Gα_q/11-mediated intracellular calcium responses to retrograde flow in endothelial cells

Abstract

Disturbed flow patterns, including reversal in flow direction, are key factors in the development of dysfunctional endothelial cells (ECs) and atherosclerotic lesions. An almost immediate response of ECs to fluid shear stress is the increase in cytosolic calcium concentration ([Ca²⁺]_i). Whether the source of [Ca²⁺]_i is extracellular, released from Ca²⁺ intracellular stores, or both is still undefined, though it is likely dependent on the nature of forces involved. We have previously shown that a change in flow direction (retrograde flow) on a flow-adapted endothelial monolayer induces the remodeling of the cell-cell junction along with a dramatic [Ca²⁺]_i burst compared with cells exposed to unidirectional or orthograde flow. The heterotrimeric G protein-α q and 11 subunit (Gα_q/11) is a likely candidate in effecting shear-induced increases in [Ca²⁺]_i since its expression is enriched at the junction and has been previously shown to be activated within seconds after onset of flow. In flow-adapted human ECs, we have investigated to what extent the Gα_q/11 pathway mediates calcium dynamics after reversal in flow direction. We observed that the elapsed time to peak [Ca²⁺]_i response to a 10 dyn/cm² retrograde shear stress was increased by 11 s in cells silenced with small interfering RNA directed against Gα_q/11. A similar lag in [Ca²⁺]_i transient was observed after cells were treated with the phospholipase C (PLC)-βγ inhibitor, U-73122, or the phosphatidylinositol-specific PLC inhibitor, edelfosine, compared with controls. Lower levels of inositol 1,4,5-trisphosphate accumulation seconds after the onset of flow correlated with the increased lag in [Ca²⁺]_i responses observed with the different treatments. In addition, inhibition of the inositol 1,4,5-trisphosphate receptor entirely abrogated flow-induced [Ca²⁺]_i. Taken together, our results identify the Gα_q/11-PLC pathway as the initial trigger for retrograde flow-induced endoplasmic reticulum calcium store release, thereby offering a novel approach to regulating EC dysfunctions in regions subjected to the reversal of blood flow.

disturbed laminar shear stress patterns, including retrograde flow, are known promoters of atherogenic signaling. These variations of hemodynamic parameters may occur naturally as a result of an increased vascular tone in the downstream resistant vessel beds, the geometrical nature of the vessels (anatomical curvature), or by the increase in diastolic retrograde flow often exacerbated with aging, hypertension, or obesity (40). It was recently shown that even at rest, older subjects exhibit greater brachial retrograde flow compared with younger subjects (30). Changes in the nature of the pulsatile laminar flow in straight segments of the vasculature may lead to the appearance of a net flow reversal in regions where the endothelium has geometrically adapted to shear by inclining in the direction of flow. Correlations between increased retrograde arterial blood flow and the severity of the congestive heart failure state or an acceleration of atherosclerotic lesion formation have been established (12, 16). In fact, increases in magnitude of retrograde shear were shown to induce dose-dependent reduction in endothelial function in vivo (38). Recently, surgical procedures such as the use of intra-aortic balloon pumping, the device most frequently used in circulatory support, or the commonly performed bypass surgeries also expose autologous vessels to nonnative flow reversal, which often leads to the development of neointimal hyperplasia (5, 6, 23).

The consequences of sustained retrograde flow include increased endothelial cell (EC) permeability and atherogenic gene expression (31). However, studies of early responses to shear stress in ECs may be key to understanding endothelium dysfunction. It was shown that flow-adapted ECs respond to sudden changes in flow direction by a restructuring of the cell-cell junction along with a rapid and significantly higher intracellular calcium concentration ([Ca²⁺]_i) than cells subjected to orthograde flow (24). Changes in [Ca²⁺]_i occur promptly and elevations in [Ca²⁺]_i have profound effects upon a host of cellular processes, including cell growth and differentiation, cell motility and contraction, intercellular coupling, and apoptosis and necrosis. Nonetheless, the molecular mechanisms linking changes in junctional inclination and the initiation of the [Ca²⁺]_i transients in response to reversal of flow remain to be determined.

The heterotrimeric G protein subunits-α q and 11 (Gα_q/11) are abundantly expressed at the endothelial cell-cell junction and are activated upon shear stress within seconds (15, 13, 29). The activation of G proteins may be direct (13) or indirect by flow-induced activation of G protein-coupled receptor (7). Upon receptor coupling, Gα_q/11 is known to activate β-isoforms of phospholipase C (PLC-β) and initiate inositol lipid signaling, which leads to calcium release from inositol 1,4,5-trisphosphate (IP₃)-regulated intracellular stores (17). Studies have shown that inhibition of Gα_q/11 significantly reduces systemic blood pressure and neointima formation in ferric chloride-induced carotid injury (19). Blocking of Gα_q was also shown to prevent pressure overload myocardial hypertrophy in transgenic mice (1). Furthermore, shear-induced nitric oxide production and Ras activation were shown to be altered by pertussis toxin-sensitive G proteins (14, 21). Nonetheless, no causal link has been clearly established to address the prominent role of Gα_q/11 in triggering the initial intracellular responses to shear. Phospholipase dependence of calcium mobilization was first suggested in bovine ECs subjected to mechanical stimulation (10). Although, mechanical stimulation was shown in the complete absence of external calcium by promoting calcium release from internal stores in ECs (9, 28, 25), a new class of plasmalemmal calcium channels, the transient receptor potential cation (TRPC) channels, has recently been described as being implicated in triggering calcium dynamics from extracellular pool (42). It is still unclear whether the source of calcium is extracellular, released from intracellular calcium stores, or both upon flow reversal in ECs.

To address to what extent the Gα_q/11-PLC pathway participates in initiating retrograde flow-induced calcium responses on flow-adapted monolayers, RNA interference techniques and inhibitory drugs were employed. Here we demonstrated the cooperative involvement of Gα_q/11 and PLC in mediating primary flow-induced calcium responses. The specific targeting of the IP₃ receptor (IP₃R) was sufficient to entirely block calcium transient in response to retrograde flow.

MATERIALS AND METHODS

Cell cultures and treatments.

Primary human coronary artery endothelial cells (HCAECs) were purchased from Lonza Walkersville (Walkersville, MD) and Cell Applications (San Diego, CA) and grown in complete EC growth media (EGM-2, Lonza). Before all experimental procedures, HCAECs were serum-starved overnight in ATP-free EBM-2 (Lonza), supplemented with 1% bovine serum albumin (BSA, Sigma-Aldrich, St. Louis, MO) to establish quiescence in the monolayer. The same starvation media was used for perfusion after overnight CO₂ equilibration. In some experiments, cells were treated with 100 nM final concentration of histamine dihydrochloride (Tocris Bioscience, Ellisville, MO) as a positive control for the activation of Gα_q/11 in ECs (32). The PLC inhibitors edelfosine and U-73122 (Tocris Bioscience) were resuspended on the day of experiment in water or DMSO (0.1% final) and used at 25 and 5 μM final, respectively, in the perfusion media during both the flow adaptation and experimental step flow. Concentrations of these inhibitors were chosen based on their abilities to significantly reduce, but not completely, histamine-induced calcium responses. Higher concentrations or longer exposure induced cytotoxicity during preincubation time as also observed by Sobolewski and colleagues (35). The inactive U-73122 analog, U-73343, was purchased from Enzo Life Sciences (Farmingdale, NY) and used at same final concentration. The IP₃R blocker xestospongin C (XeC) was purchased from Wako Chemical (Richmond, VA), and each batch was resuspended on the day of experiment in DMSO (0.5% final) at 10 μM final in perfusion media.

Silencing of Gα_q/11 was performed using a small interfering (si)RNA target sequence common for both human Gα_q and Gα₁₁ (siGα_q/11, sense: 5′-AAGATGTTCGTGGACCTGAA-3′). siRNA (375 nM) was combined with Lipofectamine siRNAmax (Invitrogen) according to the manufacturer's protocol and dropped onto subconfluent cells for 4 h before cells were reseeded at higher density onto fibronectin-coated glass slides or coverslips for 2 days, allowing cells to reestablish confluency and reform cell-cell junctions. Transfection of a nontargeting siRNA (siCTRL, Ambion, Foster City, CA) was used as a control.

Shear-stress exposure.

HCAEC monolayers were used in a parallel plate flow microchamber as described in Melchior and Frangos (24). A syringe pump delivered orthograde or retrograde flow rate via a computer-controlled PHD 2000 syringe pump (Harvard Apparatus) holding a 100-ml CO₂ gas-tight prewarmed syringe (SGE analytical science). Flow adaptation consisted of a preconditioning orthograde flow involving a 5-min slowly ramped-up flow to 5 dyn/cm², followed by a 20-min steady-flow period at 5 dyn/cm², and finally a 5-min ramped down flow. Cells then rested for 2 min, allowing time for the user to switch valves, and then were exposed to 10 dyn/cm² step retrograde flow according to experimental conditions. Sham controls were flow adapted but not exposed to retrograde flow. Static control slides were only removed from the incubator immediately before harvest.

Intracellular calcium measurement.

Cells were incubated overnight in starvation medium (1% BSA) and then incubated for 1 h with 15 ng/ml of the free intracellular calcium chelator fluo-4 AM (excitation, 494 nm; and emission, 516 nm), supplemented with 2.5 mM probenicid, 20 mM HEPES, 0.042% Pluronic F-127, and 0.1% BSA (Sigma) to facilitate the calcium indicator loading (all from Invitrogen, unless specified). Cells were then rinsed with HBSS containing calcium and incubated at 34°C for 30 min for equilibration in ATP-free media before mounting slides onto the flow chamber. Picture sequences were acquired continuously (8-bit unidirectional scan; time lapse, 1.57 s) under a confocal fluorescent microscope (Zeiss Pascal LSM5, Carl Zeiss), equipped with a ×40 Plan-NeoFluar/1.3 numerical aperture oil immersion objective, and the mean intensity of each selected cells was calculated using Volocity 5 software (Improvision/PerkinElmer, Waltham, MA). To control for cell-to-cell variation in dye loading, all fluorescence measurements were expressed as a ratio (F/F₀) of dynamic fluorescence intensity (F) to the average of basal fluorescence intensity measured in the 2-min period before onset of flow (F₀). A value of amplitude exceeding a maximum F/F₀ ratio above 1.5 was defined as a positive Ca²⁺ response. Internal controls showed there was limited photobleaching of the fluo-4 AM during the time course of our experiments.

IP₃ measurement.

IP₃ was detected on a fluorescence polarization reader (GeniosPro, Tecan) using the HitHunter IP₃-FP assay from DiscoveRx (Fremont, CA) following the manufacturer's instructions. Briefly, HCAEC slides subjected to a 10-s retrograde flow were immediately quenched in perchloric acid and put in competition with an IP₃ fluorescent tracer to tumble with an IP₃ binding protein. The polarized signal (in mP) is inversely proportional to the amount of IP₃, and measurements from triplicates were plotted on a standard curve run in parallel for each individual experiment. Concentration values were readjusted according to the final volume of lysate harvested for each samples and to a ×40 dilution factor to match our samples to the IP₃ standard concentration provided by the manufacturer. Each sample was measured in triplicate, and each well was an average of 25 readings.

Western blot analysis.

To determine Gα_q/11 expression after siRNA silencing, a separate set of slides also starved overnight in 1% BSA were scraped off at the time of experiments and cells were pelleted with ice-cold PBS, resuspended in 200 μl lysis buffer [consisting of 60 mmol/l octyl glucoside, 50 mM Tris·HCl (pH 7.5), 50 μmol/l EGTA, 125 mmol/l NaCl, 2 mmol/l dithiothreitol, 2 mmol/l Na₃VO₄, and protease inhibitors], and incubated on ice for 30 min. Lysates were centrifuged (14 000 g, 20 min, 4°C), and the detergent-soluble supernatant fractions were retained. Crude lysates were separated on NuPAGE 4–12% Bis-Tris gels (Invitrogen) and transferred to polyvinylidene difluoride membrane (Millipore, Temecula, CA). Membranes were blocked for 1 h with 5% BSA in Tris-buffered saline with 0.1% (vol/vol) Tween 20 (TBST) and then incubated with a primary antibody overnight in 3% BSA-TBST at 4°C. β-Tubulin antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA), and a rabbit polyclonal antibody (LJBI clone47) was produced in-house against the common COOH-terminal regions of both Gα_q and Gα₁₁. Bound primary antibodies were detected by horseradish peroxidase-conjugated secondary antibodies anti-rabbit IgG (Cell Signaling Technology, Danvers, MA). Band intensity was quantified on unsaturated X-ray film by a digital image analyzer (Bio-Rad, Hercules, CA).

Statistics.

Data are expressed as means ± SE from at least three independent experiments. Statistical comparisons between groups were performed using a one-tailed paired Student's t-test. A difference of P < 0.05 was judged as significant and indicated on bar graphs with an asterisk.

RESULTS

Delayed calcium response to retrograde flow in Gα_q/11-silenced ECs.

To address to what extent the Gα_q/11-PLC pathway participates in initiating retrograde flow-induced calcium responses, low-passage HCAEC monolayers were transfected with a custom-designed siRNA targeting a common sequence of both human the Gα_q and Gα₁₁ sequences. Western blot analysis on lysates harvested at time of experiments showed a >95% Gα_q/11 protein reduction compared with cells transfected with a nontargeting control siRNA (Fig. 1A, inset). Neither siRNA showed any cytotoxic effect within the 48 h following the transfection, and the integrity of the EC monolayer was confirmed by platelet endothelial cell adhesion molecule-1 immunostaining (data not shown). Application of a sudden 10 dyn/cm² retrograde step flow on flow-adapted Gα_q/11-silenced cells did not significantly reduce the magnitude of [Ca²⁺]_i response (Fig. 1, A and B, maximum F/F₀: siGα_q/11, 4.8 ± 0.7, n = 6; and siCTRL, 4.5 ± 0.5, n = 8, P = 0.35). Because individual cells within an EC monolayer may not be transfected uniformly, the ratio of cells responding to retrograde flow with a [Ca²⁺]_i increase was assessed and found not to be significantly different between the two groups (Fig. 1C; percentage of cells responding: siCTRL, 93.3 ± 12.3, n = 8; and siGα_q/11, 99.5 ± 0.7, n = 6). However, the two transfected groups showed differences in calcium dynamics at both the initiation of the calcium response and at the time to peak after onset of flow. Cytosolic calcium responses at both burst and peak were further delayed by 10 and 11 s, respectively, between the siCTRL and siGα_q/11-transfected cells (Fig. 1, A and D; siCTRL [Ca²⁺]_i peak, 24.8 s ± 1.6, n = 8; siGα_q/11 [Ca²⁺]_i peak, 35.8 s ± 2.4, n = 6; siCTRL [Ca²⁺]_i initiation, 18.3 s ± 1.7, n = 8; and siGα_q/11 [Ca²⁺]_i burst, 28.4 s ± 2.0, n = 6).

Fig. 1.

Increased latency in retrograde flow-induced intracellular calcium concentration ([Ca²⁺]_i) responses in G protein-α q and 11 subunit (Gα_q/11)-silenced endothelial cells. A: human coronary artery endothelial cells (HCAECs) were transfected with either a nontargeting small interfering (si)RNA (siCTRL) or a siRNA specially designed against a common region of the human Gα_q and Gα₁₁ nucleotide sequence (siGα_q/11). Knockdown of protein expression was monitored by Western blot analysis 48 h after transfection, and experiments were considered when the ratio of Gα_q/11 to β-tubulin protein expression was >95% of basal control (inset). After an initial 30-min period of gradual ramp-up and down flow, flow-adapted HCAEC monolayers were subjected to a 10 dyn/cm² step retrograde flow for 1 min (indicated as “Flow”). All fluorescence measurements were expressed as a ratio (F/F₀) of dynamic fluorescence intensity (F) to the average of basal fluorescence intensity measured in the 2-min period before onset of flow (F₀). F/F₀ values are means ± SE of n = 6 for siGα_q/11 (solid black) and n = 8 for siCTRL (dotted gray) experiments, and each individual experiment consists of an average of F/F₀ measurements from 24 cells for each 302 time points. Data were acquired every 1.6 s over a 5-min period. B: averages of F/F₀ peak values during the flow period for each siRNA-transfected group of experiments. C: percentage of cells responding to retrograde flow with increased [Ca²⁺]_i. Cells were considered responders when their maximum F/F₀ exceeded 1.5 after onset of flow. D: time to initiation (init) and time to peak of [Ca²⁺]_i responses. At both time points, significant lags in calcium transient were observed when cells were silenced with siGα_q/11 (*P < 0.05).

Lags in flow-induced calcium transient are PLC dependent.

One target of the Gα_q/11 subunit is the membrane-bound PLC-β (17). Using both U-73122, a nonselective inhibitor of the PLC, and the phosphatidylinositol-specific PLC specific inhibitor edelfosine, we observed a similar lag in calcium response to retrograde flow compared with their respective control [Fig. 2A: U-73122 [Ca²⁺]_i peak, 36.3 s ± 3.9, n = 6; U-73343 [Ca²⁺]_i peak, 24.8 s ± 3.0, n = 6; delay (U-73122 vs. U-73343) = 11.4 s; P = 0.019; and Fig. 2B: edelfosine [Ca²⁺]_i peak, 35.2 s ± 2.1, n = 6; control media [Ca²⁺]_i peak, 23.6 s ± 1.0, n = 8; delay (edelfosine vs. control) = 11.6 s; P = 0.00008].

Fig. 2.

Changes in retrograde flow-induced calcium dynamics after phospholipase C (PLC) inhibition. A: increased latency in [Ca²⁺]_i responses to retrograde flow after pretreatment with 5 μM of the broad PLC inhibitor U-73122 (solid black) compared with its analog U-73433 (dotted gray). [Ca²⁺]_i responses were monitored as in Fig. 1, retrograde flow (“flow”) was applied on HCAEC monolayers for 1 min at 10 dyn/cm². F/F₀ values are means ± SE; for both groups, n = 6 experiments in which 24 cells were monitored for each individual experiment. B: lags in time to peak of [Ca²⁺]_i response was also observed when perfusion media included 25 μM of the phosphatidylinositol-specific PLC inhibitor edelfosine (dotted gray, average of n = 6 experiments) compared with regular perfusion media (solid black, n = 8 experiments). Insets: averages ± SE of time to peak [Ca²⁺]_i responses for each treatment. *P < 0.05.

Gα_q/11/PLC-dependent flow-induced IP₃ levels.

Differences in the time to peak rather than the magnitude of the [Ca²⁺]_i responses in siGα_q/11 or PLC inhibitor-treated cells could imply a slower accumulation of a secondary messenger such as IP₃, which is produced in cells by PLC-mediated hydrolysis of phosphatidylinositol-4,5-biphosphate. To verify this hypothesis, we measured IP₃ accumulation in flow-adapted cells subjected to a 10-s retrograde flow (seconds before onset of the [Ca²⁺]_i burst in control cells) and compared it with a positive control exposed to 30 s of 100 nM histamine (Fig. 3). IP₃ levels were significantly increased in flow-induced siCTRL samples compared with the respective sham-treated cells [Fig. 3A, IP₃ (in nM) per 5.10⁵ cells, Sham_siCTRL, 25.32 ± 1.79, n = 5, compared with Flow_siCTRL, 60.89 ± 5.26, n = 6, P = 0.00034]. However, IP₃ levels remained at their respective sham levels when cells were silenced with siGα_q/11 (Sham_siGαq/11, 22.79 ± 9.22, n = 5, compared with Flow_siGαq/11, 27.81 ± 5.90, n = 6, P = 0.104; Flow_siCTRL vs. Flow_siGαq/11, P = 0.00096). Similarly, treatment with the PLC inhibitor U-73122 reduced retrograde flow-induced IP₃ levels compared with slides treated with the U-73433 analog [Fig. 3B; IP₃ (in nM) per 5.10⁵ cells; Flow_U-73343, 81.61 ± 11.1, n = 7 vs. Flow_U-73122, 29.13 ± 10.8, n = 7, P = 0.00395].

Fig. 3.

Effect of Gα_q/11-PLC pathway blocking on flow-induced intracellular inositol 1,4,5-trisphosphate (IP₃) concentration. A: HCAECs transfected with either siGα_q/11 or siCTRL were assessed for Gα_q/11 and β-tubulin protein content by Western blot analysis (left) or harvested for IP₃ intracellular concentration measurements (right) after cells were either subjected to 10 s of a 10 dyn/cm² retrograde flow, subjected to flow adaptation only (sham), or left under static conditions. B: similar measurements were performed after 10 s of retrograde flow and treatment with 5 μM of either U-73122 or its inactive analog U-73343. Addition of 100 μM of histamine for 30 s was used as positive control. *P < 0.05.

Inhibition of IP₃R abrogates all retrograde flow-induced calcium response.

Because [Ca²⁺]_i responses in Gα_q/11/PLC-inhibited cells were only delayed and not reduced in magnitude, we investigated the level of participation of the IP₃ pathway in triggering retrograde flow-induced [Ca²⁺]_i responses using the potent IP₃R blocker XeC. Surprisingly, 10 μM of XeC completely abrogated retrograde flow-induced [Ca²⁺]_i responses compared with the 0.5% DMSO control perfusion media (Fig. 4). Interestingly, whereas XeC-treated cells did not show any burst of [Ca²⁺]_i, a few cells noticeably showed a reduction from the basal cytosolic calcium after being subjected to retrograde flow (Fig. 4, arrowhead, and supplemental video S1), indicating a possible mechanism of flow-induced intracellular calcium extrusion, which was previously masked by the IP₃R response.

Fig. 4.

Abrogation of reversal flow-induced calcium transient after IP₃ receptor inhibition. [Ca²⁺]_i response to retrograde flow were measured as described in Fig. 1 in HCAEC monolayers with 10 μM of xestospongin C (XeC; solid black, average of n = 3 experiments) or an equivalent amount of DMSO carrier (dotted gray, average of n = 4 experiments) in the perfusion media. XeC completely abrogated retrograde flow-induced [Ca²⁺]_i responses. Calcium extrusion was detected in some cells at the onset of flow (arrowhead) as exemplified in supplemental video S1.

DISCUSSION

Using a model of flow-adapted arterial ECs, we addressed the role of the Gα_q/11-PLCβ pathway in modulating intracellular calcium responses to retrograde flow. Using both RNA silencing methods and an inhibitory drug approach, we observed an increased latency in triggering [Ca²⁺]_i responses and decreased levels of IP₃ at onset of flow. Moreover, blocking IP₃R completely inhibited calcium transients. We conclude that Gα_q/11-PLCβ activation is the primary pathway in triggering endoplasmic reticulum (ER) calcium release in response to retrograde shear stress in ECs.

In our model of retrograde shear stress on flow-adapted arterial cell monolayer, calcium transients consistently peaked around 25 s after onset of flow. Latency in building up calcium responses was first described to occur within 1 min of shear stress on bovine ECs (3), in which IP₃ was found to precede and peak after 20–30 s of flow in human umbilical vein endothelial cells (4, 27). In more recent studies, a significant lag between the onset of fluid-flow and [Ca²⁺]_i increase was also observed in intercalated, renal, and bone cells subjected to shear stress (20, 22, 33, 34, 41). Although likely to be a function of the nature of the force applied, this lag time differs between cell types, but the peak occurrence was typically found to be between 10 and 30 s after stimulus. Our data showed that reduced expression of Gα_q/11 or PLC inhibition consistently increased the time to peak calcium transients by 11 s, but the peak value was only slightly attenuated. Similar increased lags in calcium response were observed in bone cells treated with lipid raft-disrupting agent (41). Interestingly, Kou and colleagues (20) observed a 10-s increase in response time when lowering oscillatory shear stress in osteoblasts. The existence of an equivalent delay in calcium response can also be observed with a reduction of shear magnitude, and an activation threshold of the applied shear stress was suggested (2). In a model of mechanical forces applied by micropipettes on isolated ECs, the calcium response also reaches a maximum within 10–20 s and was further delayed by used of membrane calcium channel blockers (35). Overall, these data and ours strongly suggest that mechanically induced calcium responses require a threshold for activation and that any reduction in mechanical cue or treatment aiming at reducing accumulation of a second messenger only delays the threshold required for the calcium response.

To assess the participation of the second messenger IP₃ in triggering calcium release upon retrograde flow stimulation, cells were harvested seconds before initiation of flow-induced [Ca²⁺]_i responses. Levels of IP₃ were found to be reduced when the Gα_q/11-PLC pathway was compromised. In these experiments, it is hypothesized that inhibition of this pathway, albeit partial, allowed minute levels of IP₃ to accumulate to later reach a threshold for release of intracellular calcium stores. In fact, G protein-mediated IP₃ accumulation to a threshold for calcium release was previously shown in neuronal cells (39). In a similar study in Purkinje cells, calcium dynamics were shown to be threshold dependent on IP₃ concentration, which in turn was dependent on Gα_q activation (11).

Blocking of the IP₃R completely abrogated retrograde-flow [Ca²⁺]_i responses in flow-adapted arterial cell monolayers. There are reports of the nonspecific action of XeC, showing it as an equally potent blocker of the endoplasmic reticulum calcium pump (8). Whether this is the case in our experiments, our results still indicates that retrograde flow-induced calcium responses uniquely involve ER calcium store release rather than the induction of extracellular calcium influx through plasmalemmal nonselective calcium channels. This would be a unique feature of retrograde shear stress forces on arterial ECs. Indeed, other mechanical cues such as hypoosmotic stress have shown participation of extracellular calcium through stretch activation of the superfamily of TRPC channels (18, 36). With the use of mechanical vibration in ECs, it was shown that calcium influx starts first at regions of impact through membrane stretch-activated channels and then is followed by ER calcium release (26). Another study in TRPC6-overexpressing HEK293 cells suggests that simultaneous mechanical forces and receptor activation may synergistically amplify TRPC activation, but a 10 dyn/cm² shear force alone does not trigger any TRPC6 current (18). In our model of retrograde flow, it is suggested that the mechanical stimuli is likely due to membrane tension accumulation and repositioning of the cell-cell junction (24). With regard to our observations, this mechanism appears to uniquely involve ER calcium release, and it is suggested that plasmalemmal calcium channels activation, if any, is secondary upon retrograde-induced flow. Supporting this hypothesis, it was recently demonstrated that TRPC1 requires IP₃ as a prerequisite for activation (37).

Although the role of Gα_q/11 was previously demonstrated on shear-induced nitric oxide production and Ras activation (14, 21), no study had yet established a direct link between Gα_q/11-PLCβ pathway and early molecular events occurring at onset of flow stimulation in ECs. Our results here establish for the first time a causal link between Gα_q/11 activation and retrograde flow-induced [Ca²⁺]_i responses. The calcium response can be ascribed to the Gα_q/11-PLCβ pathway activation and restricted to ER calcium release in ECs. Although only delayed, calcium transients are significantly affected by the disruption of the Gα_q/11-PLCβ pathway. In this regard, Gα_q/11 can be considered as the primary component of the observed calcium transients evoked by retrograde shear stress.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute MERIT Award R37-HL-040696.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

B.M. and J.A.F. conception and design of research; B.M. performed experiments; B.M. analyzed data; B.M. and J.A.F. interpreted results of experiments; B.M. prepared figures; B.M. drafted manuscript; B.M. and J.A.F. edited and revised manuscript; B.M. and J.A.F. approved final version of manuscript.