Heme oxygenase-1 regulates cell proliferation via carbon monoxide-mediated inhibition of T-type Ca2+ channels
Induction of the antioxidant enzyme heme oxygenase-1 (HO-1) affords cellular protection and suppresses proliferation of vascular smooth muscle cells (VSMCs) associated with a variety of pathological cardiovascular conditions including myocardial infarction and vascular injury. However, the underlying mechanisms are not fully understood. Over-expression of Cav3.2 T-type Ca2+ channels in HEK293 cells raised basal [Ca2+]i and increased proliferation as compared with non-transfected cells. Proliferation and [Ca2+]i levels were reduced to levels seen in non-transfected cells either by induction of HO-1 or exposure of cells to the HO-1 product, carbon monoxide (CO) (applied as the CO releasing molecule, CORM-3). In the aortic VSMC line A7r5, proliferation was also inhibited by induction of HO-1 or by exposure of cells to CO, and patch-clamp recordings indicated that CO inhibited T-type (as well as L-type) Ca2+ currents in these cells. Finally, in human saphenous vein smooth muscle cells, proliferation was reduced by T-type channel inhibition or by HO-1 induction or CO exposure. The effects of T-type channel blockade and HO-1 induction were non-additive. Collectively, these data indicate that HO-1 regulates proliferation via CO-mediated inhibition of T-type Ca2+ channels. This signalling pathway provides a novel means by which proliferation of VSMCs (and other cells) may be regulated therapeutically.
KeywordsHeme oxygenase Carbon monoxide Calcium channel Proliferation Vascular smooth muscle
Vascular smooth muscle cells (VSMCs) control vascular tone (and hence blood flow and distribution) through regulated contraction which is highly dependent on Ca2+ influx, primarily via voltage-dependent L-type Ca2+ channels [4, 21, 33, 48, 50, 54]. VSMCs are not terminally differentiated and can undergo adaptive phenotypic changes: their ability to become non-contractile, proliferative cells is an important factor in both developmental vasculogenesis and vascular repair [35, 36, 52]. However, this switch to a proliferative state is also important in pathological situations such as atherosclerosis, restenosis, neointimal hyperplasia and hypertension [19, 36]. Given the impact of such cardiovascular diseases on global health, a greater understanding of the mechanisms underlying this phenotypic change in VSMCs has tremendous potential to reveal novel therapeutic strategies.
Although T-type Ca2+ channels are expressed in VSMCs, their role in vasoconstriction is unclear (see ). However, in proliferating VSMCs, whilst L-type Ca2+ channel expression decreases, T-type Ca2+ channel expression has long been known to increase [26, 42], and Ca2+ influx via T-type Ca2+ channels appears to be required for proliferation in vitro and neointima formation following vascular injury [26, 29, 43, 45]. The implication of a role for T-type Ca2+ channels has often been based on the use of mibefradil [29, 45], which is now known to exert effects on targets other than T-type Ca2+ channels (e.g. ). However, more recent molecular approaches have confirmed this class of channel as being of primary importance in vascular proliferation [43, 47].
Heme oxygenase (HO) enzymes catalyse the degradation of heme to biliverdin, Fe2+ and carbon monoxide (CO). Whilst HO-2 is constitutively active and widely distributed, HO-1 is induced by a variety of cellular stresses [23, 44] and regarded as protective, since heme itself is pro-oxidant, and biliverdin is rapidly converted to the powerful antioxidant, bilirubin. HO-1 induction affords protection in a variety of pathological cardiovascular conditions including myocardial infarction, hypertension, atherosclerosis and vascular injury, many of which involve VSMC proliferation [8, 44]. Indeed, HO-1 is established as being anti-proliferative, and CO may account for many of the effects of HO-1 in VSMCs [12, 13, 34]: inhalation of CO can inhibit the proliferation of VSMCs in intimal hyperplasia following vessel grafting [34, 41] and CO inhalation, as well as CO-releasing molecules (CORMs), are being developed for future cardiovascular therapy  despite the detailed mechanisms underlying its anti-proliferative effect remaining unknown. In recent years, we and others have suggested that specific ion channels are targets of regulation by CO and that their modulation may account for some of the important actions of CO [11, 22, 46, 53].
We have most recently demonstrated that recombinant and native neuronal T-type Ca2+ channels are inhibited by CO . In the present study, we have investigated the potential role of T-type Ca2+ channel regulation by CO on cellular proliferation.
A7r5 cells (a smooth muscle cell line derived from rat thoracic aorta ) were obtained from the European Collection of Cell Cultures (ECACC, Public Health England, Porton Down, UK). They were grown in A7r5 complete media, consisting of Dulbecco's Modified Eagle Medium (DMEM) containing 10 % foetal bovine serum (FBS) (Biosera, Ringmer, UK) and 1 % glutamax (Gibco, Paisley, UK). Cells were kept in a humidified incubator at 37 °C (95 % air: 5 % CO2) and passaged weekly.
Human saphenous vein smooth muscle cells (HSVSMCs)
Smooth muscle cells were isolated from the saphenous vein (SV) of anonymous patients undergoing coronary bypass graft surgery at Leeds General Infirmary following ethical approval and informed patient consent. Segments of SV, around 1 cm in length, were denuded of endothelium and adventitia and were cut open longitudinally, lumen facing upwards. The segment was then divided into two pieces. Two milliliters of complete medium (DMEM containing 10 % (v/v) FBS (Biosera, Ringmer, UK), 1 % (v/v) L-glutamine and 1 % (v/v) penicillin/streptomycin (Gibco, Paisley, UK; unless otherwise stated)) were transferred into a clean petri dish and a segment of vein placed in the media. This was then cut with a razorblade into fragments around 0.5 mm2 in size. This tissue and media mixture was then transferred to a 25-cm2 tissue culture flask and maintained in a humidified atmosphere (37 °C; 95 % air: 5 % CO2). Cells migrated out from these tissue fragments within 7–10 days, and when 80–90 % confluent, the cells were plated for experimentation. HSVSMCs were used at passages between P1 and P6.
Wild type (WT; untransfected) HEK293 cells were cultured in minimum essential medium containing Earle’s salts and L-glutamine and supplemented with 10 % (v/v) foetal bovine serum (Biosera, Ringmer, UK), 1 % (v/v) non-essential amino acids, 1 % (v/v) antibiotic/antimycotic and 0.1 % (v/v) gentamicin. HEK293 cells stably expressing Cav3.2 T-type Ca2+ channels (a kind gift from Prof. E. Perez-Reyes; University of Virginia, VA, USA) were cultured in WT HEK293 media, additionally supplemented with 1 mg/ml G-418 to maintain selection pressure (all reagents from Gibco, Paisley, UK; unless otherwise stated). HEK293/Cav3.2 cells were used at passages between P1 and P8, and WT HEK293 cells were used at passages between P1 and P12; both cell types were kept in a humidified incubator at 37 °C (95 % air: 5 % CO2) and passaged weekly.
Cells were plated in 24-well plates in complete media at 1 × 104 cells per well. HSVSMCs were allowed to adhere overnight and subjected to serum free media (SFM) for 2.5 days. A7r5 and HEK293 cells were allowed to adhere for 6 h and then subjected to SFM overnight. On day 0 of the assay, SFM was removed and 1 ml of the relevant complete media was added to each well, in addition to the required drug or compound being investigated. To count cells, media was removed, cells were washed with 1 ml of Dulbecco’s phosphate buffered saline (PBS) and 200 μl of 0.05 % trypsin-EDTA (Gibco, Paisley, UK) was added (pre-warmed to 37 °C). Post-incubation, 800 μl of complete media was added and the cell suspension centrifuged (600g for 6 min). Following removal of 950 μl of media, 50 μl of supernatant remained with the cell pellet, which was then re-suspended with 50 μl of 0.4 % trypan blue (Thermo Scientific, Rockford, USA) to exclude unviable cells. Media was retained from one well of each treatment, processed in the same manner as the cell samples, and any cells present were counted as an additional quantification of non-viable cells. Day 0 counts and media counts were performed using a hemocytometer. All other counts were performed using a TC10 automated cell counter (Bio-Rad, Hemel Hempstead, UK).
HSVSMCs, WT HEK293 and HEK293/Cav3.2 cells were grown to 80 % confluence in 6-well plates. The wells were replenished with 0.4 % serum-containing media plus the required concentration of cobalt protoporphyrin IX (CoPPIX). Post-treatment, the cells were washed with PBS and lysed via incubation for 30 min with 200 μl mammalian protein extraction reagent (M-PERTM; Thermo Scientific, Rockford, USA) containing complete mini protease inhibitors (Roche Diagnostics Ltd., Lewes, UK). Cell lysates were retrieved and protein levels determined using a BCA protein assay kit according to manufacturers’ instructions (Thermo Scientific, Rockford, USA). Protein (10–20 μg) containing 2× sample buffer (250 mM Tris/HCl, pH 6.8, 4 % (w/v) SDS, 20 % (w/v) glycerol, 1 % bromophenol blue and 10 % β-mercaptoethanol) was loaded onto 12.5 %, 0.75-mm-thick polyacrylamide-sodium dodecyl sulphate gels and separated for ~1 h at 35 mA before being transferred onto 0.2 μm polyvinyl difluoride membranes at 30 V overnight. Membranes were blocked using 5 % (w/v) non-fat dried milk powder in tris buffered saline (TBS)-tween (0.05 %) for 1 h, then incubated with rabbit anti-HO-1 antibody raised against amino acids 184–288 of human HO-1 (SC-10789; Santa Cruz, Dallas, USA) at 1:200 for 3 h at room temperature (21–24 °C). Mouse anti-β-actin raised against the N-terminal of β-actin (Sigma, Gillingham, UK) was used as a loading control at 1:4,000. The membranes were then washed in TBS-tween (0.05 %) and incubated with the corresponding anti-rabbit or anti-mouse peroxidase-conjugated secondary antibody (GE Healthcare, Amersham, UK) at 1:2,000 for 1 h at room temperature. Protein bands were detected using the enhanced chemi-luminescent method (GE Healthcare, Amersham, UK) on hyperfilm. Densitometric analysis was performed using Image J (NIH UK).
Ca2+ currents were recorded from A7r5 cells using the whole-cell configuration of the patch-clamp technique at room temperature (21–24 °C) as previously described  using an Axopatch 200A amplifier/Digidata 1300 interface controlled by Clampex 9.0 software (Molecular Devices, Sunnyvale, CA, USA). Offline analysis was performed using Clampfit 9.0. Pipettes (4–6 MΩ) were filled with (in mM) the following: CsCl 120, MgCl2 2, EGTA 10, TEA-Cl 20, HEPES 10, Na-ATP 2 and pH 7.2 (adjusted with CsOH). To optimise recording of T-type Ca2+ currents, cells were perfused with (in mM) the following: NaCl 95, CsCl 5, MgCl2 0.6, CaCl2 15, TEA-Cl 20, HEPES 5, D-glucose 10 and pH 7.4 (adjusted with NaOH). Cells were voltage-clamped at −80 mV and either repeatedly depolarized to −20 mV (200 ms, 0.1 Hz) or to a series of test potentials ranging from −100 to +60 mV. To record L-type Ca2+ currents, extracellular Ca2+ was replaced with 20 mM Ba2+ (pH 7.4, adjusted with NaOH) and a holding potential of −50 mV was employed in order to inactivate T-type Ca2+ channels. Cells were repeatedly depolarized to +10 mV (200 ms, 0.1 Hz). All currents were low-pass filtered at 2 kHz and digitised at 10 kHz.
Real-time polymerase chain reaction (RT-PCR)
To determine mRNA expression levels of Cav3.2 and Cav3.1 channels, T75 flasks (70–80 % confluency) were washed with PBS and cells dissociated using 0.5 ml 0.05 % trypsin-EDTA for 3 min (37 °C; 95 % air: 5 % CO2). Enzyme activity was halted by adding 0.5-ml ice-cold PBS; the cell suspension was then centrifuged (600g for 6 min). RNA was generated from whole cell lysates using the Aurum total RNA mini kit (Bio-Rad, Hemel Hempstead, UK) following manufacturer’s instructions. A cDNA template was generated from RNA samples using the iScript cDNA synthesis kit (Bio-Rad, Hemel Hempstead, UK) following manufacturer’s instructions (reaction profile was 5 min at 25 °C, 30 min at 42 °C, 5 min at 85 °C, 5 min at 4 °C). Rat or human Taqman probes (Applied Biosystems (ABI), UK) for Cav3.1 (CACNA1G), Cav3.2 (CACNA1H) and the endogenous housekeeper hypoxanthine phosphoribosyltransferase (HPRT1) were employed for A7r5 cells and HSVSMC, respectively. In all cases, 2 μl of sample cDNA and 18 μl of RT-PCR reaction mix (10 μl Taqman universal PCR master mix, 0.5 μl Taqman probes (both from ABI) and 7.5 μl RNase/DNase-free water (Gibco, Cambridge, UK)) were added to the required wells of a 96-well PCR plate (Applied Biosystems, Cambridge, UK). RT-PCR was carried out using an ABI 7500 real-time PCR system (reaction profile was 2 min at 50 °C, 10 min at 95 °C, 15 s at 95 °C for 60 cycles, 1 min at 60 °C). Data were analysed using the 7500 software (ABI) and relative gene expression calculated using the 2−ΔΔCT method with HPRT1 as the endogenous control.
WT HEK293 or HEK293/Cav3.2 cells were plated at the required cell density on circular glass coverslips (10 mm, thickness 0) and allowed to adhere overnight. Cells were washed and incubated with 4 μM Fura 2-AM (Invitrogen, Cambridge, UK) diluted in HEPES-buffered saline for 40 min at room temperature (21–24 °C). Composition of HEPES-buffered saline was (in mM): NaCl 135, KCl 5, MgSO4 1.2, CaCl2 2.5, HEPES 5, glucose 10, osmolarity adjusted to 300 mOsm with sucrose, and pH adjusted to 7.4. The Fura 2-containing saline was removed after 40 min and replaced with HEPES-buffered saline for 15 min to allow de-esterification. Coverslip fragments were loaded into a perfusion chamber on an inverted epifluorescence microscope, and the cells were superfused via gravity at 2–3 ml/min. [Ca2+]i was indicated by fluorescence emission measured at 510 nm as a result of alternating excitation at 340 and 380 nm using a Cairn Research ME-SE Photometry system (Cairn Research, Cambridge, UK). Baseline readings were obtained on exposure to HEPES-buffered saline, and Ca2+ homeostasis was monitored in response to the addition of a drug, or in response to Ca2+-free HEPES-buffered saline (composition as above, but with CaCl2 replaced by 1 mM EGTA).
Statistical comparisons were made using, as appropriate, paired or unpaired student’s t tests, one-way ANOVA with a multiple comparison test or repeated measures one-way ANOVA with a multiple comparison test.
CO regulation of T-type Ca2+ channels regulates proliferation in A7r5 cells
HO-1 and CO inhibit proliferation in HSVSMCs
CO inhibits augmented proliferation in Cav3.2-expressing HEK293 cells
Cav3.2 over-expression increases basal [Ca2+]i
HO-1 and CO regulate [Ca2+]i in Cav3.2-expressing cells
Although Ca2+ influx via L-type Ca2+ channels is important for VSMC contraction, a reduction in their expression is associated with the proliferative phenotypic change [16, 19], as observed in pathological models involving VSMC proliferation . However, Ca2+ influx is still needed for the progression of proliferation since it regulates the activity of numerous transcription factors, e.g. NFAT (nuclear factor of activated T-cells; ). Some studies suggest TRP (transient receptor potential) channels, particularly TRPC channels, contribute to Ca2+ influx during VSMC proliferation [19, 27]. Further evidence indicates STIM1/Orai–mediated Ca2+ entry is also involved in VSMC proliferation, migration and neointima formation in vivo [3, 56]. However, there is also compelling evidence for the involvement of voltage-gated T-type Ca2+ channels in VSMC proliferation. Indeed, in proliferating VSMCs, as L-type Ca2+ channel expression decreases, there is a concomitant increase in T-type Ca2+ channel expression [26, 42]. Evidence suggests Ca2+ influx via T-type Ca2+ channels is required for VSMC proliferation in vitro and in neointima formation observed following vascular injury [26, 29, 43, 45]. Although the implication of a role for T-type Ca2+ channels has often (but not always) been based on the use of mibefradil (which was originally proposed as a selective T-type Ca2+ channel blocker but has since been shown to exert other effects, such as inhibition of store-operated Ca2+ entry ), mibefradil clearly blocks T-type Ca2+ channels, inhibits proliferation associated with vascular injury-mediated neointima formation and NFAT-mediated transcriptional activity [29, 45]. Furthermore, in the pulmonary vasculature, evidence for T-type Ca2+ channels regulating proliferation comes also from siRNA-targeted T-type (Cav3.1) Ca2+ channel knock-down . Most convincingly, murine knockout models have recently shown beyond doubt that Cav3.1 is required for VSMC proliferation following systemic vascular injury .
In VSMCs expressing native T-type Ca2+ channels (A7r5 cells and HSVSMCs), data presented are also consistent with these channels exerting an important influence on proliferation. Consistent with previous work , we detected expression of both Cav3.1 and Cav3.2 in A7r5 cells, and also detected mRNA for both channel types in HSVSMCs (Fig. 6), and mibefradil reduced proliferation in both cell types (Figs. 1 and 5). In A7r5 cells, despite the presence of nifedipine-sensitive L-type Ca2+ channels (Fig. 3), nifedipine was without effect on proliferation (Fig. 1), which discounts the possibility that mibefradil (or indeed NNC 55-0396) reduced proliferation via a non-selective blockade of L-type Ca2+ channels. Ni2+ (studied in the presence of nifedipine) was effective at reducing proliferation only at higher (>100 μM) concentrations. This suggests that influx of Ca2+ into A7r5 cells via T-type Ca2+ channels predominantly involves Cav3.1 rather than Cav3.2 channels, since Cav0.3.2 channels would be expected to be already fully inhibited at these higher Ni2+ concentrations .
The major finding of the present study is that HO-1 induction leads to reduced proliferation in VSMCs (both A7r5 cells, Fig. 1, and HSVSMCs, Figs. 4 and 5) and that this occurs via CO formation which in turn inhibits T-type Ca2+ channels. Thus, reduced proliferation arising from HO-1 induction could be mimicked by application of the CO-donor CORM-3 in both cell types (Figs. 2 and 4), and in A7r5 cells, we were able to demonstrate directly that T-type Ca2+ channels were inhibited by CORM-2 (Fig. 3). It should be noted that we could not use CORM-2 for proliferation studies, since cells did not tolerate long-term exposure to its solvent, DMSO (data not shown). CO also inhibited L-type Ca2+ channels (as we have previously shown in cardiac myocytes ), but this appears to be without influence on proliferation, since proliferation was insensitive to nifedipine (Fig. 1b). The reason why L-type Ca2+ channels do not influence proliferation in these cells is unknown, but may be due to a lack of tonic activity at the cell’s resting membrane potential. In HSVSMCs, the lack of additive effects of HO-1 induction and mibefradil exposure on proliferation further support the idea that T-type Ca2+ channel modulation by CO accounts for the inhibition of proliferation by HO-1. These data, combined with our recent electrophysiological study directly demonstrating inhibition of all 3 isoforms of T-type Ca2+ channels by CO , and also the observation that HO-1 induction or exposure to CO reduces basal [Ca2+]i in Cav3.2-expressing cells and reduces proliferation, collectively argue strongly that VSMC proliferation can be regulated via T-type Ca2+ channel modulation by CO derived from HO-1.
T-type Ca2+ channels are also clearly associated with proliferation in other cell types, including certain cancers , where they represent viable therapeutic targets (e.g. ). The present study also demonstrates, in agreement with an earlier report , that over-expression of T-type Ca2+ channels (in this case, Cav3.2; Fig. 7) in HEK293 cells promotes proliferation. This increase is attributable to Ca2+ influx via these channels, since inhibition with mibefradil reduced proliferation rates to levels observed in WT cells (i.e. not expressing T-type Ca2+ channels). Furthermore, Cav3.2-mediated increases in proliferation were associated with increased basal [Ca2+]i (Fig. 8), suggesting that tonic Ca2+ influx via Cav3.2 provided a sustained elevation of [Ca2+]i which promoted proliferation. This presumably occurs via the well-described T-type Ca2+ channel ‘window current’  which arises from a small proportion of the total T-type Ca2+ channel population that retains tonic activity (i.e. partially activated and not fully inactivated) at or around the cell’s resting membrane potential. The presence of a window current generated by expressed Cav3.2 channels would also require that native currents set the resting membrane potential appropriately. The nature of such currents has not to date been explored in depth.
Numerous studies have indicated that HO-1 induction affords protection in a variety of pathological systemic cardiovascular conditions, many of which involve VSMC proliferation [8, 44]. Indeed, HO-1 is established as being anti-proliferative, and its induction is observed in these proliferative vascular diseases. For example, proliferation associated with hypertension is suppressed by HO-1 induction , and many studies have indicated that HO-1 protects against inflammation and oxidative stress associated with atherosclerosis . Furthermore, whilst bilirubin is a powerful antioxidant in its own right, evidence points towards CO as accounting for many of the effects of HO-1 in VSMCs [12, 13, 34]. For example, inhalation of CO inhibits VSMC proliferation in intimal hyperplasia following vessel grafting [34, 41] and CO inhalation, as well as CO-releasing molecules (CORMs), is being developed for future cardiovascular therapy [14, 32]. HO-1 is also influential in the pulmonary vasculature: hypoxic induction of pulmonary hypertension (and hence right ventricular dilation) was far worse in HO-1−/− mice , and targeted pulmonary over-expression of HO-1 prevents inflammation and pulmonary vessel wall hypertrophy (including VSMC proliferation) in response to hypoxia . Clearly, HO-1 is beneficial in providing protection against pulmonary remodelling, and speculation that CO mediates these effects of HO-1 has been made in earlier publications . More recent studies have provided direct evidence that CO inhalation in HO-1−/− mice prevented hypoxic pulmonary remodelling and inhibited hypoxic VSMC proliferation in vitro, whereas biliverdin was ineffective .
Despite these numerous accounts of the beneficial effects of CO derived from HO-1 induction in both systemic and pulmonary vascular diseases, no mechanism has been proposed to account for such effects to date. The data presented here provide compelling evidence that CO may provide protection against proliferative vascular diseases at least in part via inhibition of T-type Ca2+ channels. Since CO can regulate all three isoforms of T-type Ca2+ channels (Cav3.1, 3.2 and 3.3 ), this regulation can occur regardless of which isoform is expressed. The present data therefore provide mechanistic insight into the beneficial cardiovascular effects of HO-1 and CO and also point to a novel signalling pathway which can be targeted in future treatment strategies for vascular diseases.
This work was supported by the British Heart Foundation.
- 20.Huang L, Keyser BM, Tagmose TM, Hansen JB, Taylor JT, Zhuang H, Zhang M, Ragsdale DS, Li M (2004) NNC 55-0396 [(1S,2S)-2-(2-(N-[(3-benzimidazol-2-yl)propyl]-N-methylamino)ethyl)-6-fluoro-1,2, 3,4-tetrahydro-1-isopropyl-2-naphtyl cyclopropanecarboxylate dihydrochloride]: a new selective inhibitor of T-type calcium channels. J Pharmacol Exp Ther 309:193–199PubMedCrossRefGoogle Scholar
- 31.Minamino T, Christou H, Hsieh CM, Liu Y, Dhawan V, Abraham NG, Perrella MA, Mitsialis SA, Kourembanas S (2001) Targeted expression of heme oxygenase-1 prevents the pulmonary inflammatory and vascular responses to hypoxia. Proc Natl Acad Sci U S A 98:8798–8803PubMedCentralPubMedCrossRefGoogle Scholar
- 48.Van BC, Cauvin C, Johns A, Leijten P, Yamamoto H (1986) Ca2+ regulation of vascular smooth muscle. Fed Proc 45:2746–2751Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.