INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

BLOOD FLOW TO THE MEDULLA of the kidney is supplied by descending vasa recta (DVR). These vessels arise from juxtamedullary efferent arterioles and either traverse the center of outer medullary vascular bundles to supply blood flow to the inner medulla or branch from the bundle periphery to supply the outer medullary interbundle region (12). Thus DVR might have a role to regulate the relative distribution of blood flow to the outer vs. inner medulla of the kidney. In addition to endothelial expression of aquaporin 1 and facilitated urea transport, smooth muscle remnants known as pericytes impart a contractile function to the DVR wall (25, 26). The signaling pathways that mediate and modulate DVR vasoreactivity are largely unexplored and would be expected to play a role in the regulatory functions of the renal medulla, including urinary concentration and the control of salt and water excretion (5, 20, 21).

Calcium influx through voltage-gated channels mediates contraction of many types of smooth muscle (11, 29). In the kidney, it is generally accepted that afferent arteriolar smooth muscle responds to ANG II stimulation by depolarizing and activating L-type calcium channels. Studies in the efferent arteriole have yielded mixed results, most often failing to identify a role for depolarization and activation of voltage-gated calcium entry pathways (1, 2, 4, 7, 10, 14-17). Hansen et al. (9) recently shed new light on this subject by identifying expression of both L- and T-type calcium channels in juxtamedullary but not cortical efferent arterioles. Also, we recently showed that DVR pericytes respond to ANG II by depolarizing (23, 35), an event that is expected to presage voltage-gated calcium entry into the pericyte cytoplasm. In this study we tested whether vasoconstriction and vasodilation of DVR is accompanied, in parallel, by depolarization and hyperpolarization of pericyte membrane potential. The results confirm that depolarization by K⁺ channel blockers or high KCl induces mild vasoconstriction and that the vasodilators pinacidil and bradykinin repolarize ANG II-treated pericytes. Finally, a role for voltage-gated calcium entry pathways was confirmed because the L-type calcium channel blocker diltiazem relaxes ANG II- or KCl-constricted DVR and prevents ANG II-induced calcium entry into the pericyte cytoplasm.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation of DVR. Kidneys were harvested from Sprague-Dawley rats (70-150 g; Harlan) that had been anesthetized by an intraperitoneal injection of thiopental (50 mg/kg body wt). Tissue slices were placed in buffer and maintained at 4°C. Individual DVR were dissected from outer medullary vascular bundles and transferred to the stage of an inverted microscope for fluorescence imaging or microperfusion studies as previously described (22-24, 27). The solution used for dissection, microperfusion, and measurement of vasoreactivity contained (in mM) 140 NaCl, 10 Na acetate, 5 KCl, 1.2 MgSO₄, 1.2 Na₂HPO₄, 1 CaCl₂, 5 HEPES, 5 L-alanine, 5 D-glucose, and 0.5 g/dl bovine albumin. The pH was adjusted to 7.55 at room temperature using NaOH to yield a pH of ~7.4 at 37°C. When patch clamp studies were performed, tissue was stored in physiological saline solution (PSS). PSS also served as the extracellular solution during membrane potential recordings. PSS contained (in mM) 145 NaCl, 5 KCl, 1 MgCl₂, 1 CaCl₂, 10 HEPES, 10 glucose, pH 7.4 at room temperature.

Measurement of DVR diameter. Vasoactivity was monitored in DVR perfused in vitro as previously described (22). Vessels were recorded on videotape (Panasonic WV-BL90) and diameters were analyzed during playback. The inverted microscope supporting the chamber was equipped with a beam splitter and a side port with a C-mount for attachment of a video camera. DVR luminal diameter was observed with a ×40 dry objective; the final magnification on the videoscreen was ×1,300. Internal diameters were measured using calipers at the location of maximal constriction. Diameter changes are expressed as percentage constriction, given by [1  (D/D₀)] × 100, where D is experimental diameter and D₀ is initial baseline diameter. The effects of KCl- and TEACl-induced depolarization were tested by isosmolar substitution of NaCl to yield the desired KCl or TEACl concentration. In some protocols, either 100 mM KCl or ANG II (10⁸ M) was substituted into the bath to preconstrict vessels and permit assessment of the subsequent effects of vasodilators.

Isolation of pericytes from the DVR wall. Small wedges of renal medulla were separated from kidney slices by dissection and transferred to CaCl₂-free PSS containing collagenase 1A (0.45 mg/ml, Sigma), protease XIV (0.4 mg/ml, Sigma), and albumin (1.0 mg/ml) (23, 35). These were incubated at 37°C for 22 min and then transferred back to CaCl₂ (1 mM) containing PSS and held at 4°C in a petri dish. At intervals, vessels were isolated from the digested renal tissue by microdissection and transferred to a perfusion chamber on an inverted microscope (Nikon Diaphot). In the chamber, DVR were captured and aspirated into a microperfusion-style collection pipette with an opening of 5-10 µm. As previously described and illustrated (32), during the aspiration, pericytes strip from the abluminal surface of the vessel. Once the vessel has been completely drawn into the pipette, a preparation of isolated pericytes remains suspended in the bath, adherent to the pipette tip, available for fura-2 loading and intracellular calcium ([Ca²⁺]_i) measurements.

Measurement of [Ca²+]_i. Pericytes were loaded with fura-2 by incubating them with the fura-2 AM ester (10 µM, Molecular Probes, Eugene, OR) for 20 min at 37°C in the presence of probenecid (1 mM). This yields a strong fluorescent signal because the anion transport inhibitor probenecid prevents leakage of de-esterified fura-2 from the cytoplasm (32). A photon-counting photomultiplier assembly was employed to measure the fluorescent emission of fura-2 at 510 nm. Excitation was provided by a 75-W xenon arc lamp using a 350/380 nm wavelength combination isolated with a computer-controlled monochrometer (PTI, Lawrenceville, NJ). Fluorescent emission was isolated with a 510WB40 bandpass filter (Omega Optical, Brattleboro, VT) and collected with a Nikon CF fluor ×40 oil immersion objective (1.3 numerical aperture). The background-subtracted ratio of fluorescent emission (R_350/380) was converted to the equivalent intracellular calcium concentration assuming a dissociation constant of 224 nM for fura-2 at 37°C. R_max and R_min were measured by exposing vessels to buffer containing 5 mM CaCl₂ or 0 CaCl₂, 0.5 mM EGTA, respectively, along with 10 µM ionomycin (24).

Whole cell patch clamp recording. Membrane potential (m) was monitored by patch clamp recording from pericytes at room temperature. To accomplish this, DVR were digested with the same enzyme treatment described above for pericyte isolation (23, 35). Patch clamp studies of pericytes were always done on intact vessels, i.e., pericytes isolated by stripping were used for [Ca²⁺]_i measurements but not for electrophysiological studies. Patch pipettes were made from borosilicate glass capillaries (PG52151-4, external diameter 1.5 mm, internal diameter 1.0 mm; World Precision Instruments, Sarasota, FL), using a two-stage vertical pipette puller (Narshige PP-830) and heat polished. To obtain electrical access for whole cell perforated patch clamp recording, nystatin was used as the pore-forming agent. The pipette solution contained (in mM) 120 Kaspartate, 20 KCl, 10 NaCl, 10 HEPES, pH 7.2, and nystatin (100 µg/ml with 0.1% DMSO) in ultrapure water. Nystatin in DMSO was kept frozen at 20°C and renewed weekly. Each day, the nystatin stock was thawed, dispensed into the Kaspartate pipette solution at 37°C by vigorous vortexing for 1 min, and subsequently protected from light. Pipettes were backfilled with nystatin-containing electrode solution via a 0.2-µm filter.

Reagents. ANG II, bradykinin (BK), probenecid, pinacidil, diltiazem, ionomycin, bovine serum albumin (A2153, Cohn fraction V), nystatin, collagenase 1A, and protease XIV were from Sigma (St. Louis, MO). ANG II, BK, pinacidil, and diltiazem were stored in water in 200-µl aliquots at 20°C and diluted 1:100 or 1:1,000 on the day of the experiment. The enzyme digestion solution was prepared in 50-ml batches, frozen in 2-ml aliquots, and thawed daily as needed. Fura-2 (Molecular Probes) was stored at 1 mM in anhydrous DMSO. Reagents were thawed once and the excess was discarded at the end of the day.

Statistics. Data in the text and figures are given as means ± SE. The significance of differences between means was calculated using Student's t-test (paired or unpaired, as appropriate) and analysis of variance. Where sampling rates were high, the majority of error bars were suppressed to clarify display of data.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Vasoconstriction by K+ channel blockers. We verified that the K⁺ channel blockers BaCl₂ and TEACl were effective to depolarize DVR pericytes. During I = 0 current clamp recording of m, BaCl₂ (1 mM) or TEACl (30 mM) was introduced into the bath for 5 min and then washed out. BaCl₂ depolarized pericytes from 56.2 ± 2.8 to 36.0 ± 4.5 mV (n = 7, P < 0.05, Fig. 1, A and B). TEACl depolarized pericytes from 62.0 ± 2.4 to 35.8 ± 4.3 mV (Fig. 1, C and D, n = 7, P < 0.05). Having established that these agents depolarize the pericyte cell membrane, we tested whether they would also constrict in vitro-perfused DVR. After 2 min of baseline recording, either BaCl₂ (Fig. 2A) or TEACl (Fig. 2B) was added to the bath for 10 min and then washed out. BaCl₂ decreased luminal diameter from 12.9 ± 0.8 µm to a minimum of 11.9 ± 0.6 µm, whereas TEACl constricted from 13.4 ± 0.6 to 11.6 ± 0.6 µm. Despite similar degrees of depolarization (Fig. 1, B and D), TEACl was a more effective constrictor of DVR than BaCl₂ (P < 0.05, 6-14 min, %constriction, ordinate, Fig. 2, A and B). We speculate that TEA was a better constrictor because Ba²⁺ can compete for Ca²⁺ influx pathways and Ca²⁺-dependent intracellular signaling processes. Subsequent to washout of BaCl₂ and TEACl, to verify contractility of the vessels and compare their effectiveness to a biological constrictor, ANG II (10⁸ M) was added to the bath. ANG II reduced luminal diameter to 8.6 ± 0.5 and 8.7 ± 0.5 µm for BaCl₂ and TEACl groups, respectively. Thus both BaCl₂ and TEACl constricted DVR but were substantially less effective than ANG II.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Effect of BaCl2 on pericyte membrane potential and contractility. A: baseline membrane potential (m) (63 mV) was recorded for 1 min, after which BaCl2 (1 mM) was added to the extracellular buffer for 5 min and then removed. Recording is an example of experiments summarized in B. B: mean of m during the final 10 s of each period is shown by the ordinate (* P < 0.05 vs. resting potential, n = 7). C: baseline membrane potential (66 mV) was recorded for 1 min, after which TEACl (30 mM) was added to the extracellular buffer for 5 min and then removed. Recording is an example of experiments summarized in D. D: mean of m during the final 10 s of each period is shown by the ordinate (* P < 0.05 vs. resting potential, n = 7).

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Constriction of in vitro-perfused descending vasa recta (DVR) by BaCl2 and TEACl. A: in vitro-perfused DVR luminal diameter was recorded before and during 10-min exposure to BaCl2 or sham exchange (n = 7, each). At the end of the experiment, BaCl2 was washed out and ANG II (108 M) was added to the bath. B: in vitro-perfused DVR luminal diameter was recorded before and during 10-min exposure to TEACl (30 mM) or sham exchange (n = 8, 7, respectively). At the end of the experiment, TEACl was washed out and ANG II (108 M) was added to the bath. * P < 0.05 vs. control.

KCl-induced depolarization of DVR. To achieve pericyte membrane depolarization exceeding that resulting from BaCl₂- or TEACl-induced K⁺ channel blockade (Fig. 1), we tested the effect of raising extracellular KCl concentration to 100 mM. As expected, this maneuver strongly depolarized the pericytes from 58.2 ± 2.5 to 12.5 ± 0.2 mV (Fig. 3, A and B). We previously showed that ANG II depolarizes DVR pericytes toward the expected equilibrium potential for Cl ion, about 33 mV (see below) (23). Despite the fact that KCl depolarized pericytes to a greater extent than ANG II, it was less effective in constricting in vitro-perfused DVR (Fig. 3C). In separate experiments we tested the hypothesis that depolarization by 100 mM KCl would increase pericyte [Ca²⁺]_i. Again, KCl depolarization increased pericyte [Ca²⁺]_i, but did so less effectively than ANG II (Fig. 3D). The effect of KCl to depolarize m occurs more rapidly in Fig. 3A than the effect to vasoconstrict or increase calcium (Fig. 3, C and D). This is likely to be related to differences in the rate of bath exchange that, with our apparatus, is faster for patch clamp studies than for videomicroscopy or fluorescent microscopy.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Effect of 100 mM extracellular KCl on pericyte m, contractility, and intracellular calcium concentration ([Ca2+]i). A: baseline m (71 mV) was recorded for 1 min, after which 100 mM KCl was isosmotically exchanged into the extracellular buffer for 2 min and then washed out. Data were sampled at 10 Hz. Recordings are examples of experiments summarized in B. B: values on the ordinate show the mean of m during the final 10 s of 100 mM KCl in experiments from A (* P < 0.01 vs. resting potential, n = 8). C: in vitro-perfused DVR luminal diameter was recorded before and during exposure to 5 mM KCl (control, n = 7), 100 mM KCl (n = 5), or ANG II (108 M) (n = 8). D: effect of ANG II (108 M, n = 8) or 100 mM KCl (n = 8) on [Ca2+]i of fura-2-loaded pericytes is shown. Both increased pericyte [Ca2+]i but ANG II was more effective.

Vasodilation and membrane repolarization by pinacidil and BK. If voltage-gated Ca²⁺ entry pathways exist in DVR pericytes, agents that hyperpolarize the cells should be vasodilators. To test this hypothesis, we examined the effects of the K_ATP channel opener pinacidil. When this agent was applied to the bath of DVR pericytes in log molar increasing concentrations, m progressively declined from resting levels toward the equilibrium potential of K⁺ ion (Fig. 4), a finding that supports the expression of K_ATP channels in these cells. We tested whether this agent would both repolarize DVR pericytes and vasodilate in vitro-perfused vessels constricted by ANG II. ANG II (10⁸ M) depolarized pericytes from a resting level of 63.5 ± 3.5 to 32.0 ± 1.0 mV (n = 9, P < 0.01). Addition of pinacidil (10⁵ M) to the bath induced m oscillations in all but one cell, examples of which are shown in Fig. 5, A and B. On average, the effect of pinacidil was to repolarize the pericytes to 53.7 ± 4.3 mV (P < 0.01, Fig. 5C), an effect that was reversible after washout. As anticipated, pinacidil effectively vasodilated ANG II (10⁸ M)-preconstricted DVR (n = 8, Fig. 5D).

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Effect of pinacidil on resting pericyte m. Pericyte m was recorded for 1 min, after which pinacidil was added to the extracellular buffer in log molar increasing concentrations from 107 to 104 M at 5-min intervals. Means ± SE of n = 5 recordings is presented. m was sampled at 10 Hz, averaged to 1 Hz, and then averaged for all vessels for presentation. Most error bars have been suppressed for clarity.

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Effect of pinacidil (Pnc) on ANG II-treated pericytes. A and B: pericyte m was recorded at baseline (58 mV, 62 mV) for 1 min, after which ANG II (108 M) was added to the bath, leading to depolarization. Subsequently, pinacidil (105 M) was added to and then removed from the bath. Data acquired at 10 Hz. Panels are examples of experiments summarized in C. C: data points (n = 8) indicate m averaged during the last 10 s of acquisition during of each period (ANG II, ANG II + pinacidil, ANG II recovery). Repeated measurements were averaged and recorded as a single point per cell (* P < 0.05 vs. ANG II, n = 9). D: luminal diameter of in vitro-perfused DVR was measured before and during exposure to ANG II (108 M). From 5 to 10 min, pinacidil (105 M, n = 8) or vehicle (n = 8) was exchanged into the bath and then washed out. Pinacidil reversibly dilated ANG II-constricted DVR (* P < 0.05 vs. control).

7

8

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Effect of bradykinin (BK) on ANG II-depolarized DVR pericytes. A and B: pericyte m was recorded at baseline (54 mV, 48 mV) for 1 min, after which ANG II (108 M) was added to the bath, leading to depolarization. Subsequently, BK (107 M) was added to the bath and then removed. Data acquired at 10 Hz. Records are examples of experiments summarized in C. C: values on the ordinate show the mean ± SE of m during the final 10 s of each period (* P < 0.01 vs. ANG II, n = 7).

Evidence for voltage-gated Ca²+ entry into DVR pericytes. Having established that DVR vasoreactivity parallels the expected changes in m, we next tested whether the L-type channel blocker diltiazem would induce vasodilation and whether the L-type agonist BayK 8644 would induce constriction. Diltiazem reversibly vasodilated in vitro-perfused DVR that had been preconstricted with 100 mM KCl (Fig. 7A) or 10⁸ M ANG II (Fig. 7B). Also, as expected for functional expression of L-type calcium channels, BayK 8644 constricted DVR, however, compared with ANG II (10⁸ M), the constriction by BayK 8644 (10⁶ M) was mild (Fig. 8).

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Reversal of DVR vasoconstriction by diltiazem. A: luminal diameters of in vitro-perfused DVR were monitored before and during exchange of 100 mM KCl into the bath. Subsequently, vehicle (n = 6), diltiazem (107 M, n = 5), or diltiazem (105 M, n = 5) was added from 5 to 10 min and then removed (* P < 0.05 vs. vehicle). B: luminal diameters of in vitro-perfused DVR were monitored before and during exchange of ANG II (108 M) into the bath. Subsequently, diltiazem (106 M, n = 13) or vehicle (n = 9) was added from 6 to 11 min (* P < 0.05 vs. control).

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Constriction of DVR by BayK 8644. Luminal diameters were monitored before and during addition of BayK 8644 (106 M, n = 8), ANG II (108 M, n = 9), or vehicle (n = 8) to the bath. BayK 8644 was a mild constrictor of in vitro-perfused DVR (* P < 0.05 vs. control).

8

View larger version (23K): [in this window] [in a new window]   Fig. 9.   Effect of diltiazem (Dilt) on ANG II-stimulated pericyte [Ca2+]i. A: [Ca2+]i of fura-2-loaded isolated pericytes was measured before (1 min) and then during (10 min) addition of ANG II (108 M, n = 7) or ANG II + diltiazem (106 M, n = 6). B: [Ca2+]i measured in a fura-2-loaded pericyte preparation exposed to diltiazem (106 M) from t = 11 to 16 min. ANG II had been in the bath since t = 1 min. Data provide an example of experiments summarized in C. C: data summarize 6 experiments that are a continuation of the control group shown in A. Diltiazem (106 M, n = 6) was added to and then removed from the bath of pericytes preexposed to ANG II for 10 min. Diltiazem reduced the plateau phase of the DVR pericyte [Ca2+]i transient (P < 0.01 ANG II/diltiazem vs. ANG II).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

It is generally accepted that m plays an important role in the excitation-contraction coupling of vascular smooth muscle. Depolarization of the plasma membrane reduces the electrochemical driving force for calcium entry into the cytoplasm but activates voltage-gated calcium entry pathways to enhance transmembrane conductance. A diverse array of voltage-gated calcium channels has been identified; however, in the cardiovascular system, L type and T type predominate (11, 29). In past studies, investigators have examined the effect of L-type voltage-gated channel blockers on regional blood flow in the kidney. Greater enhancement of medullary than cortical blood flow was frequently observed (3, 6, 8, 18, 34). Whether this is due to effects at the juxtamedullary afferent arteriole, efferent arteriole, or DVR is uncertain.

The high degree of specificity of L-type calcium channel blockers has been exploited in several studies to establish the role of L channels in the modulation of vasomotor tone of resistance vessels in the kidney. In the isolated perfused kidney, diltiazem and amlodipine inhibited pressure-dependent and ANG II-induced vasoconstriction, respectively (15, 16). In the hydronephrotic kidney preparation, nitrendipine was found to preferentially vasodilate preglomerular vessels (arcuate, interlobular, and afferent arterioles) (7). L-type calcium channel blockers also effectively dilate ANG II-preconstricted afferent arterioles in the in vitro-perfused juxtamedullary nephron preparation (2, 10). The demonstration that ANG II depolarizes afferent arteriolar smooth muscle implies that this constrictor could activate L channels in the afferent circulation (14).

In addition to voltage-gated calcium entry pathways, vasoconstrictors can increase [Ca²⁺]_i via other pathways such as store-operated or receptor-operated channels. To examine functional expression of voltage-gated calcium channels, the effect of KCl-induced depolarization on vasoreactivity and [Ca²⁺]_i has been tested. In the hydronephrotic kidney preparation, 30 mM KCl constricted afferent arterioles to a much greater degree than efferent arterioles, an effect that was blocked by nifedipine (17). Similarly, diltiazem has been found to block KCl-induced vasoconstriction in isolated perfused afferent arterioles (4). Carmines and colleagues (1) showed that KCl depolarization increased [Ca²⁺]_i of afferent but not efferent arteriolar smooth muscle. Recently, the pathways through which ANG II stimulates Ca²⁺ entry into afferent and efferent arteriolar smooth muscle were examined using ratiometric detection of fura-2. ANG II increased [Ca²⁺]_i in both arteriolar segments, but nifedipine inhibited only the afferent response, whereas the receptor-operated Ca²⁺ channel blocker SKF96365 blocked the increase in efferent [Ca²⁺]_i (13). These and other studies established that L-type voltage-gated channels provide a functionally important route for calcium entry into smooth muscle of the afferent arteriole.

The recent study of Hansen and colleagues (9) provides an important new perspective on the channel architecture of juxtamedullary renal resistance vessels. Coexpression of the ₁-subunit for both L- and T-type calcium channels was identified in afferent arterioles and both efferent arterioles and DVR of the juxtamedullary circulation. In contrast, efferent arterioles arising from superficial glomeruli did not express those channel subunits. The authors found that depolarization by KCl could increase [Ca²⁺]_i of both juxtamedullary afferent and efferent smooth muscle. The data in Figs. 7-9 provide similar corroborating evidence for the functional presence of such pathways in DVR pericytes. The incomplete inhibition of ANG II-induced vasoconstriction by diltiazem in Fig. 7 raises the possibility that pathways other than L-type channels, such as T-type or receptor-operated channels, might be present in DVR pericytes.

Despite the small size of DVR pericytes, it has been possible to measure ms and cellular currents in those cells (23). It is also possible to examine [Ca²⁺]_i transients in DVR pericytes after they have been isolated from endothelial cells by stripping. Without prior isolation of the pericytes from the DVR wall, pericyte [Ca²⁺]_i cannot be measured because the adjacent endothelia strongly load fura-2 and obscure the small fluorescent emission that originates from the pericytes (32). In this study we exploited those methods to test the hypothesis that depolarization and voltage-gated [Ca²⁺]_i entry into DVR pericytes accompany vasoreactivity. The nonspecific K⁺ channel blockers BaCl₂ and TEACl depolarize the pericytes and constrict isolated DVR (Figs. 1 and 2). Similarly, elevation of extracellular K⁺ markedly depolarizes pericytes (Fig. 3, A and B) and induces DVR vasoconstriction that is reversed by L-type Ca²⁺ channel blockade (Fig. 7). In this and a prior study, we showed that ANG II depolarizes pericytes from resting levels that lie between 50 and 65 mV toward the equilibrium potential for Cl ion (Figs. 5 and 6; Ref. 23). Despite the ability of K⁺ channel blockade and 100 mM extracellular KCl to depolarize pericytes to a similar degree, they are less effective than ANG II to stimulate DVR vasoconstriction. This finding contrasts with observations in afferent arteriolar smooth muscle, where KCl is highly effective to induce vasoconstriction (1, 4). On the basis of the comparison among ANG II, Ba²⁺, TEA⁺, and KCl (Figs. 1-3), it seems likely that ANG II activates signaling events that are not mimicked by nonspecific depolarization. Downstream effects of ANG II receptor occupancy, such as tyrosine kinase activation and receptor transactivation, may be needed to fully activate the pericyte contractile response (19, 30).

In addition to verifying that depolarization induces vasoconstriction, we tested whether ANG II-induced vasoconstriction could be reversed by agents that repolarize the pericyte cell membrane. Pinacidil hyperpolarized resting pericytes and repolarized ANG II depolarized pericytes to a remarkable degree (Figs. 4 and 5). In resting cells, pinacidil (10⁵ M) reduced m to values that approached the equilibrium potential of K⁺ ion (Fig. 4). After ANG II pretreatment, the average effect of this agent was to repolarize the pericyte to a level that lies below the threshold for activation of either T- or L-type calcium channel activation (Fig. 5C). Pinacidil strongly reversed ANG II-induced vasoconstriction, a finding that is consistent with functional importance of depolarization for ANG II to constrict DVR (Fig. 5D). Given the origin of DVR in the relatively hypoxic renal outer medulla, it is not surprising to have identified a robust effect of K_ATP channel activation in these vessels. Glybenclamide is a blocker of K_ATP channels that is widely used as a hypoglycemic agent in the treatment of diabetes. Two prior studies demonstrated that glybenclamide reduces blood flow to the renal medulla (28, 33), suggesting that K_ATP channel activity exerts tonic vasodilatory effects to preserve medullary blood flow. We previously demonstrated that BK relaxes ANG II-constricted DVR, increases NO production, and increases DVR endothelial [Ca²⁺]_i (24, 27, 31). In this study, we verified that this is accompanied by pericyte repolarization to a degree that could inhibit voltage-gated calcium entry pathways (Fig. 6).

A number of investigators has examined the ability of L-type calcium channel blockers to affect renal medullary blood flow. Infusion of diltiazem into the renal interstitium resulted in enhancement of papillary blood flow (18). Similarly, intravenous infusion of verapamil selectively enhanced medullary blood flow (8). Using single vessel videomicroscopy, Yagil and colleagues (34) found an increase in vasa recta blood flow with low rates of infusion of the dihydropyridine blocker CS-905. The effects of calcium channel blockade on medullary blood flow have also been examined in pathological models. Papillary plasma flow increased when verapamil was infused into the renal artery of dogs subjected to caval constriction (3). Treatment of the spontaneously hypertensive rat with nisoldipine enhanced medullary blood flow and sodium excretion (6). Taken together, an ability of calcium channel blockade to increase renal medullary blood flow seems well established. The present findings, coupled with the recent work of Hanssen and colleagues (9), imply that DVR are a likely a site of action for L-type calcium channel blockers to induce vasodilation.