Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The transforming growth factor- (TGF-) superfamily of cytokines is composed of a group of closely related proteins, TGF-₁, TGF-₂, TGF-₃, TGF-₄, and TGF-₅, having 70 to 80% protein sequence identity, and of a number of more distantly related proteins, including the activins, inhibins, and bone morphogenetic proteins, having 30 to 40% identity with TGF-₁ at the primary sequence level (1). Analysis of the complementary DNAs (cDNAs) encoding the TGF-s has shown that each is initially synthesized as a larger precursor molecule whose carboxy-terminal portion contains the mature form of TGF- (1). After proteolytic cleavage, the two portions of the precursor remain together and are secreted as a biologically inactive, noncovalently bound complex consisting of dimers of both the amino-terminal remainder of the precursor, designated latency associated peptide (LAP), and mature TGF- (2, 3). In some cases, this complex is secreted while bound to another protein, termed latent TGF--binding protein (LTBP) (2, 4). The function of LTBPs remains to be determined, since it is clear that they are not necessary to maintain TGF- in an inactive form and do not appear to bind mature TGF-. Although LTBPs may facilitate the secretion of TGF- (5) or binding of the inactive complex to the cell surface, where activation takes place (6), they are also found as free proteins associated with components of the extracellular matrix (7).

The TGF-s manifest a wide range of biologic activities, including growth stimulation and inhibition, modulation of extracellular matrix production, and regulation of cell differentiation (1, 8). These multiple, varied effects are initiated by the binding of TGF- to specific transmembrane receptors (type I and type II), which exhibit serine/ threonine kinase activity (9, 10). Persuasive evidence indicates that TGF- first binds to the type II receptor, and that the type I receptor is then recruited to this complex and phosphorylated by the type II receptor (11). Specific features of the subsequent series of signaling events are only now being defined (10, 12).

In the present work, we investigated the effects of TGF-₁ on the expression of tropoelastin by cultured human fetal lung fibroblasts. We demonstrate that TGF-₁ dramatically increases the expression of tropoelastin at both the messenger RNA (mRNA) and protein levels. This effect is mediated by stabilization of elastin mRNA rather than by transcription of the elastin gene. The intracellular signaling pathway leading to mRNA stabilization involves phosphatidylcholine (PC)-specific phospholipase C (PLC) and protein kinase C (PKC), but not G proteins or extracellular signal regulated kinases. These results suggest that the TGF-₁ signaling pathway leading to mRNA stabilization shares components with that modulating transcription, but also contains diverse other elements.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials

Specific inhibitors were purchased from suppliers as follows: PD098059 from New England Biolabs (Beverly, MA); calphostin, staurosporin, U73122, and D609 from Biomol Inc. (Plymouth Meeting, PA); cholera and pertussis toxins from List Biologicals (Campbell, CA); and 5,6-dichlorobenzimidazole riboside (DRB) from Sigma Chemical Co. (St. Louis, MO).

Cell Culture and RNA Analysis

Human fetal lung fibroblasts (GM05389; Coriell Institute for Medical Research, Camden, NJ) were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal calf serum (FCS). Before the addition of TGF-₁, the medium of confluent cultures was replaced with DMEM containing 1% serum. After incubation with TGF-₁, total cellular RNA was extracted by the acid guanidine isothiocyanate method (13), and 15 µg were electrophoresed on formaldehyde-1.2% agarose gels, transferred to Zeta-Probe membranes (Bio-Rad, Richmond, CA), and hybridized to a 2.2-kbp human elastin cDNA probe labeled with ³²P by the random primer method (14). RNA loading and transfer were evaluated by probing with a glyceraldehyde phosphate dehydrogenase (GAPDH) cDNA probe. Equivalent loading and transfer were also verified by quantitative image analysis of ethidium bromide staining of ribosomal RNA in the blots themselves. The filters were autoradiographed, and scanning densitometry and analysis were performed to quantify the relative amounts of mRNA (MolQuant V3.1 software; Molecular Dynamics, Sunnyvale, CA). Elastin mRNA values were normalized to equivalent values for GAPDH.

Measurement of Elastin mRNA Stability

Confluent cultures were incubated with or without 3 ng/ml of TGF-₁ for 24 h in DMEM containing 1% FCS. DRB was then added to a final concentration of 50 µM. At the indicated times after DRB addition, RNA was isolated from replicate cultures, dissolved in a standard volume of 30 µl of water, and subjected to Northern analysis. Fifteen micrograms of the zero-time samples were analyzed, and a sample volume equal to this amount was analyzed for the other timed samples. Quantitation of total RNA as measured by absorbance at 260 nm demonstrated a slow decline in total RNA content, so that after 20 h, an equal sample volume contained approximately 10 µg of RNA rather than 15 µg. Image analysis of the ethidium bromide-stained ribosomal RNA in the blots confirmed this decrease (data not shown).

Measurement of Transcription Rate by Nuclear Run-on

Relative transcription rates in control and TGF--treated cells were measured as previously described, with minor modifications (15). Briefly, 3 to 4 × 10⁷ cells were lysed in 4 ml Nonidet P-40 (NP-40) buffer (10 mM Tris-HCl, pH 7.4; 10 mM NaCl; 3 mM MgCl₂; 0.5% NP-40) for 5 min on ice, and centrifuged at 500 × g for 5 min. Nuclei were washed with NP-40 buffer and resuspended in 100 (50 mM Tris-HCl, pH 8.3; 40% glycerol; 5 mM MgCl₂; 0.1 mM ethylenediamine tetraacetic acid [EDTA]) for storage in liquid nitrogen. The nuclei were thawed for the run-on transcription assay, mixed with 100 µl of reaction buffer (10 mM Tris-HCl, pH 8.0; 5 mM MgCl₂; 300 mM KCl; 5 mM dithiothreitol [DTT]; 0.4 units/µl ribonuclease [RNAse] inhibitor [Boehringer-Mannheim, Indianapolis, IN], 1.0 mM each of adenosine triphosphate [ATP], CTP, and GTP) and 100 µCi of [-³²P]uridine triphosphate ([-³²P]UTP), and reacted at 37°C for 30 min. Purified, linearized plasmid DNA (15 µg) was denatured in NaOH, neutralized in ammonium acetate, and applied to Zeta-Probe membranes. The DNA was then bound to the membrane using a Stratogene Stratolinker (Stratogene, La Jolla, CA), followed by baking at 80°C for 1 h. The individual probes contained cDNA inserts for human elastin (2.2 kbp), human fibronectin (2.0 kbp), and GAPDH (1.3 kbp). Vector plasmid DNA served as a control. The synthesized RNA molecules were isolated as previously described (13) and hybridized in 7% sodium dodecyl sulfate (SDS); 50% formamide, 1 mM EDTA; 250 mM NaCl; and 250 mM Na₂PO₄; pH 7.2, at 42°C for 48 h. After washing, the filters were treated with 10 µg/ml RNase A (Sigma Chemical Co.) for 30 min at 37°C, and rewashed, and the hybridization signals for the elastin and fibronectin transcripts were quantitated by phosphorimaging analysis (Molecular Dynamics, Inc.) and normalized to that for GAPDH.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of TGF- on Elastin mRNA Steady-state Levels, Elastin Gene Transcription, and mRNA Stability

Confluent lung fibroblast cultures were incubated with various concentrations of TGF-₁ for 24 h, after which RNA was extracted and the levels of elastin mRNA determined by Northern analysis (Figure 1). The results of this experiment demonstrated that TGF-₁ produced a substantial increase in elastin mRNA in a concentration-dependent manner, so that at 2.9 ng/ml there was an approximate 25-fold increase in elastin mRNA with no alteration in GAPDH mRNA. A time-course analysis demonstrated that this effect of TGF-₁ developed relatively slowly (Figure 2), with little increase in elastin mRNA seen during the first 6 h of treatment, and a maximal response reached at 18 to 24 h. This result suggested that this effect was not mediated through increased transcription of the elastin gene, since a much more rapid response has been observed in other genes for which TGF- regulation of transcription has clearly been demonstrated (16). However, in order to prove this, nuclear transcription run-on analysis was done. The results of this experiment demonstrated that TGF-₁ had no significant effect on elastin gene transcription, but did, as expected, significantly stimulate fibronectin transcription, which served as a positive control (Figure 3).

View larger version (32K): [in this window] [in a new window]   Figure 1.   Dose-dependent response of elastin mRNA levels to TGF-1. Confluent cultures of human fetal lung fibroblasts were incubated with the designated concentrations of TGF-1 for 24 h, after which total RNA was isolated and analyzed by Northern hybridization as described in MATERIALS AND METHODS. The plotted values represent the average of determinations from duplicate cultures. The insert shows the autoradiograph of the Northern blot after hybridization with elastin and GAPDH probes.

View larger version (11K): [in this window] [in a new window]   Figure 2.   Time course of TGF-1-mediated increase of elastin mRNA level. Confluent cultures of human fetal lung fibroblasts were incubated with 3 ng/ml of TGF-1 for the indicated times. RNA was isolated and analyzed by Northern hybridization for elastin mRNA levels as in Figure 1. The values represent the average of determinations from duplicate cultures.

View larger version (29K): [in this window] [in a new window]   Figure 3.   Nuclear run-on. Nuclear run-on was done on nuclei prepared from control cells or those incubated with 3 ng/ml of TGF-1 for 24 h, as described in MATERIALS AND METHODS. The labeled mRNA was isolated and hybridized to plasmid vector, GAPDH, fibronectin, and elastin cDNA probes that had been bound to a Zeta-Probe membrane. After washing, the filters were treated with 10 µg/ml RNase A, rewashed, and then subjected to phosphorimaging analysis, as described in MATERIALS AND METHODS. The values for elastin and fibronectin were normalized to that of GAPDH.

The effect of TGF-₁ on elastin mRNA stability was then determined. Cultures were pretreated with 3 ng/ml of TGF-₁ for 24 h, after which DRB was added to the cultures as well as to control cultures to terminate transcription, and mRNA levels were determined by Northern hybridization. Analysis of the resulting data demonstrated that under both control and TGF-₁-stimulated conditions, levels of elastin mRNA decreased with first-order kinetics. Moreover, TGF-₁ treatment markedly increased the stability of elastin mRNA as compared with that in untreated control cultures (Figure 4). Collectively, these results are consistent with those observed in human dermal fibroblasts, in which TGF- upregulated elastin mRNA levels by increasing mRNA stability with no apparent alteration in transcription (15, 17).

View larger version (31K): [in this window] [in a new window]   Figure 4.   Effect of TGF-1 on elastin mRNA stability. Confluent cultures of human fetal lung fibroblasts were incubated for 24 h in the prescence or absence of 3 ng/ml of TGF-1. DRB was then added to a final concentration of 50 µM, and total RNA was prepared from duplicate cultures at the indicated times and analyzed by Northern hybridization, as described in MATERIALS AND METHODS. The insert shows the autoradiographs of the Northern blots hybridized with elastin and GAPDH probes.

Effect of Various Signaling Pathway Inhibitors on Elastin mRNA Level

Although the biologic functions of TGF- are reasonably well understood and documented, the biochemical mechanisms involved in mediating these functions are for the most part incompletely defined. Considerable effort has recently been expended in defining crucial steps and key second messengers involved in the transduction of signals from the interaction of TGF- with its membrane receptor to the cell nucleus, resulting in modulation of transcription and cell division. However, no information is currently available on the mechanisms by which this cytokine modulates elastin mRNA stability. In order to begin to define this pathway, we tested the effects of several inhibitors known to have relatively specific modes of action. These included D609 and U73122, inhibitors of PLC (18, 19); staurosporin and calphostin, inhibitors of PKC (20, 21); and cholera and pertussis toxins, inhibitors of G-protein-mediated transduction signaling (22, 23). Each of these inhibitors was added to the cultured fibroblasts 2 h before TGF-₁, and the incubations were continued for 24 h, at which time total RNA was extracted and analyzed by Northern analysis. D609 inhibited the effect of TGF-₁ in a dose-dependent manner (Figure 5A), so that at 25 µg/ml of D609 the elastin mRNA level was equal to 30% of that of cells treated solely with TGF-, and at 50 µg/ml the level was only marginally greater than that of cells not treated with TGF-₁. In order to determine whether the changes in mRNA levels were reflected in alterations in the production of tropoelastin, the biosynthetic precursor of elastin, the cultures were analyzed with a specific enzyme-linked immunosorbent assay (ELISA) for tropoelastin (24). This analysis demonstrated that TGF-₁ markedly increased the production of tropoelastin, and that this increase was abrogated by D609 in a dose-dependent manner (Figure 5B). In contrast to the effects of D609, which has been reported to specifically inhibit PC-specific PLC (18), U73122, an inhibitor of phosphatidylinositol (PI)- specific PLC (19), had little effect (data not shown).

View larger version (14K): [in this window] [in a new window]   Figure 5.   Effect of the PC-specific PLC inhibitor D609 on the TGF-1 modulation of elastin mRNA level. Confluent cultures of human fetal lung fibroblasts were incubated for 2 h with the various indicated concentrations of D609, and TGF-1 was then added to a final concentration of 3 ng/ml. (Top) After 24 h of incubation, total RNA was isolated and analyzed by Northern hybridization for elastin mRNA levels, as in Figure 1. The values represent the average of determinations from duplicate cultures. (Bottom) The incubation media obtained after 24 h were analyzed for tropoelastin with a specific ELISA. Basal level refers to the values obtained in the absence of TGF-1.

Cleavage of PC by PC-specific PLC releases PC and diacylglycerol (DAG), a well known endogenous activator of PKC (25). In order to determine whether PKC participates in the elastin-message-stability pathway, cells were pretreated with the PKC inhibitors staurosporin, which interacts with the ATP binding site (20), and calphostin, which binds to the regulatory subunit (21). Both inhibitors abrogated the effect of TGF-₁ in a dose-dependent manner. Staurosporin at 25 nM decreased elastin mRNA levels to approximately 30% of that found in cells treated with TGF-₁ alone, and at 100 nM further reduced elastin mRNA levels to 20% (Figure 6A), and there was a commensurate decrease in tropoelastin production as well (Figure 6C). Calphostin proved to be an equally effective inhibitor (Figures 6B and 6D). Many, but not all, PKCs are activated by phorbol esters, which act as agonists of DAG (25). However, when phorbol-12-myristate-13-acetate (PMA) was tested in our system over a wide concentration range (10 to 100 ng/ml), no effect was observed on elastin expression as analyzed at both the protein and mRNA levels (data not shown).

View larger version (22K): [in this window] [in a new window]   Figure 6.   Effect of the PKC inhibitors staurosporin and calphostin on the TGF-1 modulation of elastin mRNA level. Confluent cultures of human fetal lung fibroblasts were incubated for 2 h with the various indicated concentrations of the two inhibitors, and TGF-1 was then added to a final concentration of 3 ng/ml. (A and B) After 24 h of incubation, total RNA was isolated and analyzed by Northern hybridization for elastin mRNA levels, as in Figure 1. The values represent the average of determinations from duplicate cultures. (C and D) The incubation media obtained after 24 h were analyzed for tropoelastin with a specific ELISA. Basal level refers to the values obtained in the absence of TGF-1.

TGF- has also been reported to activate the mitogen-activated protein kinase (MAPK) pathway (26). In order to test the possibility that MAPKs participate in the TGF-₁ modulation of elastin mRNA stability, cells were pretreated with the specific inhibitor PD098059 (29) before incubation with TGF-₁. This inhibitor had no effect on the TGF-₁-mediated increase in elastin mRNA, suggesting that activation of erk1 and erk2 of the MAPK system is not necessary for TGF--mediated elastin mRNA stabilization (data not shown). However, it is possible that other components of the MAPK system, such as erk3 (30), a recently described pp57 kinase (31), or some yet unidentified kinase, may be involved. Similarly, neither cholera toxin nor pertussis toxin exhibited any inhibition of the elastin-message-stabilization activity of TGF-₁ (data not shown), suggesting that, in contrast to the case with some other effects of TGF- (23), G proteins are not involved in the elastin-mRNA-stabilizing effect.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

TGF- has been shown to modulate the expression of several extracellular matrix proteins. In most cases, the effects are achieved largely through alterations in transcription rates (16). However, as described previously in other cell systems (15, 17), and as seen in the present work, TGF- appears to upregulate elastin expression not by modulating transcription of the elastin gene, but rather by stabilizing the elastin mRNA. TGF- has also been shown to increase the synthesis of tropoelastin in cultured human dermal fibroblasts (15) and porcine aortic smooth-muscle cells (32). One of the signaling pathways by which TGF- can alter transcription leads from the heterotypic receptor complex at the cell surface through PLC, PKC, and the extracellular-signal regulated kinases (26). Here, through the use of specific inhibitors, we show that the stabilization of elastin mRNA also involves PC-specific PLC and PKC. However, extracellular-signal regulated kinases do not appear to be involved. The latter result is consistent with the observation that inhibition of de novo protein synthesis did not abrogate the TGF--mediated increase in elastin mRNA level in cultured dermal fibroblasts (17). The apparent lack of MAPK pathway involvement in the present effect is not totally unexpected. A recent paper has reported that modulation of cell growth signals in Balb 3T3 and Swiss 3T3 mouse fibroblasts by TGF- was not associated with detectable phosphorylation and activation of the 41-kD (erk1) and 43-kD (erk2) MAPKs (33). In this system, TGF- modulated cellular proliferation independently of the MAPK cascade pathway in a manner distinct from most other peptide growth factors, such as platelet-derived growth factor (PDGF), epidermal growth factor (EGF), and fibroblast growth factor (FGF), whose growth-promoting abilities have been generally associated with MAPK activation.

The 11 or 12 known members of the PKC family have different activation properties, cofactor requirements, tissue and cell distributions, and cellular compartmentalization, but share sequence similarities (25, 34). On the basis of cDNA sequences and biochemical properties, all of the PKC isoforms have been operationally divided into three groups, consisting of: (1) conventional PKC (cPKC), requiring Ca²⁺ and DAG for activation, and involving mainly the classical PI pathway; (2) novel PKC (nPKC), requiring DAG but Ca²⁺-independent and involving either the PC-specific PLC or phospholipase D (PLD) hydrolytic pathway; and (3) atypical PKC (aPKC), which apparently requires neither Ca²⁺ nor DAG, and whose mode of activation is still being investigated.

The presence of so many different PKC isotypes with similar biologic functions has sugggested that distinct isotypes may be associated with specific pathways and downstream effector molecules, the choice being dictated by a variety of parameters such as cofactor requirements and subcellular location, which enables interaction with various ligand-engaged receptors. Indeed, growing evidence has begun to associate PKC isotypes with specific cellular transduction machinery. For example, a kinase belonging to the aPKC group has been reported to be crucial and sufficient for mitogenic signal transduction in Xenopus laevis oocytes (35, 36) and, more recently, a PKC (cPKC) subtype has been shown to activate erk3 of the MAPK cascade pathway (30). Such variation in expression of PKC isotypes could explain the difference between the results of the present study, in which TGF-₁, acting through a PKC, stabilized elastin mRNA in lung fibroblasts, and those previously achieved by Parks and colleagues (37), in which phorbol ester downregulated elastin expression by destabilizing elastin mRNA in cultured fetal bovine elastic chondrocytes. The PKC involved in the stabilization pathway was unresponsive to phorbol ester, clearly distinguishing it from that acting in the chondrocyte system.

Our findings in the present study indicate that PC-specific PLC participates in the signal pathway, rather than PI-specific phospholipase, and that the PLC- subgroup is unlikely to be involved, since in our system the TGF- effect on elastin mRNA stabilization is unaffected by the inhibition of the G-protein-transduction signaling pathway known to be associated with the activation of this PLC isotype (38) . Since inositol trisphosphate would not be released in this reaction, and the PI-specific PLC inhibitor U73122 had no effect, it is unlikely that intracellular Ca²⁺ levels are important in this context. These considerations, in conjunction with the phorbol ester unresponsiveness discussed earlier, suggest that the PKC isotype likely to be involved may belong to the aPKC, Ca²⁺- and DAG-independent class. This possibility is currently under study, as are the mechanisms directly involved in the achievement of message stabilization.