Androgens Stimulate Human Vascular Smooth Muscle Cell ...

0013-7227/05/$15.00/0 Printed in U.S.A.

Endocrinology 146(4):2085–2090 Copyright © 2005 by The Endocrine Society doi: 10.1210/en.2004-1242

Androgens Stimulate Human Vascular Smooth Muscle Cell Proteoglycan Biosynthesis and Increase Lipoprotein Binding Kazuhiko Hashimura, Krishnankutty Sudhir, Julie Nigro, Shanhong Ling, Maro R. I. Williams, Paul A. Komesaroff, and Peter J. Little Hormones and Vasculature (K.H., S.L., M.R.I.W., P.A.K., K.S.) and Cell Biology of Diabetes Laboratories (J.N., P.J.L.), Baker Heart Research Institute, and Department of Medicine, Monash University (J.N., P.J.L.), The Alfred Hospital, Melbourne, Victoria 8008, Australia; and Stanford University Medical Center (K.S.), Stanford, California 94304-5637 Vascular smooth muscle cell (VSMC) proliferation and proteoglycan biosynthesis are two critical contributors to the development of atherosclerosis. We investigated the effects of specific androgens, androstenedione, dihydrotestosterone, and testosterone, on proteoglycan biosynthesis in human VSMC derived from internal mammary arteries. Vascular SMCs were metabolically labeled with [35S]sulfate or [35S]methionine/cysteine to assess glycosaminoglycans (GAGs) or proteoglycan core protein, respectively. The electrophoretic migration of radiolabeled proteoglycans was assessed by SDSPAGE. Proteoglycan-low density lipoprotein (LDL) interactions were assessed using LDL affinity columns. Treatment of VSMCs with androstenedione (100 nM), dihydrotestosterone (10 nM), or testosterone (100 nM) increased [35S]sulfate incorporation into GAGs by 24.8% (P < 0.05), 22% (P < 0.05), and 32.5% (P < 0.05), respectively. Treatment of VSMCs with tes-

M

ALES HAVE A higher risk of coronary artery disease (CAD) compared with age-matched females despite a postmenopausal increase in the risk of CAD in women (1). Although much attention has focused on the protective effect of estrogens, endogenous androgens may be proatherogenic in males. Serum concentrations of testosterone are 10 times higher in males (10 –35 nm) than in females (0.5–2.3 nm) (2). To date, the role of androgens in atherosclerosis has attracted relatively little attention. Dihydrotestosterone regulates DNA synthesis in human vascular cells (3), whereas testosterone enhances apoptosis in human endothelial cells (4) and worsens endothelial dysfunction in experimental atherosclerosis (5), suggesting that androgens might be proatherogenic. In addition, testosterone treatment exacerbates diet-induced coronary atherosclerosis in female monkeys (6). Conversely, testosterone is a coronary vasodilator (7), and a study of postmenopausal women showed that endogenous androgens were inversely related to carotid intimal-medial thickness, suggesting potential beneficial vascular effects of anFirst Published Online January 20, 2005 Abbreviations: CAD, Coronary artery disease; CPC, cetylpyridinium chloride; DHT, dihydrotestosterone; FBS, fetal bovine serum; GAG, glycosaminoglycan; LDL, low density lipoprotein; T, testosterone; VSMC, vascular smooth muscle cell. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

tosterone did not alter [35S]methionine/cysteine incorporation into proteoglycan core protein, suggesting that the effect of testosterone was associated with an increase in GAG length. Dihydrotestosterone (10 nM) and testosterone (100 nM) treatment of VSMCs resulted in the synthesis of biglycan and decorin that showed reduced electrophoretic mobility by SDS-PAGE, indicating an increase in GAG length. The effect of testosterone treatment on [35S]sulfate incorporation and GAG length was reversed by pretreatment of VSMCs with flutamide (1 ␮M), an androgen receptor antagonist. Proteoglycans from VSMCs treated with testosterone showed 11% (P < 0.01) higher binding capacity to LDL compared with proteoglycans from untreated cells. These results suggest a possible proatherogenic action of androgens through an elongation of GAG chains on proteoglycans in an androgen receptor-dependent manner. (Endocrinology 146: 2085–2090, 2005)

drogens (8). Thus, the role of androgens in atherogenesis remains unclear. The response to retention hypothesis proposes that the intramural retention and accumulation of atherogenic lipoproteins by extracellular matrix, mainly proteoglycans, is a key event in the initiation and development of atherosclerotic lesions (9, 10). Low density lipoproteins (LDL) bind with high affinity to proteoglycans in the arterial wall (11). After significant retention of LDL-proteoglycan complexes, lipoprotein oxidation and cellular chemotaxis (such as smooth muscle cell migration, monocyte adhesion, and foam cell formation) follows, leading to lesion development (9). Proteoglycans are complex macromolecules that are composed of a core protein to which one or more glycosaminoglycan (GAG) chains are linked (12). The major proteoglycans synthesized by vascular smooth muscle cells (VSMC) and found in the vascular wall are versican, perlecan, biglycan, and decorin (13, 14). Proteoglycans are present within the normal arterial wall, but the proportions and structural properties of each type are altered at different stages of the atherogenic process (15–18). A major structural change is the length of GAG chains on proteoglycans (19). Established atherogenic factors, such as TGF-␤1 (20), angiotensin II (21), oxidized LDL (22), and free fatty acids (23), all increase GAG length, leading to increased LDL binding. The potential effects of androgens on GAG elongation processes are unknown. Recent evidence indicates the existence of androgen recep-

2085

2086

Endocrinology, April 2005, 146(4):2085–2090

tors in VSMCs (24); however, the biological role of androgens in proteoglycan biosynthesis and the interaction with LDL have not been investigated. Therefore, we evaluated whether androgens 1) stimulate proteoglycan biosynthesis, 2) alter proteoglycan GAG length, 3) affect proteoglycan biosynthesis in the presence of flutamide, a testosterone receptor antagonist, and 4) affect proteoglycans and their binding to LDL. Materials and Methods Materials Androstenedione, dihydrotestosterone, testosterone, and [3H]testosterone were obtained from Sigma-Aldrich Corp. (St. Louis, MO). Carrierfree sodium [35S]sulfate and [35S]methionine/cysteine were purchased from ICN Biomedicals (Irvine, CA). Chromatography paper, 3MM grade, was obtained from Whatman International Ltd. (Kent, UK). Instagel Plus scintillation cocktail was obtained from Packard (Groningen, The Netherlands). All other chemicals were of the highest grade commercially available and were purchased from Sigma-Aldrich Corp.

Human VSMC culture Human VSMCs were isolated from internal mammary arteries obtained at the time of cardiac bypass surgery at Alfred Hospital (Melbourne, Australia). The human ethics committee of Alfred Hospital approved the acquisition of tissue. The medial layer of the vessel was removed in strips under a dissecting microscope and held under a sterile glass coverslip (9 ⫻ 22 mm) in a 90-mm dish containing DMEM with 15 mm glucose and supplemented with 10% fetal bovine serum (FBS). The tissue was incubated at 37 C in a humidified atmosphere with 5% CO2 (Forma Scientific, Marietta, OH). The medium was changed two or three times per week for 2 wk. Cells that had migrated and proliferated from the tissue were passaged (1:3 split) at wk 3– 4. Stock cells from passages 5–7 were stored in liquid nitrogen until required. Eighth through 12th passages were used in the present study. VSMCs were characterized as previously described (25).

Androgen receptor binding assay Confluent VSMC cells were established in six-well plates (⬃1.04 ⫻ 106 cells/well). Cells were incubated with [3H]testosterone (0.3–5 nm), cold testosterone (1 ␮m), and triamcinolone (1 ␮m) for 90 min. After extensive washing with PBS, cells were transferred into scintillation vials with scintillation liquid and counted for 3H on a liquid scintillation counter.

Proteoglycan biosynthesis Human VSMCs were cultured at a density of 4 ⫻ 104 cells/well in 24-well plates in DMEM containing 5 mm glucose and 10% FBS and were allowed to grow to confluence. Cells were serum-deprived in DMEM containing 0.1% FBS for 48 h and treated with androstenedione (1–100 nm), dihydrotestosterone (DHT; 0.1–10 nm), and testosterone (T; 1–100 nm) for 40 h. Cells were metabolically labeled with [35S]sulfate (50 ␮Ci/well) to assess GAG synthesis or [35S]methionine/cysteine (10 ␮Ci/ well) to measure proteoglycan core protein for 40 h. To investigate whether the effect of T was through the androgen receptor, VSMCs were treated with 1 ␮m flutamide (an androgen receptor antagonist) 4 h before T treatment. To quantify total GAG synthesis and core protein, 50-␮l aliquots of the medium were spotted in duplicate on 30 ⫻ 25-mm rectangles of Whatman chromatography paper. The paper was washed five times in 1% cetylpyridinium chloride (CPC) and counted as previously described (26). Parallel plates were established to assess cell number using a Coulter particle counter and size analyzer (Coulter Corp., Miami, FL).

Electrophoretic mobility of proteoglycans by SDS-PAGE Econocolumns were filled with 50% (vol/vol) diethylaminoethylSephacel (500 ␮l) in low salt buffer containing 8 m urea, 0.25 m NaCl, 0.02 m disodium EDTA, and 0.5% Triton X-100. The columns were washed with low salt buffer, and the media of three identical treatments were

Hashimura et al. • Androgens and Vascular Proteoglycans

pooled and applied to the column. The columns were washed five times with low salt buffer to remove nonproteoglycan-associated radioactivity. Fractions were collected with high salt buffer containing 8 m urea, 3 m NaCl, 0.02 m disodium EDTA, and 0.5% Triton X-100. An aliquot from each fraction was counted. Samples were prepared for SDS-PAGE. Chondroitin sulfate (250 ␮g/ml) was added as the cold carrier. An aliquot of a cold ethanol solution containing 95% ethanol and 1.3% potassium acetate was added to each sample to precipitate proteoglycans and GAG chains. After brief vortexing, the samples were placed at ⫺20 C for 1 h. The samples were centrifuged (14 ⫻ 103 rpm) for 10 min at 4 C, the supernatant was discarded, and the pellets were resuspended in high quality water and precipitated with cold ethanol solution. After brief vortexing, the samples were placed at ⫺20 C for 1 h. The samples were centrifuged (14 ⫻ 103 rpm) for 10 min at 4 C. The supernatant was discarded, and the pellets were left to air-dry before resuspending them in 8 m urea buffer. SDS-PAGE was prepared on 4 –13% linear gradient slab gels with a 3.5% stacking gel. For estimation of the apparent relative mass of proteoglycans, 14C-labeled high molecular weight standards (Invitrogen Life Technologies, Inc., Grand Island, NY) were run in separate lanes. The labeled proteoglycans and standard proteins were exposed (3 d) to a film imaging plate (Fuji Photo Film Co., Tokyo, Japan), and the imaging plate was scanned by the Bio-Imaging analyzer BAS-1000 MacBas (Fuji Photo Film Co.).

LDL binding capacity LDL was isolated from human plasma as described previously (27). In brief, LDL (␳ ⫽ 1.019 –1.063 g/ml) was separated from normal human plasma by ultracentrifugation and purified by sequential density gradient ultracentrifugation. Purified LDL was dialyzed against 0.01 m PBS and 100 ␮m EDTA for 2–3 d. LDL affinity columns were prepared and used as previously described (22). LDL (5 mg/ml) was bound to 1 ml cyanogen bromideactivated Sepharose (Amersham Biosciences, Arlington Heights, IL) by a routine technique. Heparin (50 ␮g/ml) was included during the coupling step to protect the proteoglycan-binding regions of apolipoprotein B from being blocked during the reaction with activated gel. LDLSepharose was washed with 2 m NaCl (10 vol) to dissociate heparin from the covalently bound LDL, and the product was equilibrated in buffer (10 mm HEPES, 20 mm NaCl, and 250 ␮m butylated hydroxytoluene). Radiolabeled ([35S]SO4) samples were dialyzed against 10 mm HEPES, 20 mm NaCl, and 250 ␮m butylated hydroxytoluene. Samples from control and T-treated cells were applied to individual LDL-Sepharose columns (1 ml), and the column flow-through was collected. The columns were washed five times with 10 ml 10 mm HEPES, 20 mm NaCl, and 250 ␮m butylated hydroxytoluene. The bound material then was eluted with 1 m NaCl. Five 2-ml fractions were collected, and an aliquot of each fraction was counted for 35S on a liquid scintillation counter. The amount of radioactivity eluted from the column was expressed as a percentage of the total applied to the column. The control was set at 100% LDL binding capacity.

Statistical analysis Data are expressed as the mean ⫾ sem, and all graphics were created using Fig. P version 2.98 (Biosoft, Cambridge, UK). The T receptor binding curve and the Scatchard plot were generated using the one-site ligand binding and linear regression equations, respectively, in Fig. P. The concentration response of androgens and the effect of flutamide were assessed using a one-way ANOVA. The statistical significance for LDL binding capacity was assessed by a t test with unequal variances (df ⫽ 2.8) using Fig. P. P ⬍ 0.05 and P ⬍ 0.01 were considered significant.

Results

Androgen receptor number was quantitated by radioligand assay using [3H]T. Specific binding of [3H]T was shown in human VSMCs at a density of 2374 ⫾ 190 sites/cell, with a Kd of 0.05 nm (Fig. 1). These data indicate that the human VSMCs used in this study possess abundant androgen receptors.


FIG. 1. Human VSMCs express T receptors. Confluent VSMCs were incubated with [3H]T (0.3–5.0 nM) in the presence of nonradiolabeled T (1 ␮M). After extensive washing, the amount of bound [3H]T was assessed. Nonlinear representation of the amount of bound T with a Scatchard plot transformation (inset) is shown. The binding data showed that there were 33.7 fmol/1 ⫻ 106 cells or 20,266 sites/cell, and the Scatchard plot generated a maximum binding value of 35 fmol and a Kd of 0.44 nM.

Treatment of VSMCs with androstenedione (100 nm) increased [35S]sulfate incorporation into GAGs by 24.8% (P ⬍ 0.05) compared with untreated cells (Fig. 2A). Androstenedione (100 nm) treatment of VSMCs did not alter the electrophoretic mobility of radiolabeled proteoglycans corresponding to biglycan and decorin (Fig. 2B) (13, 14). DHT (10 nm) treatment of VSMCs increased [35S]sulfate incorporation into GAGs by 22% (P ⬍ 0.05) compared with untreated cells (Fig. 3, A). Proteoglycans from VSMCs treated with DHT (10 nm) showed reduced electrophoretic mobility compared with proteoglycans from untreated cells (Fig. 3B). These data indicate that DHT treatment of VSMCs increases proteoglycan GAG length. T (100 nm) treatment of VSMCs increased [35S]sulfate incorporation into GAGs by 32.5% (P ⬍ 0.05) compared with untreated cells (Fig. 4A). Proteoglycans from VSMCs treated with T (100 nm) showed a reduction in electrophoretic mobility by SDS-PAGE (Fig. 4B). To determine whether the effects on [35S]sulfate incorporation into GAGs and GAG length were due to T stimulation of the T receptor, we used a competitive T receptor antagonist, flutamide, in the presence of T. Flutamide was used at 1 ␮m because this concentration could completely block T (100 nm)-induced DNA synthesis (data not shown). Human FIG. 2. Treatment of VSMCs with androstenedione increases [35S]sulfate incorporation into proteoglycans. Human VSMCs were treated with androstenedione (1–100 nM) and metabolically labeled with [35S]sulfate. A, Proteoglycans secreted into the culture medium were assessed by the CPC precipitation assay. Data were expressed as counts per minute per cell and then normalized to basal values. Results are the mean ⫾ SEM from three experiments performed in triplicate. *, P ⬍ 0.05. B, Secreted proteoglycans from cells treated with or without androstenedione (AD; 100 nM) were isolated and separated by SDS-PAGE.


2087

VSMCs pretreated with flutamide (1 ␮m) and stimulated with T (100 nm) showed a 23% (P ⬍ 0.01) reduction in [35S]sulfate incorporation into GAGs compared with cells treated with T alone (Fig. 4C). Additionally, human VSMCs pretreated with flutamide (1 ␮m) and stimulated with T (100 nm) showed an increase in the electrophoretic mobility of biglycan and decorin compared with cells treated with T alone (Fig. 4B). These data suggest that T treatment increases GAG length, represented by an increase in [35S]sulfate incorporation into GAGs, and these effects are probably dependent on the T receptor. To exclude the possibility that the effect of T treatment to increase [35S]sulfate incorporation into GAGs was not due to an increase in proteoglycan core protein synthesis, human VSMCs were treated with T in the presence and absence of flutamide and metabolically labeled with [35S]methionine/ cysteine. Treatment of human VSMCs with T did not alter the incorporation of [35S]methionine/cysteine into proteoglycan core protein compared with untreated cells, suggesting that the increase in [35S]sulfate incorporation into GAGs was not due to an increase in core protein (Fig. 4D). Flutamide treatment of VSMCs in the presence of T did not alter [35S]methionine/cysteine incorporation into proteoglycan core protein compared with cells treated with T alone, indicating that the decrease in 35S incorporation into GAGs was not due to a decrease in core protein synthesis (Fig. 4D). Treatment of VSMCs with factors that increase proteoglycan GAG length, such as TGF-␤1 (20), oxidized LDL (22) or angiotensin II (21), has been correlated with an increase in LDL binding affinity. Treatment of VSMCs with testosterone reduced the electrophoretic mobility of proteoglycans suggesting an increase in GAG chain length. We tested whether or not proteoglycans from VSMCs treated with testosterone showed increased binding to LDL using an LDL affinity column. Human VSMCs treated with testosterone (100 nm) showed 11% (P ⬍ 0.01) more binding to LDL compared with proteoglycans from untreated cells (Fig. 5). Discussion

Androgen treatment of human VSMCs increased proteoglycan synthesis and GAG length, and this was reflected as an increase in binding to LDL. Treatment of VSMCs with androstenedione, DHT, and T showed concentration-related stimulation of proteoglycan synthesis. Our data showed that

2088



FIG. 3. Treatment of VSMCs with DHT increases [35S]sulfate incorporation into proteoglycans and increases GAG length. Human VSMCs were treated with DHT (0.1–10 nM) and metabolically labeled with [35S]sulfate. A, Proteoglycans secreted into the culture medium were assessed by the CPC precipitation assay. Data were expressed as counts per minute per cell and then normalized to basal values. Results are the mean ⫾ SEM from three experiments performed in triplicate. *, P ⬍ 0.05. B, Secreted proteoglycans from cells treated with or without DHT (10 nM) were isolated and separated by SDS-PAGE.

the effect of T treatment to increase proteoglycan synthesis was associated with an increase in GAG length, and these proteoglycans had a higher LDL-binding capacity than proteoglycans from untreated cells. The effect of T on vascular proteoglycans was dependent on the T receptor, because the response was inhibited by the receptor antagonist, flutamide. With a radiolabeling time of 24 h being required to assess proteoglycan biosynthesis, it is difficult to speculate on the postreceptor events, and thus it is possible that nongenomic as well as genomic effects may have contributed to the effects of the androgens on proteoglycan biosynthesis. These data

on androgen-induced GAG elongation and increased LDL binding are consistent with the findings of others showing that VSMCs treated with TGF-␤1 (20), angiotensin II (21), oxidized LDL (22), or nonesterified free fatty acids (23) synthesize proteoglycans that have longer GAG chains, and these proteoglycans show enhanced binding to LDL. The concentrations of the androgens used in our study are high relative to physiological levels. However, several considerations are appropriate: the agents in culture may be degraded over the 24-h period of the experiments, and the mean time-averaged concentration to which the cells are

FIG. 4. Treatment of VSMCs with T increases [35S]sulfate incorporation into proteoglycans and increases GAG length. Human VSMCs were treated with T (1–100 nM) and metabolically labeled with [35S]sulfate. A, Proteoglycans secreted into the culture medium were assessed by the CPC precipitation assay. Data were expressed as counts per minute per cell and then normalized to basal values. Results are the mean ⫾ SEM from three experiments performed in triplicate. *, P ⬍ 0.05. B, Secreted proteoglycans from cells treated with or without T (100 nM) and from cells treated with T (100 nM) and flutamide (1 ␮M) were isolated and separated by SDS-PAGE. C, Human VSMCs treated with T (100 nM) in the presence or absence of flutamide (1 ␮M) and metabolically labeled with [35S]sulfate were assessed for proteoglycans by the CPC precipitation assay. Data were expressed as counts per minute per cell and then normalized to basal values. Results are the mean ⫾ SEM from three experiments performed in triplicate. **, P ⬍ 0.01; ***, P ⬍ 0.001. D, Human VSMCs treated with T (100 nM) in the presence or absence of flutamide (1 ␮M) and metabolically labeled with [35S]methionine/cysteine were assessed for proteoglycan core protein by the CPC precipitation assay. Data were expressed as counts per minute cell and then normalized to basal values. Results are the mean ⫾ SEM from three experiments performed in triplicate.


FIG. 5. T-treated VSMCs synthesize proteoglycans with increased binding capacity to LDL. Human VSMCs were treated with and without T (100 nM) and metabolically labeled with [35S]sulfate. Radiolabeled proteoglycans were applied to separate LDL affinity columns and eluted with 1 M NaCl. The amount of radioactivity eluted from the column was expressed as a percentage of the amount applied to the column. The control was set at 100% LDL binding capacity and proteoglycan-LDL binding from T-treated cells was expressed as a percentage of the control. Data are the mean ⫾ SEM of three experiments. **, P ⬍ 0.01.

exposed is lower, but to an uncertain extent, than the nominal concentrations to which the cells were initially exposed. Furthermore, issues of tissue concentration and protein binding are relevant for these lipophilic agents. The effects observed are appreciable and statistically significant as well as being reversed by an antagonist. Thus, we would contend that the description of highly relevant, albeit small, effects may point to potential mechanisms operating in a model of disease where the underlying disease process develops over decades. Obviously additional studies would be required to fully evaluate and appreciate the biological and potentially pathological significance of these observations. Recent studies have highlighted the importance of proteoglycans in atherosclerosis (10). Biglycan contributes to the pathogenesis of atherosclerosis by trapping lipoproteins in the artery wall (18). Estrogens, but not androgens, reduce myointimal response to balloon injury in rat (28) or mouse (29) carotid artery, indicating different effects on the response to remodeling of the vascular wall. There are several lines of evidence that show that T or DHT stimulates proteoglycan biosynthesis in the prostate (30), testis (31), cartilage (32), and submandibular glands (33); however, the mechanism(s) by which atherogenic effects of androgens are mediated in VSMCs is unclear. One possible explanation for the atherogenic effects of androgens involves direct effects on atherogenesis and progression at the level of the arterial intima. We excluded the possibility that the increase in proteoglycan synthesis was due to an increase in core protein synthesis and, hence, additional GAG initiation and elongation sites, because total core protein synthesis did not change with T treatment. However, the increase in proteoglycan synthesis may be associated with cellular proliferation. We observed that T treatment of VSMCs increases DNA synthesis in an androgen receptor-dependent manner (Hashimura, K., and


2089

P. J. Little, unpublished observations). Proteoglycans isolated from proliferating arterial SMCs have longer GAG chains and bind with greater affinity to LDL than proteoglycans isolated from quiescent arterial SMCs (34). These data taken together with those from the current study show that T modulates not only DNA synthesis, but also proteoglycan synthesis, through the androgen receptor. Additive or synergistic effects of T on vascular cell growth and proteoglycans may enhance lipid retention in the vessel wall. Differentiated macrophages synthesize heparan sulfate proteoglycans and oversulfated chondroitin sulfate proteoglycans, allowing macrophages to interact with lipoprotein lipase, and this facilitates LDL uptake (35). Furthermore, it has been reported that proteoglycan derived from bovine aorta forms insoluble complexes with LDL, but desulfation of the proteoglycan by solvolysis in dimethylsulfoxide inhibited complex formation with LDL (36). Taken together, there is a possibility that the degree of sulfation of proteoglycans may modulate the charge density for specific interactions with LDL. Sulfation of proteoglycans was not addressed in this study, because we have shown that androgen treatment increases the length of the GAG chains on the proteoglycans. However, in addition to an elongation in GAG length, the increase in proteoglycan synthesis after treatment with androgens may reflect the production of highly sulfated proteoglycans, and if these observations were reproduced in vivo, these chemical changes may also modulate lipid retention in the vascular wall. Men at any age are at higher risk for CAD than similarly aged women (1). Although the effects observed in the current study are modest, there is a possibility that prolonged exposure to higher androgen concentrations could render men susceptible to atherosclerosis. Anabolic androgens accelerate atherosclerosis in primate models (37), markedly reduce high density lipoprotein cholesterol, and induce endothelial dysfunction in male body builders (38). Risk factors for cardiovascular disease interact negatively in a synergistic manner; thus, a small change in proteoglycans induced by elevated androgen levels in addition to the presence of one or more coronary risk factors may render males more susceptible to atherosclerosis than females. Of interest, Ng et al. (39) recently showed that androgens up-regulate atherosclerosisrelated genes in macrophages from males, but not females, confirming the complexity of hormone-vascular interactions that could account for gender-based differences in atherosclerosis. Our findings show that androgen-induced changes in vascular proteoglycans form part of this complex interaction. In summary, these studies have shown that androgens stimulate proteoglycan biosynthesis and elongate GAG chains with higher affinity binding to LDL through an androgen receptor-dependent pathway. This provides a new potential mechanism, if manifest in vivo, for supporting a proatherogenic effect of androgens on human VSMC proteoglycan biosynthesis and may provide new opportunities to explore the complex and unresolved issue of hormones and cardiovascular disease.

2090



Acknowledgments

TN, Chait A 1998 Comparison of apolipoprotein and proteoglycan deposits in human coronary atherosclerotic plaques: colocalization of biglycan with apolipoproteins. Circulation 98:519 –527 Ballinger M, Nigro J, Frontanilla K, Dart A, Little P 2004 Regulation of glycosaminoglycan structure and atherogenesis. Cell Mol Life Sci 61:1296 – 1306 Little PJ, Tannock L, Olin KL, Chait A, Wight TN 2002 Proteoglycans synthesized by arterial smooth muscle cells in the presence of transforming growth factor-␤1 exhibit increased binding to LDLs. Arterioscler Thromb Vasc Biol 22:55– 60 Figueroa JE, Vijayagopal P 2002 Angiotensin II stimulates synthesis of vascular smooth muscle cell proteoglycans with enhanced low density lipoprotein binding properties. Atherosclerosis 162:261–268 Chang MY, Potter-Perigo S, Tsoi C, Chait A, Wight TN 2000 Oxidized low density lipoproteins regulate synthesis of monkey aortic smooth muscle cell proteoglycans that have enhanced native low density lipoprotein binding properties. J Biol Chem 275:4766 – 4773 Olsson U, Bondjers G, Camejo G 1999 Fatty acids modulate the composition of extracellular matrix in cultured human arterial smooth muscle cells by altering the expression of genes for proteoglycan core proteins. Diabetes 48: 616 – 622 Oparil S, Levine R, Chen Y 1996 Sex hormones and the vasculature. In: Sowers J, ed. Endocrinology of the vasculature. Totowa, NJ: Humana Press; 225–227 Sudhir K, Hashimura K, Bobik A, Dilley RJ, Jennings GL, Little PJ 2001 Mechanical strain stimulates a mitogenic response in coronary vascular smooth muscle cells via release of basic fibroblast growth factor. Am J Hypertens 14:1128 –1134 Nigro J, Dilley RJ, Little PJ 2002 Differential effects of gemfibrozil on migration, proliferation and proteoglycan production in human vascular smooth muscle cells. Atherosclerosis 162:119 –129 Heinecke JW, Baker L, Rosen H, Chait A 1986 Superoxide-mediated modification of low density lipoprotein by arterial smooth muscle cells. J Clin Invest 77:757–761 Chen SJ, Li H, Durand J, Oparil S, Chen YF 1996 Estrogen reduces myointimal proliferation after balloon injury of rat carotid artery. Circulation 93:577–584 Sullivan Jr TR, Karas RH, Aronovitz M, Faller GT, Ziar JP, Smith JJ, O’Donnell Jr TF, Mendelsohn ME 1995 Estrogen inhibits the response-toinjury in a mouse carotid artery model. J Clin Invest 96:2482–2488 Terry DE, Clark AF 1996 Influence of testosterone on chondroitin sulphate proteoglycan in the rat prostate. Biochem Cell Biol 74:645– 651 Raychoudhury SS, Blackshaw AW, Irving MG 1993 Rat Sertoli cell extracellular matrix regulates glycosaminoglycan synthesis by peritubular myoid cells in vitro. Mol Reprod Dev 35:151–158 Franchimont P, Bassleer C 1991 Effects of hormones and local growth factors on articular chondrocyte metabolism. J Rheumatol 27(Suppl):68 –70 Kofoed JA, Tumilasci OR, Curbelo HM, Fernandez Lemos SM, Arias NH, Houssay AB 1990 Effect of testosterone on proteoglycans in the submandibular glands of the rat. J Biol Buccale 18:255–259 Camejo G, Fager G, Rosengren B, Hurt-Camejo E, Bondjers G 1993 Binding of low density lipoproteins by proteoglycans synthesized by proliferating and quiescent human arterial smooth muscle cells. J Biol Chem 268:14131–14137 Edwards IJ, Xu H, Obunike JC, Goldberg IJ, Wagner WD 1995 Differentiated macrophages synthesize a heparan sulfate proteoglycan and an oversulfated chondroitin sulfate proteoglycan that bind lipoprotein lipase. Arterioscler Thromb Vasc Biol 15:400 – 409 Vijayagopal P, Srinivasan SR, Radhakrishnamurthy B, Berenson GS 1981 Interaction of serum lipoproteins and a proteoglycan from bovine aorta. J Biol Chem 256:8234 – 8241 Obasanjo IO, Clarkson TB, Weaver DS 1996 Effects of the anabolic steroid nandrolone decanoate on plasma lipids and coronary arteries of female cynomolgus macaques. Metabolism 45:463– 468 Ebenbichler CF, Sturm W, Ganzer H, Bodner J, Mangweth B, Ritsch A, Sandhofer A, Lechleitner M, Foger B, Patsch JR 2001 Flow-mediated, endothelium-dependent vasodilatation is impaired in male body builders taking anabolic-androgenic steroids. Atherosclerosis 158:483– 490 Ng MK, Quinn CM, McCrohon JA, Nakhla S, Jessup W, Handelsman DJ, Celermajer DS, Death AK 2003 Androgens up-regulate atherosclerosisrelated genes in macrophages from males but not females: molecular insights into gender differences in atherosclerosis. J Am Coll Cardiol 42:1306 –1313

Received September 20, 2004. Accepted January 7, 2005. Address all correspondence and requests for reprints to: Dr. Peter J. Little, Cell Biology of Diabetes Laboratory, Baker Heart Research Institute, St. Kilda Road Central, P.O. Box 6492, Melbourne, Victoria 8008, Australia. E-mail: [email protected]. This work was supported by the Alfred Hospital Foundation (to P.J.L.). J.N. was supported by an Australian postgraduate award.

19.

20.

21.

References 1. Gorodeski G, Utian W 1994 Epidemiology and risk factors of cardiovascular disease in postmenopausal women. In: Lobo R, ed. Treatment of the postmenopausal woman: basic and clinical aspects. New York: Raven Press; 199 – 221 2. Vermeulen A, Verdonck L, Kaufman JM 1999 A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 84:3666 –3672 3. Somjen D, Kohen F, Jaffe A, Amir-Zaltsman Y, Knoll E, Stern N 1998 Effects of gonadal steroids and their antagonists on DNA synthesis in human vascular cells. Hypertension 32:39 – 45 4. Ling S, Dai A, Williams MR, Myles K, Dilley RJ, Komesaroff PA, Sudhir K 2002 Testosterone (T) enhances apoptosis-related damage in human vascular endothelial cells. Endocrinology 143:1119 –1125 5. Hutchison SJ, Sudhir K, Chou TM, Sievers RE, Zhu BQ, Sun YP, Deedwania PC, Glantz SA, Parmley WW, Chatterjee K 1997 Testosterone worsens endothelial dysfunction associated with hypercholesterolemia and environmental tobacco smoke exposure in male rabbit aorta. J Am Coll Cardiol 29:800 – 807 6. Adams MR, Williams JK, Kaplan JR 1995 Effects of androgens on coronary artery atherosclerosis and atherosclerosis-related impairment of vascular responsiveness. Arterioscler Thromb Vasc Biol 15:562–570 7. Chou TM, Sudhir K, Hutchison SJ, Ko E, Amidon TM, Collins P, Chatterjee K 1996 Testosterone induces dilation of canine coronary conductance and resistance arteries in vivo. Circulation 94:2614 –2619 8. Bernini GP, Moretti A, Sgro M, Argenio GF, Barlascini CO, Cristofani R, Salvetti A 2001 Influence of endogenous androgens on carotid wall in postmenopausal women. Menopause 8:43–50 9. Williams KJ, Tabas I 1995 The response-to-retention hypothesis of early atherogenesis. Arterioscler Thromb Vasc Biol 15:551–561 10. Skalen K, Gustafsson M, Rydberg EK, Hulten LM, Wiklund O, Innerarity TL, Boren J 2002 Subendothelial retention of atherogenic lipoproteins in early atherosclerosis. Nature 417:750 –754 11. Vijayagopal P, Figueroa JE, Fontenot JD, Glancy DL 1996 Isolation and characterization of a proteoglycan variant from human aorta exhibiting a marked affinity for low density lipoprotein and demonstration of its enhanced expression in atherosclerotic plaques. Atherosclerosis 127:195–203 12. Wight TN, Heinegard DK, Hascall VC 1991 Proteoglycans: structure and function. In: Hay ED, ed. Cell biology of extracellular matrix. 2nd ed. New York: Plenum Press; 45–77 13. Schonherr E, Jarvelainen HT, Sandell LJ, Wight TN 1991 Effects of plateletderived growth factor and transforming growth factor-␤1 on the synthesis of a large versican-like chondroitin sulfate proteoglycan by arterial smooth muscle cells. J Biol Chem 266:17640 –17647 14. Schonherr E, Jarvelainen HT, Kinsella MG, Sandell LJ, Wight TN 1993 Platelet-derived growth factor and transforming growth factor-␤1 differentially affect the synthesis of biglycan and decorin by monkey arterial smooth muscle cells. Arterioscler Thromb 13:1026 –1036 15. Evanko SP, Raines EW, Ross R, Gold LI, Wight TN 1998 Proteoglycan distribution in lesions of atherosclerosis depends on lesion severity, structural characteristics, and the proximity of platelet-derived growth factor and transforming growth factor-␤. Am J Pathol 152:533–546 16. Riessen R, Isner JM, Blessing E, Loushin C, Nikol S, Wight TN 1994 Regional differences in the distribution of the proteoglycans biglycan and decorin in the extracellular matrix of atherosclerotic and restenotic human coronary arteries. Am J Pathol 144:962–974 17. Camejo G, Hurt-Camejo E, Olsson U, Bondjers G 1993 Proteoglycans and lipoproteins in atherosclerosis. Curr Opin Lipidol 4:385–391 18. O’Brien KD, Olin KL, Alpers CE, Chiu W, Ferguson M, Hudkins K, Wight

22.

23.

24. 25.

26.

27.

28. 29.

30. 31.

32. 33.

34.

35.

36.

37.

38.

39.

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.