Collagen biosynthesis by human skin fibroblasts III. The effects of ascorbic acid on procollagen production and prolyl hydroxylase activity

117

Biochimica etBiophysicaActa, 675 (1981) 117-122

Elsevier/North-HollandBiomedicalPress BBA 29613 COLLAGEN BIOSYNTHESIS BY HUMAN SKIN FIBROBLASTS III. THE EFFECTS OF ASCORBIC ACID ON PROCOLLAGEN PRODUCTION AND PROLYL HYDROXYLASE A C T I V I T Y BARBARA A. BOOTHa and JOUNI UITTO b,* a Division o f Dermatology, Department o f Medicine, Washington University School o f Medicine, St. Louis, 310 63110, and b Division of Dermatology, Department o f Medicine, Harbor-UCLA Medical Center, Torrance, CA 90509 (U.S.A.J

(Received January 30th, 1981)

Key words: C•••agen synthesis; Asc•rbic acid; Pr•c•••agen;

Pr••y• hydr•xy•ase; (Human skin •br•b•ast•

Human skin fibroblasts were cultured under conditions optimized for collagen synthesis, and the effects of ascorbic acid on procollagen production, proline hydroxylation and the activity of prolyl hydroxylase were examined in cultures. The results indicated that addition of ascorbic acid to confluent monolayer cultures of adult human skin fibroblasts markedly increased the amount of [aH]hydroxyproline synthesized. Ascorbic acid, however, did not increase the synthesis of all-labeled collagenous polypeptides assayed independently of hydroxylation of proline residues, nor did it affect the amount of prolyl hydroxylase detectable by an in vitro enzyme assay. Also long-term cultures of the cells or initiation of fibroblast cultures in the presence of ascorbic acid did not lead to an apparent selection of a cell population which might be abnormally responsive to ascorbic acid. Thus, ascorbic acid appears to have one primary action on the synthesis of procollagen by cultured human skin fibroblasts: it is necessary for synthesis of hydroxyproline, and consequently for proper triple helix formation and secretion of procollagen.

Introduction

Ascorbic acid has been shown to be important in the synthesis of collagen, and specifically, in the posttranslational formation of hydroxyproline from proline (for reviews see Refs. 1-5). Although it is generally accepted that ascorbic acid serves as a cofactor in the prolyl hydroxylase reaction, there has been some controversy about possible additional effects of the vitamin; some investigators have suggested that ascorbic acid increases the synthesis of collagen polypep* To whom correspondence and reprint requests should be addressed. Abbreviations: Hepes, N-2-hydroxyethylpiperazine-N'-2ethanesulfonic acid; Tris, 2-amino-2-hydroxymethyl-l,3propanediol.

tides [6,7] while others have suggested that it increases the activity of prolyl hydroxylase [8-11]. There are disagreements, however, about both of these putative actions [ 12-17]. In a previous paper of the present series [18,19], we demonstrated that ascorbic acid increases the amount of [3H]hydroxyproline synthesized by cultured human skin fibroblasts. In this paper we have further explored the effects of ascorbic acid on procollagen synthesis, proline hydroxylation, and prolyl hydroxylase activity. Methods Materials. Unless otherwise indicated, the materials were obtained from the same sources as indicated previousiy [ 18,19].

0304-4165/81/0000-0000/$02.50 © Elsevier/North-HoUandBiomedicalPress

118 Human skin fibroblast cultures. Standard culture conditions (details in Ref. 18) included growth of fibroblasts in Dulbecco's modified Eagle's medium plus glutamine containing 30 mM Hepes buffer, pH 7.55 at 37°C, 200 units/ml penicillin, and 20 gg/ml streptomycin. Fetal calf serum (10%)was used during growth of the cultures. Cells were labeled with [aH]proline in the same medium containing 10% dialyzed fetal calf serum and 20 ktg/ml ~-aminopropionitrile. Ascorbic acid was used at varying concentrations. After preincubation, 10 gCi of [aH]proline was added and the cultures were labeled at 37°C for 20 h unless otherwise specified. At the end of the labeling period, the cell and medium fractions were separated as described previously [18] and protease inhibitors were added to give final concentrations of 20 mM Na2EDTA, 10 mM N-ethylmaleimide and 0.3 mM phenylmethylsulphonyl fluoride. Prolyl hydroxylase assay. [3H]Protocollagen was prepared from chick embryo tendons as described by Berg and Prockop [20]. The final preparation contained 170000 dpm [aH]proline per 10 /.tl. Cells were harvested from T-75 or T-150 flasks by scraping with a rubber policeman in a small volume of 0.2 M NaC1, 0.2 M Tris and 0.01% Triton X-100, pH 7.5. The cell suspension was hand-homogenized on ice with 20 strokes of a Teflon glass tissue grinder and then sonicated for 10 s. The suspension was extracted for 60 rain at 4°C with gentle stirring and then centrifuged at 3 0 0 0 0 X g for 30 rain. The supernatant was dialyzed, with one change of buffer, for 2.5 h against enzyme buffer (0.2 M NaC1/0.01 M Tris, pH 7.5) and then immediately assayed. The assay mixture contained 10 gl substrate and 50/.d cofactor mixture in 0.5 M Tris, pH 7.5.A variable amount of enzyme preparation, and buffer were added to a final volume of 0.5 ml. The final cofactor concentrations were 2 mM ascorbic acid, 0.08 M FeSO4 and 0.5 mM c~-ketoglutaric acid. The substrate was heated at 90°C for 3 rain just before use and the assay initiated by addition of the enzyme preparation. Tubes were incubated at 37°C with gentle shaking for 60 min, and the reaction was stopped by the addition of 0.5 ml concentrated HC1. The tubes were sealed, hydrolyzed and assayed for [aH]hydroxyproline [21 ]. Other assays. The radioactive hydroxyproline and total incorporated radioactivity were assayed by a

specific radiochemical method, as described previously [21]. As an independent measure of the synthesis of collagenous proteins, dialyzed samples were incubated with bacterial collagenase, Type Vt (Sigma Chemical Co., St. Louis, MO), in the presence of 0.01 M Tris, 0.15M NaC1, 0.005M CaC12, t mM N-ethylmaleimide, and 30/sM phenylmethylsulfonyl fluoride, at pH 7.6, for 2 h at 25°C. After incubation, the samples were dialyzed against a large volume of buffer containing 1 mM EDTA. Samples for scintillation counting were taken before and after this final dialysis [22]. For assay of DNA and total cell protein, aliquots of the cell homogenate were dialyzed against 0.t5 M NaC1, 1 mM Tris (pH 7.5) and 1 mM Na2EDTA. DNA was measured by a diphenylamine method [23] and protein was assayed according to Lowry et al. [24]. Results

Procollagen synthesis by cells grown in the presence or absence ofascorbic acid Fibroblasts were passed and grown to visual confluence (4 days) in the presence or absence of ascorbic acid (50 /.tg/ml) added in fresh medium every 2 days. Although most of the ascorbic acid has been metabolized after 2 days in culture [15], previous studies by us [18] have demonstrated that after the addition of 50 btg/ml of ascorbic acid, sufficient concentrations for optimal collagen synthesis still exist in cultures 48 h later. At confluence, these cultures were preincubated with 25 or 50 #g/ml ascorbic acid for 4 h and then labeled in the presence of ascorbic acid for 20 h. Growth in the presence of ascorbic acid did not increase the amount of [3H]hydroxyprotine or the amount of collagenase-sensitive material synthesized (Table I). In further experiments, primary cultures were initiated from biopsies of normal human skin; one set of cultures was grown in the presence of ascorbic acid and another set was grown in the absence of ascorbic acid. Every 2 days, fresh medium containing 50 /.tg/ml ascorbic acid was added. At passage 6, confluent cultures from each group were preincubated for 4 h with 50 ~g]ml ascorbic acid and then labeled with [3H]proline for 20 h in the presence of ascorbic acid. There was no difference in [aH]hydroxyproline synthesized or in the percentage of proline residues converted to hydroxyproline between the two groups of cell cultures.

119 TABLE I THE SYNTHESIS OF [3H]PROCOLLAGEN BY HUMAN SKIN FIBROBLASTS GROWN IN THE PRESENCE OR ABSENCE OF ASCORBIC ACID Cells were grown for 4 days in the presence or absence of ascorbic acid, 25 or 50/~g/ml, in two separate experiments with different cell lines. Media was changed every 2 days. At confluence, cultures were preincubated with 25 or 50/2g/ml ascorbic acid for 4 h and then labeled with [3H]proline in the presence of ascorbic acid for 20 h. The cultures were harvested and assayed for [3H]hydroxyproline, 3H-labeled collagenous material, cell protein and DNA. Values are expressed as (dpm/20 h) (×10-3) and are the mean ± S.D. of three parallel cultures. [3H]Hydroxyproline

Days in the presence of ascorbic acid

(per mg protein)

Collagenase-released 3H-labeled peptides (per ag DNA) (per mg protein)

(per/2g DNA)

20 19

Expt. 1 25 ug/ml ascorbic acid 1 5

464 ± 6 542 ± 29

9.6 ± 2.0 9.1 ± 0.6

1350 ± 348 1237 ± 111

Expt. 2 50 ~g/ml ascorbic acid 1 5

171 ± 5 180±26

2.9 ± 0.1 3.4±0.8

474 ± 33 456± 39

Effect of preincubation with ascorbic acid To examine whether there is a delay in the action of ascorbic acid on procollagen production, cells were preincubated for 4 h and then labeled with [3H]proline for 8 h; 50/.tg/ml ascorbic acid was added to the cultures at the beginning of the preincubation ( - 4 h), at the time of labeling (0 h) or half-way through the TABLE II THE EFFECT OF PREINCUBATION OF FIBROBLAST CULTURES WITH ASCORBIC ACID ON THE PRODUCTION OF [3H]HYDROXYPROLINE Confluent fibroblast cultures were preincubated for 4 h and then labeled with [3H]proline for 8 h. 50 ,ug/ml ascorbic acid was added to the cultures at varying times. Times for addition of ascorbic acid: - 4 , 4 h prior to labeling; 0, at the time of labeling; +4, 4 h after the start of labeling (half-way through the labeling period); + = plus ascorbic acid; - = minus ascorbic acid. Values are mean ± S.D. from three parallel cultures. Time (h) -4 0

+4

-

+

+ +

+ + +

[3H]Hydroxyproline (dpm/8 h) (×10-3) (per mg protein)

(per/2g DNA)

55±14 75±13 130±27 130213

1.4±0.4 1.9±0.2 3.6±0£ 3.6±0.1

[3H]Hydroxyproline per total 3H (%) 45±0.3 7.7±0.2 9.7±1.2 10.2±0.9

±4 ±2

8.1 ± 0.6 8.0±0.9

labeling (+4h) Table II). The results indicated that there was a marked increase in the production of nondialyzable [3H]hydroxyproline when ascorbic acid was added to the culture. No difference, however, was noted between cultures which had ascorbic acid present only during the labeling period or 4 h prior to labeling as well. On the other hand, in the cultures incubated with ascorbic acid only half-way through the labeling period, intermediate values between maximal stimulation and the control without ascorbic acid was noted. Similar observations were made in additional experiments employing lower ascorbic acid concentrations and shorter incubation times. The results suggest, therefore, that there is no appreciable lag in the initiation of the effect after addition of ascorbic acid into the culture medium.

Effect of ascorbic acid on prolyl hydroxylase activity Cell cultures in stationary phase were incubated with 50 #g/ml ascorbic acid, and prolyl hydroxylase activity was assayed and compared with the activity in cells not receiving ascorbic acid (Fig. 1 and Table III). Cells incubated for 3 h or overnight with ascorbic acid did not show increased enzyme activity. Somewhat surprisingly, in three separate experiments, the prolyl hydroxylase activity was lower in the group which was preincubated overnight with ascorbic acid than in the controls.

120 TABLE III

5.0-

THE ACTIVITY OF PROLYL HYDROXYLASE IN CELLS INCUBATED WITH OR WITHOUT ASCORBIC ACID

b

Cell cultures in visual confluency were preincubated for 3 or 18 h with 50 ~zg/ml ascorbic acid, and the activity of prolyl hydroxylase was assayed. In the assay, a blank without enzyme and another with 10 mM c~,~'-dipyridyl were included. The assays were performed on the linear range of the enzyme curve (Fig. 1). The presence of [3H]hydroxyproline in the blanks reflects pre-existing labeled hydroxyproline, since the same amount of [aH]hydroxyproline was present in substrate without incubation. Values are mean ± S.D. of three parallel assays.

4.0-

E

/

"" 3.0u.i

z D

._1 0 n," 121.. >- 2 . 0 -

x

o n.(:3 >..Iii,'I i

/

1.0-

O0

I 0.1

Sample

I 0.2

I 0.3

I 0.4

A M O U N T OF C E L L E X T R A C T (ml)

Fig. 1. Assay of prolyl hydroxylase activity in cultured human skin fibroblasts. Various amounts of enzyme preparation, obtained by extraction of cells, were incubated with [3H]prolineqabeled protocollagen substrate and the synthesis of [3H]hydroxyproline was taken as a measure of the activity of prolyl hydroxylase. A blank obtained by incubation of the substrate with c~,a'-dipyridyl (see Table III) was subtracted from the values. The results indicate that the assay is linear with 0.1-0.4 ml of the cell extract, a range used in the subsequent enzyme determinations.

[3H]Hydroxyproline formed (dpm/flask) (×10-3)

[3H]Hydroxyprolim per total 3 H-labeled substrate (%)

Blank (minus enzyme)

15.4 -+ 1.2

7.5 -+0.2

Blank (plus c~,c~'dipyridyl)

13 .5 -+2.2

7.5 + 0.3

Cells, minus ascorbic acid

35.9 -+5.6

17.4 -+ 1.1

Cells, plus ascorbic acid for 3 h

35.4 -+2.8

17.5 -+0.1

Cells, plus ascorbic acid for 18 h

22.2 -+ 3.0

10.0 -+ 1.5

Discussion Since the growth stage o f cell cultures could modulate the response to ascorbic acid (see Ref. 9), similar experiments were performed with cultures in early log and late log phase. The relative cell n u m b e r s in these cultures, based on DNA c o n t e n t o f the cultures, were 25 and 70% o f the c o n f l u e n t control cultures, respectively. Again, no stimulation of prolyl hydroxylase activity in cultures i n c u b a t e d with ascorbic acid for 3 h could be n o t e d , w h e n compared to cultures incubated w i t h o u t ascorbic acid. Specifically, the relative enzyme activities in cultures i n c u b a t e d with ascorbic acid in these two experiments were 94 and 108% of the controls, respectively.

Our experiments [ 18], along with others (see Refs. 1 - 5 ) employing different cell types, have shown that ascorbic acid increases the h y d r o x y l a t i o n o f proline residues and the secretion o f procollagen. It is generally agreed that ascorbic acid acts as a cofactor in the prolyl hydroxylase reaction [ 1 - 5 ] , although Peterkofsky et al. [25] have recently suggested that in confluent cultures o f L-929 cells, ascorbate is n o t essential for proline h y d r o x y l a t i o n . In our experiments with c o n f l u e n t cultures of h u m a n skin fibroblasts, however, ascorbic acid markedly stimulated the form a t i o n of [aH]hydroxyproline.

121 When fibroblasts were grown in the presence of ascorbic acid and compared at confluence with the same cells grown in the absence of ascorbate, but labeled in its presence, there was no difference in the amount of collagenous material synthesized when measured by [3H]hydroxyproline assay and by bacterial collagenase digestion. Other investigators found that in some cells, ascorbic acid did increase the amount of collagenous material synthesized [6,7], while in other cells [15] the rate of collagen synthesis was unaffected. Fibroblasts grown from initiation of culture in the presence of ascorbic acid did not synthesize an increased amount of collagenous material when compared with the same line grown in the absence of the vitamin. These results indicate that there is not apparent selection of a cell population, which might be abnormally responsive to ascorbic acid, during the culture in the presence of the vitamin. It has also been suggested that ascorbic acid increases the activity of prolyl hydroxylase. Stassen et al. [8] found increased enzyme activity after incubation with ascorbic acid. Levene et al. [9] found that prolyl hydroxylase activity was increased when 3T6 ceils in late log phase, but not in stationary phase, were preincubated with ascorbate for 4 h. Evans and Peterkofsky [10] found that a 2 h preincubation with ascorbate gave a 2-fold increase in enzyme activity in log phase Balb 3T3 cells. In stationary phase they found an increase in enzyme activity in one experiment but not in a second experiment [ 10]. Recently, Kuttan et al. [ 11 ] reported increased enzyme activity in human synovial fibroblasts after a 3 h incubation with ascorbate. They did not see any increase in total collagen production [11 ]. In contrast to the above studies, Berg et al. [12] assayed prolyl hydroxylase activity in L-929 cells in early and late log phase cultures and did not find an increase in enzyme activity. Similarly, Kao et al. [13, 14] found no stimulation of prolyl hydroxylase activity or increased enzyme protein after incubation of chick embryo tendon fibroblasts with ascorbic acid. In the human skin fibroblasts used in our studies, incubation with ascorbic acid did not increase the activity of prolyl hydroxylase in early log, late log or stationary phase of growth. The differences in the apparent action of ascorbic acid noted in various studies may be due to differences in cell type, in

experimental design or they may be due to the use of different assays for prolyl hydroxylase activity. In conclusion, in cultured human skin fibroblasts ascorbic acid appears to have one primary action in the synthesis of collagen: it is necessary for the hydroxylation of proline residues and thus for proper helix formation and secretion of procollagen. It does not increase the amount of collagenous material synthesized nor the activity of prolyl hydroxylase. Acknowledgements The authors would like to thank Mr. Gary Kinder, Ms. Kathy Polak, Dr. Perdita Fisher and Ms. Martha Kaplan for their excellent technical assistance. This study was supported in part by United States Public Health Service, National Institutes of Health grants GM-28833 and TO-AM 07284, and by a grant from National Foundation March of Dimes. Dr. Uitto is recipient of Research Career Development Award 7-K04-AM-00897 from National Institutes of Health. A preliminary report of part of this work has been presented at the 37th Annual Meeting of the Midwest Section, The American Federation for Clinical Research, Chicago, IL, November, 1979 [26].

References 1 Peterkofsky, B. and Udenfriend, S. (1965) Proc. Natl. Acad. Sci. U.S.A. 53,335-342 2 Kivirikko, K.I. and Prockop, D.J. (1967) Arch. Biochem. Biophys. 118,611-618 3 Barnes, M.J. and Kodicek, E. (1972) Vitam. Horm. 30, 1~43 4 Cardinale, G.J. and Udenfriend, S. (1974) Adv. Enzymol. 41,245-290 5 Ptockop, D.J., Berg, R.A., Kivirikko, K.I. and Uitto, J. (1976) in Biochemistry of Collagen (Ramachandran, G.N. and Reddi, A.H., eds.), pp. 163-273, Plenum, New York 6 Green, H., Todaro, G.T. and Goldberg, B. (1966) Nature 209,916-917 7 Faris, B., Snider, R., Levine, A., Moscaritolo, R., Salcedo, L. and Franzblau, C. (1978) In Vitro 14, 1022-1027 8 Stassen, F.L.H., Cardinale, G.J. and Udenfriend, S. (1973) Proc. Natl. Acad. Sci. U.S.A. 70, 1090-1093 9 Levene, C.I., Aleo, J.J., Prynne, C.J. and Bates, C.J. (1974) Biochim. Biophys. Acta 338, 29-36 10 Evans, C.A. and Peterkofsky, B. (1976) J. Cell Physiol. 89,355-368 11 Kuttan, R., Parrott, D.A., Kaplan, S.R. and Fttller, G.C. (1979) Res. Commun. Chem. Pathol. Pharmacol. 26, 337-345

122 12 Berg, R.A., Kao, W.W.-Y. and Prockop, D.J. (1976) Biochim. Biophys. Acta 444,756-764 13 Kao, W.W.-Y., Berg, R.A. and Prockop, D.J. (1975) Biochina. Biophys. Acta 411,202-215 14 Kao, W.W.-Y.,Haks, J.G. and Prockop, D.J. (1976) Arch. Biochem. Biophys. 173,638-648 15 Peterkofsky, B. (1972) Arch. Biochem. Biophys. 152, 318-328 16 Bates, C.J., Bailey, A.J., Pryrme, C.J. and Levene, C.I. (1972) Biochim. Biophys. Acta 278,372-390 17 Nolan, J.C., Cardinale, G.J. and Udenfriend, S. (1978) Biochim. Biophys. Acta 543,116-122 18 Booth, B.A., Polak, K.L. and Uitto, J. (1980) Biochim. Biophys. Acta 607, 145-160

19 Uitto, J., Booth, B.A. and Polak, L. (1980) Biochim. Biophys. Acta 624,545-561 20 Berg, R.A. and Prockop, D.J. (1973) Biochemistry 12, 3395-3401 21 Juva, K. and Prockop, D.J. (1966) Anal. Biochem. 15, 77 -83 22 Uitto, J., Lichtenstein, J.R. and Bauer, E.A. (1976) Biochemistry 15, 4935-4942 23 Burton, K. (1956) Biochem. J. 62, 315-323 24 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193,265-275 25 Peterkofsky, B., Kalwinsky, D. and Assad, R. (1980) Arch. Biochem. Biophys. 199,363-373 26 Booth, B.A., Polak, K.L. and Uitto, J. (1979) Clin. Res. 27,623A