Differential effects of superoxide and hydrogen peroxide on myogenic signaling, membrane potential, and contractions of mouse renal afferent arterioles

Abstract

Myogenic contraction is the principal component of renal autoregulation that protects the kidney from hypertensive barotrauma. Contractions are initiated by a rise in perfusion pressure that signals a reduction in membrane potential (E_m) of vascular smooth muscle cells to activate voltage-operated Ca²⁺ channels. Since ROS have variable effects on myogenic tone, we investigated the hypothesis that superoxide (O₂^·−) and H₂O₂ differentially impact myogenic contractions. The myogenic contractions of mouse isolated and perfused single afferent arterioles were assessed from changes in luminal diameter with increasing perfusion pressure (40–80 mmHg). O₂^·−, H₂O₂, and E_m were assessed by fluorescence microscopy during incubation with paraquat to increase O₂^·− or with H₂O₂. Paraquat enhanced O₂^·− generation and myogenic contractions (−42 ± 4% vs. −19 ± 4%, P < 0.005) that were blocked by SOD but not by catalase and signaled via PKC. In contrast, H₂O₂ inhibited the effects of paraquat and reduced myogenic contractions (−10 ± 1% vs. −19 ± 2%, P < 0.005) and signaled via PKG. O₂^·− activated Ca²⁺-activated Cl⁻ channels that reduced E_m, whereas H₂O₂ activated Ca²⁺-activated and voltage-gated K⁺ channels that increased E_m. Blockade of voltage-operated Ca²⁺ channels prevented the enhanced myogenic contractions with paraquat without preventing the reduction in E_m. Myogenic contractions were independent of the endothelium and largely independent of nitric oxide. We conclude that O₂^·− and H₂O₂ activate different signaling pathways in vascular smooth muscle cells linked to discreet membrane channels with opposite effects on E_m and voltage-operated Ca²⁺ channels and therefore have opposite effects on myogenic contractions.

renal autoregulation maintains a stable renal blood flow and glomerular capillary pressure during changes in blood pressure by adjustments of the tone of the afferent arterioles that are the main resistance vessel of the kidney (8). A breakdown of renal autoregulation in hypertensive chronic kidney disease (CKD) can accelerate renal damage by transmitting elevated blood pressure into sensitive glomerular capillaries (37), their podocytes (22), and the renal parenchyma (termed barotrauma) (32). Renal autoregulation is bolstered by tubuloglomerular feedback and a third mechanism but depends primarily on a myogenic contraction that entails a rapid intrinsic contraction of vascular smooth muscle cells (VSMCs) during increased perfusion pressure (6, 8).

An increase in perfusion pressure reduces the membrane potential (E_m) of VSMCs of resistance vessels and opens voltage-operated Ca²⁺ channels (VOCCs) to raise intracellular Ca²⁺ that initiates myogenic contractions (8). Increases in the perfusion pressure of afferent arterioles generate ROS that have been previously variously reported to mediate (26, 27), leave unchanged (41), or impair (13, 44) autoregulation or myogenic contractions. The two primary ROS are superoxide (O₂^·−) and H₂O₂. Therefore, we tested the hypothesis that O₂^·− and H₂O₂ activate different signaling mechanisms and membrane channels that differentially change E_m, VOCCs, and myogenic contractions. This has clinical impact since a defect in renal autoregulation (17), especially in the presence of hypertension (8, 17, 22), accelerates the progression of CKD and nonselective antioxidants given to models of CKD have been reported either to be successful (25), ineffective (40), or harmful (33) in protecting the kidney from further damage. The first step to therapeutic progress should be a clearer definition of the preferred target.

Paraquat (N,N′-dimethyl-4,4′-bipyridinium dichloride) is a redox-cycling quinolone. It undergoes a single electron reduction in tissues, forming a free radical that is rapidly oxidized by molecular oxygen to generate the superoxide anion radical. Therefore, paraquat was used to generate intracellular O₂^·− (42) to study the graded effects of O₂^·− on myogenic tone. PKC is activated by O₂^·− (19) and can mediate myogenic contractions (18) by activating Cl⁻ channels (30) that can reduce E_m of VSMCs (3) and afferent arterioles (21). Ca²⁺-activated Cl⁻ channels (CaCCs) have been proposed recently to play a key role in the depolarization of E_m of VSMCs during a myogenic contraction (10). However, their role needs further definition since afferent arteriolar myogenic contractions persist after nonselective blockade of Cl⁻ channels (8, 47). We tested the hypothesis that O₂^·− activates PKC signaling to open CaCCs to depolarize E_m that disinhibits VOCCs to increase cellular Ca²⁺ entry to enhance myogenic contractions.

Unlike O₂^·−, low concentrations of H₂O₂ can dilate arterioles (9), although opposite effects have also been reported in some studies (39). H₂O₂ can dimerize and activate PKG in VSMCs (5, 55) and activate large-conductance Ca²⁺-activated K⁺ (BK_Ca) channels in human coronary arterioles (49) to increase E_m and cause relaxation (55). However, although BK_Ca channels modulate autoregulation in the cerebral circulation (20), they are usually quiescent in renal afferent arterioles (12, 38), which therefore requires further study. We tested the hypothesis that H₂O₂ activates PKG to hyperpolarize E_m by activation of specific K⁺ channels to blunt myogenic contractions.

Myogenic contractions were studied directly in single afferent arterioles dissected from mouse kidneys from the measured reduction in luminal diameter during increased perfusion pressure across a physiological relevant range of 40–80 mmHg (25–27). Arteriolar O₂^·− was quantified from pegalated (PEG)-SOD-inhibitable ethidium-to-dihydroethidium (E:DHE) fluorescence, H₂O₂ from PEG-catalase-inhibitable 6-carboxy-2′,7′-dichlorodihydrofluorescein diacetate (C-H₂DCFDA) fluorescence, and E_m from bis-(1,3-dibutylbarbituric acid)trimethine oxonol [DiBAC₄(3)] fluorescence.

The results demonstrated that O₂^·− and H₂O₂ activate two distinct signaling pathways with opposite effects on VOCCs and myogenic contractions. These opposing effects of the principal ROS may underlie some of the conflicting reports of the effects of oxidative stress on renal autoregulation and myogenic contractions in disease models and, by inference, on the progression of CKD.

METHODS

Isolation and perfusion of afferent arterioles.

Male C57Bl/6 mice (Charles River, Germantown, MD), aged 3–4 mo and weighing 25–30 g, were fed regular mouse chow and euthanized. All procedures conformed with the National Institutes of Health Guide for Care and Use of Laboratory Animals. Experiments were approved by the Georgetown University Animal Care and Use Committee. Mice were euthanized, one kidney was removed from the mouse, and a single afferent arteriole with an attached glomerulus was dissected. The arteriolar lumen was perfused via a pipette containing an internal pressure pipette, and the glomerulus was stabilized by a suction pipette in an organ bath at 37°C on the stage of an inverted microscope, as previously described in detail (7, 25, 27, 52).

Measurement of arteriolar myogenic responses.

The perfusion pressure was increased from 40 to 80 mmHg while the arteriolar luminal diameter was measured directly to assess myogenic responses, as previously described (25–27). Dose-response curves for paraquat or H₂O₂ were performed in individual arterioles by adding drugs from the lowest to highest doses at 15-min intervals. To assess the effects of O₂^·−, the perfused afferent arteriole was first incubated with graded doses of paraquat (10⁻⁸–10⁻⁴ mol/l), but subsequent experiments used the ED₅₀ dose of 10⁻⁶ mol/l. To assess the effects of H₂O₂, graded doses of 10⁻⁶–10⁻⁴ mol/l were used first, but subsequent experiments used the ED₅₀ dose of 10⁻⁵ mol/l. To assess the roles of specific signaling and ion channels, perfused afferent arterioles were preincubated with 3 × 10⁻⁶ mol/l of Gö6983 to inhibit PKC, 3 × 10⁻⁵ mol/l of Rp-8-bromo-PET-cGMP to inhibit PKG, 10⁻⁵ mol/l of CaCCinh-A01 to inhibit CaCCs, 10⁻⁷ mol/l of apamin and 10⁻⁸ mol/l of charybdotoxin to inhibit BK_Ca and small-conductance Ca²⁺-activated K⁺ (SKCa) channels, 10⁻⁵ mol/l of linopirdine to inhibit voltage-gated K⁺ channel 7 (K_v7), and 10⁻⁶ mol/l of nifedipine to inhibit VOCCs. To assess the role of nitric oxide (NO), arterioles were incubated with N^ω-nitro-l-arginine methyl ester (l-NAME; 10⁻⁴ mol/l), and, to assess the role of the endothelium, arterioles were perfused with saponin (0.125 mg/ml) for 10 min to remove endothelial function, which was confirmed by loss of the response to the endothelium-dependent vasodilator ACh (46).

Measurement of O₂^·− and H₂O₂.

O₂^·− generation in the individual perfused afferent arteriole was assessed by fluorescence microscopy from the PEG-SOD-inhibitable E:DHE fluorescence ratio (f/f₀) and H₂O₂ from PEG-catalase-inhibitable C-H₂DCFDA fluorescence, as previously described (29). Separate vessels were used to assess the changes in E:DHE and H₂DCFDA fluorescence. The specificity of the signals for O₂^·− and for H₂O₂ were examined by bath addition of PEG-SOD (200 U/ml) or PEG-catalase (1,000 U/ml), respectively.

Measurement of E_m.

E_m was evaluated using an anionic membrane potential-sensitive fluorescent dye, DiBAC₄(3) (11, 48). DiBAC₄(3) partitions across cell membranes according to the Nernst equation. It enters the cytosol during reductions in E_m, where its fluorescence intensity is enhanced by binding to cytosolic proteins, whereas during increases in E_m, it detaches from cytosolic proteins and exits the cell, which loses fluorescence. Afferent arterioles were perfused with physiological salt solution containing 2 μM DiBAC₄(3) at 36°C for 30 min in the dark before the start of the experimental protocols and throughout the experiments. Microvascular fluorescence images were observed through a fluorescence microscopy, and fluorescence intensity was quantitated with PTI Felix32.

Statistics.

Data were obtained from 5–7 mice/group and analyzed by ANOVA. Where appropriate, a post hoc Bonferroni method was used to test differences between groups. Results (mean ± SE values) were considered significant at P < 0.05.

RESULTS

Basal afferent arteriolar diameter during perfusion at 40 mmHg was reduced by paraquat but increased by H₂O₂ (Table 1). An increase in perfusion pressure from 40 to 80 mmHg contracted afferent arterioles (−19 ± 4%). This myogenic contraction was increased by paraquat. This was prevented by PEG-SOD but not by PEG-catalase (Fig. 1A). Incubation with H₂O₂ reduced myogenic contractions, but this inhibition was reduced by PEG-catalase (Fig. 1B). H₂O₂ at >100 μmol/l caused variable responses, and very high concentrations caused contractions (data not shown). Incubation with 10 μmol/l H₂O₂ reduced the sensitivity of the myogenic contractions to paraquat (Fig. 1C).

Table 1.

Basal afferent arteriolar diameter at 40 mmHg of perfusion pressure

	Arteriolar Diameter, μm
Drug	Before drug addition	After drug	Change, %
Paraquat (1 μM)	8.7 ± 0.5	5.7 ± 0.4	−35.3 ± 1.3**
H2O2 (10 μM)	9.3 ± 0.8	10.0 ± 0.7	+9.1 ± 3.3*
Gö-6983 (3 μM)	10.1 ± 0.7	10.1 ± 0.8	−0.3 ± 1.2
Rp-8-bromo-PET-cGMP (30 μM)	9.0 ± 0.6	9.0 ± 0.6	−0.2 ± 1.0
CaCCinh-A01 (10 μM)	11.3 ± 0.8	11.8 ± 0.7	+4.6 ± 1.1*
Apamin (100 nM) + charybdotoxin (10 nM)	11.0 ± 0.4	10.9 ± 0.4	−1.0 ± 0.5
Linopirdine (10 μM)	9.4 ± 0.3	9.3 ± 0.3	−0.7 ± 0.7
Apamin + charybdotoxin + linopirdine	9.3 ± 0.4	9.1 ± 0.3	−2.5 ± 1.1
TRAM34 (1 μM)	11.1 ± 0.5	11.0 ± 0.5	−0.4 ± 0.3
Nifedipine (1 μM)	11.1 ± 0.5	11.1 ± 0.5	+0.1 ± 0.9
Nω-nitro-l-arginine methyl ester (100 μM)	10.4 ± 0.4	9.7 ± 0.4	−7.0 ± 0.6***

Values are means ± SE; n = 5–7 per group. Shown are values before and 15–30 min after bath addition of the drug.

^*P < 0.05,

^**P < 0.01, and

^***P < 0.005 compared with before drug addition.

Fig. 1.

Paraquat (PQ) enhances afferent arteriolar contractions during increases in perfusion pressure (PP) from 40 to 80 mmHg, whereas H₂O₂ has opposite effects. A: effects of PQ (10⁻⁶ mol/l) alone or with pegalated (PEG)-SOD (200 U/ml) or with PEG-catalase (1,000 U/ml) on myogenic contractions. B: effects of H₂O₂ (10⁻⁵ mol/l) alone or with PEG-SOD (200 U/ml) or with PEG-catalase (1,000 U/ml) on myogenic contractions. C: effects of PQ (10⁻⁶ mol/l) alone or with H₂O₂ (10⁻⁵ mol/l). D: changes in the ethidium-to-dihydroethidium (E:DHE) fluorescence ratio of afferent arterioles with PQ (10⁻⁶ mol/l) alone or with PEG-SOD (200 U/ml) or with PEG-catalase (1,000 U/ml). E: changes in 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) fluorescence of afferent arterioles by H₂O₂ (10⁻⁵ mol/l) alone or with PEG-SOD (200 U/ml) or with PEG-catalase (1,000 U/ml). F: changes in H₂DCFDA fluorescence of afferent arterioles by PQ (10⁻⁶ mol/l) alone or with PEG-SOD (200 U/ml) or with PEG-catalase (1,000 U/ml). MR, myogenic response. *P < 0.05, **P < 0.01, and ***P < 0.005 compared with vehicle; ††P < 0.01 compared with PQ or H₂O₂ alone.

The E:DHE fluorescence signal for O₂^·− was enhanced during increased perfusion pressure, as previously described (27), and was enhanced further by paraquat (Fig. 1D). This was >85% prevented by PEG-SOD but was unaffected by PEG-catalase and was unchanged by 10⁻⁵ mol/l H₂O₂ (Fig. 1D). The C-H₂DCFDA fluorescence signal for H₂O₂ was enhanced by H₂O₂ but was >85% prevented by PEG-catalase but not by PEG-SOD (Fig. 1E) and was little changed by 10⁻⁶ mol/l paraquat alone. However, paraquat with PEG-SOD increased H₂DCFDA fluorescence, whereas this was reduced by coincubation with PEG-catalase (Fig. 1F).

Incubation of afferent arterioles with 3 × 10⁻⁶ mol/l Gö6983 to block PKC inhibited basal myogenic contraction and blocked >80% of its augmentation by paraquat (Fig. 2A). Incubation with 3 × 10⁻⁵ mol/l Rp-8-bromo-PET-cGMP to block PKG had no effect on basal myogenic contraction but blocked >90% of its inhibition by H₂O₂ (Fig. 2B).

Fig. 2.

PKC mediates the effects of PQ on myogenic contractions and membrane depolarization and PKG mediates the effects of H₂O₂. A and C: effects of PQ (10⁻⁶ mol/l) and/or Gö6983 (3 × 10⁻⁶ mol/l). B and D: effects of H₂O₂ (10⁻⁵ mol/l) and/or Rp-cGMPs (3 × 10⁻⁵ mol/l). DiBAC₄(3), bis-(1,3-dibutylbarbituric acid)trimethine oxonol. ***P < 0.005 compared with vehicle; ††P < 0.01 and †††P < 0.005 compared with PQ or H₂O₂ alone.

An increase in DiBAC₄(3) fluorescence implies a reduced E_m. An increase in perfusion pressure increased fluorescence modestly (Fig. 2C), which was increased further by paraquat and was prevented by blockade of PKC (Fig. 2C). The increase in fluorescence with perfusion pressure was reversed by incubation with H₂O₂, and this was prevented by blockade of PKG (Fig. 2D).

Blockade of CaCCs with 10⁻⁵ mol/l CaCCinh-A01 increased basal afferent arteriolar diameter (Table 1), blocked the reduction in E_m with perfusion pressure, and blocked 65% of the augmentation of E_m with paraquat (Fig. 3A). Likewise, CaCCinh-A01 blunted myogenic contractions and blocked 40% of its augmentation by paraquat (Fig. 3B).

Fig. 3.

Membrane potential (E_m) is reduced by activation of Ca²⁺-activated Cl⁻ channels (CaCCs) with PP or PQ. DiBAC₄(3) fluorescence (A) was increased with PP and PQ (10⁻⁶ mol/l) (implying depolarization), but this was blunted by inhibition of CaCCs with CaCCinh-A01 (10⁻⁵ mol/l). Luminal diameter (B) was reduced by PP, and this was enhanced by PQ but blunted by CaCCinh-A01. *P < 0.05 and ***P < 0.005 compared with vehicle; †††P < 0.005 compared with PQ alone.

Pilot experiments were undertaken to determine the effects of antagonists of different K⁺ channels on myogenic contractions since variable responses have been previously reported (8). Responses were reduced or unchanged after blockade of ATP-sensitive K⁺ (K_ATP) channels, intermediate-conductance Ca²⁺-activated K⁺ channels, and inward rectifier K⁺ (Kir2) channels with glibenclamide, TRAM-34, and ML-133, respectively (data not shown). Therefore, these channels were not further investigated as mechanisms for reduced myogenic contraction with H₂O₂. However, blockade of SK_Ca and BK_Ca channels with apamin (10⁻⁷ mol/l) and charybdotoxin (10⁻⁸ mol/l) or blockade of K_v7 with 10⁻⁵ mol/l linopirdine enhanced myogenic contractions and therefore were selected for further study. These antagonists individually reduced, but did not prevent, the blunting of myogenic contractions by H₂O₂ (Fig. 4, A and B). However, incubation of arterioles with all of these three K⁺ channel blockers together prevented the effects of H₂O₂ to reduce myogenic contractions (Fig. 4C) or to hyperpolarize E_m (Fig. 4, D–F).

Fig. 4.

H₂O₂ activates small-conductance Ca²⁺-activated K⁺ (SK_Ca) channels, large-conductance Ca²⁺-activated K⁺ (BK_Ca) channels, and voltage-activated K⁺ channel 7 (K_v7) to blunt myogenic contractions. The reduction in luminal diameter with PP was inhibited by H₂O_2, but the effect of H₂O₂ was lessened by blockade of SK_Ca and BK_Ca channels with apamin (10⁻⁷ mol/l) and charybdotoxin (10⁻⁸ mol/l) (A) or blockade of K_v7 by linopirdine (10⁻⁵ mol/l; B) and was prevented by blockade of all channels by a combination of all three blockers (C). The increase in E_m by H₂O₂ was prevented by blockade of K⁺ channels. DiBAC₄(3) fluorescence was decreased with H₂O₂ (10⁻⁵ mol/l) (implying hyperpolarization), but this effect of H₂O₂ was lessened by apamin (10⁻⁷ mol/l) and charybdotoxin (10⁻⁸ mol/l; D) or linopirdine (10⁻⁵ mol/l; E) and prevented by a combination of all three blockers (F). *P < 0.05, **P < 0.01, and ***P < 0.005 compared with vehicle; †P < 0.05, ††P < 0.01, and †††P < 0.005 compared with H₂O₂ alone; ‡P < 0.05 and ‡‡‡P < 0.005 compared with antagonists alone.

Blockade of VOCCs with 10⁻⁶ mol/l nifedipine prevented myogenic contractions and blocked >85% of the enhancing effect of paraquat (Fig. 5A) yet left intact the reduction in E_m with paraquat (Fig. 5B).

Fig. 5.

Changes in E_m are upstream from activation of voltage-operated Ca²⁺ channels. Luminal diameter (A) was reduced by PP and reduced further by PQ (10⁻⁶ mol/l), but this was largely prevented by nifedipine (10⁻⁶ mol/l). E_m (B) was reduced by PP or PQ (10⁻⁶ mol/l) (implying depolarization), but these effects were unchanged by nifedipine (10⁻⁶ mol/l). ***P < 0.005 compared with vehicle; †††P < 0.005 compared with PQ alone.

Functional removal of the endothelium by intraluminal perfusion of saponin did not affect myogenic contractions (Fig. 6A) or E_m (Fig. 6B). Blockade of NO generation with l-NAME (10⁻⁴ mol/l) reduced basal afferent arteriolar diameter (Table 1) but did not prevent the enhanced myogenic contractions with paraquat (Fig. 6C) and did not modify the reduction in myogenic contractions with low concentration of H₂O₂ but did enhance the effects of higher concentrations (Fig. 6D).

Fig. 6.

Changes in DiBAC₄(3) fluorescence by PP alone or with saponin (0.125 mg/ml; A), changes in luminal diameter by by PP alone or with saponin (0.125 mg/ml; B), by PQ (10⁻⁶ mol/l) alone or with N^ω-nitro-l-arginine methyl ester (l-NAME; 10⁻⁶ mol/l; C), or by H₂O₂ (10⁻⁵ mol/l) alone or with l-NAME (10⁻⁶ mol/l; D) during increasing PP from 40 to 80 mmHg. *P < 0.05 and **P < 0.01 compared with vehicle (by two-way ANOVA); ††P < 0.01.

DISCUSSION

The details of the myogenic mechanism remain controversial despite its importance in renal, cerebral, and cardiac protection against damage from hypertension (6, 8, 22, 32). We confirmed that an increase in perfusion pressure induces strong myogenic contractions. O₂^·− is generated in afferent arterioles from normal mice or mice with genetic deletion of SOD and mediates much of the myogenic contractions (26, 27, 29). The main new findings are that the generation of O₂^·− with increased perfusion pressure is enhanced dose dependently by paraquat, which also enhances myogenic contractions. In contrast, almost no generation of H₂O₂ by perfusion pressure could be detected in normal afferent arterioles. This is consistent with the very low endogenous vascular H₂O₂ concentration of perhaps <10⁻⁶ mol/l (31). This study showed that such low concentration of H₂O₂ had no effects on basal diameter, E_m, or myogenic contractions of afferent arterioles. However, incubation with >10⁻⁶ mol/l H₂O₂ impaired myogenic contractions as in a previous study (24) of a mouse model of CKD. E_m of afferent arterioles is depolarized by increased perfusion pressure and is enhanced by paraquat via activation of PKC, but this is transformed into hyperpolarization with H₂O₂ via activation of PKG. Blockade of CaCCs blunts the reduction in E_m and the enhanced myogenic contractions with paraquat. A combined blockade of SK_Ca, BK_Ca, and K_v7 channels blocks the effects of H₂O₂ to hyperpolarize E_m and to blunt myogenic contractions. Blockade of VOCCs prevents myogenic contractions and the enhancing effects of paraquat but leaves intact the changes in E_m. The vascular endothelium is not involved in myogenic contractions. NO synthase (NOS) is not required for the blunting of the myogenic response by modest concentrations of H₂O₂.

E:DHE fluorescence in afferent arterioles was a relatively specific signal for O₂^·− since its fluorescence was increased by perfusion pressure or by paraquat but not by H₂O₂, and the effects of paraquat were greatly diminished by incubation with PEG-SOD but not by PEG-catalase. DHE is oxidized by O₂^·− to oxyethidium, which is detected with ethidium using our fluorescence system. Precise quantitation of O₂^·− requires HLPC separation of oxyethidium from ethidium (15), but this is not possible with a single afferent arteriole. Nevertheless, our results suggest that the majority of the fluorescence signal in arterioles loaded with DHE can be ascribed to O₂^·− since it is prevented by PEG-SOD but not by PEG-catalase. C-H₂DCFDA fluorescence was a relatively specific signal for H₂O₂ since its fluorescence was increased by exogenous H₂O₂ or H₂O₂ generated from paraquat-induced O₂^·− by PEG-SOD but not by paraquat alone. The increased H₂O₂ signals were prevented by PEG-catalase but not by PEG-SOD.

Most of the enhanced myogenic contractions with paraquat could be ascribed to O₂^·− since the enhanced E:DHE signal and enhanced myogenic contractions were reduced or prevented by metabolism of O₂^·− by PEG-SOD but were unaffected by PEG-catalase. Paraquat also enhances afferent arteriolar contractions during increased ROS generation by thromboxane (50). Recently, Vogel et al. (50) reported that paraquat enhances Ca²⁺ entry into arterioles via l-type VOCCs but leaves unchanged cellular Ca²⁺ derived from store-operated Ca²⁺ channels and Ca²⁺-induced Ca²⁺ release from the sarcoplasmic reticulum. This is quite consistent with our finding that paraquat reduced E_m since this should disinhibit L-type channels, raise intracellular Ca²⁺, and enhance myogenic contractions. Indeed, our data demonstrated that the reduction in E_m with paraquat is upstream from activation of VOCCs since blockade of VOCCs with nifedipine disabled myogenic contractions without perturbing the reduction in E_m. This extends findings from a previous report (27) showing that myogenic contractions were prevented by incubation in Ca²⁺-free solution without perturbing the increase in O₂^·−.

The myogenic response of the afferent arteriole is initiated by stretch or deformation of VSMCs, which is transduced into intracellular signals. The activated signals lead to a reduction in E_m that disinhibits VOCCs to increase intracellular Ca²⁺ and thereby to activate myosin light chain kinase, which mediates the contraction (8, 50). O₂^·− generated from NADPH oxidase is required for mouse afferent arteriolar myogenic contractions since these are blocked by apocynin (27). Although these arterioles express neutrophil oxidases (NOX)-2 and NOX-4 as well as p47^phox and p22^phox (2, 51), the NOX-2/p22^phox/p47^phox pathway is implicated since an increase in perfusion pressure of afferent arterioles from p47^phox gene-deleted mice fails to generate O₂^·− or contractions (26). Step increases in perfusion pressure elicit step increases in vascular inositol 1,4,5-triphosphate and diacylglycerol, which can activate PKC (34). Our results confirm that PKC is required for myogenic contractions and further show that it mediates the membrane depolarization by perfusion pressure or paraquat. This is consistent with activation of PKC by O₂^·− in renal tubules (19, 53), Moreover, the reduction in E_m and enhanced myogenic contraction with paraquat depend in part on Cl⁻ channels (CaCCs). The relationship between O₂^·− and activation of Cl⁻ channels is complex. O₂^·− may traverse membranes via Cl⁻ channels in activated neutrophils, and O₂^·− can open Cl⁻ channels in the phagocytic vacuole (43) and cell membrane (45). CaCCs mediate vascular contractions to thromboxane (1). Moreover, Cl⁻ channels in endothelial cells mediate changes in cell membrane current with O₂^·− (4), but this is the first documentation of the role of CaCCs in mediating myogenic responses to O₂^·−. Overall, the data indicate that O₂^·− generated from paraquat or p47^phox/NOX-2 activates PKC signaling that subsequently activates CaCCs, depolarizes E_m of VSMCs, disinhibits VOCCs, raises intracellular Ca²⁺, and initiates a myogenic contraction.

A novel finding was that exogenous H₂O₂ blunted myogenic contractions of afferent arterioles, but H₂O₂ apparently was not generated with increased perfusion pressure in normal mouse afferent arterioles since myogenic contractions of normal vessels were unaffected by metabolizing H₂O₂ with PEG-catalase. However, micromolar concentrations of exogenous H₂O₂ inhibited myogenic contractions and antagonized the effects of O₂^·− in a dose-dependent manner. We used the ED₅₀ dose of H₂O₂ (10⁻⁵ mol/l) that blunted about half of myogenic contraction and increased E_m as indicated by a reduction of intracellular DiBAC₄(3) fluorescence. H₂O₂ can alter vascular contractions by Ca²⁺-dependent and -independent manners (55). The finding that H₂O₂ did not completely prevent myogenic contractions while reversing the changes in E_m suggests a Ca²⁺-independent component of action. An increase in H₂O₂ can enhance K⁺ conductance in VSMCs (5) and in human coronary arterioles (55). However, H₂O₂ can block mitochondrial K_ATP channels (56), which might underlie the apparently paradoxical reduction in myogenic contractions observed after inhibition of K_ATP channels with glibenclamide (data not shown). In contrast, K_v channels are activated by H₂O₂ in coronary arteries (35) and mesenteric arterioles (36), where they mediate vasodilation. H₂O₂ activates BK_Ca channels in fibroblasts (14) and human coronary arteries (31, 55) and K_v7 channels in neurons (16). These reports are consistent with our conclusion that H₂O₂ activates BK_Ca, K_v, and K_v7 channels to inhibit myogenic contractions. Although rat afferent arteriolar K⁺ channels are normally quiescent, they can be activated by ROS in models of diabetes, where they mediate the vasodilation that contributes to diabetic hyperfiltration (49). Blockade of SK_Ca, BK_Ca, or K_v7 channels was required to prevent the responses to H₂O₂. This indicates that H₂O₂ blunts myogenic contractions by activation of specific K⁺ channels but suggests significant redundancy amongst them. Basal levels of H₂O₂ were not implicated in the modulation of K⁺ channels or contractions with perfusion pressure since there were no effects of PEG-catalase. However, exogenous H₂O₂ activated specific K⁺ channels, hyperpolarized E_m, and prevented myogenic contraction via PKG. The regulation of K⁺ channels by ROS has been closely studied in the heart, where they are implicated in ischemic dysrhythmias (54). H₂O₂ activates PKG (5) and BK_Ca channels in coronary arterioles (49, 55), as in afferent arterioles in the present study.

H₂O₂ can activate aortic endothelial NOS3 (28). However, myogenic contractions were unaltered in NOS3 gene-deleted mice (27) and usually do not depend on the endothelium (8), as shown in this study. Although blockade of NOS in this study reduced basal arteriolar diameter and enhanced myogenic contractions moderately, it did not prevent the blunting of the response by H₂O₂, except at the highest concentrations. This suggests that NO determines basal vascular diameter in afferent arteriole but does not normally mediate the effects of physiological concentrations of H₂O₂ on myogenic tone.

This study has some limitations. First, PKC has more than 10 isoforms. The specific isoform that mediates the effects of O₂^·− needs further study. Second, patch-clamp techniques would be required to precisely confirm the ion channels involved in changes in E_m, but these are hard to perform on vessels that are stretching and contracting. Third, this study did not separate effects of pressure and flow. Flow normally induces an endothelium-dependent relaxation in preconstricted coronary arterioles (55).

In conclusion, O₂^·− and H₂O₂ modulate myogenic contractions differentially. O₂^·− acts via PKC-induced activation of channels, including CaCCs, which depolarizes E_m to disinhibit VOCCs and enhance myogenic tone, whereas H₂O₂ acts via PKG-induced activation of SK_Ca, BK_Ca, and K_v7 channels to hyperpolarize E_m and reduce myogenic tone.

Perspectives

These findings may help to explain previous conflicting reports of the effects of ROS on autoregulation and myogenic contractions. Mice with SOD gene deletions generate excessive vascular O₂^·− and develop very strong myogenic contractions (7, 29). Surprisingly, mice with the reduced renal mass model of CKD that excrete sixfold more 8-isoprostane F_2α (produced by the reaction of O₂^·− with arachidonate) (23) and also generate excessive afferent arteriolar O₂^·− have severely blunted myogenic contractions (25). However, mice with reduced renal mass also excreted threefold more H₂O₂, which our present study suggests could have antagonized the effects of O₂^·− to enhance myogenic contractions. Thus, overproduction of H₂O₂ in the afferent arteriole may be the cause of the reduced myogenic contractions in some settings despite increased O₂^·−, but this remains to be investigated. Generation of H₂O₂ from metabolism of O₂^·− by SOD in the afferent arteriole during prolonged oxidative stress also may prevent excessive and prolonged contractions leading to renal ischemia and might thereby be beneficial in maintaining some renal blood flow.

GRANTS

This work was supported by National Institutes of Health Grants DK-49870, DK-36079, and HL-68086, by funds from the George E. Schreiner Chair of Nephrology, the Smith-Kogod Family Foundation, and the Georgetown University Hypertension, Kidney and Vascular Research Center, and by National Nature Science Foundation of China Grant 31471100 (to E. Y. Lai).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

L.L. and E.Y.L. conception and design of research; L.L. and E.Y.L. performed experiments; L.L. and E.Y.L. analyzed data; L.L., E.Y.L., A.W., and W.J.W. interpreted results of experiments; L.L. and E.Y.L. prepared figures; L.L. and E.Y.L. drafted manuscript; C.S.W. edited and revised manuscript; C.S.W. approved final version of manuscript.