Stretching increases calcium influx and efflux in cultured pulmonary arterial smooth muscle cells RUSSELL A. BIALECKI, THOMAS J. KULIK, AND WILSON S. COLUCCI Cardiovascular Division, Departments of Medicine, Brigham and Women’s Hospital and Harvard Medical School, and Department of Cardiology, Children’s Hospital, Boston, Massachusetts 02115 Bialecki, Russell A., Thomas J. Kulik, and Wilson S. Colucci. Stretching increases calcium influx and efflux in cultured pulmonary arterial smooth muscle cells. Am. J. Physiol. 263 (Lung Cell. Mol. Physiol. 7): L602-L606, 1992.-To determine the effect of a single static stretch on calcium fluxes in cultured pulmonary arterial smooth muscle cells (PASMC), calcium influx and efflux were evaluated in PASMC on a collagencoated silicone membrane using 45Ca2+ as a tracer. A single 20% linear stretch of the silicone membrane of 1 min in duration increased calcium uptake by 71%. This effect was partially inhibited by verapamil or gadolinium, but was not altered by staurosporine, pertussis toxin, or removal of extracellular sodium. Stretch-stimulated calcium uptake attenuated over time, such that uptake during the last minute of a 5min sustained stretch was 46% of that during the first minute of stretch. A single 20% stretch sustained for 6 min caused a 47% increase in calcium efflux, the magnitude of which was linearly related to the degree of cell stretch. Gadolinium and removal of extracellular calcium each partially inhibited stretch-induced calcium efflux. We conclude that a single static stretch of PASMC causes increases in both calcium influx and efflux. Stretchstimulated calcium influx does not require sodium influx and is mediated in part by a pathway sensitive to both gadolinium and verapamil. Stretch-stimulated calcium efflux is due to both calcium influx via a gadolinium-sensitive pathway and mobilization of intracellular stores. Because calcium is a key cellular second messenger, these effects of stretch on cellular calcium handling may play a role in the regulation of vascular smooth muscle cell phenotype and function. stretch; vascular smooth muscle; verapamil; gadolinium HAS BEEN SHOWN to stimulate contraction of pulmonary arterial smooth muscle (14), and mechanical stress appears to play a role in the pulmonary vascular remodeling that occurs with chronic alveolar hypoxia (18, 22). How mechanical stress is transduced into a physiological response is unknown, but we have recently shown (13) that cultured pulmonary arterial smooth muscle cells (PASMC) respond to stretch with a transient increase in D-myo-inositol 1,4,Strisphosphate and D-myo-inositol 1,3,4,Stetrakisphosphate, suggesting that calcium may serve as a second messenger in the transduction of stretch. In this study, we have characterized the effects of a single, static stretch on calcium homeostasis in PASMC. Calcium influx and efflux were assessed with the use of 45Ca2+ as a tracer. We found that stretch caused substantial increases in both calcium influx and efflux of a magnitude comparable to those induced by neurotransmitters and hormones. We also investigated the characteristics and mechanisms of stretch-induced calcium flux.
STRETCH
METHODS
37°C using 0.125 mg/ml elastase (type III, 90 U/mg, Sigma Chemical, St. Louis, MO), 1 mg/ml collagenase (company code CLS, type I, 150 U/mg, Worthington Biochemical, Freehold, NJ), 0.25 mg/ml soybean trypsin inhibitor (type l-S, Sigma), and 2 mg/ml crystallized bovine serum albumin (Pentex, Miles Laboratories, Elkhardt, IN). The PASMC were grown (37”C, 5% C02-95% air, in a humidified atmosphere) in medium 199 (GIBCO Laboratories, Grand Island, NY) with 20% (vol/vol) fetal calf serum (FCS; GIBCO), 100 U/ml penicillin, and 0.1 mg/ml streptomycin. After multiple passages, a clonal line was selected. By immunofluorescent staining, these cloned cells stably express smooth muscle-specific myosin [antibody courtesy of Dr. D. M. Larson (15)] and smooth muscle a-actin (Enzo Biochem, New York) over many passages. They also express mRNA for the smooth muscle a-actin isoform (by Northern blotting using a riboprobe derived from the 3’ untranslated region of the smooth muscle a-actin isoform, courtesy of Dr. J. L. Lessard) and predominantly express the smooth muscle cu-tropomyosin mRNA isoform (30). For stretch experiments, the PASMC were grown on thick (0.01 in.) silicone sheets (Silastic; Dow Corning, Midland, MI) fashioned into “tubs” (34 mm wide x 40 mm long x 10 mm deep), which were precoated with rat tail type I collagen (28) (Fig. 1). The PASMC were seeded into the tubs at an initial density of lO,OOO-20,000 cells/cm2 and grown in Dulbecco’s modified Eagle’s medium (DMEM) containing antibiotics and 10% FCS. The cells reached confluence at 2-4 days and were then used for study. Cells were incubated in DMEM containing 0.5% FCS and no antibiotics for 18-24 h before experimentation to remove potential antibiotic inhibitors of stretch-activated ion channels (19) and to reduce the tonic influence of serum factors on calcium flux. Static stretch of PASMC. To stretch the cells, a solid support fashioned at one end of the tub was moved to the appropriate stop position (Fig. 1). It took -1 s for the stretch to be accomplished. In preliminary experiments, the uniformity with which the Silastic surface is elongated by stretch was determined. The surface of the tub was divided into a grid of 15 equally spaced points arranged in five rows perpendicular to the axis of stretch, and the local elongation of the tub along the axis of stretch was measured during a 20% stretch using a microscope with a reticle. Local elongation at points located in four of the rows ranged from 18 to 25% (mean = 21 +- 1%) and was slightly reduced in the one row closest to the movable support (mean = 16 t 1%). Preliminary experiments utilizing phase-contrast microscopy demonstrated that the magnitude of deformation of the cells closely corresponded with the magnitude of elongation of the membrane. In addition, the cells did not detach from the collagen-coated Silastic surface during a 20% stretch sustained for 1 h (data not shown). Calcium uptake. Cell 45Ca2+ uptake was determined as previously described, with minor modifications (2, 3, 17, 21). The millimolar composition of the physiological salt solution (PSS) was as follows: 130 NaCl, 5 KCl, 1.5 CaCl,, 1 MgC12, 5 glucose, and 5. N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid. Zero-sodium PSS contained 130 mM choline chloride instead of NaCl. To initiate the assay, 6 ml of PSS containing 45Ca2+ (1 &i/ml) with or without other agents, was added to each tub.
CeLL culture. Main and proximal branch pulmonary arteries from adult Sprague-Dawley rats (killed by cervical dislocation) were stripped of adventitia and enzymatically dissociated at L602 1040-0605/92 $2.00 Copyright 0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
STRETCHING
ALTERS
SMOOTH
MUSCLE
L603
CALCIUM
ICN Biomedicals, Irvine, CA. Verapamil and pertussis toxin were purchased from Sigma. Gadolinium chloride was obtained from Aldrich Chemical, Milwaukee, WI. Staurosporine was purchased from Boehringer Mannheim Biochemicals, Indianapolis, IN. Drugs that required dissolution in ethanol (e.g., verapamil) were adjusted to a final vehicle concentration of l:lO,OOO at the drug concentrations indicated. Accordingly, in all experiments involving ethanol as solvent, control cells were incubated with the vehicle at a final concentration of l:lO,OOO.
a
RESULTS
Fig. 1. Line drawing of apparatus used to stretch pulmonary arterial smooth muscle cells. Plexiglas supports (b) are glued to 140-mm tissue culture dish (a). Silicone sheet is glued to Plexiglas end supports (d, e) to form the tub (c), which is 34 mm wide, 40 mm long, and 10 mm deep. The 20% linear elongation is effected by lifting movable end support (e) with a forceps so that stop (f) is lifted from slot and extended to end of support. This distance is 8 mm, which is 20% of length of silicone surface. Although not illustrated here, in some apparatuses, Plexiglas supports have slots to accommodate movable end support at intervals appropriate for 10,20, and 30% elongation, so that different magnitudes of stretch can be achieved.
Five seconds later, the tub was stretched by 20%. Uptake was terminated after various intervals by rapid washing of the cells five times with 6 ml of ice-cold PSS containing IO mM LaCl, and no CaC12, with a lo-min dwell in the final wash aliquot. The cells were then dissolved in 1 ml of 0.1 N HN03, and cellassociated 45Ca2+ was determined by counting of the radioactivity within the acid digest by standard scintillation techniques. All cell monolayers studied in the presence of IO PM verapamil, IO PM gadolinium, or 25 nM staurosporine were preincubated with the agent for 30 min before the initiation of the assay. Calcium ef’lux. The efflux of 45Ca2+ was measured as previously reported, with minor modifications (2,4,21). In brief, cell monolayers were equilibrated with 45Ca2+ for 18-24 h in DMEM containing 45CaC12 (1 &i/ml) and no antibiotics. After equilibration, the cells were washed rapidly four times with 6 ml of PSS at 37”C, and calcium efflux was initiated by the addition of 6 ml of PSS and other agents under study. The tubs were stretched (10, 20, or 30%) 5 s later. The assay was terminated after 6 min by rapid washing of the cells five times with 6 ml of ice-cold PSS containing LaCl, and no CaCl,, with a lo-min dwell in the final wash aliquot. The cells were then dissolved in 1 ml of 0.1 N HN03, and cell-associated 45Ca2+ was determined by counting of the radioactivity within the acid digest by standard scintillation techniques. Calcium efflux is expressed as the decrease in cellular calcium content over 6 min and reported as the percent change from that in unstretched control cells. In some experiments, efflux was evaluated in nominally calciumfree PSS containing 2 mM EGTA and no CaC12. Analysis of data. Statistical significance was determined by Student’s two-tailed t test for paired samples (20). P values ~0.05 were considered significant. In all cases, n equals the number of experiments performed. The data from each experiment represent the average of duplicate or triplicate tubs. Data are presented as means t SE. Reagents. 45CaC12 (Il.3 mCi/mg calcium) was obtained from
Kinetics of stretch-induced calcium uptake. To determine whether PASMC stretch is associated with an increase in calcium uptake, the early time course of calcium uptake was measured in stretched cells and, for comparison, in unstretched control cells (Fig. 2). Calcium uptake was linear for the first minute in both stretched and control cells, reflecting unidirectional calcium influx. At all time points up to 2 min, calcium uptake was greater (P < 0.01; n = 3 experiments) in stretched cells than in unstretched controls (Fig. 2). Stretch (20%) increased the average 1-min calcium uptake by 71 t 7% (P < 0.01; n = 12) compared with unstretched control cells. The kinetics of stretch-induced calcium uptake were characterized further by assessment of the duration of the increase in calcium uptake during a single sustained stretch (Fig. 3). For these studies, cells were stretched 20% for various times before cells were pulsed with 45Ca2+ for 1 min. Stretch caused a marked increase (111 t 16%; P < 0.01; n = 3) in cellular 45Ca2+ uptake at 1 min compared with uptake in control cells in unstretched tubs. During the fifth minute of a 5-min sustained stretch, calcium uptake was reduced to 46 t 17% (P < 0.05; n = 3) of the first minute response. During the last minute of a 60-min sustained stretch, calcium uptake was no different from that in unstretched cells (P = 1.18; n = 3) EffeEt of channel blockers on stretch-induced calcium uptake. To elucidate the pathway of stretch-induced cal-
cium influx, 1-min calcium uptake was measured in the presence of either 10 PM verapamil, an inhibitor of Ltype voltage-dependent calcium channels (27), or 10 PM gadolinium, a lanthanide element shown to inhibit stretch-induced calcium influx in several types of cells - 2.2 -a 2 2.0 5
--
1.8-
+I 1.6E g 1.4-
/
/
/
/
/
---
-0
/ /o-----**
**
1 .o Time
(min)
Fig. 2. Effect of stretch on calcium uptake in vascular smooth muscle cells as assessed by uptake of 45Ca2+. A sustained 20% stretch (open circles) caused a significant increase in calcium uptake at all time points compared with unstretched control cells (solid circles). Each point represents mean & SE of 3 determinations. * P < 0.05. ** P < 0.01.
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
L604
STRETCHING
z
140cn 6 voII z looc 808 Q) cl _: 60-
ALTERS
SMOOTH
MUSCLE
CALCIUM
**
* * 1
X
3 u!I
40-
20; 3 O-3 .0 3 -20 0
I 1
I 1 min
5 min
60
Fig. 3. Effect of sustained 20% stretch on calcium influx. Cells stretched for 1, 5, or 60 min were pulsed with 45Ca2+ for 1 min throughout 1-min stretch or during last minute of 5 and 60-min stretches. Bars, mean t SE of at least 3 experiments performed in triplicate. * P < 0.05 vs. unstretched controls. ** P < 0.01 vs. unstretched controls.
(3234). Verapamil decreased calcium uptake in stretched cells by 49 t 8% (P < 0.01; n = 5) compared with that in unstretched control cells. Gadolinium likewise caused a 43 t 7% reduction (P < 0.05; n = 5) in stretch-induced calcium uptake compared with that in stretched control cells. Concurrent incubation with 10 PM verapamil and 10 PM gadolinium decreased the stretch-induced calcium uptake response by 49 t 11% (P < 0.04; n = 3), an effect no greater than that of either verapamil or gadolinium alone. Role of sodium influx in stretch-induced
calcium uptake.
Replacement of extracellular sodium by 130 mM choline chloride had no effect on the magnitude of stretch-induced calcium uptake (n = 3; data not shown). Effect of staurosporine induced calcium uptake.
and pertussis
CTL
min
toxin on stretch-
Both protein kinase C and G proteins may influence calcium influx in a variety of cells (7, 21, 33). The role of protein kinase C activation in the stretch-induced increase in calcium uptake was evaluated by preincubating vascular smooth muscle cells with 25 nM staurosporine, an inhibitor of protein kinase C (25). The stretch-induced increase in calcium uptake of staurosporine-treated cells was not different from that of stretched control cells (73 t 32% vs. 59 t 8%, respectively; P = 0.68; n = 6). The possible role of a pertussis toxin-sensitive G protein in the mediation of stretch-induced calcium uptake was determined in stretched cells preincubated with pertussis toxin (20 rig/ml, 3 h). The stretch-induced increase in calcium uptake of pertussis toxin-treated cells was not different from that of control cells (79 t 47% vs. 59 t 8%, respectively; P = 0.70; n = 6). Stretch-induced calcium efflux. Stretching of the cells by 20% increased the 6-min calcium efflux by 47 t 5% (P < 0.001; n = 11) over that in unstretched control cells. The relationship between the magnitude of cell stretch and calcium efflux was linear (r = 0.984; P < 0.02; n = 3) over the range of lo-30% stretch (Fig. 4). An increase in calcium efflux reflects an increase in intracellular free calcium that may be due to the mobilization of intracellular calcium stores and/or the influx of extracellular calcium. To establish whether stretch induces mobilization of intracellular calcium stores inde-
I 20
10 Cell
Stretch
I 30
I
(percent)
Fig. 4. Effect of various degrees of stretch on 6-min calcium efflux as assessed by 45Ca 2+. Basal calcium efflux in unstretched control cells was 3.5 k 0.5 nanomol calcium. mg protein-l. 6 min. Data depict percentage increase in calcium efflux over basal in unstretched cells. Each point represents mean k SE of 4 experiments performed in triplicate.
pendent of calcium influx, stretch-induced calcium efflux was assessed in nominally calcium-free PSS containing 2 mM EGTA. Stretch-induced calcium efflux was 62 t 9% lower (P < 0.01; n = 5) in calcium-free PSS than in normal PSS (Table 1). This result suggests that ~38% of the stretch-induced calcium efflux response may be attributable to mobilization of intracellular calcium. Consistent with the role of extracellular calcium in stretch-induced calcium efflux, gadolinium (10 PM) decreased stretch-stimulated calcium efflux by 58 t 13% (P < 0.01; n = 6) in calcium-containing buffer but had no effect on stretch-induced calcium efflux in calcium-free buffer. In calcium-free buffer, stretch-stimulated calcium efflux was 20 t 2% and 23 t 2% in the absence and presence of gadolinium, respectively. Gadolinium had no effect on agonist-stimulated (thrombin, 10 U/ml) calcium efflux in unstretched control cells (20 t 1% vs. 22 t 2% without and with gadolinium, respectively), further indieating that gadolinium does not inhibit components of the pathway for agonist-mediated calcium mobilization (i.e., D-myo-inositol 1,4,5-trisphosphate generation, sarcoplasmic reticulum calcium release, or plasmalemma calcium adenosine triphosphatase). DISCUSSION
It has long been suspected that mechanical stress may cause changes in the contractile state or growth behavior of pulmonary blood vessels (9). Although it is not clear whether mechanical stress may act directly on pulmonary Table 1. Effect of gadolinium calcium efflux in the presence of extracellular calcium
on stretch-induced and absence
% Increase
Stretch Stretch Stretch Stretch
n
53t6 8 22&6* 6 20t2* 5 (10 ,uM) + [Cal, = 0 23&2-j3 Data are means t SE of percent increase in efflux over basal efflux in control cells in unstretched Silastic tubs. n, no. of experiments, each performed in triplicate. Calcium efflux over 6 min was assessed with 45Ca2+ in response to a sustained 20% elongation of Silastic culture tub. * P < 0.01 vs. stretch: -F P < 0.05 vs. stretch. + gadolinium + [Cal, = 0 + gadolinium
(10 PM)
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
STRETCHING
ALTERS
SMOOTH
vascular smooth muscle to cause growth and connective tissue production (26)) experiments using cultured aortic smooth muscle cells suggest that stretch can act directly on smooth muscle cells to modify protein production (12). However, the mechanism by which a mechanical stimulus is coupled to these or other effects is not known, and little is known about the mechanism of stretch-mediated signal transduction in PASMC. Our experiments indicate that a single sustained stretch of PASMC is a potent stimulator of both calcium influx and efflux. The magnitude of these effects is large, comparable to that stimulated by potent vasoconstrictors (2,4,21,29) in cultured vascular smooth muscle cells. Twenty percent stretch for 1 min caused an -70% increase in calcium uptake, a response that is similar in magnitude to the vasopressin- and endothelinstimulated increases in calcium uptake (69 t 3 and 85 t 14%, respectively) we observed previously in PASMC (2). Likewise, a single 20% stretch maintained for 6 min caused a 47% increase in calcium efflux, an effect that is greater than that stimulated by 10 PM norepinephrine (4) Stretch has been shown to activate nonspecific ion channels in a variety of cell types, including gastric smooth muscle cells (11). Our data extend these observations by showing for the first time in vascular smooth muscle that stretch increases calcium influx. In several types of cells, the lanthanide element gadolinium has been shown to block stretch-induced ion channels (32, 34). In PASMC, we found that stretch-induced calcium influx was inhibited w 45% by gadolinium, as well as by verapamil, which is generally regarded as an inhibitor of L-type calcium channels (27). The combination of verapamil plus gadolinium caused no more inhibition of stretch-induced calcium influx than did either agent alone. This result suggests that these agents act on the same pathway. Because gadolinium is also known to block voltage-dependent L-type calcium channels (1, 6), these data are consistent with the interpretation that the gadolinium/verapamil-sensitive component of stretch-induced calcium influx in this cell may be mediated by voltage-dependent calcium channels. An alternative possibility is that both agents act on a stretch-activated channel that is distinct from the voltage-dependent Ltype channel or an as yet undefined site. The component of stretch-induced calcium influx that is not sensitive to gadolinium and verapamil likely reflects influx by a second pathway. Because, as discussed subsequently, stretch also appears to cause mobilization of intracellular calcium stores, it is possible that some or all of this component of calcium influx is involved in the refilling of these intracellular stores (8, 24). Because stretch may increase sodium influx in cardiac myocytes (lo), we considered the possibility that sodium influx was required for stretch-induced calcium influx in PASMC. Removal of extracellular sodium did not diminish stretch-induced calcium influx, thereby suggesting that sodium influx does not play an important role in this effect. Protein kinase C has been shown to modulate the conductance of L-type calcium channels in vascular smooth muscle cells (7,21). However, the failure of staurosporine,
MUSCLE
CALCIUM
L605
an inhibitor of protein kinase C (25), to inhibit stretchinduced calcium influx does not support an important role for protein kinase C in mediating stretch-induced calcium influx. Likewise, the lack of inhibition of stretchinduced calcium influx by pertussis toxin suggests that a pertussis toxin-sensitive G protein is not required for this effect. However, this finding does not exclude involvement of one of several pertussis toxin-insensitive G proteins. Furthermore, although the protocol we used is adequate to ribosylate sensitive G proteins in other cells fully (16), we cannot exclude the possibility that the pertussis toxin treatment did not result in adequate ribosylation of the G protein in these experiments. The stretch-induced increase in calcium influx attenuates over time, such that it is reduced -50% by 5 min and is abolished after 60 min of static stretch. This pattern of attenuation suggests that stretch-induced calcium influx most likely occurs in response to a change in cellular distension rather than the absolute level of stretch. Stretch increased calcium efflux in the absence of extracellular calcium, suggesting that stretch also caused release of calcium from intracellular stores. Consistent with this interpretation, we have previously found that stretch increases D-myo-inositol trisphosphate and D-myo-inositol 1,4,5-tetrakisphosphate in PASMC (13). Stretch-induced calcium efflux was also partially inhibited by removal of extracellular calcium and the addition of gadolinium, presumably reflecting the component of efflux that is due to the influx of extracellular calcium. In the absence of extracellular calcium, gadolinium had no effect on stretch-induced calcium efflux, indicating that gadolinium does not interfere with stretch-induced mobilization of intracellular calcium. The magnitude of stretch used in these studies (20%) is similar to that used by several investigators in other cultured cell systems (5, 23, 31). Although observations by light microscopy and trypan blue exclusion suggested that stretch of PASMC by this amount does not induce detachment from the silicone membrane or cause cell injury, we cannot completely exclude the possibility that some of the increases in calcium influx and efflux reflects an unphysiological, albeit minor, disruption or the cell membrane. However, several observations argue that this is not the case. First, a substantial portion of the stretchinduced calcium uptake is inhibited by verapamil, a pharmacological agent that would not be expected to interfere with an injury-induced leakage of calcium. Second, stretch-induced calcium influx attenuates rapidly over time, whereas injury-induced influx might be expected to persist. Third, a significant component of stretch-induced calcium efflux occurs in the absence of extracellular calcium. In summary, these experiments indicate that a single static stretch of PASMC induces both calcium influx and efflux. A substantial portion of stretch-induced calcium influx occurs via a channel that is inhibited by both verapamil and gadolinium. Activation of this channel by stretch does not require sodium influx and is not inhibited by pertussis toxin or staurosporine. Although the major component of stretch-induced calcium efflux is secondary to calcium influx, a significant component does
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
L606
STRETCHING
ALTERS
SMOOTH
not require extracellular calcium, suggesting that stretch may also cause release of intracellular calcium. These data indicate that mechanical perturbations can exert a potent effect on calcium homeostasis in PASMC and that calcium mobilization may therefore play a role in the transduction of stretch in these cells. The authors thank Carolyn Kastner and Erin Smith for expert technical assistance throughout the course of these studies. This work was supported by Grants HL-429082, HL-429082, and HL-02226-02 from the National Heart, Lung, and Blood Institute. R. A. Bialecki is supported by National Research Service Award Postdoctoral Fellowship. 1-F-32-HL-08321-01. W. S. Colucci is an American Heart Association-Sandoz Established Investigator. Address for reprint requests: W. S. Colucci, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis St., Boston, MA 02115. Received
7 January
1992; accepted
in final
form
17 June
1992.
REFERENCES
B. A., and J. J. Enyeart. Gadolinium blocks low- and high-threshold calcium currents in pituitary cells. Am. J. Physiol. 259 (Cell Physiol. 28): C515-C520, 1990. 2. Bialecki, R. A., T. N. Tulenko, and W. S. Colucci. Cholesterol enrichment increases basal and agonist-stimulated calcium influx in rat vascular smooth muscle cells. J. Clin. Invest. 88: 1894-1900, 1991. 3. Brown, R. D., K. D. Berger, and P. E. Taylor. cul-Adrenergic receptor activation mobilizes cellular calcium in a muscle cell line. J. Biol. Chem. 259: 7554-7562, 1984. Jr., and R. 4. Colucci, W. S., T. A. Brock, M. A. Gimbrone, W. Alexander. Nonlinear relationship between cul-adrenergic receptor occupancy and norepinephrine-stimulated calcium flux in cultured vascular smooth muscle cells. Mol. Pharmacol. 27: 517524, 1985. 5. Dartsch, P. C., and H. Hammerle. Orientation response of arterial smooth muscle cells to mechanical stimulation. Eur. J. CeZl BioL. 41: 339-346, 1986. R. J. Gadolinium selectively blocks a component of 6. Docherty, calcium current in rodent neuroblastoma x glioma hybrid (NGl08-15) cells. J. Physiol. Lond. 398: 33-47, 1988. J. P., J. Qar, M. Fosset, C. Renterghem, and M. 7. Galizzi, Lazdunski. Regulation of calcium channels in aortic muscle cells by protein kinase C activators (diacylglycerol and phorbol esters) and by peptides (vasopressin and bombesin) that stimulate phosphoinositide breakdown. J. Biol. Chem. 262: 6947-6950, 1987. importance of alpha adrenoceptor-me8. Hester, R. K. Functional diated, DGOO-insensitive Ca2+ entry in rabbit aorta. J. Pharmacol. Exp. Ther. 247: 223-234, 1988. J. I. E., A. M. Rudolph, and M. A. Heymann. 9. Hoffman, Pulmonary vascular disease with congenital heart lesions: pathologic features and causes. Circulation 64: 873-877, 1981. 10. Kent, R. L., K. Hoober, and G. Cooper IV. Load responsiveness of protein synthesis in adult mammalian myocardium: role of cardiac deformation linked to sodium influx. Circ. Res. 64: 74-85, 1989. 11. Kirber, M. T., J. V. Walsh, Jr., and J. J. Singer. Stretchactivated ion channels in smooth muscle: a mechanism for the initiation of stretch-induced contraction. Pflugers Arch. 412: 339345, 1988. P. R., S. R. Bates, M. B. Mathews, A. L. Horwitz, 12. Kollros, and S. Glagov. Cyclic AMP inhibits increased collagen production by cyclically stretched smooth muscle cells. Lab. Invest. 56: 410-417, 1987. W. S. Colucci, A. Rothman, E. 13. Kulik, T. J., R. A. Bialecki, Stretch increases inositol T. Glennon, and R. H. Underwood. trisphosphate and inositol tetrakisphosphate in cultured pulmonary vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 180: 983-987, 1991. 14. Kulik, T. J., J. N. Evans, and W. J. Gamble. Stretch-induced contraction in pulmonary arteries. Am. J. Physiol. 255 (Heart Circ. Physiol. 24): Hl391-H1398, 1988. 1.
Biagi,
MUSCLE
15. Larson,
CALCIUM
D. M., K. Fugiwara,
R. W. Alexander,
and M. A.
Gimbrone.
16 l
17.
Heterogeneity of myosin antigenic expression in vascular smooth muscle in vivo. Lab. Invest. 50: 401-407, 1984. Nakamura, T., and M. Ui. Simultaneous inhibitions of inositol phospholipid breakdown, arachidonic acid release, and histamine secretion in mast cells by islet-activating protein, pertussis toxin. J. Biol. Chem. 260: 3584-3593, 1985.
Neary, J. T., C. van Breemen, E. Forster, berg, and M. D. Norenberg. ATP stimulates primary astrocyte 1410-1416, 1988.
18.
cultures.
Biochem.
Biophys.
L. 0. B. Norencalcium influx in Res. Commun. 157:
Rabinovitch, M., M. A. Konstam, W. J. Gamble, N. Papanicolaou, M. J. Aronovitz, S. Treves, and L. Reid. Changes
in pulmonary blood flow affect vascular response to chronic hypoxia in rats. Circ. Res. 52: 432-441, 1983. transducers. In: CeZZ Shape: 19. Sachs, F. Ion channels as mechanical Determinants, Regulation, and Regulatory Role, edited by W. D. Stein and F. Bronner. New York: Academic, 1989, p. 63-92. The Principles and 20. Sokal, R. R., and F. J. Rohlf. Biometry: Practice of Statistics in Biological Research. San Francisco, CA: Freeman, 1981, p. l-105. 21. Sperti, G., and W. S. Colucci. Phorbol ester stimulated bidirectional transmembrane calcium flux in A,& vascular smooth muscle cells. Mol. PharmacoZ. 32: 37-42, 1987. 22. Stenmark, K. R., A. A. Aldashev, E. C. Orton, A. G.
Durmowicz, F. Voelkel,
D. B. Badesch, W. C. Parks, R. P. Mecham, N. and J. T. Reeves. Cellular adaptation during
chronic neonatal iol. Suppl. (Oct.) 23.
Sumpio,
hypoxic pulmonary 261: 97-104, 1991.
B. E., and A. J. Banes.
smooth muscle cells to cyclic tensional Surg. Res. 44: 696-701, 1988. 24.
Takemura, H., A. R. Hughes, Putney, Jr. Activation of calcium
hypertension.
Am.
J. Phys-
Response of porcine aortic deformation in culture. J.
0. Thastrup,
and J. W.
entry by the tumor promoter thapsigargin in parotid acinar cells. Evidence that an intracellular calcium pool and not an inositol phosphate regulates calcium fluxes at the plasma membrane. J. BioZ. Chem. 264: 12266-12271, 1989. 25.
Tamaoki, Morimoto, phospholipid/Ca Res. Commun.
26.
T., H. Nomoto, I. Takahashi, Y. Kato, and F. Tomita. Staurosporine, a potent inhibitor 2+-dependent 135: 397-402,
protein 1986.
kinase.
Biochem.
M. of
Biophys.
Tozzi, C. A., G. J. Poiani, A. M. Harangozo, C. D. Boyd, and D. J. Riley. Pressure-induced connective tissue synthesis in pulmonary artery segments CZin. Invest. 84: 1005-1012,
is dependent 1989.
on intact
J.
D. J., and R. A. Janis. Calcium channel ligands: structure-function relationships. In: Structure and Physiology of the Slow Inward Calcium Channel, edited by D. J. Triggle. New York: Liss, 1987, p. 29-50. 28. Vandenburgh, H. H. A computerized mechanical cell stimulator for tissue culture: effects on skeletal muscle organogenesis. In Vitro Dev. Biol. 24: 609-619, 1988. 29. Wallnofer, A., C. Cauvin, and U. T. Ruegg. Vasopressin increases 45Ca2+ influx in rat aortic smooth muscle cells. Biochem. Biophys. Res. Commun. 148: 273-278, 1987. 30. Wieczorek, D. F., C. W. J. Smith, and B. Nadal-Ginard. The rat alpha-tropomyosin gene generates a minimum of six different mRNAs coding for striated, smooth and non-muscle isoforms by alternative splicing. Mol. CeZZ. BioZ. 8: 679-694, 1988. 31. Wirtz, H. R., and L. G. Dobbs. Calcium mobilization and exocytosis after one mechanical stretch of lung epithelial cells. Science Wash. DC 250: 1266-1269, 1990. 32. Yang, X.-C., and F. Sachs. Block of stretch-activated ion channels in Xenopus oocytes by gadolinium and calcium ions. Science Wash. DC 243: 1068-1071, 1989. 33. Yatani, A., J. Codina, Y. Imoto, J. P. Reeves, L. Birnbaumer, and A. M. Brown. A G protein directly regulates mammalian cardiac calcium channels. Science Wash. DC 238: 12881292, 1987. 34. Zhou, X.-L., M. A. Stumpf, H. C. Hoch, and C. Kung. A mechanosensitive channel in whole cells and in membrane patches of the fungus uromyces. Science Wash. DC 253: 1415-1417, 1991. 27.
Triggle,
endothelium.
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
STRETCH
METHODS
37°C using 0.125 mg/ml elastase (type III, 90 U/mg, Sigma Chemical, St. Louis, MO), 1 mg/ml collagenase (company code CLS, type I, 150 U/mg, Worthington Biochemical, Freehold, NJ), 0.25 mg/ml soybean trypsin inhibitor (type l-S, Sigma), and 2 mg/ml crystallized bovine serum albumin (Pentex, Miles Laboratories, Elkhardt, IN). The PASMC were grown (37”C, 5% C02-95% air, in a humidified atmosphere) in medium 199 (GIBCO Laboratories, Grand Island, NY) with 20% (vol/vol) fetal calf serum (FCS; GIBCO), 100 U/ml penicillin, and 0.1 mg/ml streptomycin. After multiple passages, a clonal line was selected. By immunofluorescent staining, these cloned cells stably express smooth muscle-specific myosin [antibody courtesy of Dr. D. M. Larson (15)] and smooth muscle a-actin (Enzo Biochem, New York) over many passages. They also express mRNA for the smooth muscle a-actin isoform (by Northern blotting using a riboprobe derived from the 3’ untranslated region of the smooth muscle a-actin isoform, courtesy of Dr. J. L. Lessard) and predominantly express the smooth muscle cu-tropomyosin mRNA isoform (30). For stretch experiments, the PASMC were grown on thick (0.01 in.) silicone sheets (Silastic; Dow Corning, Midland, MI) fashioned into “tubs” (34 mm wide x 40 mm long x 10 mm deep), which were precoated with rat tail type I collagen (28) (Fig. 1). The PASMC were seeded into the tubs at an initial density of lO,OOO-20,000 cells/cm2 and grown in Dulbecco’s modified Eagle’s medium (DMEM) containing antibiotics and 10% FCS. The cells reached confluence at 2-4 days and were then used for study. Cells were incubated in DMEM containing 0.5% FCS and no antibiotics for 18-24 h before experimentation to remove potential antibiotic inhibitors of stretch-activated ion channels (19) and to reduce the tonic influence of serum factors on calcium flux. Static stretch of PASMC. To stretch the cells, a solid support fashioned at one end of the tub was moved to the appropriate stop position (Fig. 1). It took -1 s for the stretch to be accomplished. In preliminary experiments, the uniformity with which the Silastic surface is elongated by stretch was determined. The surface of the tub was divided into a grid of 15 equally spaced points arranged in five rows perpendicular to the axis of stretch, and the local elongation of the tub along the axis of stretch was measured during a 20% stretch using a microscope with a reticle. Local elongation at points located in four of the rows ranged from 18 to 25% (mean = 21 +- 1%) and was slightly reduced in the one row closest to the movable support (mean = 16 t 1%). Preliminary experiments utilizing phase-contrast microscopy demonstrated that the magnitude of deformation of the cells closely corresponded with the magnitude of elongation of the membrane. In addition, the cells did not detach from the collagen-coated Silastic surface during a 20% stretch sustained for 1 h (data not shown). Calcium uptake. Cell 45Ca2+ uptake was determined as previously described, with minor modifications (2, 3, 17, 21). The millimolar composition of the physiological salt solution (PSS) was as follows: 130 NaCl, 5 KCl, 1.5 CaCl,, 1 MgC12, 5 glucose, and 5. N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid. Zero-sodium PSS contained 130 mM choline chloride instead of NaCl. To initiate the assay, 6 ml of PSS containing 45Ca2+ (1 &i/ml) with or without other agents, was added to each tub.
CeLL culture. Main and proximal branch pulmonary arteries from adult Sprague-Dawley rats (killed by cervical dislocation) were stripped of adventitia and enzymatically dissociated at L602 1040-0605/92 $2.00 Copyright 0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
STRETCHING
ALTERS
SMOOTH
MUSCLE
L603
CALCIUM
ICN Biomedicals, Irvine, CA. Verapamil and pertussis toxin were purchased from Sigma. Gadolinium chloride was obtained from Aldrich Chemical, Milwaukee, WI. Staurosporine was purchased from Boehringer Mannheim Biochemicals, Indianapolis, IN. Drugs that required dissolution in ethanol (e.g., verapamil) were adjusted to a final vehicle concentration of l:lO,OOO at the drug concentrations indicated. Accordingly, in all experiments involving ethanol as solvent, control cells were incubated with the vehicle at a final concentration of l:lO,OOO.
a
RESULTS
Fig. 1. Line drawing of apparatus used to stretch pulmonary arterial smooth muscle cells. Plexiglas supports (b) are glued to 140-mm tissue culture dish (a). Silicone sheet is glued to Plexiglas end supports (d, e) to form the tub (c), which is 34 mm wide, 40 mm long, and 10 mm deep. The 20% linear elongation is effected by lifting movable end support (e) with a forceps so that stop (f) is lifted from slot and extended to end of support. This distance is 8 mm, which is 20% of length of silicone surface. Although not illustrated here, in some apparatuses, Plexiglas supports have slots to accommodate movable end support at intervals appropriate for 10,20, and 30% elongation, so that different magnitudes of stretch can be achieved.
Five seconds later, the tub was stretched by 20%. Uptake was terminated after various intervals by rapid washing of the cells five times with 6 ml of ice-cold PSS containing IO mM LaCl, and no CaC12, with a lo-min dwell in the final wash aliquot. The cells were then dissolved in 1 ml of 0.1 N HN03, and cellassociated 45Ca2+ was determined by counting of the radioactivity within the acid digest by standard scintillation techniques. All cell monolayers studied in the presence of IO PM verapamil, IO PM gadolinium, or 25 nM staurosporine were preincubated with the agent for 30 min before the initiation of the assay. Calcium ef’lux. The efflux of 45Ca2+ was measured as previously reported, with minor modifications (2,4,21). In brief, cell monolayers were equilibrated with 45Ca2+ for 18-24 h in DMEM containing 45CaC12 (1 &i/ml) and no antibiotics. After equilibration, the cells were washed rapidly four times with 6 ml of PSS at 37”C, and calcium efflux was initiated by the addition of 6 ml of PSS and other agents under study. The tubs were stretched (10, 20, or 30%) 5 s later. The assay was terminated after 6 min by rapid washing of the cells five times with 6 ml of ice-cold PSS containing LaCl, and no CaCl,, with a lo-min dwell in the final wash aliquot. The cells were then dissolved in 1 ml of 0.1 N HN03, and cell-associated 45Ca2+ was determined by counting of the radioactivity within the acid digest by standard scintillation techniques. Calcium efflux is expressed as the decrease in cellular calcium content over 6 min and reported as the percent change from that in unstretched control cells. In some experiments, efflux was evaluated in nominally calciumfree PSS containing 2 mM EGTA and no CaC12. Analysis of data. Statistical significance was determined by Student’s two-tailed t test for paired samples (20). P values ~0.05 were considered significant. In all cases, n equals the number of experiments performed. The data from each experiment represent the average of duplicate or triplicate tubs. Data are presented as means t SE. Reagents. 45CaC12 (Il.3 mCi/mg calcium) was obtained from
Kinetics of stretch-induced calcium uptake. To determine whether PASMC stretch is associated with an increase in calcium uptake, the early time course of calcium uptake was measured in stretched cells and, for comparison, in unstretched control cells (Fig. 2). Calcium uptake was linear for the first minute in both stretched and control cells, reflecting unidirectional calcium influx. At all time points up to 2 min, calcium uptake was greater (P < 0.01; n = 3 experiments) in stretched cells than in unstretched controls (Fig. 2). Stretch (20%) increased the average 1-min calcium uptake by 71 t 7% (P < 0.01; n = 12) compared with unstretched control cells. The kinetics of stretch-induced calcium uptake were characterized further by assessment of the duration of the increase in calcium uptake during a single sustained stretch (Fig. 3). For these studies, cells were stretched 20% for various times before cells were pulsed with 45Ca2+ for 1 min. Stretch caused a marked increase (111 t 16%; P < 0.01; n = 3) in cellular 45Ca2+ uptake at 1 min compared with uptake in control cells in unstretched tubs. During the fifth minute of a 5-min sustained stretch, calcium uptake was reduced to 46 t 17% (P < 0.05; n = 3) of the first minute response. During the last minute of a 60-min sustained stretch, calcium uptake was no different from that in unstretched cells (P = 1.18; n = 3) EffeEt of channel blockers on stretch-induced calcium uptake. To elucidate the pathway of stretch-induced cal-
cium influx, 1-min calcium uptake was measured in the presence of either 10 PM verapamil, an inhibitor of Ltype voltage-dependent calcium channels (27), or 10 PM gadolinium, a lanthanide element shown to inhibit stretch-induced calcium influx in several types of cells - 2.2 -a 2 2.0 5
--
1.8-
+I 1.6E g 1.4-
/
/
/
/
/
---
-0
/ /o-----**
**
1 .o Time
(min)
Fig. 2. Effect of stretch on calcium uptake in vascular smooth muscle cells as assessed by uptake of 45Ca2+. A sustained 20% stretch (open circles) caused a significant increase in calcium uptake at all time points compared with unstretched control cells (solid circles). Each point represents mean & SE of 3 determinations. * P < 0.05. ** P < 0.01.
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
L604
STRETCHING
z
140cn 6 voII z looc 808 Q) cl _: 60-
ALTERS
SMOOTH
MUSCLE
CALCIUM
**
* * 1
X
3 u!I
40-
20; 3 O-3 .0 3 -20 0
I 1
I 1 min
5 min
60
Fig. 3. Effect of sustained 20% stretch on calcium influx. Cells stretched for 1, 5, or 60 min were pulsed with 45Ca2+ for 1 min throughout 1-min stretch or during last minute of 5 and 60-min stretches. Bars, mean t SE of at least 3 experiments performed in triplicate. * P < 0.05 vs. unstretched controls. ** P < 0.01 vs. unstretched controls.
(3234). Verapamil decreased calcium uptake in stretched cells by 49 t 8% (P < 0.01; n = 5) compared with that in unstretched control cells. Gadolinium likewise caused a 43 t 7% reduction (P < 0.05; n = 5) in stretch-induced calcium uptake compared with that in stretched control cells. Concurrent incubation with 10 PM verapamil and 10 PM gadolinium decreased the stretch-induced calcium uptake response by 49 t 11% (P < 0.04; n = 3), an effect no greater than that of either verapamil or gadolinium alone. Role of sodium influx in stretch-induced
calcium uptake.
Replacement of extracellular sodium by 130 mM choline chloride had no effect on the magnitude of stretch-induced calcium uptake (n = 3; data not shown). Effect of staurosporine induced calcium uptake.
and pertussis
CTL
min
toxin on stretch-
Both protein kinase C and G proteins may influence calcium influx in a variety of cells (7, 21, 33). The role of protein kinase C activation in the stretch-induced increase in calcium uptake was evaluated by preincubating vascular smooth muscle cells with 25 nM staurosporine, an inhibitor of protein kinase C (25). The stretch-induced increase in calcium uptake of staurosporine-treated cells was not different from that of stretched control cells (73 t 32% vs. 59 t 8%, respectively; P = 0.68; n = 6). The possible role of a pertussis toxin-sensitive G protein in the mediation of stretch-induced calcium uptake was determined in stretched cells preincubated with pertussis toxin (20 rig/ml, 3 h). The stretch-induced increase in calcium uptake of pertussis toxin-treated cells was not different from that of control cells (79 t 47% vs. 59 t 8%, respectively; P = 0.70; n = 6). Stretch-induced calcium efflux. Stretching of the cells by 20% increased the 6-min calcium efflux by 47 t 5% (P < 0.001; n = 11) over that in unstretched control cells. The relationship between the magnitude of cell stretch and calcium efflux was linear (r = 0.984; P < 0.02; n = 3) over the range of lo-30% stretch (Fig. 4). An increase in calcium efflux reflects an increase in intracellular free calcium that may be due to the mobilization of intracellular calcium stores and/or the influx of extracellular calcium. To establish whether stretch induces mobilization of intracellular calcium stores inde-
I 20
10 Cell
Stretch
I 30
I
(percent)
Fig. 4. Effect of various degrees of stretch on 6-min calcium efflux as assessed by 45Ca 2+. Basal calcium efflux in unstretched control cells was 3.5 k 0.5 nanomol calcium. mg protein-l. 6 min. Data depict percentage increase in calcium efflux over basal in unstretched cells. Each point represents mean k SE of 4 experiments performed in triplicate.
pendent of calcium influx, stretch-induced calcium efflux was assessed in nominally calcium-free PSS containing 2 mM EGTA. Stretch-induced calcium efflux was 62 t 9% lower (P < 0.01; n = 5) in calcium-free PSS than in normal PSS (Table 1). This result suggests that ~38% of the stretch-induced calcium efflux response may be attributable to mobilization of intracellular calcium. Consistent with the role of extracellular calcium in stretch-induced calcium efflux, gadolinium (10 PM) decreased stretch-stimulated calcium efflux by 58 t 13% (P < 0.01; n = 6) in calcium-containing buffer but had no effect on stretch-induced calcium efflux in calcium-free buffer. In calcium-free buffer, stretch-stimulated calcium efflux was 20 t 2% and 23 t 2% in the absence and presence of gadolinium, respectively. Gadolinium had no effect on agonist-stimulated (thrombin, 10 U/ml) calcium efflux in unstretched control cells (20 t 1% vs. 22 t 2% without and with gadolinium, respectively), further indieating that gadolinium does not inhibit components of the pathway for agonist-mediated calcium mobilization (i.e., D-myo-inositol 1,4,5-trisphosphate generation, sarcoplasmic reticulum calcium release, or plasmalemma calcium adenosine triphosphatase). DISCUSSION
It has long been suspected that mechanical stress may cause changes in the contractile state or growth behavior of pulmonary blood vessels (9). Although it is not clear whether mechanical stress may act directly on pulmonary Table 1. Effect of gadolinium calcium efflux in the presence of extracellular calcium
on stretch-induced and absence
% Increase
Stretch Stretch Stretch Stretch
n
53t6 8 22&6* 6 20t2* 5 (10 ,uM) + [Cal, = 0 23&2-j3 Data are means t SE of percent increase in efflux over basal efflux in control cells in unstretched Silastic tubs. n, no. of experiments, each performed in triplicate. Calcium efflux over 6 min was assessed with 45Ca2+ in response to a sustained 20% elongation of Silastic culture tub. * P < 0.01 vs. stretch: -F P < 0.05 vs. stretch. + gadolinium + [Cal, = 0 + gadolinium
(10 PM)
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
STRETCHING
ALTERS
SMOOTH
vascular smooth muscle to cause growth and connective tissue production (26)) experiments using cultured aortic smooth muscle cells suggest that stretch can act directly on smooth muscle cells to modify protein production (12). However, the mechanism by which a mechanical stimulus is coupled to these or other effects is not known, and little is known about the mechanism of stretch-mediated signal transduction in PASMC. Our experiments indicate that a single sustained stretch of PASMC is a potent stimulator of both calcium influx and efflux. The magnitude of these effects is large, comparable to that stimulated by potent vasoconstrictors (2,4,21,29) in cultured vascular smooth muscle cells. Twenty percent stretch for 1 min caused an -70% increase in calcium uptake, a response that is similar in magnitude to the vasopressin- and endothelinstimulated increases in calcium uptake (69 t 3 and 85 t 14%, respectively) we observed previously in PASMC (2). Likewise, a single 20% stretch maintained for 6 min caused a 47% increase in calcium efflux, an effect that is greater than that stimulated by 10 PM norepinephrine (4) Stretch has been shown to activate nonspecific ion channels in a variety of cell types, including gastric smooth muscle cells (11). Our data extend these observations by showing for the first time in vascular smooth muscle that stretch increases calcium influx. In several types of cells, the lanthanide element gadolinium has been shown to block stretch-induced ion channels (32, 34). In PASMC, we found that stretch-induced calcium influx was inhibited w 45% by gadolinium, as well as by verapamil, which is generally regarded as an inhibitor of L-type calcium channels (27). The combination of verapamil plus gadolinium caused no more inhibition of stretch-induced calcium influx than did either agent alone. This result suggests that these agents act on the same pathway. Because gadolinium is also known to block voltage-dependent L-type calcium channels (1, 6), these data are consistent with the interpretation that the gadolinium/verapamil-sensitive component of stretch-induced calcium influx in this cell may be mediated by voltage-dependent calcium channels. An alternative possibility is that both agents act on a stretch-activated channel that is distinct from the voltage-dependent Ltype channel or an as yet undefined site. The component of stretch-induced calcium influx that is not sensitive to gadolinium and verapamil likely reflects influx by a second pathway. Because, as discussed subsequently, stretch also appears to cause mobilization of intracellular calcium stores, it is possible that some or all of this component of calcium influx is involved in the refilling of these intracellular stores (8, 24). Because stretch may increase sodium influx in cardiac myocytes (lo), we considered the possibility that sodium influx was required for stretch-induced calcium influx in PASMC. Removal of extracellular sodium did not diminish stretch-induced calcium influx, thereby suggesting that sodium influx does not play an important role in this effect. Protein kinase C has been shown to modulate the conductance of L-type calcium channels in vascular smooth muscle cells (7,21). However, the failure of staurosporine,
MUSCLE
CALCIUM
L605
an inhibitor of protein kinase C (25), to inhibit stretchinduced calcium influx does not support an important role for protein kinase C in mediating stretch-induced calcium influx. Likewise, the lack of inhibition of stretchinduced calcium influx by pertussis toxin suggests that a pertussis toxin-sensitive G protein is not required for this effect. However, this finding does not exclude involvement of one of several pertussis toxin-insensitive G proteins. Furthermore, although the protocol we used is adequate to ribosylate sensitive G proteins in other cells fully (16), we cannot exclude the possibility that the pertussis toxin treatment did not result in adequate ribosylation of the G protein in these experiments. The stretch-induced increase in calcium influx attenuates over time, such that it is reduced -50% by 5 min and is abolished after 60 min of static stretch. This pattern of attenuation suggests that stretch-induced calcium influx most likely occurs in response to a change in cellular distension rather than the absolute level of stretch. Stretch increased calcium efflux in the absence of extracellular calcium, suggesting that stretch also caused release of calcium from intracellular stores. Consistent with this interpretation, we have previously found that stretch increases D-myo-inositol trisphosphate and D-myo-inositol 1,4,5-tetrakisphosphate in PASMC (13). Stretch-induced calcium efflux was also partially inhibited by removal of extracellular calcium and the addition of gadolinium, presumably reflecting the component of efflux that is due to the influx of extracellular calcium. In the absence of extracellular calcium, gadolinium had no effect on stretch-induced calcium efflux, indicating that gadolinium does not interfere with stretch-induced mobilization of intracellular calcium. The magnitude of stretch used in these studies (20%) is similar to that used by several investigators in other cultured cell systems (5, 23, 31). Although observations by light microscopy and trypan blue exclusion suggested that stretch of PASMC by this amount does not induce detachment from the silicone membrane or cause cell injury, we cannot completely exclude the possibility that some of the increases in calcium influx and efflux reflects an unphysiological, albeit minor, disruption or the cell membrane. However, several observations argue that this is not the case. First, a substantial portion of the stretchinduced calcium uptake is inhibited by verapamil, a pharmacological agent that would not be expected to interfere with an injury-induced leakage of calcium. Second, stretch-induced calcium influx attenuates rapidly over time, whereas injury-induced influx might be expected to persist. Third, a significant component of stretch-induced calcium efflux occurs in the absence of extracellular calcium. In summary, these experiments indicate that a single static stretch of PASMC induces both calcium influx and efflux. A substantial portion of stretch-induced calcium influx occurs via a channel that is inhibited by both verapamil and gadolinium. Activation of this channel by stretch does not require sodium influx and is not inhibited by pertussis toxin or staurosporine. Although the major component of stretch-induced calcium efflux is secondary to calcium influx, a significant component does
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
L606
STRETCHING
ALTERS
SMOOTH
not require extracellular calcium, suggesting that stretch may also cause release of intracellular calcium. These data indicate that mechanical perturbations can exert a potent effect on calcium homeostasis in PASMC and that calcium mobilization may therefore play a role in the transduction of stretch in these cells. The authors thank Carolyn Kastner and Erin Smith for expert technical assistance throughout the course of these studies. This work was supported by Grants HL-429082, HL-429082, and HL-02226-02 from the National Heart, Lung, and Blood Institute. R. A. Bialecki is supported by National Research Service Award Postdoctoral Fellowship. 1-F-32-HL-08321-01. W. S. Colucci is an American Heart Association-Sandoz Established Investigator. Address for reprint requests: W. S. Colucci, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis St., Boston, MA 02115. Received
7 January
1992; accepted
in final
form
17 June
1992.
REFERENCES
B. A., and J. J. Enyeart. Gadolinium blocks low- and high-threshold calcium currents in pituitary cells. Am. J. Physiol. 259 (Cell Physiol. 28): C515-C520, 1990. 2. Bialecki, R. A., T. N. Tulenko, and W. S. Colucci. Cholesterol enrichment increases basal and agonist-stimulated calcium influx in rat vascular smooth muscle cells. J. Clin. Invest. 88: 1894-1900, 1991. 3. Brown, R. D., K. D. Berger, and P. E. Taylor. cul-Adrenergic receptor activation mobilizes cellular calcium in a muscle cell line. J. Biol. Chem. 259: 7554-7562, 1984. Jr., and R. 4. Colucci, W. S., T. A. Brock, M. A. Gimbrone, W. Alexander. Nonlinear relationship between cul-adrenergic receptor occupancy and norepinephrine-stimulated calcium flux in cultured vascular smooth muscle cells. Mol. Pharmacol. 27: 517524, 1985. 5. Dartsch, P. C., and H. Hammerle. Orientation response of arterial smooth muscle cells to mechanical stimulation. Eur. J. CeZl BioL. 41: 339-346, 1986. R. J. Gadolinium selectively blocks a component of 6. Docherty, calcium current in rodent neuroblastoma x glioma hybrid (NGl08-15) cells. J. Physiol. Lond. 398: 33-47, 1988. J. P., J. Qar, M. Fosset, C. Renterghem, and M. 7. Galizzi, Lazdunski. Regulation of calcium channels in aortic muscle cells by protein kinase C activators (diacylglycerol and phorbol esters) and by peptides (vasopressin and bombesin) that stimulate phosphoinositide breakdown. J. Biol. Chem. 262: 6947-6950, 1987. importance of alpha adrenoceptor-me8. Hester, R. K. Functional diated, DGOO-insensitive Ca2+ entry in rabbit aorta. J. Pharmacol. Exp. Ther. 247: 223-234, 1988. J. I. E., A. M. Rudolph, and M. A. Heymann. 9. Hoffman, Pulmonary vascular disease with congenital heart lesions: pathologic features and causes. Circulation 64: 873-877, 1981. 10. Kent, R. L., K. Hoober, and G. Cooper IV. Load responsiveness of protein synthesis in adult mammalian myocardium: role of cardiac deformation linked to sodium influx. Circ. Res. 64: 74-85, 1989. 11. Kirber, M. T., J. V. Walsh, Jr., and J. J. Singer. Stretchactivated ion channels in smooth muscle: a mechanism for the initiation of stretch-induced contraction. Pflugers Arch. 412: 339345, 1988. P. R., S. R. Bates, M. B. Mathews, A. L. Horwitz, 12. Kollros, and S. Glagov. Cyclic AMP inhibits increased collagen production by cyclically stretched smooth muscle cells. Lab. Invest. 56: 410-417, 1987. W. S. Colucci, A. Rothman, E. 13. Kulik, T. J., R. A. Bialecki, Stretch increases inositol T. Glennon, and R. H. Underwood. trisphosphate and inositol tetrakisphosphate in cultured pulmonary vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 180: 983-987, 1991. 14. Kulik, T. J., J. N. Evans, and W. J. Gamble. Stretch-induced contraction in pulmonary arteries. Am. J. Physiol. 255 (Heart Circ. Physiol. 24): Hl391-H1398, 1988. 1.
Biagi,
MUSCLE
15. Larson,
CALCIUM
D. M., K. Fugiwara,
R. W. Alexander,
and M. A.
Gimbrone.
16 l
17.
Heterogeneity of myosin antigenic expression in vascular smooth muscle in vivo. Lab. Invest. 50: 401-407, 1984. Nakamura, T., and M. Ui. Simultaneous inhibitions of inositol phospholipid breakdown, arachidonic acid release, and histamine secretion in mast cells by islet-activating protein, pertussis toxin. J. Biol. Chem. 260: 3584-3593, 1985.
Neary, J. T., C. van Breemen, E. Forster, berg, and M. D. Norenberg. ATP stimulates primary astrocyte 1410-1416, 1988.
18.
cultures.
Biochem.
Biophys.
L. 0. B. Norencalcium influx in Res. Commun. 157:
Rabinovitch, M., M. A. Konstam, W. J. Gamble, N. Papanicolaou, M. J. Aronovitz, S. Treves, and L. Reid. Changes
in pulmonary blood flow affect vascular response to chronic hypoxia in rats. Circ. Res. 52: 432-441, 1983. transducers. In: CeZZ Shape: 19. Sachs, F. Ion channels as mechanical Determinants, Regulation, and Regulatory Role, edited by W. D. Stein and F. Bronner. New York: Academic, 1989, p. 63-92. The Principles and 20. Sokal, R. R., and F. J. Rohlf. Biometry: Practice of Statistics in Biological Research. San Francisco, CA: Freeman, 1981, p. l-105. 21. Sperti, G., and W. S. Colucci. Phorbol ester stimulated bidirectional transmembrane calcium flux in A,& vascular smooth muscle cells. Mol. PharmacoZ. 32: 37-42, 1987. 22. Stenmark, K. R., A. A. Aldashev, E. C. Orton, A. G.
Durmowicz, F. Voelkel,
D. B. Badesch, W. C. Parks, R. P. Mecham, N. and J. T. Reeves. Cellular adaptation during
chronic neonatal iol. Suppl. (Oct.) 23.
Sumpio,
hypoxic pulmonary 261: 97-104, 1991.
B. E., and A. J. Banes.
smooth muscle cells to cyclic tensional Surg. Res. 44: 696-701, 1988. 24.
Takemura, H., A. R. Hughes, Putney, Jr. Activation of calcium
hypertension.
Am.
J. Phys-
Response of porcine aortic deformation in culture. J.
0. Thastrup,
and J. W.
entry by the tumor promoter thapsigargin in parotid acinar cells. Evidence that an intracellular calcium pool and not an inositol phosphate regulates calcium fluxes at the plasma membrane. J. BioZ. Chem. 264: 12266-12271, 1989. 25.
Tamaoki, Morimoto, phospholipid/Ca Res. Commun.
26.
T., H. Nomoto, I. Takahashi, Y. Kato, and F. Tomita. Staurosporine, a potent inhibitor 2+-dependent 135: 397-402,
protein 1986.
kinase.
Biochem.
M. of
Biophys.
Tozzi, C. A., G. J. Poiani, A. M. Harangozo, C. D. Boyd, and D. J. Riley. Pressure-induced connective tissue synthesis in pulmonary artery segments CZin. Invest. 84: 1005-1012,
is dependent 1989.
on intact
J.
D. J., and R. A. Janis. Calcium channel ligands: structure-function relationships. In: Structure and Physiology of the Slow Inward Calcium Channel, edited by D. J. Triggle. New York: Liss, 1987, p. 29-50. 28. Vandenburgh, H. H. A computerized mechanical cell stimulator for tissue culture: effects on skeletal muscle organogenesis. In Vitro Dev. Biol. 24: 609-619, 1988. 29. Wallnofer, A., C. Cauvin, and U. T. Ruegg. Vasopressin increases 45Ca2+ influx in rat aortic smooth muscle cells. Biochem. Biophys. Res. Commun. 148: 273-278, 1987. 30. Wieczorek, D. F., C. W. J. Smith, and B. Nadal-Ginard. The rat alpha-tropomyosin gene generates a minimum of six different mRNAs coding for striated, smooth and non-muscle isoforms by alternative splicing. Mol. CeZZ. BioZ. 8: 679-694, 1988. 31. Wirtz, H. R., and L. G. Dobbs. Calcium mobilization and exocytosis after one mechanical stretch of lung epithelial cells. Science Wash. DC 250: 1266-1269, 1990. 32. Yang, X.-C., and F. Sachs. Block of stretch-activated ion channels in Xenopus oocytes by gadolinium and calcium ions. Science Wash. DC 243: 1068-1071, 1989. 33. Yatani, A., J. Codina, Y. Imoto, J. P. Reeves, L. Birnbaumer, and A. M. Brown. A G protein directly regulates mammalian cardiac calcium channels. Science Wash. DC 238: 12881292, 1987. 34. Zhou, X.-L., M. A. Stumpf, H. C. Hoch, and C. Kung. A mechanosensitive channel in whole cells and in membrane patches of the fungus uromyces. Science Wash. DC 253: 1415-1417, 1991. 27.
Triggle,
endothelium.
Downloaded from www.physiology.org/journal/ajplung at Tulane University (129.081.226.078) on February 13, 2019.
