Ca_V3.2 Channels and the Induction of Negative Feedback in Cerebral Arteries

Abstract

Rationale

T-type (Ca_V3.1/Ca_V3.2) Ca²⁺ channels are expressed in rat cerebral arterial smooth muscle. Although present, their functional significance remains uncertain with findings pointing to a variety of roles.

Objective

This study tested whether Ca_V3.2 channels mediate a negative feedback response by triggering Ca²⁺ sparks, discrete events that initiate arterial hyperpolarization by activating large-conductance Ca²⁺-activated K⁺ channels.

Methods and Results

Micromolar Ni²⁺, an agent that selectively blocks Ca_V3.2 but not Ca_V1.2/Ca_V3.1, was first shown to depolarize/constrict pressurized rat cerebral arteries; no effect was observed in Ca_V3.2^−/− arteries. Structural analysis using 3-dimensional tomography, immunolabeling, and a proximity ligation assay next revealed the existence of microdomains in cerebral arterial smooth muscle which comprised sarcoplasmic reticulum and caveolae. Within these discrete structures, Ca_V3.2 and ryanodine receptor resided in close apposition to one another. Computational modeling revealed that Ca²⁺ influx through Ca_V3.2 could repetitively activate ryanodine receptor, inducing discrete Ca²⁺-induced Ca²⁺ release events in a voltage-dependent manner. In keeping with theoretical observations, rapid Ca²⁺ imaging and perforated patch clamp electrophysiology demonstrated that Ni²⁺ suppressed Ca²⁺ sparks and consequently spontaneous transient outward K⁺ currents, large-conductance Ca²⁺-activated K⁺ channel mediated events. Additional functional work on pressurized arteries noted that paxilline, a large-conductance Ca²⁺-activated K⁺ channel inhibitor, elicited arterial constriction equivalent, and not additive, to Ni²⁺. Key experiments on human cerebral arteries indicate that Ca_V3.2 is present and drives a comparable response to moderate constriction.

Conclusions

These findings indicate for the first time that Ca_V3.2 channels localize to discrete microdomains and drive ryanodine receptor–mediated Ca²⁺ sparks, enabling large-conductance Ca²⁺-activated K⁺ channel activation, hyperpolarization, and attenuation of cerebral arterial constriction.

Cerebral arteries form an integrated network that controls the magnitude and distribution of tissue blood flow. Tone within these structures is regulated by multiple stimuli including blood flow,^1,2 neuronal activity,^3,4 tissue metabolism,⁵ and intraluminal pressure.⁶ These vasoactive stimuli alter myosin light chain phosphorylation through a dynamic process controlled by myosin light chain kinase and phosphatase.^7–9 Although the precise signaling mechanisms do vary among stimuli, a global rise in cytosolic [Ca²⁺]_i is generally thought to be a key mediating step.⁶ This global rise is in turn intimately tied to changes in the smooth muscle membrane potential (V_M) and the subsequent activation of voltage-gated Ca²⁺ channels.^6,10,11

Voltage-gated Ca²⁺ channels are heteromultimeric complexes that comprise a pore-forming α₁ subunit and auxiliary subunits that influence gating and protein trafficking to the plasma membrane.¹² In cerebral arterial smooth muscle, the L-type Ca_V1.2 channel is the dominant Ca²⁺ entry pathway by which vasoactive stimuli set cytosolic [Ca²⁺]_i and consequently tone development.^6,13 Recent studies have noted that, in addition to Ca_V1.2, T-type channels (ie, Ca_V3.1 and Ca_V3.2) are also expressed in rat and mouse cerebral arteries.^14–16 It has been argued that the T-type conductance, like that of Ca_V1.2, plays a direct role in elevating cytosolic [Ca²⁺]_i, albeit at hyperpolarized potentials attributable to a leftward shift in their voltage dependence.^14,15,17 Earlier reports have, however, suggested that the relationship between T-type channels and arterial tone is more complex with Ca²⁺ influx via Ca_V3.2 potentially acting in a discrete fashion to influence a defined target. Speculation of the downstream effector does vary, ranging from nitric oxide synthase in the endothelium to Ca²⁺-activated channels in the smooth muscle.^18–20

The large-conductance Ca²⁺-activated K⁺ channel (BK_Ca) is expressed in cerebral arterial smooth muscle, and its principal role is to feedback upon and limit excessive constriction. ^21,22 Vasoconstrictor stimuli enhance BK_Ca activity through arterial depolarization and augmentation of Ca²⁺ spark generation. Ca²⁺ sparks are discrete sarcoplasmic reticulum (SR)–driven events that arise in response to the transient opening of ryanodine receptors (RyRs).²³ Although the functional significance of Ca²⁺ sparks is recognized,^22–24 the mechanistic events that initiate repetitive SR release remain ambiguous, with current theories suggesting a role for Ca_V1.2 or transient receptor potential vanilloid 4 channel in triggering the cytosolic or luminal gate of RyR.^25–27 Decidedly absent from this discussion has been a potential role for a T-type conductance.

The present study tested whether Ca_V3.2 channel triggers Ca²⁺ spark generation, BK_Ca channel activation, and ultimately negative feedback control of cerebral arterial tone. Our examination progressed from cellular to tissue level and involved the integrative use of pressurized vessel myography, electrophysiology, confocal and electron microscopy, and computational modeling. In rat cerebral arteries, we specifically show that Ca_V3.2 and RyR colocalize within a microdomain and that steady-state depolarization activates Ca_V3.2 to trigger Ca²⁺ sparks. We subsequently show that Ca_V3.2-evoked Ca²⁺ sparks activate BK_Ca channels, contributing to hyperpolarization that attenuates myogenic constriction. We further demonstrate that this negative feedback mechanism is not limited to rat arteries but extends to the human cerebral circulation. Overall, this study is the first to illustrate that localized Ca²⁺ influx through T-type Ca²⁺ channels in vascular smooth muscle plays an important but indirect role in setting arterial tone by targeting key conductances involved in V_M regulation.

Materials and Methods

Female Sprague Dawley rats were euthanized by CO₂ asphyxiation as approved by the Animal Care and Use Committee at the University of Calgary. Rat brains were removed, placed in cold PBS, and middle and posterior cerebral arteries were isolated. Human cerebral arteries were extracted from resected brain tissue according to the University of Calgary Institutional Review Board. Structural analysis was performed using confocal, electron, and epifluorescence approaches. Vasomotor/V_M responses were subsequently assessed with the aid of pressure myography. Conventional, perforated, and on-cell patch clamp electrophysiology was used to record whole-cell and single-channel voltage-gated Ca²⁺ and BK_Ca currents. Data are presented as means ± SEM; paired or unpaired t tests were performed where appropriate, and a P<0.05 was considered statistically significant. An expanded version of the Materials and Methods can be found in the Online Data Supplement.

Results

Ca_V3.2 Inhibition: Vasomotor and Electric Responses in Rat Cerebral Arteries

Our examination began by confirming the ability of 50 μmol/L Ni²⁺ or 200 nmol/L nifedipine to block Ca_V3.2 and Ca_V1.2 currents, respectively, in tSA-201 cells. Figure 1A illustrates that Ni²⁺ effectively abolished inward Ba²⁺ current through Ca_V3.2 channels without affecting charge movement through Ca_V1.2 or Ca_V3.1 (Online Figure I). In comparison, nifedipine selectively blocked Ca_V1.2 channels (Online Figure I). Moving to rat cerebral arterial smooth muscle cells, we next monitored the nifedipine-insensitive Ba²⁺ current, a conductance that is dominated by T-type channels and is stable over time (Online Figure IIA and IIB).^28,29 Nickel partially attenuated this native inward current, a finding consistent with Ca_V3.2 channel expression (Figure 1B). On-cell recordings further denoted T-type activity in rat cerebral arterial smooth muscle. While Ca_V1.2 channels were blocked, single-channel activity was observed at hyperpolarized voltages (−50 to −20 mV), and slope conductance was 8.5 pS. The subsequent application of Ni²⁺ to endothelial-denuded arteries enhanced myogenic tone at 20 to 80 mm Hg (Figure 1C and 1D). Control experiments confirmed that arterial responses to pressure were repeatable over time (Online Figure IIC). Coincident with arterial constriction was an Ni²⁺-induced depolarization of 5 ± 0.9 mV in pressurized arteries (Figure 1E and 1F). The latter observation inferred that Ca_V3.2-mediated Ca²⁺ influx could elicit hyperpolarization and dilation through a smooth muscle signaling mechanism. In theory, this hyperpolarization could be triggered through localized Ca²⁺ entry initiating RyR-mediated Ca²⁺ sparks, SR events that activate BK_Ca channels.^18,19,23 Nickel’s effects on the nifedipine-insensitive Ba²⁺ current and arterial tone were reversible (not shown).

Figure 1. Effects of Ni²⁺ on Ca_V3.2 currents, myogenic tone, and membrane potential (V_M).

A, Representative traces and summary data of inward currents in Ca_V3.2-transfected tSA-201 cells in the absence and presence of Ni²⁺ (Ca_V3.2 blocker, 50 μmol/L). A voltage step from −90 to −10 mV was used to evoke inward Ba²⁺ current (n=5; *P<0.05, paired t test). B, T-type current in rat cerebral arterial smooth muscle before and after Ni²⁺ (50 μmol/L). Experiments were performed in the presence of nifedipine (200 nmol/L) to block L-type Ca²⁺ channels. A voltage step from −90 to 0 mV was used to elicit inward current (n=8; *P<0.05, paired t test). C and D, Single-channel recordings and summary current–voltage relationship (n=5; slope conductance=8.5 pS) of T-type Ca²⁺ currents in cerebral arterial myocytes. On-cell recording was performed at −50 to −10 mV in the presence of nifedipine (200 nmol/L), 60 mmol/L Ca²⁺, and 50 mmol/L TEA-Cl (tetraethylammonium chloride) to block K⁺ channels. E and F, Rat middle or posterior cerebral arteries were pressurized from 20 to 100 mm Hg, whereas diameter was monitored under control conditions, in the presence of Ni²⁺ (50 μmol/L) and in Ca²⁺-free medium. Representative traces (E) and summary data (F) display augmented arterial tone in response to Ni²⁺ (n=7; *P<0.05, paired t test). G and H, Arterial V_M in pressurized cerebral arteries (60 mm Hg) in the absence and presence of Ni²⁺ (50 μmol/L). Illustrative traces (G) and summary data (H) reveal the depolarizing effect of Ni²⁺ (n=6; *P<0.05, paired t test).

Microdomains and the Colocalization of Ca_V3.2 and RyR

Ca_V3.2 and RyR reside in the plasma and SR membranes, respectively. For these proteins to functionally interact, there must be regions where the 2 membranes come into close apposition. With this in mind, 3-dimensional electron tomography assayed for microdomains; image analysis and model reconstruction revealed the presence of microstructures which comprised caveolae and SR (Figure 2A–2C). These discrete regions were ≈500 to 600 nm in length and were circumferentially discontinuous. Immunogold labeling subsequently confirmed that RyR localized to regions underneath caveolae, whereas Ca_V3.2 was confined to the plasma membrane in-or-close to caveolae (Figure 2D–2F).

Figure 2. Electron microscopic imaging of rat cerebral arterial smooth muscle cells (SMCs).

A, Tissue sections (300 nm thick) were used to generate a contiguous stack of 2-dimensional photomicrographs (≈3.5 nm resolution); subcellular structures were subsequently traced on each section. B and C, Three-dimensional models of discrete membranous regions where caveolae and sarcoplasmic reticulum are in close apposition to one another. D and E, Transmission electron microscopy and immunogold labeling of ryanodine receptor (RyR; D) or Ca_V3.2 channels (E) in rat cerebral arteries. RyR labeling (arrowheads) can be observed in membranes localized beneath the plasma membrane. Ca_V3.2 labeling (arrowheads) was confined to the plasma membrane in association with caveolae. Boxed areas were magnified in the lower micrographs. F, Control experiments showed no electron-dense particles. Each photomicrograph is representative of 3 independent preparations.

To strengthen the emerging relationship between Ca_V3.2 and RyR, the preceding structural work was supplemented with an immunohistochemical analysis of fixed cerebral arteries using antibodies against actin, Ca_V3.2, and RyR. Findings in Figure 3A first illustrate that actin labeling runs lengthwise in cerebral arterial smooth muscle cells, fading every 7 to 10 μm as actin leaves the viewing plane. Ca_V3.2 staining was circumferential and often observed in regions devoid of smooth muscle actin. A similar circumferential pattern was observed for RyR2, a finding indicative although not definitive for colocalization with Ca_V3.2 (Figure 3B). Unlike Ca_V3.2 and RyR, Ca_V1.2 labeling was ribbon-like and ran lengthwise in smooth muscle (Figure 3C). A proximity ligation assay was subsequently performed, and consistent with Ca_V3.2 and RyR2 residing within 40 nm of one another, punctate red fluorescent product was observed in myocytes treated with both primary and secondary antibodies (Figure 4A). Reaction product was absent in control experiments where one or both primary antibodies were removed (Figure 4B–4D).

Figure 3. Ca_V3.2 displays localization patterns similar to ryanodine receptor 2 (RyR2) in rat cerebral arteries.

A, Cerebral arteries were labeled with antibodies against smooth muscle actin (green) and Ca_V3.2 (red). Labeling of Ca_V3.2 ran perpendicular (arrowheads) to the longitudinal axis of smooth muscle cells (SMCs; arrow) in regions devoid of smooth muscle actin. Bottom, Smooth muscle actin and Ca_V3.2 are displayed separately. B, Immunohistochemical staining of RyR (red) localized to areas where actin (green) was absent. Magnified panels show RyR was perpendicular to the longitudinal axis of SMCs. C, Ca_V1.2 labeling (arrowheads) was parallel to the longitudinal axis of SMCs (arrow). Right, The boxed area (middle) is magnified. Photomicrographs are representative of 3 independent experiments.

Figure 4. Proximity ligation assay of Ca_V3.2 and ryanodine receptor 2 (RyR2) in rat cerebral arterial smooth muscle cells.

A, A gallery representation reveals the presence of red fluorescent product consistent with Ca_V3.2 and RyR colocalization within 40 nm of one another. Nuclei were labeled with Hoechst 33342 (blue). B, Assay was performed with no primary antibodies. C and D, Assay controls were developed with 1 primary antibody. Scale bars are 10 μm, and optical section depth in each image is 0.3 to 0.5 μm. Photomicrographs are representative of ≈10 to 20 smooth muscle cells, and the assay was tested 2 to 3 times for each panel.

Ca_V3.2, RyR-Mediated Ca²⁺ Release, and the Induction of BK_Ca Activity

To ascertain at a conceptual level whether Ca²⁺ flux through Ca_V3.2 channels could activate RyR to initiate Ca²⁺ sparks, a computational model was designed. The microdomain model (Figure 5A) was developed based on the preceding structural data, measurements of Ca_V channel activity, and mathematical representations of other Ca²⁺ transporters/binding proteins. Findings illustrate that a depolarizing stimulus (from −60 to −40 mV) elicits Ca²⁺ spark-like events in the subspace between the plasma membrane and the SR (Figure 5B). These repetitive events fire at a frequency of ≈0.11 Hz and are fully abolished with RyR blockade (Figure 5C). In keeping with a role for Ca_V3.2, the elimination of this conductance attenuated these spark-like events (≈59% inhibition, Figure 5D). A broader voltage-dependent analysis also revealed that the frequency of Ca²⁺ spark-like events rose with depolarization (Figure 5E).

Figure 5. Computational modeling of the role of Ca_V3.2 in smooth muscle Ca²⁺ dynamics.

A, A computational model was developed using structural and electrophysiological data. The model consists of an 8.5-μm slice of an arterial smooth muscle cell. The microdomain is 600 nm in length and 15 nm from the sarcoplasmic reticulum (SR). Membrane proteins have been distributed, and the level of expression was set by optimization procedures. Key proteins include Ca_V1.2, Ca_V3.1, Ca_V3.2, ryanodine receptor (RyR), Na⁺/ Ca²⁺ exchanger (NCX), SERCA (sarco/endoplasmic reticulum Ca²⁺ ATPase)/PMCA (plasma membrane Ca²⁺ ATPase) pumps, calmodulin, and calsequestrin. Concentration of Ca²⁺ was calculated in the subspace region ([Ca²⁺]_ss). B, Simulations display repetitive Ca²⁺ spark-like events in response to depolarization from −60 to −40 mV. C and D, Spark-like events were fully abolished by RyR inhibition and attenuated by Ca_V3.2 blockade. E, Frequency of Ca²⁺ sparks increased with depolarization.

Moving forward to experimentally explore the Ca_V3.2/RyR relationship, Ca²⁺ imaging and line scan analysis were used to monitor Ca²⁺ sparks in rat cerebral arteries (Figure 6A). In opened tissues, Ca²⁺ sparks were observed in 57% of 291 line scans with a mean frequency of 0.0148 sparks/μm per second. Subsequent application of Ni²⁺ reduced event frequency by 53% (Figure 6A–6C) and had a significant effect on the spatial/temporal characteristics of Ca²⁺ sparks (Online Figure III). Given these positive observations, we next used perforated patch clamp electrophysiology to monitor spontaneous transient outward K⁺ currents (STOCs), BK_Ca-mediated events in response to Ca²⁺ spark generation.^21–23 Findings in Figure 6D show that STOCs were robustly observed in cerebral arterial myocytes and their frequency increased as the holding V_M was stepped from −40 to −20 mV. The subsequent application of 50 μmol/L Ni²⁺ reduced STOC frequency at −40 but not −20 mV, a finding consistent with the voltage dependence of Ca_V3.2 channels. The reduction in STOC frequency occurred without effect on amplitude (Figure 6D). In comparison, STOCs were abolished by 1 μmol/L paxilline, a BK_Ca inhibitor (Figure 6E). Control experiments (Online Figure IV) confirmed that peak inward/outward current in myocytes, slowly ramped from −60 to +20 mV, was unaffected by Ni²⁺. They also confirmed that 200 nmol/L nifedipine does not reduce STOC frequency at −40 mV (n=4: control, 70 ± 15 events/min; nifedipine, 62 ± 12 events/min). Overall, these results support the view Ca²⁺ influx via Ca_V3.2 channels drives BK_Ca activity via a mechanism involving RyR and the induction of Ca²⁺ sparks.

Figure 6. Ca_V3.2 suppression attenuates the generation of Ca²⁺ sparks and spontaneous transient outward K⁺ currents (STOCs).

A, Line scan imaging performed on posterior and middle rat cerebral arteries to ascertain Ca²⁺ spark generation under control conditions and in the presence of Ni²⁺ (50 μmol/L). Arrowheads denote the presence of Ca²⁺ sparks. B and C, Summary data highlight Ca²⁺ sparks frequency (sparks/μm per second) and the number of line scans in which Ca²⁺ sparks were detected (n=6 arteries, 291 line scans in total; *P<0.05, paired t test). D, Representative traces and summary data of STOC measurements under control conditions and in the presence of Ni²⁺ (50 μmol/L; n=8; *P<0.05, paired t test). Holding membrane potentials were set at −40 or −20 mV. E, STOCs were monitored before and after the application of paxilline (1 μmol/L; n=8; *P<0.05, paired t test). SMC indicates smooth muscle cell.

Further functional experiments were sought to emphasize the relationship between Ca_V3.2, BK_Ca activity, and the attenuation of arterial constriction. First, Figure 7A and 7B reveals that BK_Ca blockade (paxilline, 1 μmol/L) enhanced myogenic tone at intravascular pressures <80 mm Hg, akin to Ni²⁺ (Figure 1C and 1D). Second, when Ni²⁺ and paxilline were sequentially added, the first agent induced constriction, whereas the second had little or no additional effect (Figure 7C–7F). Control experiments confirmed that Ni²⁺-induced constriction at 60 mm Hg was absent in mesenteric arteries isolated from Ca_V3.2^−/− mice (Figure 7G and 7H). Overall, these results are consistent with Ca_V3.2 and BK_Ca channels working cooperatively within a common signaling pathway.

Figure 7. Effects of Ni²⁺ and paxilline on myogenic tone in rat cerebral arteries.

A, Middle and posterior cerebral arteries were gradually pressurized from 20 to 100 mm Hg while diameter was monitored. The experiment was performed under control conditions and in the presence of paxilline (1 μmol/L) or in Ca²⁺-free media. B, Summary data of the experiment in A (n=6; *P<0.05, paired t test). C and D, Traces and summary data illustrate the effects of sequential exposure to Ni²⁺ (50 μmol/L) followed by paxilline (1 μmol/L; n=7; *P<0.05, paired t test). E and F, The order of Ni²⁺ and paxilline in C was reversed (n=6; *P<0.05, paired t test). G and H, Traces and summary data illustrate the effects of Ni²⁺ (50 μmol/L) on pressurized mesenteric arteries (60 mm Hg) from wild-type (WT; n=4) and Ca_V3.2 knockout (n=6) mice (*P<0.05, unpaired t test).

Ca_V3.2 in Human Cerebral Arteries

A final set of experiments was conducted on human cerebral arteries to ascertain the translational impact of the preceding findings. Cerebral arteries were isolated from brain tissues resected from patients undergoing temporal lobectomy (Figure 8A). Polymerase chain reaction analysis on isolated smooth muscle cells, prescreened for endothelial contamination, illustrated that Ca_V3.2 mRNA was expressed (Figure 8B). Whole-cell patch clamp electrophysiology subsequently confirmed the presence of a nifedipine-insensitive current that was partially sensitive to 50 μmol/L Ni²⁺ (Figure 8C). Analogous to rat, Ni²⁺ application to human cerebral arteries elicited constriction and enhanced myogenic tone at pressure values ≤60 mm Hg (Figure 8D and 8E). These findings confirm that Ca_V3.2 is not only expressed in human myocytes, but its paradoxical role in tone development is likely akin to the rat.

Figure 8. Expression and function of Ca_V3.2 in human cerebral arteries.

A, Brain tissues were excised from patients undergoing lobectomy. Whole cerebral arteries and cerebral arterial smooth muscle cells (SMCs) were subsequently isolated for experimental assessments. B, Polymerase chain reaction (PCR) analysis of whole brain and isolated SMCs highlights the presence of Ca_V3.2. Data are representative of cells obtained from 2 human subjects. C, A voltage step from −90 to 0 mV was used to monitor inward Ba²⁺ current (n=6 cells from 4 subjects) in human cerebral arterial SMCs in the absence or presence of Ni²⁺ (50 μmol/L; *P<0.05, paired t test). Currents were monitored in the presence of nifedipine (200 nmol/L) to block L-type Ca²⁺ channels. D and E, Human cerebral arteries were pressurized from 20 to 100 mm Hg while diameter was sequentially monitored under control conditions and in the presence of Ni²⁺ (50 μmol/L) or in Ca²⁺-free media (n=4 arteries from 3 subjects; *P<0.05, paired t test). BP indicates base pairs.

Discussion

This study delineated Ca_V3.2 channels in cerebral arterial smooth muscle and determined whether this T-type conductance triggers Ca²⁺ sparks and consequently BK_Ca channels to elicit feedback control of arterial tone. Experiments progressed from cells to tissues and incorporated electrophysiology, cellular imaging, and computational modeling. Patch clamp electrophysiology confirmed the presence of a Ca_V3.2 current in cerebral arterial smooth muscle cells, a conductance selectivity blocked by micromolar Ni²⁺. In pressurized arteries, Ca_V3.2 blockade induced unexpected depolarization and constriction, a result indicative of Ca_V3.2 involvement in a dilatory process. A combination of structural and protein localization techniques revealed that Ca_V3.2 channels localize to microdomains in close apposition to RyR. Computational modeling then conceptually revealed that Ca_V3.2 could gate RyR, elicit Ca²⁺ sparks, and activate BK_Ca channels. Consistent with these predictions, Ni²⁺ inhibited Ca²⁺ spark production and STOC generation at physiological voltages. Further functional analysis reinforced this linkage by extending experiments to humans. In summary, this study is the first to demonstrate that Ca_V3.2 drives a local Ca²⁺-induced Ca²⁺ release event that restrains cerebral arterial constriction by triggering Ca²⁺ sparks and BK_Ca channel activation.

Background

The depolarization of cerebral arterial smooth muscle augments extracellular Ca²⁺ influx through the activation of voltage-gated Ca²⁺ channels. This response elevates global [Ca²⁺]_i, enhances myosin light chain phosphorylation, and augments arterial tone development.^6,9 Ca²⁺ channels are categorized according to the pore-forming α₁–subunit,¹² and in cerebral arterial smooth muscle, the L-type Ca_v1.2 is considered the primary conductance governing Ca²⁺ entry.⁶ Although Ca_V1.2 is a dominant conductance, recent studies have begun to acknowledge the expression of low-voltage activated Ca²⁺ channels in cerebral arteries.^14,15,28 T-type channels are the sole members of this subfamily, and as their name suggests, their activation/inactivation profiles are leftward shifted compared with the high-voltage activated L-type Ca²⁺ channels.^12,28 Ca_V3.1 and Ca_V3.2 are expressed in arterial smooth muscle, and recent work suggests that Ca²⁺ entry through one or both T-type channels could elevate global [Ca²⁺]_i, albeit at more hyperpolarized potentials, to modestly facilitate myogenic tone.^14–17,29 Although Ca²⁺ entry through T-type channels could drive bulk [Ca²⁺]_i changes, it could also elicit localized increases to gate conductances tied to V_M regulation.^18,19 To date, evidence of discrete Ca²⁺ signaling is limited in vascular tissue although studies have alluded to this possibility given the unexpected impairment of arterial dilation after Ca_V3.2 blockade.^19,30

Ca_v3.2 Channels in Cerebral Arteries

Studying vascular T-type channels is challenging because pharmacological tools display minimal subtype selectivity. The one exception is Ni²⁺ which, at low micromolar concentrations, selectively blocks Ca_V3.2 over Ca_V3.1 or Ca_V1.2, the primary Ca²⁺ channels in vascular smooth muscle.^28,31 We confirmed Ni²⁺ selectivity by transfecting the preceding Ca_Vx.x subunits into tSA-201 cells and monitoring the inward Ba²⁺ current (Figure 1; Online Figure I). Moving into cerebral arterial myocytes and focusing on the nifedipine-insensitive current dominated by T-type activity, ^28,29 Ni²⁺ attenuated but did not abolish this conductance, consistent with expression of both Ca_V3.2 and Ca_V3.1. On-cell electrophysiology further confirmed successful single-channel recordings with a slope conductance consistent with T-type channels.³² Although only a handful of vascular studies have exploited differential Ni²⁺ sensitivity to isolate Ca_V3.2 currents,^19,28,33 this approach is commonly used in cardiac/neuronal tissues to isolate this conductance and to ascertain its cellular function.^34,35 In this context, we show for the first time that selective Ca_V3.2 blockade augmented myogenic tone, findings paradoxical to typical vasodilatory effects of Ca²⁺ channel blockers.^6,14,15 The enhancement of tone resulted from the ability of Ni²⁺ to depolarize arterial V_M. These observations along with earlier reports¹⁹ indicate that Ca_v3.2 might elicit localized rise in cytosolic [Ca²⁺]_i that gates a K⁺ conductance that limits arterial constriction.

In the cerebral circulation, BK_Ca channels moderate vasoconstriction to agonists and elevated intravascular pressure. The channel comprises 4 pore-forming α₁-subunits and 4 β₁-subunits to confer Ca²⁺ sensitivity.^24,36 To activate BK_Ca, [Ca²⁺]_i must discretely rise to micromolar levels and this is achieved through Ca²⁺ spark generation, SR events dependent on RyR gating.^21,22,24 The opening of RyR is an integrated process, partially reliant on extracellular Ca²⁺ entry triggering the RyR cytosolic Ca²⁺ sensor. The identity of this entry channel is uncertain although past studies have alluded to candidates including transient receptor potential vanilloid 4 channel²⁵ and L-type channels.^26,27,37 Although both are plausible candidates, their intrinsic properties are somewhat inconsistent with a triggering role. Transient receptor potential vanilloid 4 channel displays voltage-independent properties, yet Ca²⁺ spark generation is graded in a voltage-dependent manner. L-type channels exhibit Ca²⁺-dependent inactivation and if positioned in a diffusion-restriction microdomain, high [Ca²⁺]_i would elicit strong inactivation, impinging on its ability to activate the RyR cytosolic gate. Because Ca_V3.2 channels are voltage gated, free of Ca²⁺-dependent inactivation, and display a voltage window that overlaps with physiological V_M,^12,28 this conductance seems best suited for microdomain localization and functioning as a trigger of Ca²⁺-induced Ca²⁺ release.

Microdomains and Ca²⁺ Channel Localization

For Ca²⁺ influx via Ca_V3.2 to trigger RyR and initiate Ca²⁺ sparks, the SR and plasma membranes must form a discrete signaling domain and then the proteins of interest must localize in close apposition to one another. In this context, we began our examination of a potential Ca_V3.2– RyR relationship using electron tomography, a structural technique that permits intercellular structures to be viewed in high 3-dimensional resolution. With this approach, microdomains that comprised caveolae and SR were readily identified (Figure 2). These discrete signaling regions were observed periodically along smooth muscle cells and were discontinuous longitudinally and circumferentially. Immunogold labeling subsequently placed Ca_V3.2 in-or-near caveolae and RyR in regions beneath these invaginated structures. In light of these findings, a broader immunohistochemical analysis was performed in which further evidence of Ca_V3.2–RyR colocalization was observed. First, using whole-mounted cerebral arteries, this study found that both Ca²⁺ pores were expressed in regions devoid of smooth muscle actin (Figure 3). Their circumferential labeling pattern was intriguing and strikingly distinct from Ca_V1.2. Second, a proximity ligation assay yielded a fluorescent product in isolated cells, consistent with Ca_V3.2 and RyR2 colocalizing within 40 nm of one another (Figure 4). Immunolabeling controls were negative, and antibody specificity was characterized, a priori, by Western blot analysis.¹⁴

Ca_V3.2 Channels, Ca²⁺ Sparks, STOCs, and Arterial Tone Development

To forward the stated hypothesis, it is important to consider Ca²⁺ flux in context with channel localization and the spatial compartments. Accordingly, our next step was to build a computational model to ascertain whether Ca²⁺ flux through Ca_V3.2 could, on a theoretical level, trigger the opening of RyR (Figure 5). Simulations revealed that at physiological V_M, Ca²⁺ spark-like events could be repetitively generated. Event frequency was coupled to voltage, an observation that aligns with experimental literature.^23,38,39 RyR blockade abolished these events, whereas Ca_v3.2 inhibition reduced frequency by 59%. In keeping with theory, analysis of opened arteries confirmed that Ca_V3.2 blockade decreased the frequency of Ca²⁺ spark (Figure 6); it also had a significant effect on their spatial/temporal characteristics (Online Figure III). The inability of Ni²⁺ to completely block discrete events indicates a complexity to RyR gating that extends beyond our focused analysis of Ca_v3.2. Future studies will need to consider whether other Ca²⁺ transporters are present in microdomains and able to trigger the cytosolic Ca²⁺ gate of RyR. Ca²⁺ transport proteins external to the microdomain could also foster Ca²⁺ sparks generation by altering SR refilling or the rate of Ca²⁺ diffusion from the subspace.^26,27

The predicted consequence of Ni²⁺ blockade and reduced Ca²⁺ spark generation should be decreased BK_Ca activity. We assessed the latter using perforated patch clamp electrophysiology to monitor STOCs in arterial myocytes. As denoted in Figure 6D and 6E, Ni²⁺ application reduced STOC frequency at physiological voltages (−40 mV) and had an insignificant effect at depolarized potentials where Ca_v3.2 channels reside in the inactive state. In comparison, paxilline abolished all STOC activity at both voltages. The voltage-dependent effect of Ni²⁺ is intriguing, and one that suggests that the ability of Ca_V3.2 to drive a feedback response might be confined to a specific V_M range. Functional observations in Figure 1 align with this perspective in that the Ni²⁺ effect on myogenic was greatest at intravascular pressures (40–60 mm Hg) where arterial V_M will overlap with the peak window current of Ca_V3.2. Given that paxilline augmented myogenic tone in an analogous manner to Ni²⁺, we can further suggest that Ca_V3.2 channels are likely a dominant trigger of BK_Ca in intact cerebral arteries (Figure 7). This view is further supported by our observations that placing one blocker on another had no additive effect on arterial tone.

In interpreting the preceding findings, it is important to consider the possible off-target effects of Ni²⁺. Past studies have noted that under certain conditions this divalent can affect voltage-gated K⁺ and depolarizing transient receptor potential currents.^40–43 Two lines of evidence indicate that such off-target effects are minimal in this study. First, electrophysiology revealed that Ni²⁺ had no effect on peak inward/outward current in smooth muscle cells ramped from −60 to 20 mV (Online Figure IV). Second, Ni²⁺ failed to alter tone in paxilline- pretreated cerebral arteries or in vessels isolated from Ca_V3.2 knockout mice (Figure 7).

Translation to Humans

Our work in rat cerebral arteries highlights a structural and functional association among Ca_V3.2, RyR, and BK_Ca, whereby voltage-dependent Ca²⁺ influx drives Ca²⁺ sparks generation and consequently arterial hyperpolarization. Although these foundational observations are unique and provocative, questions remained as to whether they translate to human tissue. In this context, we harvested human cerebral arteries from resection surgeries and repeated key experiments. We show that Ca_V3.2 mRNA is indeed present in human cerebral arterial myocytes. Further, Ca_V3.2 is functionally expressed because patch clamp electrophysiology delineated a Ni²⁺-sensitive T-type current. Finally, consistent with Ca_V3.2 driving arterial hyperpolarization, we found that Ni²⁺ constricted pressurized human cerebral arteries with an effect peaking at 60 mm Hg, where arterial V_M likely resides at −45 mV. These findings are the first to note T-type Ca²⁺ channel expression in human cerebral circulation and that it has a unique physiological role.

Summary

Vascular Ca²⁺ channels have been targets of investigative interest with Ca_v1.2 receiving particular attention given that dihydropyridines induce profound arterial dilation. With the recent isolation of T-type Ca²⁺ channels,^14,28,33 interest has begun to shift toward defining their physiological function. ^17,18,33 Vascular studies using blockers that do not discriminate among the T-type subunits have argued that they contribute modestly to global [Ca²⁺]_i albeit at hyperpolarized potentials.^14,17 The present study challenges this stereotypic view by arguing that Ca²⁺ influx through Ca_V3.2 acts in a localized manner to alter Ca²⁺-sensitive conductances involved in V_M regulation. Although this study focused specifically on Ca_V3.2, RyR, and BK_Ca, it is intriguing to speculate that Ca_V3.1 might also regulate a Ca²⁺-activated target such as transient receptor potential melastatin 4 channel or transmembrane member 16A channel, a Ca²⁺-activated Cl⁻ conductance.^44,45 Both conductances have been identified in arterial smooth muscle and linked to pressure-induced depolarization.^44–46

In summary, this study delineated Ca_V3.2 channels, explored their cellular expression, and examined their relationship to tone development in the cerebral circulation. Our examination used theoretical and experimental approaches from computational modeling to structural analysis, electrophysiology, and pressure myography. Ca_V3.2 channels were readily identified, shown to colocalize in microdomains with RyR to initiate Ca²⁺ sparks. These discrete events activate BK_Ca channels to facilitate arterial hyperpolarization and drive a feedback response that moderates constrictor events including those initiated by intravascular pressure. Because Ca_V3.2 channels are present in other vascular beds,^17,19,20,30 their feedback mechanism likely extends beyond the cerebral circulation. These findings also provide a straightforward explanation how Ca_V3.2 deletion paradoxically affects arterial relaxation.^19,30 This atypical Ca_V3.2 vasomotor response entails further attention given the emerging potential use of therapeutic T-type blockers for hypertension, cerebral vasospasm, or pain.^47–49

Supplementary Material

Supplement

Online Figure I. Effects of Ni²⁺ and nifedipine on Ca_V1.2, Ca_V3.1 and Ca_V3.2 currents. Summary data of inward currents in tSA-201 cells transfected with rCa_V1.2, rCa_V3.1 or rCa_V3.2-cDNA. Experiments were performed in the absence and presence of Ni²⁺ (50 μM) or Nifedipine (200 nM). A voltage step from −90 to 0 mV was used to evoke inward Ba²⁺ current (n=5 each, *P<0.05, paired t-test). Ni²⁺ and nifedipine selectively block Ca_V1.2 and Ca_V3.2, respectively.

Online Figure II. Time control experiments. (A, B) Representative traces and summary data of T-type current in rat cerebral arterial smooth muscle cells over time (n=8). Experiments were performed in the presence of nifedipine (200 nM) to block L-type Ca²⁺ channels. A voltage step from −90 to 0 mV was used to elicit inward current. (C) Rat middle or posterior cerebral arteries were pressurized from 20 to 100 mmHg twice under control conditions; passive responses were ascertained in Ca²⁺-free media (n=7). Electrical and vasomotor measurements were stable and repeatable over time.

Online Figure III. Effects of Ni²⁺ on the kinetic characteristics of Ca²⁺ sparks. Line scan imaging performed on posterior and middle rat cerebral arteries to ascertain Ca²⁺ spark generation under control conditions and in the presence of Ni²⁺ (50 μM). Summary data characterizes the influence of Ni²⁺ on spatial (A & B) and temporal (C & D) characteristics of Ca²⁺ sparks (n=6 arteries, 291 line scans in total, *P<0.05, paired t-test). Ni²⁺ elicited a limited but significant effect on the spatial and temporal characteristics of Ca²⁺ sparks.

Online Figure IV. Ni²⁺ control experiments. (A, B) Representative traces and summary data of inward and outward current in rat cerebral arterial smooth muscle cells under control conditions and in the presence of Ni²⁺ (50 μM, n=10). Cells were slowly ramped from −60 to +20 mV (4 mV/s). Ni²⁺ had no effect on peak inward/outward current.

Novelty and Significance.

What Is Known?

• T-type (Ca_V3.1/Ca_V3.2) Ca²⁺ channels are expressed in cerebral arterial smooth muscle.

• T-type Ca²⁺ channels are thought to mediate arterial tone development although the mechanisms remain uncertain.

What New Information Does This Article Contribute?

• Ca_V3.2 channels mediate a paradoxical dilation in pressurized cerebral arteries.

• Ca_V3.2 channels are located in microdomains in association with the sarcoplasmic reticulum and ryanodine receptors.

• Ca²⁺ influx through Ca_V3.2 channels triggers ryanodine receptors, generates transient Ca²⁺ sparks, and activates large-conductance Ca²⁺-activated K⁺ channels to elicit hyperpolarization.

T-type (Ca_V3.1/Ca_V3.2) Ca²⁺ channels are present in rat cerebral arterial smooth muscle, but their functional significance is uncertain. We tested whether Ca_V3.2 channels might mediate dilation rather than constriction by triggering Ca²⁺ sparks, discrete events that initiate arterial hyperpolarization by activating large-conductance Ca²⁺-activated K⁺ channels. Micromolar Ni²⁺, a Ca_V3.2 blocker, constricted pressurized rat cerebral arteries. Structural analysis revealed microdomains that comprised sarcoplasmic reticulum and caveolae, with Ca_V3.2 and ryanodine receptors residing next to each another. Modeling showed that Ca²⁺ influx through Ca_V3.2 could activate ryanodine receptors, and consistent with theory, Ca²⁺ imaging and electrophysiology demonstrated that Ni²⁺ suppressed Ca²⁺ sparks and downstream large-conductance Ca²⁺-activated K⁺ channel activity. Ca_V3.2 channels are also present in human cerebral arteries and drive a comparable response. In summary, we show for the first time that Ca²⁺ influx through Ca_V3.2 channels discretely activates Ca²⁺ sparks and large-conductance Ca²⁺-activated K⁺ channels to elicit arterial hyperpolarization and dilation. This feedback mechanism will prevent cerebral arteries from overly constricting to strong stimuli such as intravascular pressure. This new knowledge challenges the stereotypical view that Ca²⁺ channels are singularly involved in mediating arterial constriction.

Nonstandard Abbreviations and Acronyms

BK_Ca

large-conductance Ca2⁺-activated K⁺ channel

RyR

ryanodine receptor

SR

sarcoplasmic reticulum

STOC

spontaneous transient outward K⁺ current

V_M

membrane potential