Having previously demonstrated that glucose transporter-4 (GLUT4) expression was reduced in aortas and carotid arteries of deoxycorticosterone acetate (DOCA) salt-hypertensive rats, we hypothesized that troglitazone (TG), through activation of peroxisome proliferator-activated receptor-γ (PPAR-γ), would stabilize GLUT4 expression and possibly preserve the differentiated phenotype in vascular smooth muscle cells. In DOCA salt-hypertensive rats treated with TG (100 mg/day), there was a significant (P < 0.001) decrease in systolic blood pressure (BP; 149.9 ± 4.4 mmHg) compared with the untreated DOCA salt-hypertensive rats (202.2 ± 10.34 mmHg). Separate trials with rosiglitazone (RS; 3 mg/day) demonstrated a significant (P < 0.001) decrease in BP (DOCA salt, 164.2 ± 9.8 vs. DOCA-RS, 124.9 ± 3.7 mmHg) comparable to that with TG. Expression of GLUT4, h-caldesmon, and smooth muscle myosin heavy chain SM2 was significantly decreased in aortas of DOCA salt-hypertensive rats and was reversed by TG to levels similar to those in aortas of sham-treated rats. TG (50 μM) induced GLUT4 and h-caldesmon expression in 24-h culture of explanted carotid arteries of DOCA salt-hypertensive rats, and the endogenous PPAR-γ ligand 15-deoxy-Δ12–14-prostaglandin J2 (PGJ2; 20 μM) and TG (50 μM) similarly increased GLUT4, h-caldesmon, and SM2 protein expression in explanted aortas. The expression of activated, phosphorylated Akt was increased by PGJ2 and TG with no significant effect on total Akt levels. Inhibition of phosphorylated Akt expression using the phosphatidylinositol 3-kinase inhibitor LY-294002 (16 μM) abrogated the increased expression of h-caldesmon and SM2. These data demonstrate that PPAR-γ agonists maintain or induce expression of markers of the contractile phenotype independently of their effects on hypertension, and that this effect may be mediated through activation of phosphatidylinositol 3-kinase/Akt.
- peroxisome proliferator-activated receptor-γ
- glucose transporter isoform-4
in hypertension, vascular cells undergo adaptive changes. One of the most marked of these is arterial remodeling characterized by medial hypertrophy due to smooth muscle cell hypertrophy and hyperplasia (24). Although the processes involved in this remodeling remain poorly defined, it is clear that there is a change in the vascular smooth muscle phenotype. In normal, uninjured arteries, the vascular smooth muscle is characterized by a nonproliferating, differentiated, “contractile” phenotype. This phenotype corresponds with the expression of high levels of smooth muscle-specific actin and intermediate filaments. After exposure to elevated blood pressure (BP), there are changes in gene expression both through transcriptional regulation and alternative splicing that lead to a less differentiated “proliferative” phenotype (5, 23). When vascular smooth muscle cells are cultured, they likewise undergo a change in phenotype (29). With increasing time in culture, these cells express fewer of the contractile markers and more of those associated with the proliferative phenotype. We have found that the expression of the facilitative glucose transporter-4 (GLUT4) is decreased in vascular smooth muscle of the deoxycorticosterone acetate (DOCA) salt model of hypertension (2). Furthermore, GLUT4 expression is very low in cultured vascular smooth muscle cells. Thus as in adipose tissue and skeletal muscle, GLUT4 appears to be a marker of the mature phenotype in vascular smooth muscle.
Thiazolidinediones (TZDs) comprise a family of compounds that have various degrees of efficacy in the treatment of type 2 diabetes and insulin-resistant hypertension (12). In these cases, TZDs increase target-tissue sensitivity to insulin and also lower BP. Although there may be an association between these latter two phenomena, this is not always the case. In one report, the TZD pioglitatzone lowered BP in both insulin-resistant and insulin-sensitive hypertensive models (12). Insulin sensitivity was increased only in the insulin-resistant model with treatment, and the insulin sensitizer metformin had no effect on BP in either model. These findings demonstrate that TZDs have effects not directly related to improvement of total-body insulin sensitivity.
Peroxisome proliferator-activated receptor-γ (PPAR-γ), which is a member of the superfamily of nuclear receptor ligand-activated transcription factors that regulate gene expression (30), is expressed in vascular smooth muscle in vivo and in culture (14). TZDs bind with high affinity to PPAR-γ and exert pleiotropic responses (27) that are presumably mediated by PPAR-γ. TZDs increase glucose transporter expression and glucose uptake (11) and inhibit proliferation and migration (15, 17, 19) of vascular smooth muscle cells. The putative endogenous ligand for PPAR-γ, 15-deoxy-Δ12–14-prostaglandin J2 (PGJ2), not only inhibits vascular smooth muscle proliferation but also increases mRNA expression for SM1 and induces mRNA expression of SM2, each of which is a splice variant of the smooth muscle myosin heavy chain (SM-MHC) differentiation marker (18).
We previously demonstrated that GLUT4 expression and glucose uptake are decreased in DOCA salt hypertension (2). In addition, increased vascular reactivity was associated with decreased glucose uptake (2). Because TZDs have been demonstrated to increase GLUT4 expression in other tissues that express this transporter, we hypothesized that troglitazone (TG) would increase GLUT4 expression and glucose uptake and decrease hypertension possibly by maintaining the mature vascular smooth muscle phenotype. The results reported herein demonstrate that TZDs are effective in restoring GLUT4 expression and glucose uptake and in lowering BP. In addition, TG and PGJ2 increase the expression of several markers of the mature smooth muscle phenotype. This latter effect is independent of any ameliorative effects on BP.
METHODS AND MATERIALS
DOCA salt-hypertensive rats were generated as described previously (4). The University of Michigan Committee on Use and Care of Animals approved all procedures using rats. All work was conducted in accord with the principles and procedures outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Adult male Sprague-Dawley rats (body wt, 250–300 g) were anesthetized with pentobarbital sodium (50 mg/kg ip), and their left kidneys were removed. A DOCA (200 mg/kg)-impregnated Silastic implant was placed subcutaneously behind the skull in 50% of the rats. After surgery, DOCA salt-treated rats were given salt water (1% NaCl with 0.2% KCl), and sham-treated rats received tap water. TG and rosiglitazone (RS), which were gifts of Pfizer and Glaxo-SmithKline, respectively, were mixed into ground diet and given to both sham- and DOCA salt-treated rats at 100 and 3 mg/day, respectively. Systolic BP values in conscious rats were measured by the tail-cuff method using a pneumatic transducer. Rats remained on therapy for 4 wk before experimentation. Rats were anesthetized with pentobarbital sodium (50 mg/kg ip), and both diaphragms were cut. The aortas (from the aortic arch to just above the diaphragm) and carotid arteries were removed, snap-frozen in liquid nitrogen, and stored at −80°C.
Frozen vascular samples were ground in a chilled mortar, suspended in the presence of SDS-PAGE loading buffer (that contained 125 mmol/l Tris·HCl, pH 6.8, 4% SDS, 20% glycerol, 100 mM PMSF, 10 mg/ml aprotinin, 1 mg/ml leupeptin, and 1 mg/ml pepstatin A), and sonicated. Protein concentrations were determined by bicinchoninic acid assay (Pierce; Rockford, IL). Lysates were evaluated on 6 or 10% SDS-PAGE and immunoblotted with antisera to GLUT4, calponin, h-caldesmon, SM-MHCs SM1 and SM2, Akt, and phosphorylated Akt (pAkt) as previously described (2). The constitutively expressed β-tubulin was used as a correction for loading. Antibodies to calponin and h-caldesmon were from Sigma (St. Louis, MO), those to SM1 and SM2 were kindly provided by Dr. Robert Adelstein (National Institutes of Health; Bethesda, MD), those to Akt and pAkt were from Cell Signaling (Beverly, MA), those to proliferating cell nuclear antigen (PCNA) and tubulin were from Upstate (Charlottesville, VA), and that to GLUT4 was kindly provided by Dr. Maureen Charron (Albert Einstein College of Medicine; New York, NY).
Uptake of 2-deoxyglucose.
Uptake of the glucose analog 2-deoxyglucose (2-DOG) was performed as previously described (2). Briefly, aortas were removed from the rats, cleaned of all adipose tissue, opened longitudinally, cut into equal halves, and placed in Krebs-Ringer phosphate with 1% (wt/vol) BSA that contained (in mM) 128 NaCl, 5.2 KCl, 1.33 CaCl, and 2.64 MgSO4, 16.5 Na2HPO4 for 30 min at 37°C. The aorta halves (for each animal) were then placed in 0.1 mM unlabeled 2-DOG (Sigma) plus 25 μCi/ml 2-[3H]DOG (Amersham) in Krebs-Ringer phosphate solution with 1% (wt/vol) BSA with or without 20 nM cytochalasin B, which is an irreversible inhibitor of glucose transport, at 37°C for 5 min. The aortas were subsequently washed quickly three times with ice-cold Krebs-Ringer phosphate solution that contained 5.5 mg/dl phloretin to quench 2-DOG uptake. The samples were then flash-frozen in liquid nitrogen, ground, and finally lysed in 0.4% (wt/vol) SDS. After centrifugation, determination of protein concentration by bicinchoninic acid assay was performed, and an aliquot of each sample was scintillation counted. The 2-DOG uptake was expressed as picomoles of cytochalasin B-inhibitable 2-DOG uptake per 5 min per 1 mg of protein.
Ex vivo experiments.
Rats were uninephrectomized, implanted with DOCA, and maintained for 4 wk as described (see Animal experiments). At that time, the carotid arteries were carefully removed and cleaned of extraneous tissue and adventitia. Arteries were then placed in DMEM with 10% FBS overnight with or without 50 μM TG. In separate trials, aortas were removed from normal rats, cleaned, and cultured in the presence or absence of 25 μM PGJ2 (Biomol; Plymouth Meeting, PA) or 50 μM PGJ2 with or without the phosphatidylinositol 3-kinase (PI3K) inhibitor LY-294002 (12 μM; Calbiochem; San Diego, CA). The vessels were then flash-frozen and prepared for immunoblotting (see Immunoblot analysis). Some of the sham- and DOCA salt-treated arteries were frozen immediately upon removal from the animals; these served as controls for any effects of culturing on the expression of markers of interest.
Data presentation and statistical evaluation.
Immunoblot band density was determined using NIH Image 1.62 software. Data are expressed as mean percentages of sham-treated or control animals (arbitrarily set at 100) ± SE. Unpaired two-tailed Student's t-test was used to compare results from two populations. ANOVA with Scheffé's post hoc analysis was performed when more than two populations were analyzed. Differences were deemed to be statistically significant when P < 0.05 (P = NS, not significantly different).
Systolic BP, GLUT4 expression, and 2-DOG uptake in DOCA salt-treated rats also treated with TG or RS.
Systolic BP increased significantly after 4 wk of DOCA salt treatment compared with sham-treated animals (Fig. 1). Treatment with TG significantly (P < 0.01) decreased systolic BP in DOCA salt-treated rats (DOCA/TG, 149.9 ± 4.4; DOCA salt, 202.2 ± 10.3 mmHg). TG had no effect on BP in sham-treated rats. In separate trials, systolic BP was significantly (P < 0.01) decreased in RS-treated DOCA salt-treated rats (Fig. 1) compared with DOCA salt treatment alone (124.9 ± 3.7 vs. 164.2 ± 9.9 mmHg, respectively).
We previously reported that GLUT4 protein expression was decreased to 35.0 ± 9.2% (P < 0.05) of sham-treated levels in carotid arteries and aortas of DOCA salt-hypertensive rats with little or no effect on GLUT1 expression (2). In the present study, treatment of DOCA salt-treated rats with TG restored GLUT4 expression to 84.3 ± 11.4% (P = NS) of sham-treatment levels (Fig. 2A). As previously reported, there was no decrease in GLUT4 expression in adipose or skeletal and cardiac muscle in DOCA salt-treated rats (2). However, there was an increase in GLUT4 expression in these tissues (2) and in carotid arteries and aortas of TG-treated sham rats (Fig. 2A). There was no effect of TG on GLUT1 expression (not shown). GLUT4 expression in aortas was also increased in RS-treated DOCA salt-treated rats to 86.7 ± 14.3% (P = NS) of sham-treated levels (Fig. 2B). Both TG and RS increased GLUT4 expression in sham aortas. [Note: for Figs. 2 (A and B) and 3 (A–C), sham vs. sham/TG (or /RS) treatments were run on separate blots from those for the other comparisons. As such, the data in the bar graph for sham/TG (or /RS) treatments are only compared with sham treatment.]
As previously reported, there was a significant decrease in 2-DOG uptake in aortas from DOCA salt-treated compared with sham-treated rats (2). TG significantly (P < 0.05) increased 2-DOG uptake in aortas of DOCA salt-treated rats (Fig. 2C).
Expression of smooth muscle markers in aortas of DOCA salt-treated rats also treated with TG.
Immunoblotting for various markers of the mature smooth muscle phenotype was performed. The expression of h-caldesmon was significantly decreased to 19 ± 5.75% (P < 0.01) of sham-treatment levels in aortas (Fig. 3A) and carotid arteries (not shown) of DOCA salt-treated rats. TG treatment of DOCA salt-treated rats restored h-caldesmon expression to 92.4 ± 14.55% (P = NS) of sham-treatment levels (Fig. 3A). SM-MHC is expressed as the two isoforms SM1 (∼200 kDa) and SM2 (∼204 kDa). SM2 expression was decreased to 63.8 ± 7.6% (P < 0.05) of sham-treatment expression in aortas of DOCA salt-treated rats (Fig. 3B). TG increased SM2 expression in aortas of DOCA salt-treated rats to 93.6 ± 11.1% (P = NS) of sham treatment. Expression of h-caldesmon and SM2 was slightly but not significantly increased in the aortas (see Fig. 2, A and B) and carotid arteries (not shown) of TG-treated sham rats. The expression of the 200-kDa SM1 isoform in aortas was not affected by DOCA salt treatment; however, there was a modest increase in SM1 expression with TG in aortas of both sham- and DOCA salt-treated rats (not shown). There was no effect of DOCA salt treatment on calponin expression; however, TG dramatically increased calponin expression to 368 ± 73.4 and 349 ± 89.3% (P < 0.001) of sham levels in aortas from DOCA salt- and sham/TG-treated rats, respectively (Fig. 3C).
Although DOCA salt hypertension is reported to induce smooth muscle hypertrophy more than hyperplasia in large arteries (9, 26), we explored the effects of TG on proliferation. PCNA expression was not increased in DOCA salt-treated compared with sham-treated aortas (Fig. 3D). There were no effects of TG on PCNA expression in DOCA salt- (Fig. 3D) or sham-treated (not shown) aortas.
Effects of TG on GLUT4 and smooth muscle markers in explanted arteries.
Because TG lowered BP and restored the expression of the markers over the same time course (i.e., 4 wk), it was not clear whether the latter response was by virtue of the effect of TG on the hypertension or independent of it. Carotid arteries were removed from sham- and DOCA salt-treated rats and placed in culture media overnight with or without 50 μM TG. There were no differences in the expression of GLUT4, h-caldesmon, or SM2 between vessels that either had been frozen immediately after removal from the animals (either sham or DOCA salt treated) and those that had been placed in culture media overnight (not shown). Treatment of carotid arteries from DOCA salt-treated rats with TG in this culture system increased GLUT4 (Fig. 4A) to 70.9 ± 6.2% (P = NS) and h-caldesmon (Fig. 4B) to 72.6 ± 17.6 (P = NS) of sham-treatment levels. There were no effects of TG on SM2 or calponin expression during the time course of these experiments (not shown).
Effects of PGJ2 and TG on GLUT4 and smooth muscle marker expression in explanted arteries.
There are no identifiable PPAR-γ response-element consensus sequences in any of the genes under investigation. The serine-threonine protein kinase Akt has been identified as a factor necessary for maintenance of the mature phenotype in smooth muscle (7, 6). Furthermore, pAkt expression is decreased in vessels of DOCA salt-hypertensive rats (20). Therefore, additional trials were performed to elucidate the mechanism of the PPAR-γ agonist effect on GLUT4 and marker expression. Aortas were removed from rats and placed in culture media for 24 h with or without 25 μM PGJ2 or 50 μM TG in the presence or absence of the PI3K inhibitor LY-294002 (12 μM).
Consistent with the effects of TG in explanted carotids, PGJ2 had no effects on calponin expression, and there were no effects of LY-294002 either in combination with PGJ2 (Fig. 5A) or alone (not shown) on calponin expression in explanted aortas. Thus calponin expression demonstrates that 24-h treatment with the PI3K inhibitor was not toxic to the vessels. Activated Akt is phosphorylated on serine-473 by a PI3K-dependent mechanism (3). Therefore, an antibody specific to that species of Akt was used as a measure of the effect of the PPAR-γ agonist on Akt activity. PGJ2 increased pAkt (adjusted for total Akt expression) to 320.9 ± 30.7% (P < 0.01) of control expression (Fig. 5A). The PI3K inhibitor LY-294002 blocked the PGJ2-mediated increase in pAkt expression (Fig. 5A). PGJ2 slightly (P = NS) increased total Akt expression compared with control animals, whereas there were no differences in total Akt expression with LY-294002 either with PGJ2 (Fig. 5A) or alone (not shown). Similar albeit less robust results on pAkt expression were observed with TG treatment of explanted aortas (Fig. 5B). TG increased pAkt expression to 230.3 ± 5.2%, and this stimulation was abrogated by LY-294002.
GLUT4 expression was significantly increased by PGJ2 (181.7 ± 15.5%; P < 0.009; Fig. 6A) or TG (175.6 ± 11.3%; P < 0.02; Fig. 6B) in explanted aortas compared with control animals. Interestingly, LY-294002 increased GLUT4 expression in combination with either PGJ2 (223.8 ± 16.3%; P < 0.0003; Fig. 6A) or TG (203.1 ± 7.9%; P < 0.009; Fig. 6B) and alone (not shown). PGJ2 alone increased h-caldesmon expression 172.3 ± 6.9% (P < 0.03), and this effect was inhibited by LY-294002 (Fig. 7A). h-Caldesmon expression was stimulated by TG by 215.4 ± 17.6% (P < 0.02), and this effect was significantly decreased by LY-294002 (Fig. 7B). SM2 expression in explanted aortas was significantly increased by 190.9 ± 29.3% (P < 0.02) by PGJ2 and, as was the case with h-caldesmon, this increase was blocked by LY-294002 (Fig. 7C). TG increased SM2 expression by 177.7 ± 7.2%; however, LY-294002 partially inhibited this response (Fig. 7D). There were no effects of LY-294002 on basal h-caldesmon or SM2 expression (not shown).
Previously, PPAR-γ ligands were reported to inhibit smooth muscle proliferation (32). The present study demonstrates that these agents also exert a positive effect in maintaining the fully differentiated contractile phenotype in hypertensive arteries. This phenotype is identified by the expression of several markers that are unique to smooth muscle. Among these are calponin, h-caldesmon, and the SM-MHCs SM1 and SM2. These are structural proteins that play important roles in fully functional mature smooth muscle. With DOCA salt hypertension, there was decreased expression of h-caldesmon and SM2 but not of calponin or SM1 in aortas and carotid arteries. As later markers of the mature phenotype, the expression of h-caldesmon and SM2 would be expected to decrease before that of calponin and SM1 (29). TG treatment of DOCA salt-treated rats led to restoration of h-caldesmon and SM2 expression. This increased expression was also seen in sham-treated rats also treated with TG.
We previously found that GLUT4 expression is decreased in the DOCA salt hypertension model (2) and in the Nω nitro-l-arginine (l-NNA) model (K. Atkins et al., unpublished data) of hypertension. As reported here, the TZDs TG and RS increase GLUT4 expression in vessels of DOCA salt-hypertensive rats. TG and the endogenous PPAR-γ agonist PGJ2 also increased GLUT4 expression in explanted aortas. Basal glucose uptake was decreased in DOCA salt hypertension in line with decreased GLUT4 expression and was restored by TG treatment concomitant with increased GLUT4 expression. Because there was no effect of DOCA salt hypertension or TG on GLUT1 expression, it is possible that in vascular smooth muscle, in contrast with other GLUT4-expressing tissues, GLUT4 plays a major role in basal glucose uptake.
Ex vivo experiments were attempted to ascertain whether the effect of TG on the expression of these markers was directly or indirectly related to the effects on hypertension. A 24-h treatment of carotid arteries with TG was sufficient to demonstrate an increase in the expression of GLUT4 and h-caldesmon. The lack of effect of TG on SM2 and calponin in carotid arteries (and of TG and PGJ2 on calponin in explanted aortas) suggests that a longer incubation period or other local factors may be necessary for induction. Clearly, PPAR-γ agonists exert many of their prodifferentiation effects directly on the arteries independently of effects on hypertension.
Although PPAR-γ ligands inhibit the proliferative response associated with pathological conditions, the induction of expression of SM2 and h-caldesmon in vessels indicates that these compounds may act directly to induce differentiation. This induction is likely independent of any inhibitory effects on proliferation, which is supported by the finding that these vessels are not apparently in a proliferative state. Additionally, given the inherent heterogeneity of vascular smooth muscle cellular populations in vessel walls, it is possible that the actions of these ligands may only be exerted in a subset of the smooth muscle cell population. This would explain why the effects of PPAR-γ ligands were relatively more robust on aortas from DOCA salt-treated rats than on aortas from control rats. Aortas from the former would have a greater proportion of dedifferentiated cells than the latter.
Because there are no apparent PPAR-γ-response elements in any of the genes investigated, it is likely that PPAR-γ agonists act through mechanisms other than direct mediation of transcription. It has been demonstrated that Akt activation is necessary to maintain the differentiated phenotype of smooth muscle (6, 7). The finding of decreased pAkt expression in vascular smooth muscle of DOCA salt-hypertensive rats (20) suggests this as a possible mechanism. Furthermore, TG has been demonstrated to increase markers of skeletal muscle differentiation through enhancement of PI3K and Akt activation (9). TG and PGJ2 indeed increased Akt activity as demonstrated by increased Akt phosphorylation. This is a PI3K-dependent mechanism in that this effect was inhibited by LY-294002. That inhibition of Akt phosphorylation was associated with inhibition of induction of SM2 and h-caldesmon expression suggests that PI3K/Akt signaling mediates PPAR-γ agonist-induced expression of these genes. These results contrast with those of Miwa et al. (18), who reported that PGJ2 increases SM2 mRNA and decreases pAkt expression in cultured human umbilical artery smooth muscle cells (16). Although the reason for these different effects is not clear, the system utilized in our experiments (explants) is closer to the in vivo state than are cultured vascular smooth muscle cells. Likewise, it is possible that the intact endothelium in the explants could mediate at least some of the effects of the PPAR-γ ligands in intact aortas.
Although the specific pathway involved in PPAR-γ-regulated smooth muscle marker expression remains to be elucidated, there are several lines of evidence that support the role of activated Akt. The GATA-6 transcription factor is a regulator of the smooth muscle phenotype. PGJ2 and TG induction of SM-MHC expression is dependent upon transcriptional control exerted by GATA-6 binding to its consensus motif in the SM-MHC promoter (1). GATA-6 mediated SM-MHC expression in vascular smooth muscle is dependent upon nuclear factor of activated T-cells (NFAT)c1 (31). Activation of NFATc1 is dependent upon the activity of the serine-threonine phosphatase calcineurin (10). Inhibition of calcineurin suppressed the promoter activity of the smooth muscle-specific genes for h-caldesmon and α1-integrin (22). Furthermore, inhibition of the expression of these genes by blockade of the PI3K/Akt pathway was rescued by constitutively active calcineurin (22). Our data support the conclusion that PPAR-γ ligands stimulate NFAT/GATA-6-mediated gene expression through activation of a PI3K/Akt/calcineurin pathway. A mechanism of PPAR-γ activation of this pathway is potentially through increased expression of the p85α subunit of PI3K (25).
Interestingly, PPAR-γ agonist-induced GLUT4 expression is not mediated by a PI3K/Akt-dependent mechanism. Rather than inhibit the PPAR-γ agonist-mediated effects on GLUT4, as it did with regard to SM2 and h-caldesmon, LY-294002 increased GLUT4 expression. Our result is consistent with the finding that the TZD derivative YM268 increased GLUT4 expression in adipocytes independently of PI3K signaling (28). Furthermore, dephosphorylation of the CCAAT-enhancer binding protein (C/EBP), which coordinately regulates adipogenesis with PPAR-γ, led to decreased GLUT4 expression (30). On the other hand, treatment with the PI3K inhibitors wortmannin and LY-294002 actually increased phosphorylation of C/EBP (8). Because smooth muscle cells express C/EBP, this suggests a possible mechanism for the effect we observed with LY-294002. Additional studies will be required to ascertain the validity of this putative mechanism. Our finding does suggest that although GLUT4 expression is associated with the fully mature vascular smooth muscle phenotype, GLUT4 expression is regulated by PPAR-γ agonists independently of the other markers through a combinatorial rather than coordinated program of gene expression leading to the mature phenotype (13).
Although the effects of TZDs on marker expression were clearly independent of their effects on BP, it is possible that the increased/restored marker expression may be part of the ameliorative effects of TZDs on hypertension. This is an important finding given that amelioration of BP alone does not reverse the damage to the vasculature caused by chronic hypertension. It suggests a possible therapeutic scheme whereby the phenotypic changes associated with the vascular damage might be reversed. In addition, given that h-caldesmon and calponin negatively regulate actin-myosin ATPase, it is possible that part of the effect of PPAR-γ ligands on reducing BP may be through restoration or increased expression of these proteins. The basic molecular mechanism of smooth muscle contractility is the force-producing ATP-dependent interaction of actin and myosin upon elevation of free intracellular Ca2+ concentration. Through interaction with actin-tropomyosin, h-caldesmon inhibits myosin-ATPase activity in the absence of increased Ca2+ concentration (21). With an increase in intracellular Ca2+ concentration, calmodulin binds h-caldesmon, which results in activation of myosin-ATPase activity with concomitant contraction (21). When synthetic peptide antagonists of h-caldesmon are introduced to smooth muscle, basal contractile tone of vascular smooth muscle is increased (16). Thus restoration of caldesmon expression by PPAR-γ ligands may decrease basal contractile tone and reactivity.
Thus our findings support our original hypothesis that PPAR-γ ligands mediate the differentiation of vascular smooth muscle. The class effect demonstrated by these data indicates that PPAR-γ may play an important role in the development or maintenance of the mature vascular smooth muscle phenotype and may have ameliorative effects on arteries in hypertension and in other vascular diseases.
This work was supported by National Heart, Lung, and Blood Institute Grants RO1 HL-60156 and RO1 HL-65567 (to F. C. Brosius) and an Established Investigator Grant (to S. W. Watts) from the American Heart Association.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 by the American Physiological Society