Reactive Oxygen Species Mediate the Suppression of Arterial Smooth Muscle T-type Ca2+ Channels by Angiotensin II

Vascular T-type Ca2+ channels (CaV3.1 and CaV3.2) play a key role in arterial tone development. This study investigated whether this conductance is a regulatory target of angiotensin II (Ang II), a vasoactive peptide that circulates and which is locally produced within the arterial wall. Patch clamp electrophysiology performed on rat cerebral arterial smooth muscle cells reveals that Ang II (100 nM) inhibited T-type currents through AT1 receptor activation. Blocking protein kinase C failed to eliminate channel suppression, a finding consistent with unique signaling proteins enabling this response. In this regard, inhibiting NADPH oxidase (Nox) with apocynin or ML171 (Nox1 selective) abolished channel suppression highlighting a role for reactive oxygen species (ROS). In the presence of Ni2+ (50 µM), Ang II failed to modulate the residual T-type current, an observation consistent with this peptide targeting CaV3.2. Selective channel suppression by Ang II impaired the ability of CaV3.2 to alter spontaneous transient outward currents or vessel diameter. Proximity ligation assay confirmed Nox1 colocalization with CaV3.2. In closing, Ang II targets CaV3.2 channels via a signaling pathway involving Nox1 and the generation of ROS. This unique regulatory mechanism alters BKCa mediated feedback giving rise to a “constrictive” phenotype often observed with cerebrovascular disease.


Results
Whole cell patch clamp recordings confirmed targeting of T-type Ca 2+ channels by Ang II, a vasoconstrictive peptide present in cerebral circulation 22 . Ang II (100 nM) decreased peak inward Ba 2+ current in a time-dependent manner (−1.4 ± 0.1 pA/pF to −1.2 ± 0.05 and 0.9 ± 0.05 pA/pF at 5 and 10 min, respectively) ( Fig. 1). Additionally, Ang II drove a rightward shift trend in the voltage dependence of steady state activation (V 50 activation, from −23.1 ± 1.7 mV to −15 ± −1.2 mV) but not steady state inactivation (V 50 inactivation, from −42.1 ± 0.9 mV to −41 ± −0.9 mV). The ability of Ang II to suppress T-type activity was mediated via AT 1 receptors as preincubation with 1 µM Losartan abolished all inhibitory effects (−1.47 ± 0.08 pA/pF vs −1.44 ± 0.08 pA/pF). Losartan alone had no effect on peak inward current or voltage dependence of activation/inactivation. Note, Ba 2+ was the preferred charge carrier, as its elevated concentration promotes current flow without facilitating isolated smooth muscle cell contraction 12 .
AT 1 receptors are coupled to the classic Gq pathway which mobilizes phospholipase C, fostering diacylglycerol production and the activation of PKC 17 . Figure 2 demonstrates the conventional PKC blocker Go 6976 (100 nM) had no measurable effect on basal Ba 2+ current or the ability of Ang II to suppress Ca V 3.x channels (from −1.524 ± 0.0559 pA/pF to 1.0914 ± 0.0442 pA/pF). Similar results were obtained using the broad-spectrum PKC inhibitor GF 109203 × (100 nM) (from −1.46 ± 0.06 pA/pF to −0.98 ± 0.033 pA/pF) (Fig. 2). Neither blockers were associated with significant change in the voltage dependence of steady state activation/inactivation (Supplementary Figure I) We subsequently addressed the involvement of Nox, a ROS generating enzyme recently Peak T-type current (pA/pF) at 5 min intervals under control condition and following treatment with losartan ± Ang II (n = 8). All recordings were performed in the presence of 200 nM nifedipine and using Ba 2+ as a charge carrier. *Denotes significant difference compared to control (*P < 0.05, paired t test). tied to the AT 1 receptor in the resistance vasculature 15 . Apocynin (50 µM), a general Nox inhibitor, had little impact on basal currents or kinetics (Fig. 3). It did, however, abolish Ang II suppression (−1.54 ± 0.08 pA/pF compared to −1.58 ± 0.061 pA/pF), highlighting a role for ROS generation. While vascular smooth muscle cells express several Nox isoforms, subtype 1 has been previously implicated in mediating Ang II effects 23,24 . Consistent with this perspective, selective inhibition of Nox 1 (ML171, 1 µM) eliminated Ang II suppression of the inward T-type current (−1.7 ± 0.07 pA/pF versus −1.75 ± 0.066 pA/pF) (Fig. 3). Similar to apocynin, ML171 had little effect on the basal current or the voltage dependence of steady state activation/inactivation (Supplementary Figure II). (C) Representative trace, averaged I-V plots of the whole cell T-type current, prior to and following the application of GF 109203X (non-selective PKC inhibitor, 100 nM) (n = 9). (D) Representative trace and average I-V plot highlighting the inability of GF 109203X to abolish Ang II (100 nM) mediated inhibition of the T-type current (n = 8). (E) Peak T-type current (pA/pF) at 5 min intervals under control condition (n = 8) and in cells treated with either Go 6976 or GF 109203X ± Ang II (100 nM) (n = 8-9). *Denotes significant difference compared to control (*P < 0.05, paired t test). The T-type current in cerebral arterial smooth muscle reflects the composite activity of Ca V 3.1 and Ca V 3.2 channels 12 . These two individual currents can be differentiated from one another by applying low micromolar Ni 2+ , a divalent cation that selectively blocks Ca V 3.2 8,25 . Figure 4 demonstrates that when Ni 2+ is placed in the bath, Ang II is unable to suppress the remaining current, representative of Ca V 3.1 channels. (−0.68 ± 0.05 pA/pF compared to −0.65 ± 0.04 pA/pF). Likewise, Ni 2+ failed to augment T-current suppression in cells pretreated with Ang II compared to the effect of the vasoactive peptide alone (57.6 ± 5.84% versus 55.7 ± 6.25%, respectively). Selective targeting of Ca V 3.2 was further confirmed by monitoring STOC production and noting the inability of Ni 2+ to inhibit STOC frequency in isolated cells pretreated with Ang II (1.1 ± 0.12 Hz to 0.975 ± 0.19 Hz compared to 1.29 ± 0.095 Hz to 0.53 ± 0.092 Hz under control conditions) (Fig. 5). Functional data aligned with electrophysiological observations in that Ni 2+ application failed to increase myogenic tone (4% change) in arteries pretreated with Ang II compared to the 25% rise under standard conditions. Intriguingly, the Ni 2+ mediated increase in myogenic tone was restored in arteries that were pretreated with apocynin, a finding consistent with ROS modulating Ca V 3.2 (Fig. 5). Similar results were obtained with the Nox1 inhibitor ML171 (Supplementary Figure III). Pretreating arteries with the Ni 2+ did not alter their responsiveness to Ang II (Supplementary Figure IV). Such observations structurally imply that the ROS generating enzyme Nox 1 colocalizes with Ca V 3.2. This was confirmed with the proximity ligation assay (Fig. 6); punctate red fluorescent product was observed in cells treated with both primary antibodies (anti Ca V 3.2 and anti Nox1) in keeping with proteins residing within 40 nm of one another. Fluorescent product was absent in control experiments where one or both primary antibodies were omitted. Further control experiments reveal no similar co-localization pattern between Nox1 and Ca V 3.1, a secondary T-type Ca 2+ channel expressed in cerebral arteries (Fig. 6).

Discussion
This study examined whether and by what mechanism Ang II modulates T-type Ca 2+ channels in cerebral arterial smooth muscle cells. This constrictor peptide was initially observed to suppress T-type channels via activation of AT 1 receptors and a signaling pathway independent of PKC. The inhibition of Nox, a unique signaling protein involved in local ROS generation, abolished Ang II suppression. Further experiments revealed that Ang II specifically targets Ca V 3.2 channels via Nox 1, two proteins that localize within 40 nm of one another. Functionally, Ang II mediated T-channel suppression will impair STOC production at physiological voltages, reducing negative feedback and facilitating arterial constriction. This study is the first in vascular tissue to consider T-type Ca 2+ channels as a regulatory target of Ang II and downstream Nox signaling in the genesis of arterial constriction (Fig. 7). Background. Cerebral arterial tone is actively controlled by vasoactive stimuli present in the systemic circulation or locally produced within tissue 1,2 . Dynamic tuning is intimately tied to membrane potential and the gating of L-type Ca 2+ channels which sets cytosolic [Ca 2+ ] and the activity of myosin light chain kinase 26,27 . In addition to L-type channels, two T-type channels (Ca V 3.1 and Ca V 3.2) are expressed in cerebral arterial smooth muscle, each of which subtly contributes to tone development in a mechanistically unique manner 6,28 . Of particular note is Ca V 3.2, a channel whose opening paradoxically limits arterial constriction by discretely triggering Ca 2+ release from ryanodine receptors (RyR) localized to the sarcoplasmic reticulum 8 . These transient events activate BK Ca channels, driving STOC production, hyperpolarizing events which attenuate arterial constriction 29 . In considering the close and evolving relationship between Ca V 3.2 and STOC production, mechanistic questions logically arise as to the dynamic regulation of this feedback cascade. Vasoconstrictors for example, including those that circulate systemically or produced locally in the arterial wall (Ang II) suppress this cascade and facilitate arterial tone generation 19,30 . This suppression is often attributed to a direct BK Ca suppression with little consideration that the upstream initiators of Ca 2+ sparks may themselves be the target of inhibition 11,31 . It is this issue that will be addressed in the current manuscript. Initial Experimentation. Work began by examining the effect of Ang II on Ca V 3.x channels expressed in cerebral arterial smooth muscle cells 6 . Patch clamp electrophysiology revealed that this vasoactive peptide suppressed the T-type current in a time dependent manner, and initiated a rightward shift trend in the voltage dependency of activation (Fig. 1). The contractile effects of Ang II are typically mediated through AT 1 and AT 2 receptors, with the former expressed in smooth muscle and tied to constriction, and the latter present in endothelium and intimately tied to nitric oxide production 32,33 . Consistent with the involvement of AT 1 receptors, Losartan, a known anti-hypertensive agent, abolished Ang II suppression of the T-type current. As AT 1 receptors are G q coupled, it is logical to deduce that Ang II suppression is linked to phospholipase C-β and the mobilization of PKC, a phosphoprotein known to target multiple ion channels 17,34,35 . Surprisingly, both conventional and broad spectrum PKC inhibitor (Go 6976 and GF 109203X, respectively), previously shown to interfere with phorbol ester induced constriction, failed to abolish T-type current suppression (Fig. 2) 36 . These observations suggest that AT 1 receptor signaling is diverse in cerebral arterial smooth muscle and that other downstream effectors are uniquely involved in targeting the T-type current.

Involvement of Reactive Oxygen Species.
Ang II application in the cerebral circulation elevates ROS production through the stimulation of the AT 1 receptor 15,37 . While several ROS generating mechanisms are present in this vascular bed, Nox is of particular importance due to its presumed coupling with the AT 1 receptor 15,38 and its ability to regulate Ca 2+ conductances including L-type Ca 2+ channels 39 . Consistent with ROS generation impacting a broader range of Ca 2+ permeable pores, pretreating cells with the pan selective Nox inhibitor apocynin abolished the suppressive effects of Ang II on T-type channels (Fig. 3). While impactful on Ang II mediated effects, apocynin had no measurable impact on peak inward current or activation/inactivation kinetics under basal conditions. Of the seven known Nox isoforms, three have been identified at the molecular level in cerebral arteries, two of which (Nox1 and Nox4) are expressed in vascular smooth muscle 40,41 . Observations from knockout and transgenic mice suggest that Nox1 is a key downstream target of the AT 1 receptor and important to mediating ROS generation in vasculature 23,42,43 . In alignment with this observation, the Nox1 selective inhibitor, ML171 attenuated the ability of Ang II to suppress the T-type current in cerebral arterial smooth muscle (Fig. 3).
The T type current in vascular smooth muscle is representative of both Ca V 3.1 and Ca V 3.2 and in theory Ang II could be targeting one or both of the channel subtypes 6 . The two Ca V 3.x subtypes can be differentiated from one another using Ni 2+ , a divalent cation when applied at low micromolar concentrations will selectively blocks Ca V 3.2 12,25 . In the presence of 50 µM Ni 2+ the addition of Ang II failed to modulate the remaining T-type current, representative of Ca V 3.1 (Fig. 4). Ni 2+ had no inhibitory effect in smooth muscle cells pretreated with Ang II and together both findings suggest that AT 1 receptors and the downstream generation of ROS are selectively targeting Ca V 3.2 channels. While the precise mechanism of inhibition remains unclear, past work in neuronal tissue has noted that oxidation of histidine residue H191 potently inhibits Ca V 3.2 channels 44 . The selective targeting of Ca V 3.2 has potentially important implications in regard to resident feedback loops present in cerebral arterial smooth muscle cells. In detail, Ca V 3.2 has been reported to trigger the cytosolic gate of RyR, initiating Ca 2+ sparks that sequentially activate BK Ca channels, generating a hyperpolarization that moderates arterial constriction 8,21 . It follows that if Ang II is indeed selectively targeting Ca V 3.2 that the channel's ability to generate STOCs should be moderated and subject to negative feedback in presence of the peptide. Consistent with this perspective, Ni 2+ failed to reduce STOC frequency in cells, or enhance arterial tone in vessels, superfused with Ang II (Fig. 5). Intriguingly, Ni 2+ mediated increases in arterial tone could be restored if vessels were pretreated with apocynin or ML171, prior to Ang II addition. This particular finding further confirms a role for ROS and to our knowledge the first to demonstrate selective modulation of Ca V 3.2 in the cerebral vasculature.
As ROS molecules are remarkably labile, targeted redox signaling can only occur if a defined enzyme generating system is compartmentalized in close apposition to the selected protein 45 . We consequently presumed, given the preceding observations, that Nox1 and Ca V 3.2 colocalize with one another, perhaps in caveolae where both proteins have been reported to reside 8,46 . One means of detecting close interaction is to employ the proximity ligation assay, a technique where two secondary antibodies with attached DNA strands bound to two distinct primary antibodies form a circular template for DNA amplification if the targeted proteins are within 40 nM of one another 8 . Consistent with close apposition among Ca V 3.2 and Nox1, punctate red florescence label was detected on the smooth muscle cell membrane (Fig. 6). Fluorescence labeling was absent in experiments where one or both primary antibodies were removed from the biochemical process. This localization pattern was unique to Ca V 3.2 and didn't extend to Ca V 3.1 (Fig. 6), a result which highlights regulatory distinctiveness among T-type Ca 2+ channels.
Broader Implications. Tone development is intimately tied to changes in arterial membrane potential and the subsequent rise of intracellular [Ca 2+ ] via L-type Ca 2+ channels 26 . Membrane potential is in turn driven by a range of conductances, several of which are assigned the role of feeding back negatively upon depolarization, to prevent vessels from overly constricting 10,11,47 . As noted above, BK Ca is a key feedback element and a channel whose activity is regulated indirectly by Ca V 3.2 and a discrete influx of Ca 2+ that triggers the cytosolic gate of RyR 8 . It is clear that Ang II suppresses Ca V 3.2 activity and with it the resultant feedback loop, an observation with implications to vascular disease. High renin models of hypertension are typically associated with both an increase in arterial tone and a rise in ROS production [48][49][50] . It is intriguing to consider whether ROS generation partially drives the rise in tone through an axis sequentially entailing the AT 1 receptor, Nox1, Ca V 3.2, Ca 2+ sparks and the transiently opening BK Ca . In considering this potential mechanism, it's important to acknowledge that Ca V 3.2 is not the sole conductance that triggers the opening of RyR and the generation of Ca 2+ sparks 21 . There is a complementary role for L-type Ca 2+ channels and the Na + /Ca 2+ exchanger particularly at depolarized potentials whose resident activity may also be impacted by the hypertensive state 21 . Summary. Findings in this study support the emerging view that T-type Ca 2+ channels are a rich regulatory target for vasoactive stimuli. In addition to being modulated by dilatory protein kinases, the current study shows that particular Ca V 3.x channels are targeted by vasoconstrictors and their associated receptors/signaling pathway. This includes Nox mediated generation of ROS via the AT 1 receptor, a process that selectively suppresses Ca V 3.2 and along with it BK Ca mediated feedback in the cerebral vasculature (Fig. 7). These findings elucidate a new mechanism of arterial tone regulation, one that could contribute to the excessive constricted state observed with advancing vascular disease.

Animal procedures. Animal procedures and methods were approved by the Animal Care and Use
Committees at the University of Calgary and the University of Western Ontario and in accordance with the Canadian Institute of Health Research (CIHR) guidelines. Female Sprague-Dawley rats (10-12 weeks of age) were euthanized by CO 2 asphyxiation. The brain was carefully removed and placed in cold phosphate-buffered saline (PBS, pH 7.4) solution containing (in mM): 138 NaCl, 3 KCl, 10 Na 2 HPO 4 , 2 NaH 2 PO 4 , 5 glucose, 0.1 CaCl 2 and 0.1 MgSO 4 . Middle and posterior cerebral arteries were isolated and cut into 2-3 mm segments.

Isolation of Cerebral Arterial Smooth Muscle Cells.
Middle and posterior cerebral arteries were enzymatically digested to liberate smooth muscle cells 10 . Briefly, arterial segments were placed in an isolation medium (37 °C, 10 minutes) 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 (pH 7.4). Vessels were then exposed to a two-step digestion process that involved: 1) 14 minutes incubation in isolation medium (37 °C) containing 0.5 mg/ml papain and 1.5 mg/ml dithioerythritol; and 2) 10 minutes incubation in isolation medium containing 100 μM Ca 2+ , 0.7 mg/ml type F collagenase and 0.4 mg/ml type H collagenase. 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 within ~5 hours.
Electrophysiological Recordings. Whole cell currents were measured in isolated cerebral arterial smooth muscle cells. Current was recorded using an Axopatch 200B patch-clamp amplifier (Molecular Devices, Sunnyvale, CA), filtered at 1 kHz, digitized at 5 kHz, and were stored on a computer for offline analysis with Clampfit 10.3 software (Molecular Devices, Sunnyvale, CA). Whole-cell capacitance averaged 12-18 pF and was measured with the cancellation circuity in the voltage clamp amplifier. Access resistance was monitored every 60 s and cells with changes greater than 2 MΩ were excluded. To minimize offset potential (<2 mV), a 1 M NaCl-agar salt bridge between the reference electrode and the bath solution was used. All experiments were performed at room temperature (~22 °C).
For measuring Ca V 3.x current, conventional patch-clamp electrophysiology was utilized. Briefly, recording electrodes (5-8 MΩ) were pulled from borosilicate glass microcapillary tubes (Sutter Instruments, Novato, CA) using a micropipette puller (Narishige PP-830, Tokyo, Japan), and backfilled with pipette solution containing (in mM): 135 CsCl, 5 Mg-ATP, 10 HEPES, and 10 EGTA (pH 7.2). Cells were placed in a bath solution consisting of (in mM): 110 NaCl, 1 CsCl, 10 BaCl 2 , 1.2 MgCl 2 , 10 glucose, and 10 HEPES (pH 7.4). For whole cell current recordings, smooth muscle cells were voltage clamped at a holding potential of −60 mV and subjected to a −90 mV followed by a 10 voltage step (300 ms) starting from −50 to 40 mV. I-V relationships were plotted as current density (pA/pF) at the different voltage steps. Voltage dependence of steady-state inactivation was assessed by a step protocol: (a) prepulse to −90 mV (300 ms); (b) stepping from −70 to 20 mV (10 mV interval, 1.5 s each); (c) hyperpolarizing back to −90 mV (10 ms); and (d) stepping to a test voltage 10 mV (200 ms). Whole cell current elicited by the test voltage was normalized to maximal current to plot % I/I max versus different voltage steps. The voltage dependence of activation was evaluated by monitoring isochronal tail currents. More specifically, cells clamped at −60 mV were subjected to a prepulse (−90 mV; 300 ms) followed by voltage steps ranging from −80 to 40 mV (10 mV interval, 50 ms each), and then a final hyperpolarizing test pulse (−90 mV; 200 ms) to evoke tail currents. Normalized tail currents (% I/I max ) were plotted versus the voltage steps. All recordings were performed in the presence of 200 nM nifedipine, a concentration previously shown to fully block Ca V 1.2 channels without affecting Ca V 3.x isoforms 12 .
Vessel Myography. Arterial segments were mounted in a customized arteriograph and superfused with warm (37 °C) physiological salt solution (PSS; pH 7.4; 21% O 2 , 5% CO 2 , balance N 2 ) containing (in mM): 119 NaCl, 4.7 KCl, 20 NaHCO 3 , 1.1 KH 2 PO 4 , 1.2 MgSO 4 , 1.6 CaCl 2 and 10 glucose. To limit the influence of the endothelium, air bubbles were passed through the vessel lumen (1 min); successful removal was confirmed by the loss of bradykinin-induced dilation. Arteries were equilibrated at 15 mmHg and contractile responsiveness assessed by brief application of 60 mM KCl. Following equilibration, intravascular pressure was elevated from 20 to 60 mmHg, and arterial external diameter was monitored under control conditions and in the presence of Ang II and/or Ni 2+ (Ca V 3.2 blocker). Maximal arterial diameter was subsequently assessed in Ca 2+ -free PSS (zero externally added Ca 2+ plus 2 mM EGTA). An additional set of experiments was conducted on vessels preincubated with apocynin (NADPH oxidase inhibitor) using the same experimental protocol. Percent myogenic tone was calculated as follows: % myogenic tone = 100 * (D 0 − D)/D 0 ; where D is external diameter under control conditions (Ca 2+ PSS) or treated conditions, and D 0 is external diameter in Ca 2+ -free PSS. Proximity Ligation Assay (PLA). The Duolink in situ PLA detection kit was employed using freshly isolated smooth muscle cells. Briefly, cells were first fixed in PBS containing 4% paraformaldehyde (15 min), and then incubated with PBS containing 0.2% tween (15 min) for permeabilization. Cells were washed with PBS, blocked by Duolink blocking solution (1 hr), and incubated overnight with primary antibodies (mouse anti-Ca V 3.2 or mouse anti Ca V 3.1 and rabbit anti Nox1) in Duolink antibody diluent solution at 4 °C. Control experiments employed no primary antibody or one primary antibody. Cells were then washed with Duolink washing solution, and labelled with Duolink PLA PLUS and MINUS probes for 1 hr (37 °C). The secondary antibodies of PLA PLUS and MINUS probes are attached to synthetic oligonucleotides that hybridize if present in close proximity (<40 nm). The hybridized oligonucleotides are then ligated and subjected to amplification. The amplified products extending from the oligonucleotide arm of the PLA probe were detected using far red fluorescent fluorophore-tagged, complementary oligonucleotide sequences and Leica TCS SP8 confocal microscope.
Chemicals and Statistical Analysis. GF 109203X was purchased from Calbiochem. Ang II, losartan, Go 6976, apocynin, and ML-171 were obtained from Tocris. Primary antibodies against Ca V 3.2, Ca V 3.1 and Nox1 were purchased from Novus Biologicals. Duolink PLA detection kit and all other chemicals were obtained from Sigma Aldrich. Data are expressed as means ± S.E., and n indicates the number of cells or arteries or animals. Paired t-test was performed to compare the effects of a given condition/treatment on whole cell current or arterial diameter. P values ≤ 0.05 were considered statistically significant. Data Availability. The datasets generated during and/or analyzed during the current study are available from the corresponding author in a reasonable request.