INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DESCENDING VASA RECTA (DVR) are small generation resistance vessels that carry blood from the cortex to the medulla of the kidney. DVR originate as branches of the juxtamedullary efferent arteriole and traverse the outer medulla in vascular bundles. The DVR that supply the inner and outer medulla are radially arranged within the bundles from center to periphery, respectively. This parallel arrangement strongly suggests that the DVR distribute and regulate regional blood flow within the medulla (17, 26). DVR are enveloped by small contractile cells called pericytes (29) and are lined by a continuous endothelium. The importance of endothelial secretion of paracrine mediators to modulate the contractility of adjacent smooth muscle is well established. Motivated by the importance of nitric oxide (NO) in the maintenance of medullary perfusion and blood pressure (9, 16, 18-21, 24, 37, 38), we adapted two methods for measuring NO production by DVR that were isolated and dissected from vascular bundles of the outer medulla of the kidney. First, the probe 4,5-diaminofluoroscein (DAF-2), which increases fluorescence with covalent modification by NO, was used to monitor NO generation (5, 13). To corroborate the findings with DAF-2, we employed a second microincubation scheme to monitor the oxidation of the Fe²⁺ (ferrous) to Fe³⁺ (ferric) hemoglobin (Hb²⁺ and Hb³⁺, respectively) by NO (1, 23). The results verified that isolated DVR generate NO and that NO production is increased by supplying L-arginine or stimulating with the endothelium-dependent vasodilator bradykinin (BK). Reduction of superoxide anion with the cell-permeant superoxide dismutase mimetic 4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl (tempol) also increased NO in isolated DVR.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Isolation of outer medullary descending vasa recta. Kidneys were harvested from Sprague-Dawley rats (70-150 g body wt; Harlan), sliced, placed in buffer, and maintained on ice at 0-4°C. Outer medullary DVR (OMDVR) were dissected from vascular bundles and transferred to the stage of an inverted microscope as previously described (25-28). The buffer used for dissection and the bath in these studies included (in mM) 140 NaCl, 10 NaC₂H₃O₂, 5 KCl, 1.2 MgCl₂, 2 Na₂HPO₄/NaH₂PO₄, 1 CaCl₂, 5 alanine, 5 glucose, and 5 HEPES and 0.5 g/dl of albumin (pH = 7.4).

Fluorescent detection of NO with diaminofluorescein diacetate probe. The nonfluorescent molecule DAF-2 diacetate (DAF-2DA) becomes weakly fluorescent when loaded into cells via deesterification to form DAF-2. DAF-2 further increases in fluorescence when it is covalently modified by NO to yield a triazofluorescein (5, 13). Thus the fluorescent signal from DAF-2 produces an integrated measure of local NO concentration ([NO]) within the loaded cells. DAF-2 was excited at 485 nm using a xenon arc lamp (Photon Technology International; Lawrenceville, NJ). Fluorescent emission was isolated with a band-pass filter at 530 nm (Omega Optical; Brattleboro, VT) and measured with a photon-counting detection assembly (D104B; Photon Technology International). Each OMDVR was transferred to the imaging chamber, where the ends were aspirated into a pair of holding pipettes that had openings of 10-15 µm. The pipettes and vessel were positioned using micromanipulators (Instruments Technology and Machinery; San Antonio, TX) mounted on an inverted microscope (Nikon Diaphot). The OMDVR were positioned near a thermocouple in the narrow inlet region of a custom-built chamber. Temperature was maintained with a feedback system (CN9000A; Omega Engineering; Bridgeport, NJ). The vessel was warmed to 37°C and loaded for 20 min with 10 µM DAF-2DA ester. DAF-2 loads into both endothelia and pericytes (see Fig. 1, A-C) and increases emission at 530 nm during excitation between 450 and 500 nm without a spectral shift (see Fig. 1D).

View larger version (43K): [in this window] [in a new window]   Fig. 1.   Loading of the fluorescent dye 4,5-diaminofluoroscein (DAF-2) into isolated outer medullary descending vasa recta (OMDVR). A-C: paired white light and fluorescent images show that DAF-2 loads diffusely into the cytoplasm of endothelial cells as well as pericytes. D: normalized fluorescent emission from DAF-2 at 530 nm versus excitation wavelength at 1-min intervals after introduction of 2.5 mM sodium nitroprusside (SNP) into the bath. In OMDVR, DAF-2 increased in fluorescence but did not undergo a spectral shift.

Measurement of NO generation with oxyhemoglobin. As a second method to verify NO generation by isolated OMDVR, we employed the property of NO to oxidize Hb from the Hb²⁺ form to the Hb³⁺ form (1, 23). When this oxidation occurs, the absorption spectrum of Hb shifts, thereby permitting a measure of the extent of the oxidation. A microincubation scheme was devised to take advantage of this property. To serve as a cuvette for monitoring the conversion of Hb²⁺ by isolated OMDVR, a standard microperfusion-style holding pipette was created but markedly enlarged to ~2 mm in length and ~75 µm in diameter. The cuvette tip was heat polished to reduce the diameter of the entrance to ~50 µm. Light was passed from a fiber-optic filament through the cuvette as shown in Fig. 2. The 250-µm fiber-optic filament (Edmond Scientific; Barrington, NJ) was modified by drawing it out under heat on a microforge (Stoelting; Wood Dale, IL) and cutting the taper with a razor blade to obtain a tip diameter of ~50 µm. The end of a length of stainless steel music wire (Small Parts; Miami Lakes, FL) was then polished to a mirror finish with 0.3-µm grit sandpaper (Thomas Scientific; Swedesboro, NJ) and held at a 30° angle to reflect the transmitted light to the microscope objective. The light collected by the objective was measured with a photon-counting photomultiplier detection assembly (Photon Technology International) attached to the side port of the microscope. A coupling was machined from aluminum that permitted insertion of the far end of the filament into the exit slit of a computer-controlled monochrometer (Photon Technology International). Thus the wavelength of light for the measurement of transmittance could be continuously varied as the transmission through the microcuvette was monitored. In place of a standard perfusion pipette, the fiber-optic filament was fixed to a standard acrylic perfusion-pipette holding assembly (Instruments Technology Machinery; San Antonio, TX) so that it could be advanced in and out of the constriction by turning a screw. The filament tip was withdrawn as needed to allow aspiration of buffer containing Hb or several OMDVR into the cuvette area. When replaced into the constriction, the fiber-optic tip provided a good seal that prevented movement of buffer in and out of the cuvette.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Microincubation for measurement of nitric oxide (NO) production with hemoglobin (Hb). A cuvette for measurement of light transmittance was constructed on the end of a microperfusion-style holding pipette. Nominal dimensions of the cuvette: diameter, 75 µm; length, 2 mm; opening and constriction, 50 µm. Isolated OMDVR were drawn into the cuvette area with unreacted Fe2+ Hb (Hb2+; 5 µM). The conversion of Hb2+ to Fe3+ Hb (Hb3+) by NO was measured as the change in transmittance at 420 nm. Light was reflected from the end of a mirror-polished stainless steel wire to a microscope objective and then to a photomultiplier assembly.

Measurement of endothelial intracellular Ca²+. The OMDVR were loaded with the Ca²⁺-sensitive fluorescent indicator fura 2 by exposure to a bath containing 2 µM fura 2-acetoxymethyl ester (Molecular Probes; Eugene, OR) for 20 min (25, 27). We have previously shown (25) that fura 2 preferentially loads into endothelial cells rather than pericytes. For measurement of intracellular Ca²⁺ concentration ([Ca²⁺]_i), fura 2-loaded OMDVR were excited using 350- and 380-nm dual-wavelength combinations. The background-subtracted ratio of fluorescent emission (R_350/380) was calculated for conversion to the equivalent [Ca²⁺]_i assuming a dissociation constant for fura 2 at 37°C of 224 nM. The ratio when fura 2 is completely calcium bound (R_max) and the ratio in calcium-free solution (R_min) were measured as previously described by exposing vessels to buffer containing 5 mM CaCl₂ or 0 CaCl₂ and 0.5 mM EGTA, respectively, along with 10 µM 4-bromo-A-23187, a Ca²⁺ ionophore (25, 27).

Videomicroscopy and measurement of vessel diameters. To quantify vasoconstriction, microperfusion experiments were captured on videotape using a video camera (WV-BL90; Panasonic). Experiments were recorded on a video cassette recorder (AG-1960; Panasonic) equipped with a microphone for voice recording and were played back to enable diameter measurement via calipers. As previously described, changes in vessel outer diameter are expressed as percent constriction (25, 27, 36).

Reagents. DAF-2 DA (5 mM in DMSO) was purchased from Calbiochem (La Jolla, CA) and stored at 20°C; it was diluted to 10 µM to load OMDVR. Sodium nitroprusside (SNP), BK, N-nitro-L-arginine methyl ester (L-NAME), L-arginine, and tempol were purchased from Sigma. SNP was dissolved just before experimentation and was protected from light. BK, L-NAME, and L-arginine were dissolved in water and stored in 1 × 10⁴ M, 100 mM, and 100 mM concentrations, respectively, at 20°C and in 100-µl aliquots. Stock frozen solutions were thawed on the day of the experiment, and the excess was discarded. In a manner similar to that described by Zou and colleagues (37, 38), human Hb²⁺ was purchased from Sigma as a lyophilized powder and was dissolved in buffer on the day of experimentation.

Statistics. Data are shown as means ± SE. Statistical testing employed Student's t-test (paired or unpaired, as appropriate) and ANOVA. With ANOVA, the Student-Newman-Keuls test was used to examine significance.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Reaction of DAF-2 with exogenous NO. In the first series of experiments, we verified the ability of DAF-2 (loaded into isolated OMDVR) to increase fluorescence during exogenous NO production by SNP. As shown in Fig. 3 (left), OMDVR loaded with DAF-2 increased background-subtracted fluorescence by ~30% over 10 min. Time controls showed a slow decline in fluorescence. The decline in signal emitted from the controls cannot be interpreted as a lack of NO generation but does imply that the rate of loss of the dye due to leakage exceeded the rate of fluorescence increase due to basal rates of NO generation.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Reaction of DAF-2 with NO from SNP. OMDVR were loaded with DAF-2. Buffer containing SNP (0.95 mM) was exchanged into the bath as DAF-2 fluorescence (F) was monitored. Left: data are shown as means ± SE of F normalized to time 0 (F0). Right: summary and statistical comparison of the differences between SNP and controls are shown. Data from the final minute of observation have been normalized to the average of the controls. Number of vessels tested is shown in parentheses. *P < 0.05 controls vs. SNP.

View larger version (39K): [in this window] [in a new window]   Fig. 4.   Conversion of Hb2+ to Hb3+ in the microcuvette by SNP. A: transmittance as a function of wavelength is shown for Hb concentrations of 0-5 µM. Spectral shift with isosbestic point at 411 nm is observed. B: same data as in A, converted to units of absorbance. C: difference between absorbance of Hb2+ and Hb3+ at various wavelengths is a linear function of Hb concentration. Maximal changes are observed at 400 and 420 nm. D: transmittance (T) is shown as a function of time normalized to the value at time 0 (T0). Hb2+ solution containing SNP was drawn into the cuvette and the oxidation process was monitored at 420 nm. Hgb, hemoglobin; [Hgb], hemoglobin concentration.

Endogenous production of NO: effects of L-arginine. Measurement of endogenous NO production in DAF-2-loaded vessels proved feasible but required longer observation periods than were employed with SNP (see Fig. 3). We first asked whether the supply of the NO synthase (NOS) substrate L-arginine affected basal NO generation in this isolated vessel preparation. Overall DAF-2 fluorescence declined with time in control vessels that were not supplied with L-arginine. As illustrated (see Fig. 5, left) and summarized (see Fig. 5, right), addition of 0.1 mM L-arginine tended to enhance the DAF-2 signal compared with controls, but the effect did not achieve significance (P = 0.08). Addition of 1 mM L-arginine yielded a slightly increasing signal that achieved significance compared with the controls (P < 0.05; Fig. 5, right).

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Effect of L-arginine (L-Arg) bath on NO generation by isolated OMDVR. Left: normalized change in DAF-2 fluorescence as a function of time after introduction of L-arginine into the bath at 0, 0.1, or 1.0 mM concentrations. Right: comparison of the normalized average of the last minute of data (means ± SE). Number of vessels tested is shown in parentheses. *P < 0.05 vs. L-arginine (0 mM).

Endogenous production of NO with BK. To test the ability of the endothelium-dependent vasodilator BK to stimulate NO production, vessels loaded with DAF-2 were exposed to a bath containing 0.1 mM L-arginine with or without BK (1 × 10⁷ M). BK resulted in an enhancement of DAF-2 fluorescence compared with controls, but a relatively large number of observations were required to achieve significance (see Fig. 6, A and B). NO production was also examined at lower BK concentrations (1 × 10¹⁰ and 5 × 10⁹ M; n = 7 each). DAF-2 fluorescence was 99 and 102% of control, respectively, neither of which achieved significance. We also verified that L-NAME (1 × 10⁴ M) inhibits BK-stimulated NO production (see Fig. 6, C and D). The results during continuous excitation were similar to those obtained when excitation light was blocked with a shutter during most of the experiment, which verified that dye leakage rather than photobleaching accounted for the slow decline of the signal (see Fig. 6, E and F).

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Effect of bradykinin (BK) on NO generation by isolated OMDVR. A: normalized change in DAF-2 fluorescence is shown as a function of time after introduction into the bath of L-arginine (0.1 mM) with or without BK (1 × 107 M). B: comparison of the normalized average of the last minute of data from A (means ± SE). C: normalized change in DAF-2 fluorescence is shown as a function of time after introduction of BK (1 × 107 M) + L-arginine (0.1 mM) with or without N-nitro-L-arginine methyl ester (L-NAME; 1 × 104 M) into the bath. *P < 0.05 with vs. without BK. D: comparison of the normalized average of the last minute of data from C is provided (means ± SE). E and F: experiment shown in C and D was repeated except that excitation of DAF-2 was blocked with a shutter. *P < 0.05 with vs. without L-NAME. Number of vessels tested is shown in parentheses.

4

7

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Effect of BK on intracellular Ca2+ concentration ([Ca2+]i) in isolated OMDVR. The [Ca2+]i response of fura 2-loaded OMDVR is shown as BK (1 × 107 M) is introduced into the bath at time 0. Data are means ± SE of 9 vessels.

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Effect of L-NAME on BK-stimulated NO production, measured by oxidation of Hb2+. T is shown as a function of time normalized to the value at time 0. Hb2+ solution containing BK (1 × 107 M) was drawn into the cuvette with or without L-NAME (1 × 104 M), and the oxidation was process monitored at 420 nm. Data are means ± SE. *P < 0.05 with vs. without L-NAME.

Effect of tempol on [NO] within isolated OMDVR. Because oxygen free radicals are produced by a variety of endogenous enzymes and consume NO, we tested the hypothesis that local [NO] within cells of OMDVR would increase in the presence of a superoxide dismutase mimetic. Tempol was chosen because it is cell permeant, and it functions as a superoxide dismutase mimetic (10, 30). Inclusion of tempol (1 mM) in the bath tended to enhance DAF-2 fluorescence compared with controls, but the effect did not achieve significance (data not shown; P = 0.08; n = 6 and 9 for controls and tempol, respectively). When BK (1 × 10⁷ M) was also used to stimulate NO production, significant increases in DAF-2 fluorescence were readily observed with tempol compared with unstimulated (P < 0.01) or BK-stimulated (P < 0.05) vessels (see Fig. 9). A series of experiments were also performed to determine whether the addition of tempol (1 mM) would have an effect on the ability of SNP (0.95 mM) to increase the fluorescence of DAF-2 loaded into OMDVR. With tempol, fluorescence increased by 12.2 ± 1.3% in 10 min (n = 7). In the absence of tempol, fluorescence increased by 7.7 ± 1.7% in 10 min (n = 7; P = 0.06 vs. tempol). We also tested the ability of tempol (1 mM) to dilate OMDVR that had been preconstricted by ANG II (1 × 10⁸ M). After we obtained baseline diameter measurements, ANG II was added to the bath for 5 min. Thereafter, either tempol or vehicle was exchanged into the bath for 5 min and then washed out for 5 min. Tempol dilated preconstricted vessels whether or not L-arginine (1 × 10⁴ M) was included in the bath (see Fig. 10).

View larger version (28K): [in this window] [in a new window]   Fig. 9.   Effect of 4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl (tempol) on BK-stimulated NO generation of isolated OMDVR. Left: normalized change in DAF-2 fluorescence as a function of time after introduction of L-arginine (0.1 mM) alone (control), L-arginine + BK (1 × 107 M), or L-arginine + BK (1 × 107 M) + tempol (1 mM). Right: comparison of the normalized average of the last minute of data (means ± SE). Number of vessels tested is shown in parentheses. *P < 0.05 BK with vs. without tempol; **P < 0.01 BK + tempol vs. control.

View larger version (21K): [in this window] [in a new window]   Fig. 10.   Effect of tempol on ANG II-induced vasoconstriction of OMDVR. Isolated microperfused OMDVR were constricted with ANG II (1 × 108 M). Percent constriction is shown at the end of three sequential 5-min periods in which the bath was sham exchanged (left), exchanged to introduce and then remove 1 mM tempol (middle), or exchanged to introduce and remove tempol without L-arginine present in the bath (right). *P < 0.05. NS, not significant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Substantial evidence has been accumulated to show that perfusion of the renal medulla is sensitive to blockade of NO synthesis. Systemic infusion of L-arginine corrects abnormal pressure natriuresis and medullary autoregulation in spontaneously hypertensive rats (16, 18). Intravenous infusion of L-NAME into conscious rats induces hypertension, sodium retention, and reduction of medullary blood flow yet spares perfusion of the renal cortex (24). Selective infusion of L-NAME into the interstitium of the kidney of uninephrectomized rats also caused both hypertension and the selective reduction of medullary blood flow (20, 21). Hypertension was induced by the inhibition of neural NOS (nNOS) expression via interstitial infusion of antisense oligonucleotides or by inhibition of nNOS activity via interstitial infusion of a selective inhibitor (19). Recent experiments with the juxtamedullary nephron technique (12) confirmed roles for nNOS in affecting responsiveness of the afferent arteriole (due to alteration of tubuloglomerular feedback) and in affecting efferent arteriolar constriction (via intrinsic effects).

The study of regulation by NO most often depends on observation of the biological effects of nonspecific or isoform-specific NOS blockade. NO generation at the level of single cells or groups of cells is difficult to measure. Such measurements have been performed in isolated microvessels and nephron segments with microelectrodes (33) and through the determination of the conversion of L-[³H]arginine to citrulline by NOS (35). Isolated OMDVR are very small microvessels (diameter, 15 µm; length, 500 µm), respectively. As such, detection of NO generation presents a similar challenge. In this study, we investigated the feasibility of using the fluorescent probe DAF-2 for this purpose. The dye successfully detected exogenous NO donation by SNP (see Fig. 3) and endogenous NO production by OMDVR (see Figs. 5, 6, and 9). The greatest limitation was the tendency of DAF-2 to leak from the vessels after being trapped by deesterification. Our experience with losses of fura 2 after loading of the OMDVR with the ester was similar (25, 27). Fura 2 and similar probes are more forgiving, because dual-wavelength ratios compensate for and hide the effects of losses due to leakage. DAF-2 fluorescence is measured at a single wavelength so that leakage is directly superimposed on the data records. A competition between loss of signal due to leak and enhancement of fluorescence by NO determines the maximum observation time and therefore our overall ability to discern the effects of an experimental perturbation. Also, the reaction of DAF-2 with NO produced an irreversible covalent modification; therefore, repeated observations on a single vessel could not be readily performed. In summary, to measure the event of interest, sufficient NO modification of the dye must occur on a time scale that is shorter than the rate of loss due to leakage, and only group comparisons are possible. Despite this difficulty, we found that properly designed studies with DAF-2 can yield useful measurement of the modulation of NO generation in isolated OMDVR.

To provide independent verification of NO generation by another method, we also adapted a microincubation scheme that allows conversion of Hb²⁺ to Hb³⁺ by NO released from isolated vessels (see Fig. 2). In this approach, the electron released by Hb²⁺ to NO during the degradation causes a shift in the absorption spectrum that is most efficiently monitored at either 400 or 420 nm (see Fig. 4) or by taking the difference between the transmittance values at these two wavelengths (1, 23). Once again the method is feasible but difficult. The greatest limitation is the simple inability to obtain stable placement of the vessel(s) within the pipette holding area/cuvette. Exchange of the volume of Hb²⁺ in the cuvette generally dislodges or alters the location of the vessel so that repeated paired comparisons cannot be meaningfully performed. The effects of Hb on oxygen tension within the cuvette area are unknown and cannot be readily controlled. In view of this, there is also concern that the rapid local consumption of NO by Hb²⁺ might enhance activation of NOS independent of the experimental perturbations. Theoretically, variations in NO production due to manipulations that occur more rapidly than the time required for diffusion of NO across the cuvette might not be accurately measured, because diffusion away from the vessel would dominate over NO production as the rate-limiting step. This is unlikely to be a problem, because NO diffusivity (D) is high. For D = 3,300 µm²/s at 37°C in water (15) and a cuvette of diameter d = 50-100 µm, it is expected that a characteristic diffusion time of ~d²/D = 3 s would exist. Thus events occurring on a scale of minutes can be measured. In keeping with this notion, we were able to document the blockade of NO generation in BK-stimulated OMDVR (see Fig. 8). In general, measurements with DAF-2 are easier to perform and do not suffer from the concern over Hb activation. Our extensive use of DAF-2 in this study reflects our preference for that method. The possibility of extending either of these methods to the study of NO generation in isolated nephron segments has not been explored but does seem inviting.

Intracellular L-arginine concentrations are reported (2) to be in the range of 100-800 µM, which are values that are much greater than the ~10 µM substrate requirement of endothelial NOS. Based on this consideration, it is not anticipated that high concentrations of L-arginine would affect the rate of NO generation. In contrast, increasing the extracellular concentration of L-arginine has been repeatedly shown (6) to enhance NO generation; this finding has been dubbed the "arginine paradox" (14). In a previous study (25), we showed that BK-induced vasodilation could be augmented by including L-arginine in the bath of microperfused OMDVR. The results illustrated in Fig. 5 support this finding and verify that NO generation by isolated OMDVR can be enhanced by L-arginine. It has been proposed (14) that L-arginine might reside within intracellular pools sequestered from NOS and that NOS isoforms could be spatially linked to membrane L-arginine transporters. The finding that L-arginine infusion can modulate medullary perfusion and pressure natriuresis (16) coupled with the existence of L-arginine-facilitated transport systems in the medulla (8) raise the possibility that specific recycling mechanisms serve to regulate vasomotion of DVR within vascular bundles through local alterations of interstitial L-arginine concentration.

It is generally accepted that NO is a short-lived molecule, the effects of which are exerted within several micrometers of the site of synthesis. Local levels of superoxide anion regulate the availability of NO by reacting with NO to produce peroxynitrite (3, 7). This process has been implicated in exaggerated vasoactivity in spontaneously hypertensive rats (31) and has been shown to affect vasoconstriction of the afferent arteriole in rabbits (32). In support of the possibility that [NO] within the OMDVR wall is also modulated by superoxide, we found that treatment with the cell-permeant superoxide dismutase mimetic tempol substantially augmented the detection of NO during BK stimulation (11, 31, 32) (see Fig. 9). Tempol also increased DAF-2 fluorescence when SNP was the NO donor (P = 0.06). The latter finding is consistent with the hypothesis that tempol reduces the consumption of SNP-donated NO by superoxide anion; however, a specific effect on the release of NO from SNP by NADPH oxidase cannot be ruled out (21). The functional significance of the tempol effect was also investigated in ANG II-constricted OMDVR (see Fig. 10); tempol vasodilated the preconstricted vessels whether or not L-arginine was included in the bath.

We previously observed a tendency for ANG II to reduce endothelial [Ca²⁺]_i in fura 2-loaded OMDVR (27). The ability of tempol to vasodilate OMDVR, presumably through enhancement of [NO], implies that there is ongoing NO production in the presence of ANG II. Another possibility is that some NOS isoform is expressed within the pericytes where ANG II almost certainly increases [Ca²⁺]_i during vasoconstriction. We previously observed (36) that nonspecific NOS inhibitors, when applied alone, constrict OMDVR. That finding supports the functionally significant NO production in unstimulated vessels.

In the absence of Hb and oxygen, the half-life of NO can be quite long (3). A competing effect in this scenario is the requirement of oxygen as a substrate for NOS during NO generation from L-arginine (34). Whether local NO levels would tend to increase during hypoxia due to reduced destruction by superoxide anion or decrease due to reduced NO generation resulting from oxygen-substrate limitation is uncertain. The net result might depend locally on the expression and behavior of enzymes such as NADPH oxidase that generate reactive oxygen species (7). The potential interactions are clearly complex, however, given the low PO₂ (4) and the low Hb concentration of blood within the renal medulla (26). The possibility can be entertained that NO might accumulate to high levels and diffuse far enough to have distant effects.