JOURNAI, OF CELLULAR PHYSIOLOGY 142:236-246 (1990)

Regulation of a-Smooth Muscle Actin and Other Polypeptides in Proliferating and Density-Arrested Vascular Smooth Muscle Cells GENE LIAU,* MOHAMED F. JANAT, AND PETER J. WlRTH Laboratories of Molecular Biology, American Red Cross, lcrorne H . Holland laboratory for Riomedica/ Sciences, Rockville, Maryland 20855 (C.L., M.F.1.); Laboratory of €xperimenta/ Carcinogenesis, National Cancer Institute, Bethesda, Maryland 20892 (P.I.W.i

We have examined a-smooth muscle actin (u-SMactin) protein and mRNA levels in proliferating and density-arrested rabbit vascular smooth niuscle cells (SMC) and also studied overall polypeptide synthesis in these cells by two-dimensional (2-D) gel electrophoresis. (3’ the approximately 1,000 cellular polypeptides rcsolved by 2-D gel analysis, we consistently detected increased expression of 1 2 polypeptides in growth-arrested SMC. Thcse polypeptides, with apparent molecular weights of 24,000 to 55,000 exhibited relative increases of between fourtold to greater than tenfold. Three of these polypeptides were expressed at undetectable levels in proliferating SMC. We also detected 12 secreted polypeptides that were expressed at higher levels in growth-arrested SMC. More changes were associated with the secreted polypeptides, since they rcpresented approximately 4% of the total resolved secreted polypeptides, while only 1 % of the cellular polypeptides were increased in high-density growth-arrested cells. Under these conditions we observed no change in relative u-SM actin protein content a5 determined by 2-D gel analysis and Western blots. This was corroborated by high levels of a-SM actin mRNA levels in both proliferating and high-density growtharrested SMC. These results indicate rabbit vascular SMC maintain a high level of expression of a smooth muscle differentiation marker (a-SM actin) in a proliferation- and density-independent manner. We also examined polypeptide synthesis in SMC isolated by enzymatic digestion ot the aorta vs. cells isolated by the explant method. We found that although overall protein patterns were remarkably similar, several differences were observed. These differences were not due to increased contamination by fibroblasts, since both enzyniatically- and explantderived SMC contained high levels of a-SM actin as determined by immunofluorescence and by Northern analysis.

The migration of smooth muscle cells (SMC) from the media into the intima and their proliferation in the intima is considered a critical step in the development of human atheroma (Ross, 1986; Schwartz et al., 1986). This proliferative response by normally quiescent SMC is characterized by a series of morphological changes including the loss of cytoplasmic myofilaments and a n increase in biosynthetic organelles (Chamley-Campbell et al., 1979). In addition, a-smooth muscle actin (cY-SMactin), the predominant actin isoform expressed in normal SMC, is expressed in very low amounts by SMC in atheromatous plaques and experimental aortic lesions (Gabbiani et al., 1984; Clowes et al., 1988). This decrease is due to a rapid loss of a-SM specific actin mRNA after arterial injury (Clowes et al., 1988).These and other results are consistent with the suggestion that proliferating SMC within lesions are modulated to a “synthetic” phenotype (Chamley-Campbell et al., 1979). Proliferating SMC in culture are morphologically G 1990 WILEY-LISS, INC

similar to proliferating SMC in aortic lesions and have been used as a model for atheromatous SMC (Schwartz et al., 1986; Chamley-Campbell et al., 1979; Gabbiani et al., 1984). Furthermore, cultured rat SMC contain decreased amounts of a-SM actin and increased levels of the p- and y-actin isoforms-a switch in actin isoforms reminiscent of that seen in vivo after vascular injury (Skalli et al., 1986a). Interestingly, growth-arrest of these cells correlates with the increased synthesis of both a-SM actin and smooth muscle myosin heavy chains (Owens et al., 1986; Rovner et al., 1986). This has led to the suggestion that growth-arrest can induce “cytodifferentiation” in cultured SMC (Owens et al., 1986). Growth-arrest of vascular SMC also specifically enhances the expression of apolipoprotein E

Received July 3, 1989; accepted September 5, 1989.

“To whom reprint requestsicorrespondence should be addressed.

POLYPEFTIDE EXPRESSION IN SMOOTH MUSCLE CELLS

and collagen mRNA and their corresponding proteins, a s well as the synthesis of a n as-yet-unidentified secreted polypeptide with an approximate M, of 38,000 (Majack et al., 1988; Liau and Chan, 1989; Stepp e t al., 1986; Millis et al., 1986). In addition, postconfluent SMC specifically synthesize a cell surface-associated heparan sulfate containing glycosaminoglycan, which is a potent inhibitor of SMC proliferation (Fritze et al., 1985). In the present study, we have used two-dimensional (2-D) gel electrophoresis to identify additional polypeptides whose syntheses are positively modulated by growth-arrest of vascular SMC. We consistently detected the increased expression of 1 2 cellular and 12 secreted polypeptides. By contrast, the synthesis of proliferation-related polypeptides was decreased in the quiescent cells. Interestingly, parallel studies indicated that rabbit SMC in culture maintained a high level of a-SM actin mRNA as well as cu-SM actin protein in a proliferation-independent manner. Our results are consistent with other studies that indicate that postconfluent SMC in culture actively modulate their cellular and secretory phenotype (Majack et al., 1988; Fritze et al., 1985; Liau and Chan, 1989). Whether such remodeling represents “cytodifferentiation” is presently unclear.

MATERIALS AND METHODS Materials Collagenase, class I1 (Cooper Biomedical, Malvern, PA); elastase, soybean trypsin inhibitor, and antibody (Sigma, St. Louis, MO); guanidine isothiocynate, cesium chloride, electrophoretic grade acrylamide, and N, N1-methylene bisacylamide (Biological Research Laboratories, Rockville, MD). ampholytes (LKB Instruments, Rockville, MD); 12P-labeled deoxynucleotides and nicktranslating enzymes (Amersham Corp., Arlington Heights, IL); nitrocellulose paper (Schleicher & Schuell, Keene, NH) were obtained from the indicated sources. The actin clone was kindly provided by Dr. Bruce Paterson (National Cancer Institute, Bethesda, MD). Cell culture Vascular SMC were isolated from the thoracic aorta of New Zealand white rabbits as previously described (Liau and Chan, 1989). To isolate explant-derived SMC, the aortic media were minced into 1-2 mm segments and incubated with Medium 199 supplemented with 15%fetal bovine serum. Additional supplemented Medium 199 was added after 5 days, and the media were subsequently changed every 4 days. Extensive outgrowth of SMC was observed after 2 weeks. Cells were routinely passaged a t a 15 ratio. Immunofluorescence experiments using anti-a-SM-1 antibody, which specifically recognizes a-SM actin, were performed as previously described (Liau and Chan, 1989). Flow cytometry analysis of DNA content was performed using propidium iodide (Bartholomew et al., 1976). Approximately 3 x 10‘ cells were harvested by trypsinization. The cells were washed three times with phosphate buffered saline (PBS) containing 0.1% glucose and resuspended in 250 pl of the same buffer. Subsequently, 0.9 ml of ice cold absolute ethanol was added to the tube with vigorous vortexing. The cells were incubated a t

237

4°C overnight, and propidium iodide (2 mgiml) was added to obtain a final concentration of 2 pgiml. Samples were analyzed on a Becton Dickinson FACScan machine. Western analysis Cultured SMC were harvested in a solution containing 200 mM TrisiHCl pH 7.5 containing 25 mM NaC1, 80 mM EDTA, 2 mM phenyl methyl sulfonyl fluoride, and 0.2 U/ml aprotinin; and sonicated on ice for 15 sec. The protein concentration was determined by the method of Bradford (1976). Samples were made to 1x Laemmli sample buffer and electrophoresed according to standard protocols (Laemmli, 1970). The gel was blotted onto nitrocellulose and the filter was incubated overnight with 1%nonfat dry milk in PBS (Towbin et al., 1979). The filter was subsequently washed with PBS and incubated with the a-SM actin monoclonal antibody diluted 1:200 with PBS containing 3% bovine serum albumin. After incubating for 2 h r a t room temperature, the filter was washed three times with PBS for 1 5 min each, and then incubated with anti-mouse IgG conjugated t o horse radish peroxidase diluted 1: 1,000 in PBS with 3% bovine serum albumin for 1h r a t room temperature. The filter was washed a s above and subsequently developed by incubating with 10 mM TrisiHCl pH 7.4 containing 0.025% o-dianisidine and 0.00756 H,Oz. [35SlMethionine labeling of cultured SMC SMC between passage numbers 1and 6 were used for labeling experiments. All samples were analyzed in duplicate. Cells previously seeded in 35 mm culture dishes were incubated in Medium 199 without methionine for 30 min at 37°C in a 5% CO, incubator. For labeling of cellular proteins, 1.5 cc of Medium 199 containing 2% FBS and 150 k Ciiml (>1,000 Cimmole) of [35Slmethionine was added to the dish and incubated for 8 to 14 hr. The cells were rinsed twice in cold PBS, scraped off and pelleted in a clinical centrifuge a t 1,000 rpm. The pellet was washed once with PBS. For analysis of secreted protein, SMC were labeled with 150 p. Ciiml to 250 p Ciiml of [”Slmethionine in 0.2% FBS for 8 hr. The media was collected and centrifuged a t 5,OOOg through a Centricon-10 (Amicon, Danvers, MA) filtration device that had been soaked in PBS containing 0.2% NP 40. After two washes of 1ml water containing 0.2% NP 40, the concentrate was collected and lyophilized. Preparation of samples f o r 2-D gel electrophoresis Cells resuspended in 200 kl of O’Farrell’s lysis buffer A (O’Farrell, 1975) were sonicated on ice for four bursts of 15 sec at 60 sec, intervals using a Kontes Micro Ultrasonic Cell Disrupter a t maximum power. Sonicated cells were transferred to 1.5 ml Eppendorf tubes, saturated (9.5 MI with solid urea, and centrifuged a t 12,OOOg for 10 min. Duplicate 10 pl aliquots of each sample containing 200 pg of bovine serum albumin as carrier were precipitated with cold 10%-trichloroacetic acid (TCA). The TCA precipitates were kept a t 0°C for 60 min, after which they were collected on GF-C filters, washed three times with 2.510 aqueous TCA and three times with 70% aqueous ethanol, dried, and counted in

238

LIAU ET AL

Aquasol (New England Nuclear) to determine the amount of radioactivity incorporated into proteins.

2-D gel electrophoresis 2-D gel electrophoresis was performed as previously described (Wirth et al., 1986). First-dimension isoelectric focusing was performed using a modified O’Farrell 11975) procedure: 21 tube gels (160 x 2 mm) were simultaneously cast in a n apparatus similar to that described by Anderson and Anderson (1978). The ampholytes were pH range 5-8 a t 1.6% and pH range 3.5-10.0 a t 0.4%. Tube gels were prefocused a t constant current (0.5 mA per tube) until the voltage reached 750 V. Protein samples (300,000 dpd20-30 ~1 maximum volume) were loaded at the basic end of the gel with a Hamilton syringe and overlayed with 25 p1 of buffer K. Samples were subjected to electrophoresis at 750 V for 16 h r and finally at 1,000 v for 1h r (13,000 V-hr). The gels were extruded from the tubes into glass scintillation vials and equilibrated by shaking for 30 min in 10 ml of sodium dodecyl sulfate (SDS) equilibration buffer containing 10% glycerol, 5% 2-mercaptoethanol, 2.3% SDS, 0.005% bromophenol blue, and 0.0625 mM Tris-HCl pH 6.8. Second-dimension SDS polyacrylamide gel electrophoresis was modified from the method of O’Farrell (1975) using 0.75 mm 10% polyacrylamidc gels. Tube gels were placed onto the second dimension slab gel with their acidic end orientated t o the left, and held in place with 1% agarose in SDS equilibration buffer. The gels were subjected to electrophoresis with no external cooling a t 15 mA constant current per gel until the tracking dye reached the separating gel; the current was then increased to 20 mA per gel. Gels were then electrophoresed until the tracking dye reached the bottom of the separating gel. Total running time was approximately 4 hr. Immediately after electrophoresis the slab gels were fixed in 10% TCA:l0% acetic acid: 30% methanol for 1hr, and rehydrated by gentle shaking for 1 h r in 2% glycerol. Gels were transferred to Whatman 3 MM filter paper, overlayered with a Mylar sheet (0.005 in thick), dried under low vacuum and heat (6O-8O0C), and exposed to Kodak SB-5-X-ray developing film for 3 t o 14 days at room temperature, and developed. Analysis of percent actin content by 2-D gel electrophoresis was kindly performed by Dr. Gary K. Owens, University of Virginia, as previously described (Owens et al., 1986).

255 (black), corresponding to a n optical density of 0 t o 1.6. In quantitative autoradiography, i t is necessary to measure film exposure or darkening as a result of the amount of radioactivity (decays) of a given protein spot. Therefore, for each set of experimental gels, additional autoradiographs of 30-45 standard chips containing known amounts of radioactivity are exposed, developed, and scanned. Thus, a curve is generated relating absorbency to radioactivity, and which permits conversion of film density of experimental autoradiographs corrected for local background to radioactivity. Gels are segmented using a series of programs in which spots (polypeptides) are segmented from background, and individual spots are quantitated with a Vax 111750 computer. Additional programs are used to automatically match and compare protein spots among a series of autoradiographs of a particular experiment (Miller et al., 1984).

RNA isolation and Northern analysis Total RNA isolation and Northern analysis were performed as previously described (Liau and Chan, 1989).

RESULTS 2-D gel analysis of cellular polypeptides synthesized by proliferating and growth-arrestedSMC We have previously determined that rabbit aortic SMC in culture became confluent a t approximately 1.5 x lo4 cells/cm2 (1,iau and Chan, 1989). Once confluent, they began to divide a t a greatly reduced rate and exhibited the typical hill-and-valley appearance of SMC in culture (Chamely-Campbell et al., 1979). Five days after confluency, the cells were a t a density of approximately 4 x l o 4 cells/cm2, after which no increase in cell number was observed (Liau and Chan, 1989). We also characterized the cell cycle stage of 5 day postconfluent rabbit SMC by flow cytometry analysis of DNA content. We found that greater than 85% of these cells were in the G,iG, state, while the remaining cells were in the G,/M state (results not shown). No cells were detected in the transitional S phase of the cell cycle. This compared favorably with the cells made quiescent by incubating in 0.5% serum containing media for 72 hr-a method we use routinely for [“HIthymidine incorporation assays. Only approximately 70% of the latter cells were in G,/G,, while 10% were in S phase and 2 0 4 were in G21M. This analysis indicates that when SMC reach high density in culture, the maOf 2-D gel autoradiographs Computer jority of these cells are arrested in the Go/G, state. Autoradiographs were analyzed with a semiautoProliferating and 5 day postconfluent SMC were lamated computer analysis system developed for aiding beled with [35S]methionine,and the total cellular polyin digitization, segmentation, quantitation, and image peptides analyzed by 2-D gel electrophoresis a s decomparisons of 2-D gels. The analysis procedures have scribed in Materials and Methods. We found that a been described previously (Miller et al., 1984). Briefly, number of polypeptides that were highly expressed in the analysis system utilizes two hardware devices: a n proliferating SMC were synthesized in lower quantiEikonix flat bed scanning densitometer interfaced ties in growth-arrested SMC (Fig. 1). Some of the more with a Comptal Vision Onei20 image processor (Bed- abundantly expressed polypeptides that decrease in asford, MA) and a Digital Vax 111750 computer. Data sociation with quiescence are indicated by arrows in acquisition is accomplished by scanning back-lighted Figure 1A. By contrast, only a small set of polypeptides gel autoradiographs with the Eikonix scanner, which exhibited increased expression when SMC were consists of a n element photodiode array mounted be- growth-arrested. Some of these polypeptides are indihind a Nikon f2.8Micro lens. The array scans the focal cated by arrows in Figures 1B and 4D. Examination of plane of the lens, and the incident light is digitalized 5 independent SMC strains revealed that the expresinto one of 256 gray levels ranging from 0 (white) to sion of these 12 polypeptides was consistently higher in

239

5.0 I

60 I

7.0 I

-

116 97

-

66

-

45

-

36

-

29

3

0 r rn

-

20

0

-

14

6R

I

I

I

I

I

I

-

C

zm

t)

-

-

I

116 97

-+

66

A

X

0 W

Y

45

36

-

29

-

20

-

14 1

I

I

Fig. 1. 2-D gel analysis of L”S]niethionine labeled cellular polypeptidcs of enzymatically-derived SMC. Preconfluent and 5 day postconfluent SMC are labeled with [”Slmethionine as described in Materials and Methods, and the labeled cellular polypeptides are analyzed by isoelectric focusing in the first dimension and SDS polyacrylamide gel electrophoresis in thc second dimension. A: Cellular polypeptides of preconfluent SMC. Arrows indicate more highly expressed polypeptide as compared to panel B. Circles designate the area corre-

sponding to expressed polypeptides in panel B. Triangles indicate a constitutively expressed polypeptide that is differentially expressed in explant-derived SMC (see Fig. 3). B: Cellular polypeptides of 5 day postconfluent SMC. Arrows indicate mnre highly expressed polypeptides as cnmpared to panel A. Squares indicate region in which a polypeptide is expressed in explant-derived SMC (see Fig. 31. The blocked off rcgion is magnified in Figure 4.

growth-arrested cells (Fig. 1; and results not shown). Approximate fold increase observed for a typical experiment is summarized in Table 1 for SMC isolated by enzymatic digestion. The polypeptides ranged in appar-

ent M, from 24,000 to 55,000 with apparent isoelectric points (PI) of 5.1 to 6.8. Polypeptides of 34 kip1 6.3, 30 k/6.8, and 24 ki6.l (corresponding regions circled in Fig. 1A) were not detected in proliferating SMC. For

240

LIAU ET AL

TABLE 1. Quantitative analysis of cellular polypeptides more highly expressed in growth-arrested SMC'

Polypeptides (M,/pIl 55,00015.60 50,000i5.50 48,000i5.45 46,000i5.10 45,00015.35 37,00015.25-5.45 36,00015.30 34,00015.30 34,00016.30

31,000~5.25-5.30 30,000i6.8 24.000i6.1

Polypeptide spot intensity (5% of total cpm) Proliferating Growth-arrested

SMC

SMC

0.064 0.047 0.072 0.083 0.078 0.215 0.100 0.079 ND 0.094 ND ND

0.285 0.490 0.480 0.430 0.390 2.353 0.816 0.346 0.089 0.335 0.100 0.112

TABLE 2. Percent total actin content' Strains 586

Fold increase 4.4 10.4 6.7 5.5 5.0 10.9 8.2 4.4 NA 3.6 NA NA

' h s u l t s arc obtained from duplicate 2-D gel runs. ND, not delecled: NA, not

applicable.

polypeptides detected in proliferating SMC and increased in high density growth-arrested SMC, there was between a fourfold to greater than tenfold increase in expression (Table 1).As shown in Table 1,computerassisted quantitation of these polypeptides indicated they were prominently expressed after growth-arrest and ranged from approximately 0.3% to 2.3% of the total cpm.

Analysis of a-SM actin in proliferating and growth-arrestedSMC It has previously been reported that a-SM actin levels increase in growth-arrested rat SMC and that this increase may reflect the cytodifferentiation of these cells (Owens et al., 1986).It was, therefore, of interest to examine possible modulation of a-SM actin levels in relationship to the observed change in other polypeptides. The actin gene family consists of at least six closely related species. They are divided into a , p, and y isoforms by their isoelectric points (Vandekerckhove and Weber, 1979).Skeletal muscle, cardiac muscle, and vascular smooth muscle each contain a unique a-actin, while the p and y isoforms are found in practically all cells (Vandekerckhove and Weber, 1979).There is, in addition, a second y species that is unique to smooth muscle. We examined the relative content of a, p, and y isoforms by 2-D gel electrophoresis in three independent cell lines. The results, summarized in Table 2, indicated that density-arrested SMC contained approximately 30 t o 4 0 8 a-isoactin. This relatively high ratio of a-SM actin is similar to, although somewhat higher than, the 23.3% reported for postconfluent r a t SMC (Owens et al., 1986).However, it was surprising to find that preconfluent SMC also contained high levels of a-SM actin (Table 2). It is possible that the a-actin isoform resolved by 2-D gel analysis represents actin isoforms other than the a-SM actin. This can provide one explanation for the high level of a-actin observed in proliferating rabbit SMC. We therefore performed Western blotting analysis using a monoclonal antibody previously demonstrated t o recognize only the a-SM actin (Skalli et al., 1986b). Preconfluent cells, 3 days after plating, and high-density arrested cells were harvested, and a-SM actin analyzed by Western blotting. The results, shown

589 592

Growth condition Preconfluent Postconfluent Preconf Iuent Postconfluent Preconfluent Postconfluent

u-SM (5%) 35 31 34 34 40 43

P-SM

y-SM y-NM"

(R)

1%')

51 53

14 15 18 16 14 14

48 49 46 43

'Isoactin content was determined by densitometric analysis of Coomassie brilliant blue stained 2 - 0 gels. 'y-NM = y nonmusclr actin isoform.

in Figure 2A, indicated the presence of a single immunoreactive band with a n apparent M, of 43,000. We observed no difference in the amount of immunoreactive a-SM actin between proliferating and quiescent SMC (Fig. 2A). Proliferating SMC harvested 1 day to 5 days after low-density plating and quiescent SMC, all contained approximately equivalent amounts of a-SM actin. We also examined the steady-state actin mRNA levels in these cells. The a-actins are encoded by a 1.6 Kb mRNA, while the p and y species are encoded by a 2.1 Kb mRNA (Leavitt et al., 1985).As can be seen in Figure 2B, high levels of the 1.6Kb mRNA band were observed for both proliferating and density-arrested SMC. We have previously demonstrated that total RNA level per cell is essentially unchanged between preconfluent and 5 day postconfluent SMC (Liau and Chan, 1989). These results, taken together, indicate that cultured rabbit SMC retain high levels of a-SM actin mRNA and protein irrespective of their proliferative state.

Comparison of cellular polypeptide expression in SMC isolated by enzymatic digestion vs. explants In vitro experiments with SMC have been performed using SMC isolated either by collagenase and elastase digestion of the aortic media or by placing media sections in culture and isolating SMC outgrowths from these explants (Chamely-Campbell et al., 1979).Although major differences have been demonstrated for primary isolates of the respective SMC, it is unclear whether the two methods of isolation may affect the biological response of the subpassaged cells (ChamleyCampbell et al., 1979;Nabika et al., 1985;Gunther et al., 1982;Skalli et al., 1986a).It is possible that a particular subpopulation of SMC is maintained when SMC from explants are subpassaged. If indeed explant-derived SMC represent a population of SMC with a distinctively altered phenotype, then 2-D gel analysis of total cellular polypeptides should reflect this difference. We examined explant-derived SMC by 2-D gel electrophoresis and found a high degree of similarity in the overall polypeptide expression when compared to enzymatically-derived SMC (compare Figs. 1 and 3; also compared in Fig. 4). However, enzymatically-derived preconfluent SMC expressed several prominent polypeptides with a n apparent M, of 70,000 and PI of 6.2to 6.4 that are only weakly expressed in preconfluent, explant-derived SMC (Figs. 1A and 3A). These polypeptides were barely observed in density-arrested SMC obtained by either methods. In addition, two polypeptides (18 W5.4 and 31 ki6.0)increased only in ex-

POLYPEPTIDE EXPRESSION IN SMOOTH MUSCLE CELLS

241

indicate, in general, that the induction in explant-derived cells was only about half as great.

A 1 2

1 2

28S18S-

Fig. 2. Analysis of a-SM actin polypeptide and mRNA levels in proliferating and density-arrested SMC. A SMC are harvested and total cellular protein electrophoresed on SDS laemmli gel and transferred to nitrocellulose filter. The filter is incubated with monoclonal antibody to a-SM actin and treated as described in Materials and Methods. Lane 1, proliferating preconfluent SMC; lane 2, quiescent 5 day postconfluent SMC. B: Five micrograms of Lotal RNA are elcctrophoresed on a 1%agarose gel, transferred to nitrocellulose, and hybridized with a 1.6 Kh actin cDNA probe. Lane 1, proliferating preconfluent SMC; lane 2, quiescent, 5 day postconfluent SMC.

plant-derived growth-arrested SMC. As illustrated in Figure 1A (triangle), the 18,000 M, polypeptide appeared constitutively expressed in enzymaticallyderived SMC, while the 31,000M, polypeptide (Fig. l B , square) was not seen in proliferating o r growtharrested SMC obtained by enzymatic digestion. Our results indicate that overall cellular polypeptide expression is very similar between SMC isolated by enzymatic digestion and by explants, although a few discrete and consistent polypeptide changes can be observed by 2-Dgel analysis. The 12 cellular polypeptides induced in growth-arrested enzymatically-derived SMC were also induced in growth-arrested, explant-derived SMC (Fig. 2 and in greater detail in the mosaic images in Fig. 4). However, as seen in Figure 4,the magnitude of induction was not a s prominent. a s that observed for enzymatically-derived SMC. This was corroborated by computer-assisted quantitation of the corresponding spots, which

2-D gel analysis of secreted polypeptides synthesized by proliferating and growth-arrested enzymatically-derived SMC, and comparison with SMC obtained from explants We also examined differences in secreted polypeptides between proliferating and growth-arrested SMC. Twelve polypeptides were apparently expressed a t much higher levels in density-arrested SMC when compared to their preconfluent counterparts. Of these, three polypeptides (35ki5.8,35.5 ki6.2,and 35.5 ki6.7) were not observed in proliferating SMC. The location of these polypeptides are indicated by horizontal arrows in Figure 5B and the corresponding region indicated by open circles in Figure 5A.The first of these peptides (35 ki5.8)was not detected in explant-derived SMC, while the other two polypeptides were constitutively expressed a t low levels (results not shown). Other polypeptides that were more highly expressed after growtharrest are indicated by vertical arrows in Figure 5B. Six of these polypeptides (45 ki6.0,33 W5.6, 32.5 k/ 5.6-5.7,33 kl5.85,32 ki6.30 and 6.45,35.7 W6.5)were also increased in explant-derived SMC that are growth-arrested (results not shown). It is interesting to note that the increased expression of the 45 ki6.0polypeptide coincides with the decreased expression of a polypeptide of identical PI with a slightly lower apparent M, (indicated by a horizontal arrow in Fig. 5A). Likewise, concomitant with increased expression of the 32 k polypeptide with a PI of 6.30 and 6.45,there was the loss of a slightly lower M, protein with PI of 6.37. Two other polypeptides whose expression increased in growth-arrested cells (35ki5.3, 34 k15.27)were not observed in explant-derived SMC cultures, while a third polypeptide (34.5 kDai5.23-5.26) was constitutively expressed in explant-derived SMC (results not shown). These are labeled with short vertical arrows in Figure 5B. The M, and PI similarity of these three polypeptides suggest they may differ by changes in glycosylation patterns or other posttranslational modifications. These results suggest that more apparent changes in secreted polypeptides are associated with growth-arrest of SMC than changes in cellular polypeptides, since only about 1% of the observed cellular polypeptides showed increased expression with growth-arrest, while approximately 4% of the resolved secreted polypeptides showed a n increase. In addition, secreted polypeptides appear more sensitive to the method of SMC isolation, since only 6 of the 12 polypeptides that are increased after growth-arrest in enzymatically-derived SMC are similarly highly expressed in explant-derived SMC, while all 12 of the cellular proteins also increased in explant-derived SMC. Comparison of a-SM actin levels in SMC isolated by enzymatic digestion vs. explants The polypeptide differences that were observed between SMC isolated by enzymatic digestion vs. SMC isolated by explants could reflect the selection of a subpopulation of SMC. Alternatively, the differences could reflect contamination of the SMC population by fibroblasts or endothelial cells. Microscopic examination of the cultures indicated no obvious contamination by en-

242

LIAU ET AL.

PI 5.0

6.0

7.0

116 97

66 45

36 29

20 14

116

97 66 45

36 29

20

14

Fig. 3. 2-D gel analysis of [3sSlmethionine labeled cellular polypeptides of explant-derived SMC. Preconfluent and 5 day postconfluent cxplant-derived SMC are labeled with [35S]methionine and analyzed a s described in Materials and Methods. A Cellular polypeptides of preconfluent SMC. B: Cellular polypeptides of 5 day postconfluent

SMC. Arrows indicate more highly expressed polypeptide i n the indicated punel. Circles in pancl A indicate region corresponding to a n expressed polypeptide in panel B. Blocked off region is magnified i n Figure 4.

dothelial cells. However, contaminating fibroblasts cannot be easily distinguished, since they are morphologically similar to cultured SMC. We, therefore, compared the cell strains using the a-SM actin a s a smooth muscle specific marker. We first examined the a-SM actin mRNA levels in

both enzymatically- and explant-derived SMC. Three independent strains of both enzymatically- and explant-derived SMC were found to contain similar levels of the 1.6 kb-mRNA that encodes the a-SM actin (Fig. 6). In addition, the amount of the 1.6 kb-mRNA was equivalent or greater than the 2.1 kb-mRNA that en-

POLYPEPTIDE EXPRESSJON IN SMOOTH MUSCLE CELLS

243

derived SMC express high levels of a-SM actin. HOWever, there was a population of cells that were not stained by the a-SM actin antibody. This population of negatively stained cells did not increase in explantderived SMC and had no apparent growth advantage, since it also did not increase in late-passaged cells. A subpopulation of SMC that does not contain a-SM actin in vivo has been reported (Skalli et al., 1986b). It is possible that the observed a-SM actin negative cells in culture reflect the presence of a-SM negative SMC in the rabbit aorta.

Fig. 4. Mosiac composite of 135Slmethioninelabeled cellular peptides from Figures 1 and 3: Panel A, preconfluent explant-derived SMC; panel B, 5 day postconfluent explant-derived SMC; panel C, preconfluent enzymatically-derived SMC; panel D, 5 day postconfluent enzymatically-derived SMC.

codes the p and y actin isoforms. These results suggest that the a-SM actin gene is highly active in both enzymatically- and explant-derived SMC. The results reflect average a-SM actin gene expression in the total SMC population but do not address whether there is a subpopulation of cells that does not express a-SM actin. We addressed this second question by performing immunofluorescence analysis on the same six cell strains on which we have performed Northern analysis. These results are summarized in Table 3. We examined both early passaged cells (passages 2 to 4) and late passaged cells (passages 10 to 14) and found that between 70% to 90% of the cells examined were highly stained by the monoclonal antibody specific for u-SM actin. More importantly, there was no appreciable difference in the percentage of cells positive for a-SM actin between enzymatically- and explant-derived cells. Finally, we observed no decrease in the percentage of a-SM actin positive cells in late-passaged SMC compared to earlypassaged cells. The immunofluorescence results are consistent with the Northern analysis and indicate that the majority of both enzymatically- and explant-

DISCUSSION The proliferative response of SMC in culture is closely associated with the loss of morphological niarkers for a fully differentiated SMC (Chamley-Campbell et al., 1979). However, the relationship between levels of expression of certain biochemical markers of differentiation such a s LY-SM actin and smooth muscle-specific myosin (SM-specific myosin), and the proliferative state of these cells is more controversial (Owens et al., 1986; Rovner et al., 1986; Hammerle et al., 1988). Desmin-containing SMC are apparently rapidly lost during subpassage (Skalli et al., 1986a). However, aSM actin and SM-specific myosin are maintained at low levels in passaged SMC, and a n increased expression of these proteins is associated with growtharrested rat SMC (Owens et al., 1986; Rovner et al., 1986). This has lead to the hypothesis that growtharrest induces “cytodifferentiation” in cultured SMC (Owens et al., 1986). We have examined general polypeptide changes that are associated with density-dependent growth-arrest of SMC using quantitative 2-D gel electrophoresis. Our analysis indicates enhanced expression of at least 12 cellular polypeptides and 12 secreted polypeptides in postconfluent SMC. Six of these polypeptides are synthesized at undetectable levels in proliferating SMC. Although the identity of these polypeptides is presently unknown, the high level of expression of some of the cellular proteins suggests they may have a structural role. Indeed, it is likely that the prominent cellular polypeptides (M, 31,000 to 37,OOOipT 5.25 to 5.45) are isoforms of tropomyosin (Leavitt et al., 1986). Some of the observed polypeptide changes may be due to altered translational and posttranslational processes. Other may be due t o increased transcription of mRNAs that encode these polypeptides or due to increased stabilization of these mRNAs. Indeed, we have previously reported that both a,(III) collagen mRNA and a1(IV) collagen mRNA levels are substantially increased in high-density-arrested SMC (Liau and Chan, 1989). In addition, Apo-E protein and mRNA levels, as well a s the level of a heparan sulfate containing glycosaminoglycan and a 38,000 M, secreted polypeptide increase in quiescent vascular SMC (Fritze et al., 1985; Majack et al., 1988; Millis et al., 1986). These results, taken together with our 2-D gel results, suggest that cultured SMC substantially remodel their cellular and secretory phenotype in conjunction with high-density growth-arrest. Whether such remodeling reflects “cytodifferentiation” is presently unclear. Nor is it clear what the respective contribution of high cell density vs. growth-arrest is to these changes in polypeptide synthesis. It is likely that both of these factors are important. For example, we

244

LIAU ET AL.

PI

116

97

66

45

36 29

s

0 rrn 20 14

0

C

6n

z

m 116 97 66

0 I 4 X 4

0 w Y

45 36

29

20 14

Fig. 5. 2-D gel analysis of 135Slmethioninelabeled secreted polypeptides of enzymatically-derived SMC. Preconfluent and 5 day postconfluent SMC are labeled with ["'SJmethionine and analyzed as described in Materials and Methods. A: Secreted polypeptides of preconfluent SMC. Arrows indicate more highly expressed polypeptides in the indicated panel. Circles indicate region corresponding to an expressed polypeptide in panel B. B: Secreted polypeptides of 5 day

postconfluent SMC. Horizontal arrows indicate polypeptides that arc not expressed in panel A. Long vertical arrows indicate more highly expressed polypeptides that are also more highly expressed in postconfluent explant-derived SMC. Short vertical arrows indicate polypeptides that are more highly expressed in panel B but are differentially regulated in explant-derived SMC.

have previously determined that a n increase in a,(III) collagen mRNA levels correlated with cellular growtharrest, while a n increase in al(IV) collagen mRNA levels is due to increased cellular density (Liau and Chan, 1989).

The finding that both proliferating and high-density growth-arrested rabbit SMC contain high levels of aSM actin protein and mRN.4 clearly indicates t h a t loss of this smooth muscle differentiation marker is not necessary for the proliferation of cultured SMC. The per-

POLYPEPTIDE EXPRESSION IN SMOOTH MUSCLE CELLS

245

The high level of a-SM actin in preconfluent cells cannot be due to overexpression by a small subpopulation of cells, since by immunofluorescence the majority of the SMC were positive for a-SM actin and these cells were sparsely seeded (1.3 x lo3 cellskm') to allow determination of positive staining by individual cells. Regulation of a-SM actin expression in vascular SMC is complex. In vivo results indicate that progression in G, is closely coupled to a decrease in a-SM actin mRNA levels (Clowes et al., 1988). This decrease likely accounts for the low level of a-SM actin found in atheromatous plaques and experimental aortic lesions (Gabbiani et al., 1984; Clowes et al., 1988). However, progression through the cell cycle is not necessary in vivo and no sufficient in vitro t o induce decreased steadystate a-SM actin message (Clowes e t al., 1988; Corjay et al., 1989). In addition, Corjay et al. (1989) recently reported that high-density-induced increase in a-SM actin protein level is not accompanied by a change in a-SM actin mRNA level. This suggests that the observed increase in a-SM actin protein in postconfluent rat SMC is regulated a t the translational or posttransFig. 6. Analysis of actin mRNA level in enzymatically- and explantlational level. By contrast, Kocher and Gabbiani (1987) derived SMC. Total RNA are isolated from confluent monolayer SMC. reported both a-SM actin mRNA and protein levels inFive kg of RNA are electrophoresed on a 1%agarose gel, transferred to nitrocellulose, and hybridized with a 1.6 Kb actin cDNA probe. creased in postconfluent SMC. We have observed no Lanes 1, 2, and 3 contain RNA from explant-derived SMC strains evidence for a significant transcriptional or posttran101, 114, and 115. respectively. Lanes 4, 5 , and 6 contain RNA from scriptional regulation of a-SM actin in cultural rabbit enzymatically-derived SMC strains 106, 112, and 113. SMC strains SMC. One possibility for our observation is that the 101, 114, and 115 were all harvested at passage 5. Strain 106 was harvested at passage 3, while strains 112 and 113 were harvested a t high level of a-SM actin in preconfluent cells is due to passage 6. the relatively long half-life of this protein (Strauch and Rubenstein, 1984). However, we have examined a-SM actin protein content in proliferating cells for up to 5 TABLE 3 . Immunofluorescent assay for a-SM actin' days in culture and have found no additional decrease. In addition, although our 2-D gel analysis conditions Early passage Late passage only partially resolved the actin isoforms, there is no Enzymatic evidence of a substantial difference in a-SM actin syn93% (10) 106 644 (3) 90% (14) 95% (4) thesis between preconfluent and postconfluent SMC in 112 90% (10) 113 68% (2) these gels. We, therefore, favor the second possibility Explant that rabbit SMC maintain a high basal level of a-SM 93% (10) 101 ND actin in culture and are unable to further increase the 85% (10) 89% (2) 114 ND 115 68% (2) synthesis of a-SM actin after growth-arrest. The lack of cell density dependent modulation of a-SM actin pro'SMC strains 106, 112. and 113 were isolated by enzymatic digestion of the aorta. while SMC strains 101, 114, a n d 115 mere isolated by the explant method. tein level in rabbit SMC when compared to the rat may Numbers represent percentage of cells positively stained hy monoclonal ahbody reflect a species difference. to n-SM actin. SMC were seeded at 1.3 x lo3 celldcm', and approxmakly 100 When polypeptide and gene expression of enzymaticells w e r e counted for each assay. Numbers in parenthesis represent cell passage numbers. ND, not done. cally-derived SMC are compared to explant-derived SMC, several differences are observed. Although overall polypeptide expression is very similar, the exprescentage of a-SM actin (31 to 43%)as a function of total sion of a polypeptide (70 kip1 6.2-6.4) that is highly actin content is consistent with the Northern results, expressed in enzymatically-derived SMC is barely obwhich generally indicate approximately a l:i ratio be- served in explant-derived SMC. In addition, of the 24 tween the 1.6 Kb-mRNA that encodes the a-SM actin, cellular and secreted polypeptides that are expressed at and the 2.1 Kb-actin mRNAs. This is considerably higher levels after density-dependent growth-arrest of higher than the percentage of a-SM actin (12.4%)re- enzymatically-derived SMC, 4 are expressed a t undeported for subconfluent rat SMC. However, it is still tectable levels, while 3 other are constitutively exmuch lower than the 70% a-SM actin reported for SMC pressed in explant-derived SMC. These observations in intact vessels (Owens et al., 1986). The higher level indicate that distinct differences can be detected beof a-SM actin in rabbit SMC is also reflected in a tween SMC strains isolated by the two methods. The higher percentage of a-SM actin immunofluorescent results are consistent with other studies indicating positive cells when compared to those reported for the that certain SMC differentiation markers, such as rat. We find that approximately 80% of the rabbit SMC functional angiotensin I1 receptors and desmin expresare heavily stained by the monoclonal Ab to a-SM ac- sion, are expressed in SMC isolated by enzymatic ditin, while the value reported for fifth-passaged rat aor- gestion of the aorta but rarely expressed or not detected tic SMC is 26% strongly stained, 48% weakly stained, in explant cultures of SMC (Gunther e t al., 1982; and 26%unstained (Table 3; and Skalli et al., 1986b). Griendling et al., 1986; Skalli et al., 1986a). It is pres-

246

LIAU ET AL

ently not clear whether the explant technique selects for a subpopulation of SMC from the aorta or whether these SMC become more modulated during the process of outgrowth from the explants. It is possible that latepassaged enzymatically-derived SMC may exhibit similar changes t o , those observed for explant-derived SMC. Our analysis indicates that only relatively subtle differences exist, since the majority of the polypeptides expressed are quite similar and the level of a-SM actin polypeptide and mRNA are also similar between SMC isolated by the two methods.

ACKNOWLEDGMENTS We would like to thank Dr. G. Owens for analysis of isoactin content and for advice on culturing SMC. The excellent secretarial assistance of Ms. L. Peterson, Ms. S. Young, and Ms. K. Wawzinski is acknowledged. This work was supported in part by Grant R 0 1 HL37510 from the National Institutes of Health. LITERATURE CITED Anderson, N.G., and Anderson, N.L. (1978) Analytical techniques for cell fractions. XXI. Two-dimensional analysis o f serum and tissue proteins: Multiple isoelectric focusing. Anal. Biochem., 85:331-340. Bartholomew, J.C., Yokota, H., and Ross, P. (1976) Effect of serum on the growth of Balh 3T3 A31 mouse fibroblasts and an SV40-transformed derivative. J. Cell. Physiol., 88:277-286. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proetin dye binding. Anal. Biochem., 72348-254. Chamley-Campbell, J., Campbell, G.R., and Ross, K. 119791 The smooth muscle cell in culture. Physiol. Rev., 59t1-61. Clowes, A.W., Clowes, M.M., Kocher, O., Ropraz, P., Chaponnier, C., and Gabbiani, G. (1988) Arterial smooth muscle cells in vivo: Relationship between actin isoform expression and mitogenesis and their modulation by heparin. J. Cell Biol., 107:1939-1945. Corjay, M.H., Thompson, M.M., Lynch, K.R., and Owens, G.K. (1989) Differential effect of platelet-derived growth factor vs. serum-induced growth on smooth muscle a-actin and non-muscle p-actin mRNA expression in cultured rat aortic smooth muscle cells. J. Biol. Chem., 264:10501-10506. Fritze, L.M.S., Rcilly, C.F., and Rosenberg, R.D. (19851 An antiproliferative heparan sulfate species produced by postconfluent smooth muscle cells. J. Cell Biol., 100:1041-1049. Gabbiani, G., Locher, O., Bloom, W.S., Vandekerckhovc, J., and Weber, K. (19841 Actin expression in smooth muscle cells of rat aortic intimal thickening, human atheromatous plaque, and cultured rat aortic media. J. Clin. Invest., 73:148-152. Griendling, K.K., Rittenhouse, S.E., Brock, T.A., Ekstein, L.S., Gimbrone, M.A., and Alexander, R.W. (1986) Sustained diacylglycerol formation from inositol phospholipids in angiotensin 11-stimulated vascular smooth muscle cells. J. Biol. Chem., 261:5901-5906. Gunther, S., Alexander, R.W., Atkinson, W.J., and Gimbrone, M.A. (1982) Functional angiotensin 11 receptor in cultured vascular smooth muscle cells. J. Cell Biol., 923289-298. Hammerle, H., Fingerle, J., Rupp, J., Grunwald, J., Betz, E., and Haudenschild, H.C. (1988) Expression of smooth muscle myosin in relation to growth kinetics of cultured aortic smooth muscle cells. Exp. Cell Res., 178t390-400. Kocher, O., and Gabbiani, G. (1987) Analysis of a-smooth-muscle ac-

tin mKNA expression in rat aortic smooth muscle cells using a specific cUNA probe. Differentiation, 34301-209. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly o f the head bacteriophage T4. Nature, 227:680-685. Leavitt, J., Gunning, P., Kedes, L., and Jariwalla, K. (1985) Smooth muscle wactin is a transformation sensitive marker for mouse NIH 3T3 and Rat-2 cells. Nature, 316t840-842. Leavitt, J., Latter, G., Lutomski, L., Goldstein, D., and Burbeck, S. (1986) Tropomyosin isoform switching in tumorigenic human fibroblasts. Mol. Cell. Biol., 6:2721-2726. Liau, G., and Chan, L.M. (1989) Regulation of extracellular matrix RNA levels in cultured smooth muscle cells: Relationship to cellular quiescence. J. Biol. Chem., 264:10315-10320. Majack, R.A.. Castle, C.K., Goodman, L.V., Weisgraber, K.H., Mahley, K.W., Shooter, E.M., and Gebicke-Haerter, P.J. (1988) Expression of apolipoprotein E by cultured vascular smooth muscle cells is controlled by growth state. J Cell Biol., 107:1207-1213. Miller, M.J., Olson, A.D., and Thorgerisson, S.S.(1984) Computer analysis o f two-dimensional gels: Automatic matching. Electrophoresis, 5~297-303. Millis, A.J.T., Hoyle, M., and Kent, L. (1986) In vitro expression of a 38,000 dalton heparin-binding glycoprotein by morphologically differentiated smooth muscle cells. J. Cell. Physiol., 127:366-372. Nabika, T., Velletri, P.A., Lovenberg, W., and Beaven, M.A. (1985) Increase in cytosolic calcium and phosphoinositide metabolism induced by angiotension I1 and [Arglvasopressin in vascular smooth muscle cells. J. Biol. Chem., 260:4661-4670. OFarrell, P.H. (1975)High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem., 250:4007-4021. Owens, G.K.? Loeb, A,, Gordon, D., and Thompson, M.M. (1986) Expression o f smooth muscle-specific a-isoactin in cultured vascular smooth muscle cells: Relationship between growth and cytodifferentiation. J. Cell Biol., 102r343-362. Ross, R. (1986) The pathogenesis of atherosclerosis-an update. New Engl. J. Med., 314:488-500. Rovner, A.S., Murphy, R.A., and Owens, G.K. (1986) Expression of smooth muscle and non-muscle myosin heavy chains in cultured vascular smooth muscle cells. J. Biol. Chem.: 261:14740-14745. Schwartz, S.M.; Campbell, G.R., and Campbell, J.M. (1986) Replication of smooth muscle cells in vascular disease. Circ. Res., 58:427444. Skalli, O., Bloom, W.S., Ropraz, P., Azarrone, B., and Gabbiani, G. (1986a)Cytoskeletal remodeling o f rat aortic smooth muscle cells in vitro: Relationships to culture conditions and analogies to in vivo situations. J. Submicrosc. Cytol., 3:481-493. Skalli, O., Ropraz, P., Trzeciak, A,, Benzonana, C ., Gillesson, D., and Gabbiani, G. (198613)A monoclonal antibody against a-smooth muscle actin: A new probe for smooth muscle differentiation. J. Cell B i d , 103:2787-2796. Stepp, M.A., Kindy, M.S., Yrazblau, C., and Sonenshein, G.E. (1986) Complex regulation of collagen gene expression in cultured bovine aortic smooth muscle cells. J. Biol. Chem., 261:6542-6547. Strauch, A.R., and Rubenstein, P.A. (1984) A vascular smooth muscle a-isoactin biosynthetic intermediate in BC,,Hl cells. J. Biol. Chem., 2593'224-7229. Towbin, H., Staehelin,T., and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76: 4350-4354. Vandekerckhove, J., and Weber. K. (1979) The complete amino sequence of actins from bovine aorta, bovine heart, bovine fast skeletal muscle, and rabbit slow skeletal muscle. Differentiation, 14: 123-133. Wirth, P.J., Doniger, J., Thorgeirsson, S.S., and DiPaolo, J.A. (1986) Altered polypeptide expression associated with neoplastic transformation of Syrian hamster cells by bisulfite. Cancer Kes., 46:396399.