Transforming growth factor-β reduces the phenotypic expression of osteoblastic MP3T3-E1 cells in monolayer culture

Bone, 8, 259-262 (1987) Printed in the USA. All rights reserved.

8756-3282/87 $3.00 + .OO Copyright 0 1987 Pergamon Journals Ltd.

TransformingGrowthFactor-PReducesthe Phenotypic Expressionof Osteoblastic MC~TZI-EI Cells in Monolayer Culture P.R. ELFORD, H.L. GUENTHER, R. FELIX, M.G. CECCHINI and H. FLEISCH Department

of Pathophysiology,

Address for CorresDondence CH-3010, Bern, Sw’itzerland.

University of Bern, Murtenstrasse

35, 3010 Bern, Switzerland.

and Reonnts: Dr. P.R. Elford. Deeartment

of Pathophysiology,

University

of Bern, Murtenstrasse

35,

’

ined the action of TGF-P on bone cells, using the clonal nontransformed cell line MC3T3-El (Kodama et al., 1981). Since alkaline phosphatase activity (AP) and cyclic adenosine 3’, 5’-monophosphate (CAMP) response to parathyroid hormone (PTH) are taken as a measure of osteoblast function, we have examined the effect of TGF-p and also of EGF on both these parameters. Furthermore, we have examined the action upon cell proliferation, since osteoblast number must also be of importance in the regulation of bone formation. EGF, a systemic growth factor which previously was shown to affect bone cells, e.g. stimulation of DNA synthesis (Ng et al., 1983) and inhibition of AP activity (Kumegawa et al., 1983) was included in this study as a reference to which the effects of TGF-p shall be compared.

Abstract Transforming growth factor beta (TGF-p) regulates cell growth and differentiation. Since it is abundant in bone, we have studied the effect of the polypeptide upon the growth and phenotypic expression of murine osteoblastic cells in monolayer culture. Its actions were compared to those of epidermal growth factor (EGF), another hormonally active polypeptide known to alter bone cell function. Picogram amounts of TGF-p were found to inhibit the growth and phenotype (alkaline phosphatase and CAMP response to parathyroid hormone) of the clonal nontransformed MC3T3-El osteoblastic cell line. EGF also inhibited phenotypic expression, although at higher (nanogram) concentrations, but stimulated cell growth. The low concentration of TGF-P required to inhibit growth and phenotype of osteoblastic cells together with its abundance in bone suggest that TGF-P may be an important regulator of bone cell function.

Materials and Methods Cells of the osteoblastic clone MC3T3-El (a gift from Dr Kodama, Tohoku Dental University, Japan) were routinely cultured in aminimal essential medium (uMEM) (Flow Laboratories, Irvine, Scotland) containing 10% fetal bovine serum (FBS) (Gibco, Paisley, Scotland) with 100 pgiml penicillin/streptomycin (Seromed. Berlin, W. Germany) in 150 cm2 tissue culture flasks (2 x lo6 cells/flask) and passaged every second or third day to passage 15 as described by Kodama et al. (1981). For use, cells were passaged into 24-well multiwell plates (2 x 104/well) and cultured in the same medium. Since increasing amounts of AP is produced by MC3T3-El cells after confluence (Kodama et al., 1981). experiments to assess the effect of TGF-P or EGF upon AP activity were normally initiated between days lo- 12 Therefore, at confluence (3-4 days), the serum level was reduced to 1% to prevent cell overgrowth At this point, cell number varied between 2-4 x IO5 per well and with the reduction in serum the cells became quiescent. Cells were then cultured in this state to days 10-12. To assess the effect of porcine platelet TGF-P (R and D Systems, Minneapolis, U.S.A.) or mouse submaxillary gland EGF (Sigma, St. Louis, U.S.A.) on AP, medium was replaced with that containing various concentrations of factor. After three days, medium was removed, the cell layer was washed 3 times with 0.9% sodium chloride and the plates frozen at -20°C until assay for AP. Parallel cultures were used to determine cell number; cells were removed using 0.1% type II bacterial collagenase (Fakola, Basel, Switzerland) and 0.05% trypsin (Sigma) in solution A (Gibco) and then counted using a Coulter Counter. The AP assay was carried out as described by Felix and Fleisch (1979), using p-nitrophenylphosphate as the substrate.

Key Words: Alkaline phosphatase-Bone metabolismEpidermal growth factor-Osteoblast-Parathyroid hormone-Transforming growth factor.

Introduction Transforming growth factor beta (TGF-p) is a polypeptide originally identified by the ability to reversibly induce the anchorage independent growth of nontransformed cells in the presence of epidermal growth factor (EGF) (DeLarco and Todaro, 1978). Although first detected in the culture supernatants of neoplastic cells, it is now clear that many normal cells also produce TGF-p (Roberts et al., 1981; Roberts et al., 1984). Indeed, there are reports indicating that TGF-P can modulate a variety of normal cellular functions such as growth rate (Tucker et al., 1984; FrBterSchrtjder et al., 1986), morphology (Shipley et al., 1984) and production of various proteins (Ignotz and MassaguB, 1986; Laiho et al., 1986). The finding that the culture supernatant of fetal rat calvaria (Centrella and Canalis, 1985) and subsequently bovine bone matrix (Seyedin et al., 1986) contains TGF-P suggests that the polypeptide may influence bone cell metabolism. We have therefore exam259

P.R. Elford et al

In order to determine the effect of TGF-6 or EGF on the CAMP response to synthetic bovine 1-34 PTH (Beckman, Palo Alto, U.S.A.), cells were passaged into 24.well multiwell plates as described above. Twenty-four hours later, the medium was replaced with that containing various concentrations of factor. After three days, during which time confluence had been reached, the cells were incubated for 5 min with 0.5 U/ml PTH, using the method described by Rizzoli et al. (1983). CAMP levels were then determined using a competitive protein binding assay (Brown et al., 1971). To measure cell proliferation, cells were passed into 24.well multiwell plates and cultured for 24 h in medium containing 10% serum. This was then replaced with that containing either 2% or 10% serum, with or without factor(s), as detailed in the legends to figures 2 and 3. Cells in triplicate wells were then removed at the time points indicated in the figures, using 0.1% type II bacterial collagenase and 0.05% trypsin in solution A and counted using a Coulter Counter. As assessed by exclusion of trypan blue, 2 ng:ml TGF-8 or 10 ng/ml EGF had no effect upon cell viability.

Results Figure 1 is representative

of six experiments

where

the

EGF upon the AP activity of confluent MC3T3-El cells was examined. As can be seen from the graph, both factors decreased AP activity in what appears to be a dose-related manner, although TGF-8 was the more potent inhibitor of the two. In one experiment (results not shown), cells were exposed to TGF-8 in the growth phase. Again, a decrease in AP was observed, with 0.4 ng/ml of TGF-8 reducing activity to undetectable levels. Table I shows the CAMP response to PTH of MC3T3-El cells in the presence or absence of TGF-8 or EGF, this again being data representative of six experiments. The action of these two factors upon CAMP response to PTH was very similar to that observed upon AP activity. TGF-8 appeared more potent, 80 pg/ml virtually nullifying the ability of the cells to respond to 05/ml PTH. The effect of TGF-8 and EGF upon cell proliferation was first examined by culturing cells in 2% serum with or without factor and comparing growth to that occurring in the presence of 10% serum (Fig. 2). Cells cultured in 2% serum alone or with TGF-8 were almost quiescent, while those in 2% serum with EGF grew at a rate approaching those cultured in 10% serum. When TGF-f3 and EGF were effect

of TGF-8 and

60

f

0” 0.01

0.1 TGF/

1

10

or EGF (ng/mlb

Fig. 1. The effect of TGF-8 or EGF upon the AP of MC3T3-El cells. The exoeriment was carried out as described in materials and methods. Results represent the mean r SEM of triplicate cultures. m control: 0 TGF-8: 0 EGF. Statistics Wudent’s t-test): *significantly different from ‘control, p < 0.05; *‘*significantly different from control, p < 0.005.

TGF-8 suppresses

the osteoblastic

phenotype

Table 1. The effect of TGF-8 or EGF on the CAMP response PTH of MC3T3-El cells.

to

CAMP production pm010 O6 cells/5 min Compound ngiml

Vehicle (without PTH)

PTH 0.5 U/ml

TGF-8 0 0.016 0.08 0.4 2 10

1.4 1.3 2.1 1.6 3.0 2.8

+ -c + 2 ? 2

0.3 0.3 0.5 0.5 0.5 1.0

34.5 12.7 3 3 12 2.9 4.4

? 2 t 2 2 2

5.8 1.6* 0.1** 1.2* 0 6*

EGF 0 0.016 0.08 0.4 2 IO

3.5 2 2.7 2 3.1 I? 3.2 2 3.4 ? 18209

0.5 1.5 0.5 0.5 1.9

38.3 46.9 35.3 32.1 9.3 4 7

? 2 ? 2 2 2

5 1 7.2 18 3 5 1.9*** 3.5***

11.5

Cultures were tested for the increase in cAMP production in response to PTH following growth to confluence in the presence or absence of factor in aMEM + 10% FBS, as described in materials and methods. Results are expressed as the mean ? SEM of triplicate cultures and were statistically evaluated by use of Students’ t-test. ’ p < 0.05 * p < 0.02 (significantly different from control) *** I) < 0.005 l

present together, the stimulatory effect of EGF was partially antagonized. This suggested that the two factors could have opposing effects upon growth and that TGF-8 might inhibit cell proliferation. Figure 3 shows the effect of TGF-8 on the growth of cells in 10% serum. It can be seen that TGF-8 does indeed inhibit growth, with apparent halfmaximal concentration of around 1 ng/ml.

Discussion This study has shown that TGF-8, under the experimental conditions employed, induces a potent reduction in AP activity and CAMP response to PTH of MC3T3-El cells, both functions generally taken as markers to characterize the osteoblastic phenotype (Cohn and Wong, 1979) and also inhibits the growth of MC3T3-El cells. Although it has already been shown that EGF causes an inhibition of AP activity of MC3T3-El cells (Kumegawa et al., 1983) and now as shown here also on the CAMP response to PTH, this investigation now demonstrates that TGF-8 affects the parameters listed above and at a concentration below that required with EGF. The significance of this finding becomes intensified by recent reports in the literature which show that bone forming cells synthesize TGF-8 (Young et al., 1986; Seyedin et al., 1986). Moreover, it is currently thought that bone as a tissue, apart from blood platelets, constitutes the richest source of TGF-8 (Seyedin et al., 1985, 1986; Centrella and Canalis, 1985). Based on these facts and on the data presented here, it appears conceivable that TGF-P, functioning as an autologous factor, may play a role in the regulation of the osteoblastic phenotype. Recent data appearing in the literature would seem to lend support to this notion, as it was demonstrated with a va-

261

P R Elford et al.: TGF-p suppresses the osteoblastic phenotype

1985). Important in this context is the disthat TGF-j3 antagonizes the mitogenic effects Of EGF on bone ceils, a finding which has also been found to prevail with normal rat fibroblasts (Roberts et al., 1985). Thus, since TGF-P was found not to be mitogenic for bone cells grown in monolayer culture, on the contrary an inhibition of cell growth was observed, it would seem that TGF-P would be the more likely of the two factors to participate in riBdi6

Bt al.,

covery

regulating the osteogenic potential of an osteoblastic cell. Although our data may only be valid with bone Cells grown in monolayer culture (anchorage-dependent growth) leaving aside the important aspects of TGF-p acting on cells maintained in semisolid media (Sporn et al., 1986) it is conceivable that TGF-6, presumably in concert with other bone-related factors and/or hormones could be a part of a system whereby the function of osteoblasts is balanced according to the physiological requirements of bone. Note added in proof: During submission this manuscript,

I

I

I

I

1

I

N. Nada

and review of and G.A. Rodan (Biochem.

Biophys. Res. Comm. 140:56, 1986) published a paper about the biological effects of TGF-6 on MC3T3-El cells. The report describes a TGF-p-induced decrease of both cell proliferation and alkaline phosphatase activity, of approximately the magnitude reported here

I

0123456 day Fig. 2. The effect of TGF+ and/or EGF on the growth of MC3T3El cells. The experiment was carried out as outlined in materials and methods, the medium being replaced with the same on day 3. Results represent the mean f SEM of triplicate cultures. 0 2%

serum alone; A 2% serum + 5 nglml EGF; l 10% serum alone; q 2% serum + 1 ng/ml TGF-f3; n 2% serum + 5 ngiml EGF + 1 ng/ml TGF-p.

nety of other cell types that TGF-6 in combmation

with other factors appears to be involved in events that regulate cell differentiation (Sporn et al., 1986). Since EGF promotes proliferation of osteoblast-like cells (Ng et al., 1983), a decrease in AP activity and CAMP response to PTH rnduced by this cytokine may be a result of an increase in cell proliferation. Indeed, it was shown prevrously that a reverse relationship exists between cell proliferation and synthesis of AP with bone cells (Gngo-

ul

0 P *

0

1 TGFp

2

10

ng/ml

Fig. 3. Inhibition of MC3T3-El cell growth by TGF-B The experiment was carried out as described in materials and methods. Cells were cultured for 3 days in the presence of the concentrations of TGF-f3shown. Results represent mean 2 SEM of triplicate

cultures

Acknowledgments: We thank Ms. M Perret for excellent technical assistance and Ms. B. JBrmann and 5. Meier for typing the manuscript. This work has been supported by the Swiss National Science Foundation (grant 3.880-85)

References Brown B L , Albano J Eklns R Sgherzr A and Tampion W.. A simple and sensitrve assay method for the measurement of adenosine 3’ 5’.cyclrc monophosphate. Biochem. J. 121.561-562. 1971. Centrella M. and Canalis E Transformrng and nontransformrng growth factors are present in medium condrtioned by fetal rat calvarra Proc Nali Acad So USA 82 7335-7339. 1985. Cohn D.V and Wong G.. Isolated bone cells In Skeletal Research Summons 0 T. and AS Kunrn. eds Academic Press, New York. 1979. pp 3-20 DeLarco J.E and Todaro G J Growth factors from murlne sarcoma vrrustransformed cells Proc. Nat/ Acad SCI USA 75:4001-4005, 1978. Felix A and Flersch H Increase In alkaline phosphatase activity in calvana cells cultured with drphosphonates Biochem J. 183 73-81, 1979. Frater-Schroder M Muller G Birchmerer W and Bohlen P Transforming growth factor-beta inhibits endothelial cell proliferation Brochem Eiophys Res Commun. 137 295302, 1986 Gngonadrs P M Petrovrch R Aubrn J E and Heersche J N M.. Subclone heterogenerty in a clonally-derived osteoblast-lrke cell lrne Bone 6 249-256, 1985 lgnotz R A and Massague J Transforming growth factor-8 stimulates the expressron of frbronectrn and collagen and their IncorporatIon Into the extracellular matrix J BIO/ Chem 281 4337-4345. 1986 Kodama H , Amagar Y Sudo H Kasar S. and Yamamoto S Establishment of a clonal osteogenrc cell lrne from newborn mouse calvaria Jpn J. Oral. Bfol. 23:899-901, 1981 Kumegawa M Hrramatsu M Hatakeyama K Yafrma T Kodama H Osaki T and Kunsu K Effects of epidermal growth factor on osteoblastlc cells in vrtro Calcrf Tissue Int 35 542-548, 1983 Lalho M Saksela 0 and Keskr-Oja J Transformrng growth factor-p alters plasminogen activator actrvrty In human skrn frbroblasts Exp Ceil Res. 184 399-407, 1986 Ng K W Partridge N C.. Nrall M and Martin T J. Strmulatlon of DNA

262 snythesls by epidermal growth factor in osteoblast-like

cells Calcif. Tissue Int. 36624-628, 1983. Rizzoli R.E., Somerman M , Murray TM and Aurbach G.D Binding of radioiodinated parathyrord hormone to cloned bone cells Endocnnology 113:1832-1838, 1983 Roberts A B Anzano M.A., Lamb L.C., Smith J.M and Sporn M.B New class of transforming growth factors potentiated by epidermal growth factor: isolation from nonneoplastic tissues Proc. Nat/. Acad. Sci USA 78:5339-5343, 1981. Roberts A B., Frolik C.A , Anzano M.A., Assoian R.K and Sporn MI.: Purlflcation of type 8 transformrng growth factors from non-neoplastrc tissues In: Methods for Preparation of Media, Supplements and Substrata for Serum-Free Animal Cell Culture. Alan R. Liss, New York, 1984, pp. 181-194. Roberts A.B., Anzano M.A., Wakefield L.M., Roche N.S., Stern D F and Sporn M B. Type 8 transforming growth factor: a bifunctronal regulator of cellular growth. Proc. Nat/. Acad. Sci. USA 82: 119- 123, 1985 Seyedin S.M., Thomas T.C., Thompson A.Y., Rosen D M. and Piez K A: Purification and characterization of two cartilage-inducing factors from bovine demineralized bone. Proc. Nat/. Acad Scf USA 82:2267-2271, 1985 Seyedin S M., Thompson A.Y., Bentz H., Rosen D.M., McPherson J M., Conti A., Siegel N.R Galluppi G R and Piez K A. Cartilage inducing

P.R

Elford et al.: TGF-p suppresses

the osteoblastic

phenotype

factor-A. Apparent identity to transforming growth factor-p J. B/o/. Chem. 2815693-5695, 1986 Shipley G D., Childs C.B.. Volkenant M.E. and Moses H L : Differential effects of epidermal growth factor and insulin on DNA and protean synthesis and morphology in serum-free cultures of AKR-2B cells Cancer Res. 44.710-716, 1984 Sporn M.B , Roberts A B., Wakefield L.M. and Assoian R.K Transformtng growth factor-8 biological function and chemical structure. Science 233.532-535, 1986 Tucker R F Shipley G D and Moses H.L Growth inhrbitor from 6X-l cells closely related to platelet type beta transforming growth factor Science 228,705 707, 1984 Young M.E., Robey P G , Reddt A.H Roberts M B. and Sporn M.B TGF-8 expression in fetal bone formtng cells. J. Bone Min. Res. suppl. 1 155A, 1986.

Recewed October 13, 1986 Revised March 22, 1987 Accepted March 23, 1987