Transforming growth factor beta modulates the expression of collagenase and metalloproteinase inhibitor

Inflammatory and immunomodulatory roles of TGF-beta

Transforming growth factors (TGFs) are small polypeptides that were initially defined by their ability to induce transformation of non-neoplastic cells in culture. However, it has become increasingly clear that TGFs are not restricted in function to promoting cell growth. One type of transforming growth factor, TGF-beta, is a multifunctional molecule which has unique and potent effects on many target cells and tissues. In this article, Sharon Wahl, Nancy McCartney-Francis and Stephan Mergenhagen focus on the evolving role of TGF-beta in regulating inflammation, immune responses and tissue repair.

Identification and characterization of calcium-dependent metalloproteases in rat brain

We have begun to identify and characterize brain protease activities separated by and assayed in substrate-containing polyacrylamide gels. In the present report, we focus on four proteolytic activities identified from rat brain that are dependent on micromolar and millimolar Ca2+ concentrations for activity. In contrast to the previously described Ca2+-dependent neutral cysteine proteases (calpains), all four activities appear to be metalloproteases based on their inhibition by EDTA, EGTA, and 1,10-o-phenanthroline, but not by blockers of serine, cysteine, or aspartic proteases. In the presence of excess Ca2+ and the Zn2+-chelating inhibitor 1,10-o-phenanthroline, activity of the enzymes was reconstituted by addition of lower concentrations of Zn2+, and inhibited by higher Zn2+ concentrations. The four metalloproteases were designated MP-112, MP-92, MP-70, and MP-65 on the basis of their apparent molecular masses in kilodaltons. MP-70, the major activity detected, had an apparent kact for Ca2+ greater than 100 microM versus 10-25 microM for MP-65 and 50-100 microM for MP-92. MP-112 was a minor activity for which Ca2+ activation levels were not determined. MP-112, MP-70, and MP-65 were similar in being most active in the soluble fraction of 7-day neonate forebrain. In contrast, MP-92 activity was highest in the particulate fraction of adult forebrain. About half of the MP-92 activity and lower levels of the other three activities were still detectable in particulate fractions after detergent extraction of membrane, suggesting an association with cytoskeletal or other structural proteins.(ABSTRACT TRUNCATED AT 250 WORDS)

Transforming growth factor-beta causes partial inhibition of interleukin 1-stimulated cartilage degradation in vitro

We show that purified human transforming growth factor-beta (1-10ng/ml) inhibits interleukin 1-stimulated loss of proteoglycan from cartilage in vitro. Inhibition is incomplete, as interleukin 1 retains the ability to cause a dose dependent stimulation of proteoglycan release in the presence of high levels of transforming growth factor-beta (100ng/ml) although both basal and interleukin 1-stimulated levels can be reduced by up to 50 per cent. This observation, together with its ability to stimulate proteoglycan synthesis and to stimulate proteinase inhibitor production, suggests a possible role for transforming growth factor-beta in limiting cartilage proteoglycan loss in inflammatory conditions such as rheumatoid arthritis.

Collagenase gene expression in fibroblasts is regulated by a three-dimensional contact with collagen

Collagenase activity in fibroblasts is regulated by cytokines and the interaction with the extracellular matrix. In this study we demonstrate that fibroblasts cultured within a three-dimensional collagen gel show a strong induction of collagenase gene expression. In addition to increased de novo synthesis most of the secreted enzyme was found to be activated leading to a high collagenolytic activity and complete degradation of collagen matrices after removal of fetal calf serum. Collagen I gene expression was found to be reduced under these conditions. These data suggest a specific modulation of cellular metabolism in response to contact with a three-dimensional collagenous matrix resulting in the divergent regulation of collagen and collagenase.

Inhibitory effects of transforming growth factor-beta on laminin production and growth exhibited by endoderm-like cells derived from embryonal carcinoma cells

Previous studies have shown that two mouse embryonal carcinoma (EC) cell lines do not express cell surface receptors for transforming growth factor type-beta (TGF-beta) until they are induced to differentiate. To understand the effects of TGF-beta in this model system, we have examined the effects of TGF-beta on parietal endoderm-like cells derived from EC cells. We have determined that TGF-beta exerts three effects on these cells. TGF-beta inhibits proliferation of the parietal endoderm-like cells, and this occurs even in the presence of growth factors that stimulate their proliferation. TGF-beta also alters the morphology of the parietal endoderm-like cells by increasing their spreading. Moreover, the morphological effect of TGF-beta is observed in the presence of dibutyryl cyclic AMP (dbcAMP), which reduces the spreading of these cells. Lastly, TGF-beta, but not other growth factors, decreases the production of laminin by the parietal endoderm-like cells. This was unexpected since TGF-beta has been shown to increase the production of extracellular matrices in other systems. Thus, our findings indicate that parietal endoderm-like cells provide a useful system for broadening the study of TGF-beta. Furthermore, our findings provide additional support for the possibility that TGF-beta plays important roles during the early stages of mammalian development.

Cytokine-mediated proteolysis in tissue remodelling

Proteolytic enzymes play a key role in a variety of physiological processes in which the degradation of macromolecules is essential: angiogenesis, embryogenesis, bone and tissue remodelling, blood hemostasis and cell migration. The action of these enzymes is also crucial in the development of many pathological conditions such as wound healing, neoplasia, inflammation and arthritic disorders. The activity of proteases is negatively affected by specific protease-inhibitors. Various growth factors and other cytokines modulate the synthesis and secretion of both proteases and protease-inhibitors. The study of this regulation results in a better insight into (patho)physiology at the molecular level and promises to result in alternative treatment strategies.

Genes for extracellular-matrix-degrading metalloproteinases and their inhibitor, TIMP, are expressed during early mammalian development

Extracellular matrix (ECM) remodeling accompanies cell migration, cell-cell interactions, embryo expansion, uterine implantation, and tissue invasion during mammalian embryogenesis. We have found that mouse embryos secrete functional ECM-degrading metalloproteinases, including collagenase and stromelysin, that are inhibitable by the tissue inhibitor of metalloproteinases (TIMP) and that are regulated during peri-implantation development and endoderm differentiation. mRNA transcripts for collagenase, stromelysin, and TIMP were detected as maternal transcripts in the unfertilized egg, were present at the zygote and cleavage stages, and increased at the blastocyst stage and with endoderm differentiation. These data suggest that metalloproteinases function in cell-ECM interactions during growth, development, and implantation of mammalian embryos.

Regional localization of the TIMP gene on the human X chromosome. Extension of a conserved synteny and linkage group on proximal Xp

The gene encoding a tissue inhibitor of metallo-proteinases, TIMP, has previously been shown to be X-linked in both the human and mouse genomes. We have used a series of somatic cell hybrids segregating translocation and deletion X chromosomes to map the TIMP gene on the human X chromosome. In combination with previous data, the gene can be assigned to Xp11.23----Xp11.4. Genetic linkage analyses demonstrate that TIMP is linked to the more distal ornithine transcarbamylase (OTC) locus at a distance of about 22 centimorgans. The data are consistent with the conclusion that TIMP maps to a conserved synteny and linkage group on the proximal short arm of the human X chromosome and on the pericentric region of the mouse X chromosome, including loci for synapsin-1, a member of the raf oncogene family, OTC, and TIMP.

Liver fibrosis and extracellular matrix

Liver fibrosis and extracellular matrix play a central role in liver function impairment. Little information is available on the dynamic aspects and the natural history of fibroplasia, even if there is growing evidence that extracellular matrix accumulation (collagen I, III, IV, fibronectin, laminin, proteoglycans, etc.) is not to be considered only a passive structural support for damaged hepatic tissue, but may actively modulate liver cell behaviour. Clinicians need to date liver fibrosis and to monitor connective tissue synthesis and degradation, but attempts to develop reliable serological markers for collagen metabolism are hampered by the absence of a well defined golden standard to validate them. Nevertheless, serum type III aminoterminal procollagen peptide, at the moment, seems to be the most acceptable parameter of fibrogenesis. The data concerning the mechanisms of collagen production-degradation are becoming so precise and numerous that even if they have not, to date, led to 'routine' advantages for patients, they will end up becoming important tools in the clinical practice and management of liver fibrosis.

Differential stimulation of collagenase and chemotactic activity in fibroblasts derived from rat wound repair tissue and human skin by growth factors

Epidermal growth factor and cartilage-derived basic fibroblast growth factor (EGF and CD-bFGF) are mitogens shown to increase the rate of wound repair in animal models. In addition to being a mitogen for granulation tissue, CD-bFGF stimulates the recruitment of cells to the wound site. CD-bFGF and a closely-related chondrosarcoma-derived fibroblast growth factor stimulated chemotaxis of granulation tissue cells in vitro, each factor having a maximum activity at a concentration of 55 pM. Epidermal growth factor was also a potent chemoattractant for rat granulation tissue fibroblasts; however, maximum activity was obtained at 1.7 nM. Cells from all stages of wound repair were chemotactically responsive to these factors, but there was some attenuation of the response to bFGF in cells derived from fully-organized day 28 granulation tissue. Collagenase-catalyzed restructuring of collagen, an additional significant feature of wound repair, is probably critical to cell movement in an extracellular matrix. Cells derived from organizing (6-day old) sponge granulation tissue secreted latent collagenase constitutively in vitro. In the presence of serum, the production of collagenase was stimulated three-four fold by 1.8 nM bFGF derived either from cartilage or chondrosarcoma. When serum was present, as at a wound site, collagenase production was not enhanced by the addition of EGF. Cells from fully organized, day 21 sponge granulation tissue did not secrete latent collagenase constitutively and could not be stimulated to do so by the addition of EGF, bFGF, or phorbol ester. Human skin fibroblast collagenase production was also stimulated by bFGF and was refractory to EGF. While both classes of growth factor have the ability to promote wound healing, the varying responses they elicit in cell populations from the wound site emphasize the different pathways of cellular activation.

Biochemical evaluation of organ cultures from primate flexor tendons

Recently our laboratory has reported, in a lacerated flexor tendon model, that the "early turnover" phase of the repair process extends for a longer period of time in vivo than previously documented. The extensive turnover of the collagenous matrix was consistent with the presence of collagenolytic activity in repairing tendon tissue and suggested a possible regulatory role for neutral metalloproteinases in flexor tendon repair. However, these in vivo observations could not distinguish the relative contribution by the tendon fibroblasts from that of the surrounding sheath and vascular tissue elements. To further define these interrelationships, the present study investigates the repair process of the flexor tendon in an in vitro tissue culture environment. The sequential changes in matrix formation were defined (i.e., proteoglycans/glycosaminoglycans, glycoproteins, and collagenous proteins). The concomitant production of neutral metalloproteinases as well as prostaglandin E2 was determined in relation to net tissue repair. Profundus flexor tendon segments were obtained from young adult Macaca nemestrina monkeys and maintained in organ culture for periods from 4 days through 9 weeks. Initially (at 2 wks) there was an increase in both sulfated and nonsulfated glycosaminoglycans, which preceded the onset of maximal collagen protein formation. By 6 and 9 weeks of in vitro repair, of the lacerated tendon segments, there was a significant increase in net collagen formation. Neutral metalloproteinase activity increased early in the repair period, from the 4th to 9th day, and decreased thereafter through the 9th week of culture. Functionally the enzyme appeared to be a gelatinase. The temporal pattern of in vitro collagen synthesis in relation to the gelatinase activity support the hypothesis that regulation of this enzyme(s) may be a critical factor in mediating the flexor tendon response to injury.

Collagenase production by early and late passage cultures of human fibroblasts

Comparison of the proteins secreted by early and late passage cell cultures of human fibroblasts revealed a high level of immunoreactive collagenase (Mr = 55,000 Da and 58,000 Da) in the late passage cell culture conditioned medium. Both molecular weight species reacted with a monoclonal anticollagenase antibody and were apparently glycosylation varaents of the same protein. The question of whether the apparent age-dependent differences in collagenase synthesis reflected changes in protein synthesis or secretion was addressed by assaying immunoreactive collagenase and collagenase mRNA. Immunofluorescence microscopy of cellular collagenase revealed that the percentage of collagenase positive cells ranged from 1 to 6% (early passage) to 35 to 46% (late passage) indicating that the late passage cells had higher basal levels of collagenase synthesis. Later passage cultures also secreted higher levels of immunoprecipitable collagenase into the culture medium and Northern analysis established that the basal level of collagenase mRNA was also 10 times greater in late passage cells. High basal levels of collagenase were also observed in fibroblasts cultured from an in vivo aged donor and from donors with Werner's syndrome. Collagenase production was induced in both early and late passage cell cultures by exposure to fibroblast extracellular matrix, fibroblast conditioned media, polypeptide growth factors, or phorbol esters. The induced levels were always greater in the late passage cell cultures than in the early passage cell cultures.

Transforming growth factor-beta regulation of collagenase, 72 kDa-progelatinase, TIMP and PAI-1 expression in rat bone cell populations and human fibroblasts

Quiescent cultures of normal fetal rat calvarial bone cell populations (RC I and RC IV) and human fibroblasts were incubated with 1.0 ng/ml TGF-beta and the conditioned culture media were processed individually to separate collagenase and 72 kDa-progelatinase from TIMP, the tissue inhibitor of matrix metalloendoproteinases, using mini-columns of heparin- and gelatin-Sepharose. Collagenase synthesis was decreased progressively by TGF-beta in fibroblasts despite a 1.6-fold increase in secreted protein levels and a approximately 1.8-fold increase in 72 kDa-progelatinase synthesis. The human fibroblasts and the osteoblast-enriched RC IV cells showed a greater TGF-beta-induced stimulation in 72 kDa-progelatinase levels over controls compared with the RC I cells. In contrast to RC IV cells, in which TIMP mRNA levels were increased 2.9-fold by TGF-beta, the constitutive level of TIMP transcripts in the RC I cells was greater than 20-fold over that of the RC IV cells, but was not elevated by TGF-beta. TGF-beta also increased TIMP expression in fibroblasts approximately 1.7-fold and PAI-1 levels approximately 5-fold in RC IV cells, and greater than 10-fold in fibroblasts.

Enhancement of transformed cell growth in agar by serine protease inhibitors

We investigated the effects of three serine protease inhibitors (leupeptin, soybean trypsin inhibitor, and aprotinin) on the serum-free growth of two transformed cell lines in soft agar. Aprotinin markedly enhanced the growth of rat embryo fibroblasts that had been transformed by polyoma middle T antigen (PyMLV-REF52), while having only a slight effect on the colonial growth of SV40 transformed Balb/c 3T3 cells (SV3T3-Aga). Leupeptin and soybean trypsin inhibitor, on the other hand, significantly enhanced the growth of SV3T3-Aga cells while having little effect on PyMLV-REF52 growth. We observed no stimulatory effect of any of the protease inhibitors on serum-free monolayer growth. Under conditions of excess aprotinin, PyMLV-REF52 cells were found to be unresponsive to epidermal growth factor (EGF) at a concentration that would normally stimulate agar colony growth. However, aprotinin was not capable of supporting colony formation with transforming growth factor-beta. These results indicate that aprotinin acts primarily as a protease inhibitor in spite of its structural homology to EGF and that EGF may promote the soft agar growth of these cell lines either by inhibiting proteolysis directly or by enhancing the synthesis of a serine protease inhibitor.

Restriction fragment length polymorphism analysis and assignment of the metalloproteinases stromelysin and collagenase to the long arm of chromosome 11

Collagenase and stromelysin are two metalloproteinases produced mainly by connective tissue cells and involved in the breakdown of the extracellular matrix. cDNA clones for both of these genes have been isolated and sequencing has shown them to be closely related. The collagenase and stromeylsin cDNA clones have been used to assign these genes to the long arm of chromosome 11 in the regions 11q21-22.1 and 11q22.2-22.3, respectively. This has been achieved using somatic cell hybrids and in situ hybridization. In addition a Taq1 restriction fragment length polymorphism has been demonstrated using the stromelysin cDNA.

A maternal mRNA localized to the vegetal hemisphere in Xenopus eggs codes for a growth factor related to TGF-beta

We report that Vg1, a maternal mRNA localized to the vegetal hemisphere of frog eggs, encodes a member of the transforming growth factor-beta (TGF-beta) family of proteins. Furthermore, we show that Vg1 mRNA is distributed to presumptive endodermal cells after fertilization. Previous studies had shown that the vegetal end of a frog egg produces a signal that induces the overlying animal pole cells to form mesodermal tissue. More recently it has been shown that fibroblast growth factor (FGF) and TGF-beta can participate in the induction of muscle. Together, these results lead us to propose that the formation of mesoderm during frog development is specified by the products of localized maternal mRNAs, including Vg1.

Effects of growth factors in vivo. I. Cell ingrowth into porous subcutaneous chambers

Growth factors secreted by platelets and macrophages may play roles in atherogenesis and in wound repair. The multiple biologic effects of these factors are being studied extensively in vitro, but their roles in vivo are relatively unexplored. The cellular responses to platelet-derived growth factor (PDGF), transforming growth factor beta (TGF beta), basic fibroblast growth factor (bFGF), and epidermal growth factor (EGF) were examined in a wound chamber model in rats. Growth factors were emulsified in bovine dermal collagen suspensions, placed in 1 X 30-mm porous polytetrafluoroethylene tubes, inserted subcutaneously, and removed after 10 days. The presence of PDGF (400 ng), TGF beta (200 ng), or bFGF (100 ng) increased the DNA content of the chambers two- to sixfold, compared with controls. Regardless of dose, EGF (100-800 ng) did not affect the DNA content. The increases in DNA observed for PDGF, TGF beta, or bFGF resulted from accumulations of varying numbers of fibroblasts, capillaries, macrophages, and leukocytes in 10-day chambers. The addition of 250 micrograms/ml heparin to the collagen suspension potentiated the response to PDGF and bFGF, but not to TGF beta or EGF. The clearance of 125I-labeled growth factors from the chambers was biphasic. After an initial rapid phase, the remaining growth factor was slowly cleared. The half-life of the initial phase was rapid for PDGF (12 hours) and bFGF (9 hours) and somewhat slower for TGF beta (22 hours). There was no difference in the rate of clearance between collagen and collagen/heparin matrices for any of the growth factors examined. These studies demonstrate that PDGF, bFGF, and TGF beta can induce granulation tissue development in normal animals. The similarity in cellular responses to three peptides with differing in vitro actions suggests that the responses observed at 10 days reflect a secondary process, possibly mediated by effector cells such as macrophages, lymphocytes, or granulocytes that are attracted into the chamber by each growth factor, rather than a direct effect of the factors themselves.