Modulation by retinoic acid and corticosteroids of collagenase production by rabbit synovial fibroblasts treated with phorbol myristate acetate or poly(ethylene glycol)

Effect of retinoids on rheumatoid arthritis, a proliferative and invasive non-malignant disease

In rheumatoid arthritis synovial tissue proliferates and destroys articular cartilage, bone and tendons. Collagenase is a major mediator of the connective tissue degradation. This enzyme is produced in large quantities by rheumatoid tissue and its synthesis can be inhibited by retinoids. However, knowledge of mechanisms controlling retinoid inhibition of collagenase production and of factors possibly controlling synovial cell proliferation is limited. We found that transforming growth factor beta in combination with epidermal growth factor, epidermal growth factor alone and immune interferon increased proliferation of cultured human and rabbit synovial fibroblasts. Only transforming growth factor beta caused a piling up of cells into foci resembling those seen in primary cultures of human rheumatoid tissue. All the factors were antagonized by retinoids but not by glucocorticoids or indomethacin. Adding retinoids or glucocorticoids to collagenase-producing cells decreased hybridizable collagenase mRNA by 50% within 24 h. Oral administration of retinoids to rats with experimental arthritis decreased clinical disease without toxicity, and inhibited collagenase synthesis by synovial cells taken from treated animals. Retinoids are both antiproliferative and anti-invasive, and therefore may be potential therapeutic agents in the treatment of rheumatoid arthritis.

Retinoid-mediated suppression of tumor invasion and matrix metalloproteinase synthesis

Cancer mortality usually results from the tumor invading the local environment and metastasizing to vital organs, e.g. liver, lung, and brain. Degradation of the extracellular matrix is, therefore, the sine qua non of tumor cell invasion. this degradation is mediated mainly by MMPs, and thus, inhibition of MMP synthesis is a target for anticancer agents. Tumor cells must traverse both the basement membrane (type IV collagen) and the interstitial stroma (type I collagen). Therefore, we used scanning electron microscopy to examine the invasive behavior of several aggressive tumor cell lines, A2058 melanoma cells, and SCC and FaDu squamous cell carcinomas through these matrices; and we monitored the ability of all-trans retinoic acid and several RAR-specific ligands to block invasion. We demonstrate that several retinoids, which are specific RAR alpha, beta, or gamma agonists/antagonists, selectively inhibited MMP synthesis in the three tumor cell lines. However, there was not a common pattern of MMP inhibition by a particular retinoid. For instance, a RAR alpha antagonist suppressed MMP-1 and MMP-2 synthesis in the melanoma cell line, but not in the FaDu or SCC-25 cells. On the other hand, synthesis of MMP-1 and MMP-9 by the FaDu cells was affected hardly at all, while a RAR gamma antagonist reduced the levels of MMP-2. Only all-trans retinoic acid reduced MMP-1 synthesis in these cells. We postulate that the differences may be related to a differential pattern of RAR expression in each of these cells, and that the RARs expressed by each cell line may not be targets of these RAR specific compounds. All-trans retinoic acid is a pan ligand, binding to all three RARs and, therefore, may modulate gene expression more generally. We conclude that the power of these new ligands lies in their specificity, which can be directed towards modulating expression of certain RARs and, thus, of certain MMPs. By blocking MMP synthesis, retinoids may be effective in cancer therapy by decreasing tumor invasiveness.

Induction of matrix metalloprotease-1 gene expression by retinoic acid in the human pancreatic tumour cell line Dan-G

We have investigated the effects of retinoic acid (RA) on matrix metalloprotease-1 (MMP-1) gene expression in the human pancreatic tumour cell line Dan-G. 13-cis RA results in a time- and dose-dependent increase of MMP-1 protein concentration. These stimulatory effects were paralleled by a time- and dose-dependent increase of MMP-1 mRNA steady-state concentrations. Nuclear run-on analysis revealed that the increase of MMP-1 mRNA was partially due to an increase of MMP-1 gene transcription. In addition, 13-cis RA treatment results in an increase of MMP-1 mRNA stability. These data demonstrate that RA stimulates MMP-1 gene expression in human pancreatic carcinoma cells by transcriptional and post-transcriptional mechanisms.

Differential expression of aggrecanase and matrix metalloproteinase activity in chondrocytes isolated from bovine and porcine articular cartilage

The release of aggrecan catabolites from cartilage is an early event in the pathogenesis of degenerative joint diseases. The enzymes involved in this process are unknown, controversial, and the subject of intense investigation. In this paper we have utilized a recombinant substrate containing the interglobular domain (IGD) of aggrecan to study specifically aggrecanase versus matrix metalloproteinase (MMP) catabolism in this domain of aggrecan. Our studies have shown that (i) there are species differences in the expression of latent versus active MMP activity on the aggrecan IGD; (ii) interleukin-1alpha exposure induces both aggrecanase and MMP activities, whereas retinoic acid induces only aggrecanase activity and inhibits the MMP activity on the aggrecan IGD; (iii) activators of latent MMP activity (p-aminophenylmercuric acetate and trypsin) significantly reduce aggrecanase activity; (iv) the time course of the appearance of aggrecanase versus the MMP catabolism of aggrecan IGD differs; (v) aggrecanase is a protease with metalloprotease characteristics; however (vi) the physiological (tissue) inhibitors of MMPs show weak inhibition (TIMP-1) or no inhibition (TIMP-2) of aggrecanase activity. Collectively, these studies show that aggrecanase and MMP catabolism of the aggrecan IGD are independent and uncoupled.

The nuclear receptor corepressor SMRT inhibits interstitial collagenase (MMP-1) transcription through an HRE-independent mechanism

Nuclear receptors inhibit synthesis of collagenase-1 (matrix metalloproteinase-1; MMP-1), an enzyme that degrades interstitial collagens and contributes to joint pathology in rheumatoid arthritis. SMRT (Silencing Mediator for Retinoid and Thyroid hormone receptors) mediates the repressive effect of nuclear receptors at hormone responsive elements (HREs), prompting us to investigate whether this co-repressor could also regulate transcription of MMP-1, which lacks any known HREs. We find that primary synovial fibroblasts express SMRT. When over-expressed by transient transfection, SMRT inhibits MMP-1 promoter activity induced by interleukin-1 (IL-1), phorbol phorbol myristate acetate (PMA) or v-Src. SMRT apparently inhibits MMP-1 gene expression by interfering with one or more transcriptional elements clustered in a region between -321 and +63. We conclude that SMRT negatively regulates MMP-1 synthesis through a novel, HRE-independent mechanism that involves proximal regions of the MMP-1 promoter.

Inhibition of rabbit collagenase (matrix metalloproteinase-1; MMP-1) transcription by retinoid receptors: evidence for binding of RARs/RXRs to the -77 AP-1 site through interactions with c-Jun

Treatment of synovial fibroblasts with retinoic acid (RA) decreases their expression of collagenase (matrix metalloproteinase-1 or MMP-1), an enzyme that degrades interstitial collagens and contributes to the pathology of rheumatoid arthritis. This inhibition results, at least in part, from RA-induced decreases in the mRNA for the transactivators Fos and Jun (with concominant increases in RAR mRNA) and by sequestration of Fos/Jun by RARs/RXRs. Previously, we provided evidence that retinoid receptors are also present in complexes that bind to fragments of rabbit MMP-1 promoter DNA containing an AP-1 site at -77 (Pan et al., 1995, J. Cell. Biochem., 57:575-589). However, it was unclear whether RARs and retinoid X receptors (RXRs) were binding directly to the DNA or indirectly through another protein. We now use a sensitive MMP-1 promoter/luciferase reporter construct to confirm the transcriptional role of the AP-1 site at -77. In addition, with electrophoretic mobility shift analyses (EMSAs), antibody "supershifts" and DNAase 1 footprinting, we examine the interaction of retinoid receptors and AP-1 protein on the MMP-1 promoter. We demonstrate that RARs, RXRs, and c-Jun form a complex at the AP-1 site in which c-Jun binds directly to the DNA and apparently tethers the retinoid receptors to the complex. We conclude that retinoid receptors/AP-1 protein interactions at the DNA may provide an additional means of controlling collagenase gene transcription by retinoids.

Regulation of membrane-type matrix metalloproteinase-1 expression by growth factors and phorbol 12-myristate 13-acetate

Overexpression of membrane-type matrix metalloproteinase (MT-MMP-1) results in the activation of both endogenous and exogenous 72-kDa gelatinase. To understand the effects of MT-MMP-1 on 72-kDa gelatinase activation, we analyzed its expression in human fibroblasts and HT-1080 fibrosarcoma cells. Both cell types expressed the MT-MMP-1 mRNA constitutively at a considerable level and treatment of cells with PMA enhanced the expression about 2-3-fold. Concanavalin A treatment increased MT-MMP-1 mRNA levels in fibroblasts about 4-fold. Induction of MT-MMP-1 by phorbol 12-myristate 13-acetate (PMA) required protein synthesis as shown by cycloheximide inhibition. The induction was also inhibited by dexamethasone. Analysis of MT-MMP-1 mRNA stability using actinomycin D indicated that the half-life was rather long and not affected by PMA, suggesting transcriptional regulation. Only HT-1080 cells had significant 72-kDa gelatinase processing activity after treatment with PMA or concanavalin A, while fibroblasts were virtually negative. Immunoblotting analysis of fibroblast lysates indicated that MT-MMP-1 was present mainly in a 60-kDa form. PMA and concanavalin A caused 2-4-fold increases in its protein levels, while in HT-1080 cells PMA, concanavalin A, or overexpression of MT-MMP-1 did not significantly enhance the level of the 60-kDa protein. Instead, an immunoreactive, proteolytically processed 43-kDa form was observed, and its appearance correlated to 72-kDa gelatinase processing activity. Thus 72-kDa gelatinase activation, while enhanced by MT-MMP-1 expression, needs additional co-operating factors.

Calcium ionophores decrease pericellular gelatinolytic activity via inhibition of 92-kDa gelatinase expression and decrease of 72-kDa gelatinase activation

To understand the roles of intracellular calcium levels on gelatinase/type IV collagenase expression, we analyzed the effects of calcium ionophores on the expression of 92- and 72-kDa gelatinases (MMP-9 and MMP-2) in human fibrosarcoma cells (HT-1080). Calcium ionophores ionomycin and A23187 reduced the levels of pericellular gelatinolytic activity in both untreated and phorbol 12-myristate 13-acetate (PMA) or tumor necrosis factor-alpha (TNF alpha)-stimulated cells as determined by degradation of radiolabeled gelatin. Gelatin zymography and immunoblotting revealed a dose-dependent decrease in the levels of secreted 92-kDa gelatinase, which was paralleled by a decrease of its mRNA. Treatment of cells with thapsigargin caused similar decreases of 92-kDa gelatinase mRNA and protein. The decrease of 92-kDa gelatinase expression was due to lower transcription rate as determined by transfection assays with 92-kDa gelatinase/luciferase construct. The expression of 72-kDa gelatinase was only slightly decreased by ionophores. Treatment of HT-1080 cells with PMA, TNF alpha, or concanavalin A resulted in the conversion of 72-kDa gelatinase proenzyme to its presumed 64- and 62-kDa active forms as determined by gelatin zymography and immunoblotting. Simultaneous treatment with the ionophores or thapsigargin resulted in inhibition of PMA-induced gelatinase activation. The expression of membrane-type matrix metalloproteinase, a potential activator of 72-kDa gelatinase, was not affected by ionophores. The results indicate that calcium ionophores decrease gelatinolysis by repressing both the expression of 92-kDa gelatinase and the activation of the 72-kDa gelatinase.

Heat shock of human synovial and dermal fibroblasts induces delayed up-regulation of collagenase-gene expression

We investigated the effect of heat shock on the expression of the collagenase gene in normal human synovial and dermal fibroblasts. Heat shock (42-44 degrees C for 1 h) caused a marked increase in heat-shock protein 70 (HSP-70) mRNA levels, followed by a delayed increase in collagenase mRNA levels, in both cell types. Pretreatment with cycloheximide had no effect on the heat-shock-induced increase in HSP-70 mRNA expression, but abrogated the induction of collagenase mRNA during the recovery. To study the mechanisms of collagenase-gene induction by heat shock, the transcriptional activity of a collagenase-promoter-driven chloramphenicol acetyltransferase (CAT) reporter gene was examined in transient transfection experiments. Heat shock was followed by a > 2-fold increase in CAT activity driven by a 3.8 kb fragment of the collagenase promoter, or by a construct containing an AP-1 binding site. A mutation in the AP-1 binding site abolished the effect of heat shock. Electrophoretic-mobility-shift assays revealed a marked increase in DNA-binding activity specific for the AP-1 binding site in nuclear extracts prepared from synovial fibroblasts recovering from heat shock. These results indicate that heat shock causes a delayed increase in collagenase-gene expression in human fibroblasts, and suggests that this stimulation involves, at least in part, transcriptional activation through an AP-1 binding site. Heat shock appears to initiate a programme of cellular events resulting in collagenase-gene expression, and therefore may contribute to connective-tissue degradation in disease states.

The role of cyclic-3',5'-adenosine monophosphate in prostaglandin-mediated inhibition of basic calcium phosphate crystal-induced mitogenesis and collagenase induction in cultured human fibroblasts

Synovial fluid basic calcium phosphate (BCP) crystals are associated with severe destructive arthropathies characterised by synovial proliferation and non-inflammatory degradation of intra-articular collagenous structures. BCP crystals stimulate fibroblast and chondrocyte mitogenesis, metalloprotease secretion and prostaglandin production. As a tissue protective effect of prostaglandins has been suggested, we recently studied the effect of PGE1 on BCP crystal-induced mitogenesis and collagenase mRNA accumulation in human fibroblasts (HF). We demonstrated a dose-dependent inhibition of BCP crystal-induced mitogenesis and collagenase mRNA accumulation. The mechanism of PGE1 inhibition of BCP crystal-induced mitogenesis and collagenase mRNA accumulation was therefore explored. PGE1 (100 ng/ml) increased HF intracellular cAMP 40-fold over control. BCP alone caused no such change but inhibited the PGE1-induced increase in intracellular cAMP by at least 60%. The PGE1-induced increase in intracellular cAMP was also blocked by the adenyl cyclase inhibitor, 2',5'-dideoxyadenosine (ddA) (10 microM) and ddA reversed the PGE1-mediated inhibition of BCP crystal-induced mitogenesis. Dibutyryl cAMP also inhibited BCP crystal-induced mitogenesis in a concentration-dependent manner. Agents which increase intracellular cAMP levels such as the adenyl cyclase activator forskolin and the phosphodiesterase, inhibitor 3-isobutyl-1-methylxanthine (IBMX) mimicked the effect of PGE1 on HF collagenase mRNA levels. PGE1 inhibits the biologic effects of BCP crystals through the cAMP signal transduction pathway and such inhibition may have significant therapeutic implications.

The secretion of the tissue inhibitor of metalloproteinases (TIMP) by human synovial fibroblasts is modulated by all-trans-retinoic acid

The matrix metalloproteinases are a family of enzymes involved in the turnover of the connective tissues. The regulation of these enzymes is complex, involving the control of synthesis, the activation of proenzyme forms and the presence of specific inhibitors. Retinoids have been reported to inhibit the production of metalloproteinases by human and rabbit synovial fibroblasts and by human skin fibroblasts. The production of the highly specific tissue inhibitor of metalloproteinases (TIMP) by connective tissue cells may be crucial in the regulation of connective tissue breakdown and this present study was undertaken to determine if retinoic acid (RA) could modulate TIMP and collagenase production by synovial fibroblasts. The results show that RA at concentrations from 10(-7) to 10(-5) M significantly stimulated the secretion of TIMP by two of three human synovial cell lines. The effect of mononuclear cell factor (MCF) on TIMP and collagenase levels was also investigated. The apparent reduction of collagenase levels in the presence of RA, could result from a failure to accurately measure this enzyme in the presence of increased levels of TIMP.

Localization of collagenase mRNA in rheumatoid arthritis synovium by in situ hybridization histochemistry

Collagenase has been implicated as playing an important role in the connective tissue destruction that is a major feature of rheumatoid arthritis. Numerous cell types in the hyperplastic rheumatoid synovium are capable of synthesizing collagenase. Past studies have used predominantly synovial fibroblasts in culture as a model system for the regulation of collagenase production, but the major cellular source of the enzyme in vivo has not been determined. Using the techniques of in situ hybridization histochemistry and indirect immunofluorescence, we determined the cellular source of collagenase in frozen sections of human synovium. Collagenase mRNA production was localized to cells along the synovial lining layer in rheumatoid arthritis. These were identified as the macrophage-like Type A synovial lining cells by immunofluorescence with antibody LeuM3. Endothelial cells, fibroblasts, and T and B lymphocytes were devoid of detectable collagenase mRNA. Synovial tissue sections from patients with osteoarthritis and trauma did not contain detectable collagenase mRNA. These data identify the Type A macrophage-like synovial lining cell as the primary source of collagenase mRNA in vivo in the rheumatoid arthritis synovium and, potentially, as a major effector cell in the tissue destruction of the disease.

Molecular oncogenetics of metastasis

Metastasis is a complex non-stochastic process that is most likely the result of genetic and epigenetic interactions of a wide variety of genes. The search for a single gene which can encompass such a pleiotropic response as to account for the observed phenotypic characteristics of metastatic tumour populations has been unsuccessful. Particular studies involving gene transfection, subtractive hybridisation and cell fusion are beginning to identify specific genes which contribute to metastasis in some cell types. However, such analyses are complicated by the inherent genetic instability and phenotypic heterogeneity present in tumour populations. A more detailed understanding of the metastatic process may require an abandoning of current generalised approaches to metastasis in favour of concentrating on key components of the metastatic cascade such as adhesion and invasion.

Heat shock of rabbit synovial fibroblasts increases expression of mRNAs for two metalloproteinases, collagenase and stromelysin

Two metalloproteinases, collagenase and stromelysin, are produced in large quantities by synovial fibroblasts in individuals with rheumatoid arthritis. These enzymes play a major role in the extensive destruction of connective tissue seen in this disease. In this study, we show that heat shock of monolayer cultures of rabbit synovial fibroblasts increases expression of mRNA for heat shock protein 70 (HSP-70), and for collagenase and stromelysin. We found that after heat shock for 1 h at 45 degrees C, the mRNA expression for HSP-70 peaks at 1 h and returns to control levels by 3 h. Collagenase and stromelysin mRNA expression is coordinate, reaching peak levels at 3 h and returning to control levels by 10 h. The increase in mRNA is paralleled by an increase in the corresponding protein in the culture medium. 3 h of heat shock at a lower temperature (42 degrees C) is also effective in inducing collagenase and stromelysin mRNAs. Concomitant treatment with phorbol myristate acetate (PMA; 10(-8) or 10(-9) M) and heat shock is not additive or synergistic. In addition, all-trans-retinoic acid, added just before heat shock, prevents the increase in mRNAs for collagenase and stromelysin. Our data suggest that heat shock may be an additional mechanism whereby collagenase and stromelysin are increased during rheumatoid arthritis and perhaps in other chronic inflammatory stress conditions.

A model to account for the effects of oncogenes, TPA, and retinoic acid on the regulation of genes involved in metastasis

We have postulated that signals from the microenvironment can induce shifts in tumor cell phenotypes and that microenvironmental factors are therefore important for cancer metastasis. In this article we expand on this hypothesis and propose a model to explain (a) how extracellular signals can lead to changes in tumor phenotypes, and (b) how cytoplasmic oncogenes, which influence signal transducing pathways as well as nuclear oncogenes regulating gene expression via DNA binding transacting factors, might affect metastatic competence.

Cloning of a complementary DNA for rabbit proactivator. A metalloproteinase that activates synovial cell collagenase, shares homology with stromelysin and transin, and is coordinately regulated with collagenase

Rabbit proactivator is a neutral metalloproteinase that activates another metalloproteinase, procollagenase, and degrades noncollagenous matrix. We describe the construction of an activator complementary DNA (cDNA) clone, which is 1.9 kb, that selects a 2.1-kb messenger RNA (mRNA) in Northern blot hybridizations. Nucleic acid sequence studies of the activator cDNA indicate 1) that it encodes protein Mr 53,881, 2) that this protein exhibits approximately 80% homology with rat transin, an oncogene-induced protein with a previously unknown function, and 3) that, in the first 172 residues, it is virtually identical to the rabbit metalloproteinase, stromelysin. Homology between rabbit activator and human skin collagenase is approximately 50%. Activator and collagenase mRNA are coordinately regulated; untreated cultures of rabbit synovial fibroblasts produce low levels of each protein, but addition of phorbol myristate acetate (10(-8)M) results in an increase in mRNA for both proteins by 2.5-5 hours. Adding all-trans-retinoic acid (10(-6)M) or dexamethasone (10(-7)M) to phorbol-stimulated cells coordinately suppresses both activator and collagenase mRNA. Our data suggest the existence of coordinately regulated metalloproteinases that are important in the modulation of connective tissue metabolism.

Regulation of human peripheral blood monocyte collagenase by prostaglandins and anti-inflammatory drugs

Macrophages, which produce the collagenolytic enzyme collagenase, are commonly found at sites of connective tissue destruction in chronic inflammatory lesions. Since tissue macrophages are derived from circulating peripheral blood monocytes, we used these less-differentiated, more readily available cells to examine the production and regulation of collagenase. Human monocytes, isolated in large quantities by counterflow centrifugal elutriation, were shown to produce substantial amounts of collagenase when stimulated by concanavalin A (Con A) and to a lesser extent with lipopolysaccharide, while unstimulated monocyte cultures produced negligible collagenase. Collagenase was detected in the culture media within the first 24 hr of culture after activation with peak production at 48 hr. Analysis of the intracellular regulation of collagenase revealed that synthesis of this enzyme required a prostaglandin (PGE2)-dependent step since indomethacin-inhibited enzyme synthesis was reversed by PGE2. Additionally, dibutyryladenosine cyclic monophosphate (dBcAMP) restored collagenase synthesis in indomethacin-blocked cultures, indicating a PGE2-dependent generation of cAMP requirement for collagenase production similar to that demonstrated in experimental animals systems. In additional studies, anti-inflammatory drugs which are known to modulate connective tissue destruction were analyzed for their influence on monocyte-derived collagenase. Dexamethasone, colchicine or retinoic acid all inhibited collagenase synthesis by monocytes in a dose-dependent manner although the effect of these drugs on monocyte PGE2 synthesis differed. Dexamethasone inhibited PGE2 synthesis, which resulted in the suppression of collagenase. However, PGE2 production was unaffected by colchicine whereas retinoic acid caused a significant increase in PGE2 levels. Inhibition of collagenase synthesis by dexamethasone, but not colchicine or retinoic acid, could be reversed by PGE2 or phospholipase A2. These findings provide insight into the intracellular events regulating monocyte collagenase synthesis and also implicate monocytes as a target of anti-inflammatory agents which ameliorate connective tissue degradation associated with chronic inflammatory lesions.

Molecular cloning of human synovial cell collagenase and selection of a single gene from genomic DNA

We used a subclone of a rabbit genomic clone for collagenase that cross-hybridizes with human synovial cell messenger RNA (mRNA) to identify a human collagenase complementary DNA (cDNA) clone. The human cDNA clone is 2.1 kilobases (kb) and selects a mRNA transcript of approximately the same size from primary cultures of rheumatoid synovial cells that produce collagenase, but no mRNA is selected from control (nonproducing) synovial fibroblasts. Restriction enzyme analysis and DNA sequence data indicate that our cDNA clone is full length and that it is identical to that recently described for human skin fibroblast collagenase. The cDNA clone identified a single collagenase gene of approximately 17 kb from blots of human genomic DNA. The identity of human skin and synovial cell collagenase and the ubiquity of this enzyme and of its substrates, the interstitial collagens types I, II, and III, imply that common mechanisms controlling collagenolysis throughout the human body may be operative in both normal and disease states.

Characterization of rabbit genes for synovial cell collagenase

To provide tools for understanding collagenase gene expression in rheumatoid arthritis, we have isolated and characterized genomic clones for rabbit synovial cell collagenase. These clones represent 2 types of collagenase gene, at least 1 of which is transcribed in synovial fibroblasts. By examining the rabbit genome in situ, we provide evidence that there are only 2 different synovial cell collagenase genes found in a haploid genome. Amplification of these genes is not a mechanism for collagenase messenger RNA induction by phorbol esters.

Half-life of synovial cell collagenase mRNA is modulated by phorbol myristate acetate but not by all-trans-retinoic acid or dexamethasone

As part of our studies on the mechanisms controlling the synthesis of the neutral proteinase collagenase by rabbit synovial cells, we used a cDNA clone to measure total collagenase mRNA levels and to determine mRNA half-life. Phorbol myristate acetate was used to induce collagenase synthesis while all-trans-retinoic acid and dexamethasone were used to inhibit it. Cells stimulated with phorbol myristate acetate contained substantial amounts of collagenase mRNA, but cells treated with all-trans-retinoic acid or dexamethasone contained decreased amounts of collagenase mRNA which correlated well with levels of collagenase protein. Studies on mRNA half-life showed that the t1/2 for total poly(A+) RNA was about 25 h, while that of collagenase varied from as short as 12 h to as long as 36 h. The half-life was not affected by treatment with all-trans-retinoic acid or dexamethasone but was affected by the level of induction of collagenase mRNA: the greater the amount of collagenase mRNA induced, the longer the t1/2. We conclude that our data are consistent with the hypothesis that retinoic acid and dexamethasone act at the level of transcription to decrease collagenase production and the increased level of collagenase mRNA resulting from stimulation with phorbol esters is, in part, due to increased stability of the induced collagenase mRNA.

Suppression of hyaluronic acid synthesis in synovial organ cultures by corticosteroid suspensions

The effects of sparingly soluble corticosteroid suspensions prepared for intraarticular therapy, of their vehicles, and of hydrocortisone on synovial hyaluronic acid (HA) synthesis were compared in organ cultures of normal canine villous synovium. Both hydrocortisone and the corticosteroid suspensions suppressed HA synthesis, as did 2 vehicle components, polysorbate 80 and myristyl-gamma-picrolinium chloride. Cultures of synovium from joints of dogs which, 1 day previously, had been injected with methylprednisolone acetate suspension synthesized less HA than did control cultures from noninjected joints of the same animals. The results indicate that suppression of HA synthesis is a mechanism by which these drugs can act to reduce joint HA content and promote resolution of synovial effusions.

Retinoic acid stimulates degradation of interstitial collagen fibrils by rat mucosal keratinocytes in vitro

A large body of evidence suggests that retinoids modulate the phenotypic expression of epithelial cells of skin and mucous membranes. The purpose of this study was to investigate the effect of retinoic acid on keratinocyte-mediated collagen breakdown. Keratinocytes derived from the ventral surfaces of the tongues of 4-6 week-old male Wistar rats were established in culture under conditions which are restrictive to growth of fibroblasts, and they were eventually cloned by limiting dilution. The cells were seeded (100,000 cells/cm2) in dishes coated with 3H-labeled, reconstituted type I collagen fibrils and incubated in serum-free medium over a 3-5 day period. Dissolution of the collagen fibrils was monitored by the release of radioactivity to the culture medium. Unstimulated cells metabolized the collagen rather slowly, but addition of retinoic acid in concentrations from 10(-6)M to 10(-8)M resulted in marked acceleration of the degradative process, with complete solubilization of the collagen fibrils in four or five days. The effect of 10(-6)M retinoic acid was of the same order of magnitude as that obtained by addition of a proteolytic activating system either in the form of plasmin or of plasminogen, which is converted to catalytic plasmin by endogenous activators. The effects of retinoic acid and plasminogen/plasmin, however, were not additive. Keratinocytes rendered vitamin A-deficient by cultivation in sera from deficient rats were clearly less effective in degrading the collagen substrate than were "sufficient" cells. Addition of retinoic acid (10(-7)M) enhanced collagen breakdown in both sets of cultures and partially restored the collagenolytic activity of the deficient cells.

Coregulation of collagenase and collagenase inhibitor production by phorbol myristate acetate in human skin fibroblasts

Phorbol myristate acetate (PMA), a tumor promotor known to stimulate collagenase production in fibroblasts and endothelial cells, was examined with regard to its ability to regulate the expression of the collagenase inhibitor secreted by human skin fibroblasts. Confluent human skin fibroblasts were incubated with concentrations of PMA ranging from 10(-11) to 10(-7) M, and the conditioned medium was analyzed by enzyme-linked immunosorbent assay for both immunoreactive collagenase and collagenase inhibitor. PMA stimulated the production of both collagenase and collagenase inhibitor in several cell lines to maximal rates that were very similar, 300 to 350 vs 230 to 330 pmol 10 micrograms DNA-1 48 h-1, respectively. Due to differences in the basal levels of expression of these proteins, such rates reflected a two- to sevenfold stimulation in collagenase production, in comparison to a more uniform two- to threefold enhancement in inhibitor synthesis. Production of inhibitor was 50% of maximal at 7 X 10(-9) M and maximal at 10(-7) M phorbol. This concentration-dependent effect was very similar to that observed for collagenase expression. Total protein synthesis by the phorbol-conditioned cells, as studied by incorporation of [3H]leucine into newly synthesized protein, was not significantly increased, nor was cellular DNA content. The onset of the effect of PMA on inhibitor production occurred between 4 and 8 h, was maximal by 8 h, and continued undiminished for at least another 64 h. After the first 8 h, inhibitor production continued at a roughly constant rate of approximately 10 pmol 10 micrograms DNA-1 h-1. Interestingly, following the removal of phorbol from culture medium, such fibroblasts continued to produce increased quantities of inhibitor protein for at least 72 h. Metabolic labeling studies in which fibroblasts were exposed to [3H]leucine followed by immunoprecipitation using inhibitor-specific antibody suggested that stimulation of inhibitor production by PMA was mediated via an increased synthesis of new inhibitor protein. Therefore, in response to the tumor promoter, PMA collagenase and collagenase inhibitor expression by human skin fibroblasts appear to be coregulated.

Glucocorticoid modulation of collagenase expression in human skin fibroblast cultures. Evidence for pre-translational inhibition

Glucocorticoids inhibit collagenase accumulation in the medium of human skin explant cultures. To examine the mechanism for this process, skin fibroblasts were placed in serum-free medium containing various steroids. Dexamethasone produced a dose-dependent inhibition of trypsin-activatable collagenase in the culture medium with maximal inhibition of approx. 85% at 10(-6) M. Dexamethasone failed to inhibit collagenase activity directly. The decrease in activity in the medium was paralleled by a decrease in immunoreactive protein, suggesting inhibition of enzyme synthesis. The specificity of the effect was shown in two ways. At 10(-6) M steroid, only dexamethasone and hydrocortisone were inhibitory; estradiol, progesterone and testosterone produced less than 10% inhibition. In biosynthetic studies, exposure to 10(-7) M dexamethasone for 24 h produced approx. 50% inhibition of collagenase synthesis but caused no greater than 10% inhibition of total protein synthesis. The T1/2 for achieving the effect was approx. 16 h after initial exposure to dexamethasone. These kinetics were parallel to the inhibition caused by actinomycin D and cordycepin, two inhibitors of transcription, but were longer than that caused by cycloheximide (T 1/2 less than 3 h). To examine this process, cells were cultured in the presence or absence of 10(-6) M dexamethasone prior to harvesting mRNA for cell-free translation. In each case the inhibition or enzyme activity in the intact cells was paralleled by a reduction in translatable collagenase mRNA from the same cells. At the same time, there was no significant inhibition of total protein translation by the steroid. These data suggest that glucocorticoids regulate collagenase synthesis at a pre-translational level, possibly through inhibition of transcription.

Regulation of macrophage collagenase, prostaglandin, and fibroblast-activating-factor production by anti-inflammatory agents: different regulatory mechanisms for tissue injury and repair

Activation of macrophages results in the production of tissue destructive mediators and enzymes including prostaglandins (PGE) and collagenase. In addition, activated macrophages also generate mediators which enhance connective tissue formation through their effects on fibroblast growth. To determine whether the pro-inflammatory mediators and the mediator(s) involved in tissue repair are under the same regulatory control, guinea pig macrophage cultures were treated with various pharmacologic agents and their supernatants monitored for biologic activity. The nonsteroidal anti-inflammatory agent, indomethacin, and the glucocorticoid, dexamethasone, at pharmacologic concentrations inhibited not only prostaglandin synthesis (greater than 90%) but also the production of collagenase (greater than 90%). Colchicine, a microtubule disruptive agent, but not the inactive form, lumicolchicine, markedly diminished the production of collagenase independently of prostaglandin synthesis. In contrast to the inhibitory effects of these anti-inflammatory agents on PGE and collagenase production, indomethacin did not inhibit the production of macrophage-derived fibroblast-activating factor (FAF). Furthermore, dexamethasone at pharmacologic doses did not inhibit FAF production. Colchicine not only did not inhibit FAF, but frequently enhanced the appearance of FAF In the macrophage cultures. Thus, it appears that regulation of the production of PGE and collagenase is different than the regulation of FAF synthesis and therefore the production of these mediators can be differentially modulated. Such a dissociation may provide a basis for mononuclear cell-mediated fibroblast growth and tissue repair to occur independently of the release of PGE2 and collagenase and even following anti-inflammatory drug therapy.

Autoregulation of collagenase production by a protein synthesized and secreted by synovial fibroblasts: cellular mechanism for control of collagen degradation

Conditioned medium taken from cultures of resting rabbit synovial fibroblasts contained a protein that prevented the synthesis of the neutral proteinase collagenase. Conditioned medium was concentrated 10-fold and placed on cultures of rabbit synovial fibroblasts along with an inducer of collagenase (phorbol myristate acetate or latex particles) and [3H]leucine. Collagenase production was measured by immunoprecipitation of culture medium with monospecific antibody. Gel filtration showed that the inhibitory factor had MrS of 12,500, 25,000-50,000, and 150,000, suggesting that the protein may exist as aggregates. Activity was destroyed by boiling, by trypsin, and by dithiothreitol. Production of the inhibitory protein was prevented by cycloheximide. Isoelectric focusing purified the protein 100- to 150-fold and revealed pIs in the range of 3.2-3.7. Glycosylation was demonstrated by binding to Con A-Sepharose. Our data indicate that rabbit synovial fibroblasts autoregulate collagenase production and suggest that the low levels of collagenase seen in resting cultures result from an active suppression of collagenase synthesis.

Dose-dependent suppression by the synthetic retinoid, 4-hydroxyphenyl retinamide, of streptococcal cell wall-induced arthritis in rats

We studied the effects of oral administration of the retinoid, 4-hydroxyphenyl retinamide (4-HPR), on group A streptococcal cell wall-induced polyarthritis in the rat, a model characterized initially by exudative inflammation of peripheral joints followed by chronic proliferative/erosive synovitis. Experimental arthritis was induced in female LEW/N rats by i.p. injection of streptococcal cell walls in saline (15 micrograms/g body weight). Depending upon the experiment, continuous daily oral administration of the retinoid was begun either 14 days prior to induction of the disease, at the time of cell wall administration and/or 11 days and 31 days after cell wall injection. Dosage was either 1 or 2 mmol 4-HPR/kg of chow. During the course of the disease, severity of clinical illness was assessed by determination of clinical severity index, by histological or radiologic examination, and by measurement of production in vitro of collagenase and prostaglandin E2 by excised synovial tissue. In rats fed the retinoid prior to cell wall injection, both the acute and the chronic responses were suppressed. In rats given the retinoid at the time of cell wall injection, the acute inflammatory response was only partially suppressed on the diet containing 2 mmol 4-HPR/kg chow, but the chronic disease was impressively inhibited in a dose dependent manner. Similarly, in animals with established disease, the drug was also effective; however, the more advanced the illness, the less effective the drug. Clinical observations were paralleled by the histological, radiographical and biochemical analyses. Treated animals showed far less synovial proliferation and joint destruction, and synovial tissues taken from these rats produced lesser amounts of collagenase and prostaglandin E2. No significant toxicity of the retinoid was noted. We conclude that oral administration of 4-HPR suppresses, in a dose and time dependent manner, both the acute and chronic stages of streptococcal cell wall-induced arthritis in rats without apparent significant toxicity. Our data suggest that studies of the effects of this retinoid on patients with chronic inflammatory synovitis are warranted.

Regulation of guinea pig macrophage collagenase production by dexamethasone and colchicine

Previous studies have demonstrated that exposure of guinea pig macrophages to a primary signal, such as lipopolysaccharide (LPS), stimulates the synthesis of prostaglandin E2 (PGE2) which, in turn, elevates cAMP levels resulting in the production of the enzyme, collagenase. The potential of regulating the biochemical events in this activation sequence was examined with the anti-inflammatory agents dexamethasone and colchicine, which suppress the destructive sequelae in chronic inflammatory lesions associated with the degradation of connective tissue. The addition of dexamethasone with LPS to macrophage cultures resulted in a dose-dependent inhibition of PGE2 and collagenase production, which was reversed by the exogenous addition of phospholipase A2. Collagenase production was also restored in dexamethasone-treated cultures by the addition of products normally produced as a result of phospholipase action, such as arachidonic acid, PGE2 or dibutyryl-cAMP. Since the effect of dexamethasone was thus linked to phospholipase A2 inhibition, mepacrine, a phospholipase inhibitor, was also tested. Mepacrine, like dexamethasone, caused a dose-dependent inhibition of PGE2 and collagenase. In addition to corticosteroid inhibition, colchicine was also found to block collagenase production. However, this anti-inflammatory agent had no effect on PGE2 synthesis. Colchicine was effective only when added at the onset of culture and not 24 h later, implicating a role for microtubules in the transmission of the activation signal rather than enzyme secretion. The failure of lumicolchicine to inhibit collagenase activity provided additional evidence that microtubules are involved in the activation of macrophages. These findings demonstrate that dexamethasone and colchicine act at specific steps in the activation sequence of guinea pig macrophages to regulate collagenase production.

Isolation of a collagenase cDNA clone and measurement of changing collagenase mRNA levels during induction in rabbit synovial fibroblasts

To facilitate our studies on the mechanisms controlling collagenase production at a molecular level in rabbit synovial fibroblasts, we have constructed a cDNA library using mRNAs isolated from cells induced with crystals of monosodium urate monohydrate. We have screened this library with cDNA probes made from induced and control mRNA populations. From among 30 clones that hybridized preferentially to the induced-cell probe, 4 contained collagenase sequences. The largest, a clone of 650 base pairs, was identified by its ability to hybrid select a mRNA that could be translated in a cell-free system into a product that was precipitable with monospecific antibody to collagenase. Using this clone to probe blots of RNA from induced cells, we detected the appearance of a collagenase mRNA of 2.7 kilobases within 5 hr of addition of urate. The level of collagenase mRNA continued to increase for 35-40 hr, when it was 60 to 90 times more abundant in induced cells than in control cells. The increase in mRNA levels correlated with an increase in immunoreactive collagenase protein that was detectable in culture medium by 10 hr.

The effects of razoxane (ICRF 159) on the production of collagenase and inhibitor (TIMP) by stimulated rabbit articular chondrocytes

Monolayer cultures of rabbit chondrocytes were stimulated to produce collagenase with conditioned medium from human peripheral blood mononuclear cells (MCM), and the ability of Razoxane to modulate the production of collagenase and specific tissue inhibitor of metalloproteinases (TIMP) was studied. Collagenase production was inhibited and TIMP increased by Razoxane, in a dose-dependent manner, when cells were treated daily for 3 days. Over this period the effect of Razoxane was progressive; 50 micrograms/ml or less had no effect at day 1 but 50 micrograms/ml was effective by day 3. The effectiveness of Razoxane was inversely related to the degree of MCM stimulation and the confluency of the culture. On removal of the drug, chondrocytes stimulated with MCM recovered their ability to produce collagenase, and TIMP production returned to near normal. The results suggest that the ability of Razoxane to reduce collagenase and increase TIMP production may correlate with its effectiveness in treating psoriatic arthritis.

The cartilage-resorbing protein catabolin is made by synovial fibroblasts and its production is increased by phorbol myristate acetate

Pig synovial fibroblasts were shown to produce a protein that caused live cartilage to resorb its proteoglycan matrix in vitro. Fibroblasts were obtained either from synovial tissue digest or by allowing them to grow out of explants. The population derived from the digests was homogeneous and free of macrophage-like cells after two passages, but was still producing the cartilage-resorbing protein after seven passages. The active protein was found to have Mr 20,000 on gell filtration, and pI 4.8 on isoelectric focussing in polyacrylamide gel. It was indistinguishable from a protein with the same activity from pig mononuclear leucocytes, which has been called catabolin. Production of the protein was increased if the synovial fibroblasts were cultured with the tumour promoter phorbol 12-myristate 13-acetate. Fibroblasts from other sources (joint capsule and peritoneum) also apparently made the protein. The possibility that catabolin is the same as interleukin-1 is discussed: if they are, then the results suggest that fibroblasts can make an interleukin-1-life protein.

Inflammation and collagenase production in rats with adjuvant arthritis reduced with 13-cis-retinoic acid

Oral administration of 13-cis-retinoic acid (40 or 160 milligrams per kilogram of body weight daily) significantly reduced the inflammation associated with developing and established adjuvant arthritis, an experimentally induced arthritis in rats that resembles human rheumatoid arthritis. The amount of collagenase secreted in tissue culture by adherent cells isolated from the inflamed joints of adjuvant rats treated with 13-cis-retinoic acid also decreased as compared to the amount secreted by cells from vehicle-treated adjuvant rats. Collagenase is important in the joint destruction accompanying rheumatoid arthritis. The successful use of retinoids in the treatment of this proliferative but nonmalignant disorder demonstrates a new application of these compounds.

Retinoic acid inhibition of collagenase and gelatinase expression in human skin fibroblast cultures. Evidence for a dual mechanism

Human skin fibroblast cultures have been employed to study the effects of a variety of vitamin A analogues (retinoids) on the expression of two enzymes involved in collagen degradation in the skin, collagenase and a gelatinolytic protease. In normal and recessive dystrophic epidermolysis bullosa fibroblast cultures, retinoic acid compounds were effective inhibitors of the accumulation of both enzymes in the culture medium with half-maximal inhibitions occurring at 0.25-1 microM for collagenase and at 3-6 microM for the gelatinolytic protease. Various retinoids exhibited differing degrees of inhibitory actions, so that at a 1 microM concentration, relative inhibitions were: 13-cis-retinoic acid greater than all-trans-retinoic acid greater than aromatic retinoid (Ro 10-9359) much greater than retinol. The retinoic acid-mediated decrease in collagenase activity was accompanied by a parallel decrease in immunoreactive collagenase protein, suggesting that the retinoic acids were acting to inhibit synthesis of the enzyme. However, an additional effect of these agents was encountered. Although the retinoids themselves had no direct collagenase inhibitory action, medium derived from cultures maintained in these retinoids showed direct inhibitory capacity which was dependent both on the concentration of retinoic acid and on the length of time in culture. The results suggest that the retinoic acids modulate collagenase in vitro by two mechanisms: by decreasing the synthesis of enzyme protein and by modulating the expression of an inhibitory molecule.

Effect of retinoic acid on the acylation of phospholipids in granulocytes in vitro

The influence of retinoic acid on the incorporation of [1-14C]palmitic acid and [1-14C]arachidonic acid into phospholipids was examined in guinea pig peritoneal granulocytes. All-trans-retinoic acid inhibited the incorporation of both fatty acids into phosphatidic acid and phosphatidylinositol. However, it stimulated the incorporation of both fatty acids into phosphatidylcholine but not other phospholipids. All-trans-retinoic acid was more effective than 13-cis-retinoic acid. The influence of all-trans-retinoic acid on the acylation of phospholipids was concentration-dependent with significant effect occurring at 2.1 microM. The loss of labeled fatty acids from prelabeled phospholipids and the transport of labeled fatty acids into granulocytes were not responsive to the presence of retinoic acid in the incubation media. These results suggest that retinoic acid may affect the activities of acyltransferases involved in the synthesis of phosphatidic acid, phosphatidylinositol and phosphatidylcholine.

Effects of all-trans-retinoic acid (retinoic acid) and 4-hydroxyphenylretinamide on synovial cells and articular cartilage

We studied the effects of two retinoids, naturally occurring all-trans-retinoic acid (retinoic acid) and the synthetic 4-hydroxyphenylretinamide (4-OH-PRT) on monolayer cultures of rabbit synovial fibroblasts and on explants of rabbit articular cartilage. Treatment of fibroblasts with phorbol myristate acetate (PMA; 10(-8) M) induced the synthesis and secretion of large amounts of collagenase: this was inhibited if the cells were treated with retinoic acid (10(-6) M) or dexamethasone (10(-7 M). Combined treatment with retinoic acid and the steroid prednisolone, at concentrations as low as 19(-10) M, gave an additive inhibition of collagenase production. Both retinoids inhibited collagenase production, but only 4-OH-PRT prevented the increase in prostaglandin E2 (PGE2) induced by PMA. Levels of plasminogen activator were also increased by treatment with PMA, and concomitant addition of either retinoid further enhanced this stimulation. Possible toxicity was assessed by measuring release of glycosaminoglycans (GAG) from explants of articular cartilage. Treatment with retinoic acid induced release of 80% of the total GAG, whereas treatment with 4-OH-PRT resulted in release of 40% of the total, a finding similar to that seen with untreated samples. 4-OH-PRT inhibited production of collagenase and PGE2 by rabbit synovial fibroblasts but was not toxic to articular cartilage.

Collagenase production by cloned populations of rabbit synovial fibroblasts

Monolayers of rabbit synovial fibroblasts treated experimentally with phorbol myristate acetate produce large amounts of collagenase and prostaglandin E2 and have been a suitable experimental model for the proliferative/destructive lesion of rheumatoid arthritis. We used X-irradiation to prolong the in vitro lifespan of these cells so that cloned populations could be studied. By a number of criteria, X-irradiation did not alter the cells to make them unrepresentative of synovial fibroblasts. With limiting dilution techniques, we simultaneously isolated three clones. These clones were shown to have different growth rates and to produce different amounts of collagenase and prostaglandin E2. Rates of protein synthesis, measured by incorporation of 3H-leucine, were similar for all three clones. Our data support the concept that particular populations of synovial cells may contribute selectively to the joint destruction seen in rheumatoid disease.