CELL REGULATION, Vol. 1, 821-831, October 1990

Specific growth stimulation of cultured smooth muscle cells from spontaneously hypertensive rats by plateletderived growth factor A-chain homodimer

Th6rese J. Resink*t, Timothy Scott-Burden*, Alfred W.A. Hahn*, Marianne Rouge$, Markus Hosang$, Jerry S. Powell+, and Fritz R. Buhler* *Department of Research Basel University Hospital Basel, Switzerland *F. Hoffmann-La Roche Ltd. Basel, Switzerland Cultured vascular smooth muscle cells (VSMC)' from spontaneously hypertensive rats (SHR) possess specific cell surface receptors for both homodimeric forms of platelet-derived growth factor (PDGF-AA and PDGF-BB), in contrast to cells from normotensive Wistar Kyoto (WKY) animals, which express receptors only for the B-chain form of PDGF. Stimulation of quiescent VSMC from SHR with PDGF-AA resulted in activation of S6-kinase and induction of phosphoinositide catabolism, as well as cellular proliferation when cultures were maintained for prolonged periods with daily supplementation of the growth factor. WKY-derived VSMC showed no response to PDGF-AA, which was consistent with their lack of specific receptors for this homodimer. The responsiveness of quiescent cells from SHR and WKY to the B-chain homodimer was similar. The enhanced growth responsiveness of SHR-derived cells to fetal calf serum, as compared with cells from their normotensive counterparts, may be accounted for in part by their expression of receptors for the AA homodimer of PDGF.

Introduction

Inappropriate proliferative processes are the hallmark of the vascular pathologies of arteriosclerosis and essential hypertension (EHT) (Bevan, 1976; Ross, 1981; Owens and Reidy, t Corresponding author. ' Abbreviations: All, angiotensin 11; FCS, fetal calf serum; PDGF, platelet-derived growth factor; PDS, plasma-derived serum; SHR, spontaneously hypertensive rat; VSMC, vascular smooth muscle cell; WKY, Wistar Kyoto rat.

© 1990 by The American Society for Cell Biology

1985; Schwartz et al., 1986; Wissler, 1986). One of the experimental models for investigation of these processes has made comparative use of the genetic strain of spontaneously hypertensive rats (SHR) and their normotensive Wistar Kyoto (WKY) littermates (Okamoto and Aoki, 1963). Several laboratories have shown that vascular smooth muscle cells (VSMC) derived from SHR exhibit enhanced responsiveness to a number of growth factor and hormonal agonists, as well as to various levels of fetal calf serum (Folkow et al., 1973; Yamori et al., 1981; Haudenschild et al., 1985; Hadrava et aL, 1989; Scott-Burden etal., 1989a,b,c,d, 1990). In comparative studies designed to elucidate the underlying causes of this differential responsiveness between SHR- and WKY-derived VSMC, we have assessed the ability of a number of agonists to stimulate S6-kinase activation and phosphoinositide metabolism in quiescent cultures of the different VSMC isolates (Resink et al., 1989a; Scott-Burden et al., 1989a,b,c,d). Of particular interest was our observation that the response of SHR-VSMC to porcine platelet-derived growth factor (PDGFBB) was similar to that of WKY-derived cells (Scott-Burden et al., 1989b,c). In the light of data showing amplified responses of SHR-VSMC to all other agonists tested (Scott-Burden et al., 1988; 1989a,b,c,d; Hadrava et al., 1989; Resink et al., 1989a), the response to PDGF was surprising since this mitogen is considered as the doyen of growth factors for vascular smooth muscle (Ross et al., 1978). Growth inhibition of VSMC by conditioned medium from cultured endothelial cells has been ascribed to the "shedding" of heparin/heparin sulfate (see Scott-Burden and Buhler, 1989 for review). We also demonstrated that SHR-derived cells exhibit a decreased susceptibility to growth inhibition by added exogenous heparinoids as a consequence of their decreased (as compared with WKY cells) number of cell-surface heparin binding-sites (Resink et al., 1989b). This latter finding might imply that the increased in vivo growth of VSMC in hypertension reflects 821

T.J. Resink et al.

a reduced regulation of growth control by endothelial cell-derived heparinoids. However, neither the observation of inhibition of VSMC growth by endothelial cell conditioned medium nor a decreased sensitivity to inhibitory growth control can adequately explain our finding (Scott-Burden, unpublished data) that in coculture studies with VSMC and bovine aortic endothelial cells, and using mitogen-depleted medium, the proliferation of SHR-derived cells, but not WKY-VSMC, was stimulated. Factors relating to positive growth regulation need to be invoked. Endothelial cells in culture synthesize a number of growth factors, including PDGF-BB and PDGF-AA and also secrete significant amounts of PDGF-AA (±0.5 ng.ml-1.24 h-1) into the conditioned medium (Fox and DiCorleto, 1986; DiCorleto and Fox, 1988). Rat arterial smooth muscle cells cultured in serum-free (nonmitogenic) medium, have also been shown to express the gene for PDGF A chain (Naftilan et aL., 1989). Secretion of the growth factor, however, was undetectable in stationary cultures (serum-free), but did occur in the presence of added exogenous PDGF or in proliferating cultures (containing serum) (Sjolund et aL., 1988). Furthermore, stimulation of quiescent VSMC with angiotensin 11 (All) leads to the increased expression of the PDGF A chain gene. We have found that chronic, daily treatment with All of VSMC from SHR animals maintained in mitogen-depleted medium resulted in slow but sustained proliferation, whereas cells from WKY animals were only mildly stimulated under the same experimental conditions (Scott-Burden et aL., 1990). All these observations prompted our investigation of a possible differential responsiveness between SHR- and WKY-derived VSMC to the two homodimeric isoforms of PDGF, namely PDGF-AA and PDGF-BB. Our studies show that only the SHR-derived cells are sensitive to stimulation by PDGF-AA and that their sensitivity is a consequence of their expression of PDGF-a receptors.

Results When quiescent cultures of VSMC from SHR and WKY rats were exposed to a number of growth factors or vasoactive peptides, we observed a significant differential activation of S6kinase (Table 1) between the two cell types for all the agonists tested, with the exception of commercial preparations of human PDGF (predominantly PDGF-AB) and porcine PDGF 822

Table 1. Activation of S6-kinase in quiescent smooth muscle cells S6-kinase (pmol P4O incorporated/ 1 06 cells) Agonist

Dose

SHR-VSMC

PDGF (porcine) PDGF (human)

TGF8

5 ng/ml. 5 ng/ml. 5 ng/ml. 5 ng/ml. 5 ng/ml. 2.5 ng/ml.

1.7 ± 2.0 ± 3.2 ± 3.5 ± 3.3 ± 4.0 ±

A II VP

100 nM 100 nM

3.7 ± 0.5** 3.0 + 0.4*

2.3 ± 0.4 2.2 ± 0.4

FCS

100/0

4.7 ± 0.4

4.8 ± 0.5

IGF, EGF FGFb

0.5 0.4 0.4** 0.4** 0.5** 0.6**

WKY-VSMC 2.3 ± 1.8 ± 2.1 ± 1.9 ± 1.2 ± 1.9 ±

0.4 0.3 0.4 0.4 0.3 0.4

Quiescent VSMC from SHR-WKY rats were exposed to the agonists listed and the levels of S6-kinase activation assessed as described previously (Scott-Burden et al., 1 989a). Agonists were used at the levels indicated in the table and for their optimal activation times as previously determined (Resink et al., 1 989a; Scott-Burden et al., 1 989a,b,c,d). All determinations were performed on at least three different pairs of SHR-WKY cell isolates and the data represent the means ± SD of the values obtained after the subtraction of the levels of activation obtained with serum-free medium alone. Levels of significance (p) for each agonist (SHR vs WKY) were determined using Student's t test for unpaired data; *p < 0.01; **p < 0.001. PDGF, platelet-derived growth factor; IGF1, insulin-like growth factor; EGF, epidermal growth factor; FGF, fibroblast growth factor; TGF, transforming growth factor; A II, angiotensin 11; VP, vasopressin; FCS, fetal calf serum.

(exclusively PDGF-BB). However, when similar

experiments were performed using pure recombinant preparations of either PDGF-AA or PDGF-BB, a difference was observed (Figure 1). Whereas the B chain homodimer stimulated S6kinase to a similar extent in SHR- and WKYVSMC, only SHR-VSMC were activated by PDGF-AA (Figure 1). These differences were also seen after exposure of quiescent, myo[3H]inositol-prelabeled cells to the two homodimeric PDGF isoforms. SHR-VSMC exhibited a stimulated phosphoinositide catabolic response to both mitogens, whereas only the B chain homodimer elicited this response in WKY-derived cells (Figure 2). In dose-response studies for the homodimers, the concentrations of PDGF-BB eliciting a half-maximal increase in inositol phosphates in SHR- and WKY-VSMC were comparable (V0.8 ng/ml, Figure 3B). For PDGFAA the half-maximally effective concentration was -1.5 ng/ml for SHR-VSMC (Figure 3A) while, as shown in Figure 2, the response of WKY-VSMC to this homodimer was negligible. CELL REGULATION

PDGF receptors and responses in SHR-VSMC

1

52 G)

u

(0 -o -0 I--%

0) -

3.

.0-

I

-

0 u c

0a.

-4

@3 0

E0L

FL

1I

0

AII

PDGFAA PDGFBB

Figure 1. Activation of S6-kinase in smooth muscle cells by PDGF A and B chain homodimers. Quiescent VSMC were stimulated with either Angiotensin II (All, 1 nM), PDGFAA or PDGF-BB (both 5 ng/ml) as described previously (Scott-Burden et al., 1 989a) and in Methods. The data shown (mean ± SD, n = 3) represent the levels of incorporation of [32p]P-P4= into S6-polypeptide in response to the activation of cytosolic S6-kinase by the agonists used and were obtained using at least three separate pairs (SHR-WKY) of VSMC isolates (Scott-Burden et al., 1 989a). The levels of incorporation following stimulation with 100% FCS were -4.2-4.7 pmol P04/10 cells for both cell types, and the value for unstimulated cultures indicated by the broken line on the figure. WKY-VSMC, O; SHR-VSMC, B. Quantitatively similar data to that presented in the figure (for dose of 5 ng/ml) were obtained if VSMC were stimulated with either 2 or 10 ng/ml (data not shown).

Stimulation of the above signal transduction mechanisms may lead to the proliferation of cells and this hypothesis was tested in longterm growth experiments (Figure 4). Cells from SHR, maintained in 1 0/ plasma-derived serum (PDS)-medium, exhibited a slow but sustained growth over 10 d in the presence of either PDGF-AA or PDGF-BB. Although WKY-derived cells did not increase in cell number in the presVol. 1, October 1990

ence of the A chain homodimer (Figure 4B), supplementation of medium with PDGF-BB stimulated their proliferation to the same extent as SHR-VSMC. In the absence of added exogenous growth factor, SHR-VSMC gradually increased in cell number over the 10 d in culture (Figure 4A), whereas WKY-VSMC numbers were unchanged (Figure 4B). This finding supports the conclusion that SHR-VSMC were capable of conditioning their medium. Therefore, we tested whether the SHR-VSMC secreted a soluble factor into their medium. In two experiments (Figure 4C) VSMC were maintained for 10 d in medium containing 1% PDS either with or without medium changes every 2 d. For these experiments VSMC were seeded at a lower density than for growth studies presented in Figure 4, A and B, to avoid possible exhaustion of medium nutrients during culture without medium changes. WKY-VSMC did not increase in cell number under either condition. However, the proliferation of SHR-VSMC in 10/o PDS was greater when cultured without regular medium changes, consistent with the accumulation of a secreted growth factor in the conditioned medium. Recently it has been shown that the two PDGF receptor subunits, termed a and ,B, can dimerize to form three distinct high-affinity receptors (aa, ad, and ,3,B) with differing ligand specificity (a-subunit recognizes both, A- and B-chain, whereas the A-subunit recognizes only B-chain) (Seifert et al., 1989). Furthermore, different cell types appear to express different relative and absolute numbers of these subunits and, consequently, respond differently to the three PDGF isoforms (Hosang et al., 1989; Hosang and Rouge, 1989). We therefore examined whether the different responsiveness to PDGFAA between VSMC from SHR and WKY rats might be due to a difference in their complement of surface a-receptor subunits. Saturation binding experiments with [1251]-labeled PDGFAA and PDGF-BB indicate that SHR-derived VSMC express both a- and :3-subunits (Figure 5). In contrast, WKY-VSMC exhibit a comparable complement of f-subunits, but virtually no asubunits (Figure 5), which is similar to that previously observed for smooth muscle cells from umbilical cord vein (Hosang and Rouge, 1989). These findings were confirmed in competition binding studies with isolates of SHR and WKY cells prepared from animals obtained through a number of commercial suppliers (Table 2). It should be noted that although the values for aand fl-receptor given in Table 2 were obtained by extrapolation and at a subsaturating ligand 823

T.J. Resink et al.

400

SHRBB

I

WKYBB

300 0 C -

200

c

0 >

.0co (U

150

0

0-

100 c .40 -

c 0

160

SHRAA

WKYAA

)

C

.In 0 c

I

140 0

120 0

100o

4

0

1

2

5

0

1

7

2

5

Time (min) Figure 2. Kinetics of accumulation of inositol phosphates. Quiescent [3H]-inositol prelabeled-VSMC from SHR and WKY were treated with either PDGF-BB or PDGF-AA for the times shown, and both agonists were used at a final concentration of 5 ng/ml (a saturating concentration for stimulation of phosphoinositide catabolism in VSMC). The tritium content of the inositol phosphates in subsequent cell extracts was determined as described previously (Resink et al., 1 989a; Scott-Burden et al., 1 989a) and in Methods. Inositol monophosphate, inositol bisphosphate, and inositol trisphosphate are represented by A, 0, and *, respectively. Data express the tritium content as a percentage of that in control VSMC (in the absence of agonists) which was arbitrarily taken as 100% for each of the inositol phosphates measured. The data represent mean values from three separate experiments, each performed in triplicate, and the standard deviation both within and between experiments was generally