Low blood concentrations of 25-hydroxyvitamin D3 are associated with increased mortality, while some studies suggest improved cardiovascular outcomes with vitamin D3 supplementation in chronic kidney disease. However, the physiological effects of vitamin D3 on the cardiovascular system remain poorly understood making it difficult to determine whether vitamin D3 supplementation might provide cardiovascular benefit or even cause harm. Thus here we investigated the effects of chronic 1,25-dihydroxyvitamin D3 treatment on intracellular signaling in human coronary artery smooth muscle cells (HCASMCs) and found that 1,25-dihydroxyvitamin D3 significantly potentiated endothelin (ET-1) signaling. Specifically, 1,25-dihydroxyvitamin D3 (24-h pretreatment) caused a more than threefold enhancement in both ET-1-induced intracellular calcium mobilization and extracellular signal-regulated kinase (ERK) activation. This 1,25-dihydroxyvitamin D3-elicited signaling enhancement was not observed for either vasopressin or carbachol. With the use of endothelin receptor (ETR) isoform-selective antagonists, ETRA was found to be primarily responsible for the 1,25-dihydroxyvitamin D3-induced ET-1 responsiveness and yet ETRA mRNA expression and protein abundance were unaltered following 1,25-dihydroxyvitamin D3 treatment. While there was an increase in ETRB mRNA expression in response to 1,25-dihydroxyvitamin D3, the protein abundance of ETRB was again unchanged. Finally, ETRA/ETRB heterodimerization was not detected in HCASMCs in either the absence or presence of 1,25-dihydroxyvitamin D3. Together, these data show for the first time that 1,25-dihydroxyvitamin D3 enhances endothelin responsiveness in HCASMCs and that the effect is mediated through ETRA.
- vitamin D3
- coronary artery smooth muscle cells
- intracellular calcium
- extracellular signal-regulated kinase
it is well known that 1,25-dihydroxyvitamin D3 [1,25(OH)2D3 or calcitriol] elicits both calciotropic effects, including maintenance of calcium and phosphate homeostasis, as well as a wide range of noncalciotropic effects on target tissues, including the cardiovascular system (7, 11, 13, 31). Epidemiological studies suggest that low plasma vitamin D3 concentrations predispose a person to cardiovascular events (15, 16, 27) and that oral vitamin D3 supplementation may improve all cause mortality including by decreasing cardiovascular deaths (2). However, whether vitamin D3 supplementation should be recommended more widely to provide cardiovascular benefit remains controversial and large-scale clinical outcome trials are currently underway to address this (22, 25). In addition, there remains insufficient understanding of the cellular and physiological effects of 1,25(OH)2D3 in the vasculature and such information is necessary to define the likely physiological or therapeutic mechanism(s) of cardiovascular action for this endogenous hormone.
Vascular smooth muscle cells (VSMCs) express vitamin D receptors the activation of which has been shown in some studies to alter vascular contractility and calcification, as well as to affect VSMC proliferation and migration, and cytosolic calcium concentration (reviewed in Ref. 31). Since 1,25(OH)2D3 is generally associated with maintaining a low intracellular Ca2+ (Ca2+i) concentration in cells, particularly in epithelia responsible for active Ca2+ transport, then it is important to understand the effect of 1,25(OH)2D3 on Ca2+i signaling in human VSMCs.
Endothelin-1 (ET-1) is the most powerful vasoconstrictor peptide in the cardiovascular system and is produced largely by endothelial cells (32). It binds to two class-A G protein-coupled receptor (GPCR) subtypes, namely endothelin receptor A (ETRA) and B (ETRB) (21). The stimulation of ETRA leads to increased cardiac inotropy and vasoconstriction whereas ETRB stimulation results in vasodilation and ET-1 clearance (5, 10). ET-1 has been implicated in the pathogenesis of a variety of cardiovascular diseases not least atherosclerosis, the development of which involves several cell types including VSMCs (14). Acute ETRA stimulation elicits phospholipase C-mediated inositol trisphosphate formation leading to a rapid mobilization of Ca2+ from the sarcoplasmic reticulum, which in turn causes VSMC contraction and vasoconstriction (14). ET-1 is also associated with activation of the extracellular signal-regulated kinases (ERKs), ERK1 and ERK2, which are proline-directed protein kinases involved in proliferation and other cell functions. Thus, to better understand the action of vitamin D3 on vascular cell physiology, our objective in the current study was to investigate the effect of 1,25(OH)2D3 pretreatment on ET-1 signaling in human coronary artery smooth muscle cells (HCASMCs).
Human coronary artery cell culture.
HCASMCs (Life Technologies) were grown in M-231 medium including smooth muscle growth supplement (GIBCO) in the absence of antibiotics or antimycotics at 37°C in a humidified 5% CO2 incubator. HCASMCs, used between passage 3 and 8, were grown to ∼80% confluence on glass coverslips and then treated with or without 1,25(OH)2D3 for 24 h (unless otherwise stated). All chemicals were from Sigma unless otherwise stated.
Intracellular calcium assay.
HCASMCs cultured on glass coverslips were loaded with fura-2/AM (10 μM for 45 min) at room temperature in the dark in Ca2+ assay buffer (comprising in mM: 125 NaCl, 4 KCl, 0.5 CaCl2, 0.5 MgCl2, 5.5 glucose, and 20 HEPES pH 7.4) supplemented with 0.1% BSA. At the beginning of the experiment, cells were transferred into a perfusion chamber and washed with Ca2+ assay buffer (containing 0.5 mM Ca2+) before treatment. Cells were observed through a ×40 oil-immersion objective. Dual-excitation wave-length microfluorometry was performed using a Nikon Diaphot inverted microscope (Metafluor software). To estimate Ca2+i store content, cells incubated in extracellular Ca2+ (Ca2+o)-free buffer (+1 mM EGTA) were exposed to 1 μM thapsigargin with the area under the curve of the resulting rise in Ca2+i concentration quantified.
ERK phosphorylation assay.
Cells were grown to ∼80% confluence in 35-mm culture dishes and rinsed in PBS for 5 min before equilibration for 25 min in experimental buffer at 37°C. Following culture in the absence or presence of 100 nM 1,25(OH)2D3, cells were then incubated for up to 10 min in buffer supplemented with various experimental treatments and then lysed on ice in the following RIPA buffer: 12 mM HEPES (pH 7.6), 300 mM mannitol, 1% (vol/vol) Triton X-100, 0.1% (wt/vol) sodium dodecyl sulfate supplemented with 1.25 μM pepstatin, 4 μM leupeptin, 4.8 μM PMSF, 1 mM EDTA, 1 mM EGTA, 100 μM vanadate, 1 mM NaF, and 250 μM sodium pyrophosphate. Lysate was then mixed with 5× Laemmli buffer and boiled for 5 min before immunoblotting using phospho-specific anti-ERK monoclonal antibody (1:2,000 dilution; Cell Signaling). Total ERK levels were demonstrated using anti-ERK2 antibody (Santa Cruz).
Immunoblotting was performed as previously described (28). Briefly, cell lysates were prepared using RIPA buffer and boiled for 5 min in Laemmli buffer. Protein samples were then resolved by SDS-PAGE and transferred electrophoretically to nitrocellulose. Nonspecific binding sites were blocked using 2% (wt/vol) BSA or 5% nonfat milk solutions (according to antibody protocol), followed by a 1-h incubation in either anti-ETRA rabbit polyclonal antibody (1:1,000 dilution; Santa Cruz Biotechnology), anti-pERK monoclonal antibody (1:3,000 dilution; Chemicon), anti-ETRB rabbit polyclonal antibody (1:1,000 dilution; Chemicon), anti-sarco/endoplasmic reticulum Ca2+-ATPase (anit-SERCA; 1:1,000 dilution; Cell Signaling), or, anti-β-actin polyclonal antibody (1:5,000 dilution; Cell Signaling). After being washed, blots were exposed to either horseradish peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies (1:5,000) for 1 h and then developed using ECL reagents (Geneflow). All washes and incubations used Tween/TBS solution [15 mm Tris (pH 8), 150 mm NaCl, and 0.1% (vol/vol) Tween 20]. Immunoreactivity was detected using light-sensitive paper and quantified by densitometry.
Cell lysates were prepared as above (in the absence of SDS) and then mixed with protein A-agarose beads and incubated with rotation for 30 min at 4°C. The precleared supernatant (50 μl) was then mixed with ETRB antibody (1 μl) and incubated as before for 1 h followed by overnight incubation in the presence of protein A-agarose beads. The supernatant was preserved, and the beads were then washed three times, boiled for 5 min in Laemmli buffer and the resulting immunopreciptates and supernatants processed for immunoblotting using anti-ETRB antibody. The membranes were then stripped and reprobed using anti-ETRA antibody.
Total RNA was collected from HCASMCs (Qiagen RNeasy mini kit) including an RNase-free DNase (Qiagen) step, followed by cDNA synthesis (Bioline). Quantitative real-time PCR (qPCR) was performed in an Applied Biosystems 7500 Real Time PCR System using gene-specific primers [ETRA, ETRB, and phospholamban (PLN)] from Primer Design (PrimerDesign, Southampton, UK). The reaction components comprised SYBR Green Master mix (10 μl), 3′-primer (final concentration 300 nM), 5′-primer (final concentration 300 nM), cDNA (5 ng/μl), and nuclease-free water (to final volume of 20 μl). The protocol included a 10-min initial activation at 95°C, 15-s denaturation at 95°C, and data collection at 60°C (50 cycles).
Data are presented as means ± SE. Statistical significance was determined by one-way or repeated-measures ANOVA or by unpaired t-test, using GraphPad Prism.
Effect of 1,25(OH)2D3 on endothelin responsiveness in HCASMCs.
Pretreatment of HCASMCs with 1,25(OH)2D3 (100 nM) for 24 h significantly increased the responsiveness of the cells to subsequent endothelin (ET-1; 100 nM) exposure (Fig. 1A). Specifically, ET-1 elicited significantly greater Ca2+i mobilization following 24-h 1,25(OH)2D3 pretreatment than in its absence. This increased endothelin responsiveness appeared independent of Ca2+o concentration, being observed in either 0.5, 1.2, or 2 mM Ca2+o (data not shown) and was also observed using only 10 nM ET-1 (Fig. 1Bi). Next, to assess whether the 1,25(OH)2D3-induced increase in endothelin responsiveness is specific to this particular G protein-coupled receptor agonist or represents a more general phenomenon, the cells were alternatively stimulated acutely with either vasopressin (100 nM; Fig. 1Bii) or carbachol (100 μM; Fig. 1Biii). In these experiments, 1,25(OH)2D3 pretreatment failed to increase vasopressin or carbachol responsiveness significantly.
Effect of 1,25(OH)2D3 on endothelin responsiveness as a function of concentration and time.
The concentration of 1,25(OH)2D3 commonly used in cell studies is 100 nM, for example, to change global gene expression in HCASMCs (31). Thus we next determined the concentration dependency of the 1,25(OH)2D3 response. Exposure of the cells to increasing concentrations of 1,25(OH)2D3 (up to 100 nM, 24 h) did result in a concentration-dependent increase in ET-1 responsiveness (Fig. 2A), with 20 nM 1,25(OH)2D3 being sufficient to elicit a significant effect. Next, cells were incubated with 1,25(OH)2D3 (100 nM) for either 24 or 48 h and then stimulated with ET-1 (100 nM; Fig. 2B). Treatment with 1,25(OH)2D3 for 24 h elicited the greatest effect although even at 48 h; the endothelin responsiveness was still elevated.
Identification of the receptor subtype mediating endothelin-induced calcium mobilization.
To determine whether ETRA and/or ETRB are responsible for the 1,25(OH)2D3-induced endothelin responsiveness, HCASMCs were stimulated with endothelin (100 nM) in the absence or presence of antagonists to ETRA (BQ123) and ETRB (BQ788). Cells were pretreated with either 100 nM or 3 μM of the antagonists for 10 min and then cotreated with the endothelin. The ETRA antagonist BQ123 largely abolished the response to ET-1 in 1,25(OH)2D3-treated cells (Fig. 3A). In contrast, the ETRB antagonist BQ788 failed to significantly inhibit the same ET-1 response (Fig. 3B). Consistent with this, the ETRB-selective agonist sarafotoxin S6c (100 nM) was without effect, despite subsequent ET-1 treatment eliciting a similar response to before (Fig. 3C). Together, these results suggest that ETRA represents the main contributor to the stimulatory effect of 1,25(OH)2D3 on ET-1 responsiveness.
Effect of 1,25(OH)2D3 pretreatment on endothelin-induced ERK activation.
Next, it was necessary to determine whether the increased endothelin responsiveness reported here is limited to changes in Ca2+i mobilization or whether other downstream signaling pathways are similarly affected. To address this, the effect of 1,25(OH)2D3 on endothelin-induced ERK (extracellular signal-regulated) activation was then investigated. ET-1 produced a concentration-dependent increase in ERK phosphorylation in HCASMCs pretreated either with or without 1,25(OH)2D3; however, endothelin responsiveness was significantly greater in the 1,25(OH)2D3-pretreated cells (Fig. 4A). In particular, the responses to 1 and 10 nM ET-1 were markedly elevated following 24-h 1,25(OH)2D3 pretreatment. The time dependence of the ERK responses to ET-1 was similar in both control and 1,25(OH)2D3-treated cells, that is, they both peaked at 5 min (Fig. 4B) and declined to baseline by 20 min; however, again, the cells treated with 1,25(OH)2D3 exhibited significantly larger responses to ET-1 at 2, 5, and 10 min.
Next, to determine which receptor subtypes are responsible for the ET-1-induced ERK activation, cells were cotreated with either BQ123 or BQ788. Unlike in the experiments investigating Ca2+i mobilization, neither antagonist inhibited ET-1-induced ERK activation when used at 100 nM concentration (Fig. 5). However, in the presence of 1 μM BQ123 the ERK response to ET-1 was completely ablated whereas BQ788 was still without effect. Therefore, for both Ca2+i mobilization and ERK activation, the 1,25(OH)2D3-enhanced ET-1 responses in HCASMCs appear to be mediated via ETRA.
Effect of 1,25(OH)2D3 on ETR expression in HCASMCs.
Next, total RNA and protein lysates were collected from HCASMCs cultured in the absence or presence of 1,25(OH)2D3 (100 nM, 24 h) and the relative expression of ETRA and ETRB determined by qPCR and immunoblotting. Despite the apparent increase in ETRA-mediated endothelin responsiveness elicited by 1,25(OH)2D3, no change in ETRA expression was observed either at the transcriptional level (Fig. 6A) or in terms of protein abundance (Fig. 6B). In contrast, despite the apparent lack of ETRB involvement in endothelin responsiveness (Figs. 3 and 5), qPCR analysis revealed a 150% increase in ETRB expression in the 1,25(OH)2D3-treated HCASMCs (Fig. 6A). However, the ETRB protein abundance was unchanged by the 1,25(OH)2D3 pretreatment (Fig. 6B). Next, to determine whether the apparent increase in ETRA responsiveness is due to altered ETRA/ETRB heterodimerization, ETRB was immunoprecipitated from HCASMCs pretreated with or without 1,25(OH)2D3 and then probed with anti-ETRA antibodies. However, no ETRA was detected in the ETRB precipitates under either condition, suggesting that (altered) heterodimerization does not explain 1,25(OH)2D3-enhanced ET-1 responsiveness. Finally, to investigate whether 1,25(OH)2D3 may be modulating the Ca2+i stores more directly, we examined the effect of 1,25(OH)2D3 pretreatment on the expression in HCASMCs of the SERCA pump and of its negative regulator PLN as well as on the relative amounts of thapsigargin-mobilizable stored Ca2+i within the cell. 1,25(OH)2D3 (100 nM, 24 h) failed to alter protein expression of SERCA (Fig. 6Di) but did inhibit PLN gene expression by 80% (Fig. 6Dii), consistent with a previous microarray study (30); however, this could not be confirmed at the protein level. If PLN downregulation did occur in the presence of unaltered SERCA expression, then we might expect to see increased Ca2+i store filling (due to decreased SERCA inhibition) and interestingly the total amount of stored Ca2+i in the HCASMCs estimated using thapsigargin treatment (in the absence of extracellular Ca2+) was indeed significantly greater in cells pretreated for 24 h with 1,25(OH)2D3 (Fig. 6Diii).
In HCASMCs, ET-1 elicited increases in Ca2+i concentration and ERK phosphorylation consistent with activation of ETRs on the cell membrane. Specifically, endothelin exerts its vasoactive effects via two GPCRs, namely ETRA and ETRB both of which can couple to Gαq/11-induced PLC activation leading to Ca2+i mobilization and also to the Raf/MEK/ERK pathway (see 20). In the current study, the presence of both ETRA and ETRB in the HCASMCs was confirmed both by qPCR and immunoblotting. Furthermore, with the use of isoform-selective antagonists, BQ123 for ETRA and BQ788 for ETRB, it was found that ETRA mediated the responses of ET-1 on Ca2+i mobilization and ERK activation in the HCASMCs.
Next, 24-h exposure to 1,25(OH)2D3 significantly and consistently enhanced ET-1-induced ERK activation and Ca2+i mobilization in HCASMCs. To our knowledge, this represents the first report of 1,25(OH)2D3-mediated stimulation of endothelin responsiveness. In contrast, 1,25(OH)2D3 failed to alter ET-1-induced ERK activation in neonatal rat VSMCs (4), while in rat osteoblastic osteosarcoma cells, 1,25(OH)2D3 actually lowered ET-1-induced inositol phosphate metabolism (18). Interestingly, the 1,25(OH)2D3-induced increase in ET-1-mediated Ca2+i mobilization seen here in HCASMCs appeared relatively specific to endothelin since it failed to increase the responsiveness of vasopressin or carbachol, vasoactive substances that also induce Gαq/11-coupled GPCRs to increase Ca2+i concentration (19, 26).
The apparent specificity of the 1,25(OH)2D3 effect for ET-1 responsiveness could be potentially explained by increased expression of ETRA or perhaps ETRB or other downstream signaling mediators. However, there was no change in the gene expression of ETRA and neither was there a change in the protein abundance of either ETRA or ETRB. Interestingly, there was a significant increase in ETRB gene expression, but in the absence of a change in the resulting protein abundance, this may then reflect increased protein turnover rather than a sustained rise in the number of ETRB molecules present at any time. Therefore, the heightened responsiveness to ET-1 induced by 1,25(OH)2D3 cannot be explained by gross changes in ETRA expression. Another possible explanation is that 1,25(OH)2D3 somehow influences heterodimerization of ETRA/ETRB such that it alters ET-1 responsiveness but without changing receptor expression levels. Indeed, heterodimerization of ETRA and ETRB has been demonstrated by fluorescence resonance energy transfer analysis (9), while in ETRA-transfected HEK-293 cells, coexpression with ETRB causes the ETRA signaling to become more sustained (8). However, in the current study, there was no evidence of ETRA/ETRB heterodimerization in either control or 1,25(OH)2D3-treated cells, possibly consistent with the differential subcellular localization of ETRA and ETRB seen recently in human VSMCs (3). Therefore, we were unable to identify the mechanism by which 1,25(OH)2D3 increased ET-1 responsiveness although we were at least able to rule out changes in ETRA expression, ETRA/ETRB heterodimerization, or a general increase in GPCR responsiveness. However, we were further able to clarify that 1,25(OH)2D3 does not alter SERCA expression but we did observe decreased gene expression of PLN together with an apparent increase in the amount of thapsigargin-mobilizable Ca2+i stored within the cells. It should be noted that the area under the curve measurements used herein approximate to, although are not definitive for, Ca2+i store content as they are also affected by plasma membrane Ca2+-ATPase activity. Nevertheless, these data do point to a possible effect of 1,25(OH)2D3 pretreatment on the Ca2+i handling machinery in HCASMCs.
Although the concentration of 1,25(OH)2D3 used here in vitro is higher than its plasma concentration in vivo, it should be noted that VSMCs express 1α-hydroxylase and can actually synthesize 1,25(OH)2D3 (1, 12, 17, 24, 33), in addition to that delivered via the bloodstream. Therefore, it remains to be determined whether the 1,25(OH)2D3-induced increase in endothelin responsiveness reported here represents the physiological consequence of vitamin D3 repletion or is instead more indicative of the pharmacological consequence of 1,25(OH)2D3 therapy. It is interesting in this regard that the therapeutic use of calcitriol and its analogs is associated with a bimodal, U-shaped relationship between plasma 1,25(OH)2D3 concentration and its subsequent vascular effects (23), such that both low and high 1,25(OH)2D3 concentrations can result in increased intima-media thickness and coronary artery calcification (23). Similarly, with regards to ET-1 sensitivity in the HCASMCs, 1,25(OH)2D3 pretreatment increased responsiveness to 1 nM ET-1 by ERK assay and to 10 nM ET-1 by Ca2+i assay. This is consistent with most vascular studies in vitro, which require low-to-moderate nM concentrations of ET-1 to elicit functional responses (6, 29) despite circulating ET-1 levels being in the low-to-moderate pM range in vivo. It is unclear whether this consistent discrepancy is due to the lack of endogenous positive allosteric modulators in the simple salt buffers used in vitro (not least ET-2) or to the absence of signal crosstalk from other circulating growth factors, neurotransmitters or integrins.
Currently, the potential cardiovascular benefit of vitamin D analog therapy remains controversial (22) with a recent study finding that it failed to alter left ventricular mass index or improve certain measures of diastolic dysfunction in chronic kidney disease patients (25). However, a number of large-scale clinical outcome trials are underway to resolve this issue conclusively. In any case, as an endogenous circulating hormone it is important to understand the physiological cellular consequences of 1,25(OH)2D3 on the vasculature. Thus, in light of our observation that 1,25(OH)2D3 heightens HCASMC responsiveness to the endogenous vasoactive peptide endothelin, it will now be necessary to determine whether these effects can also be observed in vivo.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: M.A. and D.T.W. conception and design of research; M.A. performed experiments; M.A. analyzed data; M.A. interpreted results of experiments; M.A. drafted manuscript; M.A. and D.T.W. approved final version of manuscript; D.T.W. prepared Figs.; D.T.W. edited and revised manuscript.
We thank A. Canfield for providing the HCASMCs. M. Absi was a recipient of a United Nations Educational, Scientific and Cultural Organization-L′Oreal Women in Science Fellowship.
- Copyright © 2013 the American Physiological Society