Contrasting vascular effects caused by loss of Bardet-Biedl syndrome genes

Abstract

Bardet-Biedl syndrome (BBS) is a genetically heterogeneous, autosomal-recessive disorder associated with several clinical features including obesity, hypertension, and cardiovascular abnormalities. BBS proteins play an important role in the function of cilia, a mechanosensory organelle in endothelial cells, but whether these proteins are directly involved in the regulation of vascular function is unclear. Here, we show that Bbs genes (1–12) are expressed in endothelial and smooth muscle cell lines and tissues enriched in endothelial (lung) and smooth muscle (stomach) cells as well as the aorta. Next, we used aortic rings to examine the vascular function of two BBS mouse models that recapitulate the human phenotype, namely Bbs2^−/− (obese and normotensive) and Bbs6^−/− (obese and hypertensive) mice. Interestingly, the endothelium-dependent relaxation (induced by ACh) was significantly enhanced in Bbs2^−/− but not Bbs6^−/− mice. In contrast, the endothelium-independent relaxation (induced by sodium nitroprusside) was unaltered in both BBS mouse models. In addition, the contractile responses to serotonin and endothelin-1 were attenuated in Bbs2^−/− but not Bbs6^−/− mice. Of note, the NO-producing enzymes (eNOS and iNOS) were upregulated in the aorta of Bbs2^−/− but not Bbs6^−/− mice. On the other hand, the expression level of membrane subunits of NADPH oxidase (p22^phox and p47^phox) in the aorta was decreased in Bbs2^−/− mice but increased in Bbs6^−/− mice. In conclusion, these data implicate Bbs genes in the regulation of vascular function and demonstrate that disrupting Bbs2 and Bbs6 genes affect differentially the vascular function.

bardet-biedl syndrome (BBS) is a genetically heterogeneous disorder characterized by obesity, learning disabilities, polydactyly, renal abnormalities, hypogonadism, and retinopathy (34). Hypertension and cardiovascular disease are also common in BBS patients (20). The BBS results from mutations in several genes, 12 described thus far, that encode proteins that are critical for the structure and function of cilia (33). In keeping with this, highly conserved homologs of BBS proteins are found in ciliated organisms but not in nonciliated species (19, 22). In addition, Bbs gene products were shown to localize to the centriolar satellites of centrosomes and basal bodies of primary cilia where they interact with components of the dynein transport machinery of intraflagellar transport (17, 30).

Recent studies provide important clues on how BBS proteins are involved in ciliary function and why mutations in any Bbs gene lead to the same phenotype. Indeed, 7 BBS proteins (BBS1, BBS2, BBS4, BBS5, BBS7, BBS8, and BBS9) were found to form a core complex, known as the BBSome, which functions as a unit that mediates protein/vesicle trafficking to cilia (23). On the other hand, 3 chaperonin-like BBS proteins (BBS6, BBS10, and BBS12) form a complex in combination with other chaperonin proteins of the CCT/TRiC family and mediate the assembly of the BBSome (28).

The development of animal models of BBS has allowed a remarkable progress in understanding the specific role of BBS proteins in various tissues and to gain mechanistic insight into the pathophysiology of this syndrome. Of note, BBS mouse models exhibit major components of the human phenotype including obesity, retinal degeneration, and neurological deficits (10, 13, 22, 24). We (10, 26) also demonstrated that Bbs4- and Bbs6-null mice have elevated blood pressure, whereas Bbs1 M390R knockin and Bbs2 knockout mice do not.

Disruption of Bbs genes in the neurons appears to affect the membrane localization of several receptors including the leptin receptor (4, 29). Consequently, loss of BBS proteins in the hypothalamic neurons perturbs the mechanisms that control energy balance leading to obesity (26, 29). On the other hand, inactivation of Bbs genes in retinal tissue leads to mislocalization of rhodopsin, which precedes apoptotic death of the photoreceptors and retinal degeneration (24). Despite this progress, there are a large number of cell types and tissues in which the role of BBS proteins is still undermined.

The key role of Bbs genes in ciliary function led us to hypothesize that BBS proteins are critically involved in the regulation of vascular function. The expression of Bbs genes in the endothelial and smooth muscle cells as well as the vasculature was examined. In addition, to clarify the role of Bbs genes in the control of vascular functions, we evaluated the vascular effects of deleting Bbs2 or Bbs6 genes in mice.

MATERIALS AND METHODS

Drugs and reagents.

ACh, A-23187, sodium nitroprusside (SNP), KCl, and 5-HT were obtained from Sigma-Aldrich and dissolved in physiological saline solution. Prostaglandin F_2α (PGF_2α; Lutalyse; Pfizer Animal Health) was supplied by the University of Iowa Pharmacy. Endothelin-1 (ET-1) was purchased from Peninsula Laboratories and dissolved in physiological saline. All concentrations are stated as the final concentrations in the organ chamber. We used antibodies against phosphorylated endothelial nitric oxide synthase (phospho-eNOS; Ser1177) and inducible nitric oxide synthase (iNOS) both from Cell Signaling Technology, total eNOS (Santa Cruz Biotechnology), and β-actin (Abcam). Secondary rabbit antibody was purchased from Santa Cruz Biotechnology. Enhanced chemiluminescence (ECL) detection kit was purchased form Amersham GE Healthcare. Smooth muscle cell basal medium and endothelial cell basal medium were from Lonza.

Animals.

Generation of Bbs2^−/− and Bbs6^−/− mice and genotyping were described previously (13, 24). Littermate wild-type mice were used as controls. All mice were fed standard mouse chow (LM-485; Teklad Premier Laboratory Diets) and water ad libitum. Care of the mice used in the experiments met the standards set forth by the National Institutes of Health (NIH) in their guidelines for the care and use of experimental animals. All procedures were approved by the Animal Care and Use Committee of the University of Iowa.

Aortic ring preparations.

Aortic rings were prepared for measurements of contraction and relaxation as described previously (5, 6). In brief, 4- to 7-mo-old Bbs2^−/− and Bbs6^−/− mice and littermate controls were given a lethal dose of pentobarbital (50–100 mg/mouse ip), and the thoracic aorta was quickly removed and placed in Krebs buffer containing the following (in mmol/l): 118.3 NaCl, 4.7 KCl, 2.5 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, 25 NaHCO₃, 11 glucose. Connective tissue and remaining fat in the adventitia were removed, and the vessel was cut into four equal, straight pieces of 4–5 mm in length. Vascular rings were suspended in a myograph chamber (Multi Wire Myograph System Model 610M; Danish Myo Technology) containing 5 ml of oxygenated Krebs solution maintained at 37°C. The rings were connected to a force transducer via two built-in steel hooks to measure isometric tension (contraction and relaxation). Resting tension was increased stepwise to reach the final tension of 0.5 g, and the rings were allowed to equilibrate for 45 min. This amount of resting tension is optimal for contraction in these arteries as we (5, 6) have shown previously. After an initial precontraction (submaximally, 40–50% of maximum) with PGF_2α (10–50 μmol/l) and relaxation with ACh (10⁻⁴ M) and SNP (10⁻⁴ M) to establish the integrity and functionality of the vessel, dose-response curves were preformed. Vessels were contracted submaximally with PGF_2α as described above. After a stable contraction plateau was reached, dose-response curves were obtained for ACh and SNP (10 nM to 0.3 μM). Contractile response in the aorta to ET-1 (0.1 nM to 0.3 μM), KCl (4–160 mM), and 5-HT (10 nM to 0.3 μM) in a dose-responsive manner were examined. A maximum of five dose-response curves were performed per segment.

Gene expression assays.

Expression of Bbs genes was examined in cultured mouse brain microvascular endothelial (BMVEC) and smooth muscle (BMSMC) cells as well as in the aorta, lung, and stomach of wild-type mice using a standard RT-PCR assay. BMVEC and BMSMC (obtained from Dr. Steven Moore, University of Iowa) were grown in endothelial cell basal medium and smooth muscle cell basal medium, respectively, as described previously (9). Mice were killed by CO₂ asphyxiation, and the tissues were dissected, freeze-clamped, and stored at −80°C. Total RNA was purified by TRIzol method from 100-mm dishes of BMVEC and BMSMC or from 50-mg tissues. cDNA was amplified by reverse transcriptase II (Invitrogen). The following PCR conditions were used: step 1, 95°C, 5 min; step 2, 95°C, 30 s, 55°C, 30 s, 72°C, 1 min, 30 cycles; step 3, 72°C, 10 min. Real-time RT-PCR was carried out with SYBR Green (Bio-Rad) using primers shown in Table 1. Samples were analyzed in 2% agarose gels.

Table 1.

PCR primers used for gene expression studies

Gene	Sense Primer	Antisense Primer
Bbs1	5′-TTGTCTGTGCAGTCACTCAG-3′	5′-CGTCCTCATCTGCCAGGTT-3′
Bbs2	5′-ATGCTGCTGCCTGTGTTCACT-3′	5′-TGCAGAGCACAGTTTCCACC-3′
Bbs3	5′-CCACAATCATTAACAAGCTGA-3′	5′-CTGACATGTCAAACACTGTA-3′
Bbs4	5′-ACTCAGGTTCCTGCATCTGTT-3′	5′-CACTGTGCAGTTGGTCTTGTG-3′
Bbs5	5′-GGAGCATGTGTATGATAAGAT-3′	5′-TCCTCCACTCTGCTGAGAGCTT-3′
Bbs6	5′-AGGACGATACCACTGAGCTGG-3′	5′-CCTCAGACCTCACAG ACTCTT-3′
Bbs7	5′-GTCTTAGCACTCATAATGGA-3′	5′-TCAAAGCTGTCAACACACAAG-3′
Bbs8	5′-CTATCACTAGCTCATCTGGAAG-3′	5′-ACATCGTTTTCATGGTGGAGAA-3′
Bbs9	5′-CCGTGCTGCAAGTGGAAGTC-3′	5′-GGTTTCCATGTTCCACATTTC-3′
Bbs10	5′-CTACCAGAGGTTGCTTCACA-3′	5′-AGAGTGAGGTATAAATGCACATG-3′
Bbs11	5′-TGTCACCTGTGATGCTGAAG-3′	5′-CTTCATTCTCAGAGAAGAAGTGG-3′
Bbs12	5′-GGACAGATAATAGCCACTGG-3′	5′-CTTGCTGCAGACACAGACTCG-3′
M3R	5′-ACACACACACAGAGACTCT CC-3′	5′-CATTGTGACTCTCTGACATAG-3′
β-Actin	5′-CATCCTCTTCCTCCCTGGAGAAGA-3′	5′-ACAGGATTCCATACCCAAGAAGGAAGG-3′

Western blotting.

Bbs2^−/− and Bbs6^−/− mice and littermate controls (4–7 mo old) were killed with CO₂ asphyxiation, and the thoracic aorta was quickly removed. Proteins were extracted by homogenizing the aorta in tissue lysate buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1 mM MgCl₂, 1 mM CaCl₂, 10 mM NaF, 5 mM EDTA, 1% Triton X-100, 2 mM sodium orthovanadate, and Roche cocktail protease inhibitor tablet). Twenty-microgram protein samples were subjected to SDS-PAGE, electrotransferred on a polyvinylidene fluoride membrane, and then probed with primary antibodies at the following concentrations: phospho-eNOS (Ser1177, 1:1,000), eNOS (1:1,000), iNOS (1:1,000), and β-actin (1:10,000), followed by a secondary anti-rabbit antibody (1:10,000). Protein expression was visualized with ECL detection kit.

Data analysis.

All data are expressed as means ± SE. Western blot data were quantified using NIH ImageJ. SigmaStat (Systat Software) was used to analyze the data. For concentration-response curves, comparisons were made with two-way repeated-measures ANOVA followed by a Tukey test or t-test when appropriate. The remaining data were compared by one-way ANOVA followed by Tukey post hoc test. P < 0.05 was considered significant.

RESULTS

Bbs genes are expressed in the vasculature.

To address the potential role of Bbs genes in the vascular function, we first tested whether Bbs genes are expressed in the cells that form the vasculature (smooth muscle and endothelial cells). As shown in Fig. 1, the 12 Bbs genes are expressed with various intensities in cultured endothelial (BMVEC) as well as smooth muscle (BMSMC) cells. These Bbs genes were also expressed with different degrees in the tissues enriched in endothelial (lung) and smooth muscle (stomach) cells as well as in a blood vessel, the aorta (Fig. 1).

Fig. 1.

Bbs genes are expressed in the vasculature. Analysis of the expression of 12 established Bbs genes in mouse brain microvascular endothelial (BMVEC) and smooth muscle (BMSMC) cells and wild-type mice tissues including lung, stomach, and aorta by RT-PCR. M, marker.

Endothelial-dependent relaxation is differentially altered in BBS mice.

To characterize the role of Bbs genes in the control of vascular function, we performed vascular studies in the aorta of Bbs2^−/−, Bbs6^−/−, and wild-type control mice. Wild-type littermates of Bbs2^−/− and Bbs6^−/− mice did not exhibit any difference and, therefore, were combined and used as controls throughout the studies described here. Bbs2^−/− and Bbs6^−/− mice were obese as indicated by the increased body weight and fat mass (Supplemental Fig. S1, available in the data supplement online at the AJP-Heart and Circulatory Physiology web site). We (26) have previously shown that whereas the obese Bbs6^−/− are hypertensive, the obese Bbs2^−/− are normotensive.

Aortic rings of obese normotensive Bbs2^−/− mice showed a significantly enhanced relaxation to ACh, but the maximal relaxations were not different compared with wild-type controls (Fig. 2A). The obese hypertensive Bbs6^−/− mice had a normal response to ACh (Fig. 2A). To test the possibility that alterations in the muscarinic receptors may contribute to the enhanced vasodilatory response to ACh, we used quantitative RT-PCR to compare the expression level of the muscarinic receptor subtype (M₃R) that mediates the vascular effects of ACh (2, 18) in the aorta of BBS and control mice. Consistent with our findings, the expression levels of M₃R were significantly increased in Bbs2^−/− but not Bbs6^−/− mice (Supplemental Fig. S2).

Fig. 2.

Comparison of vascular relaxation between Bbs2^−/−, Bbs6^−/−, and wild-type control mice (WT). Vasodilatation to endothelial-specific dilator ACh (A), endothelial- and NO-specific ionophore A-23187 (B), and endothelial-independent NO donor sodium nitroprusside (SNP; C) were tested in aortic rings from Bbs2^−/− and Bbs6^−/− mice and wild-type littermate controls. *P < 0.05 vs. wild-type mice (n ≥ 8).

In contrast to ACh responses, the endothelium-dependent relaxations induced by the calcium ionophore A-23187 were significantly impaired in Bbs6^−/−, but not Bbs2^−/−, mice (Fig. 2B). On the other hand, the aortic response to an endothelium-independent stimulus (SNP) was unchanged in both BBS mouse models (Fig. 2C) suggesting that the changes in the endothelium-dependent relaxations are not due to smooth muscle dysfunction.

Contractile responses are markedly decreased in BBS mice.

To test whether the ability of the vessel to contract is also altered in BBS mice, we compared the effect of several vasoactive substances on the aorta of Bbs2^−/− and Bbs6^−/− mice and littermate controls. Strikingly, the aortic rings of Bbs2^−/− mice showed a significantly attenuated contractile response to ET-1, serotonin (5-HT), and KCl (Fig. 3, A–C). The vasoconstrictions induced by KCl were also significantly attenuated in Bbs6^−/− mice relative to wild-type controls but less compared with Bbs2^−/− mice (Fig. 3C). In contrast, there was no difference in the responses induced by PGF_2α (Fig. 3D) indicating that the changes in the contractile responses are selective.

Fig. 3.

Characterization of contractile vascular responses in Bbs2^−/−, Bbs6^−/−, and wild-type control mice. Contractions were induced in aortic rings from Bbs2^−/− and Bbs6^−/− mice and wild-type littermate controls using receptor-dependent agonists including serotonin (5-HT; A), endothelin-1 (ET-1; B), and prostaglandin F_2α (PGF_2α; D) or the receptor-independent constrictor potassium chloride (KCl; C). *P < 0.05 vs. wild-type mice, †P < 0.05 vs. wild-type and Bbs2^−/− mice (n ≥ 8).

Absence of structural alterations in the vasculature of BBS mice.

We examined the possibility that structural changes may contribute to the alterations in vascular reactivity in BBS mice. For this, we analyzed the vascular cross-section area, inner and outer diameter, and wall thickness of BBS mice carotids. However, there was no significant difference in all of these parameters between Bbs2^−/−, Bbs6^−/−, and wild-type controls (Supplemental Fig. S3).

Contrasting molecular changes in the vasculature of BBS mice.

To gain insight into the molecular mechanisms that account for vascular changes in BBS mice, we examined potential alterations in the levels of NO-producing enzymes in the vasculature of BBS mice. Given that eNOS is considered as a major source of NO in the vascular wall, we measured the activity level (phosphorylation of Ser1177) of this enzyme in the aorta. As shown in Fig. 4A, compared with wild-type control littermates, aortas of Bbs2^−/− animals exhibited marked increase in phospho-eNOS levels (normalized to total eNOS), whereas Bbs6^−/− mice showed only a modest (not statistically significant) increase. Of note, normalization of phospho-eNOS to β-actin, instead of total eNOS, yielded similar results (wild-type control mice, 100 ± 9%; Bbs2^−/− mice, 162 ± 9%; and Bbs6^−/− mice, 117 ± 13%). Next, we compared the expression (protein level) of another NO-producing enzyme, iNOS, between BBS and control mice. The iNOS level was significantly larger (∼4-fold) in the aortas of Bbs2^−/− but not Bbs6^−/− mice (Fig. 4B). Such upregulation in the NO-producing enzymes in Bbs2^−/− is consistent with the exaggerated ACh-induced vasorelaxation in this mouse model.

Fig. 4.

Expression level of NO-producing enzymes in the aorta of Bbs2^−/−, Bbs6^−/−, and wild-type littermate control mice. Levels of phosphorylated endothelial nitric oxide synthase (p-eNOS) and eNOS (A) and inducible nitric oxide synthase (iNOS) and β-actin (B) were measured by Western blot. Representative Western blot are shown on the top, and quantification results on the bottom. *P < 0.05 vs. wild-type mice (n ≥ 6). AU, arbitrary units.

Oxidative stress is an important causative factor of endothelial dysfunction and is thought to play a major role in the occurrence and complications of the metabolic syndrome (1, 12, 27). Therefore, we evaluated the expression (protein level) of membrane subunits of NADPH oxidase (p22^phox and p47^phox) in the aorta of BBS mice. Of note, whereas the expression of p22^phox and p47^phox was decreased in Bbs2^−/− mice, the expression of these proteins was significantly increased in Bbs6^−/− mice (Fig. 5).

Fig. 5.

Expression level of membrane subunits of NADPH oxidase in the aorta of Bbs2^−/−, Bbs6^−/−, and wild-type littermate control mice. Protein levels of p22^phox and p47^phox were measured in aortas by Western blot, and β-actin was used as loading control. Representative Western blots are shown on the top, and quantification results on the bottom. *P < 0.05 vs. wild-type mice (n ≥ 6).

DISCUSSION

We describe several new findings in the present study. First, we show that Bbs genes are expressed in the endothelial and smooth muscle cells, the tissues enriched in these cells, and the aorta. Second, we found that despite obesity, two mouse models of BBS show no defect in vasodilatation. In fact, the hypertensive Bbs6^−/− showed normal endothelial function, whereas the normotensive Bbs2^−/− model showed increased endothelial-dependent vascular relaxation. Third, aortic rings from both Bbs2^−/− and Bbs6^−/− mice exhibited significant reduction in the contractile responses induced by various stimuli. More specifically, Bbs2^−/− showed decreased contractile response to serotonin, ET-1, and KCl, whereas Bbs6^−/− exhibited a significant decrease in the contraction induced by KCl but no changes in the response to ET-1. Finally, we found contrasting changes in the expression of key molecular pathways involved in vascular regulation in the aorta of Bbs2^−/− and Bbs6^−/− mice. These findings demonstrate that Bbs2 and Bbs6 genes alter differentially the vascular tone in mice.

Despite the presence of obesity in both mouse models studied, Bbs2^−/− and Bbs6^−/− mice showed contrasting vascular changes. However, the enhanced endothelial function (as indicated by the effect of ACh) and reduced contractile responses in Bbs2^−/− mice are consistent with the lack of hypertension in this mouse model (26). On the other hand, the vascular changes we described in Bbs6^−/− mice are in line with the presence of hypertension in this model. These changes in vascular reactivity may contribute to the difference in arterial pressure in these two mouse models.

Similar to our findings in mice, in humans several BBS genotypes including BBS6 were found to be accompanied by hypertension except for patients with BBS2 who remain normotensive (20). In addition, variants of Bbs6, but not Bbs2, gene have been reported to increase the risk of raised blood pressure in common obese individuals (3). Thus elucidation of the mechanisms that protects some obese patients from the development of endothelial dysfunction and hypertension as in BBS2 will help understand the pathophysiology of obesity-associated vascular abnormalities and high blood pressure.

We identified several molecular changes that may account for the contrasting vascular changes in BBS mice. We show that aorta of Bbs2^−/− mice exhibit higher expression levels of the muscarinic receptor subtype (M₃) that mediates the endothelium-dependent relaxation caused by ACh (2, 18). This finding is consistent with enhanced ACh-induced vasodilatation in this mouse model. We further show increased levels of eNOS phosphorylation in position Ser1177 in the aorta of Bbs2^−/− mice. Although regulation of eNOS activity is complex, phosphorylation of Ser1177 of eNOS is known to enhance the activity of the enzyme leading to an elevation of NO production (11, 15). Bbs2^−/− mice also exhibited a large (∼4-fold) increase in iNOS expression in the aorta. Elevated iNOS expression was previously reported in the aorta of diet-induced obese mice (25). The expression profile of NO-producing enzymes in Bbs2^−/− mice may contribute to improved endothelial function and reduced receptor-mediated contractile responses in this mouse model. NO is critical for vascular relaxation and considered as the major endothelium-derived relaxing factor in the conduit vessels such as the aorta (14, 32). Release of NO by the vascular endothelium causes the vascular smooth muscle to relax by activating soluble guanylate cyclase, increasing the cGMP concentration, and activating the cGMP-dependent protein kinase G, resulting in vasodilation (21, 32). However, an increase in iNOS such as that observed in the aorta of Bbs2^−/− mice may promote inflammation (7, 8). Such an inflammatory state may offset the beneficial effects of higher eNOS leading to normal arterial pressure in this mouse model.

The expression level of NADPH oxidase, a key reactive oxygen species-producing enzyme, is a critical determinant of vascular function (14, 16). Of note, vascular NADPH oxidase appears to be a major source of increased oxidative stress in obesity (31). We found that the expression level of membrane subunits of NADPH oxidase (p22^phox and p47^phox) in the aorta was decreased in the normotensive Bbs2^−/− mice but increased in the hypertensive Bbs6^−/− animals.

We recognize several limitations to the present studies that need to be addressed. First, vascular functions were examined ex vivo. Whether the changes we found using aortic rings sustain in vivo remain to be determined. Further studies are also needed to assess the relevance of the vascular reactivity to the control of arterial pressure in BBS mice. Second, we performed molecular assays using RNA and proteins obtained from the whole vessel. The contributions of the various cell types that form the vessel to the changes we report require further investigation. However, we demonstrated that Bbs genes are expressed in the cells that are critical for the control of vascular tone (smooth muscle and endothelial cells). Third, although some of the molecular alterations we found appear consistent with the variations in vascular reactivity and arterial pressure in the two BBS mouse models, additional studies are needed to functionally link the alterations in the expression of eNOS, iNOS, and subunits of NADPH oxidase to the changes in vascular reactivity and arterial pressure. Finally, additional studies are also required to determine the mechanisms by which loss of Bbs2 and Bbs6 genes lead to contrasting molecular changes in the vasculature.

In conclusion, our data implicate Bbs genes in the regulation of vascular tone and demonstrate that disrupting Bbs2 and Bbs6 genes alter differentially the vascular function in mice. The vascular changes may contribute to the contrasting blood pressure effects caused by mutations in these Bbs genes.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grant HL-084207. A. M. Beyer was supported by NIH National Research Service Award (NRSA) Research Fellowship HL-07121. V. C. Sheffield is an investigator of the Howard Hughes Medical Institute.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).