FASEB J. Cell Migration Consortium
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by DUNCAN, M. R.
Right arrow Articles by GROTENDORST, G. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by DUNCAN, M. R.
Right arrow Articles by GROTENDORST, G. R.
(The FASEB Journal. 1999;13:1774-1786.)
© 1999 FASEB

Connective tissue growth factor mediates transforming growth factor ß-induced collagen synthesis: down-regulation by cAMP

MATTHEW R. DUNCAN, KEN S. FRAZIER1, SUSAN ABRAMSON*, SHAWN WILLIAMS, HELENE KLAPPER, XINFAN HUANG{dagger} and GARY R. GROTENDORST2

Department of Cell Biology and Anatomy, University of Miami School of Medicine, Miami, Florida 33136 USA;
{dagger} FibroGen, Inc., South San Francisco, California 94080, USA; and
* Cleveland Clinic Florida, Ft. Lauderdale, Florida 33309, USA

2Correspondence: Department of Cell Biology and Anatomy, University of Miami School of Medicine (R-124), 1600 NW 10th Ave., Miami, FL 33136, USA. E-mail: ggrotend{at}mednet.med.miami.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Connective tissue growth factor (CTGF) is a cysteine-rich peptide synthesized and secreted by fibroblastic cells after activation with transforming growth factor beta (TGF-ß) that acts as a downstream mediator of TGF-ß-induced fibroblast proliferation. We performed in vitro and in vivo studies to determine whether CTGF is also essential for TGF-ß-induced fibroblast collagen synthesis. In vitro studies with normal rat kidney (NRK) fibroblasts demonstrated CTGF potently induces collagen synthesis and transfection with an antisense CTGF gene blocked TGF-ß stimulated collagen synthesis. Moreover, TGF-ß-induced collagen synthesis in both NRK and human foreskin fibroblasts was effectively blocked with specific anti-CTGF antibodies and by suppressing TGF-ß-induced CTGF gene expression by elevating intracellular cAMP levels with either membrane-permeable 8-Br-cAMP or an adenylyl cyclase activator, cholera toxin (CTX). cAMP also inhibited collagen synthesis induced by CTGF itself, in contrast to its previously reported lack of effect on CTGF-induced DNA synthesis. In animal assays, CTX injected intradermally in transgenic mice suppressed TGF-ß activation of a human CTGF promoter/lacZ reporter transgene. Both 8-Br-cAMP and CTX blocked TGF-ß-induced collagen deposition in a wound chamber model of fibrosis in rats. CTX also reduced dermal granulation tissue fibroblast population increases induced by TGF-ß in neonatal mice, but not increases induced by CTGF or TGF-ß combined with CTGF. Our data indicate that CTGF mediates TGF-ß-induced fibroblast collagen synthesis and that in vivo blockade of CTGF synthesis or action reduces TGF-ß-induced granulation tissue formation by inhibiting both collagen synthesis and fibroblast accumulation.—Duncan, M. R., Frazier, K. S. Abramson, S., Williams, S., Klapper, H., Huang, X., Grotendorst, G. R. Connective tissue growth factor mediates transforming growth factor ß-induced collagen synthesis: down-regulation by cAMP.


Key Words: fibrosis • wound repair • granulation tissue • cholera toxin • transgenic


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE PROCESS OF WOUND REPAIR involves a cascade of coordinately linked, overlapping phases that, after disruption of tissue homeostasis, includes inflammation, granulation tissue formation, extracellular matrix deposition and assembly, and termination (1) . Peptide factors appear to orchestrate the repair cascade not only by controlling platelet function, leukotaxis, cytokine synthesis, and angiogenesis, but also by directing the progression of fibroblast phenotypes that ultimately results in the formation of mature scar tissue by regulating the ability of fibroblasts to proliferate and to quantitatively and qualitatively change their extracellular matrix component production profiles. One of the primary regulatory factors involved in initiating the wound healing cascade is transforming growth factor beta (TGF-ß).3 Although TGF-ß is involved in many diverse physiological processes including inflammation, neoplastic progression, cell cycle regulation, and development, its central role in the wound healing cascade has been confirmed by numerous in vitro and in vivo studies. In vitro, the ability of TGF-ß to stimulate the proliferation and extracellular matrix component synthesis of cultured fibroblasts is well documented (2 3 4 5 6 7) . Similarly, injection of TGF-ß into the subcutis of neonatal mice induces the formation of granulation tissue (4 , 8) and TGF-ß or TGF-ß mRNA has been detected at sites of normal wound repair as well as in the lesions of numerous fibrotic disorders (9 10 11 12) . Previous studies have indicated that the TGF-ß stimulation of extracellular matrix synthesis is not shared by other growth factors such as fibroblast growth factor or platelet-derived growth factor (6 , 13) . The observation that protein synthesis is required for TGF-ß to induce extracellular matrix protein gene expression indicates that some other intermediate factor or factors may be involved in these signaling pathways (6) .

One protein with demonstrated potential as a downstream mediator for TGF-ß signaling in fibroblastic cells is the cysteine-rich peptide, connective tissue growth factor (CTGF) (14) . CTGF is not only mitogenic and chemotactic for fibroblasts (15) , but also stimulates the synthesis of at least two extracellular matrix components: type I collagen and fibronectin (8) . Fibroblast CTGF gene expression is strongly induced by TGF-ß but not by other growth factors (16) . Protein synthesis is not required for TGF-ß induction of CTGF gene expression (16 , 17) , and a novel TGF-ß response element has been identified in the CTGF promoter (17) . This novel TGF-ß response element shares partial sequence homology with the consensus sequence for the cAMP response element, and its activation by TGF-ß is inhibited by cAMP analogs and agents elevating intracellular cAMP (14 , 18) . Coordinate expression of TGF-ß and CTGF mRNAs has been observed during wound repair (16) , and CTGF transcripts are induced in skin fibroblasts at the site of intradermal injection of TGF-ß in neonatal mice (8) . Moreover, CTGF injection sites develop granulation tissue histologically similar to that induced by TGF-ß injection (8) . Thus, CTGF mimics many of the fibroblast-activating/matrix-forming activities of TGF-ß both in vitro and in vivo; CTGF gene expression is directly stimulated by TGF-ß, strongly suggesting that CTGF may function as a downstream mediator of TGF-ß action on fibroblasts.

Two of our recent in vitro studies support this hypothesis, demonstrating that CTGF acts as a downstream mediator of one TGF-ß regulated fibroblast function: TGF-ß-induced, anchorage-independent proliferation of a continuous line of cultured normal rat kidney (NRK) fibroblasts (18 , 19) . We investigated whether CTGF is also essential for TGF-ß-induced fibroblast collagen synthesis in vitro and if agents that block CTGF synthesis or action can inhibit TGF-ß-induced granulation tissue formation in vivo.

Our results confirm that CTGF by itself is a potent inducer of cultured fibroblast collagen synthesis and demonstrate that blocking CTGF action with an anti-CTGF antibody or blocking CTGF synthesis with an antisense CTGF gene or agents elevating cAMP prevent TGF-ß-induced collagen synthesis in NRK and human foreskin fibroblast cultures. Using in vivo studies, we found that cAMP-elevating agents can effectively block TGF-ß induction of a CTGF promoter transgene in neonatal murine dermis as well as selectively reduce TGF-ß-induced granulation tissue collagen formation in wound chambers subcutaneously (s.c.) implanted in rats. However, inhibition of fibroblast influx/proliferation may also contribute to the decreased collagen content of granulation tissue, as cAMP elevation significantly decreased the number of fibroblastic cells present at TGF-ß injected dermal sites in neonatal mice. Collectively, these data demonstrate that CTGF mediates TGF-ß-induced collagen synthesis and suggest that CTGF may be a useful target for antifibrotic therapies, as anti-CTGF antibodies and cAMP-elevating agents both have the potential to act as antifibrotic therapeutics by inhibiting CTGF action or TGF-ß-induced CTGF gene expression.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Sources of growth factors and reagents
Recombinant human CTGF was prepared in our laboratory using a baculovirus expression system (8) . Recombinant human TGF-ß1 and TGF-ß2 and CTX were obtained from Gibco BRL (Gaithersburg, Md.) and 8-Br-cAMP and 8-Br-cGMP from Sigma Chemical (St. Louis, Mo.). Chicken anti-human rCTGF was raised against recombinant human CTGF as described previously (19) .

Goat anti-CTGF immunoglobulin G production and purification
Goat anti-human CTGF antibodies were prepared using a standard IACUC approved protocol where the animals were challenged initially (day 1) with 20 µg of pure recombinant human CTGF suspended in 150 µg of Freund's complete adjuvant, followed by subsequent challenges of 20 µg CTGF suspended in 100 µg of Freund's incomplete adjuvant on days 28, 35, 42, 49, 56, 77, 84, etc. The animals were bled on days 42, 49, 56, 63, 91, 98, 105, etc. Anti-CTGF titers were initially screened using Western blot assays and the highest titers of serum were pooled. The total immunoglobulin G (IgG) fraction from the serum was isolated using protein G Sepharose; this material was further purified using CTGF-Affi-Gel 10 as an affinity matrix to yield CTGF-specific IgG.

Cell cultures
NRK fibroblasts, clone NRK-49F, a continuous line of cultured normal rat kidney fibroblasts (20) , were obtained from American Type Culture Collection (ATCC; Rockville, Md.) and from Dr. R. Assoian, University of Pennsylvania. Human foreskin fibroblasts were established from explant cultures. Cell cultures were maintained in Dulbecco's modified Eagle media (DMEM) containing 2.5% fetal bovine serum (FBS) and 2.5% Nu-Serum I (NS) (Collaborative Biomedical, Bedford, Mass.) and passaged prior to confluence. Numerous investigators (2 3 4 5 6 7) have shown that both NRK and foreskin fibroblasts are sensitive target cells for TGF-ß regulation of fibroblast proliferation and collagen synthesis. NRK fibroblasts stably transfected with an antisense CTGF gene (NRK-ASCTGF) were transfected and cloned in our laboratory as described previously (18) .

Assay of fibroblast collagen synthesis
Growth-arrested monolayers of NRK or foreskin fibroblasts were prepared by seeding 10,000 cells/well in 48-well plates and allowing fibroblasts to grow to confluence in 5 to 7 days in DMEM + 2.5% FBS/NS. Fibroblast monolayers were then serum starved in DMEM containing 25 mM HEPES and ITS premix (Collaborative Biomedical) for 1 to 8 days before initiating assays for the effect of CTGF and other biological agents on collagen synthesis. To evaluate the effect of CTGF and other biological agents on collagen synthesis, biological agents along with 10 ng/ml of insulin-like growth factor II and 50 µg/ml ascorbic acid were added to the starved culture media; collagen synthesis during the terminal 24 h of a 48 h treatment period was assessed by measuring 3H-proline incorporation into pepsin-resistant, salt-precipitated extracellular/cell surface-associated collagen using a quantitative assay (21) we have used extensively (22 23 24 25) . Each experimental condition was done in duplicate or triplicate wells; results are expressed as cpm/well ± SE, as the tested reagents caused only minor changes in the cell numbers of these growth-arrested cultures over the 2 day assay period. When comparing cell types with differing confluent cell densities, results are expressed as cpm/103 cells after counting trypsinized cell monolayers. Moreover, the length of growth arrest had no quantitative effect on responses to any test reagent except CTGF, whose collagen synthesis stimulatory activity was most notable in cultures arrested for the longest time periods. We previously noted a similar effect for CTGF induction of fibroblast DNA synthesis (15) .

Western blot analysis
Western blots of supernatant media CTGF were performed as described previously (8) . Briefly, cell culture supernatant media were subject to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 12% acrylamide gels and transferred to nitrocellulose filters by electroblotting. Blots were incubated with 2 µg/ml chicken anti-CTGF antibodies and then with peroxidase-conjugated affinity-purified rabbit anti-chicken IgY (1:1000 dilution, Organon Teknika-Cappel, West Chester, Pa.). Antigens were detected with a chemiluminescent substrate system (ECL, Amersham, Arlington Heights, Ill.).

RNA isolation and Northern blotting
Total cellular RNA was extracted from cell monolayers by a single-step method of RNA isolation (Trizol Reagent, Gibco BRL) and Northern blot analysis was performed largely as described previously (8 , 25) . Hybridization signals were compared using a scanner and quantitated based on a standard area of pixels after normalizing for 18s rRNA controls. Double-stranded cDNA fragments used for probes were labeled by random priming with digoxigenin-modified dUTP and hybridization signals were detected with an anti-digoxigenin antibody coupled to alkaline phosphatase and a chemiluminescent substrate detection system (Boehringer Mannheim, Indianapolis, Ind.). The CTGF probe was derived from a 1.1 kb human cDNA fragment that encompassed the open reading frame of the CTGF transcript. The pro{alpha}1(I)collagen probe was a 1.5 kb fragment at the 3' end of the open reading frame cleaved from plasmid Hf677 (ATCC).

CTGF promoter/LacZ constructs and transgenic mice development
A 965 bp SacI/XhoI fragment containing the CTGF promoter was cloned into a pSV-ß-galactosidase plasmid (Promega, Madison Wis.). The final construct, PRM1, contained the CTGF promoter, the bacterial ß-galactosidase gene, and an SV40 polyadenylation signal, and was digested with SmaI/BamHI and separated from the plasmid by agarose gel electrophoresis prior to injection into pronuclei. The final 897 bp CTGF promoter sequence was composed of bases -823 to +74 relative to the transcription start site and is identical to the P0 fragment of the human CTGF promoter described previously that exhibits a strong induction by TGF-ß (17) .

The PRM1 DNA fragment were submitted to the University of Miami Transgene Facility, and pronuclear injections and breeding were performed routinely (26) . Briefly, ova were obtained from superovulated C57BL/6J females mated to SJL males. Transgenic mice were generated by introducing 1.5 ng/ml of linearized DNA into one of the pronuclei of recently fertilized mouse ova. Zygotes were transferred into the oviducts of pseudopregnant B6SJL mice. Tail genomic DNA was obtained using prescribed techniques (27) , and Southern blotting was performed routinely using a probe consisting of a 3 kb HindIII fragment of the original plasmid. Two positive founder mice were separated and each line was bred to homozygosity. The resulting PRM1 transgenic lines were named TgN(CTGGLRp)1Grg and TgN(CTGGLRp)2Grg according to the standardized nomenclature for transgenic animals (28) . Although both lines expressed detectable levels of lacZ at sites of endogenous CTGF expression in the embryos, the TgN1Grg line exhibited a more intense signal and was used for the studies reported here.

In vivo CTGF promoter/LacZ transgene induction studies
One-day-old transgenic neonatal mice were injected once daily for 3 days with 800 ng TGF-ß in a 20 µl saline vehicle in the dermis at the nape of the neck and simultaneously with 800 ng TGF-ß and 100 ng of cholera toxin in the dermis at a site on the top line over the lumbar vertebrae. Transgenic neonatal mice were assayed for ß-galactosidase activity as described previously, with X-gal staining producing a grossly visible blue stain in the cells and tissues expressing lacZ (29) . Briefly, embryos and neonates were fixed for 1 h in paraformaldehyde, then washed three times for 30 min each in phosphate-buffered saline containing 2 mM MgCl2, 4 mM potassium ferrocyanide, 5 mM potassium ferricyanide, 0.2% Nonidet P-40, and 0.1% deoxycholate. After X-gal staining, the neonates were washed in phosphate-buffered saline and soaked in formalin for 24 h and 70% ethanol for 24 h.

Wound chamber studies
Wound chamber studies were performed using the previously described Schilling-Hunt wound chamber model (30 31 32 33) . Briefly, a 1 cm longitudinal incision was made on the middle of the dorsum of anesthetized and shaved male Sprague-Dawley rats (300 g). After blunt scissor dissection beneath the panniculus carnosus on both sides of the incision, wound chambers (1x3 cm), with medical grade silicone elastomer plugs on each end, were implanted into the two dorsal s.c. pockets. The chambers used in these experiments were composed of high-density nylon mesh, which does not induce any foreign body reaction. Animals were randomized into eight treatment groups; beginning on day 0, six chambers per treatment group were injected daily with either saline/1% bovine serum albumin control vehicle, CTX (0.1 and 1.0 mg), or 8-Br-cAMP (10 and 100 mg) with or without 1 µg of TGF-ß2, which was injected every other day starting on day 1. This schedule was maintained for 14 days, on which day the animals were killed and the chambers removed by dissection. All tissue was carefully removed from the outside of the chamber and the chamber contents were solubilized by pepsin digestion. Aliquots of the pepsin digests were then subjected to SDS-PAGE and Coomassie staining, and compared by densitometry to Coomassie stained collagen standards to biochemically quantitate chamber collagen content. Results were averaged for each treatment group and expressed as mg of collagen/chamber ± SE.

s.c. Administration of factors to mice
Neonatal NIH Swiss mice were injected in the nape of the neck daily for 3 days with 20 µl total volume of either saline control, 800 ng TGF-ß, or 400 ng CTGF with or without the addition of 10, 50, or 100 ng of cholera toxin to the injection mixture (4 , 8) . At least three mice were used for every injection set. Animals were killed 24 h after the last injection and large biopsies of the area containing and surrounding the injection site were removed. Sections were prepared for histopathologic examination after routine formalin fixation, processing, and staining with either Mayer's hematoxylin and eosin or tissue-specific Giemsa. Immunohistochemical staining with rabbit anti-mouse CD45 (leukocyte common antigen) was performed routinely on additional 5 µM sections to further delineate monocytes from fibroblasts.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Antisense CTGF gene transfection and anti-CTGF IgG block TGF-ß-induced collagen synthesis by cultured NRK and human foreskin fibroblasts
We previously observed and reported that CTGF stimulates pro{alpha}1(I)collagen mRNA expression and type I collagen protein synthesis in NRK fibroblasts when RNA expression is assessed by Northern blotting and collagen synthesis by PAGE analysis after [35S]methionine biosynthetic labeling (8) . However, such data are not quantitative and we have now analyzed the effect of CTGF on NRK fibroblast synthesis using a quantitative microwell assay (21) . The dose-response curve presented in Fig. 1 demonstrates that CTGF can stimulate large increases in collagen synthesis over a range of low concentrations from 0.04 to 1 ng/ml, with maximal activity observed at only 1 ng/ml (27 pM) of CTGF. Higher CTGF concentrations of up to 25 ng/ml did not further stimulate collagen synthesis (data not shown). Although we have not investigated here the individual collagen types whose synthesis is stimulated by CTGF, our previous report indicates that type I collagen, the major collagen type produced by NRK fibroblasts, is stimulated by CTGF treatment (8) . Whether CTGF stimulates the synthesis of other collagen types and their contribution to the collagen levels measured by our quantitative microwell assay remain to be investigated. Since the levels of collagen synthesis induced by 1 ng/ml or less of CTGF were comparable to that induced by 5 ng/ml TGF-ß (Fig. 1) and CTGF gene expression is known to be induced by TGF-ß (16 , 17) , we wanted to determine whether CTGF was required for TGF-ß stimulation of fibroblast collagen synthesis. To do this, we examined TGF-ß-induced collagen synthesis in NRK fibroblasts stably transfected with an antisense CTGF gene (NRK-ASCTGF), which effectively blocks TGF-ß-activated CTGF gene expression (18) . As shown in Fig. 1 , the collagen synthesis of these NRK-ASCTGF fibroblasts was only minimally stimulated by TGF-ß treatment, suggesting that CTGF synthesis is essential for TGF-ß induction of collagen synthesis.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 1. CTGF and TGF-ß stimulate collagen synthesis in NRK fibroblast cultures and transfection with an antisense CTGF gene blocks TGF-ß-induced collagen synthesis. Cultures of NRK fibroblasts or NRK-ASCTGF fibroblasts in DMEM + ITS were treated for 48 h with listed concentrations of CTGF or TGF-ß1 (5 ng/ml). Collagen synthesis was assessed by measuring 3H-proline incorporation into pepsin-resistant, salt-precipitated extracellular/cell surface-associated collagen during the terminal 24 h of treatment.

We also prepared an anti-CTGF-specific neutralizing antibody by immunizing goats with purified recombinant human CTGF and investigated whether anti-CTGF IgG could block TGF-ß-induced fibroblast collagen synthesis. Using a similar strategy we previously demonstrated that an anti-CTGF antibody raised in chickens blocks the TGF-ß-induced, anchorage-independent growth of NRK fibroblasts (18) . Because our chicken anti-CTGF exhibited only limited neutralizing activity, requiring 500 µg/ml of IgY for 50% neutralization of NRK fibroblast proliferation, we developed the goat anti-CTGF antibody described here, which has similar properties to the chicken anti-CTGF IgY but is 10-fold more potent, causing a 50% neutralization of TGF-ß-induced, anchorage-independent growth of NRK fibroblasts at only 50 µg/ml (data not shown). As shown in Fig. 2 , 50 µg/ml of this affinity-purified goat anti-CTGF IgG almost totally neutralized the collagen synthesis stimulatory activity of 25 ng/ml of CTGF in NRK fibroblast cultures. An equivalent concentration of purified nonimmune goat IgG had no neutralizing activity, and the goat anti-CTGF IgG had little effect on NRK fibroblast basal collagen synthesis. When the goat anti-CTGF IgG was added to the media of NRK fibroblast cultures stimulated with 5 ng/ml of TGF-ß, the anti-CTGF blocked the TGF-ß-induced collagen synthesis in a concentration-dependent manner, causing approximately a 50% neutralization at 10 µg/ml and a complete (100%) inhibition at 50 µg/ml (Fig. 3 A). Similar results were obtained when goat anti-CTGF was added to TGF-ß stimulated human foreskin fibroblast cultures (Fig. 3B ). These data demonstrate that CTGF mediates TGF-ß-induced fibroblast collagen synthesis.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. Goat anti-CTGF IgG neutralizes CTGF stimulation of collagen synthesis in NRK fibroblast cultures. Cultures in DMEM + ITS were treated for 48 h with CTGF (25 ng/ml), goat anti-CTGF IgG (50 µg/ml), or nonimmune goat IgG (50 µg/ml), singly or in combination, and collagen synthesis was assessed by 3H-proline incorporation during the terminal 24 h of treatment.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 3. Goat anti-CTGF IgG blocks TGF-ß stimulation of collagen synthesis in NRK and human foreskin fibroblast cultures. A) NRK fibroblast cultures in DMEM + ITS were treated for 48 h with TGF-ß1 (5 ng/ml) ± goat anti-CTGF IgG (5–100 µg/ml) or nonimmune goat IgG (100 µg/ml). B) Foreskin fibroblasts in DMEM + ITS were treated for 48 h with TGF-ß1 (5 ng/ml) ± goat anti-CTGF IgG (25 µg/ml) or nonimmune goat IgG (25 µg/ml). Collagen synthesis was assessed by 3H-proline incorporation during the terminal 24 h of treatment.

Elevation of intracellular cAMP blocks TGF-ß-induced collagen synthesis by cultured NRK and human foreskin fibroblasts
We previously reported that elevation of intracellular cAMP levels can block the DNA synthesis needed for cell proliferation in TGF-ß-stimulated NRK fibroblasts by suppressing CTGF gene expression (19) . To determine what effect suppressing CTGF gene expression by elevating levels of cAMP has on TGF-ß-induced collagen synthesis, we examined growth-arrested cultures of NRK and foreskin fibroblasts. As detailed in Fig. 4 A, 100 ng/ml of the adenylyl cyclase activator, cholera toxin (CTX), and a membrane-permeable analog of cAMP, 8-Br-cAMP (1 mM), inhibited TGF-ß (5 ng/ml) -induced collagen synthesis in NRK fibroblast cultures by 71% and 76%, respectively. This inhibitory effect appeared to be selective for agents elevating cAMP as 1 mM 8-Br-cGMP did not decrease TGF-ß stimulation of collagen synthesis. Moreover, CTX and cAMP had no effect on basal collagen synthesis, suggesting that only TGF-ß-induced collagen synthesis is subject to cAMP-mediated down-regulation. These results were confirmed in several additional experiments with NRK fibroblasts as well as with two separate strains of human foreskin fibroblasts (Fig. 4B ).



View larger version (19K):
[in this window]
[in a new window]
  		Figure 4. CTX and cAMP inhibit TGF-ß-stimulated collagen synthesis in NRK and human foreskin fibroblast cultures. A) NRK fibroblast cultures. B) NRK fibroblasts and human foreskin fibroblast (hFSF) cultures showing three repeat NRK fibroblast studies, and studies with two different hFSF strains. Cultures in DMEM + ITS were treated for 48 h with CTX (100 ng/ml), 8-Br-cAMP, or 8-Br-cGMP (1 mM) with or without TGF-ß1 (5 ng/ml); collagen synthesis was assessed by 3H-proline incorporation during the terminal 24 h of treatment.

To determine whether levels of CTGF production correlated with fibroblast collagen synthesis, as we have previously found in NRK fibroblast proliferation studies (19) , we compared foreskin fibroblast pro{alpha}1(I)collagen and CTGF mRNA levels by Northern blotting and assayed supernatant media for CTGF peptides by Western blotting. The data presented in Fig. 5 for one strain of foreskin fibroblasts show that after 24 h of treatment, TGF-ß-induced increases in the 4.8 and 5.8 kb mRNA transcripts for pro{alpha}1(I)collagen. These increases were paralleled by enhanced levels of the 2.4 kb CTGF mRNA transcript and immunoreactive 38 kDa CTGF peptides. Moreover, 8-Br-cAMP almost completely inhibited the TGF-ß induction of CTGF peptides and mRNA transcripts and caused a 50–60% decrease in pro{alpha}1(I)collagen mRNA transcripts. Taken together, these data indicate that elevated cAMP levels selectively inhibit TGF-ß-induced fibroblast collagen synthesis via transcriptional mechanisms, depressing CTGF production. Consequently, TGF-ß-induced fibroblast collagen synthesis appears to be dependent on the direct action of increased levels of newly synthesized CTGF.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 5. 8-Br-cAMP blocks TGF-ß-induced increases in pro{alpha}1(I)collagen and CTGF mRNA transcripts and CTGF protein synthesis in human foreskin fibroblast cultures. Foreskin fibroblast cultures were treated in DMEM + ITS for 24 h with 5 ng/ml of TGF-ß1 with or without 1 mM 8-Br-cAMP. A) Northern blot analysis for mRNA transcripts and B) Western blot analysis of media supernatant CTGF protein were performed as described in Materials and Methods. Western blot CTGF standard (10 ng/lane).

In our previously reported studies investigating NRK fibroblast mitogenesis, we were able to confirm the requirement for CTGF in TGF-ß-stimulated mitogenesis by overcoming the cAMP or CTX blockade of TGF-ß-induced proliferation by the addition of exogenous CTGF to the cultured cells (19) . Accordingly, we tested the ability of CTGF to restore collagen synthesis in the TGF-ß-activated cell cultures that had been treated with either CTX or 8-Br-cAMP. Surprisingly, we could not restore the collagen synthesis, even with CTGF concentrations as high as 25 ng/ml (Fig. 6 ), although CTGF alone in concentrations of 1 ng/ml can clearly induce maximal collagen synthesis (Fig. 1) . This apparent paradox was investigated by assaying the effect elevated intracellular cAMP levels had on collagen synthesis induced by CTGF itself. The data reported in Fig. 6 indicate that in addition to inhibiting collagen synthesis by blocking TGF-ß-induced CTGF gene expression, elevated intracellular cAMP levels also directly suppress CTGF-induced collagen synthesis, as both CTX and 8-Br-cAMP treatment blocked NRK fibroblast collagen synthesis stimulated by treatment with CTGF itself. Thus, cAMP appears to influence TGF-ß stimulation of collagen production in two ways: 1) by inhibiting CTGF synthesis and 2) by blocking the CTGF signaling pathway that regulates collagen synthesis.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 6. cAMP and CTX block CTGF-stimulated collagen synthesis in NRK fibroblast cultures. Cultures of NRK fibroblasts in DMEM + ITS were treated for 48 h with CTGF (25 ng/ml), TGF-ß1 (5 ng/ml), 8-Br-cAMP (1 mM) or CTX (100 ng/ml), singly or in combination, and collagen synthesis was assessed by 3H-proline incorporation during the terminal 24 h of treatment.

CTX inhibits expression of a CTGF promoter/lacZ reporter transgene in vivo
Because the adenylyl cyclase activator CTX was capable of significantly inhibiting the induction of the CTGF gene by TGF-ß in vitro, we wanted to determine whether it could be used in vivo to block induction of the CTGF gene. For this purpose, we used a line of transgenic mice we constructed that contains a transgene composed of the human CTGF promoter controlling the expression of a bacterial ß-galactosidase (lacZ) reporter gene. When stained with X-gal, these transgenic mice produce a grossly visible blue stain in the cells and tissues expressing lacZ. Using in situ hybridization techniques, we have demonstrated that s.c. injection of TGF-ß in the nape of the neck of neonatal mice induced CTGF transcripts in the fibroblasts, forming granulation tissue at the site of injection (8) . We repeated this study with our CTGF promoter/lacZ transgenic mouse line. Neonatal transgenic mice were injected as described previously (4 , 8) with TGF-ß (800 ng) alone or TGF-ß (800 ng) and CTX (100 ng) daily for 3 days. Individual mice were injected with the two mixtures at sites separated by 1 cm so that each individual could serve as an internal control. The mice were killed 6 h after the last injection and stained with X-gal to detect expression of the transgene. In the region of the TGF-ß injected skin, a circumscribed area of blue stain was grossly visible through the epidermis, and the dermis and superficial subcutis exhibited blue staining granulation tissue (Fig. 7 A–C). In contrast, the site injected with both TGF-ß and CTX lacked any detectable blue staining (Fig. 7A ). Saline-injected transgenic controls did not produce granulation tissue at the site, and the dermis and subcutis did not stain (data not shown). These results indicate that CTX can effectively inhibit CTGF expression in vivo, as predicted from our cell culture experiments, and suggest that our transgenic mice may be a convenient model with which to study regulation of CTGF gene expression in vivo.



View larger version (101K):
[in this window]
[in a new window]
  		Figure 7. CTX inhibits TGF-ß-induced expression of a CTGF promoter/lacZ reporter transgene. Neonatal TgN1Grg mice were injected intradermally with 800 ng of TGF-ß1 and 800 ng TGF-ß1 + 100 ng CTX daily for 3 days and stained with X-gal to detect CTGF expression. A) Note blue stain evident at area of reflected epidermis in TGF-ß1 injection site (arrow) vs. negative stain in dermis of site injected with TGF-ß1 + CTX and zone of clear skin surrounding it (arrowhead). B) Low-power microscopic view of section of TGF-ß injected dermis, stained with X-gal and cut with vibratome at 50 µM. Positive blue staining is evident at arrows in granulation tissue under break in epidermis caused by hypodermic needle penetration. C) High-power view of sectioned dermis injected with TGF-ß. Arrow marks positive staining.

CTX and 8-Br-cAMP selectively block TGF-ß-induced collagen deposition in vivo
To quantitatively evaluate the effects of CTX and cAMP analogs on TGF-ß stimulation of extracellular matrix formation in vivo, we used the Schilling-Hunt wound chamber model (30 , 31) . We and others have used this model to evaluate the ability of various growth factors, including TGF-ß, to influence the ongoing wound repair process that occurs in response to s.c. implantation of the wound chambers (16 , 32 , 33) . Because the chambers are all of uniform size and shape, the tissue formed within the chamber can be harvested precisely and quantitative measurements can be made on the contents. The results of these experiments are illustrated in Fig. 8 . As expected, TGF-ß injection stimulated a quantitative increase of 42% in the amount of collagen present within harvested chambers compared to control saline-injected chambers. Coinjection of either 1 µg of CTX or 10 µg or 100 µg of 8-Br-cAMP with TGF-ß completely blocked the increase in collagen deposition induced by TGF-ß, reducing collagen deposition to levels comparable to or less than saline-injected controls. The collagen levels in chambers treated with either 0.1 or 1 µg of CTX without TGF-ß were relatively unaffected and exhibited no more than an 18% reduction from saline-injected control chambers. These results demonstrated that elevation of in vivo fibroblast cAMP levels totally inhibits the wound chamber collagen deposition induced by exogenous TGF-ß. Elevation of cAMP appeared to only partially suppress the collagen deposition that occurs in the absence of exogenous TGF-ß. This suggests that the collagen deposition occurring in untreated wound chambers is only partially induced by endogenous TGF-ß and is consistent with previous observations that wounds in rats treated with a neutralizing antibody to TGF-ß heal effectively, but have a partially reduced collagen content and a less scar-like appearance (34) . This ability of cAMP elevating agents to selectively inhibit only the collagen synthesis induced by TGF-ß suggests such agents may have potential as antifibrotic therapeutics, since they would effectively inhibit TGF-ß-induced, CTGF-mediated collagen synthesis while sparing both basal and TGF-ß-independent induced collagen synthesis.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 8. CTX and 8-Br-cAMP selectively block TGF-ß-induced wound chamber collagen deposition. The wound tissue isolated from Schilling-Hunt chambers implanted in rats for 13 days and injected with either the saline/bovine serum albumin vehicle or listed doses of TGF-ß2, 8-Br-cAMP, and CTX was analyzed for collagen content, as described in Materials and Methods. Bars are the mean of six chambers ± SE

CTX blocks granulation tissue fibroblast population increases induced by TGF-ß in vivo
Although the wound chamber studies provide an accurate assay of in vivo collagen synthesis, they do not assess the contribution fibroblast influx/proliferation has on granulation tissue collagen formation. In contrast, neonatal mice injection studies do provide an accurate means of assessing the recruitment of fibroblasts during granulation tissue formation but do not allow for practical quantitative biochemical analysis of matrix components due to difficulties in obtaining uniform samples. Since our previous studies indicate that both TGF-ß and CTGF stimulate a fibroblast-rich granulation tissue when injected into the dermis of neonatal mice, we wanted to determine whether CTX could inhibit this effect of TGF-ß. The formation of granulation tissue in this model of early wound repair involves both new matrix protein synthesis and a simultaneous increase in the matrix fibroblast population by both chemotactic recruitment and subsequent proliferation of fibroblasts at the site, all of which are demonstrated biological activities of CTGF (4 , 8 , 15) . The results of our study are presented as hematoxylin- and eosin-stained sections in Fig. 9 and as quantitated data counting the number of fibroblasts and inflammatory neutrophils present in Fig. 10 . Monocytes were differentiated from fibroblasts by immunohistochemistry for CD45 (leukocyte common antigen) and the number of monocytes was found to be similar in all injection samples, with counts averaging 1–3 per high-power field. As shown, intradermal injection of 800 ng of TGF-ß induced accumulation of a loose extracellular matrix and the presence of a significant fibroblast population without substantial inflammatory cell infiltration (Fig. 9A , Fig. 10 ). When 100 ng of CTX was injected, a different response, consisting of neutrophilic inflammation with edema and few fibroblasts, was observed (Fig. 9B , Fig. 10 ). Since CTX is a bacterially derived protein used as a vaccine adjuvant due to its leukotactic and lymphokine-inducing activities, the leukocytic infiltrate was expected (35) . At a dose of 100 ng of CTX, the neonates showed no adverse systemic clinical signs, had full stomachs at necropsy, and were similar in size to noninjected littermates. When 800 ng TGF-ß and 100 ng CTX were coinjected, the markedly increased fibroblast population induced by TGF-ß was greatly reduced and the histological appearance more closely resembled that injected with CTX alone (i.e., a neutrophilic infiltrate but few fibroblasts) (Fig. 9C , Fig. 10 ). We also injected doses of 10 ng and 50 ng of CTX with 800 ng of TGF-ß and obtained a relative dose response in the number of fibroblasts present at the site (data not shown).



View larger version (104K):
[in this window]
[in a new window]
  		Figure 9. CTX blocks granulation tissue fibroblast population increases induced by TGF-ß. Photomicrographs show high-power (400x) fields of sections of skin and subcutis from NIH Swiss neonatal mice after injection of TGF-ß1, CTGF, and/or CTX daily for 3 days. A) 800 ng of TGF-ß1. Note large fibroblast population (arrows). B) 100 ng of CTX. Note many infiltrating neutrophils (arrowheads) and minimal fibroblasts (arrow) as compared to panel A. C) 800 ng TGF-ß1 and 100 ng CTX. Note increased neutrophils as in panel B, but decreased fibroblast numbers from panel A. D)400 ng CTGF and 100 ng CTX. Note large fibroblast and neutrophil populations. E) 800 ng TGF-ß1, 100 ng CTX, and 400 ng CTGF. Note large population of fibroblasts and numerous inflammatory neutrophils.



View larger version (23K):
[in this window]
[in a new window]
  		Figure 10. Quantitation of granulation tissue fibroblast populations present at dermal/subcuticular injection sites in neonatal mice. Numbers of neutrophils and fibroblasts present in sections of injected tissue illustrated in Fig. 9 were quantitated per 400x high-power field (HPF) averaged for 20 multiple fields. Injections were performed as described in Fig. 9 .

To determine whether CTX was acting exclusively by inhibiting CTGF gene expression or was also blocking CTGF activity, we coinjected 400 ng of CTGF with CTX. The resulting granulation tissue (Fig. 9D , Fig. 10 ) contained the expected neutrophilic infiltrate and an enhanced number of fibroblasts comparable to sites injected with CTGF alone (8) . This indicates that CTX does not directly block the ability of CTGF to increase fibroblast population numbers during the formation of granulation tissue. Accordingly, when CTGF was coinjected with TGF-ß and CTX, it prevented CTX from down-regulating TGF-ß-stimulated fibroblast population numbers by replacing the endogenous CTGF whose synthesis was blocked by CTX injection (Fig. 9 E, Fig. 10 ). Thus, it appears that cAMP-elevating agents inhibit TGF-ß-induced chemotactic and mitogenic responses that lead to increased in vivo granulation tissue fibroblast populations solely by blocking CTGF gene expression. In contrast, the synthesis of matrix components such as collagen is inhibited by the blockade of both CTGF synthesis and CTGF action.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
A complex network of cell–matrix–cytokine interactions is thought to regulate the initiation and progression of normal wound repair as well as the development of the pathological fibrosis associated with numerous diseases (1 , 11) . Nonetheless, many believe that TGF-ß plays a major role in initiating these complex biological processes, as TGF-ß stimulates both fibroblast proliferation and the synthesis and secretion of several extracellular matrix components (1 2 3 4 5 6 7 8 9 10 11 12) . Consequently, there has been an intense effort to identify agents that selectively block the biological actions of TGF-ß, much of which has focused on TGF-ß signaling pathways. One signaling molecule selectively induced in connective tissue cells by TGF-ß is the 38 kDa cysteine-rich peptide, CTGF (16 , 17) . CTGF mimics many of TGF-ß actions on connective tissue cells, acting as a fibroblast chemoattractant, mitogen, and inducer of extracellular matrix component synthesis (8 , 15 , 18) . In previous studies with cultured fibroblasts, we demonstrated that the blocking of CTGF action with anti-CTGF antibodies or blocking CTGF gene expression with an antisense CTGF gene or with agents elevating cAMP inhibits TGF-ß-induced fibroblast proliferation (18 , 19) .

In the studies reported here, we examined whether blockade of CTGF action or synthesis could also inhibit TGF-ß-induced fibroblast collagen synthesis. Our results demonstrate that similar to cell proliferation, blockade of CTGF action with neutralizing antibodies or inhibition of CTGF synthesis by gene repression significantly reduces fibroblast collagen synthesis induced by TGF-ß. Specific goat anti-CTGF antibodies reduced collagen protein production induced by TGF-ß by 85–100%, demonstrating that in the fibroblast cultures we have studied, TGF-ß induces collagen synthesis exclusively via a CTGF-dependent pathway. Whether TGF-ß induces collagen synthesis in all fibroblasts or other mesenchyme-derived cells, such as chondrocytes and vascular smooth muscle cells, by CTGF-dependent pathways remains to be investigated. We also found that NRK fibroblasts stably transfected with an antisense CTGF gene were virtually unresponsive to TGF-ß induction of collagen synthesis and that agents that block CTGF gene induction by TGF-ß, such as CTX and 8-Br-cAMP, reduced TGF-ß induction of pro{alpha}1(I)collagen mRNA transcripts by 50–60% and reduced collagen (primarily type I) protein production by 70–80%. These data are consistent with previous studies demonstrating that TGF-ß stimulation of collagen gene expression requires protein synthesis (6) . The current work demonstrates that CTGF is one of the proteins that must be synthesized for TGF-ß up-regulation of type I collagen gene expression. Whether CTGF mediates TGF-ß-induced expression of other collagen types or other TGF-ß-regulated fibroblast genes is currently unknown. However, some TGF-ß-inducible effects on fibroblasts appear to be CTGF independent, as CTX and 8-Br-cAMP do not block TGF-ß-induced changes in NRK and NIH/3T3 fibroblast morphology (19) . Such CTGF-independent actions may be mediated by other TGF-ß signaling pathways, such as Smads, which have recently been implicated in TGF-ß activation of the human type VII collagen gene (36) . Nevertheless, elevation of intracellular cAMP levels has been reported to down-regulate the expression of several additional TGF-ß inducible genes, including c-sis, c-myc, collagenase, procathepsin L, laminin, and plasminogen activator inhibitor (13 , 37 38 39 40 41) . Thus, sensitivity to cAMP appears to be a common feature for many TGF-ß-regulated genes. Whether the expression of any of these genes is regulated via a CTGF-dependent signaling pathway, however, remains to be investigated.

Even though this paper appears to be the first to demonstrate that cAMP inhibits TGF-ß-induced collagen synthesis in fibroblasts, several studies have shown that elevation of intracellular cAMP levels with phosphodiesterase inhibitors such as pentoxifylline inhibit fibroblast growth and collagen synthesis induced by serum and/or fibroblast-activating cytokines (23 , 24 , 42) . Similarly, prostaglandin E2, a known inducer of intracellular cAMP, has long been known to inhibit serum-driven fibroblast collagen synthesis (43) . Since pentoxifylline inhibition of fibroblast collagen synthesis in the presence of serum has generally been reported to be greater than in fibroblasts deprived of serum stimulation for 48 h, one potential mechanism consistent with our observations is that pentoxifylline-elevated cAMP levels may be inhibiting collagen synthesis by blocking the induction of CTGF synthesis by TGF-ß contained in serum. This may be the case, as we found that 8-Br-cAMP inhibits the collagen synthesis of NRK fibroblasts cultured in 5% FBS (data not shown). Moreover, pentoxifylline has been reported to inhibit collagen synthesis in animal models of hepatic fibrosis thought to be dependent on TGF-ß for development, demonstrating that in addition to 8-Br-cAMP and CTX, other agents that elevate intracellular levels of cAMP in connective tissue cells can also function as effective in vivo inhibitors of fibrotic tissue formation (44) . The decrease in collagen synthesis caused by pentoxifylline has been linked to the down-regulation of the transcription factor nuclear factor 1 (25) , which has been reported to be required for TGF-ß-induced activation of the type I and III procollagen promoters (45 , 46) . The role of CTGF, whether any, in regulating nuclear factor l expression is currently unknown.

In previous studies where we identified elevated intracellular cAMP as an important repressor of TGF-ß-induced, CTGF gene expression, we were able to overcome the blockade of TGF-ß-induced cell proliferation by cAMP or CTX with the addition of exogenous CTGF to the NRK fibroblast cultures (19) . In contrast to these observations, our current studies revealed that addition of CTGF to NRK fibroblast activated with TGF-ß in the presence of cAMP or CTX did not restore collagen production because elevated intracellular cAMP levels also blocked collagen synthesis directly induced by CTGF itself. Combined, our data suggest that there are cAMP-sensitive differences in the signaling pathways from the CTGF receptor for cell proliferation vs. collagen synthesis, as schematically illustrated in Fig. 11 . The nature of the signaling pathways distal to the CTGF receptor that regulate collagen synthesis and are affected by cAMP remain to be determined. However, they likely involve complex indirect interactions, since no collagen gene promoters are known to contain the cAMP responsive element (47) .



View larger version (14K):
[in this window]
[in a new window]
  		Figure 11. Schematic diagram summarizing cAMP blockade of CTGF-mediated TGF-ß signaling pathways in fibroblastic cells.

Our animal study findings parallel our tissue culture observations and suggest that cAMP-elevating agents may have therapeutic potential as agents selectively inhibiting CTGF-induced fibrosis. Intradermal coinjection of CTX with TGF-ß suppressed TGF-ß activation of a human CTGF promoter/lacZ reporter transgene in transgenic mice and both 8-Br-cAMP and CTX blocked TGF-ß-induced collagen deposition in a wound chamber model of fibrosis in rats. CTX also reduced granulation tissue fibroblast population increases induced by intradermal injections of TGF-ß in neonatal mice, but not increases induced by CTGF or TGF-ß combined with CTGF. Thus, blockade of in vivo CTGF synthesis reduces TGF-ß-induced granulation tissue formation by inhibiting both collagen synthesis and fibroblast accumulation. Whereas the use of two separate in vivo wound repair models was necessary to accurately quantitate granulation tissue collagen content and fibroblast population numbers, their use also allowed us to examine the effects of elevated cAMP levels on granulation tissue formation induced by both short- and long-term TGF-ß treatment. The short-term treatment of three daily TGF-ß intradermal injections used in our neonatal mice studies causes the formation of a loose, fibroblast-rich granulation tissue closely resembling that seen in early normal wound repair. In contrast, the long-term treatment schedule of TGF-ß injections every other day for 14 days used in our wound chamber studies induces wound chambers filled with large quantities of a dense extracellular matrix rich in organized collagen, more closely resembling the matrix observed in tissue derived from sites of fibrotic disease. Our results demonstrate that elevation of cAMP levels blocks granulation tissue formation and matrix collagen accumulation regardless of whether tissue fibroblast exposure to TGF-ß is acute, as occurs in normal wound repair, or is chronic, as likely occurs in fibrotic diseases, provided that cAMP elevating agents are coadministered with TGF-ß.

The progression of wound repair from inflammation to early granulation tissue formation through active matrix synthesis to remodeled mature scar involves a cascade of cytokines, matrix components, and other factors operating in a specific temporal sequence to drive fibroblasts and other wound site cell populations through a series of phenotypic changes (1) . Collectively, the current and previously reported data strongly suggest that after initial release of TGF-ß by activated platelets and infiltrating monocytes, newly synthesized CTGF stimulates a fibroblast chemotactic response and subsequent proliferation of fibroblasts, resulting in a markedly increased wound site fibroblast population. This expanded fibroblast population can then be stimulated by paracrine/autocrine CTGF to produce enhanced quantities of collagen and possibly other matrix components. That TGF-ß synthesis precedes CTGF expression during wound repair has been aptly demonstrated in wound chamber studies in rats (16) . Though TGF-ß was found to be maximally expressed 3 days after wounding, CTGF peak expression did not occur until days 6–9 and persisted through day 15, well into the active matrix synthesis phase of wound repair. Thus, it appears that the prolonged expression of CTGF and its long time course of action allow wound site fibroblasts to continue to accumulate, proliferate and synthesize enhanced quantities of collagen or other matrix components for a considerable time after a relatively brief exposure to TGF-ß. Thus, TGF-ß/CTGF-mediated tissue formation after injury may function in many ways like a developmental process during embryogenesis. That is, brief exposure of responsive cells in an appropriate environment to an inducer is sufficient to initiate a complex biological response, such as limb formation during embryogenesis or, in this case, connective tissue formation after activation with TGF-ß.

The inappropriate or over expression of tissue-associated TGF-ß has long been linked with the development of the pathological fibrosis characteristic of numerous fibrotic disorders (11 , 12) . More recently the presence of CTGF expression, often along with TGF-ß, has also been noted in tissue sampled from several diseases in which fibrosis represents the major or an accompanying pathology. These diseases include scleroderma (48) , keloids and other localized skin fibrotic diseases (49) , atherosclerosis (50) , glomerulosclerosis (51) , pulmonary fibrosis (52) , inflammatory bowel disease (53) , and mammary tumors associated with the development of a fibrous stroma (54) . The elucidation of downstream mediators of TGF-ß, like CTGF in pathways of granulation tissue formation and fibrosis may provide therapeutic targets for diminishing these fibrotic effects in a more selective manner. Our results indicate that agents that elevate intracellular cAMP have the potential to be the first generation of therapeutics effectively targeting one of these downstream mediators (i.e., CTGF). Our observations from cell culture systems and animal models indicate that basal in vitro collagen synthesis and that portion of induced in vivo collagen synthesis occurring independent of TGF-ß activation are both insensitive to down-regulation by cAMP elevating agents. These results are both surprising and unexpected. They suggest that distinct molecular mechanisms may regulate basal/TGF-ß-independent vs. TGF-ß/CTGF-induced collagen gene expression and subsequent collagen synthesis. Consequently, it may be possible in the future to selectively inhibit the collagen synthesis induced by TGF-ß at tissue sites of pathological fibrosis with therapeutic agents designed to block the synthesis or action of CTGF while sparing basal collagen synthesis and that induced independently of TGF-ß.


   ACKNOWLEDGMENTS
 
This research was supported by National Institutes of Health (NIH) grant GM37223, and a grant from FibroGen, Inc. to GRG. K.F. was supported by NIH training grant #RR07057–01A1 and an institutional fellowship from the University of Miami. We appreciate the assistance of the University of Miami Transgenic Facility with the establishment of the transgenic mice and Dr. Brian Masters for assistance with macrophotography.


   FOOTNOTES
 
1 Current address: University of Georgia, Veterinary Diagnostic and Investigational Lab, P.O. Box 1389, Tifton, GA 31793, USA.

3 Abbreviations: CTGF, connective tissue growth factor; CTX, cholera toxin; DMEM, Dulbecco's modified Eagle media; FBS, fetal bovine serum; Ig, immunoglobulin; NRK, normal rat kidney; NS, Nu-Serum I; s.c., subcutaneously; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TGF, transforming growth factor.

Received for publication February 18, 1999. Accepted for publication April 17, 1999.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Clark, R. A. (1993) Basics of cutaneous wound repair. J. Dermatol. Surg. Oncol. 19,693-706[Medline]
  2. DeLarco, J. E., Todaro, G. J. (1978) Growth factors from murine sarcoma virus-transformed cells. Proc. Natl. Acad. Sci. USA 75,4001-4005[Abstract/Free Full Text]
  3. Ignotz, R. A., Massague, J. (1986) Transforming growth factor-beta stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J. Biol. Chem. 261,4337-4345[Abstract/Free Full Text]
  4. Roberts, A. B., Sporn, M. B., Assoian, R. K., Smith, J. M., Roche, N. S., Wakefield, L. M., Heine, U. I., Liotta, L. A., Falanga, J. H., Fauci, A. S. (1986) Transforming growth factor type ß: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc. Natl. Acad. Sci. USA 83,4167-4171[Abstract/Free Full Text]
  5. Raghow, R., Postlethwaite, A. E., Keski-Oja, J., Moses, H. L., Kang, A. H. (1987) Transforming growth factor beta increases steady state levels of type 1 procollagen and fibronectin mRNAs posttranslationally in cultured fibroblasts. J. Clin. Invest. 79,1285-1288
  6. Roberts, A. B., Sporn, M. B. (1990) The transforming growth factor Betas. Sporn, M. B. Roberts, A. B. eds. Peptide Growth Factors and their Receptors—Handbook of Experimental Pathology ,419-472 Springer-Verlag Heidelberg, Germany.
  7. Penttinen, R. P., Kobayashi, S., Bornstein, P. (1988) Transforming growth factor ß increases mRNA for matrix proteins both in the presence and in the absence of changes in mRNA stability. Proc. Natl. Acad. Sci. USA 85,1105-1108[Abstract/Free Full Text]
  8. Frazier, K. S., Williams, S., Kothapalli, D., Klapper, H., Grotendorst, G. R. (1996) Stimulation of fibroblast cell growth, matrix production and granulation tissue formation by connective tissue growth factor. J. Invest. Dermatol. 107,404-411[Medline]
  9. Peltonen, J., Kahari, L., Jaakkola, S., Karai, V.-M., Varga, J., Uitto, J., Jimenez, S. A. (1990) Evaluation of TGF-ß and type I procollagen gene expression in fibrotic skin diseases by in situ hybridization. J. Invest. Dermatol. 94,365-371[Medline]
  10. Nagy, P., Schaff, Z., Lapis, K. (1991) Immunohistochemical detection of TGF-ß1 in fibrotic liver disease. Hepatology 14,269-273[Medline]
  11. Smith, E. A., LeRoy, E. C. (1990) A possible role for transforming growth factor beta in systemic sclerosis. J. Invest. Dermatol. 95,125s-127s[Medline]
  12. Border, W. A., Noble, N. A. (1994) Transforming growth factor ß in tissue fibrosis. N. Engl. J. Med. 331,1286-1292[Free Full Text]
  13. Leof, E. B., Proper, J. A., Gouston, A. S., Shipley, G. D., Dicorleto, P. E., Moses, H. L. (1986) Induction of c-sis mRNA and activity similar to PDGF by TGFß: a proposed model for indirect mitogenesis involving autocrine activity. Proc. Natl. Acad. Sci. USA 83,2453-2457[Abstract/Free Full Text]
  14. Grotendorst, G. R. (1997) Connective tissue growth factor: a mediator of TGF- ß action on fibroblasts. Cytokine Growth Factor Rev 8,171-179[Medline]
  15. Bradham, D., . M,Igarashi, A., Potter, R. L., Grotendorst, G. R. (1991) Connective tissue growth factor: a cysteine-rich mitogen secreted by human vascular endothelial cells is related to the SRC-induced immediate early gene product CEF-10. J. Cell Biol. 114,1285-1294[Abstract/Free Full Text]
  16. Igarashi, A., Okochi, H., Bradham, D. M., Grotendorst, G. R. (1993) Regulation of connective tissue growth factor gene expression in human skin fibroblasts and during wound repair. Mol. Biol. Cell 4,637-645[Abstract]
  17. Grotendorst, G. R., Okochi, H., Hayashi, N. (1996) A novel transforming growth factor ß response element controls the expression of the connective tissue growth factor gene. Cell Growth Differ 7,1-12
  18. Kothapalli, D., Frazier, K. S., Welply, A., Segarini, P. R., Grotendorst, G. R. (1997) Transforming growth factor-ß induces anchorage-independent growth of NRK fibroblasts via a connective tissue growth factor-dependent signaling pathway. Cell Growth Differ 8,61-68[Abstract]
  19. Kothapalli, D., Hayashi, N., Grotendorst, G. R. (1998) Inhibition of TGF-ß-stimulated CTGF gene expression and anchorage-independent growth by cAMP identifies a CTGF-dependent restriction point in the cell cycle. FASEB J 12,1151-1161[Abstract/Free Full Text]
  20. DeLarco, J. E., Todaro, G. J. (1978) Epithelioid and fibroblastic rat kidney cell clones: epidermal growth factor receptors and the effect of mouse sarcoma virus transformation. J. Cell. Physiol. 94,335-342[Medline]
  21. Webster, D. F., Harvey, W. (1979) A quantitative assay for collagen synthesis in microwell fibroblast cultures. Anal. Biochem. 96,220-224[Medline]
  22. Duncan, M. R., Berman, B. (1989) Differential regulation of collagen, glycosaminoglycan, fibronectin and collagenase activity production in cultured human adult dermal fibroblasts by interleukin 1-alpha and beta and tumor necrosis factor-alpha and beta. J. Invest. Dermatol. 92,699-706[Medline]
  23. Berman, B., Duncan, M. R. (1989) Pentoxifylline inhibits normal human fibroblast in vitro proliferation, collagen, glycosaminoglycan and fibronectin production and increases collagenase activity. J. Invest. Dermatol. 92,605-610[Medline]
  24. Berman, B., Wietzerbin, J., Sanceau, J., Merlin, G., Duncan, M. R. (1992) Pentoxifylline inhibits certain constitutive and tumor necrosis factor-{alpha}-induced activities of human normal fibroblasts. J. Invest. Dermatol. 98,706-712[Medline]
  25. Duncan, M. R., Hasan, A., Berman, B. (1995) Pentoxifylline, pentifylline, and interferons decrease type I and III procollagen mRNA levels in dermal fibroblasts: evidence for mediation by nuclear factor 1 downregulation. J. Invest. Dermatol. 104,282-286[Medline]
  26. Balkan, W., Colbert, M., Bock, C., Linney, E. (1992) Transgenic indicator mice for studying activated retinoic acid receptors during development. Proc. Natl. Acad. Sci. USA 89,3347-3351[Abstract/Free Full Text]
  27. Hanrahan, D. (1989) A protocol for isolating high-molecular weight DNA from mouse tails. Hogan, B. Constantini, F. Lacy, E. eds. Manipulating the Mouse Embryo: A Laboratory Manual ,174-176 Cold Spring Harbor Laboratory Press Cold Spring Harbor, N.Y..
  28. Gordon, J. W., Coffin, J. M., Davisson, M. T., Markert, C. L., Woychick, R. P. (1992) Standardized nomenclature for transgenic animals. ILAR News 34,45-52
  29. Wassarman, P. M., DePamphilis, M. L. (1992) Postimplantation stages and LacZ reporters. Methods Enzymol 225,459-469
  30. Schilling, J. A., Joel, W., Shurley, H. M. (1959) Wound healing: a comparative study of the histochemical changes in granulation tissue contained in stainless steel wire mesh and polyvinyl sponge cylinders. Surgery 46,702-710
  31. Hunt, T. K., Twomey, P., Zederplatt, B., Dunphy, J. E. (1967) Respiratory gas tension and pH in healing wounds. Am. J. Surg. 114,302-307[Medline]
  32. Sporn, M. B., Roberts, A. B., Shull, J. H., Smith, J. M., Ward, J. M., Sodek, J. (1983) Polypeptide transforming growth factors isolated from bovine sources and used for wound healing in vivo. Science 219,1329-1331[Abstract/Free Full Text]
  33. Grotendorst, G. R., Martin, G. R., Pencev, D., Sodek, J., Harvey, A. K. (1985) Stimulation of granulation tissue formation by platelet-derived growth factor in normal and diabetic rats. J. Clin. Invest. 76,2323-2329
  34. Shah, M., Forman, D. M., Ferguson, M. W. J. (1992) Control of scarring in adult wounds by neutralizing antibody to transforming growth factor beta. Lancet 339,213-214[Medline]