The intracellular calcium ([Ca2+]i) response of outer medullary descending vasa recta (OMDVR) endothelia to ANG II was examined in fura 2-loaded vessels. Abluminal ANG II (10− 8 M) caused [Ca2+]i to fall in proportion to the resting [Ca2+]i (r =0.82) of the endothelium. ANG II (10− 8 M) also inhibited both phases of the [Ca2+]i response generated by bradykinin (BK, 10− 7 M), 835 ± 201 versus 159 ± 30 nM (peak phase) and 169 ± 26 versus 103 ± 14 nM (plateau phase) (means ± SE). Luminal ANG II reduced BK (10− 7 M)-stimulated plateau [Ca2+]i from 180 ± 40 to 134 ± 22 nM without causing vasoconstriction. Abluminal ANG II added to the bath after luminal application further reduced [Ca2+]i to 113 ± 9 nM and constricted the vessels. After thapsigargin (TG) pretreatment, ANG II (10− 8 M) caused [Ca2+]i to fall from 352 ± 149 to 105 ± 37 nM. This effect occurred at a threshold ANG II concentration of 10− 10 M and was maximal at 10− 8 M. ANG II inhibited both the rate of Ca2+ entry into [Ca2+]i-depleted endothelia and the rate of Mn2+ entry into [Ca2+]i-replete endothelia. In contrast, ANG II raised [Ca2+]i in the medullary thick ascending limb and outer medullary collecting duct, increasing [Ca2+]i from baselines of 99 ± 33 and 53 ± 11 to peaks of 200 ± 47 and 65 ± 11 nM, respectively. We conclude that OMDVR endothelia are unlikely to be the source of ANG II-stimulated NO production in the medulla but that interbundle nephrons might release Ca2+-dependent vasodilators to modulate vasomotor tone in vascular bundles.
- fura 2
blood is supplied to the medulla of the kidney via the efferent flow of juxtamedullary glomeruli (16). Juxtamedullary efferent arterioles branch in the outer stripe of the outer medulla to form outer medullary descending vasa recta (OMDVR), which coalesce in the inner stripe to form vascular bundles. The orientation of OMDVR within vascular bundles is specific. Those on the bundle periphery give rise to the dense capillary plexus that supplies the outer medullary interbundle region and its nephrons. Those in the bundle center traverse the outer medulla to supply the inner medulla. Smooth muscle remnants known as pericytes surround the OMDVR wall, imparting contractile properties to these vessels. It is likely that OMDVR regulates regional perfusion of the outer versus inner medulla of the kidney. For example, preferential constriction of vessels in the center of vascular bundles or, equivalently, preferential dilation of OMDVR on the periphery, would tend to redirect blood flow away from the inner medulla toward the outer medullary interbundle capillary plexus. Such an effect would favor delivery of oxygen and nutrients to the metabolically active medullary thick ascending limb (MTAL) and outer medullary collecting duct (OMCD) (6).
A major role for production of nitric oxide (NO) in the control of medullary blood flow has been proposed based on the observation that nitric oxide synthase (NOS) inhibition leads to reduction of medullary blood flow and hypertension (20) and that a high-sodium diet results in upregulation of NOS isoforms (19). A reciprocal relationship between medullary ANG II and NO production is supported by the observation that interstitial infusion of ANG II leads to increased NO generation as measured by microdialysis with oxyhemoglobin trapping (37). Because OMDVR are the only microvessels with a continuous endothelium in the outer medulla, we tested the hypothesis that ANG II would elevate endothelial intracellular calcium concentration ([Ca2+]i) as required to stimulate NO production by the calcium-sensitive forms of NOS. In contrast with this, we found that ANG II lowers endothelial [Ca2+]i, blocks the elevation of [Ca2+]i induced by treatment with thapsigargin (TG), suppresses calcium signaling by bradykinin (BK), and inhibits divalent cation entry into the endothelia. Based on these observations, we hypothesize that in OMDVR ANG II inhibits [Ca2+]i for the purpose of decreasing production of Ca2+-dependent endothelial vasodilatory autacoids. This would tend restrict control of OMDVR vasomotor tone to vasodilators that diffuse to the vascular bundles from the adjacent surroundings. Given the radial arrangement of OMDVR within vascular bundles, dilation of OMDVR on the vascular bundle periphery preferentially favors distribution of juxtamedullary efferent blood flow to the interbundle region during times of increased metabolic demand.
In vitro microperfusion.
Tissue for microdissection was obtained from 67 Sprague-Dawley rats (80–150 g) (Harlan) anesthetized before nephrectomy by an intraperitoneal bolus of thiopental (50 mg/kg body wt). OMDVR, MTAL, OMCD, and the descending thin limb of long looped nephrons (LDLu) were dissected from the outer medulla as previously described (25). Vessels and nephron segments were mounted on glass pipettes and perfused using standard apparatus (ITM, San Antonio, TX) (22, 23, 25). For OMDVR, microperfusion pipettes were 4–6 μm (inner diameter), holding pipettes 13–15 μm (inner diameter) with constrictions of 6–9 μm, and collection pipettes ∼10 μm inner diameter at the inlet. Pipette diameters were increased as appropriate to accommodate nephrons segments. Vessels and tubules were maintained at 37°C in custom-built microperfusion chambers during microperfusion. The following buffer was used for dissection, bath, and perfusate (in mM): 5 HEPES, 140 NaCl, 10 sodium acetate, 5 KCl, 1.2 MgCl2, 1.71 Na2HPO4, 0.29 NaH2PO4, 1 CaCl2, 5 alanine, 5 glucose, and 0.5 g/dl albumin, pH 7.4.
Videomicroscopy and measurement of vessel diameters.
To quantify vasoconstriction, microperfusion experiments were captured on videotape using a video camera (Panasonic WV-BL90). Experiments were recorded on a Panasonic model AG 1960 videocasette recorder equipped with a microphone for voice recording and played back to enable diameter measurement calipers. As previously described, changes in vessel outer diameter are expressed as percent constriction (22).
Measurement of endothelial intracellular calcium.
OMDVR were loaded with the Ca2+-sensitive fluorescent indicator fura 2 by exposure to bath containing 2 μM fura 2-AM (Molecular Probes, Eugene, OR). At the time the bath was exchanged to contain fura 2-AM, the feedback controller was turned on, gradually warming the vessel to 37°C over about 5 min. Total loading time was 20 min. We have previously shown that fura 2 preferentially loads into endothelial cells rather than pericytes (23). For measurement of [Ca2+]i, fura 2-loaded OMDVR were briefly excited at 340 and 380 nm or at 350 and 380 nm dual-wavelength combinations. The background-subtracted ratio of fluorescent emission (R340/380 or R350/380) was calculated for conversion to the equivalent [Ca2+]i assuming a dissociation constant for fura 2 at 37°C of 224 nM (10). Rmax and Rmin were measured as previously described by exposing vessels to buffer containing 5 mM CaCl2, or 0 CaCl2-0.5 mM EGTA along with 10 μM 4-Br-A-23187 (23).
Either a digital fluorescent imaging system or a photon counting photomultiplier assembly (PMT) was employed to measure fluorescent emission at 510 nm from fura 2-loaded OMDVR endothelia. Light for excitation of fura 2 was provided from a 75-W xenon arc lamp directed through a computer-controlled shutter (Uniblitz, Rochester, NY). The excitation wavelengths were isolated using bandpass filters (Omega Optical, Brattleboro, VT) or with a computer-controlled monochrometer (PTI, South Brunswick, NJ). OMDVR were observed through a 1.3 NA Nikon CF fluor ×40 oil immersion objective. Fluorescent emission was isolated with a 510WB40 filter (Omega Optical) and directed to the intensified charge-coupled device (CCD) camera or PMT (23). Excitation (340 and 380 nm) was used with CCD detection, whereas later experiments employed 350- and 380-nm wavelengths with PMT detection. Output of the CCD camera was captured with a 4-MB GRB silicon video MUX framegrabber (Epix, Northbrook, IL) with a resolution of 8 bits/pixel. When the PMT was employed to measure the fluorescent emission, the output was monitored with a multichannel scaling card (ACE-MCS, EG&G Ortec, Oak Ridge, TN), using custom software or with commercial hardware and software (PTI).
Measurement of Ca2+ influx into OMDVR endothelia using fura 2.
To obtain a measurement of the rate of net Ca2+ influx into OMDVR endothelia, [Ca2+]i was measured at 6-s intervals as vessels were exposed to TG (10− 7 M) and nominally Ca2+-free buffer containing 100 μM EGTA to deplete calcium stores. After the data acquisition software was reset to increase the sampling rate, [Ca2+]i was measured as frequently as possible (approximately every 1.7 s) as the bath was exchanged from zero-Ca2+ EGTA to 2.5 mM CaCl2 without EGTA.
Measurement of Mn2+ influx into OMDVR endothelia using fura 2.
A method for examining the effect of agents on plasmalemmal influx of divalent cations is to exploit the effect of Mn2+ to quench the fluorescence of fura 2 (4). For these protocols, OMDVR were loaded with fura 2 and excited at the calcium-insensitive (isosbestic) wavelength, 360 nm, while fluorescent emission (F360) was monitored at 510 nm with a PMT. After MnCl2 (500 μM) was added to the bath, the rate of decline of F360 provides a measure of plasmalemmal Mn2+ influx through all available pathways.
BK, ANG II, and 4-bromo-A-23187 calcium ionophore (4-Br-A-23187) were purchased from Sigma. These agents were dissolved in water at 10− 3–10− 5M and stored frozen at −20°C. Aliquots were thawed on the day of the experiment, and the excess was discarded at the end of each day. 4-Br-A-23187 was dissolved in ethanol at 10 mM. TG and fura 2-AM ester was purchased from Molecular Probes (Eugene, OR) and stored frozen in anhydrous DMSO. TG was always protected from light.
Experimental results are reported as means ± SE. Statistical comparisons employ paired t-test, unpaired t-test, or repeated-measures ANOVA as appropriate. For ANOVA, significance was determined by the Student-Newman-Keuls test. P < 0.05 was used for significance.
The effect of abluminal application of ANG II (10− 8 M) on endothelial [Ca2+]i in shown in Fig.1. As previously observed, baseline endothelial [Ca2+]i in these vessels generally lies between 50 and 150 nM. Baseline [Ca2+]i was measured for 1 min after which the vessel was exposed to ANG II for an additional 5–10 min. The mean [Ca2+]i for the endothelia of these vessels before ANG II exposure was 89 ± 10 nM. After application of ANG II, [Ca2+]i fell by an average of 46 ± 9 nM (P < 0.05). An experimental record from the OMDVR with the highest resting [Ca2+]i is shown in Fig. 1 A, and the results from 20 vessels are summarized in Fig. 1 B. On average, the change in [Ca2+]i was proportional to the preexisting [Ca2+]i level of the endothelium (n = 20, r = 0.82, P < 0.05). Equivalent results are obtained whether or not the vessel was perfused (n = 13) or simply mounted on pipettes for immobilization (n = 7). To increase the technical ease of obtaining large numbers of observations, and to ensure equal luminal and abluminal concentrations of the agents exchanged into the bath, subsequent experiments were performed without perfusion except where otherwise specified.
Most vessels have a low baseline [Ca2+]i (77 ± 33 nM, means ± SD for all vessels in this report), and the extent of the ANG II induced fall in [Ca2+]i is small (Fig. 1). We next performed a series of experiments to determine whether stimulation by ANG II modulates the increase in [Ca2+]i generated by the endothelium-dependent vasodilator BK. Baseline [Ca2+]i was recorded for 2 min after which BK or BK + ANG II was added to the bath for 10 min. BK ± ANG II was then removed from the bath for 3 min. To quantify the baseline, plateau, and recovery phases of the response, [Ca2+]i concentrations were averaged for five data points before bath exchange or termination of the experiment. A Ca2+ response typical of that observed on abluminal application of BK (10− 7 M) by exchange into the bath is shown in Fig.2 A, and the smaller response typical of BK application in the presence of ANG II (10− 8 M) are shown in Fig. 2 B. The means ± SE of a group comparison (n = 9 each) is shown in Fig. 2 C. In a manner similar to that previously described (23), BK increased [Ca2+]i from ∼70 nM to a peak of 835 ± 201 nM. The peak phase is followed by a sustained and reversible plateau of 169 ± 26 nM in control vessels. ANG II significantly inhibited both the peak (159 ± 30 nM) and plateau (103 ± 14 nM) phases of the BK-induced Ca2+ response. In a separate series of experiments, [Ca2+]i was first raised by BK (10− 7 M, 5 min) after which abluminal ANG II (10− 8 M, 4 min) was applied and then removed (5 min). With this protocol, ANG II again consistently and reversibly reduced the plateau phase of the [Ca2+]i response (n = 10, Fig. 3). Time controls (n = 8) showed no effect.
We next studied the effect of luminal versus abluminal ANG II on Ca2+ signaling and vasomotion (Fig.4). To raise [Ca2+]i, vessels were perfused and bathed with BK (10− 7 M) throughout the experiment. Baseline diameter was recorded, after which [Ca2+]i was measured by fluorescent microscopy with fura 2 as ANG II (10− 8 M) was exchanged into the lumen (5 min). Subsequently, white light images were again recorded by videomicroscopy to determine the effect of luminal ANG II on vessel outer diameter. [Ca2+]i measurements were restarted as ANG II (n = 8) or vehicle (n = 6) was exchanged into the bath (5 min). After this, vessel diameters were recorded a third time to measure the effect of concomitant abluminal ANG II on vasomotion. In these experiments, to perform diameter measurements, vessel images were captured on videotape for 2 min. The total time required to suspend fluorescent data acquisition, redirect white light to the camera, and then restart data acquisition was ∼4–5 min. Luminal ANG II reduced [Ca2+]i, an effect that was enhanced by subsequent, simultaneous addition of ANG II to the bath (Fig. 4 A). In contrast, luminal ANG II alone failed to alter vessel diameter but concomitant abluminal ANG II produced a typical vasoconstrictor response (Fig. 4 B). White light images of OMDVR perfused and bathed in 10− 8 M ANG II are shown in Fig. 5. Time controls with sham exchange of the bath in the second period produced no additional effect on [Ca2+]i(n = 7, data not shown) or vessel diameters (Fig. 4 B). This selective effect of abluminal ANG II to induce vasoconstriction is consistent with the location of pericytes on the vessel periphery. The ability of luminal ANG II to modulate endothelial [Ca2+]i suggests the presence of luminal ANG II receptors on OMDVR endothelia, activation of which modulates Ca2+ signaling.
TG is a sesquiterpene lactone that blocks active uptake of Ca2+ into internal stores by irreversibly binding to the sarcoplasmic-endoplasmic Ca2+ ATPase (34). It is generally accepted that depletion of the stores activates a plasmalemmal pathway for “capacitative” Ca2+ entry, which augments a secondary increase in cytoplasmic [Ca2+]i (26). In a manner consistent with this, an increase in [Ca2+]i of OMDVR endothelia was generally observed on addition of TG to the bath (Fig.6, vehicle). Rarely, vessels failed to respond to TG and were excluded from further experimentation and analysis. In the protocol shown in Fig. 6, endothelial [Ca2+]i was measured for a total of 11 min. After a 1-min baseline observation period, TG (10− 7 M) was exchanged into the bath. Five minutes later ANG II (n = 8) or vehicle (n = 7) was also added to the bath and [Ca2+]i measured for another 5 min. In the presence of TG and vehicle, [Ca2+]i continued to rise in all vessels. In contrast, ANG II reversed the effect of TG by nearly returning endothelial [Ca2+]i to baseline levels. The concentration dependence of ANG II to modulate endothelial [Ca2+]i was also examined by measuring its effects on TG-treated vessels. For this protocol, OMDVR were only tested if the endothelial [Ca2+]i increased and stabilized within 10 min of application of abluminal TG. If [Ca2+]i was stable after 10 min of TG, the experiment was continued by increasing ANG II concentration in the bath by log molar increments from 10− 10 to 10− 7 M at 3-min intervals. A small reduction of [Ca2+]i was observed at 10− 10 M, and the effect to suppress [Ca2+]i was maximal at 10− 8 M (Fig.7).
The remarkable ability of ANG II to lower [Ca2+]i in TG-treated vessels suggests the existence of mechanism(s) that inhibit conductance pathways for plasmalemmal Ca2+ influx. To test this we determined whether the rate of Ca2+ influx into Ca2+-depleted OMDVR was blocked by ANG II. [Ca2+]i was monitored for 2 min in vessels previously exposed to TG (10− 7 M) with or without ANG II (10− 8 M). The bath was changed to nominally Ca2+-free EGTA after which [Ca2+]i fell dramatically (Fig.8 A). After the rate of data acquisition was increased, CaCl2 was replaced in the bath as [Ca2+]i was monitored. The rate of rise of [Ca2+]i was much slower in ANG II-treated OMDVR, suggesting inhibition of Ca2+influx into the endothelia (Fig. 8 B).
Another method for examining the effect of agents on plasmalemmal influx of divalent cations is to exploit the effect of Mn2+to quench the fluorescence of fura 2 (4). As fluorescence was measured, vehicle (n = 12), MnCl2 (500 μM, n = 12), ANG II (10− 8 M, n = 7), or ANG II + MnCl2 (n = 7) was exchanged into the bath. An equal rate of fall of fura 2 fluorescence was observed in control and ANG II-treated OMDVR, likely due to leak of fura 2 from the endothelial cells. In the presence of Mn2+, the rate of decline of fluorescence was markedly enhanced due to quench of fura 2 by Mn2+ (Fig. 9 A). Consistent with inhibition of plasmalemmal Mn2+ influx, the quench rate by Mn2+ was markedly reduced by ANG II (Fig. 9 B).
Because ANG II does not raise endothelial [Ca2+]i of isolated OMDVR but is known to enhance generation of NO in the renal medulla, we reasoned that it might elevate [Ca2+]i in outer medullary nephrons which express calcium-sensitive isoforms of NOS. With this in mind, we tested the ability of ANG II applied from the lumen or bath to increase [Ca2+]i in MTAL, OMCD, and LDLu (Fig. 10). In response to ANG II, [Ca2+]i increased in the MTAL and OMCD from baseline of 99 ± 33 and 53 ± 11 to peaks of 200 ± 47 and 65 ± 11, respectively. The increase in [Ca2+]i achieved significance (P < 0.05) during the time period from 3.25 through 6.58 min in the MTAL and 3.17 through 7.08 min in the OMCD. We recognize that both principal and intercalated cells are present in the OMCD so that the magnitude of the OMCD [Ca2+]iresponse might be underestimated if fura 2 loads into both cell types and ANG II affects only one of them. ANG II had no effect on the LDLu. The effect on MTAL and OMCD [Ca2+]i was observed when ANG II was applied from the bath but not the lumen. The effect of ANG II on short-looped thin descending limbs could not be reliably examined due to the difficulty of dissection and identification of that structure.
Measurement of endothelial [Ca2+]iin isolated OMDVR.
The OMDVR wall is composed of two cell types, pericytes and endothelial cells. We have found that fura 2 loading occurs largely, if not exclusively, into OMDVR endothelia. We have previously provided high-resolution white light and fluorescent images of fura 2-loaded vessels that show no fluorescent emission from pericytes (23). We also found that vasopressin constricts OMDVR (36) but fails to raise [Ca2+]i in fura 2-loaded vessels (23). Vasoconstriction via the vasopressin V1 receptor is mediated by the inositol trisphosphate (IP3) Ca2+ signaling pathway, reinforcing the notion that fura 2 is not present to detect [Ca2+]ichanges in pericyte cytoplasm. Initially, motivated by the desire to localize the fura 2 signal by digital imaging, we quantified [Ca2+]i by measuring ratio images with an intensified CCD camera (23). After determining that the fluorescent signal is endothelial in origin, we abandoned this approach in favor of photon-counting PMT detection. The latter is more accurate because image analysis in our system is limited to 8-bit resolution (0–28 gray scale), whereas photon counting PMT detection is a 16-bit operation. The greater sensitivity of PMT detection also enables reduction of excitation intensity, limiting the risk of photobleaching. Our investigations with either detection method have yielded similar results (Fig. 1).
Suppression of Ca2+signaling in OMDVR endothelia.
We have shown that treatment with ANG II reduces basal [Ca2+]i in OMDVR endothelium (Fig.1), peak and plateau [Ca2+]iresponses in BK-stimulated vessels (Figs. 2-4), and endothelial [Ca2+]i in TG-treated OMDVR (Figs.6 and 7). If [Ca2+]i in the cytoplasm is at a steady level, the rate of plasmalemmal influx must equal the rate of Ca2+ extrusion. ANG II could exert its actions on OMDVR endothelia by influencing the set point of either or both of these transport pathways. As part of this study we tested the hypothesis that plasmalemmal influx might be inhibited by ANG II. Under conditions that favor intracellular Ca2+ store depletion (TG, zero bath CaCl2, and external EGTA), ANG II reduced the rate of increase of [Ca2+]iwhen CaCl2 was replaced in the bath (Fig. 8). The precise route involved in this process is unknown, but the conditions of the experiment favor activation of capacitative Ca2+ entry pathway(s) resulting from intracellular store depletion (26). Under Ca2+-replete conditions, the rate of influx of Mn2+ was also reduced by ANG II (Fig. 9). This divalent cation can traverse nonselective cation channels in some endothelia, suggesting modulation of some similar pathway(s) that conduct divalent cations in OMDVR. Another possibility is that ANG II lowers the electrochemical driving forces for Ca2+influx by depolarizing the endothelia. In this case, modulation of K+ or Cl− channels might account for the actions of ANG II (1).
The principal finding of this study, that [Ca2+]i signaling can be inhibited by ANG II, is unusual but not without precedent. ANG II exerts its actions through type 1 (AT1a and AT1b) and type 2 (AT2) receptors. AT1a and AT1breceptors exhibit 95% amino acid sequence homology and are seven-membrane spanning G protein-coupled receptors. The majority of literature, based primarily on responses in smooth muscle, indicates that AT1 receptors act through phospholipase C (PLC)-mediated IP3 generation and [Ca2+]i signaling (3, 9). Specific reports of AT1-mediated signaling in endothelial cells also exist (27). AT1-mediated inhibition and activation of cyclic nucleotide formation has also been described (14, 32). Phosphorylation events have been shown to be important in AT1 receptor-mediated signaling. The intracellular carboxyl terminus of AT1 has multiple phosphorylation sites that appear to modulate desensitization, downregulation, and internalization of the receptor (29). In addition to acting as a substrate for phosphorylation, AT1 receptor stimulation leads to phosphorylation of various intracellular proteins. For example, the Ca2+-sensitive isoform of PLC (PLC-β) is lacking in many cell types in which ANG II increases IP3. Instead, ANG II stimulation activates the Ca2+-insensitive PLC-γ1 isoform by tyrosine phosphorylation. Interestingly, given the results of this study, PLC-γ1 generation of IP3 can sometimes be blocked through tyrosine kinase inhibition (2, 7, 8, 18). cGMP-dependent protein kinase (cGK) activation has been shown to block [Ca2+]i responses of various endothelia. The cGK1 isoform is known to phosphorylate and inactivate the IP3 receptor, inhibiting Ca2+ release, and to phosphorylate phospholamban, leading to enhancement of Ca2+ compartmentalization (5, 17).
Effect of ANG II on renal blood flow and NO production.
The effect of ANG II on perfusion of the renal medulla has been the subject of many investigations. The majority favor tonic constriction of the juxtamedullary microcirculation (24, 28). A reciprocal relationship between ANG II stimulation and intrarenal NO production has been observed by several investigators. Ito and colleagues (12) showed that NOS inhibition enhances the effects of ANG II in the afferent but not the efferent arteriole of the rabbit. Similarly, studies with NO trapping by oxyhemoglobin microdialysis have shown that infusion of ANG II into the medullary interstitium leads to an increase in global NO generation implying a possible role for NO to abrogate ANG II vasoconstriction of juxtamedullary resistance vessels (37). Direct evidence for stimulation of NO production was provided by Thorup et al. (35) who showed that renal cortical microvessels generate NO when stimulated with ANG II.
It is known that perfusion of the renal medulla is under the influence of kinins and is highly sensitive to NO. Chronic inhibition of NO production reduces medullary blood flow and induces hypertension in rats (20). Enhancement of NO production in the spontaneously hypertensive rat by l-arginine infusion normalized renal medullary hemodynamics (15). ANG AT1 receptors and the Ca2+-sensitive forms of NOS (NOS I and NOS III) have been localized to thick ascending limb, vasa recta, and collecting ducts (11, 13). Because the major endothelial isoform of NOS (eNOS or NOS III) is stimulated by elevation of [Ca2+]i, we reasoned that NO generation might self-limit ANG II-induced OMDVR constriction by raising endothelial [Ca2+]i to activate NOS III. In contrast to this, we found that ANG II reduces OMDVR endothelial [Ca2+]i and inhibits Ca2+ signaling by the endothelium-dependent vasodilator bradykinin. ANG II does, however, raise [Ca2+]i in OMCD and MTAL (Fig. 10). In the absence of specific hormonal stimulation, [Ca2+]i levels found in isolated OMDVR are in the range of 50 to 100 nM, concentrations which are probably too low to activate Ca2+/calmodulin-sensitive forms of NOS. In contrast, steep dependence of NOS is observed as [Ca2+]i is increased between 100 and 1,000 nM (30). BK elevates OMDVR endothelial [Ca2+]i to plateau values between 150 and 400 nM (Figs. 2-4), levels that are expected to stimulate NO generation. Siragy et al. (33) found that blockade of ANG II receptors led to reduction of renal interstitial BK concentrations in dogs. The ability of ANG II to modulate BK-stimulated OMDVR endothelial [Ca2+]i levels provides additional insight into possible mechanisms by which these hormones interact to influence intrarenal hemodynamics.
ANG II concentrations and possible effects on medullary blood flow.
We previously observed that ANG II constricts OMDVR with threshold and one-half maximal effects at ∼10− 11 M and 5 × 10− 10 M, respectively. Interestingly, the threshold and one-half maximal effects of ANG II on OMDVR endothelial [Ca2+]i occur at ∼10-fold higher concentration (Fig. 7). Circulating ANG II concentrations are much lower than this, typically ∼10− 11 M (21, 31). OMDVR originate from the branching of the efferent arterioles of juxtamedullary glomeruli and local ANG II concentrations within the kidney can be quite high. ANG II concentrations in proximal tubule fluid and star vessel plasma downstream of superficial glomeruli can exceed circulating levels by as much as 1,000-fold (31). Juxtamedullary nephrons are inaccessible to micropuncture; however, renin content of juxtamedullary arterioles is probably high (24). Taken together, it would seem that concentrations of ANG II within vascular bundles in vivo as high as 10− 9–10− 8M cannot be ruled out.
On the basis of the observations in this study we hypothesize that high concentrations of ANG II inhibit endothelial production of vasodilators (e.g., NO and prostacyclin) to favor more intense OMDVR vasoconstriction. This concept is at odds with the notion that ANG II typically activates compensatory production of endothelial vasodilators as a self-limiting action. It should be noted, however, that OMDVR are arranged within vascular bundles with a specific radial arrangement. OMDVR on the bundle periphery give rise to the capillary plexus that perfuses the interbundle region while those near the bundle center traverse the inner stripe to perfuse the inner medulla. It is inviting to speculate that ANG II-induced Ca2+ activation of NOS I, NOS III, and phospholipase A2 within the cytoplasm of MTAL and OMCD might lead to production of eicosanoids and NO that diffuse to the vascular bundle periphery to preferentially dilate the OMDVR that supply them with oxygen and nutrients. Such an action would be expected to abrogate ischemia of the more metabolic, actively transporting nephron segments by redistributing blood flow from the inner medulla to the outer medulla of the kidney. Preferential reduction of blood flow to the inner medulla by ANG II would still explain the observations of investigators who have employed videomicroscopy or laser-Doppler measurements on the exposed papilla (24, 28).
This work was supported by National Institutes of Health Grants DK-42495 and HL-62220.
Address for reprint requests and other correspondence: T. L. Pallone, Division of Nephrology N3W143, Univ. of Maryland at Baltimore, 22 S. Greene St., Baltimore, MD 21201-1595.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society