FEBS 10925 Volume 301, number 1, 107-l IO Q 1992 Federation of European Riochcmicul Socictics 00115793/92/SS,OO

April 1992

Phorbol ester-induced secretion of human hepatocyte growth factor by human skin fibroblasts and its inhibition by dexamethasone Eiichi Gohda”, Hirotoshi Kataoka”, Hirohito Tsubouchib, Yasushi Daikilara’ and Itaru Yamamoto’

Received 17 February 1992 Humanskin Rbroblusts sccretcd u cc&n MIIIOUII~ of human hcp~locytcgrowlh fxtor (hHGF), asdetermined by an enzyme-linked immunosorbcnt assay for hHGF. This hHGF sccrction W;IS rcmurkubly stimulutcd by prolcin kinxse C (PKC)-activating phorbol esters. which was inhibited by ~he simuhancous addition ol’ dcxnmc~hasons. Prtztrcufmcnl wiih phorbol 12.myristntc 13.acctnu (PMA) caused a down-rcgulaiion in hHGF sccrciion. hHGF sccrcicd by the PMA-lrcatcd cells showed B polcnr hcpalocylc growth-promoting activity which wu ncutralizcd by un ami-hHGF tiniiscrum. Thcsc rcsulis indicaic both that PMA-trcatcd humnn skin fibroblusls product biologically aciivc hHGF and the possible involvcmcm of PKC uctivation in this process. Human

hcpu~ocytc growth fncior: Phorbol cstcr: Protein kinasc C; Human skin fibroblnst; Dcxamcthasonc: DNA

1. INTRODUCTION Hepatocyte growth factor (HGF). which stimulates DNA synthesis on adult rat hepatocytes in primary culture, was purified from the plasma of patients with fulminant hepatic failure [I], rat platetcs [2], rabbit serum [3], normal human plasma [3], and conditioned human embryonic lung fibroblast medium [4], HGF is a heterodimcr which consists of a heavy chain of about 60,000 Da, and a light chain of about 35,000 Da. linked together, probably by a single disulfide bond [ I,-41.Molecular cloning of human HGF (hHGF) has revealed that it is synthesized as a single poiypeptide chain precursor of 728 amino acids with a signal peptide at the N-terminal [5,6]. HGF stimulates the growth of primary cultured rat hcpatocytcs at less than one-tenth the molar concentrations of transforming growth factor-a and epidermal growth factor, other potent mitogcns for hepatocytes [7]. Many lines of evidence support the concept that HGF is a physiological hepatotrophic factor in liver regcncration. Levels of HGF or an HGF-like factor in the serum Abbreriurium: hHGF, human hcpatocytc growth bctor: HGF. hepatocytc growth factor: PMA. phorbol 12.myristatc 13.ecciae PDBu. phorbol 12. I3-dibutyratc; 4~.PDD, 4a-phorbol I2,I3-didecunonte; PKC. protein kinasc C; hSF. human scatter factor; MEM, (Eugle’sj minimm essemial medium: EtfSA. cnzymc-linked immunosorbent assay; BSA. bovine serum albumin; PBS, phosphate-burcrcd saline.

C~~rr?~pohvr~ dh~,~,~: 1.Ynmamoto, Dcpa:tm:nt

of Immunochemistry, Faculty of Pharmaccuticul Scicnccs. Oknyama University, Tsushima. Okayama 700. Japan. Fax: (81) (862) 55 7456.

synthesis

and liver of mice and rats that were treated with carbon tetrachloride or partially hcpatectomized increased mtirkcdly prior to liver regeneration [S-IO]. These increases were accompanied by an elevation of the mRNA levels of this factor in the liver [l l-141. It was recently shown that cells stimulated to express the HGF gene in hepatotoxin-damaged rat liver were Kupffer and endothelial cells [ 151.However, little is known regarding the induction of HGF production by these cells. In this report we show that human skin fibroblasts secreted hHGF and that the secretion was markedly stimulated by addition of protein kinasc C (PKC)-activating phorbol esters to the cultures. We also found that this stimulation was inhibited by dexamethasone. 2. MATERlALS AND METHODS Eagle’s minimum csscntial medium (MEM) was purchased from Nissui Pharmacculical Co., Tokyo; fetal bovine serum wtis from Gibco. Grand Island. NY: phorbol Il-myristak? 13.acctntc (PMA). phorbol 12,lj-dibulyrate (PDBu), 4lr-phorbol 12.13.didculanoatc (4% PDD). and dexamcthasonc wcrc from Sigma Chum&l Co.. St. Louis. MO: Cellmatrix- I P was from Nhta Gclatinc Co.. Yao: and mcthyl[“Hlthymidinc (0.74 TBq/mmol) was from Du Pont-New Eqlund Nuclear, Boston. MA. Enzyme-linked immunosorbcm assay (El-ISA) kits for hHGF [ 161 and rabbit anti-hHGF antiserum [7] were pcncrously supplied by Olsuku Assay Laboratories, Otsuka Pharmaccuticvl Co, Tokushimn. hHGF was purified from the plasma patientswhh fulminam hepatic failure, as dcscrihed previously [I].

of

2.2. C~lrurc u~kurnu~r skirt fihroblusts ml IIWI/I~IO~~ q/ cell.s Human abdominal hkin fibroblasts. obtained from a normal baby (female, 4 months old) 1171. and kindly provided bF.‘hc Dcpartmcnt of Dermatology, Sllinshu University School of Mcdianc. Matsumoto,

107

Volulllr 301. lIulIlber

I

FEBS

LEI-I-ERS

April

wcrc used at population doubling Icvcls of I S-30. The cells wcrc grown as monolaycrs in MEM supplcmentcd with 10% fciill bovine serum nt 37°C in ;I humiditied :ttmospheru of 5% CO2 in air. After growth IO conlluc~~cy in 24.well plastic dishes (Nunc), the medium (I ml) was rcpluccd with ;I fresh 011~’ to which the test compounds were added. Culture mrdiu wcrc collcctcd after incubittion for various times and w’c’rc immudi~tcly frozen at -30°C. Ccl1 monoluycrs wcrc then washed 4 times with plr~splretc-bull’crcd saline (PBS) und solubilized in I ml of Lowry solution C [18,19], and cellular protein was determined by the method of Lowry et al. [IS]. In some cases. rftcr being washed with PBS, ccl1 layers wcrc scrnpod in ice-cold PBS containing I M NKI and 0.039% Triton X- 100. An aliquot oi the ccl) suspension wus removed for protein ;ISS~. Bovine strum ;tlbumin (BSA) was thon added at a concentration of 025%. following which 111~ccl1 suspension wils snnicutcd und ccntrifuped iIt ZO,Orjir x ,gt:-~ 20 mill tit 4°C. The supernutilnt M’OSstored at -30°C for hHGF ELISA. hHGF lcvcls in the culturr media and cell estructs bvcrc cnpresscd YS ng hHGF per mg of cellulx prolcin.

PMA

1992

(nM)

2.3. EfJS/I Lll‘lfHGF A s;ind\Vich hHGF ELISA was pcrl’ormcd ut room tcmpernlure. as described previously [ 161, Culculiition of the ~imounl of hHGF used us ;l stand;lrd was bxd on the result of its itmino ucid analysis and molecular weights dctermincd by SDS-PAGE [I].

Ths x;ivity of hHGF was dctcrmincd by meusuring its stimulatory cl’fccts on DNA synihcsis in adult ml hcp;llocytcs in primary culture, :IS described previously [ZO]. klcpiltocytes. isolated from adult m;k Wistur rilts (about 100 g) by ;I collugenasc perfusion method, wcrc pl~cd in 21.well plastic dishes (Nuac). prc-coutcd with collagen (Ccllmatrix-IP). at ;I density of25 x lO’ccllslO.2 ml/cm2 (0.38 ml/well), and wcrc incubated ilt 37°C in ;I humiditird ulmosphcre of 5% CO, in air. The plating mediuni w:is Williams mcdiuni E. supplemented 6ith 5% fall bovine ssrum. IO nM dcxumcthusonc. 100 U/mI ofpcnicillin, and 100 pg/ml ol’strcptomycin. The medium WiIs rcplxcd with fresh medium containing growth Ibctors 4 h uftcr plating. DNA synthesis, dctcrmincd by lub~ling cultured hqxtocytcs with [‘H]thymidine (74 kBq/ml. 37 GBtl/mmol) for I8 h at 37°C bctwecn 29 and 47 h after plutinp. was exprcsscd us incorpomtcd [‘H]thymidinc pcrpg ofccllular pro1ciii.

3. RESULTS When confluent human skin tibroblasts were cultured for 5 days, the conditioned medium contained a measurable amount of hHGF as determined by hHGF ELISA, as shown in Fig. ! A. Secretion of hHGF from the cells was markedly stimulated by the addition of protein kinase C-activating phorbol ester, PMA (Fig. 1A). The effect of PMA was maximal at 10 nM, showing about I 1-fold stimulation. The amounts of hHGF in extracts of cell layers untreated and treated with IO nM PMA for 5 days were 4 and 15 ngmg cellular protein, respectively, which were less than one-seventh and one-twentieth of those in their own culture media. The cells treated with IO nM PMA secreted hHGF continuously during 5 days of culture, but in the presence of 100 nM PMA maximal amounts were reached after 24 h of incubation (Fig. 16). Both PDBu and mezerein, other PKC activators, also stimulated hHGF secretion, but 4 CIPDD, known to be inactive for PKC, was without effect, as shown in Fig. 2. Prolonged pretreatment of cultured cells with PMA has been reported to down108

Time

after addition

(days)

Fig. I. Dose-rcsponsc (A) and lime-course (B) I’or PMA4nduccd hHGF secretion by hunxm skin fibroblasts, Confluent cells wcrc incubated without or with the indicated concentrations of PMA (A) for 5 days and (B) without (0) or with IO nM PMA (0) or with 100 nM PMA (m) for the indicated days. hHGP lcvcls in the culture media wcrc dctcrniined by ELISA. Vlrlucs are means ? S.D. for triplicate cultures.

regulate PKC by a depletion of this enzyme [21]. As shown in Fig. 3, pretreatment of human skin fibroblasts with 100 or 500 nM PMA for 24 h resulted in a marked decrease in hHGF secretio,l. Next, we examined this hHGF to determine whether it was biologically active As shown in Fig. 4, the hHGF, partially purified from the culture medium of PMA-treated cells by heparin-Sepharose affinity chromatography, had potent hepatocyte growth-stimulating activity. It showed a dose-response similar to that of hI-IGF purified from the plasma of patients with fulminant hepatic failure, and the activity of both hHGF was neutralized by an anti-hHGF antiserum. We tested the effects of some other biologically active compounds on hHGF secretion from human skin fibroblasts during 24 h incubation. These compounds were dibutyryladenosine 3’,5’-cyclic monophosphate (I-100 PM), dibutyrylguanosine 3’,5’-cyclic monophosphate (l-100 PM), dibutytylcytidine 3’5’.cyclic monophosphate (l-100 yM), ionomycin (30-X0 nM),

Volume 301, number I hHGF (ng/mg

Wl)

-

3

PMA

’

PDBu

8 :8

Mfxerein

i8

protein) 100

50

0

None

April 1992

FEBS LETTERS

I

1

b -II

b

t’ )r

Fip. 2. Ell’cc~s ol’vurious protcin kimlsc C x~ivutors on hHGF sccrction l’rom human skin Rbrobklsts. Conllucnt cells were insubutcd without or with the indicutcd protein kimtsc C ;)ctivu)0rs Ibr 24 h. hHGF levels in the cul~urc mcdiu were dctcrmincd by ELI!%. Vslues xc IIIWIIS f SD. Ibr triplicate culmrcs.

A23187 (IO-100 nM). dcxamcthasot~c (0.01 I-IM-IO PM), and insulin (0.01-i ,uM), None of them signifi-

cantly stimulated hHGF secretion by the cells (data not shown). We also examined the effects of these compounds in PMA-stinulatcd hHGF secretion. hJost compounds had no appreciable effect (data not shown), but dexamethusonc remarkably inhibited the sthnulatcd IIHGF secretion, as shown in Fig. 5. This inhibitory effect, observed at concentrations as low as IO nM, was maximal (about 60% inhibition) at I ,uM.

’

I’

I

lrn”’

0.5 Secreted or

I

III

1

5

authentic

hHGF

(ng/ml)

Fig. 4. The hrpntocytc growth-stimulating activity of hHGF sccrctcd from PMA-trctttcd humnn skin libroblasts, and its inhibition by anlihHGF an)iscrum. IlHGF in the 5.duy-culture mctiia of the fibroblasts trcatcd with PMA was purtially purilied by hcpxin-Scpharosc chromutoymphy. as dcscribcd previously [El. Tl~celuatc, with 1.75 M N;rCI in PBS comaining 0.013% Triton X~IOO. wns dialyrcd against PBS ttlicr BSA wus added al II concentration or2.5 mLr/ml. R~I hcpatocyws wcrc incubated with the indicdtcd concentrations of this hIiGF (3. a) or 01’ hHGF purified from the plasma or patients with I’ulminunt hcputic l%lurc ( ‘, A), Filled symbols show DNA synthesis in hep;rtocvtcs cultured with each hHGF prc-incubutcd with an antihHGF u&crum for 72 h ai 4°C. The final conccntralioii or the iinliscrum in the hcp;~tocytc culture wets O.OGf% Conccntrdtions 0r hHGF in both preparations were determined by ELISA. Values arc mcuns for duplicate cultures.

4. DISCUSSION The present study demonstrated thal the secretion of hHGF by human skin fibroblasts was markedly stimulated by the tumor-promoting phorbol ester, PMA. The following lines of evidc;;ce were supportive of a role for PKC activation in the regulation of hHGF secretion. First, the effective concentration of PMA (IO nM) was similar to that required for the activation of PKC [Xj, Second, other PKC activators were also effective, whereas inactive 4r-PDD exerted no effect. Third, pretreatment of cells with PMA. which has been reported PlrlA (nM)

hHGF

Preincubation

Incubation

0

;mj;

j

F

(Rg/mg

PmtOin)

OkI 0

Fig. 3. ElTccts or pretreatment or ~~III~II skin fibroblasts with PMA on phorbol ester-induced hHGF secretion. Confluent cells were prctreated without or with IO0 or 500 nM PMA Tar 74 h. The ~11s were then washed and incubated with IO nM PMA Ibr ;m additional 24 h. hHGF levels in the culture media wcrc determined by ELISA. Values arc means & SD, for triplicate cultures.

’

1

1

1

10 14 Dexemethasone

!

103 (nh!I

I

104

Fig. 5. Inhibition bydcxamcthasone orPMA-induced hHGFsccrction %nSm\ cells rmc i::cub~tL& with or iiom human skin tibr0blasts. c without the indicated conccntrdtions or dcxamcthasonc in the prcscnc’c or IO nM PMA ror 24 h. hHGF lcvcls in the culture media wrc determined by ELISA. Values arc means 2 SD. for triplicatccullurcs,

IO9

Volume 301, number 1

FEBS LETTERS

to down-regulate PKC activity via depletion of this enzyme, caused a marked decrease in the secretion of hHGF. While this study was in progress, Nishino et al. reported that a promyelocytic leukemia cell line, HL-60, stimulated with PMA produced hHGF '9"~ t_.~j, indicating that the stimulatory effect of PMA on hHGF secretion is not limited to its effect on human skin fibroblasts. It has recently been reported that hHGF and human scatter factor (hSF), which stimulates the motility of epitheliai cells, are identical proteins encoded by a single gene [24--26]. Large amounts of hSF are produced by human embryonic lung fibroblasts such as MRC5 and W138 [27,28]. We found that the level of hHGF/hRF in the conditioned medium of human skin fibroblasts treated with PMA for 24 h was comparable to that of untreated MRC5 cells (our unpublished results). It would be of interest to examine whether PKC is also involved in active hHGF/hSF secretion by MRC5 fibroblasts. Dexamethasone inhibited the PMA-stimulated secretion of hHGF. This fact is noteworthy, since hHGF may play an important role in liver regeneration and since regenerating rat liver under the influence ofglucocorticoids shows pronounced inhibition of mitosis and D N A synthesis [29]. Although dexamethasone itself has been reported, in some cases, to inhibit DNA synthesis in primary hepatocyte cultures [30], it is possible that the inhibition of hHGF secretion by this steroid contributes to its suppressive effect on liver regeneration. Normal human skin fibroblasts cultures may be suitable for the identification of the physiological stimulators of hHG F production. The inhibitors of h HGF production can also be detected by the use of such cultures treated with PMA. Acknowledgements: We are grateful to Otsuka Assay Laboratories for supplying hHGF ELISA kits and rabbit anti-hHGF antiserum. This work was supported, in part. by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan.

REFERENCES [1] Gohda, E,, Tsubouchi, H,, Nakayama° H., Hirono, S., Sakiyama, O,, Takahashi, K,, Miya~,aki, H., Hashimoto, S, and Daikuhara, Y. (1988) .I, Clin, Invest, 81,414-419, [2] Nakamura. T,, Nawa, K., Icl'Jihara. A,, Kaise, N, and Nishino, T, (1987) FEBS Lett, 224, 311--316. [3] Zarnegar, R, and Mid'ta!opoulos. G, (1989) Cancer Res, 49, 3314--3320, [4] Rubin, J.S., Chan, A.M,-L,, Bottaro, D.P,, Burgess, W.H., "ray. lor. W.G.. Cech, A.C,, Hirsehfield, D.W.. Wong, J.. Miki, T., Finch, P.W, and Aaronson, S.A, (1991) Proc, Natl. Acad. Sci. USA 88, 415--419. [5] Miyazawa, K,, Tsubouchi, H,, Naka, D., Takahashi, K., Okigaki, M,. Arakaki, N.. Nakayama, H,, Hirono, S,, Sakiyama,

II0

April 1992

O., Takahoshl, K,. Gohdo. E., Daikulmra. Y. and Kitamura, N. (1989) Biochen't, Biophys, Res, Commun, 163, 967-~73. [6] Nakamura, T,, Nishizawa, T,, Hagiya, .M,, Seki, T,, Shimonishi, M,, Sagimura. A,. Tashiro. K, and Sh!mizu, S, (1989) Nature 342, 440-443, [7] Gohda, E,, Yarnasaki, T., Tst~bouchi, H,, Kurobe, M., Sakiyan~a, O,, Aoki, H,, Niidani, N., Shin, S., Hayashi, K., Hashimoto. S,, Daikuhara. '¢. and Yamamoto, 1, (1990) Biochlm, Biophys, Acttt 1053, 21-26. [8] Gohda, E,, Haya~hi, Y,, Kawaida, A,, Tsubouchi, H, clad Yamamoto, I, (1990) Life Sci. 46, i801-1808. [9] Asami, O,, lhara, I., Shimidzu, N., Shimizu, S.. Tomita, Y., lchihara, A. and Nakamura, T. (1991) J. Biochem. 109, 8-13. [10] Lindroos, P,M., Zarnagar, R. and Michalopoalos. G.K. (1991) Hepatology 13. 743-749, [11] Kinoshita, T,, Tashiro, K. and Nakamura, T. (1989) Biochern. Biophys. Res, Commun, 165, 1229-1234, [12] Okajima, A., Miyazawa, K, al,d Kitamura, N. (1990) Eur. J. Biochcm, 193, 375-381, [13] Selden. C.. Jones, M,. Wade, D. and Hodgson, H. (1990) FEBS Left, 270, 81-84. [14] Zarnegar. R,, DeFranecs. M.C,, Kost. D,P,, Lindroos, R, and Miehalopoulos, G,K. (1991) Biochem. Biophys. Res. Commun. 177, 559-565. [15] Noji. S.. Tashiro. K.. Koyama, E.. Nohrm. T., Ohyama. K., Tat'figuchi. S, and Nakamura, T, (1990) Biochem. Biophys, Res. Commun. 173, 42--47. [16] Tsubouehi, H., Niitani, Y.. Hirono. S,. Nakayama, H.. Gohda. E., Arakaki, N,. Sakiyama. O,, Takahushi, K,, Kimoto, M,. Kawakami. S., Seloguchl. M,. Tachlkawa, T,. Shin, S,. Arima, T. and Daikuhara. Y, (1991) HepatoloBy 13, 1-5, [17] Shindo. Y.. Akiyama. J,. Yamazaki, Y.° Saito, K, and Takase, Y. (1991) Exp, Gerontol. 26, 29-35. [18] Lowry, O,H,, Rosebrough. N.J,, Farr, A,L. and Randall, R.J. (1951) J. Biol. Chem, 193, 265-275. [19] Patterson Jr,, M,K. (1979) Methods Enzymol, 58, 141-152, [20] Gol~da. E.. Tsubouehi. H., Nakayama, H., Hirono. S.. Takahaski, K,, Koura, M,, Hashimoto. S, and Daikuhara, Y. (1986) Exp. Cell Res. 166, 139-1:50. [21] Kishimoto. A,, Takai, Y., Moti, T., Kikkawa. U. and Nishizuka. Y. (1980) J. Biol, Chem, 255, 2273-2276. [22] Castagna, M., Tukai, Y,. Kaibuchi. K,. Sano, K.. Kikkawa, U. and Nishizuka. Y, (1982) J. Biol, Chem, 257. 7874-7851, [23] Nishino, T., Kaise, N., Sindo. Y.. Nishino. N.. Nishida. "I'., Yasuda, S. and Masui, Y, (1991) Bioehem, Biophys. Res. Commun. 181. 323-330. [24] Gherardi, E,, Gray, J., Stoker. M.. Perryman, M. and Furlong, R. (1989) Proc, Natl, Acad, Scl, USA 86, 5844--5848, [25] Gherardi, E, and Stoker, M. (1990) Nature 346. 228. [26] Weidncr, K,M., Arakaki, N,, Hartmann, G., Vandekerckhove, J.. Weingart. S.. Rieder, H., Fonatseh, C,, Tsubouchi, H,, Hishida° T., Daikuhara° Y. and Birchmeier, W. ( 1991) Proe, Natl. Aead, Sci. USA 88. 7001-7005. [27] Stoker, M.. Gherardi, E.. Perryman. M, and Gray. J. (1987) Nature 327, 239-242. [28] Weidner, K,M,, Behrens. J,, V~andekerckhove,J, and Birchmeier, W, (1990) J, Cell Biol, Ill, 2097-2108. [29] Buchner, N,L.R. (1963) in: Intcrnatlonal Review of Cytology (G,H, Bourne and J.F, Danielli, eds,) vol, 15. pp. 245-300, Academic Press, New York. [30] Vintermyr, O.K.a.ad D~skeland. S,O. (1989) 3, Cell. Physiol. 138. 29-37.