CaV3.1 channels facilitate calcium wave generation and myogenic tone development in mouse mesenteric arteries

The arterial myogenic response to intraluminal pressure elicits constriction to maintain tissue perfusion. Smooth muscle [Ca2+] is a key determinant of constriction, tied to L-type (CaV1.2) Ca2+ channels. While important, other Ca2+ channels, particularly T-type could contribute to pressure regulation within defined voltage ranges. This study examined the role of one T-type Ca2+ channel (CaV3.1) using C57BL/6 wild type and CaV3.1−/− mice. Patch-clamp electrophysiology, pressure myography, blood pressure and Ca2+ imaging defined the CaV3.1−/− phenotype relative to C57BL/6. CaV3.1−/− mice had absent CaV3.1 expression and whole-cell current, coinciding with lower blood pressure and reduced mesenteric artery myogenic tone, particularly at lower pressures (20–60 mmHg) where membrane potential is hyperpolarized. This reduction coincided with diminished Ca2+ wave generation, asynchronous events of Ca2+ release from the sarcoplasmic reticulum, insensitive to L-type Ca2+ channel blockade (Nifedipine, 0.3 µM). Proximity ligation assay (PLA) confirmed IP3R1/CaV3.1 close physical association. IP3R blockade (2-APB, 50 µM or xestospongin C, 3 µM) in nifedipine-treated C57BL/6 arteries rendered a CaV3.1−/− contractile phenotype. Findings indicate that Ca2+ influx through CaV3.1 contributes to myogenic tone at hyperpolarized voltages through Ca2+-induced Ca2+ release tied to the sarcoplasmic reticulum. This study helps establish CaV3.1 as a potential therapeutic target to control blood pressure.


DAPI
4′,6-Diamidino-2-phenylindole IP 3  Inositol 1,4,5-trisphosphate PE Phenylephrine PLA Proximity ligation assay 2-APB 2-Aminoethoxydiphenyl borate Smooth muscle cells in the arterial wall actively contract to intravascular pressure, maintaining organ blood flow under dynamic conditions 1,2 .This "myogenic response" was first described by Bayliss 3 and is intimately tied to arterial depolarization, the activation of voltage-gated Ca 2+ channels and the concomitant rise in cytosolic Ca 2+ concentration ([Ca 2+ ] i ), which complexes with calmodulin driving myosin light chain phosphorylation 4 .Three subclasses of voltage-gated Ca 2+ channels (Ca V 1-3), are encoded in the mammalian genome and each displays unique voltage-dependent properties 5 .In arterial smooth muscle, Ca V 1.2 (L-type) Ca 2+ channels are principally responsible for extracellular Ca 2+ entry and their blockade is notable for attenuating a range of constrictor responses, including those induced by pressure 6 .
L-type Ca 2+ channels are classified as high-voltage-activated and dominate the setting of smooth muscle [Ca 2+ ] i when intravascular pressure is elevated and arteries depolarized 7 .Their activity, however, markedly drops with hyperpolarization as pressure is reduced or when endothelial cell K + channels are activated by selected

Isolation of mesenteric arterial smooth muscle cells
Third and fourth-order mesenteric arteries were placed in an isolation medium (37 °C, 10 min) containing (in mM): 60 NaCl, 80 Na-glutamate, 5 KCl, 2 MgCl 2 , 10 glucose and 10 HEPES with 1 mg/mL bovine serum albumin (BSA, pH 7.4).Vessels were then exposed to a two-step digestion process that began with 14-min incubation (37 °C) in media containing 0.5 mg/mL papain and 1.5 mg/mL dithioerythritol, followed by 10-min incubation in media containing 100 μM Ca 2+ , and collagenases type H (0.7 mg/mL) and type F (0.4 mg/mL).Following incubation, tissues were washed repeatedly with ice-cold isolation medium and triturated with a fire-polished pipette.Liberated cells were stored in ice-cold isolation medium for use the same day.

Immunohistochemistry
Ca V 3.1 expression was assessed in mesenteric arterial smooth muscle cells isolated from C57BL/6 and Ca V 3.1 −/− mice.Briefly, isolated cells were fixed onto a microscope cover glass in PBS (pH 7.4) containing 4% paraformaldehyde and 0.2% Tween 20.Fixed cells were blocked (1 h, 22 °C) with a quench solution (PBS supplemented with 3% donkey serum and 0.2% Tween 20) and subsequently incubated overnight (4 °C, humidified chamber) with rabbit anti-Ca V 3.1 primary antibody diluted in quench solution (1:100).In the following morning, cells were washed 3× in PBS-0.2%Tween 20 and then incubated (1 h, 22 °C) in a PBS-0.2%Tween 20 buffer containing Alexa Fluor 488 donkey anti-rabbit IgG-secondary antibody (1:200).After further washing, isolated cells and whole-mount preparations were mounted with Prolong Diamond Antifade Mountant with DAPI.Immunofluorescence was detected through a 63× oil immersion lens coupled to a Leica-TCS SP8 confocal microscope equipped with the appropriate filter sets.Smooth muscle cells isolated from C57BL/6 cerebral arteries were used as Ca V 3.1 positive controls.Secondary antibody controls were performed and were negative for nonselective labelling.

Electrophysiological recordings
Conventional patch-clamp electrophysiology was utilized to measure voltage-gated Ca 2+ currents in smooth muscle cells isolated from mesenteric arteries.Cell averaged capacitance was 12-18 pF.Recording electrodes (pipette resistance, 5-8 MΩ) were fashioned from borosilicate glass using a micropipette puller (Narishige PP-830, Tokyo, Japan) and backfilled with pipette solution containing (in mM): 135 CsCl, 5Mg-ATP, 10 HEPES, and 10 EGTA (pH 7.2).To attain a whole-cell configuration, the pipette was lowered onto a cell while applying negative pressure to rupture the membrane and garner intracellular access.Cells were voltage clamped (holding potential: − 60 mV) and subjected to − 90 mV followed by 10 mV voltage steps (300 ms) starting from − 50

Agonist-induced constriction
Endothelium-denuded mesenteric arteries were cannulated in a pressure myograph as explained above.Following the high potassium challenge, arteries were equilibrated at 60 mmHg, then subjected to the administration of phenylephrine (PE) into the bath solution.Increasing concentrations of PE (in M) 10 -7 , 3 × 10 -7 , 10 -6 , 3 × 10 -6 , 10 -5 , and 3 × 10 -5 were superfused into the bath containing PSS in the absence (control) or presence of 0.3 µM nifedipine.Changes in diameter in response to each concentration were recorded and percentage of maximum constriction was calculated as follows: where D is the external diameter after each agonist concentration application, D0 is the external diameter in Ca 2+ PSS, and Dm is the external diameter after the highest concentration of agonist under control condition.

Calcium imaging
Freshly isolated arteries were incubated with the Ca 2+ indicator Fluo-8 and placed on the stage of a Nikon sweptfield confocal microscope with enclosed Agilent 3B laser attached to Andor camera (iXon Ultra).Fluo-8 working solution (19.1 µM) was freshly prepared by dissolving 5 μL of stock solution (1.91 mM) in 5 μL pluronic acid plus 490 μL HBSS buffer consisting of (in mM): 134 NaCl, 6 KCl, 1 MgCl 2 , 2 CaCl 2 10 HEPES, and 10 Glucose (pH 7.4).Incubation was done for 75 min at 37 °C in the dark.Vessels were then cannulated and equilibrated at an intraluminal pressure of 15 mmHg for 15 min in Ca 2+ PSS solution.Intraluminal pressure was then raised to 60 mmHg, and Ca 2+ waves were recorded in the presence and absence of 0. www.nature.com/scientificreports/or 50 µM 2-APB or 3 µM xestospongin C. Fluo-8-loaded arteries were excited at 488 nm and emission spectra at 510 nm viewed through a 60× water immersion objective (1.2 WI) and were monitored and analysed using Nikon NIS Elements software (AR 4.20.01).To limit laser-induced tissue injury, image acquisition was set to 45 s at 5 fps.A series of regions of interest (1 × 1 µm), created within the analysis software, was placed on 10 successive cells that were in sharp focus using the first visibly loaded smooth muscle cell as a starting point.A Ca 2+ wave was defined as local fractional fluorescence ( F/F 0 ) increase above the noise level of 1.1, which spans the whole cell and lasts for at least 1 s.Ca 2+ waves were assessed by the number of firing cells in an array of the 10 adjacent cells and the frequency of Ca 2+ waves propagation per cell per minute.

Proximity ligation assay
To test the spatial proximity of Ca V 3.1 and IP 3 R, the Duolink in situ PLA detection kit was employed as previously described 19 .Briefly, freshly isolated mesenteric arterial smooth muscle cells underwent successive steps of fixation (10% paraformaldehyde in PBS, 15 min), permeabilization (0.2% Tween 20 in PBS, 15 min) and blocking (Duolink blocking solution, 1 h).Cells were then washed with PBS then incubated with primary antibodies (anti-Ca V 3.1, anti-IP 3 R1) in Duolink antibody diluent solution at 4 °C overnight.Cells were then incubated with secondary Duolink PLA PLUS and MINUS probes for 1 h at 37 °C.If target proteins are within 40 nm of each other, synthetic oligonucleotides attached to the probes hybridize enabling their subsequent amplification and binding to complementary fluorescent oligonucleotide sequences, detected using Leica-TCS SP8 confocal microscope.

Statistical analysis
Data are expressed as means ± SD, and n indicates the number of cells, arteries, or animals.Power analysis was performed a priori to assess the sample size sufficient for obtaining statistical significance.No more than 1 experiment was performed on cells/tissues from any given animal.Where appropriate, paired/unpaired t-tests or two-way analysis of variance (ANOVA) were performed to ascertain significant differences in mean values to a given condition/treatment.P values ≤ 0.05 were considered statistically significant.

Solutions and chemicals
Fluo-8 was acquired from Abcam.Primary antibodies against Ca V 3.1 and IP

Results
Characterization of Ca V 3.1 −/− genotype and phenotype Genetic deletion of Ca V 3.1 channels was confirmed by PCR, immunohistochemical analysis and conventional whole-cell patch-clamp electrophysiology.In detail, PCR amplification of C57BL/6 and Cav3.1 −/− mouse DNA with Ca V 3.1 primers resulted in different sized PCR products.C57BL/6 mice had a PCR product of 288 bp corresponding to the wild type allele, while the PCR product for Ca V 3.1 −/− mice was seen at 385 bp (Fig. 1Aa).
Figure 1Ab shows Ca V 3.1 protein expression is punctate in smooth muscle cells isolated from C57BL/6 but not Ca V 3.1 −/− mesenteric arteries (n = 4 mice per group).This analysis aligned with whole-cell electrophysiology, which noted dampened Ca V 3.1 activity in smooth muscle cells from Ca V 3.1 −/− mice relative to C57BL/6 controls (P = 0.013).Note, Ca V 3.1 activity was measured by first monitoring the total inward Ba 2+ current, the collective sum of Ca V 1.2, Ca V 3.1, and Ca V 3.2 currents 20 .Based on past studies, nifedipine and Ni 2+ were then applied to abolish Ca V 1.2 (L-type) and Ca V 3.2 (T-type) activities, respectively, and the residual current was then assigned to Ca V 3.1 21,22 .The current-voltage relationship of each Ca 2+ channel is illustrated in Fig. 1B,C (Ca V 3.1 current in green), with peak current (at + 10 mV) summarized in Fig. 1D.Recordings were attained from mesenteric smooth muscle cells (9 cells per group) isolated from 6 C57BL/6 and 8 Ca V 3.1 −/− mice.
Metabolic and blood pressure measurements in C57BL/6 and Cav3.1 −/− mice Metabolic caging assessed O 2 consumption, CO 2 production, the respiratory exchange rate, along with cumulative food and water consumed during normal activity.No significant difference was observed among C57BL/6 and Ca V 3.1 −/− mice (Table 1) in these parameters.However, Ca V 3.1 −/− mice displayed a disrupted night-time sleeping pattern and were modestly but significantly heavier than the C57BL/6 controls.Subsequent tail-cuff measurements revealed that systolic, diastolic, and consequently mean arterial pressures were reduced in Ca V 3.1 −/− mice compared to C57BL/6 mice (Fig. 1E).

Ca V 3.1 channels contribute to the myogenic response
Mesenteric arteries from C57BL/6 and Ca V 3.1 −/− mice were mounted in a myograph and exposed to increasing intraluminal pressures (20 to 100 mmHg) in physiological saline solutions, with Ca 2+ and Ca 2+ -free + 2 mM EGTA.Traces and summative data are presented in Fig. 2A-C, and findings reveal that Ca V 3.1 −/− arteries displayed reduced myogenic tone compared to C57BL/6 controls, a trend that was statistically significant at lower intraluminal pressures (20 mmHg: P = 0.002, 40 mmHg: P = 0.014, 60 mmHg: P = 0.008, 80 mmHg: P = 0.188, 100 mmHg: P = 0.108, unpaired t test).Arterial distensibility, a surrogate of vessel stiffness and defined as the percentage change in passive arteriolar diameter per change in intravascular pressure 23 , was comparable among the two www.nature.com/scientificreports/groups of arteries (Fig. 2D).Control experiments using PE as a vasoconstrictor noted a comparable vasomotor tone among C57BL/6 and Ca V 3.1 −/− arteries across a full concentration range (Fig. 3A,B).This statement applies equally to tone generated in the absence and presence of nifedipine, except at the higher agonist concentrations where the L-type Ca 2+ channel blocker initially appeared to be less impactful in Ca V 3.1 −/− arteries (Fig. 3A,B).Note, however, when this data was normalized to the % maximal constriction, the nifedipine-sensitive and insensitive components of agonist-induced constriction were comparable among the two groups of arteries (Fig. 3C).

Ca V 3.1 enable myogenic tone by facilitating Ca 2+ wave generation
To assess whether Ca 2+ flux through Ca V 3.1 triggers Ca 2+ wave generation, mesenteric arteries from C57BL/6 and Ca V 3.1 −/− mice were loaded with Fluo-8, and rapid Ca 2+ imaging was assessed by swept field confocal microscopy.Ca 2+ waves in C57BL/6 mice were readily observed in 80% of smooth muscle cells (8 of 10 per vessel) at a frequency of 9 waves/cell/min, each with a duration of 3-4 s (Fig. 4A,B).Similar to rat vessels, nifedipine application had little discernible effect on Ca 2+ wave generation 24 .The deletion of Ca V 3.1 markedly reduced the number of firing smooth muscle cells (P = 0.0002) along with firing frequency (P < 0.0001) by 55% and 65%, respectively (Fig. 4A,B); the Ca 2+ waves that remained were insensitive to nifedipine.Control experiments in C57BL/6 mesenteric arteries (Fig. 4C,D) subsequently confirmed that 2-APB, a blocker of IP 3 Rs, notably attenuated the number of firing cells (P < 0.0001) and Ca 2+ wave frequency (P = 0.0002) by 80% and 75%, respectively.Owing to the non-selective nature of 2-APB, and its reported inhibition of store-operated Ca 2+ entry, the previous control experiments were repeated in the presence of xestospongin C, a selective IP 3 R blocker.Similar to 2-APB, xestospongin C attenuated the number of firing cells (P = 0.011) and Ca 2+ wave frequency (P = 0.002) (Fig. 4E,F).With this functional evidence indicating that Ca 2+ flux through Ca V 3.1 triggers IP 3 Rs and the induction of Ca 2+ waves, the PLA was employed to assess whether these two proteins sat closely to one another.Consistent with Ca V 3.1 and IP 3 R1 residing within 40 nm of one another, we observed red punctate labelling in smooth muscle cells isolated from C57BL/6 but not Ca V 3.1 mesenteric arteries (Fig. 5).Controls were performed on cells treated with anti-Ca V 3.1, anti-IP 3 R1, or secondary antibodies alone, and revealed no evidence of nonspecific binding and false product amplification.Given the preceding observation, a final set of functional experiments were performed to address the contributory role of IP 3 R-dependent Ca 2+ waves to pressure-induced constriction.Using mesenteric arteries from C57BL/6 and Ca V 3.1 −/− mice, myogenic tone was examined over a full pressure range in the absence and presence of nifedipine (0.3 µM) ± 2-APB (50 µM).Of particular note, was the nifedipine-resistant tone that was present in C57BL/6 but not Ca V 3.1 −/− arteries, particularly at lower intravascular pressures (Fig. 6A,B,D,E).That tone per se was largely eliminated with the further application of 2-APB (20 mmHg: P = 0.013, 40 mmHg: P = 0.008) consistent with IP 3 Rs and Ca 2+ waves playing a role in its genesis (Fig. 6C,I).Note that IP 3 R inhibition in Ca V 3.1 −/− arteries had no discernible effect on nifedipine insensitive tone at any pressure (Fig. 6D-F).In a set of control experiments, xestospongin C (3 µM, a more selective IP 3 R antagonist) was used in place of 2-APB in C57BL/6 and it generated a contractile (Fig. 6G,H, 20 mmHg: P = 0.015, 40 mmHg: P = 0.001, 60 mmHg: P = 0.02, 80 mmHg: P = 0.013), and Ca 2+ wave (Fig. 4E,F) phenotype akin to Ca V 3.1 −/− arteries.In a final set of controls, myogenic tone and Ca 2+ waves were assessed in C57BL/6 vessels, with and without kurtoxin (Fig. 7) to block Ca V 3 channels.Kurtoxin reduced myogenic tone at 20 and 40 mmHg when place on top of nifedipine (20 mmHg: P = 0.04, 40 mmHg: P = 0.008).Also note, whereas nifedipine had no impact on Ca 2+ wave generation www.nature.com/scientificreports/(Fig. 4), kurtoxin + nifedipine had an inhibitory effect (Fig. 7: number of firing cells (P = 0.004) and firing frequency (P = 0.002)).

Discussion
Bayliss first described the intrinsic ability of resistance arteries to constrict to a rise in intravascular pressure 3 .This foundational response is now known to set basal tone in key organs and stabilizes organ perfusion as blood  pressure fluctuates.Further, this response has been intimately tied to arterial depolarization and the rise in [Ca 2+ ] i enabled by graded Ca 2+ entry principally through L-type Ca 2+ channels 9 .Vascular L-type Ca 2+ channels are encoded by the Ca V 1.2 α 1 pore-forming subunit whose steady-state voltage-dependent properties are shifted rightward to more depolarized potentials 25 .Recent work has revealed that L-type Ca 2+ channels are not alone in vascular smooth muscle and that T-type Ca 2+ channels are also expressed, with Ca V 3.1 being key to this investigation 9 .Its steady-state activation profile is hyperpolarized, and as such enables Ca 2+ entry when L-type Ca 2+ channels are deactivated.In theory, Ca 2+ entry via T-type Ca 2+ channels could impact tone development by directly contributing to the cytosolic Ca 2+ pool or by acting locally and indirectly to trigger Ca 2+ waves.Ca 2+ waves are slow asynchronous events that spread from end to end and whose triggering is tied to the opening of sarcoplasmic reticulum IP 3 Rs by IP 3 and Ca 2+24 .Using a Ca V 3.1 −/− model, we tested whether Ca 2+ entry through this particular T-type channel does indeed facilitate myogenic tone at hyperpolarized voltages and if this functional response is coupled to the governance of Ca 2+ waves.
Our examination of Ca V 3.1 began with experiments to confirm the absence of Ca V 3.1 in mesenteric arterial smooth muscle cells from genetic deletion mice (Fig. 1).Three approaches were used, the first being PCR which confirmed Cacna1g gene (Ca V 3.1) modification in Ca V 3.1 −/− mice.Second, protein analysis using immunohistochemistry showed that surface expression of Ca V 3.1 was notably lacking in Ca V 3.1 −/− but not C57BL/6 cells.These observations aligned with the results from the third, functional approach (whole-cell electrophysiology), which revealed that the nifedipine/Ni 2+ resistant Ba 2+ current, previously ascribed to Ca V 3.1 21,22 was also absent in mesenteric arterial smooth muscle cells harvested from genetic deletion mice.The absence of this T-type Ca 2+ channel coincided with a reduction in systolic and diastolic blood pressure, a finding consistent with a role in hemodynamic control.Past observations are limited, with one study reporting no difference in blood pressure, although values were unrealistically low for both Ca V 3.1 −/− and C57BL/6 mice 26 .A second showed blood pressure trending lower in Ca V 3.1 −/− mice, along with a more substantive reduction in blood pressure variability 27 .While the mechanism driving the blood pressure change is unclear, it's reasonable to assert a role for diminished myogenic tone, an idea we tested in isolated mesenteric arteries across a full pressure range.Consistent with expectations, a reduction in myogenic tone was observed in Ca V 3.1 −/− arteries, particularly at lower pressure (Fig. 2) when vessels are hyperpolarized and T-type Ca 2+ channels more active in the steady state 28 .In considering these observations, prudent controls are key, the first being an assessment of an artery's passive structural properties.In this regard, we observed no change in the arterial distensibility in vessels harvested from Ca V 3.1 −/− or C57BL/6 mice.Likewise, in a second set of controls, this study did not observe a change in arterial contractility to PE across a full concentration range in the absence or presence of nifedipine, an L-type Ca 2+ channel blocker (Fig. 3).These results confirm that the molecular machinery mediating PE-induced constriction remains intact in Ca V 3.1 −/− animals, as does the signalling pathways downstream from the α 1 -adrenoreceptor.Past studies have performed similar agonist controls and findings have been somewhat conflicted, with Ca V 3.1 deletion notably reducing mesenteric arterial responsiveness in one study 29 , yet having the markedly opposite effect in another, presumptively increasing the Ca 2+ sensitivity of the contractile apparatus 20 .
In contextualizing the preceding observations, one should recognize past inferential work linking T-type Ca 2+ channels to myogenic tone using pharmacology with known off-target effects.This approach typically entailed the probing of myogenic tone at rest and in the presence of an L-type Ca 2+ channel blocker to isolate residual tone whose sensitivity to T-type Ca 2+ channel inhibition was then tested 12,30,31 .One should also consider www.nature.com/scientificreports/past work with Ca V 3.1 −/− mice 16 highlighting a role for Ca V 3.1 channels in tone development (at low pressure), although without defining mechanism 20 .Finally, in using this genetic deletion model, acknowledgement of other cardiovascular effects is key, in particular bradycardia 26 and impaired blood pressure regulation through impaired NO formation 32 .Ca 2+ waves are slow-spreading, end-to-end events initiated by a stimulus that drives the release of Ca 2+ from the sarcoplasmic reticulum 33,34 .The initiation and spread of these asynchronous events are tied to IP 3 R, Ca 2+ -permeable channels whose activation depends on IP 3 and Ca 2+ binding to cytosolic sites 25 .Past work in rat cerebral arteries has shown that Ca 2+ waves are present at low intravascular pressure and that frequency rises as pressure is elevated into the lower physiological range 24 .Pharmacological attenuation of Ca 2+ waves through IP 3 R blockade or impairment of store refilling results in diminished pressure-induced constriction particularly at low intravascular pressure when arteries are more hyperpolarized 24 .Respectful of these results, it follows that low threshold Ca V 3.1 channels provide the Ca 2+ needed to trigger Ca 2+ waves and foster myogenic tone when L-type Ca 2+ channels are decidedly less active.This concept was tested three ways, the first examining Ca 2+ wave generation in Ca V 3.1 −/− arteries, the second ascertaining if Ca V 3.1 colocalized with IP 3 Rs, and the final determining if Ca 2+ wave inhibition in C57BL/6 mice results in a Ca V 3.1 −/− contractile phenotype.Findings in Fig. 4 first reveal that Ca 2+ wave generation is robust in control mesenteric arteries as defined by the number of firing cells and the rate of Ca 2+ waves per firing cell.Analogous to past work in rat cerebral arteries, nifedipine didn't impact Ca 2+ wave generation, consistent with L-type Ca 2+ channels playing little role in initiating or maintaining these events 24 .Ca 2+ waves were significantly reduced in Ca V 3.1 −/− arteries and abolished in control arteries by 2-APB, and xestospongin C, IP 3 R inhibitors, findings consistent with this T-type Ca 2+ channel driving sarcoplasmic reticulum dependent events.These intriguing findings aligned with results from the PLA that note a close spatial association between Ca V 3.1 and IP 3 R.In detail, this assay involves the binding of primary antibodies to two target proteins and then uses secondary antibodies with conjugated DNA strands which form a circular DNA template for amplification if proteins are < 40 nm apart 10 .The amplified product, detected as bright red puncta, is clearly visible in Fig. 5, thus, it is logical to conclude that Ca 2+ flux via Ca V 3.1 should be sufficient to open IP 3 Rs.In light of both results, final experiments assessed whether reduced Ca 2+ wave production in C57BL/6 vessels generate a functional phenotype akin to Ca V 3.1 −/− arteries.In this regard, we monitored myogenic tone in mesenteric arteries (as a percentage; C57BL/6 and Ca V 3.1 −/− ) at rest and following treatment with nifedipine alone or with 2-APB or xestospongin C (Fig. 6).We observed residual tone in nifedipine-treated C57BL/6 arteries but not Ca V 3.1 −/− arteries, a difference that could be abolished, particularly at low intravascular pressures (20-40 mmHg) through IP 3 R blockade.This loss of tone parallels a similar loss in tone, and likewise Ca 2+ waves in C57BL/6 mice when kurtoxin, a Ca V 3.x blocker was applied on top of nifedipine, an L-type Ca 2+ channel blocker (Fig. 7).While some caution is warranted when drawing a relationship between Ca 2+ waves and myogenic tone, the preceding interpretation does align with other published observations.They include: (1) 2-APB treatment having no impact on global [Ca 2+ ] while reducing myogenic tone 35 ; (2) 2-APB only dilating arteries which prior to treatment were generating Ca 2+ waves 36 ; and (3) Ca 2+ waves abrogation corelating with reduced myosin light chain phosphorylation particularly at lower pressure 24 .
Two final points in this study require further consideration.First, while differences in myogenic tone between Ca V 3.1 −/− and C57BL/6 arteries were evident at lower pressures (20-60 mmHg), the same trend was present at higher pressures, although without statistical significance.This finding is perhaps unsurprising as L-type Ca 2+ channels rise to dominate [Ca 2+ ] i as arteries depolarize with pressurization.Second, while our work noted Ca 2+ wave insensitivity to L-type Ca 2+ channel blockade, like the cerebral vasculature 24 , it lies in contrast to cremaster arterioles where nifedipine attenuated Ca 2+ wave formation 37 .This discrepancy suggests there may be mechanistic uniqueness among vascular beds, which to date is unappreciated.Alternatively, one could potentially argue the higher concentration of nifedipine (1 μM) used on cremaster arteries may be blocking Ca V 3.1 and consequently the triggering of IP 3 R [38][39][40] .This perspective is consistent with electrophysiology observations noting that T-type Ca 2+ channels are partially blocked by low micromolar nifedipine 41,42 .

Conclusion
This study presents three key findings, summarized in Fig. 8: First, Ca V 3.1 −/− mice have lower blood pressure, and mesenteric arteries display diminished myogenic tone compared to controls.Second, immunohistochemical analysis reveals that Ca V 3.1 lies within 40 nm of IP 3 R1, and when this arrangement is genetically disrupted, arteries generate fewer Ca 2+ waves.Third, a pharmacological blockade of IP 3 Rs in C57BL/6 arteries produces a phenotype similar to Ca V 3.1 −/− vessels, that being diminished myogenic tone at lower intravascular pressure.By establishing a clear sequential relationship between Ca V 3.1, Ca 2+ waves and myogenic tone, this study advances the understanding of vascular contractility and highlights a new target for therapeutic control.In this regard, one could provocatively suggest that development of selective Ca V 3.1 blockers could be of value in the management of hypertension.

3 Figure 1 .
Figure 1.Absence of Ca V 3.1 expression and current in SMCs isolated from mesenteric arteries of Ca V 3.1 −/− mice, and lower arterial blood pressure indices in Ca V 3.1 −/− mice compared to C57BL/6.(Aa) Polymerase chain reaction of Cacna1g gene (Ca V 3.1).DNA was extracted from ear notches (C57BL/6 and Ca V 3.1 −/− mice) and amplified; the different product sizes confirm the gene modification leading to functional knockout.Illustration created with BioRender.com.(Ab) Ca V 3.1 (green) in cerebral arterial myocytes from control mice with nuclei stained with DAPI (blue) detected with immunohistochemistry.This signal was absent in Ca V 3.1 −/− mice.Secondary antibody controls were negative for nonselective labelling (n = 4 cells pooled from 4 animals/group).(B) Averaged Ca V currents were assessed by whole-cell patch clamp in C57BL/6 cells showing a residual current remaining (highlighted green) after blocking L-type and Ca V 3.2 currents by nifedipine and Ni 2+ , respectively.(C) Recordings of whole-cell Ca V currents in Ca V 3.1 −/− cells showing no residual current after nifedipine and Ni 2+ treatment.(D) Peak current (I) plots of whole-cell Ba 2+ (10 mmol/L) current before and after the application of nifedipine to C57BL/6 and Ca V 3.1 −/− smooth muscle cells.n = 9 SMCs from 6 mice in control group and n = 9 SMCs from 8 mice in knockout group.(* P = 0.013, unpaired t test).(E) Systolic, diastolic, and mean arterial pressure (mmHg) of Ca V 3.1 −/− and C57BL/6 mice were measured using the CODA6 tail-cuff system.25-min recordings daily for one week were performed on both groups (n = 5).(Systolic: *P = 0.028, Diastolic: **P = 0.005, MAP: **P = 0.008, unpaired t test).

Figure 3 .
Figure 3. Ca V 3.1 deletion has no impact on phenylephrine-induced constriction.Increasing concentrations of phenylephrine were applied onto pressurized arteries isolated from C57BL/6 and Ca V 3.1 −/− mice in the presence and absence of nifedipine (L-type Ca 2+ channel blocker).Experiments were conducted at an intraluminal pressure of 60 mmHg.(A,B) Representative traces (Left) and summary data (Right) of changes in diameter in response to phenylephrine showing a decrease in constriction in nifedipine-treated vessels from both strains.(C) %Maximal phenylephrine-induced constriction relative to KCl-induced constriction shows no significant difference in agonist-induced constriction between C57BL/6 and Ca V 3.1 −/− mice.(n = 6 arteries from 6 animals).P values for increasing PE concentrations in PSS: 0.529, 0.790, 0.763, 0.957, 0.554, 0.719 and in PSS + nifedipine: 0.565, 0.343, 0.074, 0.396, 0.925, 0.837 (Paired t test).

Figure 5 .
Figure 5. Ca V 3.1 channels colocalize with IP 3 Rs.Proximity ligation assay was employed using isolated mesenteric arterial SMCs from C57BL/6 (n = 35 cells from 6 animals) and Ca V 3.1 −/− (n = 36 cells from 6 animals) mice to determine the close association (< 40 nm) of Ca V 3.1 and IP 3 R1 proteins (red, denoted by white arrows).Nuclei were stained with DAPI (Blue).Control experiments used only one primary antibody or no primary antibody.Note, dots were averaged across cells within each animal, and represented as a data point in the bar graph to facilitate statistical comparison.**P = 0.0033.

Figure 8 .
Figure 8. High-voltage-activated Ca V 1.2 channels control [Ca 2+] i when intravascular pressure is elevated and membrane potential depolarized.In contrast, Ca V 3.1 channels display a hyperpolarized profile with more negative activation/inactivation properties compared to Cav1.2 channels.Ca V 3.1 channels foster Ca 2+ wave generation likely through sarcoplasmic reticulum IP 3 R activation as the two proteins lie in close proximity.Ca 2+ waves are known to induce a Ca 2+ -calmodulin (CAM)-dependent activation of myosin light chain kinase (MLCK) which regulates myosin phosphorylation leading to myogenic tone control.Ca V 3.1 deletion is coupled to reduced blood pressure and hemodynamic control thus bearing clinical importance.Created with BioRender.com.
3 R1 were purchased from NovusBio and Alomone Laboratories, respectively.PCR kit was obtained from Qiagen.Secondary antibody, Alexa Fluor 488 Donkey Anti-Rabbit IgG (H+L), and 2-APB were obtained from ThermoFisher.Duolink PLA detection kits, nifedipine, PE hydrochloride, DAPI, donkey serum kurtoxin, and all other chemicals were obtained from Sigma-Aldrich unless stated otherwise.In cases where DMSO was used as a solvent, the maximal DMSO concentration after application did not exceed 0.5%.Please see the Major Resources Table in the Supplemental Materials.

Table 1 .
There are no discernible metabolic differences among strains, except sleep time and weight.C57BL/6 and Ca V 3.1 −/− mice were placed into individual CLAMS metabolic chambers for 48 h.Metabolic parameters including VO 2 , VCO 2 , respiratory exchange rate (RER), food and water consumption, and sleep times, in both light and dark conditions were measured (n = 6 for each group).Data were presented as means ± SE and compared using unpaired two-tailed t-tests.Numbers in red indicate a significant difference of P < 0.05.