PPARγ in endothelial cells influences high fat diet-induced hypertension

AJH

2005; 18:549 –556

PPAR␥ in Endothelial Cells Influences High Fat Diet-Induced Hypertension Christopher J. Nicol, Masahiro Adachi, Taro E. Akiyama, and Frank J. Gonzalez Background: Peroxisome proliferator-activated receptor ␥ (PPAR␥) ligands improve human hypertension. However, the mechanism and site of this effect remains unknown, confounded by PPAR␥ expression in many cell types, including endothelial cells (ECs). Methods: To evaluate the vascular role of PPAR␥ we used a conditional null mouse model. Specific disruption of PPAR␥ in ECs was created by crossing Tie2-Cre⫹ transgenic (T2T⫹) and PPAR␥-floxed (fl/fl) mice to generate PPAR␥ (fl/fl)T2T⫹ (PPAR␥ E-null) mice. Conscious 8- to 12-week-old congenic PPAR␥ (fl/fl)Cre⫺ (wild type) and PPAR␥ E-null mice were examined for changes in systolic blood pressure (BP) and heart rate (HR), untreated, after 2 months of salt-loading (drinking water), and after treatment for 3 months with high fat (HF) diet alone or supplemented during the last 2 weeks with rosiglitazone (3 mg/kg/d). Results: Untreated PPAR␥ E-nulls were phenotypically indistinguishable from wild-type littermates. However, compared to similarly treated wild types, HF-treated

PPAR␥ E˜-nulls had significantly elevated systolic BP not seen after normal diet or salt-loading. Despite sex-dependent baseline differences, salt-loaded and HF-treated PPAR␥ E˜-nulls of either sex had significantly elevated HR versus wild types. Interestingly, rosiglitazone improved serum insulin levels, but not HF diet-induced hypertension, in PPAR␥ E˜-null mice. Conclusions: These results suggest that PPAR␥ in ECs not only is an important regulator of hypertension and HR under stressed conditions mimicking those arising in type 2 diabetics, but also mediates the antihypertensive effects of rosiglitazone. These data add evidence supporting a beneficial role for PPAR␥-specific ligands in the treatment of hypertension, and suggest therapeutic strategies targeting ECs may prove useful. Am J Hypertens 2005;18:549 –556 © 2005 American Journal of Hypertension, Ltd. Key Words: PPAR␥, conditional null mice, endothelial cells, blood pressure, type 2 diabetes.

he peroxisome proliferator-activated receptors (PPARs), of which there are three family members (␣, ␤, and ␥), first discovered in the early 1990s, are members of the orphan nuclear receptor superfamily.1 PPAR␥ is expressed in a number of cell types, including adipocytes, colonic epithelial cells, and monocytes/macrophages, as well as in vascular smooth muscle cells (VSMCs) and endothelial cells (ECs).2 Known ligands for PPAR␥ include synthetic insulin-sensitizing thiazolidinedione compounds (TZDs), such as troglitazone, BRL-49653 (aka rosiglitazone), and pioglitazone, as well as natural compounds, such as 15-deoxy-⌬12,14-prostaglandin J2, 9- and 13-hydroxyoctadecanoic acid (9- and 13-HODE), 15-hydroxyeicosatetraenoic acid (15-HETE), linoleic acid, and the recently discovered putative endogenous PPAR␥ agonist, lysophosphatidic acid (LPA).3,4

T

Use of both natural and synthetic ligands has provided evidence associating PPAR␥ not only with adipocyte differentiation, but also the onset of type 2 diabetes and colon cancer. Expression of PPAR␥ in cells of the vascular system has led to numerous studies in an attempt to clarify the physiologic role of this receptor in ECs. Several reports have shown that the use of TZDs, such as troglitazone and rosiglitazone, is correlated with marked decreases in hypertension in humans resulting from the onset of insulin resistance in adult type 2 diabetic patients.5,6 Interestingly, the decrease in mean blood pressure (BP) was also found to be significantly associated with stabilization of plasma insulin levels.6 Together, these reports suggest that PPAR␥ activation by TZDs can normalize BP through their insulin-sensitizing effects. In contrast, the most common human PPAR␥ polymorphism,

Received May 17, 2004. First decision September 13, 2004. Accepted October 19, 2004. From the Laboratory of Metabolism, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland. Present address of Christopher J. Nicol is Cancer Research Institute,

Queens University, 10 Stuart St., Rm 317, Kingston, Ont, Canada. Present address of Dr. Taro E. Akiyama is Merck Research Laboratories, 126 East Lincoln Ave., Rahway, NJ. Address correspondence and reprint requests to Dr. Frank J. Gonzalez, Building 37, Room 3106B1, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; e-mail: [email protected]

© 2005 by the American Journal of Hypertension, Ltd. Published by Elsevier Inc.

0895-7061/05/$30.00 doi:10.1016/j.amjhyper.2004.10.032

550

PPAR␥ IN ENDOTHELIAL CELLS REGULATES HYPERTENSION

a Pro:Ala change in codon 12 within the unique ligandindependent activation domain of PPAR␥2,7 has been correlated with increased systolic and diastolic BPs and body mass index (BMI) in a Utah population.8 More recently, two independent studies of Spanish and Swedish populations have similarly identified an increased risk of hypertension associated with the Pro/Pro polymorphic genotype.9,10 However, the nature and ability of either BMI or insulin resistance to influence the association between this variant and hypertension is controversial, as several studies have yielded conflicting results.7,8,10 –12 To directly evaluate the role of this nuclear receptor in the cardiovascular system, previously described PPAR␥ floxed (fl/fl)Cre⫺ mice13 were bred with a transgenic mouse line expressing the Cre recombinase protein driven by the Tie2 promoter (T2T⫹),14 specifically disrupting PPAR␥ in endothelial (E) cells and generating PPAR␥(fl/fl)T2T⫹ (henceforth referred to as PPAR␥ E-null) mice. The PPAR␥ E-nulls are viable and not phenotypically different from congenic PPAR␥(fl/fl)Cre⫺ (wild-type) littermate mice. However, after a high fat diet (HF) treatment, PPAR␥ E-null mice are markedly susceptible to increased systolic BP, a condition that is not improved by administration of rosiglitazone, suggesting a central role for PPAR␥ in the regulation of hypertension from within ECs.

Methods Animals All mice were housed and treated in accordance with protocols approved by the National Cancer Institute Animal Care and Use Committee. Mice were housed in microisolator cages on a 12-h light/dark cycle, with food and water provided ad libitum. Male and female PPAR␥(fl/fl)Cre⫺ (wildtype) were generated, as described previously,13 and crossed with Tie2-Cre⫹ transgenic mice (T2T⫹), as described elsewhere,14 to generate PPAR␥(fl/fl)T2T⫹ (PPAR␥ E-null) mice. These mice are of mixed C57Bl6/N;Sv129;FVB/N background and were bred using sibling matings for at least 12 generations before the start of the study. Genotypes were confirmed by polymerase chain reaction (PCR) analysis as described previously.15 Untreated and Salt-Loading Studies Eight- to 12-week-old PPAR␥ E-null or wild-type littermates were treated ad libitum from weaning with control diet (NIH-07 diet; Zeigler Bros, Inc., Gardeners, PA) (CD) and normal drinking water (Control). After 1 week of training and 9 days of BP analysis, these same mice were then treated weekly ad libitum with control diet and a substituted 2% NaCl solution as drinking water (2% salt water) for up to 60 days to achieve dietary salt loading, as described previously.16 Mice were then retested from 2 weeks before the end of the study for changes in hypertension.

AJH–April 2005–VOL. 18, NO. 4, Part 1

High Fat Diet Study Eight- to 12-week-old wild-type or congenic PPAR␥ Enull littermates were treated with AIN-93G-modified 35% high fat diet (F4048, BIO-SERV; Frenchtown, NJ) (HF) for 3 months. Two weeks before the end of the study, a set of mice were started on a blended HF diet containing the PPAR␥-specific agonist rosiglitazone (Avandia, 3 mg/kg/d; Smith Kline Beecham, West Sussex, UK), as described previously.17 Nonfasted serum was obtained at the end of the study through the retro-orbital sinus. Serum was then frozen in liquid nitrogen and stored in a ⫺80°C freezer until further analysis. Biochemical Assays Nonfasted serum insulin was measured by radioimmunoassay (RIA) (SRI-13K; Linco Research Inc., St. Charles, MO) according to manufacturer’s instructions, as described previously.17 BP Analysis Systolic BP and heart rate (HR) were analyzed on conscious (nonanesthetized) mice, as previously described.16 Mice were trained on the apparatus for 1 week before the start of recording and measurements were made at the same time in the morning for the time periods displayed. One complete cycle was performed before recording values to allow the mice to acclimate before actual readings were taken. Statistical Analysis Statistical evaluation was performed using one-way, twoway, or repeated measures ANOVA as appropriate, followed by a Student Neuman-Keuls post-hoc test. A significance level of P ⬍ .05 was used for all analyses.

Results To determine the role of PPAR␥ in ECs, PPAR␥ E-null mice were generated as described in the Methods section. Disruption of the PPAR␥ gene, after recombination of exon 2, was confirmed by Southern blotting (Fig. 1) and PCR analysis13,15 (data not shown). Note that the recombined allele was detected in both sexes and all tissues examined as expected, based on the wide distribution of blood vessels. Interestingly, recombination in the spleen appeared to be complete. The PPAR␥ E-null mice were born at the normal mendelian frequencies and without any outward phenotypic or biochemical differences in body weight or serum profiles compared to congenic PPAR␥(fl/ fl)Cre⫺ (wild-type) littermates (data not shown). To test for alterations in systolic BP and HR, conscious age-matched male and female wild-type and PPAR␥ Enull mice were examined using tail-cuff plethysmography. Before the start of analysis, control-treated mice were conditioned to the BP/recording apparatus by performing readings at the same time every morning for 7 days to

PPAR␥ IN ENDOTHELIAL CELLS REGULATES HYPERTENSION

AJH–April 2005–VOL. 18, NO. 4, Part 1

A

Female PPARγ E-null WAT Aorta BAT Liver Kidney (fl/-)(fl/+) Epi Ing Rpr Heart Skm Colon Spleen

Wild Flox Null

B

Male PPARγ E-null WAT Aorta BAT Liver Kidney (fl/-)(fl/+) Epi Ing Rpr Heart Skm Colon Spleen

Wild Flox Null FIG. 1 (A, B) Endothelial-specific conditional inactivation of PPAR␥. Southern blot analysis of recombination in tissues from female and male PPAR␥ E-null mice. Genomic DNA, except for aorta (3 ␮g) was isolated from tissues of 12-week-old mice, digested with BamHI, and analyzed as described in Reference 13. Lanes represent 10 ␮g of isolated DNA for all tissues except for aorta (3 ␮g). WAT ⫽ white adipose tissue; Epi ⫽ epididymal; Ing ⫽ inguinal; Rpr ⫽ retroperirenal; BAT ⫽ brown adipose tissue; Skm ⫽ skeletal muscle; (fl/⫺) ⫽ PPAR␥ (fl/⫺) positive control; (fl/⫹) ⫽ PPAR␥ (fl/⫹) positive control. Wild, Flox, and Null indicate expected sizes of PPAR␥ (⫹/⫹) (14 kb), PPAR␥ (fl/fl) (10 kb), and PPAR␥ (⫺/⫺) (8 kb) alleles, respectively.

minimize the influence of stress (data not shown). Because analysis of repeated measurements revealed no significant difference within any individual treatment group over the respective period of recordings, mean values of each group were averaged to obtain a representative physiologic measurement for that group. As shown in Fig. 2A and Table 1, no differences in systolic BP were observed between control-treated wild-type and PPAR␥ E-null mice of either sex. However, in contrast to control-treated female mice that exhibited no consistent difference in HR between genotypes, control-treated male PPAR␥ E-null mice had consistent significantly increased HR compared to their respective wild types, which was perhaps more apparent when respective means were compared (male PPAR␥ Enull versus wild type, mean bpm ⫾ SEM: 539.5 ⫾ 20.4 v 499.0 ⫾ 16.1, P ⬍ .05) (Fig. 2B and Table 1). To determine whether PPAR␥ plays a role in salt loading-induced hypertension, these same groups of mice were continued on a 2% salt water treatment for 60 days. Salt-loading resulted in enhanced systolic BP and HR over respective baseline values for all genotypes and sexes examined (P ⬍ .001) (Fig. 2 and Table 1). However, unlike systolic BP, which was not significantly different between respective salt-loaded genotypes of either sex, HR showed a trend toward significantly increased levels among salt-loaded PPAR␥ E-nulls compared to respective wild-type mice

551

over the measured treatment period, albeit this was more pronounced among males (Fig. 2). Moreover, HR was significantly increased in both sexes of PPAR␥ E-null mice compared to wild-type controls when mean values were determined (P ⬍ .05) (Table 1). To assess whether HF diet treatment would increase hypertension, systolic BP and HR were analyzed after 1 month of HF diet treatment (1 month). In fact, HF treatment significantly increased both systolic BP and HR in all genotypes and sexes examined as compared to respective untreated levels (P ⬍ .001) (Fig. 3A and Tables 1 and 2). Notably, HF-treated PPAR␥ E-null mice showed a consistent and significantly higher mean systolic BP compared to similarly treated wild-type mice, which was independent of gender (P ⬍ .05) (Fig. 3A and Table 2). Similarly, HR were also significantly higher among the HF-treated PPAR␥ E-null mice of either sex compared to wild-type controls (P ⬍ .05) (Fig. 3B and Table 2). This pattern was repeated when systolic BP and pulse were examined at the end of 3 months of HF treatment (3 months) where wild-type, as well as one-half of randomly segregated female and male PPAR␥ E-null mice were maintained on HF diet alone, whereas the other half of PPAR␥ E-null mice were treated during the final 2 weeks of the study with HF diet supplemented with rosiglitazone (P ⬍ .05) (Fig. 3 and Table 2). Consistent with published reports in the literature, data from separate previously unpublished studies in conscious congenic wildtype mice on normal chow diets respectively 1 day prior, and 4 and 10 days after oral gavage administration of rosiglitazone (3 mg/kg/d) or its vehicle control (Table 3), or with conscious congenic male mice treated for 2 months on a HF diet alone or with 2-week dietary supplementation

FIG. 2 Systolic blood pressure and heart rate of wild-type and PPAR␥ E-null mice. Systolic BP (A) and HR (B) were simultaneously measured using tail-cuff plethysmography in conscious wild-type and PPAR␥ E-null mice during the indicated periods of drinking water administration of water alone (control) or water supplemented with 2% NaCl (2% salt water). Values shown are the means ⫾ SEM. n ⫽ 7 each for female and n ⫽ 3 and 6 for male wild-type and PPAR␥ E-null groups, respectively. Statistical significance: *PPAR␥ E-null versus treated same sex wild-type controls, P ⬍ .05.

552

PPAR␥ IN ENDOTHELIAL CELLS REGULATES HYPERTENSION

AJH–April 2005–VOL. 18, NO. 4, Part 1

Table 1. Mean systolic BP and HR for wild-type and PPAR␥ E-null mice Mean Systolic BP (mm Hg) Group Female Wild-type PPAR␥ E-null Male Wild-type PPAR␥ E-null

Mean HR (beats/min)

Control

2% Salt

Control

2% Salt

117.6 ⫾ 2.3 116.7 ⫾ 1.9

144.7 ⫾ 2.4 138.8 ⫾ 6.3

547.4 ⫾ 8.3 553.6 ⫾ 7.9

555.9 ⫾ 8.9 586.0 ⫾ 11.0*

124.4 ⫾ 4.4 123.2 ⫾ 3.0

147.9 ⫾ 8.7 148.5 ⫾ 2.1

499.0 ⫾ 16.1 539.5 ⫾ 20.4*

554.2 ⫾ 7.1 580.9 ⫾ 8.6*

BP ⫽ blood pressure; HR ⫽ heart rate; PPAR␥ ⫽ peroxisone proliferator-activated receptor ␥. Mean values for each parameter were calculated by averaging the daily values for each animal determined over the 6 days of administration of normal drinking water (Control), and the first 3 and last 10 days of administration of 2% NaCl (2% salt water) as drinking water. Values shown are the means ⫾ SEM. n ⫽ 7 each for female, and 3 and 6 for male wild-type and PPAR␥ E-null mice, respectively. * Significantly different from respective same sex wild-type controls, P ⬍ .05.

of rosiglitazone (3 mg/kg/d) (systolic BP reported as mean mm Hg ⫾ SEM of HF alone, n ⫽ 5 v HF ⫹ rosiglitazone, n ⫽ 5: 123.8 ⫾ 4.7 v 110.6 ⫾ 3.0) (P ⬍ .05), indicated that rosiglitazone was able to significantly decrease systolic BP. Interestingly, rosiglitazone treatment of PPAR␥ E-null mice did not alter the significant HF-mediated increase of systolic BP compared to wild-type controls, irrespective of sex (Fig. 3A and Table 2). Strikingly though, rosiglitazone did appear to significantly reduce

175 150

Female

* * * * ** *

PPARγ E-null

† † †† * †† *†**† *†** †*

125

800 700 600

1 month

175 150

3 months

**

* **

* * † † ** * * *

1 month

3 months

0 Male

* ** * * * ** *

† † * †† † ** * *† * *

125

0

Female

B

500

0 200

PPARγ E-null + Rosi

Mean HR (bpm) Mean Pulse (bpm)

200

Wild-type

1 month 22

3 months

30 76 80 84 88 26 Time (Days)

800

Mean HR (bpm) Mean Pulse (bpm)

Mean SBP (mm of Hg)

Mean SBP (mm of Hg)

A

700 600

Male

* *** * *

†† † †* ** *

Discussion

500

0

1 month

22

26

HR among HF-treated female, but not male, PPAR␥ Enull mice to the levels of HF-treated wild types, a pattern that was consistent when overall means for the treatment period were determined and compared (P ⬍ .01) (Fig. 3B and Table 2). To determine whether in fact rosiglitazone administration was being consumed and able to have an effect, nonfasted serum insulin levels were examined at the end of BP analysis studies. After 3 months of HF treatment alone, serum insulin levels were not different between female (mean ng/mL ⫾ SD 8.22 ⫾ 2.05 v 6.84 ⫾ 1.16) and male (mean ng/mL ⫾ SD 9.99 ⫾ 2.82 v 8.40 ⫾ 1.23) PPAR␥ E-null versus wild-type control mice treated with HF alone (Fig. 4). However, as expected HF ⫹ rosiglitazone treatment did significantly lower serum insulin in both PPAR␥ E-null female (1.48 ng/mL ⫾ 1.08) and male (3.22 ng/mL ⫾ 3.77) mice compared to either same sex wild-type (P ⬍ .05) or PPAR␥ E-null mice treated with HF alone (P ⬍ .01) (Fig. 4).

3 months

30 76 80 84 88 Time (Days)

FIG. 3 Effect of HF diet on systolic blood pressure and heart rate of wild-type and PPAR␥ E-null mice. Systolic BP (A) and HR (B) were simultaneously measured in conscious wild-type and PPAR␥ E-null mice during the indicated periods of treatment as in Fig. 2. 1 month ⫽ all mice were treated for 1 month with HF diet alone. n ⫽ 4 and 7 for female, and n ⫽ 4 and 11 for male wild-type (●) and PPAR␥ E-null (Œ) groups, respectively. 3 months ⫽ after treatment with HF diet alone for 2.5 months, all PPAR␥ wild type (●, n ⫽ 4 each for females and males), and approximately half of randomly segregated female (n ⫽ 4) and male (n ⫽ 6) PPAR␥ E-null (Œ) mice groups were continued on HF diet alone for 2 additional weeks; whereas, the other half of PPAR␥ E-null mice (, n ⫽ 3 and 5 for females and males, respectively) were treated for the final 2 weeks with a HF diet supplemented with rosiglitazone (3 mg/kg body weight/d). Total duration of treatment with HF diet was 3 months. Values shown are the means ⫾ SEM. Statistical significance: *HF-treated PPAR␥ E-null versus same sex HF-treated wild-types, P ⬍ .05; †HF ⫹ rosiglitazone-treated PPAR␥ E-null versus same sex HF-treated wild types, P ⬍ .05.

Previous work has elucidated a clear role for PPAR␥ in adipogenesis, glucose and lipid homeostasis, and transport.1,13,18 This study shows, by using the first reported PPAR␥ conditional E-null mouse strain, that PPAR␥ in ECs plays an important role in HR and hyperlipidemicmediated hypertension. Because classic PPAR␥ null mice die in utero around gestational day 10.5 attributed to embryonic defects in placental vascularization and myocardial thinning,18 –20 PPAR␥ floxed mice13 and an EC targeting strategy, previously characterized as active and sustained in mice from as early as embryonic day 7.5 through adulthood,14 were used to generate conditional PPAR␥ E-null mice, resulting in spontaneous disruption of PPAR␥ in cells of endothelial origin.14,21 Interestingly, recombination in the spleen of PPAR␥ E-null mice appeared to be complete, suggesting that immune cells might also be a major site of PPAR␥ disruption. Strikingly, these mice were viable at birth and

PPAR␥ IN ENDOTHELIAL CELLS REGULATES HYPERTENSION

AJH–April 2005–VOL. 18, NO. 4, Part 1

553

Table 2. HF diet effect on mean systolic BP and HR for wild-type and PPAR␥ E-null mice Mean Systolic BP (mm Hg) Group

1 mo HF

Female Wild-type PPAR␥ E-null PPAR␥ E-null ⫹ Rosi Male Wild-type PPAR␥ E-null PPAR␥ E-null ⫹ Rosi

Mean HR (beats/min)

3 mo HF

1 mo HF

3 mo HF

n 4 135.1 ⫾ 2.3 7 153.7 ⫾ 2.0*** — —

n 4 4 3

129.9 ⫾ 3.1 152.6 ⫾ 4.1*** 152.2 ⫾ 4.1

599.5 ⫾ 15.1 640.8 ⫾ 8.2** —

4 137.0 ⫾ 3.8 11 156.4 ⫾ 1.9*** — —

4 6 5

137.6 ⫾ 3.4 150.7 ⫾ 2.1*** 153.8 ⫾ 3.8

581.9 ⫾ 7.8 595.8 ⫾ 8.5 628.6 ⫾ 3.5*** 636.5 ⫾ 4.5*** — 639.5 ⫾ 6.5

601.2 ⫾ 8.0 644.9 ⫾ 10.0*** 613.3 ⫾ 12.1‡

1 moHF ⫽ 1 month of treatment with high fat diet; 3 moHF ⫽ 3 months of treatment with high fat diet; n ⫽ number of mice; Rosi ⫽ rosiglitazone. Mean values for each parameter were calculated by averaging the daily values for each animal determined over the last 9 days leading up to treatment with 1 month of HF diet (HF), and during the final 2 weeks of treatment with 3 months of HF diet with or without rosiglitazone supplementation (HF ⫹ Rosi). Values shown are the means ⫾ SEM. ** Significantly different from respective same sex wild-type controls, P ⬍ .01; *** significantly different from respective same sex wild-type controls, P ⬍ .001; ‡ significantly different from same sex PPAR␥ E-null HF alone group, P ⬍ .01.

indistinguishable phenotypically or biochemically from their congenic wild-type littermate controls, suggesting that disruption of PPAR␥ in ECs alone is insufficient to induce embryonic or early neonatal lethality as found in the classic PPAR␥ null mice.19 It is therefore likely that disruption of PPAR␥ in other vascular cell types such as VSMCs may, either alone or in cooperation with disruption of both alleles in endothelial and other cell types, contribute to the cardiovascular defects and lethality seen in classic PPAR␥ null embryos. Because several reports using PPAR␥ ligands have suggested a role for this receptor in regulating BP in both humans and animals,22,23 it was hypothesized that PPAR␥ E-null mice may exhibit differences in hypertension as compared to wild-type controls. However, neither sex of untreated PPAR␥ E-null mice demonstrated any differences in mean systolic BP averaged over the course of the study. When mice were salt-loaded by the drinking water, a method previously shown to induce elevations in hypertension in mice,24 the results showed that compared to baseline, all groups had significant elevations in the mean systolic BP,

averaged over the course of 8 weeks of treatment. However, there were no genotypic differences in mean systolic BP between PPAR␥ E-null and wild-type controls for either sex. Together, these data suggest that disruption of PPAR␥ in endothelial cells alone or in combination with long-term salt-loading is insufficient to cause significant elevations in systolic BP over wild-type controls. In contrast, the observed increased mean HR of both untreated PPAR␥ E-null males and long-term salt-loaded PPAR␥ E-nulls of either sex suggests an important role in HR for PPAR␥ expressed in ECs. These effects are inconsistent with a study reporting no change in the HR of a small group of patients, with mild forms of both hypertension and type 2 diabetes, after treatment with the PPAR␥ ligand troglitazone.25 Reasons for this discrepancy may include strain differences or the inability to detect a significant decrease in HR after troglitazone treatment given the reportedly mild forms of the diseases and the small number of patients examined, although this remains speculative. However, consistent with the results presented here, a study comparing wild-type and whole body PPAR␥

Table 3. Rosiglitazone effect on mean systolic BP of wild-type mice Mean systolic BP (mm Hg) Wild type Female Vehicle Rosi Male Vehicle Rosi

Day 1 n 7 7 n 4 4

Day 4

Day 10

148.0 ⫾ 8.7 141.1 ⫾ 7.0

148.8 ⫾ 5.7 135.3 ⫾ 9.0

159.2 ⫾ 6.0 125.0 ⫾ 6.0*

143.6 ⫾ 10.0 163.7 ⫾ 9.3

151.5 ⫾ 15.1 143.7 ⫾ 4.3*

140.4 ⫾ 10.3 141.2 ⫾ 7.4*

Abbreviations as in Tables 1 and 2. Female and male systolic BP were simultaneously measured, using tail-cuff plethysmography, in conscious wild-type mice on normal chow diets respectively 1 day before, and 4 and 10 days after oral gavage administration of the PPAR␥ ligand rosiglitazone (3 mg/kg/d) or its vehicle control (0.5% carboxymethyl cellulose). Values shown are the means ⫾ SEM. * Significantly different from respective pretreatment controls, P ⬍ .05.

554

PPAR␥ IN ENDOTHELIAL CELLS REGULATES HYPERTENSION

FIG. 4 Nonfasted serum insulin after HF treatment of wild-type and PPAR␥ E-null mice. Serum insulin values (in nanograms per milliliter) after 3 month of HF treatment among nonfasted female and male wild-type and PPAR␥ E-null mice treated as described in Fig. 3. As described, n ⫽ 4 each for HF-treated female and male PPAR␥ wild-type mice, n ⫽ 4 and 6 for female and male HF-treated PPAR␥ E-null mice and n ⫽ 3 and 5 for female and male HF ⫹ rosiglitazonetreated PPAR␥ E-null mice. Values represent mean ⫾ SD. *Significantly different from respective wild-type controls, P ⬍ .05; †significantly different from PPAR␥ E-null HF diet alone controls, P ⬍ .01.

heterozygous mice observed no differences in BP either in untreated or pressure overloaded animals, whereas increases in the pressure overload-induced cardiac hypertrophy were more prominent among PPAR␥ heterozygous mice.26 Furthermore, although there are conflicting reports,27 several studies have shown that PPAR␥ ligands decrease cardiac hypertrophy,26,28 possibly through the PPAR␥-dependent antagonistic action of transcription factors known to be involved in this condition. In support, PPAR␥ ligands also have been shown to suppress the AP-1-mediated induction of endothelin-1 (ET-1), the latter of which is capable of inducing cardiac hypertrophy.22,28,29 Hence, the increased HR observed here may be the result of increased cardiac hypertrophy induced by disruption of PPAR␥ in ECs of PPAR␥ E-nulls, and warrants future study. Because several human studies have reported sex-dependent differences with respect to the susceptibility of developing hypertensive disorders,30 –32 male and female mice in this study were examined separately. It was reported that the presence of the PPAR␥ Pro12Ala polymorphism, and more specifically the Pro/Pro genotype, was associated with significantly increased hypertensive risk among males but less so in females.9,10 A more recent study has further identified polymorphisms within the estrogen receptors ER␣ and ER␤ associated with significantly elevated BP among males but not females.30 Thus, based on the literature, it was expected that male PPAR␥ E-null mice may exhibit more pronounced effects than their female counterparts. However, although untreated males of either genotype were observed to have a nonsignificant elevation in systolic BP when compared to respective untreated females, both sexes were observed to respond to a similar degree with respect to systolic BP after salt-loading or HF treatment. In addition, both sexes of PPAR␥ E-null mice developed significantly elevated hypertension after HF treatment as compared to their same sex wild-type controls. This suggests that both sexes of the

AJH–April 2005–VOL. 18, NO. 4, Part 1

PPAR␥ E-null model may be useful for understanding the mechanisms of hypertension. However, the apparent discrepancy regarding the lack of sex differences here with respect to hypertension may be due to inherent species differences between mouse and human studies. In contrast, untreated male, but not female, PPAR␥ E-null mice did have significantly increased HR versus wild-type controls, a difference that was masked in the presence of saltloading or HF treatment. This suggests an underlying sex difference among PPAR␥ E-null mice that is overcome under stressed conditions, although the source of this effect is unclear and warrants further study. Type 2 diabetes is often associated with an increase in hypertension, and is suggested to be linked to insulin resistance.6,33 Thus we sought to characterize whether an established mouse protocol of HF diet-mediated insulin resistance, previously shown to induce insulin resistance18 (and our unpublished observations) and essentially mimicking the human type 2 diabetic syndrome, would have an effect on our PPAR␥ conditional E-null mice. The results here suggest that PPAR␥ in ECs normally acts to prevent HF-induced hypertension. The difference in effects seen here between salt-induced and HF-mediated hypertension among PPAR␥ E-null mice suggests a differential susceptibility to the mechanisms involved. This pattern is consistent with recent reports of the ␣1D-adrenergicreceptor-deficient mouse, which is sensitive to central salt-loading-mediated, but not angiotensin II (ANG II)mediated, hypertension.34 Previous studies have suggested that PPAR␥ agonists are capable of downregulating ANG II type 1 receptors35 and that activated PPAR␥ suppresses the type 1 ANG II gene at the transcriptional level.36 Whether the hypertensive effects seen after HF treatment may be mediated in whole or in part by an unregulated ANG II response among PPAR␥ E-null mice remains to be determined. Moreover, it is noted that salt and ANG II plasma levels are inversely related, with increased salt intake associated with suppression of the renin-angiotensin system,24 and hence a decreased dependence on ANG II-mediated regulation of BP. Thus, the salt-loading induced here may suppress levels of ANG II thereby masking the genotypic differences observed in hypertension during HF treatment and warrants further study. Alternatively, PPAR␥ has been shown to inhibit expression of a number of targets in the endothelium, including thrombinmediated ET-1 production.29 Because plasma ET-1 levels have been shown to be elevated among type 2 diabetic patients,37,38 and given its vasoconstrictor and mitogenic effects as discussed, the unregulated expression of ET-1 among PPAR␥ E-null mice may contribute to the HFmediated hypertensive effects observed in this study. However, a recent report39 suggests that thromboxane blockade decreases hypertension independent of ET-1, suggesting the latter may not play a critical role in our results. Interestingly, one report suggests that both PPAR␥ ligands and PPAR␥ overexpression were able to increase the secretion of nitric oxide (NO) from ECs derived from

AJH–April 2005–VOL. 18, NO. 4, Part 1

PPAR␥ IN ENDOTHELIAL CELLS REGULATES HYPERTENSION

various species and tissue sites in a transcriptionally regulated, but in an endothelial nitric oxide synthase (eNOS)independent manner.40 In light of the fact that NO contributes to vascular tone, platelet adhesion, and VSMC proliferation, the results shown here suggest that the elevation of systolic BP may be a direct result of disruption of the homeostatic mechanisms by which PPAR␥-mediated NO secretion helps to inhibit hypertension. The inability of the PPAR␥-specific agonist rosiglitazone to reduce the HF-mediated hypertension among PPAR␥ E-null mice suggests its well-documented antihypertensive properties are exerted through activation of PPAR␥ in ECs. During the course of publishing this work, Ryan et al41 have recently reported that in a mouse model of lifelong hypertension and vascular dysfunction, rosiglitazone reduces BP in a manner that was independent of alterations in either eNOS, soluble guanylate cyclase, or genes thought to be important for vascular contraction such as angiotensin receptor 1 (AT1) and PPET-1 (precursor of the potent vasoconstrictor endothelin). It was further suggested that part of improvements seen in the vascular function and BP41 may be attributed to the PPAR␥-independent effects of rosiglitazone, such as through the attenuation of inward calcium currents and enhancement of calcium-activated potassium currents resulting in vascular relaxation as previously identified in isolated VSMCs.42 Interestingly, these findings differ somewhat from the findings in other cell and tissue culture systems, and rat studies35,43,44 and may suggest that the differences may be intrinsic to the differences in the models of hypertension used, although it was suggested that activation of PPAR␥ may indirectly control expression of the formerly mentioned genes.41 However, perhaps most striking is the inability of rosiglitazone to reduce hypertension among PPAR␥ E-null mice of either sex in the present study, which suggests that PPAR␥-dependent effects of rosiglitazone, as well as similar acting TZDs, within ECs play a direct and central role in improving HF-mediated hypertension. Nevertheless, the mechanism by which this occurs is still not clear. In conclusion, a viable endothelial-specific PPAR␥ conditional null mouse strain was created to enable direct examination in vivo of the role of PPAR␥ expression in these cells. Treatment with a specific PPAR␥ agonist, rosiglitazone, was unable to lower the high fat diet-induced marked increase in systolic BP among PPAR␥ E-null mice. The present study demonstrates that the reduction in hypertension attributed to TZDs, such as rosiglitazone, is the result of action on PPAR␥ expressed within ECs, as distinct from its expression within other cells of not only the vascular system such as smooth muscle cells, but also other tissues such as adipocytes or skeletal muscle cells that also express PPAR␥, which may play a role in the TZD-mediated insulin-sensitizing effects of these compounds. Therefore these studies have important clinical implications for the treatment of patients with hyper-

555

tension resulting from exposure to a high fat diet or type 2 diabetes, and suggest a therapeutic strategy targeting ECs, and perhaps PPAR␥ directly, may prove useful. Together these results suggest that PPAR␥ in ECs may act as an important regulator of hypertension and HR under stressed conditions that mimic those arising in type 2 diabetic patients. These data also add further evidence in support of a beneficial role for PPAR␥specific ligands in the therapeutic treatment of hypertension, and suggest therapeutic strategies focusing on ECs may prove useful in the treatment and prevention of this condition.

Acknowledgment We thank Dr. Jeffrey Kopp for advice and the generous use of plethysmographic apparatus during the course of this study.

References 1.

Lee CH, Olson P, Evans RM: Minireview: lipid metabolism, metabolic diseases, and peroxisome proliferator-activated receptors. Endocrinology 2003;144:2201–2207. 2. Braissant O, Foufelle F, Scotto C, Dauca M, Wahli W: Differential expression of peroxisome proliferator-activated receptors (PPARs): tissue distribution of PPAR-alpha, -beta, and -gamma in the adult rat. Endocrinology 1996;137:354 –366. 3. Bishop-Bailey D, Wray J: Peroxisome proliferator-activated receptors: a critical review on endogenous pathways for ligand generation. Prostaglandins Other Lipid Mediat 2003;71:1–22. 4. McIntyre TM, Pontsler AV, Silva AR, St Hilaire A, Xu Y, Hinshaw JC, Zimmerman GA, Hama K, Aoki J, Arai H, Prestwich GD: Identification of an intracellular receptor for lysophosphatidic acid (LPA): LPA is a transcellular PPARgamma agonist. Proc Natl Acad Sci USA 2003;100:131–136. 5. St John Sutton M, Rendell M, Dandona P, Dole JF, Murphy K, Patwardhan R, Patel J, Freed M: A comparison of the effects of rosiglitazone and glyburide on cardiovascular function and glycemic control in patients with type 2 diabetes. Diabetes Care 2002; 25:2058 –2064. 6. Ghazzi MN, Perez JE, Antonucci TK, Driscoll JH, Huang SM, Faja BW, Whitcomb RW: Cardiac and glycemic benefits of troglitazone treatment in NIDDM. The Troglitazone Study Group. Diabetes 1997;46:433– 439. 7. Beamer BA, Yen CJ, Andersen RE, Muller D, Elahi D, Cheskin LJ, Andres R, Roth J, Shuldiner AR: Association of the Pro12Ala variant in the peroxisome proliferator-activated receptor-gamma2 gene with obesity in two Caucasian populations. Diabetes 1998;47: 1806 –1808. 8. Hasstedt SJ, Ren QF, Teng K, Elbein SC: Effect of the peroxisome proliferator-activated receptor-gamma 2 pro(12)ala variant on obesity, glucose homeostasis, and blood pressure in members of familial type 2 diabetic kindreds. J Clin Endocrinol Metab 2001;86:536 – 541. 9. Rodriguez-Esparragon FJ, Rodriguez-Perez JC, Macias-Reyes A, Alamo-Santana F: Peroxisome proliferator-activated receptor-gamma2-Pro12Ala and endothelial nitric oxide synthase-4a/bgene polymorphisms are associated with essential hypertension. J Hypertens 2003;21:1649 –1655. 10. Ostgren CJ, Lindblad U, Melander O, Melander A, Groop L, Rastam L: Peroxisome proliferator-activated receptor-gammaPro12Ala polymorphism and the association with blood pressure in type 2 diabetes: skaraborg hypertension and diabetes project. J Hypertens 2003;21:1657–1662.

556

PPAR␥ IN ENDOTHELIAL CELLS REGULATES HYPERTENSION

11. Mancini FP, Vaccaro O, Sabatino L, Tufano A, Rivellese AA, Riccardi G, Colantuoni V: Pro12Ala substitution in the peroxisome proliferator-activated receptor-gamma2 is not associated with type 2 diabetes. Diabetes 1999;48:1466 –1468. 12. Deeb SS, Fajas L, Nemoto M, Pihlajamaki J, Mykkanen L, Kuusisto J, Laakso M, Fujimoto W, Auwerx J: A Pro12Ala substitution in PPARgamma2 associated with decreased receptor activity, lower body mass index and improved insulin sensitivity. Nat Genet 1998; 20:284 –287. 13. Akiyama TE, Sakai S, Lambert G, Nicol CJ, Matsusue K, Pimprale S, Lee YH, Ricote M, Glass CK, Brewer HB Jr, Gonzalez FJ: Conditional disruption of the peroxisome proliferator-activated receptor gamma gene in mice results in lowered expression of ABCA1, ABCG1, and apoE in macrophages and reduced cholesterol efflux. Mol Cell Biol 2002;22:2607–2619. 14. Kisanuki YY, Hammer RE, Miyazaki J, Williams SC, Richardson JA, Yanagisawa M: Tie2-Cre transgenic mice: a new model for endothelial cell-lineage analysis in vivo. Dev Biol 2001;230:230 – 242. 15. Nicol CJ, Yoon M, Ward JM, Yamashita M, Fukamachi K, Peters JM, Gonzalez FJ: PPAR␥ influences susceptibility to DMBA-induced mammary, ovarian and skin carcinogenesis. Carcinogenesis 2004;25:1747–1755. 16. Sinal CJ, Miyata M, Tohkin M, Nagata K, Bend JR, Gonzalez FJ: Targeted disruption of soluble epoxide hydrolase reveals a role in blood pressure regulation. J Biol Chem 2000;275:40504 – 40510. 17. Matsusue K, Haluzik M, Lambert G, Yim SH, Gavrilova O, Ward JM, Brewer B Jr, Reitman ML, Gonzalez FJ: Liver-specific disruption of PPARgamma in leptin-deficient mice improves fatty liver but aggravates diabetic phenotypes. J Clin Invest 2003;111:737–747. 18. Kubota N, Terauchi Y, Miki H, Tamemoto H, Yamauchi T, Komeda K, Satoh S, Nakano R, Ishii C, Sugiyama T, Eto K, Tsubamoto Y, Okuno A, Murakami K, Sekihara H, Hasegawa G, Naito M, Toyoshima Y, Tanaka S, Shiota K, Kitamura T, Fujita T, Ezaki O, Aizawa S, Nagai R, Tobe K, Kimura S, Kadowaki T: PPAR gamma mediates high-fat diet-induced adipocyte hypertrophy and insulin resistance. Mol Cell 1999;4:597– 609. 19. Barak Y, Nelson MC, Ong ES, Jones YZ, Ruiz-Lozano P, Chien KR, Koder A, Evans RM: PPAR gamma is required for placental, cardiac, and adipose tissue development. Mol Cell 1999;4:585–595. 20. Rosen ED, Sarraf P, Troy AE, Bradwin G, Moore K, Milstone DS, Spiegelman BM, Mortensen RM: PPAR gamma is required for the differentiation of adipose tissue in vivo and in vitro. Mol Cell 1999;4:611– 617. 21. Liao Y, Day KH, Damon DN, Duling BR: Endothelial cell-specific knockout of connexin 43 causes hypotension and bradycardia in mice. Proc Natl Acad Sci USA 2001;98:9989 –9994. 22. Iglarz M, Touyz RM, Amiri F, Lavoie MF, Diep QN, Schiffrin EL: Effect of peroxisome proliferator-activated receptor-alpha and -gamma activators on vascular remodeling in endothelin-dependent hypertension. Arterioscler Thromb Vasc Biol 2003;23:45–51. 23. Raji A, Seely EW, Bekins SA, Williams GH, Simonson DC: Rosiglitazone improves insulin sensitivity and lowers blood pressure in hypertensive patients. Diabetes Care 2003;26:172–178. 24. Leckie BJ: The action of salt and captopril on blood pressure in mice with genetic hypertension. J Hypertens 2001;19:1607–1613. 25. Ogihara T, Rakugi H, Ikegami H, Mikami H, Masuo K: Enhancement of insulin sensitivity by troglitazone lowers blood pressure in diabetic hypertensives. Am J Hypertens 1995;8:316 –320. 26. Asakawa M, Takano H, Nagai T, Uozumi H, Hasegawa H, Kubota N, Saito T, Masuda Y, Kadowaki T, Komuro I: Peroxisome proliferator-activated receptor gamma plays a critical role in inhibition of cardiac hypertrophy in vitro and in vivo. Circulation 2002;105: 1240 –1246. 27. Arakawa K, Ishihara T, Aoto M, Inamasu M, Kitamura K, Saito A: An antidiabetic thiazolidinedione induces eccentric cardiac hypertrophy by cardiac volume overload in rats. Clin Exp Pharmacol Physiol 2004;31:8 –13.

AJH–April 2005–VOL. 18, NO. 4, Part 1

28. Sakai S, Miyauchi T, Irukayama-Tomobe Y, Ogata T, Goto K, Yamaguchi I: Peroxisome proliferator-activated receptor-gamma activators inhibit endothelin-1-related cardiac hypertrophy in rats. Clin Sci (Lond) 2002;103(Suppl 48):16S–20S. 29. Delerive P, Martin-Nizard F, Chinetti G, Trottein F, Fruchart JC, Najib J, Duriez P, Staels B: Peroxisome proliferator-activated receptor activators inhibit thrombin-induced endothelin-1 production in human vascular endothelial cells by inhibiting the activator protein-1 signaling pathway. Circ Res 1999;85:394 – 402. 30. Ellis JA, Infantino T, Harrap SB: Sex-dependent association of blood pressure with oestrogen receptor genes ER␣ and ER␤. J Hypertens 2004;22:1127–1131. 31. Reckelhoff JF: Gender differences in the regulation of blood pressure. Hypertension 2001;37:1199 –1208. 32. Harrap SB, Stebbing M, Hopper JL, Hoang HN, Giles GG: Familial patterns of covariation for cardiovascular risk factors in adults: The Victorian Family Heart Study. Am J Epidemiol 2000;152:704 –715. 33. Chen YE, Fu M, Zhang J, Zhu X, Lin Y, Akinbami MA, Song Q: Peroxisome proliferator-activated receptors and the cardiovascular system. Vitam Horm 2003;66:157–188. 34. Chu CP, Kunitake T, Kato K, Watanabe S, Qiu DL, Tanoue A, Kannan H: The alpha 1D-adrenergic receptor modulates cardiovascular and drinking responses to central salt loading in mice. Neurosci Lett 2004;356:33–36. 35. Takeda K, Ichiki T, Tokunou T, Funakoshi Y, Iino N, Hirano K, Kanaide H, Takeshita A: Peroxisome proliferator-activated receptor gamma activators downregulate angiotensin II type 1 receptor in vascular smooth muscle cells. Circulation 2000;102: 1834 –1839. 36. Sugawara A, Takeuchi K, Uruno A, Ikeda Y, Arima S, Kudo M, Sato K, Taniyama Y, Ito S: Transcriptional suppression of type 1 angiotensin II receptor gene expression by peroxisome proliferatoractivated receptor-gamma in vascular smooth muscle cells. Endocrinology 2001;142:3125–3134. 37. Sarman B, Toth M, Somogyi A: Role of endothelin-1 in diabetes mellitus. Diabetes Metab Rev 1998;14:171–175. 38. Kawamura M, Ohgawara H, Naruse M, Suzuki N, Iwasaki N, Naruse K, Hori S, Demura H, Omori Y: Increased plasma endothelin in NIDDM patients with retinopathy. Diabetes Care 1992;15: 1396 –1397. 39. Lariviere R, Moreau C, Rodrigue ME, Lebel M: Thromboxane blockade reduces blood pressure and progression of renal failure independent of endothelin-1 in uremic rats. Prostaglandins Leukot Essent Fatty Acids 2004;71:103–109. 40. Calnek DS, Mazzella L, Roser S, Roman J, Hart CM: Peroxisome proliferator-activated receptor gamma ligands increase release of nitric oxide from endothelial cells. Arterioscler Thromb Vasc Biol 2003;23:52–57. 41. Ryan MJ, Didion SP, Mathur S, Faraci FM, Sigmund CD: PPAR(gamma) agonist rosiglitazone improves vascular function and lowers blood pressure in hypertensive transgenic mice. Hypertension 2004;43:661– 666. 42. Eto K, Ohya Y, Nakamura Y, Abe I, Fujishima M: Comparative actions of insulin sensitizers on ion channels in vascular smooth muscle. Eur J Pharmacol 2001;423:1–7. 43. Iglarz M, Touyz RM, Viel EC, Paradis P, Amiri F, Diep QN, Schiffrin EL: Peroxisome proliferator-activated receptor-␣ and receptor-␥ activators prevent cardiac fibrosis in mineralocorticoiddependent hypertension. Hypertension 2003;23:45–51. 44. Diep QN, El Mabrouk M, Cohn JS, Endemann D, Amiri F, Virdis A, Neves MF, Schiffrin EL: Structure, endothelial function, cell growth, and inflammation in blood vessels of angiotensin II-infused rats: role of peroxisome proliferator-activated receptor-gamma. Circulation 2002;105:2296 –2302.