Life Sciencca, Vol. 61, No. 7, pp. 722-738, 1997 Copyright 0 1997 Ekmicr Science Inc.

Printed in the USA. All rights reserved om-32osp7 s17.00 t .I0

PII s0024-3205(97)00537-7

ENHANCED SYNTHESIS OF PRGI-EOGLYCANS BY VASCULAR ENDOTHELIAL CELLS TREATED WITH PHORBOL ESTER

Zhuo Tao 4~. Frank W. Smart a, Julio E. Figueroa b, D. Luke Glancy c, and Pa&at Vijayagopal c

Cardiac Transplantation Unit, Ochsner Medical Institutions a, and Sections of Infectious Diseases b and Cardiology c, Department of Medicine, Louisiana State University Medical Center, New

Orleans, Louisiana U.S.A.

(Received in final form May 20,1997)

We investigated the biosynthesis of proteoglycans (PG) in endothelial cells following their treatment with phorbol 1Zmyristate 13-acetate (PMA). Confluent cultures of bovine aortic endothelial cells were incubated in the presence and absence of PMA (100 @ml) and then pulsed with [s%]sulfate, [sH]glucosamine, or [3sS]sulfate plus [sH]leucine for varying times in the absence of PMA. Alternatively, confluent endothelial cells were simultaneously incubated with PMA and [%]sulfate for varying times. The metabolically labeled PG in the cell layer and medium were analyzed. Both short-term and prolonged exposure of endothelial cells to PMA significantly stimulated PG synthesis, regardless of the experimental conditions. [%]sulfate incorporation into newly synthesized PG in PMA-treated cells also increased by 1.7-fold and 3.6-fold over control cells, following a 15min and 30-min pulse, respectively. Cycloheximide markedly inhibited the increased synthesis of PG in PMA-treated cells, while actinomycin D produced a moderate inhibition. PG secretion was increased in PMA-treated cells compared with control cells, while there was no significant difference in PG degradation between the two cultums. PG from control and PMA-treated endothelial cell cultures did not differ in composition or hydrodynamic sixes. The incorporation of [sH]leucine into total cellular proteins decreased significantly following exposure of endothelial cells to PMA. Endothelial cells exposed to PMA for 3 h had significantly more protein kinase C (PKC) activity than did control cells. Inhibition of PKC by calphostin C abolished the PMA-mediated stimulation of PG synthesis in endothelial cells. The results indicate that PMA stimulates PG synthesis in endothelial cells either directly or indirectly through a PKC dependent mechanism.

I&Y Words: endothelial cells, Ph4A stimulation, proteoglyciur synthesis

Proteoglycans are important structural components of the vasculature that also maintain vascular permeability and influence lipid metabolism, hemostasis, and thrombosis ( 1,2). Several recent studies also support a role for vascular proteoglycans in atherogenesis (3-6).

Endothelial cells and smooth muscle cells synthesize the majority of arterial wall proteoglycans. Endothelial cells alter their proteoglycan synthesis in response to many stimuli and growth conditions. For example, we have shown that low-density lipoprotein (LDL) stimulates promoglycan synthesis by human umbilical vein endothelial cells and that LDL-treated cultures secrete a high-molecular-weight chondtoitin sulfate proteoglycan, compared with cells not exposed

Correspondence: Pa&at Vijayagopal, Ph.D., Department of Medicine, Louisiana State University Medical Center, 1542 Tulane Avenue, New Orleans, LA 70112, Telephone (504) 568-8305, FAX (504) 568-2127

724 Effect of Ph4A on Proteoglycm Synthesis Vol. 61, No. 7, 1997

to LDL(7). Kinsella and Wight (8) observed stimulation of chondroitin sulfate/dermatan sulfate proteoglycan synthesis by bovine aortic endothelial cells during migration. Similarly, endothelial cell sprouting results in the loss of high-molecular-weight heparan sulfate proteoglycan (9).

In vim, arterial wall endothelial cells are in a contact-inhibited quiescent state. Immune reactions can activate endothelial cells, resulting in a number of morphologic and functional changes(l0). The activated endothelial cells express increased levels of many proteins (11). However, the effect of endothelial cell activation on proteoglycan synthesis is not known. In vim, cytokines and growth factors, acting via activation of protein kinase C (PKC) (12), are the main signals that stimulate endothelial cell activation. Since phorbol 1Zmyristate 13-acetate (PMA) also stimulates PKC in several cells, including endothelial cells( 11, 13-U), in the present study we investigated the effect of PMA on proteoglycan synthesis by cultured endothelial cells. Gur findings indicate that both short-term (l-3 h) and long-term (24 h) exposure of endothelial cells to PMA results in a significant stimulation of proteoglycan synthesis. Although short exposures to PMA am known to activate PKC(ll, 15). prolonged exposure down-regulates this enzyme (16, 17). Therefore, the stimulation of proteoglycan synthesis in endothelial cells exposed to PMA for 24 h is not likely to be directly mediated by a PKC mechanism, but may be mediated through other downstream pathways modulated by PKC.

Materials Dulbecco’s modified Eagle’s medium, newborn calf serum, and Dulbecco’s

phosphate-buffered saline were purchased from Gibco BRL Life Technologies (Grand Island, NY). We obtained tissue-culture plasticware from Corning (Coming, NY) and collagenase (Type 1) from Worthington (Freehold, NJ). [ssS]sulfate, [sH]glucosamine, [sH]leucine, and Ecolite scintillation fluid came from ICN (Irvine, CA). The PKC enzyme assay kit was obtained from Amersham Life Science (Arlington Heights, IL). All other reagents were purchased from Sigma Chemical (St. Louis, MO).

Cell culture Endothelial cells were isolated from bovine thoracic aortas by collagenase digestion (18).

Briefly, the endothelial cell layer was scraped off the intima with a sterile scalpel. The cells were then dispersed by incubation in 0.1% collagenase for 20 min at 37°C and collected by centrifugation. Following resuspension in culture medium (Dulbecco’s modified Eagle’s medium, containing 10% newborn calf serum, 100 unit/ml penicillin, 100 @ml streptomycin, 0.25 l@rnl amphotericin B; medium A), the cells were incubated in a 25 cm2 plastic tissue cultute flask at 37°C in a humidified CO2 (5%) incubator. After 24 h the culture medium was replaced with fresh medium to remove nonadherent cells. The endothelial nature of the cells was established by their cobblestone morphology at confluence and the expression of factor VIII-related antigen determined by immunofluorescence assay. The cells showed negative staining with smooth muscle cell-specific a -actin antibody.

PMA treatment of endothelial cells Endothelial cells of the 5th-7th passage were plated in six-well plastic tissue culture plates

and allowed to grow to confluence. Cultures were then incubated at 37“C in medium A in the presence or absence of PMA (100 ng/ml) for varying duration as indicated under individual experiments. Previous studies have used this PMA concentration to activate various cultured cells, including vascular endothelial cells( 11,14,15). Some of the cultures were also incubated in media containing dimethylsulfoxide (DMSO), the vehicle for PMA.

Radiolabeling and isolation of proteoglycans Control and PMA-treated endothelial cell cultures were rinsed once with Dulbecco’s

phosphate-buffered saline and then incubated in 1 ml of medium A containing either 20 @i/ml of [ssS]sulfate or 20 @i/ml of [ sH]glucosamine at 37’C for varying times. In some experiments cells were also incubated with [3?3]sulfate (20 pCi/ml) and [ sH]leucine (20 @i/ml) in order to radiolabel the glycosaminoglycans and protein core, respectively, of the newly synthesized proteoglycans. At the end of each labeling period, the culture medium from each well was transferred to a test tube and

Vol. 61, No. 7, 1997 Effect of PMA on Proteoglycan Synthesis 725

Chilled on ice. To extract proteoglycans, solid guanidine hydrochloride was added to the medium to a fmal concentration of 4.0 M, together with protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 20 mM EDTA, 50 mM knzamidine hydrochloride, 10 mM n-ethylmah$mide, and 100 mM a-aminocaproic acid). The cell layer was washed once with Dulbecco’s phosphate-buffered-saline, and representative cultures were used for cell count or protein assay. Cell-associated proteoglycans were extracted with 1 ml of 4 M guanidine hydrochloride, 0.05 M sodium acetate containing 0.1% Triton X-100, and protease inhibitors (as mentioned above) for 24 h at 4’C. The extraction procedure removed virtually all (99%) of the labeled proteoglycans from the cell layers of both control and PMA-treated cultures. Ptoteoglycan extracts of medium and cell layer were stored at -20°C until processed.

Unincorporated radioactive precursors were removed from the proteoglycan extracts by Sephadex G-50 chromatography (19). An aliquot of excluded G-50 fraction was counted in a scintillation counter (Beckman LS 3801) to quantify labeled proteoglycans. Earlier we determined that the [ssS]-radioactivity in the excluded fraction from the Sephadex G-50 column represented only proteoglycans (19). Proteoglycans in the remainder of the excluded G-50 fraction were purified by ion-exchange chromatography on DEAE-Sephacel. The solutions were dialyzed against a buffer containing 7 M urea, 0.15 M NaCl, 0.05 M sodium acetate (pH 6.0), 0.5% (w/v) Triton X-100, and protease inhibitors and then applied to a column of DEAE-Sephacel equilibrated in the above buffer. The column was washed with 20 ml of the same buffer, and labeled proteoglycans were then eluted with a continuous NaCl gradient (from 0.15 M to 1.0 M) in the same buffer at a flow rate of 10 ml/h. Fractions (1 ml) were collected, and aliquots were counted for [:ssS] radioactivity. Carrier chondroitin sulfate (300 pg) was added to the purified fractions, and proteoglycans were precipitated in 1% c&y1 pyridinium chloride (CPC) at 25°C. The proteoglycans were then reisolated from their CPC complexes (20). dissolved in phosphate-buffered saline, and analyzed for radioactivity.

Pulse-chase experiment We performed a pulse-chase analysis to determine the effect of PMA on proteoglycan

turnover in endothelial cells. Confluent endothelial cells in six-well plates were incubated for 24 h at 37°C in the presence or absence of 100 @ml PMA. Cultures were then washed with phosphate-buffeted saline and incubated in fresh medium containing 50 pCi/ml of [ssS]sulfate at 37°C for 2 h. At tire end of the pulse period, the medium was discarded, and the cells were washed six times with phosphate-buffeted saline containing 0.5 M Na2S04. Separate welts were chased in isotope-free medium for up to 8 h. After each chase period, the proteoglycans were extracted as described above. Labeled proteoglycans secreted into the medium and remaining in the cell layer, as well as [ssS]sulfate generated by intracellular degradation of proteoglycans, were quantified after fractionation by Sephadex G-50 chromatography.

Protein synthesis The effect of PMA treatment on general protein synthesis was determined by measuring the

incorporation of [sH]leucine into secreted and cell-associated proteins. Control and 24-h PMA- treated cells were incubated in medium A containing 10 pCi/ml of [sH]leucine. After 24 h, total secreted and cell-associated proteins were precipitated with trichloroacetic acid (19). The precipitated proteins were dissolved in 1 N NaOH, and aliquots were counted for [3H] radioactivity.

PKC assay PKC activity was assayed according to the manufacturer’s instructions.

Analysis of proteoglycans Confluent endothelial cells in 25 cm2 culture flasks were incubated in the presence and

absence of 100 @ml of PMA for 24 h at 37’C. Cultures were then washed and incubated in fresh medium (no PMA) containing 50 pCi/ml of [ssS]sulfate for another 24 h. Labeled proteoglycans in the medium and cell layer were isolated as described above and purified by DEAE-Sephacel chromatography. Aliquots of the purified proteoglycans were analyzed for component glycosaminoglycans. The individual glycosaminoglycans were determined by treatment with nitrous acid (21) or chondroitin ABC lyase(22). The reaction products from each treatment were then

726 Effect of PMA on Proteoglycan Synthesis Vol. 61, No. 7, 1997

chromatographed on a Sephadex G-SO column with a buffer of 0.05 M Tris-HCl, 0.2 M NaCl, 1 mM NaN3, and 0.2% triton X-100, pH 7.4. Fractions of 1 ml were collected and analyzed for radioactivity.

We used analytical Sepharose CL-2B chromatography to determine the molecular size distribution of proteoglycans. After sample application, the column was eluted with 4.0 M guanidine hydrochloride and 0.05 M sodium acetate (pH 5.8) containing 0.2% Triton X-100 at a flow rate of 9.6 ml/h. Fractions (1 ml) were. analyzed for [35S] radioactivity, and the approximate molecular weight of the proteoglycan was calculated from the elution position on the column (23).

Effect of PMA on PKC activity Figure 1 compares PKC activity in control endothelial cells and cells exposed to PMA for 3

h. PMA-treated cells had significantly more enzyme activity than did control cells.

6- l

Cl Control q PMA

6’

4’

2’

0.

Fig. 1 Effect of exposure to PMA on endothelial cell PKC activity. Confluent bovine aortic endothelial cells were incubated in the presence or absence of 100 @ml of PMA for 3 h and then washed with PBS. PKC activity was then assayed. The bars represent the mean f SEM of triplicate cultures. PKC activity in PMA-treated cells was significantly different from that in control cells.* = pd.05

Effect of PMA on endothelial cell proteoglycan synthesis Figure 2 shows the incorporation of [35S] radioactivity into secreted and cell-associated

proteoglycans as a function of the length of exposure to PMA. In these experiments the cells were exposed to PMA for different times, washed, and then pulsed with [35S]sulfate for 24 h in the absence of PMA. Compared with control cells, cells that were exposed to PMA for 1,2,3, and 24 h produced a significant increase in secreted proteoglycans (Panel A). The cell-associated proteoglycans increased significantly in cultures exposed to PMA for l-3 h (Panel B). Although cellular proteoglycans also increased in cultures treated with PMA for 24 h compared with the 3-h exposure cultures, the increase was less than that of the 24-h control cultures. Exposure to PMA for up to 24 h did not affect cell viability as determined by trypan blue exclusion. In all PMA-treated cultures, the total proteoglycan synthesis (media + cell) was significantly greater than that of the corresponding control cultures. In the control cultures, secreted and cell-associated proteoglycans represented 72% and 2896, respectively, of the total radiolabeled proteoglycans. Proteoglycan secretion was increased in the corresponding PMA-treated cultures so that the media now contained

Vol. 61, No. 7, 1997 Effect of PMA on Proteoglycan Synthesis 721

84% and the cell layer 16% of the total labeled proteoglycans. The incorporation of [sH]glucosamine into proteoglycans also increased significantly in PMA-treated endothelial cells compared with control cells (data not shown). DMSO, tbe solvent for PMA, had no effect on proteoglycan production by endothelial cells. In preliminary studies, we have determined that exposure of endothelial cells to PMA for 24 h increases versican, biglycan, decorin, and heparan sulfate proteoglycan (data not shown).

10000 -

8000 .

6000.

2000

1500

1000

500

0

A

. . .

1 2

I

0 Control q PMA

24

Cl Control q PMA

24

TIME (Ii)

Fig. 2 Incorporation of [ssS]sulfate into secreted (A) and cell-associated (B) proteoglycans as a function of exposure time to PMA. Confluent bovine aortic endothelial cells were incubated in the presence or absence of PMA (100 @ml) for the indicated times. At the end of each time period, cultures were washed and pulsed with 20 lKX/ml of [W]sulfate for 24 h in the absence of PMA. Labeled proteoglycans in the medium and cell layer were extracted with 4 M guanidine hydrochloride, isolated by Sephadex G-50 chromatography and quantitled. The bars represent the mean f SEM of three experiments, each performed in triplicate. PMA-treated cells were significantly different from control cells. * = ~~0.01, ** = p<O.OOl, *** = p<O.OOOl.

728 Effect of PMA on Proteoglycan Synthesis Vol. 61, No. 7, 1997

In a modification of the time-course experiment, we incubated confluent endothelial cells with [WJsulfate in the presence or absence of PMA for 1, 2, 3, and 24 h and determined total proteoglycan synthesis. Compared with control cells, total proteoglycans increased significantly in PMA-treated cells at all time points (data not shown). These results, together with the results presented in Figure 2, clearly indicate that both short and prolonged exposure of endothelial cells to PMA significantly stimulate the incorporation of radiolabeled sulfate into newly synthesized proteoglycans.

l l

3000 Cl Control

0 35 S 3H

Fig. 3 Incorporation of [%]sulfate and [sH]leucine into secreted proteoglycans by endothelial. cells treated with PMA. Confluent endothelial cells were treated with PMA for 3 h, washed and then pulsed with [ W]sulfate (2OpWml) and [WJleucine (2OpWml) for 6 h. Labeled proteoglycans were isolated and quantified. The values are the means f SEM of triplicate cultures. Both [3%]sulfate and [sH]leucine incorporation were significantly increased in PMA-treated cells when compared with the control. ** p=O.Ol.

To determine whether the above increase in [ 35S] incorporation into proteoglycans in PMA- treated cells was due to increased proteoglycan synthesis, we performed double labeling experiments to measure [Wlsulfate and [sH]leucine incorporation into glycosaminoglycan chains and protein core, respectively. As shown in Figure 3, both [s%]sulfate and [Wjleucine incorporation into newly synthesized proteoglycans significantly increased in PMA-treated cells compared with control cells. Additionally, in separate experiments, control and 24-h PMA-treated cells were incubated with [%]sulfate for 15 min or 30 min, and proteoglycans were then isolated and quantitated. The results are presented in Figures 4A and B. After 15 min of labeling, when secretion was very low, [%I incorporation into newly synthesized proteoglycans increased by 1.7-fold in PMA-treated endothelial cells over control (Fig. 4A, p&01). Likewise, after a 30-min pulse, there was a 3.6-fold increase in [%I incorporation into newly synthesized proteoglycans in PMA-treated cells over the control (Fig. 4B, pcO.001). PMA stimulation also significantly increased [35S] radioactivity in secreted proteoglycans (Fig. 4B, p&02). Furthermore, cycloheximide, a specific inhibitor of protein synthesis, was also used to determine whether the stimulation of proteoglycan production by PMA required the de novo synthesis of proteoglycan core protein. Figure 5 shows that the enhanced synthesis of proteoglycans in PMA-tmated cultures was inhibited significantly by cycloheximide, although the inhibition did not reduce synthesis to the control levels. Actinomycin D, a specific inhibitor of RNA transcription, inhibited proteoglycan synthesis in both control and PMA-treated cells, although the extent of inhibition was greater in the controls than in the cells exposed to PMA (52% vs 24%; Figure 6). Consequently, inhibition of transcription affected the proteoglycan synthesis of control cells more significantly than that of the PMA-treated cells. Taken together, these results show that exposure of endothelial cells to PMA stimulates de novo proteoglycan synthesis.

Vol. 61, No. 7, 1997 Effext of PMA on Proteoglycaa Synthesis 729

200 -

MEDIUM 15007

B

Cl Control rapMA

1000.

CELL

MEDIUM CELL

Fig. 4 Proteoglycan synthesis by PMA treated endothelial cells after short-term pulse. Endothelial cells were exposed to PMA for 24 h, washed and then incubated with 50 @/ml of [ssS]sulfate for 15 min (A) or 30 mitt (B). Labeled proteoglycans (medium and cell layer) were then isolated and quantifkd. The values am the mean f SEM of triplicate cultums. PMA-treated cells were signiticantly different from control cells. * = pd.02, ** = p&.01, *** = p<O.OOl.

Pulse-chase studies Figure 7 shows the results of the pulse-chase experiment conducted to examine the effect of

PMA on proteoglycan turnover. Release of [ssS]-labeled proteoglycans from control and PMA-treated cells proceeded at about the same rate during the fust 30 min of the chase. After that, however, labeled proteoglycans disappeared more rapidly from the PMA-treated cells than from the control cultures. Consequently, at the end of the 8-h chase, only 30% of the total labeled proteoglycans present at time zero in PMA-treated cultures still remained in the cell layer. For control cultures the value was 44%. These data further indicate that proteoglycan secretion is increased in PMA-treated endothelial cells. The rate of degradation of [WI-labeled proteoglycans into free sulfate was similar in control and PMA-treated cultures up to 4 h of chase. After that, the rate of proteoglycan degradation was greater in PMA-treated culhues than in control cultures.

730 Effect of PMA on Proteoglycan Synthesis Vol. 61, No. 7, 1997

600.

r-- ::::_:.: 400. ;y;i;;

::fi;f:; :::::;: ::I:;:.:

200. ::::::j: 1.’ .:.::::: . .:.:.. I:::.:. ::.:;::: ::::::.:\ \ \,

0’ :i;;::;:\>‘,;,; I

CONTROL PMA

Fig. 5 Effect of cycloheximide (CYC) on total proteoglycan synthesis by endothelial cells exposed to PMA. Control and 24 h PMA-treated endothelial cells were incubated in the presence or absence of CYC (10 @ml) for 6 h. During the last 4 h of the incubation, each well received 20 pCi/ml of [s%]sulfate. Labeled proteoglycans in the cell layer and medium were isolated and assayed for radioactivity. Data represent the mean + SEM of three cultures.

500 -

400.

=

;j $ 300 .

z

8 200 .

5

ZL loo-

I$ -y; + c

CONTROL PMA

Fig. 6 Effect of actinomycin D (Act-D) on total proteoglycan synthesis by endothelial cells treated with PMA. Confluent endothelial cells were incubated for 20 h with or without PMA in the presence or absence of 50 ng/ml of Act-D. 20 pCi/ml of [35S]sulfate was then added and the cultures were incubated for another 4 h. Secreted and cell-associated proteoglycans were then quantified. Data represent the mean f SEM of thme cultures.

Vol. 61, No. 7, 1997 Effect of PMA on Proteoglycan Synthesis 731

\\’ ,\\X\Xb

.’ ,\\’

\“‘* L Free [SSS]sulfate (PMA) x:::: Free [dCS]suWate (control)

. . . . . . . . . . . . . . . .“%::... ..::::... ..:::i.. -::>

- Cell PG (PMA)

hw Cell PG (control)

0 1 2 3 4 5 6 7 6

CHASE TIME (H)

Fig. 7 Representative pulse-chase analysis. Control and PMA-treated endothelial cells (24 h PMA) were pulsed with 50 @i/ml of [%]sulfate for 2 h, washed extensively, and chased for up to 8 h in isotope-free medium. After each chase period, the amounts of labeled proteoglycans remaining in the cell layer and free [ssS]sulfate released into the medium from cellular degradation of proteoglycans were measured.

3000

cn =

P 2000

:

8 1000

L

0

q -calphostin q +calphostin

-t- :;:;::: l

:.::;I: ;.;:.::. . . ::::;:i .‘\‘\‘\’ :.:.:. .’ ” ’ .::.:.: /‘I’,‘, I:::.:. /‘/‘/‘, .I.::’ .: /‘#‘I’, :.:.:.I ,‘,‘,‘, :::;:;: /‘1’/‘, ::I.‘.’ .:.: ,‘/‘/‘, ::::.:. 1’,‘1’, I::.:.’ .:.:.:I L ,‘/‘/‘I #‘1’1’# :::::::. \. \ I

l .

I,,

.‘I’/’ ,‘/‘/’ .‘/‘/’ I’,‘,’ 1’1’1’ I’,‘,’ C’,‘,’ 1 I’#‘/’ l X,‘,< l ‘,‘,’ .‘,‘,’ ,‘,‘,’ t’,‘,’ .\\ I

CONTROL PMA

Fig. 8 Effect of calphostin C on proteoglycan synthesis by endothelial cells treated with PMA. Endothelial cells were incubated with or without PMA in the presence or absence of calphostin C (0.5 pM) for 3 h. The cells were washed and pulsed with 20 pCi/ml of [ssS]sulfate for 6 h. Labeled proteoglycans in the culture media were then isolated and quantified. The values are the means f SEM of triplicate cultures. Calphostin C significantly reduced proteoglycan synthesis in control and PMA- treated endothelial cells. *: p&05, **: p&01.

Effect of calphostin C on proteoglycan synthesis We conducted studies to determine the effect of the PKC inhibitor calphostin C on PMA-

732 Effect of PMA on Proteoglycan Synthesis Vol. 61, No. 7, 1997

mediated stimulation of proteoglycan synthesis by endotheiial cells. The results are presented in Figure 8. Calphostin C significantly inhibited the PMA-mediated stimulation of proteoglycan synthesis (~~0.01) and brought it down to the control level. In addition, calphostin also reduced proteoglycan synthesis in control cells (~~0.05). Proteoglycan synthesis in the presence of calphostin in PMA-treated cells was more than that in control cells exposed to calphostin (~4.05).

Effect of PMA on cell proliferation and [3H]leucine incorporation into total proteins

PMA did not stimulate endothelial cell proliferation determined by cell count. To investigate whether the increase of proteoglycan synthesis in PMA-treated cells was part of a general stimulation of protein synthesis, we determined protein synthesis by measuring the incorporation of [sH]leucine into secreted and cellular proteins. Compared with control cells, there was a significant decrease in both secreted protein (p&05) and cellular protein (p&01) in PMA-treated cells after 24 h of [3H] labeling (Fig. 9).

30000

m = 8 20000

2

-z

8 f 10000

L

0

0 Control

raPMA

JIl%- MEDIUM CELL

Fig. 9 Incorporation of [sH]leucine into cell-derived proteins. Control and 24 h PMA-treated endothelial cells were pulsed with [ sH]leucine (10 @X/ml) for 24 h. After labeling, the secreted and cell-associated proteins were precipitated with trichloroacetic acid, redissolved in 1 N NaOH, and analyzed for radioactivity. The bars represent the mean k SEM of triplicate cultures. PMA-treated cells were significantly different from control cells. * p=O.O5,** = p&01.

Characterization of proteoglycans MetaboIicalIy labeled proteoglycans secreted into the culture medium and associated with the

cell layer were fractionated by DEAE-Sephacel chromatography. Two radioactivity peaks were eluted from both the culture medium (Fig. 10A) and the cell layer (Fig. 1OB) of control and PMA-treated cultures. Secreted and cell-associated proteoglycans from control cultures were eluted at similar NaCl concentrations, peak 1 at 0.45 M and peak 2 at 0.53 M. Peak 1 (secreted and cell-associated) in PMA-tmated cultures eluted at a lower NaCl concentration (0.41 M), indicating a shift to a lower charge density. On the other hand, peak 2 eluted at a NaCl concentration similar to that of peak 2 in control cultures.

The fractions in peak 1 and peak 2 were pooled separately, and aliquots were treated with nitrous acid or chondroitin ABC lyase. Undigested and digested fractions were then chromatographed on Sephadex G-50 columns. Almost all the [WS]-radioactivity in peak 1 proteoglycans from the medium of control and PMA-treated cells was susceptible to nitrous acid degradation (Fig. 11). indicating that the proteoglycans contained predominantly heparan sulfate chains. Peak 2 of medium proteoglycans were resistant to nitrous acid treatment, but all the

Vol. 61, No. 7, 1997 Effect of PMA on Proteoglycan Synthesis 733

radioactivity from both control and PMA-treated cultures was susceptible to chondroitin ABC lyase digestion (Fig. 12), indicating that the proteoglycans in peak 2 contained chondroitin/dermatan sulfates chains. Peak 1 and peak 2 proteoglycans in the cell layer of control and PMA-treated cultutes were also comprised of heparan sulfate and chondroitinklermatan sulfates, respectively.

A -- NaCl Molar@ 2500.

\ /’

0’ 2000. /’

/’ 0’

1500.

1 000. - - Control

- PMA \

\

,

1.15

0.95

0.75

0.55

0.35

0.15

rl.15

- - NaCl Molarity 4

/’ ’ -0.95

4” .’ -0.75

/

- - Control

- PMA

uu - - - 10 20 30 40 50 60 50 80 50 100

FRACTION NUMBER

Fig. 10 DEAE-Sephacel chromatography of [%I-labeled proteoglycans. Control and PMA-treated endothelial cells were labeled with 50 pCi/ml of [ s%]sulfate for 24 h. Proteoglycans from medium and cell layer were isolated by Sephadex G-50 chromatography and fractionated on DEAE-Sephacel with a continuous NaCl gradient. Fractions were analyzed for radioactivity.

Molecular weight determination by analytical Sepharose CL-2B chromatography indicated that the molecular size of proteoglycans from control and PMA-treated endothelial cells were similar (data not shown).

734 Effect of Ph4A on Proteoglycan Synthesis Vol. 61, No. 7, 1997

800

600

400

200

0

Control - Untreated - Nitrous Acid

t 1

0 10 20 30 40

800

600

400

200

0 0 IO 20 30 40

FRACTION NUMBER

Fig. 11 Sephadex G-50 chromatography of secreted proteoglycans before and after treatment with nitrous acid. Aliquots of proteoglycans in peak 1 from DEAE-Sephacel chromatography were digested with nitrous acid. The reaction products were then chromatographed on a Sephadex G-50 column along with undigested material. Fractions were collected and analyzed for radioactivity.

Our results show that both short and prolonged exposure of bovine aortic endothelial cells to PMA stimulate the incorporation of radiolabeled precursors into proteoglycans. This could result from increased synthesis and/or decreased degradation, increased glycosaminoglycan chain length, or more glycosaminoglycan chains per core protein. However, the following lines of evidence indicate that the increase in labeled proteoglycans in PMA-treated cultures is due to increased de novu synthesis. First, there was a significant increase in glycosaminoglycans and proteoglycan core proteins in PMA-treated cells as indicated by the increased incorporation of both [%]sulfate and [sH]leucine, respectively. At the same time, the sulfate to protein ratio between control (2.07) and PMA-treated cells (2.10) did not change, indicating that the proteoglycans from both cultures contained the same number of glycosaminoglycan chains per core protein. Second, in PMA-treated endothelial cells, the incorporation of [WS]sulfate into newly synthesized proteoglycans was increased by 1.7-fold and 3.6-fold over the control following a 15min and 30-min labeling,

Vol. 61, No. 7, 1997 Effect of Ph4A on Proteoglycan Synthesis 735

respectively. Third, cycloheximide, an inhibitor of protein synthesis, significantly reduced the PMA-mediated stimulation of proteoglycan synthesis. Fourth, actinomycin D, a specific inhibitor of transcription, inhibited PMA-mediated synthesis of proteoglycans by 24%. Last, the rate of proteoglycan degradation in PMA-treated cultures was either similar to that of the control cultures or greater, indicating that the increased proteoglycan accumulation was not due to decreased degradation.

Control

500 - - Untreated - AE3Case

400 -

E

e 300 -

s L 200.

100’

07 1 0 10 20 30 40

PMA 600 -

500 -

400 *

300 *

200 -

100’

FRACTION NUMBER

Fig. 12 Sephadex G-50 chromatography of secreted proteoglycans before and after digestion with chondroitin ABC lyase. Aliquots of the second proteoglycan peak from DEAE-Sephacel chromatography were chromatographed on a -Sephadex G-50 column before and after chondroitin ABC lyase treatment. Fractions were collected and analyzed for radioactivity.

Recently Porcionatto et al(24) reported that exposure of rabbit aortic endothelial cells to PMA for 5 min significantly stimulated heparan sulfate proteoglycan synthesis. Short-term exposure of bovine endothelial cells to PMA in our study also stimulated total proteoglycan synthesis. Moreover, our data as well as data from the literature (11,15) indicate that short-term exposure of endothelial cells to PMA stimulates PKC. In addition, calphostin C, a specific inhibitor of PKC abolished the PMA-mediated stimulation of proteoglycan synthesis. Taken together, these data suggest that PKC-mediated signaling pathways are involved in the acute increase in proteoglycan

736 Effect of PhIA on Proteoglycan Synthesis Vol. 61, No. 7, 1997

synthesis following short exposure to PMA. Interestingly, calphostin C also reduced proteoglycan synthesis in control cells. Since endotbelial cells express PKC activity even under basal conditions (Fig. 1). the above result indicates that PKC stimulates proteoglycan synthesis in control cells also. We did not address the precise mechanism by which PKC stimulates proteoglycan synthesis in endotbelial cells. PKC likely induces changes in the transcription or translation of the genes involved in proteoglycan synthesis. This needs to be determined. It should be pointed out that inhibition of PKC in control and PMA-treated cells did not result in total inhibition of proteoglycan synthesis. This indicates that other mechanisms that do not involve PKC-dependent signaling pathways are also important in regulating proteoglycan synthesis in endothelial cells.

Our data show that proteoglycan synthesis in the presence of calphostin C in PMA-treated cells is more than that in the control cells exposed to calphostin. This may be due to incomplete inhibition of PKC by calphostin in PMA-treated cells. If, on the other hand, PKC was inhibited in these cells to the same extend as in control cells, the observed difference in proteoglycan synthesis indicates the involvement of additional mechanisms. Obviously, further studies are needed to resolve these issues.

Another interesting observation of our study pertains to the increased synthesis of proteoglycans following prolonged exposure of endothelial cells to PMA. Because prolonged exposure to PMA down-regulates PKC (16, 17), this elevation in proteoglycan synthesis under the condition cannot be a direct PKC effect. The initial increase in PKC following exposure to PMA may stimulate the production of autocrine molecules such as transforming growth factor-8 (TGF-8) and basic fibroblast growth factor, which in turn may modulate proteoglycan synthesis. Such PKC-mediated up-regulation of these cytokines has been reported in vascular smooth muscle cells(25)and mesangial cells( 16). We and others have shown that TGF-i3 is a potent stimulator of proteoglycan synthesis in endothelial cells (26,27).

The indirect stimulatory effect of PKC on endotbelial cell proteoglycan synthesis involves both transcription and translation of proteoglycan-related genes. This is consistent with the suppression of PMA-mediated stimulation of proteoglycan synthesis by actinomycin D and cycloheximide. The marked reduction of proteoglycan synthesis by cycloheximide indicates that the PMA effect on proteoglycan metabolism involves increased translation of proteoglycan-specific mRNA and not increased secretion of a pie-formed intracellular pool of proteoglycans. Likewise, the incomplete inhibition of synthesis by actinomycin D implies continued translation of pm-existing mRNA. The effect of actinomycin D was greater on control cells than on PMA-treated cells, which suggests that proteoglycan-specific mRNA in PMA-treated cells may be more stable than it is in control cells.

Exposing endothelial cells to PMA for 24 h produced a significant decrease in the incorporation of [ sH]leucine into proteins. Therefore, the effect of prolonged exposure to PMA on proteoglycan synthesis by endothelial cells appears to be specific and occurs independently of the stimulation of general protein synthesis.

Results from the present study show that when stimulated by PMA, endothelial cells increase their secretion of proteoglycans. Consequently, at the end of a 24-h pulse, a major shift occurred in the distribution of proteoglycans into media and cell layer compartments. About 85% of the proteoglycans were secreted, as compared with about 65% in the control culture. The decreased secretion accounted for the increase in cell-associated proteoglycans in control cultures. The increased secretion of proteoglycans in PMA-treated cells could be a normal consequence of increased synthesis.

Although exposure of endothelial cells to PMA stimulates proteoglycan synthesis, this does not alter proteoglycan structure or composition. In contrast, the studies of Kinsella and Wight(8) indicate that compositional changes accompany the stimulation of proteoglycan synthesis following endothelial cell injury in v&o. This is chiefly the result of endotbelial cell migration, and to some extent, of increased cell proliferation after wounding. In our study, PMA treatment of endothelial cells did not stimulate cell proliferation nor, probably, migration.

Vol. 61, No. 7, 1997 Effect of PMA on Protenglycan Synthesis 737

In the intact blood vessel, although smooth muscle cells are the predominant proteoglycan-synthesizing cells, the contribution of endothelial cells cannot be underestimated. Our study shows that PMA stimulates proteoglycan synthesis in endothelial cells either directly or indirectly through a PKC-dependent mechanism. Endothelial cells constitute the bulk of normal arterial wall intima, and elevation of PKC in the vessel wall by any mechanism could stimulate proteoglycan synthesis. Consistent with earlier reports(3-6, 28, 29), a general increase in proteoglycan synthesis in the blood vessel may cause increased binding and retention of plasma LDL.

This study was supported by grant HL-42993 from the National Institutes of Health (PV) and the American Heart Assn-La Affiliate (JEF). We thank Anne Compliment for expert editorial assistance.

1.

32:

4.

5.

6. 7.

::

10.

:;* 13:

14.

15.

16.

17.

18.

19.

it?:

2232:

24.

G.S. BERENSON, B. RADHAKRISHNAMURTHY, S.R. SRINIVASAN, P. VIJAYAGOPAL, E.R. DALFERES Jr. and C. SHARMA, Exp. Mol. Pathol. 4-l 267- 287 (1984). T.N. WIGHT, Arteriosclerosis 2 l-20 (1989). P. VIJAYAGOPAL, S.R. SRINIVASAN, K.M. JONES, B.RADHAKRISHNAMURTHY and G.S. BERENSON, Biochim. Biophys. Acta &XZ 251-261 (1985). B.G.J. SALISBURY, D.J. FALCONE and C.R. MINICK, Am. J. Pathol. l2Q 6- 1 l( 1985). S. YLA-HERTTUALA, 0. JAAKKOLA, T. SOLAKIVI, H. KUIVANIEMI and TNIKKARI, Atherosclerosis 62 73-80 (1986). E. HURT, G. BONDJERS and G. CAMEJO, J. Lipid Res. 2 443-454 (1990). P. VIJAYAGOPAL, S. R. SRINIVASAN, E.DALFERES Jr., BRADHAKRISHNAMURTHY and G.S. BERENSON, Biochem. J. m 639-646 (1988). M.G. KINSELLA and T.N. WIGHT, J. Cell Biol. m 679-687 (1986). A. OOHIRA, T.N. WIGHT and P. BORNSTEIN, J. Biol. Chem. 2.58 2014-2021 (1983). M.E. GERRITSEN and C.M. BLOOR, FASEB. J. 2 523-532 (1993). E.M. SCARPATI and J.E. SADLER, J. Biol. Chem. 264 20705-207 13 (1989). Y. NISHIZUKA, Nature 3128 693-698 (1984). J. H. AUWERX, A. CHAIT, G. WOLFBAUER and S.S. DEEB, Mol. Cell Biol. 2 2298-2302 (1989). M. MOINAT, J.M. CHEVEY, P. MUZZIN, J.P. GIACOBINO and M. KOSSOVSKY, J. Lipid Res. ti 329-334 (1990). B.E. SLACK, R.M. NlTSCH, E. LIVNEH, G.M. KUNZ Jr, J. BREU, H. ELDAR and R.J. WURTMAN, J. Biol. Chem. 26.8 21097-21101 (1993). R.K. STUDER, P.A. CRAVEN and F.R. DERUBERTIS, Kidney Int. 46 1074-1082 (1994). R.K. STUDER, H. NEGRETE, P.A. CRAVEN and F.R. DERUBERTIS, Kidney Int. 48 422-430 (1995). D. GOSPODAROWICZ, J. MORAN, D. BRAUN and C. BIRDWELL, Proc. Natl. Acad. Sci. U.S.A. n 41204124 (1976). P. VIJAYAGOPAL, B&hem. J. 2% 603-611 (1993). I.J. EDWARDS, W.D. WAGNER and R.T. OWENS, Am. J. Patbol. l3fj 609-621 (1990). D.C. GOWDA, V.P. BHAVANANDAN and E.A. DAVIDSON, J. Biol. Chem. 26L 4926-4934 (1986). H. SAITO, T. YAMAGATA and S. SUZUKI, J. Biol. Chem. m 1536-1542 (1968). T. OHMO, J. BLACKWELL, A. JAMIESON, D. CARRINO and A. KAPLAN, Biochem. J. m 553-557 (1986). M.A. PORCIONATTO, C.R.M. PINTO, C.P. DIETRICH and H.B. NADER, Brazilian J. Med. Res. 22 2185-2190 (1994).

738 Effect of PMA on Proteoglycan Synthesis Vol. 61, No. 7, 1997

25.

26.

27. 28.

29.

A. ALI, M.G. DAVIS, M.W. BECKER and G.W. DORN, J. Biol. Chem. 2h8 17397- 17403 (1993). P. VIJAYAGOPAL, H.P. CIOLINO and G.S. BERENSON, B&him. Biophys. Acta m 129-40 (1992). M. MERRILEES and L. SCOTT, Atherosclerosis 81255-265 (1990). S.R. SRINIVASAN, P. VIJAYAGOPAL, E. DALFERES Jr., B. ABBATE, B. RADHAKRISHNAMURTHY and G.S. BERENSON, Atherosclerosis 62 201-208 (1986). S.R. SRINIVASAN, J.H. XU, P. VIJAYAGOPAL, B. RADHAKRISHNAMURTHY and G.S. BERBNSON, B&him. Biophys. Acta 1l68 158-166 (1993).