism whereby chondrocyte production of matrix degrading proteases is initiated by chondrocyte derived IL-I. © 1986AcademicPress, Inc.
The members of the interleukin-i (IL-I) family of cytokines share the property of being stimulatory in the murine thymocyte comitogenic assay also known as lymphocyte activating factor, LAF, assay (I).
Other activities
reported for these cytokines are diverse ranging from sleep induction and endogenous pyrogen activity in vivo to stimulation of cell proliferation and synthesis and release of specific macromoleucles in a variety of target cells (i).
Synthesis of IL-I like factors, initially thought to be restricted to
cells of the monocyte macrophage lineage, has now been demonstrated for a wide variety of cell types (2-7).
iThis work was supported in part by grant #AM16265. 2To whom correspondence should be addressed. 0006-291 X/86 $1.50 Copyr~ht © 1986 by Academk' P~ss, Inc. All r~h~ ~" reproduction in any jbrm reserve~
904
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Recent reports from a number of laboratories using recombinant IL-I in the LAF assay and other IL-I test systems indicate that many of the diverse activities can be attributed to proteins encoded by one or two specific genes (8,9). We recently demonstrated that recombinant IL-1 and synovial-derived interleukin-i like factor(s) stimulate chondrocyte production of a latent metalloprotease which activates latent collagenase (10,11).
Here we report on the
production of factor(s) having interleukin-i activity by bovine chondrocytes. We also demonstrate for the first time in chondrocytes isolated from human articular cartilage the presence of mRNAs identical to human monocytic IL-I~ and ~ mRNA.
MATERIALS AND METHODS Mouse recombinant I L - I ~ was a gift from Dr. P. Lomedico, Hoffman La Roche, Nutley, NJ. All chromatographic media were from Pharmacia Fine Chemicals, Upsala, Sweden. All radioactive materials were from New England Nuclear, Boston, MA. Fresh calf joints from Trelegan's slaughter house, Cambridge, MA. Dulbecco's modified Eagle's Medium, Grand Island Biological Company, Grand Island, New York. All culture plates and flasks were Falcon, obtained from Fisher Scientific. Protein determination was according to the method of Bradford (12). Preparation of Cartilage Condition Me diu m Prepared as described in reference Ii. Partial Purification of Chondrocyte IL-I Chromatographic Procedures All procedures were carried out as described in reference i0. Assay for IL-I: Stimulation of Synthesis of Collagenase Activator Protein Cartilage cores from articular surface of calf joints (radial carpal) were obtained with a 3.5 mm diameter cork borer. The upper 1.0 mm (superficial layer) was sliced off using a specially constructed template. The remaining cartilage was discarded. The cartilage slices were washed in DMEM and placed one per well into a 96 well microwell plate containing 200 ul per well of DMEM and 20 uCi of [35S] methionine (S.A. 1200-1400 Ci/rmnole). Recombinant IL-I or partially purified synovial or cartilage factors diluted in 5 ul DMEM were added where indicated. The samples were incubated at 37 ° in an atmosphere of 95% air, 5% CO 2 for 18 h. Medium from each well was removed and the secreted protein precipitated by the addition of two volumes (400 ul) cold acetone. The precipitated protein was collected by centrifugation (I0,000 x g i0 min), lyophilized to dryness and solubilized in sample buffer containing 2.5% SDS, I% glycerol, 1% 2-mercaptoethanol, i0 mM Tris HCI, pH 7.0 and 20 ug/ml phenol red (approximately 50 ul/pellet). The solubilized pellets were applied to a 11% polyacrylamide slab gel prepared according to the procedure of Laemmli (14). Electrophoresis was carried out in the cold room at a constant current of 40 m amp/gel. The proteins were fixed and visualized by staining in Coomassie blue and destaining in acetic acid. Homogeneous collagenase activator protein purified from cartilage conditioned medium by the procedure of Treadwell et al. (10,11) was also run for identification purposes. Radioactive samples run on SDS polyacrylamide gels were visualized after drying the gel under vacuum and placing the dried gel in contact with x-ray film for 18-24 h.
905
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Lylnphocyte Activation Assay - IL-I Assay This assay was performed according to the procedure of Kaye et al. (13). Preparation of Human Chondrocytes Human articular cartilage was resected from the knee of a 29 year old male immediately after the limb was surgically removed. The cartilage appeared normal on histological examination; the limb was amputated to prevent further metastasis of a highly malignant chondrosarcoma of the hip. The cartilage (5 gm) was placed in DMEM (200 ml) containing 10% fetal calf serum and 0.1% clostridial collagenase (188 u/mg, Cooper Biomedical, Malvern, PA). The tissue was stirred for 18 h at 37 ° in an atmosphere of 95% 5% CO 2. Non-digested cartilage was separated from cells by passage through sterile cheese cloth. The cells were washed in PBS and microscopically examined for viability using the trypan blue exclusion method (.025% trypan blue). The results showed greater than 90% viable chondrocytes. The cells were pelleted by centrifugation and frozen in liquid nitrogen. Isolation of RNA from Human Chondrocytes and S 1 N u c l e a s e Protection Assay Total RNA from frozen human chondrocytes was isolated by the procedure of Chirgwin et el. (15). The S1 nuclease protection assay requires the use of an end labelled probe complementary to the mRNA to be detected (16). To achieve this, cDNA clones for human IL-I~ (8) and IL-I~ (17) were first modified by deleting one of the homopolymeric tails used in the original cloning. SI probes free of homopolymeric tails were then generated for IL-I~ by labelling a 960 bp Hindlll - Pst I fragment at the Hindlll site in the strand complementary to the mRNA and for IL-I~ by labeling an 855 bp Hindlll - Taq I fragment at the Hind III site in the strand complementary to the mRNA. The specific activities of these probes were determined for each preparation and were 1 to 2 x 104 dpm/fmole. 2 to 5 fmoles of end-labelled ~ - and E-probe were hybridized separately to 5 - 50 ug of total RNA sample in 80% deionized formamide, 0.4 M NaCI, 1 mM EDTA, 40 mM Pipes pH 6.8 at 52°C overnight. Following this incubation each sample was digested for 60 minutes at 37°C with i00 units of S1 nuclease (18). The products were subsequently analyzed on a 5% sequencing gel. The sizes of the protected fragments in this assay are 210 nucleotides for IL-I~ (corresponding to the coding region for amino acids no. 64 to 134 of the complete IL-IO¢ precursor) and 525 nucleotides for IL-I~ corresponding to the coding region for amino acids no. 133 to 269 of the complete IL-I precursor plus 116 nucleotides of 3' untranslated region). RESULTS We previously reported that IL-I like factor(s) synthesized by bovine synovial tissue stimulate cartilage to produce a protease which activates latent collagenase (Ii). This protease migrates as a doublet on sodium dodecyl sulfate polyacrylamide gel electrophoresis with apparent molecular weight of 55-57,000.
Coomassie blue stained collagenase activator protein, purified
according to the procedure of Treadwell (i0) is shown in lane 1 of fig. I. Synthesis of the 55-57,000 protein in control cartilage and in cartilage stimufated by bovine synovial factor and recombinant murine IL-I is shown in the autoradiogram of [35S]methionine labeled proteins (fig. I, lane 2-4). Partially purified cartilage factor(s) also stimulate secretion of this collagenase activator protein (fig. i, lane 5).
Fractions determined by autoradio-
graphy to be most active in stimulating the synthesis of the 55-57,000 proteins
906
Vol. 141. No. 3. 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
1
2
3
4
5
.,.,90 ,,.,67
,,,. 43
-.30
Figure i. Sodium dodecyl sulfate polyacrylamide gel electrophoresis and autoradiography showing stimulated synthesis of 55-57K collagenase activator protein in cartilage incubated in the presence of IL-I and related factors. Lane 1 of the figure shows a Coomassie blue stained gel lane with I u g purified collagenase activator protein. Lanes 2-5 are autoradiograms of [35S]methionine labeled proteins in 200 ul medium conditioned by cartilage in the absence of added factor (lane 2) or in the presence of 4.5 ng purified bovine synovial factor (lane 3), 1.0 ng mouse recombinant IL-I~ (lane 4) and 500 ng partially purified bovine cartilage derived factor(s) (lane 5). The arrowhead indicates the position of the collagenase activator protein.
eluted from the Sephacryl
S200 column
in the molecular
weight region between
18-25,000. Active
fractions were pooled and analyzed
thymocyte comitogenic protein production.
assay and for stimulation
for IL-I activity using the of collagenase
Table 1 shows that preparations
cartilage conditioned medium have both activities factor and murine recombinant by bovine cartilage contains
IL-I.
Production determine
as do synovial
These results
activities
of IL-I by chondrocytes
whether cartilage
partially purified
from
IL-I like
show that medium conditioned
factor(s) with molecular
and having two of the biological
activator
weights
similar
to IL-I
of interleukin-1.
has not been reported previously.
cells have the capacity
907
to synthesize
authentic
To
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table 1 IL-I Activity in Cartilage and Synovial IL-I Like Factors IL-I
LAF*
Cartilage IL-I A Synovial I L - I A Mouse rlL-1
CAP Induction?
660 2040 108
* LAF - lymphocyte activation
500 2000 2 x 106
factor activity, units per mg.
? CAP (collagenase activator protein) induction - units per mg, one unit is the amount required to stimulate superficial cartilage to double the control level of 55-57K protein synthesis. A Bovine factor at sephacryl $200 stage of purification.
1
2
3
4
5
6
7
8
q53
Figure 2. S1 nuclease protection assay. Lanes 1-4 indicate RNA samples reacted with IL-I~ probe, lanes 5-8 with IL-IB probe. M = marker (pBR 322 cut with Hinfl, sizes in nucleotides). Arrowheads indicate protected fragments. Lanes 1 , 5 : 3 0 ug of pig gut RNA (negative control); lanes 2,6: no RNA (negative control); lanes 3 , 7 : 8 ug RNA from human LPS-induced peripheral blood leukocytes (positive control); lanes 4 , 8 : 3 0 ug of human articular chondrocyte RNA. In the experiment shown here, the specific activity of the IL-I~ probe was twice as high as that of the IL-I~ probe. The nature of the fragment appearing in lanes 5-8 at a position just below 298 nucleotides is unknown.
908
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
IL-1, total RNA from human chondrocytes was subjected to S1 nuclease protection assays, using end labelled probes for human monocytic I L - I ~
and IL-I~ .
shown in figure 2, human chondrocytes contain mRNAs for both I L - l ~ a n d
As
IL-I~ .
In addition, similar to monocytes, chondrocytes appear to contain higher levels of IL-I~
mRNA.
Given the stringency of the Sl-assay conditions, we can
conclude that human monocytic and chondrocytie IL-I mRNAs are identical since single base mismatches in the cDNA: mRNA hybrids would result in the degradation of the labeled probes by S1 nuclease. DISCUSSION
Interleukin-i was originally believed to be produced by neutrophils. Later it was shown that the monocytes in the preparation of leukocytes were actually responsible for the production of a factor that stimulated thymocyte proliferation (I). More recent evidence demonstrates the production of this factor by a variety of cells (2-7). We provide evidence that articular cartilage synthesizes IL-I which is active both in the LAF assay and in stimulating the synthesis of a protein identical in electrophoretic mobility with collagenase activator protein. Detection of protease activity in cartilage conditioned medium does not yield reproducible results unless the medium is fractionated to remove cartilage derived protease inhibitors.
Therefore, production of enzymatic activity was
not used to assay column fractions.
We have previously shown increased produc-
tion of collagenase activator protein by cartilage stimulated by recombinant IL-I and IL-I like factor purified from synovial conditioned medium (i0). The evidence that chondrocytes synthesize IL-I is strengthened by the results presented here showing that human chondrocytes contain mRNA for O/ and IL-I.
Our results indicate that human monocytic and chondrocytic IL-I mRNAs
are identical.
In the positive control RNA (isolated from LPS-stimulated human
leukocytes), the Ii-I~
probe gives rise to two protected fragments.
This
observation is very reproducible and is probably due to an S1 sensitive site rich in AT-basepairs located 125 nucleotides away from the labelled end of the probe.
The reason why the chondrocyte RNA does not give rise to the two
909
Vol, 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
fragments is not completely clear.
One interpretation for the difference could
be the presence of a population polymorphism of the IL-I~
gene since leuko-
cytes were isolated from pooled donor blood whereas the chondrocytes were derived from a single donor.
However, sequencing of a number of cDNA and
genomic clones (8,20) has thus far revealed only one population polymorphism for the IL-I~
gene and that does not map to the area in question.
Further
studies will be required to resolve this question. The deterioration of cartilage matrix which is a characteristic feature in the pathology of rheumatoid and osteoarthritis probably results from the action of secreted proteases.
The rate-limiting step in degradation of the collagen-
ous matrix is the initial cleavage of triple helical collagen molecules by a specific collagenase.
A variety of cultured connective tissue cells secrete an
inactive procollagenase; crude preparations of this enzyme can be activated by treatment with a variety of agents including mercurials and trypsin.
We (i0)
and others (19) have demonstrated that the "latent" collagenase purified from synovial conditioned medium is absolutely dependent for its activity on a second neutral metalloprotease.
IL-I stimulates cartilage cells to produce
this collagenase activator protein. Previous reports have emphasized the role of synovial IL-I like factors in cartilage degradation.
We provide evidence that cartilage cells produce IL-I,
suggesting an autocrine mechanism whereby production of matrix degrading proteases may be initiated by the chondrocytes themselves.
ACKNOWLEDGEMENT We thank Bill Benjamin of Hoffman-LaRoche for performing the lymphocyte activation assays. REFERENCES
i.
Dinarello, C.A. (1984) Reviews of Infectious Diseases 6, 51-95.
2.
Hanazawa, S., Ohmori, Y., Amano, S., Miyoshi, T., Kumegawa, M., Kitano, S. (1985) Biochem. Biophys. Res. Comm. 131, 774-779.
3.
Iribe, H., Koga, T., Kotani, S., Kusumoto, S., and Shiba, T. (1983) J. Exp. Med. 157, 2190-2195.
910
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
4.
Luger, T.A., Stadler, B.M., Luger, B.M., Mathieson, B.J., Mage, M., Schmidt, J.A., Openheim, J.J. (1982) J. Immunol. 128, 2147-2152.
5.
Fontana, A., Kristensen, F., Dubs, R., Gemsa, D., and Weber, E. (1982) J. Immunol. 129, 2413-2419.
6.
Miossec, P., Cavender, D., ziff, M. (1986) J. Immunol. 136, 2486-2491.
7.
Lovett, D.H., Szamel, M., Ryan, J.L., Sterzel, R.B., Gemsa, D., Resch, K. (1986) J. Immunol. 136, 3700-3705.
8.
Gubler, U., Chua, A.O., Stern, A.S., Hellmann, C.P., Vitek, M.P., Dechiara, T.M., Benjamin, W.R., Collier, K.J., Dukovich, M., Familletti, P.C., Fiedler-Nagy, C., Jenson, J., Kaffka, K., Kilian, P.L., Stremlo, D., Wittreich, B.H., Woehle, D., Mizel, S.B., Lomedico, P.T. (1986) J. Immunol. 136, 2492-2497.
9.
Fiedler-Nagy, C., Batula-Bernardo, C., Semionow, R., Stern, A.S. (1985) Arth. Rheum. 28, $54.
i0.
Treadwell, B.V., Towle, C.A., Ishizue, K., Mankin, K.P., Pavia, M., Ollivierre, F.M., Gray, D.H. (1986) Arch. Biochem. Biophys. Accepted for publication.
ii.
Treadwell, B.V., Neidel, J., Pavia, M., Towle, C.A., Trice, M.E., Mmnkin, H.J. (1986) Arch. Biochem. Biophys. Accepted for publication.
12.
Bradford, M.D. (1976) Anal. Biochem. 72, 248-254.
13.
Kaye, J., Gillis, S., Mizel, S.B., Shevach, E.M., Malek T.R., Dinarello, C.A., Lachman, L.B., and Janeway, C.A., Jr. (1984) J. Immunol. 133, 1339.
14.
Laemmli, U.K. (1970) Nature 22, 680-684.
15.
Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J., Rutter, W.J. (1979) Biochemistry 18:5294.
16.
Berk, A.J., Sharp, P.A. (1977) Cell 12:721.
17.
Auron, P.E., Webb, A.C., Rosenwasser, L.J., Mucci, S.F., Rich, A., Wolff, S.M., Dinarello, C.A. (1984) Proc. Natl. Acad. Sci. USA 81:7907.
18.
Maniatis, T., Fritsch, E.F., Sambrook J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory.
19.
Vater, C.S., Nagase, H., and Harris, E.D., Jr. (1983) J. Biol. Chem. 258, 9374-9382.
20.
Furutani, Y., M. Notake, M. Yamayoshi, J.-l. Yamagishi, H. Nomura, M. Ohue, R. Furuta, T. Fukui, M. Yamada, and S. Nakamura. 1985. Cloning and characterization of the cDNAs for human and rabbit interleukin-I precursor. Nucl. Acids Res. 13:5869.
911
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
December 30, 1986
Pages 904-911
Expression of IL-I Genes in Human and Bovine Chondrocytes: A Mechanism for Autocrine Control of Cartilage Matrix Degradation I
Felicia Ollivierre*, Ueli Gublert, Christine A. Towle*, Cato Laurencin*, and Benjamin V. Treadwel1.2
*Orthopaedic Research Laboratories, Massachusetts General Hospital and, Harvard Medical School, Boston, MA 02114 tDepartment of Molecular Genetics, Hoffmann-LaRoche, Inc., Nutley, NJ 07110 Received November 3, 1986
In this report we describe the presence of interleukin-i activity in medium conditioned by bovine articular cartilage. Preparations partially purified by Sephacryl $200 chromatography (Mr 18000-25000) stimulate murine thymocyte proliferation in the lymphocyte activation factor assay. Furthermore, the factor(s) activate cartilage tissue to secrete a protease which is essential for the activity of purified synovial collagenase. We also demonstrate the presence of mRNA coding for IL-I~ and ~ in human articular chondrocytes and conclude that the human monocytic and chondrocytic mRNAs are identical. Our results demonstrating cartilage expression of IL-I genes suggest the possibility of an autocrine mechanism whereby chondrocyte production of matrix degrading proteases is initiated by chondrocyte derived IL-I. © 1986AcademicPress, Inc.
The members of the interleukin-i (IL-I) family of cytokines share the property of being stimulatory in the murine thymocyte comitogenic assay also known as lymphocyte activating factor, LAF, assay (I).
Other activities
reported for these cytokines are diverse ranging from sleep induction and endogenous pyrogen activity in vivo to stimulation of cell proliferation and synthesis and release of specific macromoleucles in a variety of target cells (i).
Synthesis of IL-I like factors, initially thought to be restricted to
cells of the monocyte macrophage lineage, has now been demonstrated for a wide variety of cell types (2-7).
iThis work was supported in part by grant #AM16265. 2To whom correspondence should be addressed. 0006-291 X/86 $1.50 Copyr~ht © 1986 by Academk' P~ss, Inc. All r~h~ ~" reproduction in any jbrm reserve~
904
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Recent reports from a number of laboratories using recombinant IL-I in the LAF assay and other IL-I test systems indicate that many of the diverse activities can be attributed to proteins encoded by one or two specific genes (8,9). We recently demonstrated that recombinant IL-1 and synovial-derived interleukin-i like factor(s) stimulate chondrocyte production of a latent metalloprotease which activates latent collagenase (10,11).
Here we report on the
production of factor(s) having interleukin-i activity by bovine chondrocytes. We also demonstrate for the first time in chondrocytes isolated from human articular cartilage the presence of mRNAs identical to human monocytic IL-I~ and ~ mRNA.
MATERIALS AND METHODS Mouse recombinant I L - I ~ was a gift from Dr. P. Lomedico, Hoffman La Roche, Nutley, NJ. All chromatographic media were from Pharmacia Fine Chemicals, Upsala, Sweden. All radioactive materials were from New England Nuclear, Boston, MA. Fresh calf joints from Trelegan's slaughter house, Cambridge, MA. Dulbecco's modified Eagle's Medium, Grand Island Biological Company, Grand Island, New York. All culture plates and flasks were Falcon, obtained from Fisher Scientific. Protein determination was according to the method of Bradford (12). Preparation of Cartilage Condition Me diu m Prepared as described in reference Ii. Partial Purification of Chondrocyte IL-I Chromatographic Procedures All procedures were carried out as described in reference i0. Assay for IL-I: Stimulation of Synthesis of Collagenase Activator Protein Cartilage cores from articular surface of calf joints (radial carpal) were obtained with a 3.5 mm diameter cork borer. The upper 1.0 mm (superficial layer) was sliced off using a specially constructed template. The remaining cartilage was discarded. The cartilage slices were washed in DMEM and placed one per well into a 96 well microwell plate containing 200 ul per well of DMEM and 20 uCi of [35S] methionine (S.A. 1200-1400 Ci/rmnole). Recombinant IL-I or partially purified synovial or cartilage factors diluted in 5 ul DMEM were added where indicated. The samples were incubated at 37 ° in an atmosphere of 95% air, 5% CO 2 for 18 h. Medium from each well was removed and the secreted protein precipitated by the addition of two volumes (400 ul) cold acetone. The precipitated protein was collected by centrifugation (I0,000 x g i0 min), lyophilized to dryness and solubilized in sample buffer containing 2.5% SDS, I% glycerol, 1% 2-mercaptoethanol, i0 mM Tris HCI, pH 7.0 and 20 ug/ml phenol red (approximately 50 ul/pellet). The solubilized pellets were applied to a 11% polyacrylamide slab gel prepared according to the procedure of Laemmli (14). Electrophoresis was carried out in the cold room at a constant current of 40 m amp/gel. The proteins were fixed and visualized by staining in Coomassie blue and destaining in acetic acid. Homogeneous collagenase activator protein purified from cartilage conditioned medium by the procedure of Treadwell et al. (10,11) was also run for identification purposes. Radioactive samples run on SDS polyacrylamide gels were visualized after drying the gel under vacuum and placing the dried gel in contact with x-ray film for 18-24 h.
905
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Lylnphocyte Activation Assay - IL-I Assay This assay was performed according to the procedure of Kaye et al. (13). Preparation of Human Chondrocytes Human articular cartilage was resected from the knee of a 29 year old male immediately after the limb was surgically removed. The cartilage appeared normal on histological examination; the limb was amputated to prevent further metastasis of a highly malignant chondrosarcoma of the hip. The cartilage (5 gm) was placed in DMEM (200 ml) containing 10% fetal calf serum and 0.1% clostridial collagenase (188 u/mg, Cooper Biomedical, Malvern, PA). The tissue was stirred for 18 h at 37 ° in an atmosphere of 95% 5% CO 2. Non-digested cartilage was separated from cells by passage through sterile cheese cloth. The cells were washed in PBS and microscopically examined for viability using the trypan blue exclusion method (.025% trypan blue). The results showed greater than 90% viable chondrocytes. The cells were pelleted by centrifugation and frozen in liquid nitrogen. Isolation of RNA from Human Chondrocytes and S 1 N u c l e a s e Protection Assay Total RNA from frozen human chondrocytes was isolated by the procedure of Chirgwin et el. (15). The S1 nuclease protection assay requires the use of an end labelled probe complementary to the mRNA to be detected (16). To achieve this, cDNA clones for human IL-I~ (8) and IL-I~ (17) were first modified by deleting one of the homopolymeric tails used in the original cloning. SI probes free of homopolymeric tails were then generated for IL-I~ by labelling a 960 bp Hindlll - Pst I fragment at the Hindlll site in the strand complementary to the mRNA and for IL-I~ by labeling an 855 bp Hindlll - Taq I fragment at the Hind III site in the strand complementary to the mRNA. The specific activities of these probes were determined for each preparation and were 1 to 2 x 104 dpm/fmole. 2 to 5 fmoles of end-labelled ~ - and E-probe were hybridized separately to 5 - 50 ug of total RNA sample in 80% deionized formamide, 0.4 M NaCI, 1 mM EDTA, 40 mM Pipes pH 6.8 at 52°C overnight. Following this incubation each sample was digested for 60 minutes at 37°C with i00 units of S1 nuclease (18). The products were subsequently analyzed on a 5% sequencing gel. The sizes of the protected fragments in this assay are 210 nucleotides for IL-I~ (corresponding to the coding region for amino acids no. 64 to 134 of the complete IL-IO¢ precursor) and 525 nucleotides for IL-I~ corresponding to the coding region for amino acids no. 133 to 269 of the complete IL-I precursor plus 116 nucleotides of 3' untranslated region). RESULTS We previously reported that IL-I like factor(s) synthesized by bovine synovial tissue stimulate cartilage to produce a protease which activates latent collagenase (Ii). This protease migrates as a doublet on sodium dodecyl sulfate polyacrylamide gel electrophoresis with apparent molecular weight of 55-57,000.
Coomassie blue stained collagenase activator protein, purified
according to the procedure of Treadwell (i0) is shown in lane 1 of fig. I. Synthesis of the 55-57,000 protein in control cartilage and in cartilage stimufated by bovine synovial factor and recombinant murine IL-I is shown in the autoradiogram of [35S]methionine labeled proteins (fig. I, lane 2-4). Partially purified cartilage factor(s) also stimulate secretion of this collagenase activator protein (fig. i, lane 5).
Fractions determined by autoradio-
graphy to be most active in stimulating the synthesis of the 55-57,000 proteins
906
Vol. 141. No. 3. 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
1
2
3
4
5
.,.,90 ,,.,67
,,,. 43
-.30
Figure i. Sodium dodecyl sulfate polyacrylamide gel electrophoresis and autoradiography showing stimulated synthesis of 55-57K collagenase activator protein in cartilage incubated in the presence of IL-I and related factors. Lane 1 of the figure shows a Coomassie blue stained gel lane with I u g purified collagenase activator protein. Lanes 2-5 are autoradiograms of [35S]methionine labeled proteins in 200 ul medium conditioned by cartilage in the absence of added factor (lane 2) or in the presence of 4.5 ng purified bovine synovial factor (lane 3), 1.0 ng mouse recombinant IL-I~ (lane 4) and 500 ng partially purified bovine cartilage derived factor(s) (lane 5). The arrowhead indicates the position of the collagenase activator protein.
eluted from the Sephacryl
S200 column
in the molecular
weight region between
18-25,000. Active
fractions were pooled and analyzed
thymocyte comitogenic protein production.
assay and for stimulation
for IL-I activity using the of collagenase
Table 1 shows that preparations
cartilage conditioned medium have both activities factor and murine recombinant by bovine cartilage contains
IL-I.
Production determine
as do synovial
These results
activities
of IL-I by chondrocytes
whether cartilage
partially purified
from
IL-I like
show that medium conditioned
factor(s) with molecular
and having two of the biological
activator
weights
similar
to IL-I
of interleukin-1.
has not been reported previously.
cells have the capacity
907
to synthesize
authentic
To
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table 1 IL-I Activity in Cartilage and Synovial IL-I Like Factors IL-I
LAF*
Cartilage IL-I A Synovial I L - I A Mouse rlL-1
CAP Induction?
660 2040 108
* LAF - lymphocyte activation
500 2000 2 x 106
factor activity, units per mg.
? CAP (collagenase activator protein) induction - units per mg, one unit is the amount required to stimulate superficial cartilage to double the control level of 55-57K protein synthesis. A Bovine factor at sephacryl $200 stage of purification.
1
2
3
4
5
6
7
8
q53
Figure 2. S1 nuclease protection assay. Lanes 1-4 indicate RNA samples reacted with IL-I~ probe, lanes 5-8 with IL-IB probe. M = marker (pBR 322 cut with Hinfl, sizes in nucleotides). Arrowheads indicate protected fragments. Lanes 1 , 5 : 3 0 ug of pig gut RNA (negative control); lanes 2,6: no RNA (negative control); lanes 3 , 7 : 8 ug RNA from human LPS-induced peripheral blood leukocytes (positive control); lanes 4 , 8 : 3 0 ug of human articular chondrocyte RNA. In the experiment shown here, the specific activity of the IL-I~ probe was twice as high as that of the IL-I~ probe. The nature of the fragment appearing in lanes 5-8 at a position just below 298 nucleotides is unknown.
908
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
IL-1, total RNA from human chondrocytes was subjected to S1 nuclease protection assays, using end labelled probes for human monocytic I L - I ~
and IL-I~ .
shown in figure 2, human chondrocytes contain mRNAs for both I L - l ~ a n d
As
IL-I~ .
In addition, similar to monocytes, chondrocytes appear to contain higher levels of IL-I~
mRNA.
Given the stringency of the Sl-assay conditions, we can
conclude that human monocytic and chondrocytie IL-I mRNAs are identical since single base mismatches in the cDNA: mRNA hybrids would result in the degradation of the labeled probes by S1 nuclease. DISCUSSION
Interleukin-i was originally believed to be produced by neutrophils. Later it was shown that the monocytes in the preparation of leukocytes were actually responsible for the production of a factor that stimulated thymocyte proliferation (I). More recent evidence demonstrates the production of this factor by a variety of cells (2-7). We provide evidence that articular cartilage synthesizes IL-I which is active both in the LAF assay and in stimulating the synthesis of a protein identical in electrophoretic mobility with collagenase activator protein. Detection of protease activity in cartilage conditioned medium does not yield reproducible results unless the medium is fractionated to remove cartilage derived protease inhibitors.
Therefore, production of enzymatic activity was
not used to assay column fractions.
We have previously shown increased produc-
tion of collagenase activator protein by cartilage stimulated by recombinant IL-I and IL-I like factor purified from synovial conditioned medium (i0). The evidence that chondrocytes synthesize IL-I is strengthened by the results presented here showing that human chondrocytes contain mRNA for O/ and IL-I.
Our results indicate that human monocytic and chondrocytic IL-I mRNAs
are identical.
In the positive control RNA (isolated from LPS-stimulated human
leukocytes), the Ii-I~
probe gives rise to two protected fragments.
This
observation is very reproducible and is probably due to an S1 sensitive site rich in AT-basepairs located 125 nucleotides away from the labelled end of the probe.
The reason why the chondrocyte RNA does not give rise to the two
909
Vol, 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
fragments is not completely clear.
One interpretation for the difference could
be the presence of a population polymorphism of the IL-I~
gene since leuko-
cytes were isolated from pooled donor blood whereas the chondrocytes were derived from a single donor.
However, sequencing of a number of cDNA and
genomic clones (8,20) has thus far revealed only one population polymorphism for the IL-I~
gene and that does not map to the area in question.
Further
studies will be required to resolve this question. The deterioration of cartilage matrix which is a characteristic feature in the pathology of rheumatoid and osteoarthritis probably results from the action of secreted proteases.
The rate-limiting step in degradation of the collagen-
ous matrix is the initial cleavage of triple helical collagen molecules by a specific collagenase.
A variety of cultured connective tissue cells secrete an
inactive procollagenase; crude preparations of this enzyme can be activated by treatment with a variety of agents including mercurials and trypsin.
We (i0)
and others (19) have demonstrated that the "latent" collagenase purified from synovial conditioned medium is absolutely dependent for its activity on a second neutral metalloprotease.
IL-I stimulates cartilage cells to produce
this collagenase activator protein. Previous reports have emphasized the role of synovial IL-I like factors in cartilage degradation.
We provide evidence that cartilage cells produce IL-I,
suggesting an autocrine mechanism whereby production of matrix degrading proteases may be initiated by the chondrocytes themselves.
ACKNOWLEDGEMENT We thank Bill Benjamin of Hoffman-LaRoche for performing the lymphocyte activation assays. REFERENCES
i.
Dinarello, C.A. (1984) Reviews of Infectious Diseases 6, 51-95.
2.
Hanazawa, S., Ohmori, Y., Amano, S., Miyoshi, T., Kumegawa, M., Kitano, S. (1985) Biochem. Biophys. Res. Comm. 131, 774-779.
3.
Iribe, H., Koga, T., Kotani, S., Kusumoto, S., and Shiba, T. (1983) J. Exp. Med. 157, 2190-2195.
910
Vol. 141, No. 3, 1986
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
4.
Luger, T.A., Stadler, B.M., Luger, B.M., Mathieson, B.J., Mage, M., Schmidt, J.A., Openheim, J.J. (1982) J. Immunol. 128, 2147-2152.
5.
Fontana, A., Kristensen, F., Dubs, R., Gemsa, D., and Weber, E. (1982) J. Immunol. 129, 2413-2419.
6.
Miossec, P., Cavender, D., ziff, M. (1986) J. Immunol. 136, 2486-2491.
7.
Lovett, D.H., Szamel, M., Ryan, J.L., Sterzel, R.B., Gemsa, D., Resch, K. (1986) J. Immunol. 136, 3700-3705.
8.
Gubler, U., Chua, A.O., Stern, A.S., Hellmann, C.P., Vitek, M.P., Dechiara, T.M., Benjamin, W.R., Collier, K.J., Dukovich, M., Familletti, P.C., Fiedler-Nagy, C., Jenson, J., Kaffka, K., Kilian, P.L., Stremlo, D., Wittreich, B.H., Woehle, D., Mizel, S.B., Lomedico, P.T. (1986) J. Immunol. 136, 2492-2497.
9.
Fiedler-Nagy, C., Batula-Bernardo, C., Semionow, R., Stern, A.S. (1985) Arth. Rheum. 28, $54.
i0.
Treadwell, B.V., Towle, C.A., Ishizue, K., Mankin, K.P., Pavia, M., Ollivierre, F.M., Gray, D.H. (1986) Arch. Biochem. Biophys. Accepted for publication.
ii.
Treadwell, B.V., Neidel, J., Pavia, M., Towle, C.A., Trice, M.E., Mmnkin, H.J. (1986) Arch. Biochem. Biophys. Accepted for publication.
12.
Bradford, M.D. (1976) Anal. Biochem. 72, 248-254.
13.
Kaye, J., Gillis, S., Mizel, S.B., Shevach, E.M., Malek T.R., Dinarello, C.A., Lachman, L.B., and Janeway, C.A., Jr. (1984) J. Immunol. 133, 1339.
14.
Laemmli, U.K. (1970) Nature 22, 680-684.
15.
Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J., Rutter, W.J. (1979) Biochemistry 18:5294.
16.
Berk, A.J., Sharp, P.A. (1977) Cell 12:721.
17.
Auron, P.E., Webb, A.C., Rosenwasser, L.J., Mucci, S.F., Rich, A., Wolff, S.M., Dinarello, C.A. (1984) Proc. Natl. Acad. Sci. USA 81:7907.
18.
Maniatis, T., Fritsch, E.F., Sambrook J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory.
19.
Vater, C.S., Nagase, H., and Harris, E.D., Jr. (1983) J. Biol. Chem. 258, 9374-9382.
20.
Furutani, Y., M. Notake, M. Yamayoshi, J.-l. Yamagishi, H. Nomura, M. Ohue, R. Furuta, T. Fukui, M. Yamada, and S. Nakamura. 1985. Cloning and characterization of the cDNAs for human and rabbit interleukin-I precursor. Nucl. Acids Res. 13:5869.
911