Vasopressin Stimulates Ca²⁺ Spiking Activity in A7r5 Vascular Smooth Muscle Cells via Activation of Phospholipase A₂

Materials and Methods

Results

The concentration dependence for release of intracellular Ca²⁺ stores by AVP was assessed by treating cells with varying concentrations of AVP in Ca²⁺-free medium (Fig 1⇓). An increase in [Ca²⁺]_i was detected in individual cells with exposure to AVP at concentrations ≥100 pmol/L (Fig 1⇓, left). The EC₅₀ for the peak [Ca²⁺]_i response to AVP, reflecting release of intracellular Ca²⁺ stores, was 9.8±2.1 nmol/L (n=4; Fig 1⇓, right). A previous study¹⁰ reported that the concentration dependence for the peak [Ca²⁺]_i response to AVP was similar in the presence or absence of extracellular Ca²⁺.

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

AVP stimulates release of intracellular Ca²⁺ stores with an EC₅₀ of ≈5 nmol/L. Left, Representative [Ca²⁺]_i responses from individual cells stimulated with the indicated [AVP] in Ca²⁺-free medium. Arrowheads denote the onset of perfusion of the cell chamber with AVP (extracellular Ca²⁺ was absent beginning 2 minutes before the addition of AVP). Traces are drawn to the same scale but are spaced apart vertically for clarity. Starting [Ca²⁺]_i values for each of the traces shown were between 17 and 54 nmol/L (mean, 35±4 nmol/L). Right, Concentration-dependent effects of AVP on the peak [Ca²⁺]_i in the absence of extracellular Ca²⁺. Results shown are mean±SE for eight cells at each [AVP]. Similar results were obtained in four separate experiments. Hyperbolic curves fit to the data for each experiment (Sigma Plot) were used to determine EC₅₀ values (9.8±2.1 nmol/L, n=4) and maximum peak [Ca²⁺]_i (862±43 nmol/L, n=4).

In Ca²⁺-containing medium, A7r5 cells display spontaneous Ca²⁺ spikes, which are dependent on activation of L-type voltage-sensitive Ca²⁺ channels.⁷ In the absence of AVP, the spontaneous Ca²⁺ spiking activity was sporadic and occurred at a very low frequency in most of the coverslips examined (0.02±0.01 min⁻¹, n=47). Treatment of A7r5 cells with increasing concentrations of AVP between 10 and 100 pmol/L (in 10-pmol/L increments) resulted in a marked increase in Ca²⁺ spike frequency, with little or no change in spike amplitude (Fig 2⇓, top). Ca²⁺ spike frequency usually increased abruptly when [AVP] was increased above a threshold of 20 to 60 pmol/L and then increased further with each step in [AVP] (Fig 2⇓). The increase in frequency was reversible by lowering [AVP] (not shown). A near-maximal [AVP] (100 nmol/L) still elicited the characteristic Ca²⁺ transient attributed to the release of intracellular Ca²⁺ stores but inhibited Ca²⁺ spiking, although in some experiments the inhibition was transient and spiking resumed at high frequency after ≈10 minutes (Fig 2⇓, top). An increase in spike frequency with increasing [AVP] in the 10- to 150-pmol/L range was observed in every case in >35 experiments.¹ In additional experiments, increasing [AVP] in larger increments (100 to 200 pmol/L) revealed Ca²⁺ spiking activity, which increased up to a frequency of ≈17 min⁻¹ at 500 pmol/L AVP. In experiments using both single cells and cell populations, Ca²⁺ spiking usually ceased abruptly when [AVP] was increased to 400 or 500 pmol/L (Fig 3⇓, top). Spiking continued at [AVP] up to 500 pmol/L in only two of six experiments on cell populations. In one of these experiments, spike frequency continued to increase up to 19 min⁻¹ at 700 pmol/L AVP, with no further increase when [AVP] was raised to 800 or 900 pmol/L. The relationship between frequency of Ca²⁺ spiking and [AVP] is shown in Fig 3⇓, bottom. One-way repeated-measures ANOVA followed by pairwise multiple comparison of spike frequency values with control spike frequency (Dunnett’s method, SigmaStat) revealed that frequency of Ca²⁺ spiking was significantly increased above control values (P<.05) for all AVP concentrations tested that were ≥30 pmol/L.

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

AVP stimulates Ca²⁺ spiking, increasing frequency with increasing [AVP] between 10 and 100 pmol/L. Top, A representative trace showing Ca²⁺-spiking activity in a population of A7r5 cells treated with increasing [AVP] (indicated by boxes along top axis). The lower trace is a continuation of the upper trace. Note the increasing frequency of spiking with little change in amplitude. Similar results were obtained in >35 similar experiments. A maximal [AVP] (100 nmol/L) produces the characteristic peak [Ca²⁺]_i response, followed after a lag of ≈10 minutes by a resumption of spiking at a higher frequency. Bottom left, Frequency of spiking from trace in the top panel, measured during last 10 minutes with each [AVP].

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Ca²⁺ spiking increases in frequency with [AVP] up to ≈500 pmol/L; higher concentrations are inhibitory. Top, Recording from a single cell in a monolayer treated with increasing [AVP] is shown. The frequency of Ca²⁺ spiking increases with [AVP] up to 300 pmol/L. Further increasing [AVP] to 400 pmol/L results in a stable elevated [Ca²⁺]_i. Similar results were obtained in two other experiments with single cells and in six experiments with cell populations (summarized in bottom panel). Bottom, Combined results from >40 experiments with cell populations reveal an increase in spike frequency in response to AVP, with an EC₅₀ of ≈300 pmol/L. For each [AVP] tested, the number of measurements averaged appears above each stippled circle. Error bars, representing SE, are absent when smaller than the symbol. In ≈10% of the coverslips tested, no Ca²⁺ spiking was observed during the recording period, either in the presence or absence of AVP. These data are included in the results shown, where a frequency of 0 was entered for each case in which no Ca²⁺ spikes occurred during exposure to a given [AVP] (measured beginning 5 minutes after switching perfusing solutions, for a period of 5 to 15 minutes). In four of six experiments in which [AVP] of >100 pmol/L was tested, spiking ceased when [AVP] was increased to 500 pmol/L. In one experiment, spiking continued with a modest increase in frequency, which stabilized at 19 min⁻¹ at [AVP] between 700 and 900 pmol/L.

The different mechanism involved in generating the Ca²⁺-spiking activity versus the release of intracellular Ca²⁺ stores was demonstrated by treating cells with nimodipine (50 nmol/L) to block L-type voltage-sensitive Ca²⁺ channels. Increasing [AVP] up to 150 pmol/L in the presence of nimodipine produced no Ca²⁺-spiking activity, but a near-maximal [AVP] of 100 nmol/L elicited a typical peak [Ca²⁺]_i response (Fig 4⇓). Peak [Ca²⁺]_i responses following acute exposure to 1 μmol/L AVP in Ca²⁺-free medium were similar in the presence or absence of 50 nmol/L nimodipine (387±34 nmol/L [n=5] and 411±82 nmol/L [n=4], respectively). Both the Ca²⁺-spiking response (Fig 5⇓) and the peak [Ca²⁺]_i response to AVP (not shown, but see Reference 10) were inhibited by d(CH₂)₅[Tyr(CH₃)²,Ala-NH₂⁹]AVP, a V_1a vasopressin receptor antagonist. These results suggest that although both the Ca²⁺-spiking effects and the peak [Ca²⁺]_i response to AVP are mediated by V_1a vasopressin receptors, only the Ca²⁺-spiking activity requires L-type voltage-sensitive Ca²⁺ channels; the peak [Ca²⁺]_i response is independent of these channels.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Nimodipine blocks Ca²⁺ spiking without inhibiting the peak [Ca²⁺]_i response to AVP. A population of A7r5 cells was exposed to increasing [AVP] in the presence of nimodipine (50 nmol/L), an inhibitor of L-type voltage-sensitive Ca²⁺ channels. These results illustrate the different mechanisms involved in the two [Ca²⁺]_i responses to AVP: Ca²⁺ spiking requires L-type Ca²⁺ channels, whereas the peak [Ca²⁺]_i response is independent of these channels. Results are representative of three independent experiments.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Stimulation of Ca²⁺ spiking by AVP is blocked by a V_1a vasopressin antagonist. No change in [Ca²⁺]_i was detected in a population of cells treated with [AVP] up to 150 pmol/L in the presence of d(CH₂)₅[Tyr(CH₃)²,Ala-NH₂⁹]AVP (10 nmol/L), a selective peptide antagonist of V_1a vasopressin receptors. Inhibition was overcome by increasing [AVP] to 1 μmol/L, resulting in high-frequency spiking and an elevated baseline [Ca²⁺]_i, similar to a typical response to a much lower [AVP] (100 to 400 pmol/L; see Fig 3⇑). Similar results were observed in three independent experiments.

The different concentration dependencies for AVP-stimulated Ca²⁺ spiking and release of intracellular Ca²⁺ stores also suggest that different signaling pathways may be involved in these two effects. AVP has recently been shown to be more potent in releasing arachidonic acid than in formation of InsP₃ in rat aortic smooth muscle cells,⁶ suggesting that activation of PLA₂ may occur in the [AVP] range that stimulates Ca²⁺ spiking. To examine the involvement of PLA₂ in AVP-stimulated Ca²⁺ spiking, ONO-RS-082, a PLA₂ inhibitor, was tested. ONO-RS-082 (1 μmol/L) had no effect on resting [Ca²⁺]_i levels but completely abolished AVP-stimulated Ca²⁺ spiking (Fig 6⇓, top and middle); a lower concentration of ONO-RS-082 (100 nmol/L) had no discernible effect (not shown). ONO-RS-082 (1 μmol/L) did not block the peak [Ca²⁺]_i response to 100 nmol/L AVP (Fig 6⇓, middle). Peak [Ca²⁺]_i responses following acute exposure to 1 μmol/L AVP in Ca²⁺-free medium were similar in the presence or absence of 1 μmol/L ONO-RS-082 (712±128 nmol/L [n=3] and 762±77 nmol/L [n=3], respectively). ONO-RS-082 had no effect on the [Ca²⁺]_i response to a depolarizing solution of high extracellular K⁺ (59 mmol/L extracellular KCl; Fig 6⇓, bottom), indicating that the inhibition of Ca²⁺ spiking by ONO-RS-082 was not due to inhibition of voltage-sensitive Ca²⁺ channels. Peak [Ca²⁺]_i responses following acute exposure to high extracellular K⁺ were the same in the presence or absence of 1 μmol/L ONO-RS-082 (212±3 nmol/L [n=3] and 212±20 nmol/L [n=3], respectively).

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

Inhibition of PLA₂ prevents stimulation of Ca²⁺ spiking by AVP. Top, The PLA₂ inhibitor ONO-RS-082 (1 μmol/L) reversibly blocks Ca²⁺ spiking in the presence of 20 pmol/L AVP. Middle, ONO-RS-082 (1 μmol/L) prevents stimulation of Ca²⁺ spiking by AVP with little effect on resting [Ca²⁺]_i, spontaneous Ca²⁺ spiking, or the peak [Ca²⁺]_i response to maximal [AVP] (100 nmol/L). Bottom, ONO-RS-082 (1 μmol/L) does not inhibit the [Ca²⁺]_i increase in response to an increase in extracellular K⁺ (high K⁺, 59 mmol/L extracellular KCl). This result suggests that ONO-RS-082 does not inhibit voltage-gated Ca²⁺ channels, which are activated by the high-K⁺ treatment. Each of the panels presents results representative of at least three independent experiments.

Inhibition of PLA₂ by ONO-RS-082 was evaluated by measuring arachidonic acid released from A7r5 cells during a 30-minute incubation with AVP in cells pretreated for 30 minutes in the presence or absence of ONO-RS-082 (1 μmol/L, Fig 7⇓). Arachidonic acid release was significantly increased in cells treated with 500 pmol/L AVP, and this increase was prevented by pretreatment with ONO-RS-082. 1 μmol/L AVP produced a slightly larger increase in arachidonic acid release, which was also prevented by pretreatment with ONO-RS-082 (Fig 7⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 7.

ONO-RS-082 inhibits AVP-stimulated arachidonic acid release. [³H]Arachidonic acid release was measured in A7r5 cells as described in “Materials and Methods.” ONO-RS-082 (1 μmol/L) was absent (open bars) or present (hatched bars) during a 30-minute pretreatment and a 30-minute treatment with control medium, 500 pmol/L AVP, or 1 μmol/L AVP. Results are mean±SE from three experiments with quadruplicate samples. Counts were normalized against the mean value for control samples pretreated without ONO-RS-082 for each experiment; this value was set to 100%. Both AVP-treated groups were significantly different from control (*P<.05 and **P<.01, Student’s t test). ONO-RS-082 treatment significantly reduced arachidonic acid release compared with AVP alone (both 500 pmol/L and 1 μmol/L AVP, P<.05).

Another PLA₂ inhibitor, HELSS, a specific inhibitor of Ca²⁺-independent PLA₂, has been found to inhibit AVP-stimulated arachidonic acid release without inhibiting PLC activation in A10 VSM cells.¹¹ HELSS (5 μmol/L) also prevented AVP-stimulated Ca²⁺ spiking in A7r5 cells (at [AVP] up to 100 pmol/L) without inhibiting the peak [Ca²⁺]_i response to 100 nmol/L AVP but was found to produce a slight elevation of [Ca²⁺]_i in the absence of AVP (not shown), suggesting that it may have other effects in addition to its inhibition of PLA₂.

Discussion

Proliferation of VSM cells is a key event in wound healing and, pathologically, in the development of atherosclerotic lesions and vascularization of tumors. Spontaneous increases in cytosolic [Ca²⁺] (Ca²⁺ spikes) in VSM are believed to be important for physiological regulation of vascular tone and may also play a role in hypertension and vasospastic disorders. AVP regulates both proliferation and spontaneous Ca²⁺ spiking in VSM cells, but probably via distinct mechanisms. Proliferation of human aortic smooth muscle cells was stimulated by AVP with an EC₅₀ of 3 nmol/L,¹² similar to the concentrations that elicit phosphatidylinositol turnover and Ca²⁺ release.^{4 5} In contrast, the effects of AVP on spontaneous Ca²⁺ spiking occur at much lower concentrations (EC₅₀ ≈300 pmol/L; Fig 3⇑). This effect correlates more closely with concentrations of AVP that stimulate arachidonic acid release.⁶

[AVP] in the plasma of normal healthy adults is ≈1 pmol/L,¹³ but baroreceptor responses triggered by lowering blood pressure by up to 50% in adult volunteers results in an increase in plasma [AVP] in direct proportion to the change in mean arterial blood pressure.¹⁴ Concentrations of AVP in volunteers whose mean arterial pressure was decreased by 16% to 44% corresponds to the levels found to stimulate arachidonic acid release in primary cultures of rat aortic smooth muscle⁶ and Ca²⁺ spiking in A7r5 cells (10 to 500 pmol/L, Figs 2⇑ and 3⇑). Similarly, a reduction of blood volume of 30% in rats increased plasma [AVP] to 300 to 500 pmol/L within 5 minutes.¹⁵ The higher concentrations of AVP required to stimulate proliferation of cultured cells are probably never attained in the systemic circulation, but the recent discovery that AVP may be synthesized by VSM cells³ raises the possibility that locally released AVP may reach concentrations high enough to stimulate proliferation in an autocrine or paracrine manner.

Activation of PLA₂ by AVP will lead to production of arachidonic acid and lysophospholipids. Arachidonic acid and its eicosanoid metabolites are ubiquitous compounds with complex effects on vascular tissues. A recent study using endothelium-denuded strips of rat thoracic aorta revealed that spontaneous rhythmic contractions of this preparation were increased in a concentration-dependent manner by arachidonic acid and prostaglandin F_2α and inhibited by cyclooxygenase and lipoxygenase inhibitors.¹⁶ Other arterial tissues tested in that study also produced rhythmic contractions, leading the authors to suggest that “rhythmic contractions are physiological characteristics of many and perhaps all blood vessels and may play a role in blood flow and turbulence; the likely cause of these oscillations is the cyclic release of one or more eicosanoids.”¹⁶ It remains to be determined which products of PLA₂ activation are responsible for the Ca²⁺-spiking effects in A7r5 cells and what the molecular targets are for these signals.

Previous studies have suggested that AVP stimulates PLA₂ independently of PLC, by activation of a distinct G protein–coupled signal transduction pathway.^{6 11 17} The different concentration dependencies for AVP-stimulated PLA₂ and PLC also suggest distinct pathways for activating these enzymes. Activation of PLC has an EC₅₀ similar to the K_d for AVP binding to V_1a vasopressin receptors (≈2 nmol/L), suggesting a close correlation between this response and the fraction of occupied receptors. In contrast, activation of PLA₂ and Ca²⁺ spiking occur at much lower [AVP] (EC₅₀ ≈50 to 300 pmol/L), corresponding to a fractional occupancy of only 2.5% to 13%, assuming a K_d of 2 nmol/L. Li et al¹⁸ found that a lower concentration of a V₁ vasopressin receptor antagonist was required to inhibit AVP binding than was required to inhibit Ca²⁺ signals or mitogen-activated protein kinase activation in rat aortic smooth muscle cells, suggesting that binding to a fraction of the receptors generates maximal levels of second messengers. The existence of “spare receptors” for AVP-stimulated signaling may be important for VSM cells, enabling these cells to detect very low circulating [AVP] and to respond rapidly to changes in [AVP].¹⁹ Circulating [AVP] is closely regulated by baroreceptor and osmosensor mechanisms. AVP-stimulated Ca²⁺ spiking in VSM may therefore represent an important physiological mechanism for the moment-to-moment regulation of vascular tone in response to changes in blood pressure or osmotic balance. A hypothetical scheme for AVP-stimulated Ca²⁺ signaling pathways is shown in Fig 8⇓.

• Download figure
• Open in new tab
• Download powerpoint

Figure 8.

Hypothetical AVP-stimulated Ca²⁺-signaling pathways. In A7r5 cells, AVP stimulates both Ca²⁺ spiking and the release of intracellular Ca²⁺ stores. Two distinct pathways may be involved in these effects of AVP: one involving activation of PLA₂ by low [AVP] (10 to 500 pmol/L), which leads to increased Ca²⁺ spiking, and another that is activated by higher [AVP] (0.5 to 100 nmol/L) and results in the activation of PLC, InsP₃ formation, and the release of intracellular Ca²⁺ stores.

Several features distinguish AVP-stimulated Ca²⁺ spiking from the release of intracellular Ca²⁺ in A7r5 cells: (1) Ca²⁺ spiking requires extracellular Ca²⁺⁷ and is abolished by nimodipine (Fig 4⇑), whereas AVP-stimulated Ca²⁺ release occurs in the absence of extracellular Ca²⁺ (Fig 1⇑) and in the presence of nimodipine (Fig 4⇑). (2) AVP-stimulated Ca²⁺ spiking is abolished by PLA₂ inhibitors ONO-RS-082 (Fig 6⇑) and HELSS (not shown), whereas Ca²⁺ release is resistant to these agents (Fig 6⇑, middle). (3) The Ca²⁺-spiking response to AVP is a frequency-modulated (FM) response; spike frequency changes dramatically while the amplitude of the Ca²⁺ spiking is fairly constant over a wide range of AVP concentrations (Figs 2⇑ and 3⇑). The release of intracellular Ca²⁺ by AVP is an amplitude-modulated (AM) response, which is graded with increasing [AVP], producing increases in [Ca²⁺]_i that reach much higher levels than occur during Ca²⁺ spiking (Fig 1⇑). (4) AVP-stimulated Ca²⁺ spiking occurs at much lower [AVP] (10 to 500 pmol/L, Figs 2⇑ and 3⇑) than is required to elicit the Ca²⁺-release response (100 pmol/L to 1 μmol/L, Fig 1⇑).

Single-cell imaging experiments also reveal that Ca²⁺ spiking occurs synchronously among all of the cells in a population, whereas the Ca²⁺-release response occurs asynchronously in individual cells (not shown, but see References 7 and 20). Synchronization of Ca²⁺ spiking presumably occurs because the cells are electrically coupled via gap junctions,^{7 21 22} and the spikes are triggered by an electrical stimulus that rapidly propagates between adjacent cells. The differences in the [Ca²⁺]_i signals may have important functional consequences for contraction of VSM cells. A synchronous [Ca²⁺]_i increase may be expected to produce a coordinated contraction within an artery, which may then propagate electrically away from the initial stimulus. The electrical propagation of the Ca²⁺-spiking signal may also serve to further sensitize VSM cells to AVP, because even a small localized increase in [AVP] detected by a few cells may increase spike frequency in neighboring cells that did not detect the change in [AVP]. The Ca²⁺-release response, in contrast, is apparently not communicated between cells via gap junctions and may be considered, therefore, to be a localized response that will occur only in those cells that detect a large increase in [AVP]. The latter response may be appropriate to produce a profound vasoconstriction and initiation of cell proliferation locally, eg, at the site of an injury. The more coordinated Ca²⁺-spiking response may be better suited to regulation of vascular tone in response to changes in plasma osmolality and blood pressure, both because it is exquisitely sensitive to concentrations of AVP found in the systemic circulation and because the frequency-modulated nature of the signal may more finely tune the contractile response than the amplitude-modulated signal that results from release of intracellular Ca²⁺.

Selected Abbreviations and Acronyms