Translocon closure to Ca²⁺ leak in proliferating vascular smooth muscle cells

Abstract

Vascular smooth muscle cells have a proliferative phenotype that is important in vascular development, adaptation, and disease. Intracellular calcium handling is thought to play pivotal roles in determining the properties of these cells, and thus previously unrecognized mechanisms for transmembrane calcium movement are of potential interest. An unsolved question is the mechanism of constitutive (passive) calcium leak from the intracellular stores. Studies of other cell types have suggested that the translocon is a calcium leak pathway. Here we investigated the contribution of the translocon in proliferating vascular smooth muscle cells. Calcium leak into the cytoplasm was measured using fura-2, and protein synthesis was measured using radioactive methionine. Puromycin, emetine, and anisomycin are chemicals that inhibit protein synthesis, acting via the translocon; all three agents strongly inhibited protein synthesis in the smooth muscle cells within 1 h. Puromycin, which opens the translocon, evoked a transient increase in cytoplasmic calcium that was similar in amplitude to the calcium rise evoked by thapsigargin. The puromycin effect was abolished by thapsigargin. The treatment of cells for 1 h with emetine or anisomycin, which close the translocon, inhibited the calcium release evoked by puromycin but not the calcium release evoked by extracellular ATP, endothelin-1, or the calcium ionophore ionomycin. Thapsigargin-evoked calcium rises were slightly suppressed by emetine but unaffected by puromycin or anisomycin. The data suggest that the translocon has the capacity to act as a calcium leak pathway in the ribosomal endoplasmic reticulum but that it is normally closed and lacks relevance to physiological calcium leak mechanisms.

the proliferation and migration of vascular smooth muscle cells (VSMCs) are important in physiology and diseases (3, 25). They are essential for vasculogenesis and physiological adaptation as well as contributing significantly to atherosclerotic plaques, neointimal hyperplasia, in-stent restenoses, and other maladaptive formations of blood vessels (1, 22, 25). As with many cell behaviors, Ca²⁺ plays pivotal roles. Intriguingly, there are significant Ca²⁺-handling differences in proliferating compared with contractile VSMCs, including intracellular Ca²⁺ storage and Ca²⁺ entry after store depletion (2, 4, 11, 13, 31). Here we address an aspect of the Ca²⁺ storage system, the Ca²⁺ leak mechanism from the intracellular stores formed by the sarco(endo)plasmic reticulum.

Generally, in mammalian cells, the Ca²⁺ content of stores is thought to be governed by Ca²⁺ uptake driven by sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA), counterbalanced by constitutive Ca²⁺ leak through Ca²⁺-permeable channels of the endoplasmic reticulum (ER) membrane. A distinction is suggested to exist between Ca²⁺-leak channels and Ca²⁺-release channels. The well-recognized Ca²⁺-release channels are the inositol 1,4,5-trisphosphate (IP₃) and ryanodine receptors (19, 35). Although the activity of the release channels is not excluded as a contributor to Ca²⁺ leak, blocking SERCA with thapsigargin reveals the importance of a constitutive Ca²⁺ leak in the absence of elevated IP₃ concentrations (28, 29). The elevated Ca²⁺ concentration upon the application of thapsigargin may secondarily stimulate the ryanodine receptors, but the ryanodine receptors are not thought to be the leak mechanism (10, 17, 32). Consequently, there have been investigations to identify whether separate proteins might account for the constitutive Ca²⁺ leak. Two candidates are the translocon complex and presenilin-1 (9, 23, 24, 30, 32). Here we focus on the translocon.

The translocon is an ER membrane protein complex mediating the translocation of nascent polypeptide into the ER. It, therefore, has a critical role in the synthesis and topological organization of transmembrane proteins (33). Various polypeptide synthesis inhibitors act by modifying the activity of the translocon. One inhibitor is puromycin, which opens the translocon by mimicking aminoacyl tRNA, releasing the nascent polypeptide chain (26). The antibiotic anisomycin and anti-protozoal agent emetine, by contrast, close the translocon by inhibiting the peptidyl transferase (9, 24, 26).

There has been interest in whether the translocon might confer not only a polypeptide pathway across the ER membrane but also a pathway for ions and small molecules, particularly when the polypeptide chain releases from the translocon (10, 14, 16, 17). In this regard it can be envisaged that ion permeability, and particularly Ca²⁺ permeability, of the translocon has the potential to be a mechanism linking protein synthesis to ER Ca²⁺, regulating protein synthesis by changing the Ca²⁺ content of the ER (27). Biochemical studies of the translocon have suggested that leak is prevented by the BiP protein sealing the translocon at the moment of polypeptide release (12, 13). However, functional studies have led to the opposite conclusion, suggesting that the translocon is a major pathway for Ca²⁺ leak in cell types including a prostate cancer cell line (9, 32) and salivary gland cells (24). Here we investigated whether the translocon contributes to Ca²⁺ leak in proliferating VSMCs. The translocon must be present in VSMCs but has not been previously investigated as a contributor to Ca²⁺ handling. While we are interested in the general principles for VSMCs, we specifically focused on VSMCs from the human saphenous vein because this blood vessel is a common coronary artery bypass graft that is prone to failure because of neointimal hyperplasia (1).

METHODS

Cells.

Freshly discarded human saphenous vein VSMCs were obtained with informed consent from patients undergoing open heart surgery in the General Infirmary at Leeds. Approval was granted by the Leeds Teaching Hospitals Local Research Ethics Committee. As previously described (34), VSMCs were grown in Dulbecco's modified Eagle's medium (DMEM) containing GlutaMAX-1, supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin (GIBCO) at 37°C in a 5% CO₂-95% room air incubator. VSMCs were used for experiments at passage 2–5. For measurement of protein synthesis, the VSMCs were replated at 70–80% confluence in six-well plates (Sarstedt) for 24–36 h before experimentation. For Ca²⁺ measurements, VSMCs were replated on Nunclon Delta Microwel plates (96 well) at 70–80% confluence for 24–36 h before experimentation.

Chemicals and ionic solutions.

General salts and puromycin, anisomycin, emetine, thapsigargin, ionomycin, ATP, endothelin-1 were from Sigma. Radioactive [³⁵S]methionine was from MP Biomedicals Europe. The puromycin stock solution was 200 mM in water, stored at −20°C. The emetine stock solution was 40 mM in water, stored at 4°C. The anisomycin stock solution was in dimethylsulfoxide at 200 mM, stored at 4°C. Standard bath solution (SBS) contained (in mM) 130 NaCl, 5 KCl, 1.2 MgCl₂, 1.5 CaCl₂, 8 d-glucose, and 10 HEPES; osmolarity was adjusted to 290–300 mosmol/l with mannitol (pH 7.4). Ca²⁺-free SBS was SBS without the addition of CaCl₂.

Protein synthesis measurement.

VSMCs were preincubated for 1 h with cysteine and methionine-deprived DMEM and then incubated for a further 1 h in DMEM supplemented with radioactive [³⁵S]methionine and puromycin, emetine, anisomycin, or vehicle at 37°C. VSMCs were scraped, collected, lightly centrifuged, and washed three times with phosphate-buffered saline. VSMCs were then treated with 2% sodium dodecyl sulfate and proteins precipitated with 10% trichloroacetic acid, using bovine serum albumin as a carrier. Pellets were dissolved in 0.5 M NaOH for 1 h. Radioactive emission was determined in a liquid scintillation counter.

Ca²⁺ measurement.

VSMC Ca²⁺ measurements were made on a FlexStation II³⁸⁴ (Molecular Devices), which is a real-time fluorimeter suitable for multiwell live-cell assays. Our experiments used 96-well plates, enabling direct multiwell comparisons of test and control conditions on the same batch of cells. The fluorescence indicator fura-2 acetoxymethyl ester (AM) (Invitrogen) was used to measure changes in intracellular ionized Ca²⁺ concentration. The VSMCs were incubated with 2 μM fura-2 AM and 0.1% pluronic acid in SBS for 1 h at 37°C. VSMCs were washed two to three times in SBS and then treated with agents (protein synthesis inhibitors) at 37°C for another 1 h in a Ca²⁺-free SBS. Once loaded with fura-2, VSMCs were excited with 340- and 380-nm light, and the emission was collected at 510 nm. Changes in intracellular ionized Ca²⁺ concentration are indicated as the change in the fura-2 fluorescence ratio for the two excitation wavelengths. Measurements were made in Ca²⁺-free SBS at room temperature (21 ± 2°C), unless indicated otherwise.

Data analysis.

Data are expressed as means ± SE, where n is the number of independent experiments and N is the number of wells used in multi-well plates. In each case, Ca²⁺ responses were determined from the average of five data points at the peak of the response. A Student's t-test was used to compare control and test data sets, which were from paired experiments. A statistically significant difference is indicated by P < 0.05. Origin software was used for data analysis.

RESULTS

Effectiveness of translocon pharmacology.

Studies of the acute functional impact of the translocon on Ca²⁺ signaling depend on pharmacological agents that alter the activity of the translocon. These agents are protein synthesis inhibitors. We selected three agents (puromycin, emetine, and anisomycin) and confirmed that each is an inhibitor of protein synthesis in VSMCs under the conditions necessary for the study of Ca²⁺ leak. Protein synthesis was quantified by measuring [³⁵S]methionine incorporation. Within 1 h, each agent strongly inhibited protein synthesis (Fig. 1). The same 1-h incubation time was used for subsequent Ca²⁺ measurement experiments, except when acute responses to puromycin were studied.

Fig. 1.

Chemical inhibition of protein synthesis. Radioactive methionine incorporation in vascular smooth muscle cells (VSMCs) treated for 1 h with vehicle control, puromycin (200 μM), anisomycin (200 μM), or emetine (40 μM) (n = 3 independent experiments; N = 8 wells; P < 0.05 for each inhibitor relative to control). cpm, Counts per minute.

Puromycin-evoked Ca²⁺ release.

An application of puromycin to VSMCs loaded with fura-2 evoked a transient rise in cytoplasmic Ca²⁺ (Fig. 2A, and Table 1); the effect was similar at 37°C (n = 3; supplemental Fig. 1; note: supplemental material may be found posted with the online version of this article). The effect of puromycin occurred in the absence of extracellular Ca²⁺ (Fig. 2A) and was lost in VSMCs pretreated with thapsigargin (Fig. 2B, and Table 1), which depletes intracellular Ca²⁺ stores. Emetine and anisomycin inhibited the puromycin-evoked Ca²⁺ signal (Fig. 3, and Table 1). The data suggest that puromycin evokes Ca²⁺ release via the translocon.

Fig. 2.

Puromycin-evoked Ca²⁺ release. All graphs show measurement of intracellular Ca²⁺ (Ca

) from VSMCs in the absence of extracellular Ca²⁺ (0 Ca²⁺). Data in A and B are each from single paired-independent experiments. A: effect of extracellular application of puromycin (200 μM; a) or vehicle control (b). B: effect of extracellular application of puromycin (200 μM) after pretreatment with vehicle control (a) or 1 μM thapsigargin (b) for 0.5 h at room temperature. ΔF, change in fluorescence; T, time.

Table 1.

Summary of Ca²⁺ measurement data

	Control Response, ΔF Ratio	Test Response, ΔF Ratio	Inhibition, %	P Value	n	N
Puromycin
Thapsigargin	0.313±0.061	0.019±0.002	93.9	0*	3	60
Emetine	0.306±0.014	0.119±0.037	61.2	0.0070*	3	32
Anisomycin	0.226±0.012	0.080±0.048	64.6	0.0079*	3	24
Thapsigargin
Emetine	0.341±0.072	0.278±0.058	18.5	0.0012*	5	36
Puromycin	0.274±0.062	0.294±0.068	−7.3	0.5829	5	28
Anisomycin (200 μM)	0.262±0.048	0.273±0.032	−4.2	0.4307	3	24
Anisomycin	0.211±0.086	0.197±0.079	6.5	0.1416	3	36
Anisomycin (37°C)	0.244±0.017	0.249±0.030	−2.0	0.8963	5	76
Anisomycin (1.5 mM Ca2+, 37°C)	0.183±0.038	0.155±0.054	15.5	0.6823	4	48
Anisomycin (10 mM Ca2+, 37°C)	0.166±0.040	0.150±0.031	9.8	0.7599	4	48
Ionomycin
Emetine	0.669±0.065	0.608±0.065	9.2	0.0967	3	18
Puromycin	0.654±0.046	0.681±0.046	−4.1	0.7671	3	18
Anisomycin	1.264±0.289	1.148±0.233	9.2	0.0732	3	40
Anisomycin (37°C)	0.502±0.087	0.383±0.077	23.7	0.0725	4	60
ATP
Emetine	0.214±0.081	0.207±0.060	3.3	0.7546	4	64
Anisomycin	0.226±0.077	0.235±0.095	−3.2	0.8792	3	24
Puromycin	0.323±0.028	0.253±0.087	21.8	0.6235	3	24
Endothelin-1
Emetine	0.504±0.186	0.466±0.058	7.5	0.6530	3	44

Values are means ± SE; n, number of independent experiments; N, number of wells of the 96-well plate. Puromycin, thapsigargin, emetine, ionomycin, ATP, and endothelin-1 were used at 200, 3, 40, 5, 100 and 0.1 μM, respectively. Anisomycin was used at 400 μM unless indicated. Experiments were performed in the absence of extracellular Ca²⁺ and at room temperature, unless at 37°C as indicated. Preincubation with inhibitors was in the absence of Ca²⁺ for 1 h unless the preincubation included 1.5 or 10 mM Ca²⁺ where indicated. ΔF, change in fluorescence.

^*P < 0.05.

Fig. 3.

Inhibition of puromycin-evoked Ca²⁺ release by translocon inhibitors. All graphs show measurement of intracellular Ca²⁺ from VSMCs in the absence of extracellular Ca²⁺ (0 Ca²⁺). Data in A and B are each from single paired-independent experiments. A: effect of puromycin (200 μM) after pretreatment with vehicle control (a) or 40 μM emetine (b) for 1 h. B: effect of puromycin (200 μM) after pretreatment with vehicle control (a) or 400 μM anisomycin (b) for 1 h.

Constitutive Ca²⁺ leak.

Because the translocon is capable of conferring Ca²⁺ leak, we explored whether it is relevant to the constitutive Ca²⁺ leak that is evident when the SERCA of the store is blocked by thapsigargin, applied to VSMCs in the absence of extracellular Ca²⁺. Although there was a small inhibitory effect of emetine on the thapsigargin response (Table 1), puromycin had no effect (Table 1) and anisomycin had no effect, even at the high concentration of 0.4 mM (Fig. 4A, and Table 1). An independent method for investigating constitutive Ca²⁺ leak is to measure the total amount of Ca²⁺ releasable from intracellular stores because the total content of the stores will be greater if the constitutive Ca²⁺ leak is inhibited. To determine the total Ca²⁺ content, we applied ionomycin, which permeabilizes the membrane of the stores to Ca²⁺, causing rapid Ca²⁺ leak. Ionomycin evoked a large Ca²⁺ transient in the absence of extracellular Ca²⁺, which was not affected by emetine (Fig. 4B) or anisomycin (Table 1).

Fig. 4.

Lack of effect of translocon inhibitors on constitutive Ca²⁺ leak. All graphs show measurement of intracellular Ca²⁺ from VSMCs in the absence of extracellular Ca²⁺ (0 Ca²⁺). Data in A and B are each from single paired-independent experiments. A: effect of thapsigargin (3 μM) after pretreatment with vehicle control (a) or 400 μM anisomycin (b) for 1 h. B: effect of ionomycin (5 μM) after pretreatment with vehicle control (a) or 40 μM emetine (b) for 1 h.

The above experiments were performed at room temperature and after a 1-h preincubation in Ca²⁺-free medium (to avoid contamination from Ca²⁺-entry signals). Although Ca²⁺ leak occurred, it was conceivable that the conditions were biased against the contribution of the translocon. However, a repetition of the experiments at 37°C and the preincubation in Ca²⁺-containing medium revealed no effects of anisomycin on the thapsigargin and ionomycin responses (Table 1). In one set of experiments, we increased the Ca²⁺ concentration in the preincubation medium to 10 mM in an effort to overload the Ca²⁺ stores (20), but again there was no effect of anisomycin (Table 1).

The data suggest that the translocon has no role in the constitutive Ca²⁺ leak mechanism.

Ca²⁺ release evoked by physiological agonists.

Two physiological agonists that evoke Ca²⁺ release in saphenous vein VSMCs are endothelin-1 and ATP (5, 7). Both agonists act through G protein-coupled receptors that induce the production of IP₃. Neither endothelin-1 nor ATP responses were affected by emetine or anisomycin (Fig. 5, and Table 1). Therefore, the data suggest that the Ca²⁺ release evoked by physiological agonists does not involve the translocon.

Fig. 5.

Lack of effect of translocon inhibitors on Ca²⁺ release evoked by physiological agonists. All graphs show measurement of intracellular Ca²⁺ from VSMCs in the absence of extracellular Ca²⁺ (0 Ca²⁺). Data in A and B are each from single paired-independent experiments. A: effect of endothelin-1 (100 nM) after pretreatment with vehicle control (a) or 40 μM emetine (b) for 1 h. B: effect of ATP (100 μM) after pretreatment with vehicle control (a) or 400 μM anisomycin (b) for 1 h.

DISCUSSION

The study shows a striking Ca²⁺-release event when VSMCs are exposed to puromycin. The event is suggested to take place as a result of the opening of the translocon because it is suppressed by the translocon inhibitors anisomycin and emetine. Importantly, studies of both the puromycin Ca²⁺-release signal and direct measurements of protein synthesis enabled us to confirm that anisomycin and emetine inhibit the translocon under our experimental conditions. Under these conditions, passive Ca²⁺ leak and Ca²⁺ release evoked by physiological agonists were not affected by translocon inhibition. Therefore, although Ca²⁺ leak via the translocon is possible in VSMCs, the translocon is not normally a passive Ca²⁺ leak mechanism of these cells and does not act as a Ca²⁺ pathway influencing responses to endothelin-1 or ATP. Therefore, the data suggest that there is translocon closure to Ca²⁺ leak unless there is a premature release of the nascent polypeptide chain caused by an agent such as puromycin.

Relatively rapid puromycin-evoked release of ER Ca²⁺ has been previously suggested based on direct measurements of luminal Ca²⁺ or other ER permeability assays (14, 17, 24, 32). Our observation of a cytoplasmic Ca²⁺-release signal is consistent with these previous results, as well as the rapid effect of puromycin on protein synthesis (18). We were, however, surprised to observe that the release signal is comparable in amplitude and time course to signals evoked by other agents including thapsigargin and ATP. Superficially, the similarity suggests a release from a common intracellular Ca²⁺ store. However, although thapsigargin prevented the puromycin response, puromycin failed to inhibit the thapsigargin response. Therefore, the data suggest that the translocon is contained in ER elements that depend on Ca²⁺ uptake via SERCA but that it is absent from other SERCA-dependent elements. The latter translocon-independent elements may be sub-plasma membrane Ca²⁺ stores involved in IP₃-dependent Ca²⁺ signaling, a conclusion that is consistent with the statistically insignificant effect of puromycin on the ATP response. That thapsigargin depletes the puromycin-sensitive store yet evokes puromycin-resistant Ca²⁺ leak tells us that Ca²⁺ leak from the translocon-containing ER (ribosomal ER) also does not occur through the translocon.

Our findings suggest that the translocon is tightly sealed against Ca²⁺ flux in VSMCs, even when the cells are proliferating and actively synthesizing new proteins. The question arises, therefore, as to whether leakiness of the translocon is different in VSMCs compared with the LNCaP prostate cancer cell line or salivary gland cells where other investigators have concluded that the translocon is a physiological Ca²⁺ leak mechanism (9, 24). Ong et al. (24) indirectly observed Ca²⁺ release in response to a translocon opener, an effect that was strongly inhibited by emetine. However, thapsigargin-evoked Ca²⁺ release was only slightly affected by emetine. We observed almost exactly the same result with emetine, but our data additionally show that emetine lacks specificity for the translocon because its effect on the thapsigargin response was not mimicked by anisomycin or puromycin. Flourakis et al. (9) observed that 200 μM anisomycin caused about 30% inhibition of Ca²⁺ release evoked by thapsigargin in LNCaP cells; unfortunately, using the same protocol, we have not reproduced this result in VSMCs (Table 1) or LNCaP cells (M. S. Amer, unpublished data). Our interpretation of the observations is that we cannot yet be certain there are major differences in leakiness of the translocon to Ca²⁺ in different cell types.

To the best of our knowledge, there is no physiological or pathological situation that leads to a premature release of the nascent polypeptide chain. Therefore, it seems appropriate to conclude that the translocon is normally efficiently closed to Ca²⁺ leak. Nevertheless, the translocon can readily be made leaky to Ca²⁺ by puromycin. Puromycin is used therapeutically as an antiprotozoal agent and has been compared with cycloheximide as a potential pharmacological approach to stabilizing atherosclerotic plaques (6). Intriguingly, the latter study showed that puromycin, but not cycloheximide, evoked VSMC apoptosis. Our data may help to explain this differential effect because puromycin causes Ca²⁺ release in VSMCs, whereas cycloheximide would be expected to have no effect on Ca²⁺ release because it is a translocon inhibitor.

Our data suggest that the translocon does not explain the constitutive (passive) leak of Ca²⁺ from intracellular stores of VSMCs. Consequently, the molecular mechanism of Ca²⁺ leak in these cells remains unclear. It is important to consider the potential contribution of ryanodine receptors, even though such a contribution appears absent in some other cell types (10, 17, 32). Significantly, we have observed no effect of 0.1 mM ryanodine on the thapsigargin response of human saphenous vein VSMCs, and an activator of ryanodine receptors (10 mM caffeine) failed to evoke Ca²⁺ release, even following efforts to overload the Ca²⁺ stores using 10 mM Ca²⁺ in the extracellular medium (supplemental Figs. 2 and 3). The same caffeine was shown to be effective as a Ca²⁺-release agent on ventricular myocytes (supplemental Fig. 4). Such observations do not encourage a further consideration of ryanodine receptors as the leak mechanism in VSMCs. We have no data on the roles of IP₃ receptors in the leak mechanism, but IP₃ receptor inhibitors (2-aminoethoxydiphenylborate and xestospongin C) either enhanced or had no effect on passive Ca²⁺ leak from stores of the A7r5 VSMC line (8, 21), arguing against IP₃ receptors playing a role.

In conclusion, our observations are not consistent with the translocon acting as a physiological Ca²⁺ leak pathway or with Ca²⁺ permeability of the translocon serving as a mechanism linking Ca²⁺ handling to protein synthesis. The data are consistent with the translocon having the capacity to enable Ca²⁺ leak from ribosomal ER but, nevertheless, suggest that the translocon is not the normal Ca²⁺ leak mechanism of the ribosomal ER. The capacity for translocon-dependent Ca²⁺ leak is not apparent in the subplasma membrane Ca²⁺ stores (non-ribosomal ER) that enables IP₃-dependent Ca²⁺ release. We emphasize that our study focused on VSMCs and so does not exclude differential Ca²⁺ leakiness of the translocon in other cell types and contexts. Furthermore, pharmacologically, the translocon can be made leaky to Ca²⁺, which may explain the functional differences in effects of protein synthesis inhibitors that either open or close the translocon.

GRANTS

This study was supported by the Wellcome Trust and a scholarship from the Egyptian Ministry of Higher Education (to M. S. Amer).