ARCHIVES
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 289, No. 1, August 15, pp. 6-11, 1991
Modulation of Collagen Synthesis by Growth Factors: The Role of Ascorbate-Stimulated Lipid Peroxidation1f2 Jeffrey C. Geesin,*,? Laura J. Hendricks,?
Joel S. Gordon,? and Richard A. Berg*p3
*Department of Biochemistry, University of Medicine & Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, New Jersey 08854; and TJohnson and Johnson Consumer Products, Inc., 199 Grandview Road, Skillman, New Jersey 08558
Received June 18, 1990, and in revised form April 22, 1991
Ascorbic acid has been shown to stimulate collagen synthesis through induction of lipid peroxidation leading to increased transcription of the collagen genes. The mechanism by which lipid peroxidation stimulates collagen transcription is unknown; however, an alteration of cell membranes may affect the activity of serum growth factors leading to a change in gene expression. To test this hypothesis, we treated dermal fibroblasts with transforming growth factor-beta (TGF-/3),4 epidermal growth factor (EGF), interleukin1 (IL-l), platelet-derived growth factor (PDGF), or fibroblast growth factor (FGF) in the presence of lipid peroxidation stimulating (200 PM) and nonstimulating (1 pM) concentrations of ascorbic acid. EGF and IL-1 had no effect on collagen synthesis at either concentration of ascorbic acid. FGF affected collagen synthesis only in the presence of 200 FM ascorbic acid, producing both a stimulation (0.4-2 rig/ml) and an inhibition (>50 rig/ml). PDGF and TGF-/l stimulated collagen synthesis in the presence of both concentrations of ascorbic acid, with TGF-fl producing an 1 l-fold increase in collagen synthesis in the presence of ascorbate. This synergism produced by the combination of ascorbic acid and TGF-/I was inhibitable by the lipid peroxidation inhibitor, propyl gallate. These results indicate that regulation of collagen synthesis by
i This work was supported in part by AM 31839. ’ An abstract of this work was presented at the Federation of American Societies for Experimental Biology, Atlanta, June, 1990. s To whom correspondence should be sent at Department of Biochemistry, University of Medicine & Dentistry of New Jersey, 675 Hoes Lane Piscataway, NJ 06654. ’ Abbreviations used: TGF-/3, transforming growth factor-beta; EGF, epidermal growth factor; IL-l, interleukin-1; PDGF, platelet-derived growth factor; FGF, fibroblast growth factor; DMEM, Dulbecco’s modified Eagle’s medium; HIDCS, heat-inactivated dialyzed calf serum; PBS, phosphate-buffered saline; NF-1, Nuclear Factor 1. 6
ascorbic acid does not occur through altering the response to EGF or 11-l. Ascorbate has no effect on PDGF but the effects of TGF-/3 and FGF on collagen synthesis o 1s~ Academic appear to be sensitive to lipid peroxidation. Press,
Inc.
Ascorbic acid has been shown to stimulate collagen synthesis in cultured human dermal fibroblasts through stimulation of the expression of collagen genes for the al(I), (~2(1), al(III), and al(W) chains without changing the rate of intracellular degradation (l-5). This process has recently been shown to involve the induction of lipid peroxidation by ascorbic acid (6,7). The process by which lipid peroxidation leads to a stimulation of collagen synthesis is unknown. Several possible mechanisms can be proposed. First, biologically active products of lipid peroxidation may be released from the cell membrane as a result of membrane oxidation. These released molecules may enter the interior of the cells and display specific activity, as has been proposed for malondialdehyde (7). This proposed mechanism requires ascorbic acid not only as an inducer of lipid peroxidation, but also as a reducing agent to stabilize the malondialdehyde modification of proteins or DNA (7). A second mechanism might involve the regulation of membrane bound molecules involved in signal transduction. These molecules could include enzymes such as protein kinases or membrane spanning proteins such as receptors leading to the induction of a signaling system in the absence of the normal route of activation. There are numerous examples in the literature for the role of lipid peroxidation in perturbing such signal transduction systems, including adenylate cyclase (8-ll), 5’-nucleotidase (11) Ca2+/Mgz+ ATPase (12), CAMP-dependent protein kinase (12), phospholipid hydrolysis (13), and protein kinase C (14), Ascorbic acid-induced lipid 0003-9861/91$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
GROWTH
FACTORS,
ASCORBATE,
peroxidation has previously been demonstrated to affect the binding of extracellular ligands, including phorbol esters (Xi), dopamine antagonists (16), and dihydroalpren0101 (17). A third mechanism may involve disruption of cell surface receptors for various extracellular matrix ligands which comprise the cell adhesion receptors and their ligands (18). The alteration of the membrane lipids or proteins may lead to reduced or enhanced binding between these receptors and their ligands. In the present studies, we chose to test the activities of various growth factors on human dermal fibroblasts in the presence of lipid peroxidation stimulating (200 PM) and nonstimulating (1 PM) concentrations of ascorbic acid. The growth factors TGF-& FGF, PDGF, EGF, and the cytokine IL-l have all been shown to affect cell proliferation, cell differentiation, or matrix synthesis, which are mediated through cell surface receptors (19-24). Therefore, we tested the ability of ascorbic acid to alter the response of dermal fibroblasts to these factors by a lipid peroxidation-dependent mechanism. EXPERIMENTAL
PROCEDURES
Materials. Ascorbic acid and propyl gallate were purchased from Sigma Chemical Co. (St. Louis, MO). All media containing ascorbic acid were made fresh daily due to the instability of ascorbate in solution. All growth factors were obtained from Collaborative Research (Bedford, MA). TGF-fl and PDGF were isolated from human platelets, IL-l was from normal human peripheral blood leukocytes, FGF was from bovine pituitary, and EGF was from male mouse submaxillary glands. Cell culture. Primary cultures of newborn human dermal fibroblasts (American Type Culture Collection) were grown and treated as previously described for the determination of collagen synthesis (4). Briefly, cells were seeded at 100,000 per 35-mm diameter plastic tissue culture dish and grown to confluence in Dulbecco’s modified Eagle’s medium (DMEM) buffered with 24 mM sodium bicarbonate and 25 mM Hepes and supplemented with 20% heat-inactivated dialyzed calf serum (HIDCS) (Grand Island Biological). Plates were then washed three times
15--
oQ 0
”
”A
fi
0.016
0.06
0.4
[IL-l]
in U/ml
”
4)
2
10
AND
COLLAGEN
SYNTHESIS
g
7
0 30--
8 !J
0 1 @A ASCORBATE I 15.- 0 200 /A4 ASCORBATE
0 OQ 0
” .a
T I 4
t
A 20
v 100
1’ 500
[EGF] in rig/ml
FIG. 2.
The effect of EGF on collagen synthesis. Collagen synthesis is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
with phosphate-buffered saline (PBS) and treated for 3 days in DMEM containing 0.5% HIDCS with the appropriate concentration of ascorbic acid and the growth factor to be tested. Cells were incubated for the final 6 h with the test media containing 20 &i/ml 2,3-[sH]-l-proline (Amersham). Measurement of coL!ngen synthesis. By a previously described protocol (4), media from treated plates were combined with a solution of protease inhibitors and stored at -20°C until analysis. The cells were counted on a Coulter counter after trypsinization. The media were then dialyzed extensively versus 0.33 M calcium acetate, 25 mM Tris, pH 7.4, 0.02% sodium azide with 0.6 mM N-ethylmaleimide. Measured aliquots of the samples were combined with bovine serum albumin and digested with highly purified collagenase (Collagenase form III, Advanced Biofactures, Inc.) by an incubation at 37°C for 4 h. The samples were then precipitated with 25% trichloroacetic acid, 1.25% tannic acid and centrifuged at 12,000g for 3.5 min. The collagenase solubilized counts were determined by diluting the supernatant with Liquiscint (National Diagnostics) and quantifying on an Intertechnique SL 4000 scintillation spectrometer. The radioactivity of an undigested sample was similarly quantified. The collagenase soluble counts were used to determine the amount of newly synthesized collagen protein and are reported as collagenase soluble counts per cell. The difference between the collagenase-digested and collagenase-undigested samples was used to determine the counts in noncollagen protein. For those treatments which produced a change in collagen synthesis, the effect on the percentage collagen synthesis is also reported as percentage collagen synthesis which was computed using a correction for the high incorporation of proline into collagen as compared to other proteins, as previously described (25). In experiments where no effect was seen on collagen synthesis per cell, no effect was seen in percentage collagen synthesis. All data shown are the results of duplicate assays on each of three plates per condition with the standard deviation of these six determinations represented by error bars.
RESULTS
FIG. 1. The effect of IL-1 on collagen synthesis. Collagen synthesis is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
Treatment of cultured dermal fibroblasts with IL-l or EGF produced no effect on collagen synthesis (Figs. 1 and 2) in the presence of an ascorbic acid concentration reported to stimulate lipid peroxidation (6, 7) or in the presence of a low concentration of ascorbate which does not produce lipid peroxidation (6). The highest concen-
GEESIN
ET AL.
of TGF-P was selective for collagen synthesis. The baseline level of percentage collagen synthesized was much higher in 200 PM ascorbate than 1 PM ascorbate as would be expected from previous studies owing to the effect of ascorbate itself on collagen synthesis (1). Treatment of the dermal fibroblasts with PDGF produced a dose-dependent increase in collagen synthesis at both levels of ascorbic acid that reached twofold at the highest concentration of PDGF tested (Fig. 4A). The increase seen at 1 PM ascorbic acid was small, but statistically significant (8.63 + 0.38 cpm/lOOO cells for 1.43 U/ ml PDGF versus 3.86 + 0.38 cpm/lOOO cells in the absence of PDGF). The increase in response to PDGF was dose dependent for the entire concentration curve tested (0.004-1.43 U/ml). The effect on percentage collagen synthesis produced an increase from 58 to 72% collagen synthesis with increasing PDGF concentration in 200 PM ascorbic acid; however, no statistical effect was seen in 1 PM ascorbic acid (Fig. 4B). These results indicate that the effect of PDGF on collagen synthesis was partially loo? A 80..
-. 0
.OiU
.ci2
.i
.‘5
0
2:s
f,TOF-b] in nM FIG. 3. The effect of TGF-fl on collagen synthesis. Collagen synthesis (A) is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Percentage collagen synthesis (B) was determined by accounting for the increased incorporation of proline into collagen compared to noncollagen proteins as previously described (25). Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
20.. Y
OY
n
”
4)
”
1
100
tration of EGF used (500 rig/ml) produced a slight inhibition of collagen synthesis in the presence of 200 I.LM ascorbate, but since this dose is much higher than required to saturate the receptors (26), the observation may not relate to the physiological role of EGF (Fig. 2). TGF-P produced a dose-dependent increase in collagen synthesis per cell in both concentrations of ascorbic acid tested (Fig. 3A). The greatest change was seen between 0.2 and 2.5 nM TGF-/3 and the effect did not saturate at the highest level tested. The increase in collagen synthesis was most dramatic with high ascorbate where the increase with TGF-P was 11-fold. At 1 PM ascorbate, TGF-P increased collagen synthesis fourfold (Fig. 3A). In order to determine if the effect of TGF-P was specific for collagen, the data were also calculated for the percentage of the total protein synthesis that was devoted to collagen (25). The effect on the percentage collagen is demonstrated in Fig. 3B. Both concentrations of ascorbic acid produced increases in the percentage collagen synthesis in response to increasing amounts of TGF-B, indicating that the effect
B
I
A 1
* Y
”
.DD4
.02
.l
A
3
-iP I 1.43
[PDCF] in U/ml FIG. 4. The effect of PDGF on collagen synthesis. Collagen synthesis (A) is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Percentage collagen synthesis (B) was determined by accounting for the increased incorporation of proline into collagen compared to noncollagen proteins as previously described (5). Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
GROWTH
FACTORS.
ASCORBATE,
AND
COLLAGEN
9
SYNTHESIS
oxidation inhibitor, propyl gallate, (16, 17, 27, 28) to inhibit the synergism observed with the combination of ascorbic acid and TGF-B was tested. Propyl gallate has previously been shown to inhibit ascorbate-stimulated collagen synthesis and lipid peroxidation in these cells (6). In the present studies (Fig. 6), propyl gallate inhibited ascorbate-stimulated collagen synthesis as previously reported (6) and also inhibited the synergistic effect of ascorbic acid and TGF-/3 at a propyl gallate concentration of 10 PM and above. DISCUSSION
.”
I
le
04 0
A
2
10
50
t 143
[FGFj in &ml FIG. 6. The effect of FGF on collagen synthesis. Collagen synthesis (A) is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Percentage collagen synthesis (B) was determined by accounting for the increased incorporation of proline into collagen compared to noncollagen proteins as previously described (25). Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
accounted for by its effect on protein synthesis and partially accounted for by its effect on collagen synthesis. FGF has a striking effect on collagen synthesis in fibroblasts. In the cultures treated with 1 PM ascorbic acid, no statistically significant change was seen in either collagen synthesis/cell or percentage collagen synthesis (Figs. 5A and 5B). However, in the presence of 200 PM ascorbic acid, collagen synthesis and percentage collagen synthesis were increased and then decreased depending upon the concentration of FGF used (Figs. 5A and 5B). At 0.4 rig/ml FGF, collagen synthesis/cell was twice that seen in the absence of FGF (Fig. 5A). There was only a slight increase in the percentage collagen synthesized at these levels of FGF (Fig. 5B). At concentrations of FGF above 10 rig/ml, both collagen synthesis/cell and percentage collagen synthesis were significantly less than that seen in the absence of FGF. In order to demonstrate that lipid peroxidation was responsible for the effect of ascorbate on the induction of collagen synthesis by TGF-8, the ability of the lipid per-
In the present studies, we tested the ability of ascorbateinduced lipid peroxidation to interfere with receptor-mediated effects of various growth factors. The effects of these growth factors in low (1 PM) and high (200 PM) concentrations of ascorbic acid were compared. These concentrations were selected so the effects of lipid peroxidation could be differentiated. In previous studies, 200 PM ascorbate was shown to stimulate both lipid peroxidation and collagen synthesis, whereas at 1 PM ascorbic acid, there was shown to be little or no induction of lipid peroxidation (6). We chose to use 1 PM ascorbate because, at this concentration, the activity of the enzyme prolyl hydroxylase is not limited by the availability of ascorbate, a required cofactor, and all collagen molecules are hydroxylated (3) permitting the formation of the characteristic triple helical conformation of collagen which is necessary for secretion (29). Therefore, under the conditions of these experiments, the effect of ascorbic acid induced lipid peroxidation is not compromised by the rate of hydroxylation of proline residues in collagen. Both IL-l (30-33) and EGF (34) have previously been shown to have effects on collagen synthesis by fibroblasts
MS;""" +PRO3 LUTE
i (@id) +
i
i
i
i
(100)
+
(Ott,
(:,
i
i
(To, $0)
FIG. 6. The effect of propyl gallate on the synergism produced with ascorbic acid and TGF-P on collagen synthesis. Collagen synthesis is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Those samples receiving ascorbate as indicated received 200 FM, those with TGF-/3 received 800 pM, and those with propyl gallate received the concentration indicated in parentheses.
10
GEESIN
in culture. Increases in collagen synthesis by IL-l have been reported to be caused by culture conditions including the presence of serum (35) or various added or endogenously generated agents such as PGE2 (32,33) or PDGFAA (21, 36). EGF is believed to induce fibroblast proliferation in healing wounds (34,37) rather than stimulating collagen synthesis directly. In the studies presented here, no effect of IL-l was seen in the presence of 0.5% serum (Fig. l), while only the highest concentration of EGF used produced an inhibition of collagen synthesis in the presence of either concentration of ascorbate (Fig. 2). TGF-P has been consistently shown to stimulate collagen synthesis in fibroblasts in a large number of reports both in viva and in vitro (38-44). Although the detailed mechanism for this effect of TGF-p on collagen synthesis has not been elucidated, there is evidence to support the suggestion that it may be mediated directly through the binding of the DNA binding protein, Nuclear Factor 1 (NF-1) to the (~2 type I collagen gene promoter (45) or indirectly in combination with other factors such as PGE2 (42). In our experiments, an enhancement of collagen synthesis was seen with TGF-@ treatment in both concentrations of ascorbic acid. This induction is much greater with the addition of both TGF-/3 and high ascorbate than with either stimulant alone (Fig. 3A). Our results are consistent with the previous observation that TGF-P and ascorbate act synergistically (43), a phenomenon which we now show to be dependent upon ascorbateinduced lipid peroxidation as shown by the ability of the lipid peroxidation inhibitor, propyl gallate (27,28) to block the effect. Since enhanced collagen synthesis by TGF-P was seen in both concentrations of ascorbic acid, the observation supports the presence of two different pathways for the induction of collagen synthesis in dermal fibroblasts. One pathway is responsive to TGF-/3 and may involve the transcription factor NF-1, while ascorbic acid acts on a different pathway mediated by lipid peroxidation. An alternative explanation is that ascorbate-mediated lipid peroxidation affects a step in the TGF-P signal transduction pathway. PDGF has been reported to stimulate connective tissue synthesis in wounds through the induction of chemoattraction of macrophages and fibroblasts (34,46). In these wounds, high levels of TGF-P were found, apparently synthesized by the PDGF-induced macrophages or fibroblasts, thus producing a TGF-P-mediated effect on connective tissue synthesis (46). In our experiments, PDGF was shown to be able to produce a moderate stimulation of collagen synthesis in the presence of low levels of serum. The effect of PDGF differed from that of TGF-/3 in that it was not selectively affected by ascorbate. FGF, like TGF-/3, is selectively affected by ascorbate at concentrations that produced lipid peroxidation. It differs from TGF-/3 in that it stimulates collagen synthesis at low FGF concentrations only and does not stimulate collagen synthesis in the presence of ascorbate concen-
ET AL.
trations insufficient to stimulate lipid peroxidation. Therefore, FGF stimulation of collagen synthesis appears to occur only in the presence of ascorbate-induced lipid peroxidation and represents a cell membrane associated mechanism capable of being manipulated by ascorbateinduced lipid peroxidation. FGF has been implicated in wound healing through its ability to cause rapid neovascularization and fibroplasia in skin (47) which are assumed to be receptor-mediated (48,49). The recently described interactions between FGF and the extracellular matrix have implied that this growth factor plays an integral role in the wound healing response (50, 51). Our findings indicating that FGF can regulate collagen synthesis in a lipid peroxidation environment has interesting implications for its action in tissue repair. In instances where lipid peroxidation can occur, such as after uv irradiation (52) and oxidant-mediated inflammatory responses (53-56) in wound healing, small amounts of FGF can enhance repair through stimulation of collagen synthesis. FGF itself has been shown to be sensitive to oxidation but is completely prevented from oxidative inactivation when bound to heparin sulfate proteoglycan on the cell surface (57). In these studies we have shown that ascorbate-induced lipid peroxidation can affect the activity of growth factors in dermal fibroblasts. In particular, our results raise the possibility that some step in the pathways by which TGFp and FGF regulate collagen synthesis is sensitive to lipid peroxidation. Whether this mechanism is involved in the increase in collagen transcription seen with ascorbate is not known, but these results support that possibility. The observations here do suggest that any studies performed in the presence of high concentrations of ascorbic acid should be viewed in light of these possible effects of ascorbate-induced lipid peroxidation. REFERENCES 1. Murad, S., Grove, D., Lindberg, K. A., Reynolds, G., Sivarajah, A., and Pinnell, S. R. (1981) Proc. Natl. Acad. Sci. USA 78,2879-2882. 2. Geesin, J., Murad, S., and Pinnell, S. R. (1986) Biochim. Biophys. Acta 886, 272-274. 3. Pinnell, S. R., Murad, S., and Darr, D. (1987) Arch. Dermatol. 123, 1684-1686. 4. Geesin, J. C., Darr, D., Kaufman, R., Murad, S., and Pinnell, S. R. (1988) J. Invest. Dermatol. 90, 420-424. 5. Tajima, S., Phillips, C., Murad, S., Kaufman, R., and Pinnell, S. (1988) J. Invest. Dermatol. 90, 611A. 6. Geesin, J. C., Gordon, J. S., and Berg, R. A. (1990) Arch. Biochem. Biophys. 278, 350-355. 7. Chojkier, M., Houglum, K., Solis-Herruzo, J., and Brenner, D. A. (1989) J. Biol. &em. 264, 16,957-16,962. 8. Baba, A., Lee, E., Ohta, A., Tatsuno, T., and Iwata, H. (1981) J. Biol. Chem. 256, 3679-3684. 9. Lee, E., Baba, A., Ohta, A., and Iwata, H. (1982) Biochim. Biophys. Acta 689,370-374. 10. Sobolev, A. S., Rosenkranz, A. A., and Kasarov, A. R. (1983) J. Radiut. Biol. 44, 31-39.
GROWTH
FACTORS,
ASCORBATE,
AND
COLLAGEN
SYNTHESIS
11
11. Paradisi, L., Panagini, C., Paroloa, M., Barrera, G., and Dianzani, M. U. (1985) Chem. Biol. Interact. 53, 209-217.
34. Buckley, A., Davidson, J. M., Kameroth, C., Wolt, T. B., and Woodward, S. C. (1985) Proc. N&l. Acad. Sci. USA 82, 8340-8344.
12. Lyzlova, L. V., Persianova, V. R., Antipenko, S. N. (1987) Biochem. Int. 14,1079-1086.
35. Elias, J. A., Freundlich, B., Adams, S., and Rosenbloom, J. (1990) Ann. N. Y. Acud. Sci. 580, 233-244. 36. Raines, L. G., Dower, S. K., and Ross, R. (1989) Science 243,393396. 37. Canalis, E., and Raisz, L. G. (1979) Endocrinology 104, 862-869. 38. Ignotz, R. A., Endo, T., and Massague, J. (1987) J. Biol. Chem. 262,
A. E., and Lyzlova,
13. Shasby, D. M., Yorek, M., and Shasby, S. S. (1988) Blood 72,491499. 14. O’Brian, C. A., Ward, N. E., Weinstein, I. B., Bull, A. W., and Marnett, L. J. (1988) Biochem. Biophys. Res. Commun. 155,1374-1380. 15. Delclos, K. B., and Blumberg, P. M. (1982) Cancer Res. 42, 12271232. 16. Heikkila, R. E., Cabbat, F. S., and Manzino, L. (1982) J. Neurochem.
38,1000-1006. 17. Heikkila, R. E. (1983) Eur. J. Pharmucol. 93, 79-85. 18. Hynes, R. 0. (1987) Cell 48, 549-554. 19. Ullrich, A., Coussens, L., Hayflick, J. S., Dull, T. J., Gray, A., Tam, A. W., Lee, J., Yarden, Y., Libermann, T. A., Schlessinger, J., Downward, J., Mayes, E. L. V., Whittle, N., Waterfield, M. D., and Seeburg, P. H. (1984) Nature 309, 418-425. 20. Rifkin, D. B., and Moscatelli, D. (1989) J. Cell Biol. 109, 1-6. 21. Singh, J. P., Adams, L. D., and Bonin, P. D. (1988) J. Cell Biol.
106,813-819. 22. Yarden, Y., Escobedo, J. A., Kuang, W.-J., Yang-Feng, T. L., Daniel, T. O., Tremble, P., Chen,E. Y., Ando, M. E., Harkins, R.N., Francke, U., Friend, V. A., Ullrich, A., and Williams, L. T. (1986) Nature
323,226-232. 23. Sporn, M. B., Roberts, A. B., Wakefield, L. M., and de Crombrugghe, B. (1987) J. Cell Biol. 105,1039-1045. 24. Cheifetz, S., and Massague, J. (1989) J. Biol. Chem. 264, 12,02512,028. 25. Breul, S. D., Bradley, K. H., Hance, A. J., Schafer, M. P., Berg, R. A., and Crystal, R. G. (1980) J. Biol. Chem. 255, 5250-5260. 26. Paulsson, Y., Bywater, M., Heldin, C.-H., and Westermark, B. (1987) Exp. CellRes. 171, 186-194. 27. Heikkila, R. E., and Manzino, L. (1987) Ann. N. Y. Acud. Sci. 498,
63-76. 28. Heikkila,
R. E., and Cabbat, F. S. (1983) J. Neurochem. 41,13841392. 29. Berg, R. A., Steinmann, B., Rennard, S. I., and Crystal, R. G. (1983) Arch. Biochem. Biophys. 226, 681-686. 30. Bhatnagar, R., Penfornis, H., Mauviel, A., Loyau, G., Saklatvala, J., and Pujol, J.-P. (1986) Biochem. Znt. 13, 709-720. 31. Kahari, V.-M., Heino, J., and Vuorio, E. (1987) Biochim. Biophys. Actu 929, 142-147.
32. Goldring, M. B., and Krane, S. M. (1987) J. Biol. Chem. 262,16,72416,729. 33. Postlethwaite, A. E., Raghow, R., Stricklin, G. P., Poppleton, Seyer, J. M., and Kang, A. H. (1988) J. Cell Biol. 106,311-318.
H.,
6443-6446. 39. Fine, A., and Goldstein, R. H. (1987) J. Biol. Chem. 262, 38973902. 40. Penttinen, R. P., Kobayashi, S., and Bornstein, P. (1988) Proc. Nutl. Acud. Sci. USA 85, 1105-1108. 41. Keski-Oja, J., Raghow, R., Sawdey, M., Loskutoff, D. J., Postlethwaite, A. E., Kang, A. H., and Moses, H. L. (1988) J. Biol. Chem. 263, 3111-3115. 42. Diaz, A., Varga, J., and Jimenez, S. A. (1989) J. Biol. Chem. 264, 11,554-11,557. 43. Appling, W. D., O’Brien, W. R., Johnston, D. A., and Duvic, M. (1989) FEBS L&t. 250, 541-544. 44. Roberts, A. B., Heine, U. I., Flanders, K. C., and Sporn, M. B. (1990) Ann. N. Y. Acud. Sci. 580, 225-232. 45. Rossi, P., Karsenth, G., Roberts, A. B., Roche, N. S., Sporn, M. B., and de Crombrugghe, B. (1988) Cell 52, 405-414. 46. Pierce, G. F., Mustoe, T. A., Lingelbach, J., Masakowski, V. P., Griffin, G. L., Senior, R. M., and Deuel, T. F. (1989) J. Cell Biol.
109,429-440. 47. Folkman, J., and Klagsbrun, M. (1987) Science 235, 442-447. 48. Olwin, B. B., and Hauschka, S. D. (1986) Biochemistry 25, 34873492. 49. Neufeld, G., and Gospodarowicz, D. (1986) J. Biol. Chem. 261,5631-
5637. 50. Bashkin, P., Doctrow,
S., Klagsbrun, M., Svahn, C. M., Folkman, J., and Vlodavsky, I. (1989) Biochemistry 28, 1737-1743. 51. Saksela, O., and Rifkin, D. B. (1990) J. CelZBiol. 110, 767-775. 52. Danno, K., Horio, T., Takigawa, M., and Imamura, S. (1984) J. Znuest. Dermutol. 83, 166-168. 53. Fantone, J. C., and Ward, P. A. (1982) Am. J. Puthol. 107, 397-
418. 54. Cathcart, M. K., Morel, D. W., and Chisolm, G. M. (1985) J. Leukocyte Biol. 38, 341-350. 55. Cook, R. M., Ashworth, R. F., and Musgrove, N. R. J. (1987) Znt. Archs. Allergy Appl. Immun. 83,428-431. 56. Yui, S., and Yamazaki, M. (1990) J. Immunol. 144, 1466-1471. 57. Ortega, S., Schaeffer, M. T., Soderman, D., Di Salvo, J., Linemeyer, D. L., Gimmenez-Gallego, G., and Thomas, K. A. (1991) J. Biol. Chem. 266, 5842-5846.
OF BIOCHEMISTRY
AND
BIOPHYSICS
Vol. 289, No. 1, August 15, pp. 6-11, 1991
Modulation of Collagen Synthesis by Growth Factors: The Role of Ascorbate-Stimulated Lipid Peroxidation1f2 Jeffrey C. Geesin,*,? Laura J. Hendricks,?
Joel S. Gordon,? and Richard A. Berg*p3
*Department of Biochemistry, University of Medicine & Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, New Jersey 08854; and TJohnson and Johnson Consumer Products, Inc., 199 Grandview Road, Skillman, New Jersey 08558
Received June 18, 1990, and in revised form April 22, 1991
Ascorbic acid has been shown to stimulate collagen synthesis through induction of lipid peroxidation leading to increased transcription of the collagen genes. The mechanism by which lipid peroxidation stimulates collagen transcription is unknown; however, an alteration of cell membranes may affect the activity of serum growth factors leading to a change in gene expression. To test this hypothesis, we treated dermal fibroblasts with transforming growth factor-beta (TGF-/3),4 epidermal growth factor (EGF), interleukin1 (IL-l), platelet-derived growth factor (PDGF), or fibroblast growth factor (FGF) in the presence of lipid peroxidation stimulating (200 PM) and nonstimulating (1 pM) concentrations of ascorbic acid. EGF and IL-1 had no effect on collagen synthesis at either concentration of ascorbic acid. FGF affected collagen synthesis only in the presence of 200 FM ascorbic acid, producing both a stimulation (0.4-2 rig/ml) and an inhibition (>50 rig/ml). PDGF and TGF-/l stimulated collagen synthesis in the presence of both concentrations of ascorbic acid, with TGF-fl producing an 1 l-fold increase in collagen synthesis in the presence of ascorbate. This synergism produced by the combination of ascorbic acid and TGF-/I was inhibitable by the lipid peroxidation inhibitor, propyl gallate. These results indicate that regulation of collagen synthesis by
i This work was supported in part by AM 31839. ’ An abstract of this work was presented at the Federation of American Societies for Experimental Biology, Atlanta, June, 1990. s To whom correspondence should be sent at Department of Biochemistry, University of Medicine & Dentistry of New Jersey, 675 Hoes Lane Piscataway, NJ 06654. ’ Abbreviations used: TGF-/3, transforming growth factor-beta; EGF, epidermal growth factor; IL-l, interleukin-1; PDGF, platelet-derived growth factor; FGF, fibroblast growth factor; DMEM, Dulbecco’s modified Eagle’s medium; HIDCS, heat-inactivated dialyzed calf serum; PBS, phosphate-buffered saline; NF-1, Nuclear Factor 1. 6
ascorbic acid does not occur through altering the response to EGF or 11-l. Ascorbate has no effect on PDGF but the effects of TGF-/3 and FGF on collagen synthesis o 1s~ Academic appear to be sensitive to lipid peroxidation. Press,
Inc.
Ascorbic acid has been shown to stimulate collagen synthesis in cultured human dermal fibroblasts through stimulation of the expression of collagen genes for the al(I), (~2(1), al(III), and al(W) chains without changing the rate of intracellular degradation (l-5). This process has recently been shown to involve the induction of lipid peroxidation by ascorbic acid (6,7). The process by which lipid peroxidation leads to a stimulation of collagen synthesis is unknown. Several possible mechanisms can be proposed. First, biologically active products of lipid peroxidation may be released from the cell membrane as a result of membrane oxidation. These released molecules may enter the interior of the cells and display specific activity, as has been proposed for malondialdehyde (7). This proposed mechanism requires ascorbic acid not only as an inducer of lipid peroxidation, but also as a reducing agent to stabilize the malondialdehyde modification of proteins or DNA (7). A second mechanism might involve the regulation of membrane bound molecules involved in signal transduction. These molecules could include enzymes such as protein kinases or membrane spanning proteins such as receptors leading to the induction of a signaling system in the absence of the normal route of activation. There are numerous examples in the literature for the role of lipid peroxidation in perturbing such signal transduction systems, including adenylate cyclase (8-ll), 5’-nucleotidase (11) Ca2+/Mgz+ ATPase (12), CAMP-dependent protein kinase (12), phospholipid hydrolysis (13), and protein kinase C (14), Ascorbic acid-induced lipid 0003-9861/91$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
GROWTH
FACTORS,
ASCORBATE,
peroxidation has previously been demonstrated to affect the binding of extracellular ligands, including phorbol esters (Xi), dopamine antagonists (16), and dihydroalpren0101 (17). A third mechanism may involve disruption of cell surface receptors for various extracellular matrix ligands which comprise the cell adhesion receptors and their ligands (18). The alteration of the membrane lipids or proteins may lead to reduced or enhanced binding between these receptors and their ligands. In the present studies, we chose to test the activities of various growth factors on human dermal fibroblasts in the presence of lipid peroxidation stimulating (200 PM) and nonstimulating (1 PM) concentrations of ascorbic acid. The growth factors TGF-& FGF, PDGF, EGF, and the cytokine IL-l have all been shown to affect cell proliferation, cell differentiation, or matrix synthesis, which are mediated through cell surface receptors (19-24). Therefore, we tested the ability of ascorbic acid to alter the response of dermal fibroblasts to these factors by a lipid peroxidation-dependent mechanism. EXPERIMENTAL
PROCEDURES
Materials. Ascorbic acid and propyl gallate were purchased from Sigma Chemical Co. (St. Louis, MO). All media containing ascorbic acid were made fresh daily due to the instability of ascorbate in solution. All growth factors were obtained from Collaborative Research (Bedford, MA). TGF-fl and PDGF were isolated from human platelets, IL-l was from normal human peripheral blood leukocytes, FGF was from bovine pituitary, and EGF was from male mouse submaxillary glands. Cell culture. Primary cultures of newborn human dermal fibroblasts (American Type Culture Collection) were grown and treated as previously described for the determination of collagen synthesis (4). Briefly, cells were seeded at 100,000 per 35-mm diameter plastic tissue culture dish and grown to confluence in Dulbecco’s modified Eagle’s medium (DMEM) buffered with 24 mM sodium bicarbonate and 25 mM Hepes and supplemented with 20% heat-inactivated dialyzed calf serum (HIDCS) (Grand Island Biological). Plates were then washed three times
15--
oQ 0
”
”A
fi
0.016
0.06
0.4
[IL-l]
in U/ml
”
4)
2
10
AND
COLLAGEN
SYNTHESIS
g
7
0 30--
8 !J
0 1 @A ASCORBATE I 15.- 0 200 /A4 ASCORBATE
0 OQ 0
” .a
T I 4
t
A 20
v 100
1’ 500
[EGF] in rig/ml
FIG. 2.
The effect of EGF on collagen synthesis. Collagen synthesis is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
with phosphate-buffered saline (PBS) and treated for 3 days in DMEM containing 0.5% HIDCS with the appropriate concentration of ascorbic acid and the growth factor to be tested. Cells were incubated for the final 6 h with the test media containing 20 &i/ml 2,3-[sH]-l-proline (Amersham). Measurement of coL!ngen synthesis. By a previously described protocol (4), media from treated plates were combined with a solution of protease inhibitors and stored at -20°C until analysis. The cells were counted on a Coulter counter after trypsinization. The media were then dialyzed extensively versus 0.33 M calcium acetate, 25 mM Tris, pH 7.4, 0.02% sodium azide with 0.6 mM N-ethylmaleimide. Measured aliquots of the samples were combined with bovine serum albumin and digested with highly purified collagenase (Collagenase form III, Advanced Biofactures, Inc.) by an incubation at 37°C for 4 h. The samples were then precipitated with 25% trichloroacetic acid, 1.25% tannic acid and centrifuged at 12,000g for 3.5 min. The collagenase solubilized counts were determined by diluting the supernatant with Liquiscint (National Diagnostics) and quantifying on an Intertechnique SL 4000 scintillation spectrometer. The radioactivity of an undigested sample was similarly quantified. The collagenase soluble counts were used to determine the amount of newly synthesized collagen protein and are reported as collagenase soluble counts per cell. The difference between the collagenase-digested and collagenase-undigested samples was used to determine the counts in noncollagen protein. For those treatments which produced a change in collagen synthesis, the effect on the percentage collagen synthesis is also reported as percentage collagen synthesis which was computed using a correction for the high incorporation of proline into collagen as compared to other proteins, as previously described (25). In experiments where no effect was seen on collagen synthesis per cell, no effect was seen in percentage collagen synthesis. All data shown are the results of duplicate assays on each of three plates per condition with the standard deviation of these six determinations represented by error bars.
RESULTS
FIG. 1. The effect of IL-1 on collagen synthesis. Collagen synthesis is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
Treatment of cultured dermal fibroblasts with IL-l or EGF produced no effect on collagen synthesis (Figs. 1 and 2) in the presence of an ascorbic acid concentration reported to stimulate lipid peroxidation (6, 7) or in the presence of a low concentration of ascorbate which does not produce lipid peroxidation (6). The highest concen-
GEESIN
ET AL.
of TGF-P was selective for collagen synthesis. The baseline level of percentage collagen synthesized was much higher in 200 PM ascorbate than 1 PM ascorbate as would be expected from previous studies owing to the effect of ascorbate itself on collagen synthesis (1). Treatment of the dermal fibroblasts with PDGF produced a dose-dependent increase in collagen synthesis at both levels of ascorbic acid that reached twofold at the highest concentration of PDGF tested (Fig. 4A). The increase seen at 1 PM ascorbic acid was small, but statistically significant (8.63 + 0.38 cpm/lOOO cells for 1.43 U/ ml PDGF versus 3.86 + 0.38 cpm/lOOO cells in the absence of PDGF). The increase in response to PDGF was dose dependent for the entire concentration curve tested (0.004-1.43 U/ml). The effect on percentage collagen synthesis produced an increase from 58 to 72% collagen synthesis with increasing PDGF concentration in 200 PM ascorbic acid; however, no statistical effect was seen in 1 PM ascorbic acid (Fig. 4B). These results indicate that the effect of PDGF on collagen synthesis was partially loo? A 80..
-. 0
.OiU
.ci2
.i
.‘5
0
2:s
f,TOF-b] in nM FIG. 3. The effect of TGF-fl on collagen synthesis. Collagen synthesis (A) is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Percentage collagen synthesis (B) was determined by accounting for the increased incorporation of proline into collagen compared to noncollagen proteins as previously described (25). Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
20.. Y
OY
n
”
4)
”
1
100
tration of EGF used (500 rig/ml) produced a slight inhibition of collagen synthesis in the presence of 200 I.LM ascorbate, but since this dose is much higher than required to saturate the receptors (26), the observation may not relate to the physiological role of EGF (Fig. 2). TGF-P produced a dose-dependent increase in collagen synthesis per cell in both concentrations of ascorbic acid tested (Fig. 3A). The greatest change was seen between 0.2 and 2.5 nM TGF-/3 and the effect did not saturate at the highest level tested. The increase in collagen synthesis was most dramatic with high ascorbate where the increase with TGF-P was 11-fold. At 1 PM ascorbate, TGF-P increased collagen synthesis fourfold (Fig. 3A). In order to determine if the effect of TGF-P was specific for collagen, the data were also calculated for the percentage of the total protein synthesis that was devoted to collagen (25). The effect on the percentage collagen is demonstrated in Fig. 3B. Both concentrations of ascorbic acid produced increases in the percentage collagen synthesis in response to increasing amounts of TGF-B, indicating that the effect
B
I
A 1
* Y
”
.DD4
.02
.l
A
3
-iP I 1.43
[PDCF] in U/ml FIG. 4. The effect of PDGF on collagen synthesis. Collagen synthesis (A) is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Percentage collagen synthesis (B) was determined by accounting for the increased incorporation of proline into collagen compared to noncollagen proteins as previously described (5). Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
GROWTH
FACTORS.
ASCORBATE,
AND
COLLAGEN
9
SYNTHESIS
oxidation inhibitor, propyl gallate, (16, 17, 27, 28) to inhibit the synergism observed with the combination of ascorbic acid and TGF-B was tested. Propyl gallate has previously been shown to inhibit ascorbate-stimulated collagen synthesis and lipid peroxidation in these cells (6). In the present studies (Fig. 6), propyl gallate inhibited ascorbate-stimulated collagen synthesis as previously reported (6) and also inhibited the synergistic effect of ascorbic acid and TGF-/3 at a propyl gallate concentration of 10 PM and above. DISCUSSION
.”
I
le
04 0
A
2
10
50
t 143
[FGFj in &ml FIG. 6. The effect of FGF on collagen synthesis. Collagen synthesis (A) is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Percentage collagen synthesis (B) was determined by accounting for the increased incorporation of proline into collagen compared to noncollagen proteins as previously described (25). Points which appear to lack error bars have standard deviations which do not extend beyond the size of the data point.
accounted for by its effect on protein synthesis and partially accounted for by its effect on collagen synthesis. FGF has a striking effect on collagen synthesis in fibroblasts. In the cultures treated with 1 PM ascorbic acid, no statistically significant change was seen in either collagen synthesis/cell or percentage collagen synthesis (Figs. 5A and 5B). However, in the presence of 200 PM ascorbic acid, collagen synthesis and percentage collagen synthesis were increased and then decreased depending upon the concentration of FGF used (Figs. 5A and 5B). At 0.4 rig/ml FGF, collagen synthesis/cell was twice that seen in the absence of FGF (Fig. 5A). There was only a slight increase in the percentage collagen synthesized at these levels of FGF (Fig. 5B). At concentrations of FGF above 10 rig/ml, both collagen synthesis/cell and percentage collagen synthesis were significantly less than that seen in the absence of FGF. In order to demonstrate that lipid peroxidation was responsible for the effect of ascorbate on the induction of collagen synthesis by TGF-8, the ability of the lipid per-
In the present studies, we tested the ability of ascorbateinduced lipid peroxidation to interfere with receptor-mediated effects of various growth factors. The effects of these growth factors in low (1 PM) and high (200 PM) concentrations of ascorbic acid were compared. These concentrations were selected so the effects of lipid peroxidation could be differentiated. In previous studies, 200 PM ascorbate was shown to stimulate both lipid peroxidation and collagen synthesis, whereas at 1 PM ascorbic acid, there was shown to be little or no induction of lipid peroxidation (6). We chose to use 1 PM ascorbate because, at this concentration, the activity of the enzyme prolyl hydroxylase is not limited by the availability of ascorbate, a required cofactor, and all collagen molecules are hydroxylated (3) permitting the formation of the characteristic triple helical conformation of collagen which is necessary for secretion (29). Therefore, under the conditions of these experiments, the effect of ascorbic acid induced lipid peroxidation is not compromised by the rate of hydroxylation of proline residues in collagen. Both IL-l (30-33) and EGF (34) have previously been shown to have effects on collagen synthesis by fibroblasts
MS;""" +PRO3 LUTE
i (@id) +
i
i
i
i
(100)
+
(Ott,
(:,
i
i
(To, $0)
FIG. 6. The effect of propyl gallate on the synergism produced with ascorbic acid and TGF-P on collagen synthesis. Collagen synthesis is expressed as the total amount of collagenase-sensitive protein counts per cell as described under Experimental Procedures. Those samples receiving ascorbate as indicated received 200 FM, those with TGF-/3 received 800 pM, and those with propyl gallate received the concentration indicated in parentheses.
10
GEESIN
in culture. Increases in collagen synthesis by IL-l have been reported to be caused by culture conditions including the presence of serum (35) or various added or endogenously generated agents such as PGE2 (32,33) or PDGFAA (21, 36). EGF is believed to induce fibroblast proliferation in healing wounds (34,37) rather than stimulating collagen synthesis directly. In the studies presented here, no effect of IL-l was seen in the presence of 0.5% serum (Fig. l), while only the highest concentration of EGF used produced an inhibition of collagen synthesis in the presence of either concentration of ascorbate (Fig. 2). TGF-P has been consistently shown to stimulate collagen synthesis in fibroblasts in a large number of reports both in viva and in vitro (38-44). Although the detailed mechanism for this effect of TGF-p on collagen synthesis has not been elucidated, there is evidence to support the suggestion that it may be mediated directly through the binding of the DNA binding protein, Nuclear Factor 1 (NF-1) to the (~2 type I collagen gene promoter (45) or indirectly in combination with other factors such as PGE2 (42). In our experiments, an enhancement of collagen synthesis was seen with TGF-@ treatment in both concentrations of ascorbic acid. This induction is much greater with the addition of both TGF-/3 and high ascorbate than with either stimulant alone (Fig. 3A). Our results are consistent with the previous observation that TGF-P and ascorbate act synergistically (43), a phenomenon which we now show to be dependent upon ascorbateinduced lipid peroxidation as shown by the ability of the lipid peroxidation inhibitor, propyl gallate (27,28) to block the effect. Since enhanced collagen synthesis by TGF-P was seen in both concentrations of ascorbic acid, the observation supports the presence of two different pathways for the induction of collagen synthesis in dermal fibroblasts. One pathway is responsive to TGF-/3 and may involve the transcription factor NF-1, while ascorbic acid acts on a different pathway mediated by lipid peroxidation. An alternative explanation is that ascorbate-mediated lipid peroxidation affects a step in the TGF-P signal transduction pathway. PDGF has been reported to stimulate connective tissue synthesis in wounds through the induction of chemoattraction of macrophages and fibroblasts (34,46). In these wounds, high levels of TGF-P were found, apparently synthesized by the PDGF-induced macrophages or fibroblasts, thus producing a TGF-P-mediated effect on connective tissue synthesis (46). In our experiments, PDGF was shown to be able to produce a moderate stimulation of collagen synthesis in the presence of low levels of serum. The effect of PDGF differed from that of TGF-/3 in that it was not selectively affected by ascorbate. FGF, like TGF-/3, is selectively affected by ascorbate at concentrations that produced lipid peroxidation. It differs from TGF-/3 in that it stimulates collagen synthesis at low FGF concentrations only and does not stimulate collagen synthesis in the presence of ascorbate concen-
ET AL.
trations insufficient to stimulate lipid peroxidation. Therefore, FGF stimulation of collagen synthesis appears to occur only in the presence of ascorbate-induced lipid peroxidation and represents a cell membrane associated mechanism capable of being manipulated by ascorbateinduced lipid peroxidation. FGF has been implicated in wound healing through its ability to cause rapid neovascularization and fibroplasia in skin (47) which are assumed to be receptor-mediated (48,49). The recently described interactions between FGF and the extracellular matrix have implied that this growth factor plays an integral role in the wound healing response (50, 51). Our findings indicating that FGF can regulate collagen synthesis in a lipid peroxidation environment has interesting implications for its action in tissue repair. In instances where lipid peroxidation can occur, such as after uv irradiation (52) and oxidant-mediated inflammatory responses (53-56) in wound healing, small amounts of FGF can enhance repair through stimulation of collagen synthesis. FGF itself has been shown to be sensitive to oxidation but is completely prevented from oxidative inactivation when bound to heparin sulfate proteoglycan on the cell surface (57). In these studies we have shown that ascorbate-induced lipid peroxidation can affect the activity of growth factors in dermal fibroblasts. In particular, our results raise the possibility that some step in the pathways by which TGFp and FGF regulate collagen synthesis is sensitive to lipid peroxidation. Whether this mechanism is involved in the increase in collagen transcription seen with ascorbate is not known, but these results support that possibility. The observations here do suggest that any studies performed in the presence of high concentrations of ascorbic acid should be viewed in light of these possible effects of ascorbate-induced lipid peroxidation. REFERENCES 1. Murad, S., Grove, D., Lindberg, K. A., Reynolds, G., Sivarajah, A., and Pinnell, S. R. (1981) Proc. Natl. Acad. Sci. USA 78,2879-2882. 2. Geesin, J., Murad, S., and Pinnell, S. R. (1986) Biochim. Biophys. Acta 886, 272-274. 3. Pinnell, S. R., Murad, S., and Darr, D. (1987) Arch. Dermatol. 123, 1684-1686. 4. Geesin, J. C., Darr, D., Kaufman, R., Murad, S., and Pinnell, S. R. (1988) J. Invest. Dermatol. 90, 420-424. 5. Tajima, S., Phillips, C., Murad, S., Kaufman, R., and Pinnell, S. (1988) J. Invest. Dermatol. 90, 611A. 6. Geesin, J. C., Gordon, J. S., and Berg, R. A. (1990) Arch. Biochem. Biophys. 278, 350-355. 7. Chojkier, M., Houglum, K., Solis-Herruzo, J., and Brenner, D. A. (1989) J. Biol. &em. 264, 16,957-16,962. 8. Baba, A., Lee, E., Ohta, A., Tatsuno, T., and Iwata, H. (1981) J. Biol. Chem. 256, 3679-3684. 9. Lee, E., Baba, A., Ohta, A., and Iwata, H. (1982) Biochim. Biophys. Acta 689,370-374. 10. Sobolev, A. S., Rosenkranz, A. A., and Kasarov, A. R. (1983) J. Radiut. Biol. 44, 31-39.
GROWTH
FACTORS,
ASCORBATE,
AND
COLLAGEN
SYNTHESIS
11
11. Paradisi, L., Panagini, C., Paroloa, M., Barrera, G., and Dianzani, M. U. (1985) Chem. Biol. Interact. 53, 209-217.
34. Buckley, A., Davidson, J. M., Kameroth, C., Wolt, T. B., and Woodward, S. C. (1985) Proc. N&l. Acad. Sci. USA 82, 8340-8344.
12. Lyzlova, L. V., Persianova, V. R., Antipenko, S. N. (1987) Biochem. Int. 14,1079-1086.
35. Elias, J. A., Freundlich, B., Adams, S., and Rosenbloom, J. (1990) Ann. N. Y. Acud. Sci. 580, 233-244. 36. Raines, L. G., Dower, S. K., and Ross, R. (1989) Science 243,393396. 37. Canalis, E., and Raisz, L. G. (1979) Endocrinology 104, 862-869. 38. Ignotz, R. A., Endo, T., and Massague, J. (1987) J. Biol. Chem. 262,
A. E., and Lyzlova,
13. Shasby, D. M., Yorek, M., and Shasby, S. S. (1988) Blood 72,491499. 14. O’Brian, C. A., Ward, N. E., Weinstein, I. B., Bull, A. W., and Marnett, L. J. (1988) Biochem. Biophys. Res. Commun. 155,1374-1380. 15. Delclos, K. B., and Blumberg, P. M. (1982) Cancer Res. 42, 12271232. 16. Heikkila, R. E., Cabbat, F. S., and Manzino, L. (1982) J. Neurochem.
38,1000-1006. 17. Heikkila, R. E. (1983) Eur. J. Pharmucol. 93, 79-85. 18. Hynes, R. 0. (1987) Cell 48, 549-554. 19. Ullrich, A., Coussens, L., Hayflick, J. S., Dull, T. J., Gray, A., Tam, A. W., Lee, J., Yarden, Y., Libermann, T. A., Schlessinger, J., Downward, J., Mayes, E. L. V., Whittle, N., Waterfield, M. D., and Seeburg, P. H. (1984) Nature 309, 418-425. 20. Rifkin, D. B., and Moscatelli, D. (1989) J. Cell Biol. 109, 1-6. 21. Singh, J. P., Adams, L. D., and Bonin, P. D. (1988) J. Cell Biol.
106,813-819. 22. Yarden, Y., Escobedo, J. A., Kuang, W.-J., Yang-Feng, T. L., Daniel, T. O., Tremble, P., Chen,E. Y., Ando, M. E., Harkins, R.N., Francke, U., Friend, V. A., Ullrich, A., and Williams, L. T. (1986) Nature
323,226-232. 23. Sporn, M. B., Roberts, A. B., Wakefield, L. M., and de Crombrugghe, B. (1987) J. Cell Biol. 105,1039-1045. 24. Cheifetz, S., and Massague, J. (1989) J. Biol. Chem. 264, 12,02512,028. 25. Breul, S. D., Bradley, K. H., Hance, A. J., Schafer, M. P., Berg, R. A., and Crystal, R. G. (1980) J. Biol. Chem. 255, 5250-5260. 26. Paulsson, Y., Bywater, M., Heldin, C.-H., and Westermark, B. (1987) Exp. CellRes. 171, 186-194. 27. Heikkila, R. E., and Manzino, L. (1987) Ann. N. Y. Acud. Sci. 498,
63-76. 28. Heikkila,
R. E., and Cabbat, F. S. (1983) J. Neurochem. 41,13841392. 29. Berg, R. A., Steinmann, B., Rennard, S. I., and Crystal, R. G. (1983) Arch. Biochem. Biophys. 226, 681-686. 30. Bhatnagar, R., Penfornis, H., Mauviel, A., Loyau, G., Saklatvala, J., and Pujol, J.-P. (1986) Biochem. Znt. 13, 709-720. 31. Kahari, V.-M., Heino, J., and Vuorio, E. (1987) Biochim. Biophys. Actu 929, 142-147.
32. Goldring, M. B., and Krane, S. M. (1987) J. Biol. Chem. 262,16,72416,729. 33. Postlethwaite, A. E., Raghow, R., Stricklin, G. P., Poppleton, Seyer, J. M., and Kang, A. H. (1988) J. Cell Biol. 106,311-318.
H.,
6443-6446. 39. Fine, A., and Goldstein, R. H. (1987) J. Biol. Chem. 262, 38973902. 40. Penttinen, R. P., Kobayashi, S., and Bornstein, P. (1988) Proc. Nutl. Acud. Sci. USA 85, 1105-1108. 41. Keski-Oja, J., Raghow, R., Sawdey, M., Loskutoff, D. J., Postlethwaite, A. E., Kang, A. H., and Moses, H. L. (1988) J. Biol. Chem. 263, 3111-3115. 42. Diaz, A., Varga, J., and Jimenez, S. A. (1989) J. Biol. Chem. 264, 11,554-11,557. 43. Appling, W. D., O’Brien, W. R., Johnston, D. A., and Duvic, M. (1989) FEBS L&t. 250, 541-544. 44. Roberts, A. B., Heine, U. I., Flanders, K. C., and Sporn, M. B. (1990) Ann. N. Y. Acud. Sci. 580, 225-232. 45. Rossi, P., Karsenth, G., Roberts, A. B., Roche, N. S., Sporn, M. B., and de Crombrugghe, B. (1988) Cell 52, 405-414. 46. Pierce, G. F., Mustoe, T. A., Lingelbach, J., Masakowski, V. P., Griffin, G. L., Senior, R. M., and Deuel, T. F. (1989) J. Cell Biol.
109,429-440. 47. Folkman, J., and Klagsbrun, M. (1987) Science 235, 442-447. 48. Olwin, B. B., and Hauschka, S. D. (1986) Biochemistry 25, 34873492. 49. Neufeld, G., and Gospodarowicz, D. (1986) J. Biol. Chem. 261,5631-
5637. 50. Bashkin, P., Doctrow,
S., Klagsbrun, M., Svahn, C. M., Folkman, J., and Vlodavsky, I. (1989) Biochemistry 28, 1737-1743. 51. Saksela, O., and Rifkin, D. B. (1990) J. CelZBiol. 110, 767-775. 52. Danno, K., Horio, T., Takigawa, M., and Imamura, S. (1984) J. Znuest. Dermutol. 83, 166-168. 53. Fantone, J. C., and Ward, P. A. (1982) Am. J. Puthol. 107, 397-
418. 54. Cathcart, M. K., Morel, D. W., and Chisolm, G. M. (1985) J. Leukocyte Biol. 38, 341-350. 55. Cook, R. M., Ashworth, R. F., and Musgrove, N. R. J. (1987) Znt. Archs. Allergy Appl. Immun. 83,428-431. 56. Yui, S., and Yamazaki, M. (1990) J. Immunol. 144, 1466-1471. 57. Ortega, S., Schaeffer, M. T., Soderman, D., Di Salvo, J., Linemeyer, D. L., Gimmenez-Gallego, G., and Thomas, K. A. (1991) J. Biol. Chem. 266, 5842-5846.
