Our prior work shows that cultured BR cells derived from dog mastocytomas secrete the 92-kDa proenzyme form of gelatinase B. We provided a possible link between mast cell activation and metalloproteinase-mediated matrix degradation by demonstrating that α-chymase, a serine protease released from secretory granules by degranulating mast cells, converts progelatinase B to an enzymatically active form. The current work shows that these cells also secrete gelatinase A. Furthermore, gelatinases A and B both colocalize to α-chymase-expressing cells of canine airway, suggesting that normal mast cells are a source of gelatinases in the lung. In BR cells, gelatinase B and α-chymase expression are regulated, whereas gelatinase A expression is constitutive. Progelatinase B mRNA and enzyme expression are strongly induced by the critical mast cell growth factor, kit ligand, which is produced by fibroblasts and other stromal cells. Induction of progelatinase B is blocked by U-73122, Ro31-8220, and thapsigargin, implicating phospholipase C, protein kinase C, and Ca2+, respectively, in the kit ligand effect. The profibrotic cytokine TGF-β virtually abolishes the gelatinase B mRNA signal and also attenuates kit ligand-mediated induction of gelatinase B expression, suggesting that an excess of TGF-β in inflamed or injured tissues may alter mast cell expression of gelatinase B, which is implicated in extracellular matrix degradation, angiogenesis, and apoptosis. In summary, these data provide the first evidence that normal mast cells express gelatinases A and B and suggest pathways by which their regulated expression by mast cells can influence matrix remodeling and fibrosis.
Gelatinase A (MMP-2,3 72-kDa gelatinase) and gelatinase B (MMP-9, 92-kDa gelatinase) are MMPs that share similar domain structures and in vitro matrix substrate specificities (1), yet appear to have unique, temporally defined roles during tissue injury and repair. Analysis of biologic specimens reveals asynchronous up-regulation of gelatinases, characterized by either a transient or sustained increase in expression of one gelatinase, while that of the other remains at basal levels (2, 3). Fluids and tissues may be rich in diverse populations of inflammatory cells that secrete MMPs and may also contain pro-MMPs extravasated from inflamed vessels (4). Thus, the cellular sources of gelatinases and the roles of individual cytokines or growth factors that influence gelatinase expression in specific types of injury remain obscure. Inflammatory cells such as neutrophils, macrophages, lymphocytes, and eosinophils secrete gelatinase B (5, 6, 7, 8), while myriad cell types release gelatinase A (9). Soluble mediators such as IL-1β, TNF-α, TGF-β (10), lymphotoxin, and LPS (11) have all been shown to modulate gelatinase expression in vitro. However, net enzymatic activity of MMPs in a given milieu depends not only on their levels of expression, but also on the relative coexpression of tissue inhibitors of metalloproteinases (TIMPs) (9) and activator proteases that convert the secreted pro-MMPs to active forms via proteolytic removal of the propeptide domain (12, 13, 14, 15, 16, 17, 18).
Certain inflammatory disorders of the lungs, skin, and gut involve mast cells, which degranulate to release stored mediators such as serine and cysteine proteases, histamine, and heparin (19). Increased numbers of degranulating mast cells in tissues with excessive collagen deposition implicate them in matrix remodeling processes leading to fibrosis (20). Coincident with mast cell hyperplasia are a paucity of other inflammatory cells in the fibrotic areas and an abundance of fibroblasts in close proximity to mast cells. While specific contributions of mast cells to fibrotic processes remain unclear, these observations suggest that mast cells may act independently or depend on interactions with collagen-secreting fibroblasts (21). Both cell types secrete factors that influence each other’s migration, proliferation, differentiation, and survival. Fibroblasts secrete kit ligand (KL, stem cell factor) that binds to Kit, a mast cell receptor tyrosine kinase, whose ligation initiates a cascade of intracellular signaling events regulating these mast cell functions (22, 23, 24). We previously reported that BR dog mastocytoma cells constitutively secrete progelatinase B, which is activated upon cleavage by α-chymase, a mast cell secretory granule-associated chymotryptic serine protease (25, 26). Additional data demonstrated that incubation of mastocytoma cells with phorbol ester increases expression of gelatinase B while decreasing that of α-chymase, predicting that regulation of expression of these proteases is linked (26, 27). In the present work, we demonstrate that resident mast cells in tissues constitutively produce gelatinase B and that expression of gelatinase B by cultured mast cells is regulated by not only c-kit ligand, but also TGF-β, an inflammatory cytokine implicated in tissue fibrosis.
Materials and Methods
Dog BR mastocytoma cells were maintained in continuous suspension culture, as previously described (25), in DMEM-H16 supplemented with 2% bovine calf serum. Cells were harvested and washed in PBS, and then incubated in serum-supplemented or serum-free medium alone, in the presence of 2 mM PMSF, 25 ng/ml 12-O-tetradecanoylphorbol-13-acetate (TPA), or 5 ng/ml TGF-β1 (R&D Systems, Minneapolis, MN), or with various concentrations of recombinant canine KL (kindly provided by Dr. Keith Langley, Amgen, Thousand Oaks, CA), U73122, Ro31-8220, or EGTA (Calbiochem, La Jolla, CA) for different times at 37°C. The conditioned medium was harvested and centrifuged at 500 × g for 5 min to remove cells and debris; decanted supernatant was stored at −20°C.
Gelatinase A was partially purified from media conditioned by BR cells using a protocol similar to that previously used to purify mastocytoma gelatinase B (26). Crude media conditioned by BR cells were brought to a final concentration of 0.5 M NaCl and subjected to sequential chromatography using DEAE-cellulose and gelatin-agarose. Bound gelatinases were eluted with 10% DMSO in 50 mM Tris-HCl (pH 7.5), containing 1 M NaCl and 0.02% NaN3. The eluate was concentrated using Centricon-10 spin columns (Amicon, Beverly, MA) and stored at −20°C.
Gelatinase activity was detected by gelatin zymography, as previously described (25). Briefly, aliquots of medium conditioned by cells under various conditions were subjected to electrophoresis in 10% polyacrylamide gels containing gelatin (1 mg/ml). Gels were washed in 2.5% Triton X-100 to permit renaturation of gelatinases and stained with Coomassie blue after overnight incubation. Destaining visualized clear zones of lysis against a blue background, indicating gelatinase activity (7).
Development of α-chymase Ig
α-Chymase was purified from BR dog mastocytoma cells, as previously described (28). Polyclonal anti-α-chymase Ig was raised by injection of the purified α-chymase into rabbits. IgG was purified from antisera produced by Caltag Laboratories (South San Francisco, CA) using Affigel A (Bio-Rad, Hercules, CA), according to the manufacturer’s protocol.
Aliquots of partially purified mastocytoma gelatinase A, purified recombinant human gelatinase A (Chemicon International, Temecula, CA), sonicated cell lysates, purified dog α-chymase (28), or purified dog tryptase (29) were electrophoresed on a 10% SDS-polyacrylamide gel and blotted onto PVDF membrane. The membrane was washed in 10 mM Tris-HCl (pH 7.5) containing 150 mM NaCl and 0.3% Tween-20 (TTBS) and then incubated with polyclonal sheep anti-human gelatinase A (The Binding Site, San Diego, CA) or polyclonal rabbit anti-α-chymase for 1 h. After washing in TTBS, the membrane was incubated with goat anti-rabbit IgG alkaline phosphatase conjugate (Sigma, St. Louis, MO) for 1 h. The membrane was washed twice in TTBS, and then in 10 mM Tris-HCl (pH 7.5) containing 150 mM NaCl. Immunoreactive protein was visualized by standard colorimetric protocols using Fast Red TR/Naphthol AS-MX (Sigma), according to the manufacturer’s protocols.
Lung tissues were obtained from normal euthanized dogs, sacrificed as part of experimental protocols in unrelated projects, via a tissue sharing program at the University of California, San Francisco. Tissues were washed in PBS (pH 7.40), fixed in PBS containing 4% paraformaldehyde for 1 h, and incubated in PBS containing 30% sucrose for 18 h at 4°C. Specimens were washed in PBS before freezing in Tissue-Tek OCT compound (Miles, Elkhart, IN) at −70°C.
To identify mast cells by metachromatic staining of sulfated proteoglycans in secretory granules, 5-μm-thick frozen airway sections were washed in PBS and incubated in 10 mM Tris-HCl (pH 8) buffer containing 5 mM magnesium acetate, 0.1 mM EDTA, and 0.1% methylene blue (w/v) for various periods of time. Tissue sections were washed in PBS, dehydrated in graded ethanols, cleared in xylenes, and mounted with a coverslip.
To detect mast cell proteases, sections were washed in PBS and blocked for 10 min at 18°C in PBS containing 5% dehydrated milk, 3% nonimmune goat serum, 0.1% Triton X-100, and 1% glycine. Sections were washed in PBS and incubated for 18 h at 4°C with Abs diluted in PBS containing 0.05% Tween-20. Sections were incubated alone or with combinations of rabbit anti-dog α-chymase or commercially obtained sheep anti-human gelatinase A or sheep anti-human gelatinase B (The Binding Site; sheep polyclonal Abs are specific for either gelatinase A or B and do not cross-react with other MMPs or Ags in the manufacturer’s quality control assays). Sections were washed and incubated with either Texas Red-conjugated goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA) or FITC-conjugated donkey anti-sheep IgG (Sigma) for 10 min at 18°C. Slides were washed in PBS containing 0.05% Tween-20 and coverslipped after treatment of sections with Vectashield Antifade reagent (Vector Laboratories, Burlingame, CA). To colocalize either gelatinase A or gelatinase B to α-chymase-containing mast cells, sections were incubated simultaneously with the appropriate primary Abs, followed by simultaneous incubation with Texas Red- and FITC-conjugated secondary Abs. Control experiments were performed as indicated to detect signals resulting from recognition of nonspecific Ags by the secondary Abs. Immunofluorescence signals were detected and photographed with a fluorescence microscope.
Poly(A)+ RNA was isolated from BR cells incubated alone or with various combinations of TPA, KL, and TGF-β using the Poly(A)+ RNA MicroFast Track extraction kit (Invitrogen, Carlsbad, CA). RNA blotting was performed as previously described (26). Following size fractionation of poly(A)+ RNA and transfer to Nytran Plus nylon membrane, the membrane was prehybridized for 2 h at 42°C in 50% formamide containing 5× Denhardt’s reagent (1 g/L polyvinylpyrolidone, Ficoll, and BSA) and 5× SSPE (0.75 M NaCl, 50 mM NaH2PO4, and 5 mM EDTA) with 0.1% SDS and 150 μg/ml salmon sperm DNA. The 2.3-kb dog gelatinase B cDNA (26), the 2.2-kb human gelatinase A cDNA (American Tissue Type Collection, Manassas, VA), or the 747-bp dog α-chymase cDNA (30) was random-prime labeled with [α-32P]ATP. Labeled probes were individually hybridized to the filter at 42°C overnight. Filters were washed twice in 6× SSPE with 0.1% SDS at room temperature for 15 min, twice in 1× SSPE with 0.1% SDS at 37°C for 15 min, and once in 0.1× SSPE with 0.1% SDS at 55°C for 15 min. To remove previously bound probe, blots were incubated in 5 mm Tris (pH 8), 0.2 mM EDTA, 0.05% pyrophosphate, and 0.1× Denhardt’s reagent at 65°C for 5 h. Densitometric data were obtained by analysis of autoradiographic signals generated by hybridizing the blot with a labeled probe. To account for possible variations in signal intensity due to differing concentrations of total mRNA present in each lane, the blot was also hybridized with a labeled probe for γ-actin. Densitometric data were then compared with control values obtained with the γ-actin probe.
Differences with a p value of <0.05 using Student’s two-tailed t test were considered significant.
Mastocytoma gelatinase A expression
The gelatin zymogram depicted in Fig. 1 A reveals the presence of two major bands (bands 1 and 2 at ∼92 and ∼84 kDa, respectively) and one minor band (band 3 at ∼66 kDa) of gelatinolytic activity in medium conditioned by dog BR mastocytoma cells, as seen in lane 1. Incubation of cells with PMSF abolishes activity of band 2, revealing a serine protease-dependent mechanism for its generation. As revealed in our prior work (25), bands 1 and 2 represent gelatinolytic activity of the zymogen and active forms of gelatinase B, respectively, with extracellular proteolytic processing mediated by α-chymase released either by spontaneous degranulation or death of cultured cells. Activity of bands 2 and 3 is ablated by Zn2+ or Ca2+ chelators (data not shown), indicating that both are MMPs. However, band 3, unlike band 2, is generated by BR cells even in the presence of PMSF.
To facilitate identification, band 3 was partially purified from medium conditioned by dog mastocytoma cells. Immunoblotting using a polyclonal anti-human gelatinase A Ab identifies an immunoreactive band at ∼62 kDa, as seen in lane 1 of Fig. 1 B. By contrast, immunoreactive bands at ∼72 and ∼66 kDa are seen in lane 2, which contains purified human gelatinase A. Work by other investigators (31) has shown that purified dog progelatinase A (which is identical in 13 of 15 N-terminal residues to human progelatinase A) also yields a ∼62-kDa band by immunoblot analysis. The basis for the differences in the electrophoretic migration of dog and human progelatinase A is unclear. Thus, BR dog mastocytoma cells, an extensively characterized mast cell model (25, 26, 27, 28, 29, 30, 32), constitutively secrete progelatinase A.
Tissue mast cell gelatinase expression
To perform cellular coimmunolocalization studies of tissue mast cell proteases, polyclonal anti-α-chymase Ig was purified from rabbit antisera. Purified anti-α-chymase Ig detects a ∼30-kDa purified α-chymase band and a band of similar size in BR cell lysates, but it does not detect tryptase, the other major mast cell serine protease (data not shown). To determine whether tissue mast cells express gelatinases A and B, we performed immunohistochemistry on sections of normal dog airway. Metachromatically staining mast cells in the lung reside predominantly in bronchial airway subepithelium (33), as shown in Fig. 2,A. Immunohistochemical analysis localizes gelatinase B and gelatinase A signals to individual subepithelial cells, as shown in Fig. 2, D and G, respectively. (Immunoreactivity localized to the epithelium in Fig. 2,D results partially from nonspecific binding of FITC anti-sheep IgG to dog tissues that is not enhanced by nonimmune sheep or rabbit serum; similar background signals are generated by Texas Red anti-rabbit IgG in the presence of sheep or rabbit nonimmune serum. Alternatively, epithelial immunoreactivity in Fig. 2, D and G, may also represent signals for gelatinases in airway epithelial cells, as previously described in vitro (34).) Detection of α-chymase, a secretory granule-associated serine protease that cleaves and activates progelatinase B (25, 26), unequivocally reveals mast cells. Immunofluorescence colocalization of signals for either gelatinase B (Fig. 2,E) or gelatinase A (Fig. 2,H) to α-chymase-positive subepithelial cells (as seen in Fig. 2, F and I) establishes that α-chymase-containing mast cells express gelatinases in vivo. Thus, like other cells resident in or recruited to the airways, mast cells may contribute gelatinases A and/or B to the local milieu. These data substantiate our prior work, which showed that mastocytoma cells secrete both progelatinase B and α-chymase (25), and support the notion that α-chymase-dependent progelatinase B activation may be a mechanism whereby tissue mast cells acting independently can influence local gelatinase B activity.
Induction of mast cell gelatinase B expression
The importance of KL in mast cell migration, proliferation, and survival suggested its candidacy as a regulator of gelatinase B expression. Expression of Kit, the receptor for KL, is restricted primarily to mast cells in connective tissue with limited expression in skin melanocytes and intestinal interstitial cells of Cajal (24). As seen in Fig. 3,A, KL increases levels of mast cell gelatinase B, although to a lesser extent than those induced by phorbol. Levels of mast cell gelatinase B increase in a dose-dependent fashion, with maximal induction at concentrations of 50–100 ng/ml, as shown in Fig. 3,B. As demonstrated in Fig. 3 C, addition of KL to cells results in induction of gelatinase B expression, with increased levels noted at 4 h compared with those of controls.
Selective induction of gelatinase B by KL
Although gelatinases A and B share similar in vitro matrix substrate specificities, wound injury models suggest that their expression is temporally distinct and thus may be subject to regulation by different extracellular signals. Zymographic analysis of medium conditioned by cells shows that KL stimulation increases levels of progelatinase B and its active, truncated forms, while the level of progelatinase A remains unchanged, as seen in Fig. 4,A. Autoradiograms and densitometric data in Fig. 4, B and C, show that KL increases the steady state level of gelatinase B mRNA by ∼5-fold, although to a lesser extent than that induced by phorbol. By contrast, neither KL nor phorbol alters steady state levels of gelatinase A mRNA. These data suggest that KL:Kit interactions result in selective induction of mast cell gelatinase B.
Divergent regulation of gelatinase B and α-chymase expression by KL
As shown in Fig. 5, incubation of cells with KL decreases α-chymase mRNA levels by ∼3-fold. Thus, KL stimulation of cells results not only in selective up-regulation of the expression of progelatinase B, but also divergent regulation of α-chymase. Whether decreased α-chymase synthesis occurs remains unclear.
Signal transduction in KL up-regulation of gelatinase B expression
Ligation of Kit by KL alters expression of mast cell mediators by initiating a cascade of intracellular signaling events (23). To explore the role of early signal transduction pathways involved in KL-induced mast cell gelatinase B expression, medium was harvested from cells coincubated with KL and individual inhibitors selective for signal transduction mediators, and analyzed by gelatin zymography. As seen in Fig. 6,A, treatment of cells with increasing concentrations of U73122, an inhibitor of phospholipase C activation, blocks KL-induced gelatinase B expression. Coincubation with Ro31-8220, a bis-indoylmaleimide inhibitor selective for protein kinase C, also blocks up-regulation of gelatinase B expression by KL, as seen in Fig. 6,B. As shown in Fig. 6 C, incubation of cells with KL in the presence of EGTA, a selective nonmembrane-permeable Ca2+ chelator, also abolishes the increase in gelatinase B induced by KL. Thus, these data suggest that phospholipase C, protein kinase C, and an influx of extracellular Ca2+ play a role in the induction of mast cell gelatinase B expression by KL.
TGF-β attenuates KL-induced gelatinase B expression
Fibrotic tissue disorders demonstrate mast cell hyperplasia (20) and increased levels of TGF-β (35), an inflammatory cytokine that antagonizes the effects of KL (36) and also modulates gelatinase B expression (10, 37, 38). To explore the effect of TGF-β on KL-induced mast cell gelatinase B expression, medium conditioned by cells incubated with combinations of KL and TGF-β was analyzed by gelatin zymography. As depicted in Fig. 7,A, incubation of cells with TGF-β alone for 24 or 48 h (lanes 3 and 4) decreases gelatinase B levels. Preincubation of cells with TGF-β for 24 h also decreases the up-regulation of gelatinase B expression following stimulation by KL (lane 5), an effect enhanced by prolonged incubation with TGF-β for 48 h (lane 6). Whereas incubation with KL increases the gelatinase B mRNA signal by 3-fold, treatment with TGF-β down-regulates expression of gelatinase B by 5-fold, as seen in Fig. 7, B and C. Moreover, TGF-β attenuates the increase in the gelatinase B mRNA signal induced by KL. Thus, these data suggest that ligation of TGF-β to its receptor diminishes KL-induced gelatinase B expression by stabilizing levels of the protease’s mRNA.
Net MMP activity in a given milieu most likely depends on local concentrations of pro-MMPs and TIMPs, and on the presence of proenzyme activators. Extracellular pro-MMPs may accumulate by secretion from cells either normally resident in or recruited to injured sites, augmented by a potential contribution from proforms present in extravasated plasma. Certain fibrotic tissue disorders manifest hyperplasia of degranulating mast cells whose role in extracellular matrix turnover remains an enigma. Our prior work demonstrated that mast cells secrete a TIMP-free progelatinase B that undergoes extracellular cleavage and activation after they release α-chymase, a secretory granule-associated serine protease exocytosed only in response to an appropriate stimulus. The current work establishes that tissue mast cells secrete both gelatinases A and B, and defines the role of KL (a fibroblast mediator that binds to Kit) and the profibrotic cytokine, TGF-β, in mast cell gelatinase B expression.
Colocalization of gelatinases to α-chymase-expressing cells in normal dog airway tissues provides new insights into mast cell biology. Immunohistochemical identification of mast cell gelatinase A and gelatinase B expression in airway mast cells establishes that production of these MMPs is not restricted to the malignant phenotype of cultured mastocytoma cells. Whether all mast cells express gelatinases and whether preformed gelatinases exist in the cytosol or are sequestered in specific granules remain to be determined. Expression of both progelatinase B and α-chymase by individual mast cells is consistent with our hypothesis that exocytosed α-chymase is an important activator of progelatinase B secreted either by mast cells or by other cells in their vicinity (25). The activation mechanism of mast cell progelatinase A, shown in other cell types to be dependent on a trimolecular complex of the proform with TIMP-2 and membrane-type MMP-1 (39), remains to be determined. Other investigators have identified collagenase (MMP-1) and stromelysin (MMP-3) in rodent and murine mast cells, respectively (40, 41). Collectively, these data suggest that resident mast cells may contribute a variety of MMPs, which may not only degrade target substrates in tissues, but may also participate in overlapping proteolytic cascades favoring pro-MMP activation (13, 14, 42, 43). In the lungs, the relative paucity of mast cells in comparison with MMP-expressing epithelial, stromal, and inflammatory cells (5, 6, 7, 8, 9, 34) in normal or injured tissues suggests that mast cells may not be the major source of gelatinases. However, hyperplasia of mast cells in matrix-remodeling disorders such as pulmonary fibrosis (20) predicts that contributions of MMPs from mast cells may be significant during tissue injury and repair.
The critical role of KL:Kit interactions in the regulation of mast cell proliferation, migration, and survival (44) suggested that KL might also modulate mast cell expression of the gelatinase class of MMPs. Our data demonstrate that KL induces mast cell gelatinase B expression in a dose-dependent and selective manner. By contrast, gelatinase A expression in response to KL remains unchanged, thus substantiating the noncoordinate expression of gelatinases in tissue injury models (3). The frequent apposition of mast cells to fibroblasts in fibrotic tissues (21) suggests that fibroblast-derived KL (24) may not only contribute to the development of mast cell hyperplasia, but may also up-regulate gelatinase B expression, whose roles in tissue injury and remodeling remain speculative. Expression of KL by skin keratinocytes (24), alveolar macrophages (45), and airway epithelial cells (46) suggests that additional sources of KL may influence mast cell numbers and activities in a tissue-specific manner. Thus, interaction of fibroblasts with mast cells via KL may be important in the regulation of tissue expression of gelatinase B.
Phorbol stimulation studies were predictive of the effect of KL on mast cell expression of α-chymase, a chymotryptic serine protease that activates progelatinase B by cleaving its catalytic domain (26). Our prior work demonstrated that phorbol stimulates divergent expression of these proteases: a multifold increase in gelatinase B expression (26) contrasts with almost complete down-regulation of α-chymase expression (27). KL’s effect is similar, but of lesser magnitude.
Biochemical and physiologic consequences of KL’s effect on mast cell expression of gelatinase B and α-chymase are difficult to predict because much about progelatinase B activation and the roles of both proteases in vivo remain unknown. Convergent regulation of these proteases would have suggested that KL stimulation favors conversion of progelatinase B to active, mature forms. By contrast, divergent regulation predicts that a local accumulation of progelatinase B might occur because less α-chymase is available to activate it. Whether KL alters production of α-chymase by mature mast cells is unknown. Since progelatinase B may also be activated indirectly by a proteolytic cascade initiated by mast cell tryptase or stromelysin (MMP-3) (12, 47), decreased availability of α-chymase may not necessarily result in an excess of mast cell progelatinase B. Therefore, definitive assessment of the regulatory influence of KL on mast cell participation in matrix remodeling awaits further clarification of the interdependence of the in vivo activities of gelatinase B and α-chymase.
KL influences differentiation of bone marrow precursors into mast cells with distinct phenotypes based on protease expression. However, tissue-dependent heterogeneity of mast cells suggests that factors in addition to KL also direct mast cell protease expression (48). Effector cells recruited to inflamed or injured tissues contribute soluble mediators that most likely alter normal cell-cell and cell-matrix communication. Of the myriad cytokines released, sustained production of TGF-β has been shown to contribute to abnormal tissue remodeling that progresses to fibrosis (35). Our data demonstrate that coincubation with TGF-β attenuates mast cell production of gelatinase B in response to KL stimulation. How TGF-β suppresses KL-induced gelatinase B expression is unclear. Prior studies demonstrate that TGF-β neither disrupts binding of KL to Kit, nor alters mast cell Kit expression (36). It is possible that TGF-β may interfere in KL-initiated intracellular signal transduction via competition for downstream mediators such as mitogen-activated protein kinases (49, 50) or transcription factor AP-1 (51).
Relative concentrations of TGF-β and KL may therefore be important determinants of mast cell protease expression in tissues undergoing repair. Availability of soluble TGF-β depends on release of its bound, latent complex from extracellular matrix and its subsequent conversion from latent to active forms (52, 53, 54). TGF-β may alter mast cell numbers by suppressing the ability of KL to rescue mast cells from apoptosis (36), a cellular process that may also be regulated by gelatinase B (55). Since α-chymase also releases matrix-bound latent TGF-β (56), a complex feedback loop may regulate the effects of KL on mast cells in the presence of TGF-β. Moreover, mast cells themselves may secrete TGF-β (32, 57). Thus, persistent production of TGF-β may drive fibrotic processes not only by stimulating stromal cell collagen deposition (58, 59, 60), but also by altering mast cell numbers and their expression of gelatinase B.
In summary, our results demonstrate that resident tissue mast cells express gelatinase B, and that its expression in cultured mast cells may be regulated by both KL and TGF-β. The data suggest that KL:Kit interactions may regulate gelatinase B expression in a manner that is both mast cell and tissue specific. Perturbations in signaling mediated by KL due to excess TGF-β in inflammatory processes may alter mast cell gelatinase B expression and contribute to the development of fibrosis.
We thank Dr. Keith Langley for providing recombinant canine kit ligand and J. Caleb Rossall for his expert technical assistance.
This work was supported in part by a grant from the Committee on Research of the Academic Senate at the University of California, San Francisco, and by Grants HL-03345 and HL-24136 from the National Institutes of Health. K.C.F. is the recipient of a Dalsemer Research Scholar Award from the American Lung Association.
Abbreviations used in this paper: MMP, matrix metalloproteinase; KL, recombinant canine kit ligand; TIMP, tissue inhibitor of metalloproteinase; TPA, 12-O-tetradecanoylphorbol-13-acetate.