Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The transforming growth factor  (TGF-) family of peptides consists of four closely related proteins (70-80% homology) that are synthesized as larger precursor molecules containing the mature form of TGF- at the carboxy-terminal portion. 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 precursor remainder, designated latency associated peptide, and mature TGF-. In some cases, this complex is secreted bound to another protein termed latent TGF--binding protein (LTBP) (1, 2). The various LTBPs have also been found associated in many cases with the extracellular matrix (3). The TGF-s are multipotent in their biologic activity, modulating cell growth and differentiation as well as extracellular matrix deposition and degradation (4). This wide range of activities is initiated through transmembrane receptors (Types I and II), which possess both constitutive (Type II) and activated (Type I) serine-threonine kinase activity (5). Although the varied biologic activities of the TGF-s have been well documented, the mechanisms through which their multiple effects are transduced remain only partially understood. The discovery that the Smads are key intermediates in the signaling process, linking TGF- receptors to cellular responses, has revealed important new information on the way cells respond to this cytokine (8).

Whereas most Smad-mediated effects are associated with transcriptional regulation, by contrast the TGF- upregulation of the extracellular matrix protein, elastin, is achieved at the post-transcriptional level through mRNA stabilization (12). Our laboratory has begun to investigate the mechanisms by which TGF- regulates elastin expression in cultured human fetal lung fibroblasts as a model for expression in vivo. Previous evidence has suggested that a phosphatidylcholine-specific phospholipase C (PC-PLC), a protein kinase C (PKC), prenylated and acylated protein(s), and one or more tyrosine kinases are required for TGF- to increase elastin expression (13, 15). However, the role of the Smads in this process has not previously been addressed. In the present work, we show that the signaling pathway resulting in elastin mRNA stabilization does, in fact, require the activity of the Smads. In addition, de novo protein synthesis as well as active PKC- and the stress-activated protein kinase (SAPK), p38, are necessary for TGF- to achieve this effect.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

Specific inhibitors were purchased from suppliers as follows: PD98059 from New England Biolabs (Beverly, MA); rottlerin and GÖ 6976, from Biomol Inc. (Plymouth Meeting, PA); goat anti-rabbit peroxidase conjugate and cycloheximide from Sigma Chemical Co. (St. Louis, MO); SB203580 from Cal Biochem (La Jolla, CA).

Cell Culture, RNA, and Tropoelastin Analyses

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 as previously described (13). Before the addition of TGF-1, the medium of confluent cultures was replaced with DMEM containing 1% serum. After incubation with TGF-1_, the supernatants were reserved for tropoelastin quantification by ELISA (13). The standard curve was linear between 1 and 50 ng/ml recombinant tropoelastin (16) and each sample was analyzed in triplicate. Total cellular RNA was extracted by the acid guanidine isothiocyanate method (17). Fifteen micrograms of RNA 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 (18). 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 analyzed by phosporimaging and results were quantified to determine the relative amounts of mRNA (Image-Quant V3.1 software; Molecular Dynamics, Sunnyvale, CA). Elastin mRNA values were analyzed in duplicate and normalized to equivalent values for GAPDH to compensate for loading and transfer.

Electroporation of Fetal Lung Fibroblasts

We have found that general transfection strategies are not feasible in the present situation because we are examining expression of the endogenous elastin gene, and transfection efficiencies by the classic CaPO₄ method are too low and more efficient lipid-based transfection agents perturb the TGF-1 response in the lung fibroblasts. We have found, however, that the cells can be efficiently transfected by electroporation while maintaining the TGF-1 response. We have optimized the conditions for electroporating our cells (250 volts, 500 µF) and routinely obtain ~ 40% transfection. Ten million cells in 200 µl were co-transfected with 5 µg of plasmid expressing green lantern protein under control of the CMV promoter and 15 µg of plasmid expressing Smad7 also under control of the CMV promoter. The electroporated cells were incubated for 24 h to allow for recovery and then incubated for a 16-h period either in the absence or presence of TGF-1. After the cells were recovered by trypsinization and segregated by fluorescence activated cell sorting (FACS), total RNA isolated from the fractionated cell populations was analyzed by Northern hybridization.

Statistical Analysis

Where indicated, numerical data are presented as means ± SE from replicate cultures. Statistical evaluation was performed using a paired Student's t test. The mean values were said to be significantly different when the probability of the differences, assuming the null hypothesis to be correct, fell below 5% (i.e., P < 0.05).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Role of Smads in TGF- Stabilization of Elastin mRNA

Many of the effects of TGF- are mediated through the activities of the Smads acting as second messengers, as appears to be the case in the stimulation of expression of extracellular matrix genes such as COL1A2 (19). In most instances, TGF- results are achieved at the transcriptional level by Smad complexes acting in concert with other transcription factors (20). However, we and others have shown that TGF- dramatically increases elastin expression by stabilization of elastin mRNA (12). Therefore, it was of some interest to determine whether the Smads are involved in TGF- regulation of elastin expression. To answer this question, we co-transfected, by electroporation, a vector expressing Smad7 along with a vector expressing green lantern protein into human fetal lung fibroblasts. After allowing the cells to recover, they were incubated in the presence or absence of TGF- and then the transfected cells were separated from the nontransfected ones by FACS. Total RNA from the fractionated cell populations was analyzed by Northern hybridization. These determinations demonstrated that the Smad7 construct markedly inhibited (> 75%) the TGF-1 stimulation of elastin expression (Figure 1). Interestingly, the constitutive level of elastin expression (no TGF-1 treatment) was also depressed by Smad7 expression. This may be due to low levels of active TGF- in the serum-containing culture media or Smad involvement in basal expression. Northern analysis confirmed that Smad7 was being expressed in the transfected cells (data not shown). Cells transfected with Smad6, an inhibitor of bone morphogenetic protein action, did not alter the response to TGF-1 (data not shown).

View larger version (18K): [in this window] [in a new window]   Figure 1.   Effect of Smad7 on the TGF1 modulation of elastin expression in human fetal lung fibroblasts. Fibroblasts were electroporated as described in MATERIALS AND METHODS, with plasmids expressing green lantern protein (5 µg) and Smad7 (15 µg) and cultured for 24 h. After the cultured cells were incubated with or without 3 ng/ml TGF-1 for a further 16 h, the cells were sorted by FACS, and total RNA isolated from the fractionated cell populations was analyzed by Northern hybridization. The values represent the mean of three separate experiments and the standard deviation of these values. A representative Northern blot is shown. The Smad7 value plus TGF-1 differs from the TGF-1 alone at the P < 0.01 level while the Smad7 alone differs from the control (no Smad7, no TGF-1) at the P < 0.01 level.

Requirement for De Novo Protein Synthesis

Many of the transcriptional events initiated by TGF- can be detected within a matter of minutes after the addition of the cytokine. However, ~ 5-6 h are required before any increase in elastin mRNA can be observed (13). This long delay suggests that, unlike many of the transcription-mediated events that do not require de novo protein synthesis, elastin mRNA stabilization may require synthesis of previously unexpressed protein. To determine whether this is so, fibroblasts were preincubated for 1 h before the addition of TGF-1 with the protein synthesis inhibitor, cycloheximide, at a concentration (25 µg/ml) which achieved greater than 95% inhibition of protein synthesis. These experiments showed that cycloheximide completely blocked the action of TGF-1 (Figure 2). Therefore, we conclude that de novo protein synthesis is required for achievement of elastin mRNA stabilization.

View larger version (16K): [in this window] [in a new window]   Figure 2.   Effect of the protein synthesis inhibitor, cycloheximide, on the TGF-1 stimulation of elastin expression. Cycloheximide (25 µg/ml) was added to confluent cultures of lung fibroblasts 1 h before the addition of TGF-1 to a final concentration of 3 ng/ml. After 16 h, RNA was prepared from control and treated cultures and subject to Northern analysis as described in MATERIALS AND METHODS. The values represent the average of determinations from duplicate cultures.

The Role of Protein Kinase C

We had previously found that the pan-specific PKC inhibitors, calphostin and staurosporin, virtually abolished the increased expression of elastin caused by TGF-1 in the lung fibroblasts, strongly indicating the involvement of a PKC in the signaling pathway (13). To explore this possibility further, the effects of several more-specific PKC inhibitors were tested. These included GÖ 6976 which inhibits the classic diacylglycerol, Ca⁺⁺-dependent group of PKCs (21). However, GÖ 6976 had minimal effect on the TGF-1 response (Figure 3). In contrast, rottlerin, which has been reported to be a quite specific inhibitor of PKC- (IC₅₀ 3.6 µM; all other PKCs are inhibited only at concentrations an order of magnitude or more higher) (22) when used at the proper low concentrations, proved to be a very effective inhibitor of the TGF-1 stimulation of elastin expression. It blocked the increased expression at both the protein and mRNA levels in a dose-dependent fashion (Figure 3), producing > 60% inhibition at 2 µM. This result suggested that PKC- may be an essential component in the TGF-1-initiated signaling pathway leading to elastin mRNA stabilization. The slope of the inhibition curve was somewhat greater at the protein level compared with the mRNA level (Figure 3), suggesting that PKC- may also be involved in the regulation of the synthesis of tropoelastin.

View larger version (28K): [in this window] [in a new window]   Figure 3.   Effect of the PKC- inhibitor, rottlerin, and GÖ 6976, an inhibitor of the classic diacylglycerol, Ca++ dependent group of PKCs, on the TGF-1 stimulation of elastin expression. Confluent cultures of human fetal lung fibroblasts were preincubated for 2 h with the indicated concentrations of inhibitors and then TGF-1 was added to a final concentration of 3 ng/ml and the incubation continued for 16 h. Total RNA was prepared from the cells and subject to Northern analysis using a human elastin cDNA probe as described in Materials and Methods. The values represent the average of determinations from duplicate cultures. Basal level refers to the values obtained in the absence of TGF-1. The incubation media obtained after 16 h were analyzed for tropoelastin with a specific ELISA as described previously (13). Values represent the average of duplicate cultures determined in triplicate. Basal level refers to the values obtained in the absence of TGF-1.

Effect of Mitogen-Activated Protein Kinase Inhibitors

Because TGF- has been reported to activate the mitogen-activated protein kinase (MAPK) pathway (23), and because we have previously shown that SAPK p38 may be involved in the stimulation of fibronectin expression by TGF-1 in lung fibroblasts (26), we tested for the possible participation of MAPK in the present case. Cells were pretreated before incubation with TGF-1 with the specific inhibitors PD98059, which inhibits MEK1 and prevents the phosphorylation of erk1 and erk2 (27) or with SB203580, which inhibits SAPK p38 activity (28). PD98059 had no effect, whereas SB203580 inhibited the response in a dose-dependent manner, producing 50% inhibition at 3 µM (Figure 4). These results suggest that activation of erk1 and erk2 of the MAPK system is not necessary for the TGF--mediated response, whereas p38 activity is essential.

View larger version (26K): [in this window] [in a new window]   Figure 4.   Effect of MAPK inhibitors on the TGF-1 modulation of elastin expression. Confluent cultures of human fetal lung fibroblasts were incubated for 2 h with the various indicated concentrations of PD98059 or SB203580 and TGF-1 was then added to a final concentration of 3 ng/ml. The incubation media obtained after 16 h were analyzed for tropoelastin with a specific ELISA. Values represent the average of duplicate cultures determined in triplicate. Total RNA was prepared from the cells and subject to Northern analysis using a human elastin cDNA probe as described in Materials and Methods. The values represent the average of determinations from duplicate cultures.

Length of Time Required for De Novo Protein Synthesis to Achieve the TGF-1 Response

The findings that Smad participation is essential and that de novo protein synthesis is also required suggested that Smad signaling involved immediate early, probably transcriptional, events. Although this possibility could not be tested directly, it was important to determine the time frame in which the required new proteins were being synthesized. This was accomplished by adding cycloheximide at various times after the addition of TGF-1. These experiments demonstrated that de novo protein synthesis was required for an extended period of time after the addition of TGF-1 to achieve a maximal response (Figure 5). Only if cycloheximide was added 4 to 6 h after TGF-1 was maximal stimulation obtained.

View larger version (31K): [in this window] [in a new window]   Figure 5.   Effect of adding cycloheximide at various times after the addition of TGF-1 on the expression of elastin. TGF-1 was added to replicate cultures at zero time and cycloheximide was added to a final concentration of 25 µg/ml at the indicated times either before or after the addition of TGF-1. Sixteen hours after the addition of TGF-1, 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.

Requirement for PKC- and p38 Activities for Extended Time after TGF- Addition

We have previously observed in these cultured lung fibroblasts that the phosphorylation of p38 after the addition of TGF-1 is biphasic. An initial rapid phosphorylation reaches a maximum by 5 min, but declines to baseline by 10 min. This is followed by a second higher level of phosphorylation which reaches a maximum at 2 h, but remains elevated for at least 4 h. This finding, combined with the 4- to 6-h delay from the time of addition of TGF-1 before any increase in elastin mRNA is observed, suggested that extended activity of p38 and possibly PKC- might be required to effect a robust response in elastin expression by TGF-1. To test this possibility, SB203580 or rottlerin were added at various times (0 to 10 h) after the addition of TGF-1 and elastin expression was determined 16 h after TGF-1 addition. These experiments demonstrated that active p38 and PKC- were both required for several hours after the addition of TGF-1 to achieve an effective response. In the case of p38, maximal inhibition was observed even when SB203580 was added 4 h after the addition of TGF-1 and the response did not approach the no-inhibitor, control value until the drug was added 10 h after TGF-1 (Figure 6). Similar results were observed when PKC- was inhibited. It was only when rottlerin was added 4 h after TGF-1, and not before, that stimulation of elastin expression increased dramatically, rising thereafter to the control value (Figure 7).

View larger version (32K): [in this window] [in a new window]   Figure 6.   Effect of adding SB203580 at various times after the addition of TGF-1 on the expression of elastin. TGF-1 was added to replicate cultures at zero time and SB203580 was added to a final concentration of 25 µM at the indicated times, either before or after the addition of TGF1. Sixteen hours after the addition of TGF-1, 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.

View larger version (33K): [in this window] [in a new window]   Figure 7.   Effect of adding rottlerin at various times after the addition of TGF-1 on the expression of elastin. The protocol for this experiment was identical to that described in Figure 6 except 3 µM rottlerin was used instead of SB203580.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Because the TGF- family is involved in the regulation of a wide variety of fundamental cell processes, it is to be expected that the signaling mechanisms by which these effects are achieved will be complex. The identification of the Smad family of proteins as critical intermediates in TGF- signal transduction has contributed greatly to our understanding of TGF- action. The preponderant body of information has documented and analyzed the role of the Smads in mediating alterations in gene transcription. To determine whether the Smads are involved in the stabilization of elastin mRNA, we transfected a construct expressing inhibitory Smad7 that has the capacity to block the signaling activity of receptor-regulated Smads2 and 3, and Co-Smad4, in response to TGF-1 (29). Expression of inhibitory Smad7 dramatically diminished the TGF-1-mediated increase in elastin expression (Figure 1). In contrast, expression of Smad6, an inhibitor of bone morphogenetic protein (BMP) action that does not, however, block the activity of TGF-, had no effect. This result strongly suggests that TGF--specific Smads are required for elastin mRNA to be stabilized. However, it is probable that the Smads are not directly involved in this process since there is a 4- to 6-h lag period before any change is observed in the steady state level of elastin mRNA, and de novo protein synthesis is required for TGF-1 to be effective (Figure 2). A likely scenario is that expression of one or more proteins essential to the stabilization mechanism are induced through Smad activity and synthesized during the lag period. Furthermore, the data presented in Figure 5 using cycloheximide indicates that synthesis of the newly induced protein must continue for at least 4 h of this lag period to achieve a maximal response. Figure 8 presents a working model summarizing our findings of events essential for TGF- to stimulate elastin expression. and formulating them into a sequential scenario.

View larger version (37K): [in this window] [in a new window]   Figure 8.   Schematic illustration of the signaling events required for TGF- to stabilize elastin mRNA. Where indicated, particular steps are inhibited by designated drugs (see text for discussion).

Although the scheme depicted in Figure 8 is provisional, emerging evidence indicates that cellular components in addition to the Smads participate in TGF--initiated signaling events and that several separate pathways can converge to modulate gene expression (20). Our previous studies have suggested that a small GTPase belonging to the Ras superfamily, a phosphatidylcholine-specific phospholipase C, as well as a PKC, are required constituents in the signaling pathway leading to elastin mRNA stabilization (13, 15). The present work using a more specific inhibitor, rottlerin, extends these findings by provisionally identifying PKC- as the PKC likely to be involved (Figure 3). TGF--activated kinase (TAK1) has been shown to be involved in TGF- signaling and it is known to activate SAPKs including p38 through the activity of MKK6 or MKK3 and possibly through MKK4 (30). Thus, our finding that SB203580, a relatively specific inhibitor of p38, can effectively block the response to TGF-1 in a dose-dependent manner (Figure 4) is consistent with an essential role for p38 in the lung fibroblast system. Significantly, based upon inhibitor studies (Figures 6 and 7), the requirement for active PKC- and p38 appears to extend for hours after the addition of TGF-. This observation raises several interesting questions concerning the role of these kinases in the elastin mRNA stabilization process, about which very little is known. For example, are they acting directly or through downstream effectors that need to be identified? Are the kinases simply required for late downstream signaling events or are they more intimately involved in the stabilization mechanism? Work is currently in progress to address these questions as well as to identify those cis elements within the elastin transcript and their cognate trans-acting factors that are involved in regulation of elastin mRNA stability.