Regulation of smooth muscle a-actin expression and hypertrophyin cultured mesangial cells

LEAH A. STEPHENSON, LISA B. HANEY, ISA M. HUSSAINI, LARRY R. KARNS, and WILLIAM F. GLASS II

Department of Pathology, Eastern Virginia Medical School, Norfolk, and Department of Pathology, University of Virginia HealthSciences Center, Charlottesville, Virginia, USA

Regulation of smooth muscle a-actin expression and hypertrophyin cultured mesangial cells.

Background. Mesangial cells during embryonic developmentand glomerular disease express smooth muscle a-actin (a-SMA).We were therefore surprised when cultured mesangial cellsdeprived of serum markedly increased expression of a-SMA.Serum-deprived mesangial cells appeared larger than serum-fedmesangial cells. We hypothesized that a-SMA expression may bemore reflective of mesangial cell hypertrophy than hyperplasia.

Methods. Human mesangial cells were cultured in mediumalone or with fetal bovine serum, thrombin, platelet-derivedgrowth factor-BB (PDGF-BB) and/or transforming growth fac-tor-b1 (TGF-b1). a-SMA expression was examined by immuno-fluorescence, Western blot, and Northern blot analysis. Cell sizewas analyzed by forward light scatter flow cytometry.

Results. a-SMA mRNA was at least tenfold more abundantafter three to five days in human mesangial cells plated withoutserum, but b-actin mRNA was unchanged. Serum-deprived cellscontained 5.3-fold more a-SMA after three days and 56-fold moreafter five days by Western blot. Serum deprivation also increaseda-SMA in rat and mouse mesangial cells. The effects of serumdeprivation on a-SMA expression were reversible. Mesangial cellmitogens, thrombin or PDGF-BB, decreased a-SMA, but TGF-b1increased a-SMA expression and slowed mesangial cell prolifer-ation in serum-plus medium. Flow cytometry showed that serumdeprivation or TGF-b1 treatment caused mesangial cell hypertro-phy. PDGF-BB, thrombin, or thrombin receptor-activating pep-tide blocked hypertrophy in response to serum deprivation.

Conclusions. We conclude that increased a-SMA expression inmesangial cells reflects cellular hypertrophy rather than hyperpla-sia.

Smooth muscle a-actin (a-SMA) is expressed in mesan-gial cells in normal development [1–3] and in experimental[4–9] and human [10, 11] glomerular disease, but little orno a-SMA is expressed in normal adult mesangial cells in

vivo. Consequently, expression of a-SMA is considered amarker of mesangial cell activation [7, 10–14]. Hyperplasiaand hypertrophy are two specific mesangial cell responsesthat may result from mesangial cell activation [7]. Somestudies have shown that a-SMA correlates with mesangialcell proliferation [4, 6, 9, 10], but other studies have foundexceptions to this correlation [8, 11, 15, 16].

Serum contains extracellular matrix proteins [17–19],platelet factors [20–28], and other agents derived fromactivation of coagulation (such as, thrombin) that poten-tially regulate mesangial cell function in variety of ways. Infact, products of platelet activation during coagulation areresponsible for the ability of serum to cause cell prolifera-tion [21, 22]. Many of these same factors, either released byplatelets or produced by glomerular cells, are thought to beimportant mediators of disease and responsible for mesan-gial cell activation.

Factors responsible for the expression of a-SMA inglomerular disease are poorly understood. a-SMA expres-sion was associated with increased expression of platelet-derived growth factor (PDGF) and/or PDGF receptor insome investigations [6, 7, 29]. In vitro studies are alsolimited. In one study, growth arrest of rat mesangial cellsfor seven days in serum-free medium did not change thepercentage of a-SMA positive cells [30]. In another study,serum stimulated transcription of a reporter gene fused toa sequence spanning 2894 to 11 of the a-SMA promoterin transfected rat mesangial cells [12], but the effect ofserum directly on a-SMA was not determined. In anotherstudy, the role of the serum response element in regulatinga-SMA gene expression in rat mesangial cell hillocks wasrecently examined [31]. These investigators found that boththe expression of smooth muscle actin and the activity of ab-galactosidase reporter were decreased in hillocks com-pared to surrounding cells. a-SMA mRNA and reportermRNA were also decreased when cells were embedded incollagen gels. All of these studies suggested that serumshould increase a-SMA in routinely cultured mesangialcells.

Recently, we showed that human mesangial cells survive

Key words: human development, cytoskeleton, cell size, microfilaments,flow cytometry.

Received for publication October 17, 1996and in revised form April 10, 1998Accepted for publication May 14, 1998

© 1998 by the International Society of Nephrology

Kidney International, Vol. 54 (1998), pp. 1175–1187

1175

for several weeks in medium without serum or addedgrowth factors [32]. We expected serum deprivation tocause a loss of a-SMA expression. In contrast to ourexpectations, serum deprivation increased a-SMA expres-sion by an order of magnitude. The effects of PDGF-BB,thrombin and transforming growth factor b1 (TGF-b1)were also investigated. PDGF-BB and thrombin are mito-genic in human mesangial cells [33] and, like serum,inhibited a-SMA expression. TGF-b1 is anti-proliferative inmesangial cells [34] and stimulated a-SMA expression. Byflow cytometry, increased cell size accompanied increaseda-SMA expression. These observations suggest that in-creased a-SMA expression in disease may be evidence ofmesangial cell hypertrophy rather than hyperplasia.

METHODS

Materials

Antibody to a-SMA (clone 1A4) was from Sigma Chem-ical Co. (St. Louis, MO, USA). Recombinant PDGF-BB,recombinant TGF-b1 and monoclonal antibody toPDGF-BB were from R&D Systems (Minneapolis, MN,USA). Thrombin was a gift of Dr. J. Fenton (WadsworthCenter Labs, Albany, NY, USA). Complementary DNAprobes used in these studies were a-SMA, graciously givenby Dr. C. Chandra Kumar (Schering-Plough ResearchInstitute, Kenilworth, NJ, USA) and b-actin [35] (ATCC#37997), glyceraldehyde phosphate-3-dehydrogenase(GAPD) [36] (ATCC#57091), and 18S ribosomal RNA[37] (ATCC# 77242) from The American Type CultureCollection (Rockville, MD, USA).

Cell culture

Human mesangial cells were isolated as described [32].The cells were routinely grown in RPMI 1640 with fetalbovine serum (FBS 5:1) without antibiotics or other addi-tives and were used during the first 10 passages. Serum-freecells were prepared by washing three times and resuspend-ing in RPMI 1640 prior to plating. Rat mesangial cells wereisolated by us as described [32] from the kidneys ofSprague-Dawley rats sacrificed for hepatectomy [38] in thelaboratory of Dr. Steven L. Gonias (Department of Pathol-ogy, University of Virginia Health Sciences Center). Ratmesangial cells were positive for Th1.1 by both indirectimmunofluorescence and Western blot. Mouse mesangialcells from non-diabetic NOD mice [39] were a gift of Dr.Oleh Pankewycz (Department of Medicine, University ofVirginia Health Sciences Center).

Typical plating conditions included: (1) in RPMI 1640with 16.7% (5:1 dilution) serum, (2) in RPMI 1640 withoutother additives, but onto serum-coated substrata, or (3) inRPMI 1640 alone onto non-coated substrata. Cells wereplated on serum-coated (condition 2) and non-coated(condition 3) coverslips to compare the effects of differentsubstrata on a-SMA expression. Under both conditions,

cells become fully growth arrested after about one week[32]. On serum-coated substrata, adhesion at focal contactsis mediated via vitronectin receptors, but on non-coatedsubstrata adhesion at focal contacts is mediated by fi-bronectin receptors [32].

RNA preparation and Northern blot

Total RNA was extracted using TRIzol reagent fromGibco BRL (Gaithersburg, MD, USA) and analyzed usingultraviolet spectrophotometry at 260 and 280 nm. TheA260/A280 ratios of all samples used for Northern blotanalysis were between 1.6 and 2.0. RNA samples (10 mg perlane) were denatured with glyoxal and electrophoreticallyseparated in a 1.2% agarose gel [40]. The RNA waselectrophoretically transferred to a nylon membrane andUV-crosslinked. The blots were prehybridized in 50%formamide, 5 3 SSPE, 5 3 Denhardt’s, 1 mg/ml salmontestes DNA, and 0.5% SDS and then hybridized in thesame solution containing 4% dextran sulfate and one ormore labeled cDNA probes. Probes were labeled with[a-32P]d-cytidine triphosphate by random primer labelingusing a commercially available kit from Gibco BRL. Thehybridization temperature was 42°C for all probes used.The blots were washed twice in 5 3 SSPE/0.5% SDS at25°C and then twice in 0.5 3 SSPE/1% SDS at 65°C. A 0.24to 9.5 kb ladder from Gibco BRL was used for molecularsize determinations. The signal was detected by phospor-imaging, and the bands were quantified using ImageQuantsoftware. When necessary, the membrane was hybridizedafter washing again at 100°C in water for 30 minutes.

Immunofluorescence

Immunofluorescence was performed as previously de-scribed [32]. In brief, cells on coverslips were inverted ontodiluted antibodies overnight. The coverslips were subse-quently incubated with biotinylated secondary antibody andthen with Texas Red-avidin and bodipy-phallacidin, to-gether. The coverslips were mounted in Gelvatol [41]. Cellswere photographed using Kodak ecktachrome 400 film withan Olympus BH-2 microscope equipped for epifluores-cence.

Western blot

Western blot analysis was performed as described previ-ously [32]. To summarize, reduced proteins were trans-ferred from gels to PVDF membranes from Millipore(Bedford, MA, USA) essentially as described by Towbin,Staehelin and Gordon [42]. a-SMA on membranes wasstained using alkaline phosphatase-avidin-biotin complex(ABC) according to the manufacturer’s instructions (Vec-tor Laboratories, Inc., Burlingame, CA, USA). Blots weredeveloped with NBT-BCIP, BioRad (Hercules, CA, USA),prepared according to instructions.

Stephenson et al: a-SMA and hypertrophy in mesangial cells1176

Analysis of mesangial cell size by flow cytometry

Mesangial cells were plated in 100-mm dishes in RPMI1640 with serum at 200,000 cells/dish or 106 cells/ml. Cellsplated at the latter density were changed to mediumwithout serum after four hours. Cells in medium withserum received either no addition (1serum) or 10 ng/mlTGF-b1 (1serum, 1TGF-b1). Cells in RPMI 1640 withoutserum received either no addition (-serum), 10 ng/mlTGF-b1 (2serum, 1TGF-b1), 10 ng/ml PDGF-BB (2se-rum, 1PDGF-BB), both PDGF-BB and TGF-b1 (2serum,1PDGF, 1TGF-b1), 10 nM thrombin (2serum, 1throm-bin), or 200 mM thrombin receptor-activating peptide (2se-rum, 1SFLLRN). These agents were added on days 0, 2,and 5 after plating. On day 7 the cells were released withtrypsin and resuspended in 10 ml of RPMI 1640 with FBS.The cells were placed on ice. Aliquots of the cells werecounted using a Coulter Counter. The cells were centri-fuged at 600 3 g at 4°C for five minutes and resuspended ata concentration of 106 cells/ml. The cells were returned toice and then analyzed with a Becton Dickinson Facscanfluorescence-activated cell sorter using forward light scattermode. Results were obtained as plots of number of cellsversus cell size (proportional to cell surface area) inarbitrary units. Cells not used for flow analysis were rinsedtwice in PBS and extracted for Western blot analysis ofa-SMA.

Preparation of images

Images prepared from blots or photomicrographs offluorescence microscopy were printed on a Epson StylusColor 600 printer. Original blots or ektachrome slides werescanned with either a LaCie flatbed scanner or Nikon filmscanner using Adobe Photoshop software. Densitometricmeasurements of Western blot band densities (arbitraryunits) were determined using NIH Image software version1.59 by Wayne Rasband. All images of a particular exper-imental condition, blot or gel were processed identicallyunless stated otherwise.

RESULTS

Expression of a-smooth muscle actin in serum-deprivedmesangial cells

Increased a-SMA expression has been observed to ac-company mesangial cell proliferation in glomerular disease[4, 6, 9, 10]. Based on this observation, we hypothesizedthat serum-deprivation would arrest proliferation andcause loss of a-SMA expression in cultured mesangial cells.Human mesangial cells were plated under three conditionsto test this hypothesis: in medium with serum, and inmedium without serum on either serum-coated or non-coated coverslips. Figure 1 shows the effect of plating cellswith or without serum on a-SMA expression after threedays. The cells were fixed with paraformaldehyde anddouble-labeled with bodipy-phallacidin and a-SMA-spe-

cific antibody (indirect immunofluorescence; Texas Red).Phallacidin binds to filamentous actin (f-actin) of any typebut not to globular actin (g-actin) [43]. Mesangial cells inserum were small and often spindle-shaped, but cells inserum-free medium were broad and polygonal. In serum,the actin filaments were thinner and less abundant (Fig.1A) than in serum-free medium (Fig. 1 C, E). Contrary toour hypothesis, mesangial cells in RPMI 1640 with serumstained very weakly, if at all, for a-SMA by immunofluo-rescence, but mesangial cells in RPMI 1640 alone staineduniversally and intensely. Cells in serum-free mediumexpressed a-SMA regardless of whether they were platedonto serum-coated (Fig. 1D) or non-coated (Fig. 1F) glasscoverslips.

The reversibility of the effects of serum deprivation ona-SMA expression was investigated by both Northern andWestern blotting (Fig. 2). Cells were washed and plated in100 mm culture dishes either in medium with serum orwithout serum. Three days after plating, cells in some of thedishes were released with trypsin, washed, and extracted foranalysis of mRNA and protein. Media were changed on theremaining dishes. Some were continued in the same type ofmedium, either RPMI 1640 with 16.7% FBS (serum-plus)or without FBS (serum-minus), but others were changedfrom either serum-plus to serum-minus or from serum-minus to serum-plus. Two days later the remaining disheswere extracted. Figure 2A shows hybridization with probesfor a-SMA, b-actin, GAPD and 18S ribosomal RNA.Equal amounts of RNA were analyzed. Marked differencesin a-SMA mRNA levels were immediately apparent be-tween cells in medium with or without serum. Band inten-sities were quantified using a phosphor-imager. Relativeband intensities (summarized in Table 1) were normalizedto their values in serum on day 3 (S1). b-actin, GAPD, and18S RNA differed in other lanes by twofold or less, buta-SMA mRNA increased tenfold with serum deprivationfor three days or five days. Two days after changing themedium from serum-minus to serum-plus (S2 . S1), thelevel of a-SMA mRNA decreased from 10.6-fold to only2.2-fold of the level in S1 cells on day 3. Therefore, theeffects on a-SMA expression of adding or removing serumcould be reversed.

One dish each of cells treated as in Figure 2 wasextracted for analysis of a-SMA by Western blot (Fig. 2B).Serum-plus medium decreased a-SMA protein levels as itdid mRNA levels. After three days, the a-SMA banddensity in serum-minus cells (S2) was 5.3-fold its density inserum-plus cells (S1). The level of a-SMA protein contin-ued to decline in serum-plus medium, and after five days(S1 . S1), was only one fifth of its level after three days(S1). The band density of a-SMA from cells serum-minusfor five days (S2 . S2) was 56-fold that of cells serum-plusfor five days (S1 . S1). In 16 of 16 experiments, cells inmedium with serum for one to three days containedmarkedly less a-SMA than cells in medium without serum.

Stephenson et al: a-SMA and hypertrophy in mesangial cells 1177

Expression of a-smooth muscle actin in serum-deprivedrat and mouse mesangial cells

Many published studies of a-SMA expression in diseasehave involved animal models, primarily rats. Figure 3compares expression of a-SMA by rat and mouse mesan-gial cells treated as in Figure 2 in serum-plus and serum-minus medium. Interestingly, expression of a-SMA in thesespecies in the presence of serum greatly exceeded theexpression of a-SMA in human mesangial cells in serum.Nonetheless, expression of a-SMA was less in the presenceof serum than in serum-free medium. Similar results wereobtained by immunofluorescence (not shown).

Effect of platelet derived growth factor-BB and thrombin

The abilities of two serum constituents, PDGF-BB andthrombin, to affect a-SMA expression in human mesangialcells were determined. Figure 4 shows that two days afteradding thrombin the band density of a-SMA by Westernblot was about 50% of control, and after adding recombi-nant human PDGF-BB, the level was about 40% of control.PDGF-BB decreased a-SMA expression by Western blot in8 of 8 experiments, and thrombin decreased it in 6 of 7experiments. The effects of these agents on a-SMA mRNAlevels were also determined. RNA was extracted at 0, 4, 8,and 24 hours after the addition of either thrombin or

Fig. 1. Fluorescence staining of total F-actin and a-smooth muscle actin (a-SMA) in human mesangial cells. Human mesangial cells were washed andplated (40,000 cells/well) in RPMI 1640 with (A and B) and without (C through F) 16.7% serum onto non-coated (A, B, E, F) and serum-coated glasscoverslips (C, D). Three days after plating, the cells were fixed and double-labeled with bodipy-phallacidin (A, C, E) for total f-actin and by indirectimmunofluorescence (Texas Red) for a-SMA (B, D, F). Bar 5 116 mm.

Stephenson et al: a-SMA and hypertrophy in mesangial cells1178

PDGF-BB and analyzed by Northern blot (Fig. 5). Bandintensities were quantified using a phosphor-imager. Rela-tive band intensities (summarized in Table 2) were normal-ized to the values at time 0. The effects of thrombin andPDGF-BB on a-SMA mRNA followed a similar timecourse. The a-SMA mRNA level declined slightly in the

first eight hours, and after 24 hours was about one-fifth ofcontrol (0 hr) in both cases.

Effect of transforming growth factor-b1

Transforming growth factor-b has been reported tostimulate a-SMA expression in smooth muscle cells [44]

Fig. 2. Effect of serum or a-SMA mRNA inhuman mesangial cells by Northern andWestern blots. Cells were plated either inmedium with serum (S1) at 200,000 cells/100mm dish or without serum in non-coated dishes(S2) at 106 cells/100 mm dish. Three days afterplating RNA (5 dishes) and protein (one dish)were extracted from some dishes in each group(S1, S2). The medium was changed on theremaining dishes. Six dishes for each conditionwere continued in the same medium, eitherwith serum (S1 . S1) or without serum (S2. S2). Six dishes for each were changed frommedium with serum to medium without serum(S1 . S2) or from medium without serum tomedium with serum (S2 . S1). On day 5,RNA and protein were extracted from theremaining dishes. The RNA (10 mg/lane) wasanalyzed by Northern blot with cDNA probesfor a-SMA and b-actin, and then stripped andreprobed for GAPD, followed by 18S ribosomalRNA (A). Quantified results are summarized inTable 1. One dish each of cells treated asdescribed above was extracted and analyzed byWestern blot for a-SMA (B). Band densities inarbitrary units determined using NIH ImageSoftware are shown below each a-SMA band.

Table 1. Relative levels of a-smooth muscle actin (a-SMA) mRNA in mesangial cells in serum-plus (S1) and serum-minus (S2) mediumdetermined by Northern blot (Fig. 2) normalized to their values in serum-plus cells on day 3

S1 S1 . S1 S1 . S2 S2 S2 . S2 S2 . S1Day 0 serum 1 serum 1 serum 1 serum 2 serum 2 serum 2Day 3 extract serum 1 serum 2 extract serum 2 serum 1Day 5 extract extract extract extract

b-actin 1.0 1.0 0.7 0.6 0.5 0.9a-SMA 1.0 1.3 7.0 10.6 10.7 2.3GAPD 1.0 1.5 0.8 0.6 0.5 1.218S 1.0 1.0 0.9 0.9 1.0 1.0

Stephenson et al: a-SMA and hypertrophy in mesangial cells 1179

and fibroblasts [45]. The hypothesis that TGF-b1 increasesa-SMA expression in human mesangial cells was tested.Cells in either serum-plus or serum-minus medium forthree days were treated with 10 ng/ml TGF-b1 for two days.Next the cells were extracted and analyzed by Northern blotfor b-actin, a-SMA, GAPD, and 18S ribosomal RNA (Fig.6A). Analysis by phosphor imaging, summarized in Table 3,revealed that TGF-b1 caused about a threefold (range 2.4-to 4.0-fold) increase in the a-SMA mRNA level in thepresence of serum, but had little or no effect on the a-SMAmRNA level in serum-minus cells. Although the initialintent had been to normalize the data to GAPD or 18SRNA, the levels of these RNA species were noted todecrease with serum deprivation to varying degrees indifferent experiments. In this experiment, with samplesprepared in triplicate, serum-minus cells contained aboutone-fifth as much GAPD mRNA and about one-third asmuch 18S RNA as serum-plus cells. The cause for theobserved variability in the effects of serum deprivation onGAPD and 18S RNA is not known. However, within eitherserum-plus or serum-minus conditions, none of the agentsstudied (thrombin, PDGF, and TGF-b1) appeared to affectGAPD or 18S levels. Nonetheless, in response to serumdeprivation, these RNA species decrease while a-SMAmRNA increases. Consequently, normalization of the datato either of theses species would indicate even greaterincreases (about 30- to 60-fold) in a-SMA mRNA levelsdue to serum deprivation.

Extracts of cells from the same experiment were ana-lyzed by Western blot for a-SMA (Fig. 6B). a-SMAprotein, like its mRNA, increased in cells treated withTGF-b1 in medium with serum, about two-fold, but did notincrease in cells treated in medium without serum.

Regulation of cell size and a-smooth muscle actinexpression

Transforming growth factor-b stimulates a-SMA expres-sion in smooth muscle cells [44] and fibroblasts [45], andcauses hypertrophy of rat mesangial cells [46]. Humanmesangial cells in RPMI 1640 alone appeared larger thanmesangial cells in serum (Fig. 1). These observationssuggested that the serum-free mesangial cells underwenthypertrophy. The possibility that agents that alter mesan-gial cell a-SMA content also alter cell size was investigatedby analyzing mesangial cells with a Becton DickinsonFacscan fluorescence-activated cell sorter using forwardlight scatter mode. The cells were treated three times overthe course of a week with fresh medium containing eitherno additional agent or 10 ng/ml TGF-b1, 10 ng/ml PDGF-BB, 10 nM thrombin or 200 mM SFLLRN peptide, alone orin combination. At the end of one week, the cells wereharvested by trypsinization and analyzed. Figure 7 showsgraphically the distribution of mesangial cell sizes under

Fig. 3. Effect of serum on a-SMA protein in rat and mouse mesangialcells by Western blot. Rat and mouse mesangial cells were cultured andthe cells were treated as described in Figure 2. Lane numbers and legendsare the same as Figure 2.

Fig. 4. Western blot analysis of the effects of thrombin and PDGF-BB ona-SMA. Washed cells (200,000 cells/well of 6-well plates) were plated inRPMI 1640 without serum. Five days after plating, the cells were changedto fresh RPMI 1640 alone (control) or containing either 5 nM thrombin or10 ng/ml PDGF-BB. After seven days the cells were extracted in electro-phoresis sample buffer and analyzed by Western blot (2 mg/lane). Extractsfrom three wells per condition are shown.

Stephenson et al: a-SMA and hypertrophy in mesangial cells1180

each condition. Cells from the same experiment wereextracted and analyzed by Western blot for a-SMA (Fig. 8).Regardless of the agent added, cells in serum-minus me-dium contained more a-SMA than cells in serum-plusmedium. To avoid saturated staining (as in Fig. 6), lessprotein extracted from cells treated under serum-minusconditions (0.2 mg, panels/lanes C-H) than serum-plusconditions (1 mg, panels/lanes A, B) was analyzed byWestern blot. In each case, cellular enlargement was ac-companied by increased a-SMA expression. Cells platedwithout serum (panel/lane C) were larger and containedmore a-SMA than cells plated with serum (panel/lane A).Cells treated with TGF-b1 in serum-plus medium (panel/lane B) were larger and contained more a-SMA than cellsin serum-plus medium alone (panel/lane A). The additionof TGF-b1 to serum-minus cells (panel/lane D) had noeffect on either cell size or a-SMA content compared toserum-minus cells alone (panel/lane C). PDGF-BB pre-vented enlargement of mesangial cells in serum free me-dium (Fig. 7E) and antagonized a-SMA expression (Fig. 8,lane E). Cells treated with both PDGF-BB and TGF-b1(panel/lane F) contained an amount of a-SMA intermedi-ate between that of cells treated with either PDGF-BB(panel/lane E) or TGF-b1 (panel/lane D) alone. Thrombin(panel/lane G), like PDGF-BB, reduced mesangial cell sizeand a-SMA content compared to serum-minus medium

alone (panel/lane C). Thrombin receptor-activating peptide(SFLLRN) also decreased the size and actin content ofmesangial cells (panel/lane H), but to a lesser extent thanthrombin.

The effect of TGF-b1 and PDGF-BB in combination wasfurther investigated. Three separate experiments weredone following the same experimental protocol as above,but varying the concentration of PDGF-BB. Both forwardlight scattering and a-SMA expression were analyzed. Theresults of each experiment were normalized to the valuesfor serum-minus cells (defined as 100%) and ranged from80% to 127%. Figure 9 shows average results from thethree experiments. Although absolute values varied be-tween experiments, the trends shown were present in eachexperiment. Cells in serum-plus medium were smallest(88%) and contained the least amount of a-SMA. Cellstreated with TGF-b1 in the presence of serum were thelargest, and on average, contained four times as mucha-SMA. Cells in serum-minus medium contained the mosta-SMA, about 20 times as much as serum-plus cells; theaddition of TGF-b1 to serum-minus cells had little effect oncell size or a-SMA expression. Addition of PDGF-BB toserum-minus medium caused a dose-dependent decrease incell size and a-SMA expression that was offset by additionof TGF-b1. The effect of TGF-b1 appeared greater at lowerPDGF-BB concentrations.

DISCUSSION

Our studies show that expression of a-SMA mRNA andprotein in cultured mesangial cells correlates better withhypertrophy rather than proliferation. Human mesangialcell mitogens (serum, PDGF-BB [33] and thrombin [33])inhibited a-SMA expression and antagonized hypertropy.Conditions that inhibited proliferation (TGF-b1 [34] orserum-free medium) increased a-SMA expression andcaused hypertrophy.

Fig. 5. Expression of a-SMA mRNA inmesangial cells in serum-free mediumcontaining thrombin or PDGF. Washed cells(106 cell/100 mm dish) were plated withoutserum. Three days after plating, the cells weretreated with either 10 nM thrombin or 10 ng/mlPDGF-BB. RNA was extracted from the cellsafter 0 hours (0), four hours (4), eight hours(8), and 24 hours (24). The RNA (10 mg/lane)was analyzed by Northern blot with cDNAprobes for a-SMA and b-actin, and thenstripped and reprobed for GAPD, followed by18S ribosomal RNA. Quantified results aresummarized in Table 2.

Table 2. Relative levels of a-SMA mRNA in mesangial cells treatedwith thrombin or PDGF-BB determined by Northern blot (Fig. 5)

normalized to their values at time 0

Time

Thrombin 10 nM PDGF-BB 10 ng/ml

0 hr 4 hr 8 hr 24 hr 0 hr 4 hr 8 hr 24 hr

b-actin 1.0 0.8 1.6 0.5 1.0 1.2 1.1 0.8a-SMA 1.0 0.9 0.6 0.2 1.0 0.9 0.3 0.2GAPD 1.0 1.0 0.7 0.7 1.0 0.9 0.4 0.518S 1.0 1.0 1.1 1.2 1.0 0.9 0.8 0.9

Stephenson et al: a-SMA and hypertrophy in mesangial cells 1181

The finding that serum deprivation increases a-SMAprotein and mRNA levels was surprising given the resultsof studies using reporter constructs that contain a-SMApromoter regions [12, 31]. The rat promoter contains twoCC(A/T)6GG sequences (CArG box elements) responsiblefor binding of serum response factor. Thus, one wouldexpect that a-SMA mRNA transcription would be in-creased in medium containing serum compared to medium

without serum. Nevertheless, the effect of serum depriva-tion on a-SMA mRNA and protein is not described ineither of these papers. One possible explanation for ourunexpected finding is that although serum increasesa-SMA transcription, it increases a-SMA mRNA degrada-tion to a greater extent. Thus, serum deprivation mayincrease a-SMA mRNA and protein levels even though itdecreases a-SMA transcription by a compensatory increasein a-SMA message stability. This is currently under inves-tigation.

Conversion of plasma to serum endows it with its mito-genic properties [21, 22]. The factors are probably diverse;however, PDGF isoforms have probably been the beststudied. PDGF activity in serum varies widely betweenspecies. In one study, human serum contained 13.3 6 4.5ng/ml total PDGF activity (6.25 6 2.2% BB isoform) whileFBS contained 0.93 6 0.3 ng/ml (82 6 12% BB isoform)[47]. The mitogenic potency of human and bovine sera weresimilar despite these differences, suggesting that othersubstances were as important as PDGF in giving serum its

Fig. 6. Effect of TGF-B1 on a-SMA in human mesangial cells by Northern and Western blots. Mesangial cells were plated at a density of either400,000/100 mm dish (6 dishes) in RPMI 1640 plus serum or at 2 3 106/100 mm dish (6 dishes) in RPMI 1640 alone. After three days, the cells weretreated with fresh medium (serum-minus or serum-plus) with or without 10 ng/ml TGF-B1 for two days. The cells were detached with trypsin and cellsfrom duplicate dishes were pooled and extracted, giving triplicate samples for each condition. The RNA (10 mg/lane) was analyzed by Northern blotwith cDNA probes for a-SMA and b-actin, and then stripped and reprobed for GAPD, followed by 18S ribosomal RNA. Quantified results aresummarized in Table 3. An aliquot of cells from each sample was extracted in electrophoresis sample buffer and 5 mg protein/lane was analyzed byWestern blot (B).

Table 3. Relative levels of a-SMA mRNA in mesangial cells treatedwith TGF-b1 determined by Northern blot (Fig. 6) in triplicate

normalized to their average values in serum-plus medium without TGF-b1

TGF-b1

Serum-plus Serum-minus

2 2 2 1 1 1 2 2 2 1 1 1

b-actin 0.9 0.8 1.2 1.0 0.9 1.2 0.8 0.5 0.8 0.7 0.9 0.9aSMA 1.1 0.7 1.2 2.9 2.4 4.0 12.8 7.9 9.4 9.9 13.6 12.4GAPD 1.0 0.8 1.1 0.7 0.6 0.9 0.2 0.1 0.2 0.2 0.3 0.218S 1.1 0.9 1.0 0.8 0.7 0.9 0.4 0.3 0.4 0.4 0.5 0.6

Stephenson et al: a-SMA and hypertrophy in mesangial cells1182

Fig. 7. Forward light scatter flow cytometry of human mesangial cells treated with TGFb1, PDGF-BB, thrombin, and/or SFLLRN. Cells were in RPMI1640 with (A and B) or without serum (C through H) containing no addition (A), 10 ng/ml TGFb1, (B, D, and F), 10 ng/ml PDGF-BB (E and F), 10nM thrombin (G) and/or 200 mM SFLLRN-peptide (H). Media containing the appropriate agent(s) were changed twice and the cells were released withtrypsin and analyzed after one week. The vertical line provides a reference for comparison to the position of the peak in cells grown in medium withserum (A). The number of cells counted (y-axis; 15 cells/division) is plotted against cell size (x-axis; arbitrary units). The median cell size (arbitrary units)is noted for each condition.

mitogenic properties. These unidentified substances couldalso affect a-SMA expression and mesangial cell size.

Thrombin is present in serum. Whether thrombin con-tributes significantly to the mitogenic activity of serum isunclear because of the abundance of protease inhibitors inserum. However, the active concentration of thrombin infresh 10% FBS was 10 ng/ml (0.3 nM) and after three daysat 37°C, it was 28 ng/ml (0.8 nM) in one investigation [48].Our experiments were conducted with 5 to 10 nM thrombin.PDGF-BB and/or thrombin may contribute to inhibition ofa-SMA expression by serum, but inhibition of a-SMAexpression by serum is probably the sum of diverse compo-nents.

Thrombin causes mesangial cells to secrete PDGF [33],and thus, inhibition of a-SMA expression by thrombinmight have been the result of autocrine stimulation byPDGF. However, neutralizing-antibody to PDGF-BB (alsoPDGF-AB) did not block the effect of thrombin on a-SMAexpression at concentrations in excess of that necessary toblock 10 ng/ml PDGF-BB (not shown). Hence, the effect ofthrombin on a-SMA expression seems to be independentof PDGF-BB secretion.

If a similar relationship between hypertrophy anda-SMA expression in mesangial cells exists in vivo, thenexpression of a-SMA in conditions where there is littleproliferation can be explained. For example, angiotensin IIcaused increased expression of a-SMA in the mesangialcells of treated rats without causing mesangial cellproliferation [8]. Increased expression of a-SMA hasbeen observed in human [11] and experimental diabeticglomerular injury [49]. The authors of the latter study[49] suggested that increased expression of contractileproteins in such diseases might serve as a marker for

glomerular hypertrophy. Thus, expression of a-SMAunder these in vivo conditions may indicate mesangialcell hypertrophy.

Cells may undergo hyperplasia and hypertrophy simulta-neously. TGF-b1 caused mesangial cell hypertrophy andincreased a-SMA expression in serum-plus medium with-out fully antagonizing the hyperplastic response of the cellsto serum (Fig. 8). After one week, the dish with serum-plusmedium alone contained 16-times as many cells as plated,while the dish with serum-plus medium and 10 ng/mlTGF-b1 contained 10 times as many cells as plated. There-fore, cells in serum-plus medium with TGF-b1 underwenthyperplasia (albeit at a slower rate) and hypertrophysimultaneously. Increased a-SMA expression in prolifera-tive diseases may reflect concurrent hypertrophy and hy-perplasia in vivo. TGF-b production in diseases could beresponsible. TGF-b is known to stimulate a-SMA synthesisin smooth muscle cells [44] and fibroblasts [45], and itcauses hypertrophy of rat mesangial cells [46].

Cells in serum-plus medium alone were smallest, butcells in serum-plus medium with TGF-b1 were largest (evenlarger than serum-minus cells with TGF-b1). This suggeststhat serum contains a factor or factors that have a syner-gistic effect on TGF-b1-induced hypertrophy. TGF-b1 in-creased expression of a-SMA in cells in serum-plus me-dium but not serum-minus medium. However, the lattercontained more a-SMA, with or without TGF-b1, than cellsin serum-plus medium with TGF-b1. These differences inthe response to TGF-b1 of cells in medium with or withoutserum may be due to the presence of a complex mixture offactors in serum, including some that inhibit a-SMA ex-pression. The observation that no clear response to TGF-b1was observed in medium without serum could be due to

Fig. 8. Western blot analysis of a-SMA inhuman mesangial cells treated with TGF-b1,PDGF-BB, thrombin, and/or SFLLRN. Cellsfrom the experiment described in Figure 7 werewashed and extracted in electrophoresis samplebuffer. More sample was analyzed for cells inserum-plus medium (1 mg), than for cells inserum-minus medium (0.2 mg) in order to allowa more accurate relative quantitation of a-SMA. Values below bands show the results ofdensitometric analysis using the NIH ImageSoftware. Values in parentheses have beencorrected for differences in the total proteinanalyzed.

Stephenson et al: a-SMA and hypertrophy in mesangial cells1184

absence of serum factors that enhance the responsive-ness of cells to TGF-b1. From published observations[47], the concentration of PDGF-BB in 16.7% serum canbe estimated at 1.3 ng/ml. Although the cells were notresponsive to TGF-b1 when serum deprived, TGF-b1nonetheless antagonized the effect of PDGF-BB (Figs. 7,8 and 9). This suggests that PDGF-BB may partiallyrestore TGF-b responsiveness to the serum-deprivedmesangial cells, perhaps through affecting TGF-b1 re-ceptor expression. Clearly, diverse substances in serummay alone or in combination be responsible for the effectof TGF-b1 and serum together on cell size and a-SMAexpression.

Increased expression of a-SMA has been observed inother cell types undergoing hypertrophy. Hypertrophy ofvascular smooth muscle cells treated in culture with angio-tensin II or arginine vasopressin is accompanied by aselective increase in a-SMA expression [50]. During car-diac hypertrophy in response to increased load caused byaortic coarctation there is reactivation of a-SMA expres-sion [51]. Hence, increased a-SMA expression may be ageneral manifestation of hypertrophy in cells of mesenchy-mal origin.

In summary, human mesangial cells deprived of serumenlarge and develop dense bundles of stress fibers rich ina-SMA. This process, hypertrophy, was antagonized revers-ibly by serum, thrombin, and PDGF-BB. TGF-b1 causedhypertrophy and increased a-SMA expression in the pres-ence of serum. Diverse factors probably act on mesangialcells cultured in the presence of serum or in vivo in thesetting of inflammation or sclerosis to regulate the level ofa-SMA expression, but enhanced expression of a-SMA isprimarily reflective of conditions favoring hypertrophyrather than hyperplasia.

ACKNOWLEDGMENTS

This work was supported by a NIH Grant DK-46735 to W.F. Glass II.The authors express their appreciation to Catharine S. Fritschi, Depart-ment of Pathology, Sentara Norfolk General Hospital, Norfolk, Virginiafor assistance with flow cytometry.

Reprint requests to William F. Glass II, Department of Pathology, EasternVirginia Medical School, P. O. Box 1980, 700 Olney Road, Norfolk, Virginia23501-1980, USA.E-mail: wfg@evms.edu.

APPENDIX

Abbreviations used in this article are: a-SMA, a-smooth muscle actin;f-actin, filamentous actin; FBS, fetal bovine serum; g-actin, globular actin;GAPD, glyceraldehyde phosphate-3-dehydrogenase; PDGF, platelet de-rived growth factor; S2, serum-minus cells; S1, serum-plus cells;SFLLRN, thrombin receptor-activating peptide; TGF-b1, transforminggrowth factor-b1.

Fig. 9. Forward light scatter flow cytometry and a-SMA expression inmesangial cells treated with TGF-b1 and various concentrations ofPDGF-BB. Cells were treated with 10 ng/ml TGF-b1 in RPMI 1640 with orwithout 16.7% FBS (serum-plus and serum-minus, respectively) or inRPMI 1640 containing 1, 2, 4, or 10 ng/ml PDGF-BB. The same protocolwas followed as in Figure 8. The results are the averages of threeexperiments. Median forward light scattering and a-SMA band densitywere normalized to the values of serum-minus cells (100%) beforeaveraging. Symbols are: (j) without TGF-b1; (u) with TGF-b1.

Stephenson et al: a-SMA and hypertrophy in mesangial cells 1185

REFERENCES

1. HOLTHOFER H, SAINIO K, MIETTINEN A: The glomerular mesangium:Studies of its developmental origin and markers in vivo and in vitro.APMIS 103:354–366, 1995

2. ALPERS CE, SEIFERT RA, HUDKINS KL, JOHNSON RJ, BOWEN-POPEDF: Developmental patterns of PDGF B-chain, PDGF-receptor, andalpha-actin expression in human glomerulogenesis. Kidney Int 42:390–399, 1992

3. CAREY AV, CAREY RM, GOMEZ RA: Expression of alpha-smoothmuscle actin in the developing kidney vasculature. Hypertension19:II168–II175, 1992

4. ROY-CHAUDHURY P, WU B, MCDONALD S, HAITES NE, SIMPSON JG,POWER DA: Phenotypic analysis of the glomerular and periglomerularmononuclear cell infiltrates in the Thy 1.1 model of glomerulonephri-tis. Lab Invest 72:524–531, 1995

5. ROMANO LA, FERDER L, INSERRA F, ERCOLE L, GOMEZ RA: Intra-glomerular expression of alpha-smooth muscle actin in aging mice.Hypertension 23:889–893, 1994

6. FLOEGE J, BURNS MW, ALPERS CE, YOSHIMURA A, PRITZL P,GORDON K, SEIFERT RA, BOWEN-POPE DF, COUSER WG, JOHNSONRJ: Glomerular cell proliferation and PDGF expression precedeglomerulosclerosis in the remnant kidney model. Kidney Int 41:297–309, 1992

7. FLOEGE J, JOHNSON RJ, COUSER WG: Mesangial cells in the patho-genesis of progressive glomerular disease in animal models. (Review)Clin Invest 70:857–864, 1992

8. JOHNSON RJ, ALPERS CE, YOSHIMURA A, LOMBARDI D, PRITZL P,FLOEGE J, SCHWARTZ SM: Renal injury from angiotensin II-mediatedhypertension. Hypertension 19:464–474, 1992

9. JOHNSON RJ, IIDA H, ALPERS CE, MAJESKY MW, SCHWARTZ SM,PRITZI P, GORDON K, GOWN AM: Expression of smooth muscle cellphenotype by rat mesangial cells in immune complex nephritis.Alpha-smooth muscle actin is a marker of mesangial cell proliferation.J Clin Invest 87:847–858, 1991

10. MACPHERSON BR, LESLIE KO, LIZASO KV, SCHWARZ JE: Contractilecells of the kidney in primary glomerular disorders: An immunohis-tochemical study using an anti-alpha-smooth muscle actin monoclonalantibody. Hum Pathol 24:710–716, 1993

11. ALPERS CE, HUDKINS KL, GOWN AM, JOHNSON RJ: Enhancedexpression of “muscle-specific” actin in glomerulonephritis. Kidney Int41:1134–1142, 1992

12. SIMONSON MS, WALSH K, KUMAR CC, BUSHEL P, HERMAN WH: Twoproximal CArG elements regulate SM alpha-actin promoter, a geneticmarker of activated phenotype of mesangial cells. Am J Physiol268:F760–F769, 1995

13. ALPERS CE, HUDKINS KL, FERGUSON M, JOHNSON RJ, SCHATTE-MAN GC, BOTHWELL M: Nerve growth factor receptor expression infetal, mature, and diseased human kidneys. Lab Invest 69:703–713,1993

14. HUGO C, PICHLER R, GORDON K, SCHMIDT R, AMIEVA M, COUSERWG, FURTHMAYR H, JOHNSON RJ: The cytoskeletal linking pro-teins, moesin and radixin, are upregulated by platelet-derivedgrowth factor, but not basic fibroblast growth factor in experimen-tal mesangial proliferative glomerulonephritis. J Clin Invest 97:2499 –2508, 1996

15. FLOEGE J, ENG E, YOUNG BA, ALPERS CE, BARRETT TB, BOWEN-POPE DF, JOHNSON RJ: Infusion of platelet-derived growth factor orbasic fibroblast growth factor induces selective glomerular mesangialcell proliferation and matrix accumulation in rats. J Clin Invest92:2952–2962, 1993

16. GROMA V, MARCUSSEN N, OLSEN S: A quantitiative immunohisto-chemical study of the expression of mesangial alpha—smooth muscleactin and the proliferative marker KI-67 in glomerulonephritis in man.Virchows Arch 431:345–350, 1997

17. HAYMAN EG, ENGVALL E, A’HEARN E, BARNES D, PIERSCHBACHERM, RUOSLAHTI E: Cell attachment on replicas of SDS polyacryl-amide gels reveals two adhesive plasma proteins. J Cell Biol95:20 –23, 1982

18. KLEINMAN HK, LUCKENBILL-EDDS L, CANNON FW, SEPHEL GC: Useof extracellular matrix components for cell culture. (Review) AnalBiochem 166:1–13, 1987

19. BURRIDGE K, FATH K: Focal contacts: Transmembrane links between

the extracellular matrix and the cytoskeleton. (Review) Bioessays10:104–108, 1989

20. BALK SD: Calcium as a regulator of the proliferation of normal, butnot of transformed, chicken fibroblasts in a plasma-containing me-dium. Proc Natl Acad Sci USA 68:271–275, 1971

21. ROSS R, GLOMSET J, KARIYA B, HARKER L: A platelet-dependentserum factor that stimulates the proliferation of arterial smoothmuscle cells in vitro. Proc Natl Acad Sci USA 71:1207–1210, 1974

22. KOHLER N, LIPTON A: Platelets as a source of fibroblast growth-promoting activity. Exp Cell Res 87:297–301, 1974

23. POGGI A, VICENZI E, CIOCE V, WASTESON A: Platelet contribution tocancer cell growth and migration: The role of platelet growth factors.(Review) Haemostasis 18:18–28, 1988

24. GEE AP, MINTA JO: Migration of porcine endothelial and smoothmuscle cells in response to platelet-associated factors. Am J Hematol17:29–38, 1984

25. MORDAN LJ: Induction by growth factors from platelets of thefocus-forming transformed phenotype in carcinogen-treated C3H/10T1/2 fibroblasts. Carcinogenesis 9:1129–1134, 1988

26. PAUL D, PIASECKI A: Rat platelets contain growth factor(s) distinctfrom PDGF which stimulate DNA synthesis in primary adult rathepatocyte cultures. Exp Cell Res 154:95–100, 1984

27. RUSSELL WE, MCGOWAN JA, BUCHER NL: Biological properties of ahepatocyte growth factor from rat platelets. J Cell Physiol 119:193–197,1984

28. RUTHERFORD RB, ROSS R: Platelet factors stimulate fibroblasts andsmooth muscle cells quiescent in plasma serum to proliferate. J CellBiol 69:196–203, 1976

29. COUSER WG, JOHNSON RJ: Mechanisms of progressive renal diseasein glomerulonephritis. (Review) Am J Kidney Dis 23:193–198, 1994

30. ELGER M, DRENCKHAHN D, NOBILING R, MUNDEL P, KRIZ W:Cultured rat mesangial cells contain smooth muscle alpha-actin notfound in vivo. Am J Pathol 142:497–509, 1993

31. KITAMURA M, ISHIKAWA Y: Three-dimensional matrix primes mesan-gial cells to down-regulation of alpha-smooth muscle actin via deac-tivation of CarG box elements. Kidney Int 53:690–697, 1998

32. GLASS WF, TENG PR, HANEY LB: Extracellular matrix distributionand hillock formation in human mesangial cells in culture withoutserum. J Am Soc Nephrol 7:2230–2243, 1996

33. SHULTZ PJ, KNAUSS TC, MENE P, ABBOUD HE: Mitogenic signals forthrombin in mesangial cells: Regulation of phospholipase C andPDGF genes. Am J Physiol 257:F366–F374, 1989

34. JAFFER F, SAUNDERS C, SHULTZ P, THROCKMORTON D, WEINSHELL E,ABBOUD HE: Regulation of mesangial cell growth by polypeptidemitogens. Inhibitory role of transforming growth factor beta. Am JPathol 135:261–269, 1989

35. ADAMS MD, KELLEY JM, GOCAYNE JD, DUBNICK M, POLYMEROPOU-LOS MH, XIAO H, MERRIL CR, WU A, OLDE B, MORENO RF,KERLAVAGE AR, MCCOMBIE WR, VENTER JC: Complementary DNAsequencing: Expressed sequence tags and human genome project.Science 252:1651–1656, 1991

36. TSO JY, SUN XH, KAO TH, REECE KS, WU R: Isolation andcharacterization of rat and human glyceraldehyde-3-phosphate dehy-drogenase cDNAs: Genomic complexity and molecular evolution ofthe gene. Nucleic Acids Res 13:2485–2502, 1985

37. FINNEGAN MC, HAMMOND DW, HANCOCK BW, GOYNS MH: Activa-tion of MYCN in a case of non-Hodgkin’s lymphoma. Leukemia &Lymphoma 18:511–514, 1995

38. HEMBROUGH TA, VASUDEVAN J, ALLIETTA MM, GLASS WF, GONIAS

SL: A cytokeratin 8-like protein with plasminogen-binding activity ispresent on the external surfaces of hepatocytes. HepG2 cells andbreast carcinoma cell lines. J Cell Sci 108:1071–1082, 1995

39. PANKEWYCZ OG, GUAN JX, BOLTON WK, GOMEZ A, BENEDICT JF:Renal TGF-beta regulation in spontaneously diabetic NOD mice withcorrelations in mesangial cells. Kidney Int 46:748–758, 1994

40. SAMBROOK J, FRITSCH EF, MANIATIS T: Molecular Cloning: A Labo-ratory Manual (2nd ed). Cold Spring Harbor, Cold Spring HarborLaboratory Press, 1989

41. HARLOW E, LANE D: Antibodies: A Laboratory Manual. Cold SpringHarbor, Cold Spring Harbor Laboratory Press, 1988

42. TOWBIN H, STAEHELIN T, GORDON J: Electrophoretic transfer of

Stephenson et al: a-SMA and hypertrophy in mesangial cells1186

proteins from polyacrylamide gels to nitrocellulose sheets: Proce-dure and some applications. Proc Natl Acad Sci USA 76:4350 – 4354,1979

43. FAULSTICH H, ZOBELEY S, RINNERTHALER G, SMALL JV: Fluorescentphallotoxins as probes for filamentous actin. (Review) J Muscle ResCell Motil 9:370–383, 1988

44. ORLANDI A, ROPRAZ P, GABBIANI G: Proliferative activity and alpha-smooth muscle actin expression in cultured rat aortic smooth musclecells are differently modulated by transforming growth factor-beta 1and heparin. Exp Cell Res 214:528–536, 1994

45. DESMOULIERE A, GEINOZ A, GABBIANI F, GABBIANI G: Transforminggrowth factor-beta 1 induces alpha-smooth muscle actin expression ingranulation tissue myofibroblasts and in quiescent and growing cul-tured fibroblasts. J Cell Biol 122:103–111, 1993

46. CHOI ME, KIM EG, HUANG Q, BALLERMANN BJ: Rat mesangial cellhypertrophy in response to transforming growth factor-beta 1. KidneyInt 44:948–958, 1993

47. BOWEN-POPE DF, HART CE, SEIFERT RA: Sera and conditionedmedia contain different isoforms of platelet-derived growth factor

(PDGF) which bind to different classes of PDGF receptor. J BiolChem 264:2502–2508, 1989

48. GURWITZ D, CUNNINGHAM DD: Thrombin modulates and reversesneuroblastoma neurite outgrowth. Proc Natl Acad Sci USA 85:3440–3444, 1988

49. MAKINO H, KASHIHARA N, SUGIYAMA H, KANAO K, SEKIKAWA T,OKAMOTO K, MAESHIMA Y, OTA Z, NAGAI R: Phenotypic modulationof the mesangium reflected by contractile proteins in diabetes. Dia-betes 45:488–495, 1996

50. TURLA MB, THOMPSON MM, CORJAY MH, OWENS GK: Mechanismsof angiotensin II- and arginine vasopressin-induced increases inprotein synthesis and content in cultured rat aortic smooth musclecells. Evidence for selective increases in smooth muscle isoactinexpression. Circ Res 68:288–299, 1991

51. BLACK FM, PACKER SE, PARKER TG, MICHAEL LH, ROBERTS R,SCHWARTZ RJ, SCHNEIDER MD: The vascular smooth muscle alpha-actin gene is reactivated during cardiac hypertrophy provoked by load.J Clin Invest 88:1581–1588, 1991

Stephenson et al: a-SMA and hypertrophy in mesangial cells 1187