INTRODUCTION

TOP ABSTRACT INTRODUCTION PRODUCTION OF NO BY... ROLE OF NO IN... ROLE OF NO IN... PERSPECTIVES SUMMARY REFERENCES

IN THE LAST FEW YEARS, a series of in vivo and in vitro studies have begun to reveal a close relationship between nitric oxide (NO) and the proximal tubule and a significant role of NO in proximal tubule physiology and pathophysiology. The proximal tubule, reabsorbing approximately two-thirds of the Na⁺ and water filtered by the glomerulus, is quantitatively the most important nephron site for Na⁺ and water reabsorption. The proximal tubule also plays pivotal roles in the reabsorption of many other substances, including amino acids, glucose, bicarbonate, and phosphate. Abnormalities of proximal tubule function are involved in several important diseases such as acute renal failure.

NO is a small hydrophobic gas molecule. Since being recognized as the molecule accounting for the biological activity of endothelium-derived relaxing factor a little more than a decade ago (25, 54), an amazingly wide variety of biological activities, ranging from vasorelaxation to immune response to neurotransmission, have been shown to involve NO.

Here we reviewed current knowledge and identified gaps in the knowledge regarding production and functional roles of NO in the proximal tubule, with emphasis on the physiology and pathophysiology of proximal tubule function.

	PRODUCTION OF NO BY THE PROXIMAL TUBULE

TOP ABSTRACT INTRODUCTION PRODUCTION OF NO BY... ROLE OF NO IN... ROLE OF NO IN... PERSPECTIVES SUMMARY REFERENCES

It is controversial whether the proximal tubule produces NO under basal conditions. NO is synthesized from L-arginine, catalyzed by the enzyme NO synthase (NOS). At least three isozymes of NOS have been identified: endothelial NOS (eNOS), neuronal NOS (nNOS), and inducible NOS (iNOS). eNOS and nNOS are traditionally called constitutive NOS. Generally iNOS is only expressed after induction by appropriate stimuli. However, several tissues, including the proximal tubule, constitutively express iNOS mRNA. In an in situ hybridization study in normal rat kidneys using iNOS cRNA, the S3 segment of the proximal tubule was found to be one of the most intensely labeled nephron segments (2), whereas the labeling of the S1 and S2 segments of the proximal convoluted tubule was much weaker. In addition to iNOS, eNOS mRNA was also detected by RT-PCR in some, but not all, microdissected rat proximal tubule segments (65). NOS activity measured as the conversion of L-[³H]arginine to L-[³H]citrulline and/or accumulation of NO end-metabolites NO₂ or NO₂/NO₃ was detected in isolated rat proximal tubules, primary cultures of rat or human proximal tubule cells, and proximal tubule cell lines (8, 19, 47, 70). In proximal tubules isolated from rat kidneys using Percoll centrifugation, NOS activity was measured to be ~4-5 nmol L-[³H]citrulline · mg protein¹ · h¹ (70). In contrast, no studies have shown the presence of NOS proteins in the proximal tubule under basal conditions, which makes it questionable whether the proximal tubule is able to produce NO constitutively. Indeed, using an amperometric NO sensor, Yaqoob et al. (69) failed to detect NO signals in isolated normoxic rat proximal tubules. In studies by Terada et al. (61) and Wu et al. (67), mRNAs for nNOS, eNOS, or iNOS were not detected in microdissected rat proximal tubules by the RT-PCR technique. It is interesting to note that, unlike NOS, the mRNA for soluble guanylate cyclase, which mediates many effects of NO, appears to be more abundant in the proximal tubule than in most other tubular segments (61). It was reported that the NOS activity in isolated normoxic rat proximal tubules was inhibited by the NOS inhibitor N^G-nitro-L-arginine methyl ester (L-NAME) (70). However, at concentrations of up to 10 mM, L-NAME or the relatively selective iNOS inhibitor aminoguanidine failed to affect the basal accumulation of NO₂ in the media of opossum kidney (OK) cells or LLC-PK₁ cells, both of which are proximal tubule cell lines (unpublished data). In any case, one should be cautious when interpreting the data obtained from isolated proximal tubules or proximal tubule cell cultures. Proximal tubules in vivo are exposed to complex neurohormonal microenvironments as well as various biophysical forces that are mostly lost ex vivo or in vitro. These factors might affect the ability of proximal tubules to produce NO. Moreover, isolation procedures and in vitro culturing might alter certain properties of proximal tubular cells.

Interestingly, a recent microdialysis study showed that there was detectable NO, measured by hemoglobin trapping, in the microdialysate from the normal rat renal cortex and medulla, which presumably reflects NO concentrations in renal interstitial fluid (71). NO as well as NO₂/NO₃ concentrations in the renal microdialysate was decreased by intravenous infusion of L-NAME and increased by L-arginine. The presence of NO metabolites in renal cortical and medullary microdialysate was confirmed in our studies (A. Pfluger and F. G. Knox, unpublished data). These data indicated that proximal tubules in vivo were constantly exposed to NO that potentially could affect function of the proximal tubule. The origin of NO may be from proximal tubules and/or nonproximal tubule sources, such as the vasculature or other nephron segments. That vasculature-derived NO may affect proximal tubular functions is supported by an in vivo study by Amorena and Castro (4), which suggested that NO released from the endothelium of peritubular capillaries participated in the regulation of proximal tubular acidification. A coincubation study by Linas and Repine (42) also suggested that endothelial cell-derived NO could regulate the sodium reabsorption by proximal tubular epithelial cells.

Although it is less certain whether the proximal tubule produces NO under basal conditions, studies repeatedly showed that the proximal tubule is able to produce large quantities of NO after appropriate stimulation. An early study by Markewitz et al. (44), demonstrated that treatment with tumor necrosis factor (TNF)- and interferon- stimulated NO generation from primary cultures of rat proximal tubule cells, accompanied by substantial increase in iNOS mRNA. The induction of iNOS and/or NO production in proximal tubule cells by various combinations of lipopolysaccharide (LPS), interferon-, interleukin-1, or TNF- was confirmed in several studies performed in primary cultures of mouse (19, 23) and human (47) proximal tubule cells. In rat kidneys in vivo, treatment with LPS failed to elicit significant induction of iNOS mRNA in proximal tubules assessed by in situ hybridization (2). However, a two- to threefold increase of NOS activity was reported in proximal tubules isolated from LPS-treated rats (45). Induction of iNOS expression by LPS plus interferon- in mouse proximal tubule cells was inhibited by osteopontin, a secreted Arg-Gly-Asp-containing phosphoprotein (23). 2,4-Diamino-6-hydroxypyrimidine, a tetrahydrobiopterin synthesis inhibitor, blunted the increase of NO synthesis in mouse proximal tubule cells induced by LPS plus interferon- (3). The ₁₃-subunit of G protein, which is coupled to proinflammatory substances thrombin and thromboxane, also appears to play a role in the induction of iNOS in the proximal tubule cells because overexpression of ₁₃ in a mouse proximal convoluted tubule (MCT) cell line, results in increased expression of the macrophage type of iNOS (29).

In addition to LPS and cytokines, several other factors have also been shown to stimulate NO production in the proximal tubule. One of these factors is hypoxia. With the use of isolated rat proximal tubules, Yu et al. (70) demonstrated that 15-min hypoxia caused significant increase in NOS activity, which was not affected by Ca²⁺ chelation with EGTA. Hypoxia-induced elevation of NO production from isolated proximal tubules was confirmed by measurements using an amperometric NO sensor (69). This elevation was prevented by extracellular acidosis. No conclusive evidence was provided in these studies as regard to which isoform(s) of NOS was involved. The rather rapid response did not rule out the possible involvement of iNOS, because iNOS mRNA might be constitutively present in proximal tubules and upregulation of iNOS might occur through rapid mechanisms independent of de novo synthesis of the enzyme. On the other hand, lack of effect of Ca²⁺ chelation with EGTA did not rule out the possible involvement of constitutive NOS, because constitutive forms of NOS independent of free Ca²⁺ exist (14). In a study by Hwang et al. (24), it was shown that constitutive NOS mRNA in human proximal tubule cells was enhanced during hypoxia. With the use of a video imaging technique, Edelstein et al. (12) observed an early rise in intracellular Ca²⁺ concentration in rat proximal tubule cells after hypoxic challenge, which is followed by activation of constitutive NOS. These studies suggest that constitutive NOS is involved in hypoxia-stimulated NO production in proximal tubules. However, in a study performed in proximal tubules isolated from mice with knockout of various NOS isoforms, only the knockout of iNOS increased the resistance of the proximal tubule to hypoxic damage (43), which favors the involvement of iNOS rather than eNOS or nNOS in hypoxia-stimulated NO production in proximal tubules.

NO production and NOS activity in rat proximal tubule cells was shown to increase after treatment with iron for at least 8 h (8). The approximate twofold increase in NO₂/NO₃ accumulation after iron treatment was abolished by the relatively selective iNOS inhibitor aminoguanidine. The mechanism of this upregulation is unclear. Atrial natriuretic factor and ANG II have also been shown to increase NO production in proximal tubule cells (48).

It is interesting to note that the proximal tubule, particularly the proximal convoluted tubule, is the primary site in the kidney that synthesizes arginine from citrulline (35, 36, 50), suggesting that the substrate of NO synthesis is readily available in this nephron segment. Moreover, the proximal tubule also has locally elevated activity of arginase that degrades arginine to urea and ornithine (37). Therefore, it appears that the proximal tubule plays an important role in renal metabolism of arginine, the substrate for NO synthesis.

	ROLE OF NO IN THE REGULATION OF SODIUM AND FLUID TRANSPORT IN THE PROXIMAL TUBULE

TOP ABSTRACT INTRODUCTION PRODUCTION OF NO BY... ROLE OF NO IN... ROLE OF NO IN... PERSPECTIVES SUMMARY REFERENCES

Effects on proximal tubular sodium reabsorption. It has been shown in several studies performed in humans that systemic administration of competitive NOS inhibitors decreased fractional excretion of sodium (FE_Na) and fractional excretion of lithium, the latter being an index for sodium transport in the proximal tubule (5, 7, 49). These results suggest an inhibitory effect of basal NO on proximal tubular sodium reabsorption. In studies performed in rats, however, it was shown that systemic administration of NOS inhibitors was natriuretic (20, 27, 28), which appeared to be dependent on the elevation of renal perfusion pressure and the subsequent increase in renal interstitial hydrostatic pressure (20). It is noteworthy that the doses of NOS inhibitors used in the above human and rat studies were rather different. In the human studies, N^G-monomethyl-L-arginine (L-NMMA) was given at 3 mg/kg over 10 min (5). Alternatively, L-NAME was given at 1-25 µg · kg¹ · min¹ for 30 min (7) or at 3 µg · kg¹ · min¹ for 90 min (49). In the studies performed in rats, L-NMMA was given as a 15 mg/kg bolus followed by continuous infusion at 500 µg · kg¹ · min¹ (20, 27, 28). In a study performed in rats by Lahera et al. (34), it was shown that L-NAME at the dose of 0.1 or 1 µg · kg¹ · min¹ decreased urinary sodium excretion (U_NaV) without affecting mean arterial pressure (MAP), whereas 10 or 50 µg · kg¹ · min¹ L-NAME increased MAP and reversed the initial fall in U_NaV. In accordance with these data, it seems that the effect of systemic administration of NOS inhibitors on sodium reabsorption is affected by other systemic effects of this maneuver. Therefore, these studies, even with measurements of lithium excretion, did not provide definitive information regarding direct effects of NO on the sodium reabsorption in the proximal tubule.

In a microperfusion study in rat kidneys by Wang (66), a biphasic effect of NO on proximal tubular transport of sodium and bicarbonate was suggested. An intravenous bolus injection of L-NAME followed by addition of L-NAME to the luminal perfusate modestly decreased proximal tubular fluid reabsorption (J_v) and bicarbonate reabsorption (J_HCO3). When 1 µM sodium nitroprusside (SNP) or S-nitroso-N-acetylpenicillamine, both of which are NO donors, was added to the luminal perfusion solution, J_V and J_HCO3 in proximal convoluted tubules were increased by 30~50%. These increments appeared to be dependent on cGMP and mediated by stimulation of Na⁺/H⁺ exchangers. When 1 mM SNP was added, however, proximal tubular J_V and J_HCO3 were decreased by 50~70%. No data were provided regarding the mechanism of this inhibition of proximal tubular reabsorption. On the basis of these data, it was suggested that NO stimulated proximal tubular reabsorption at lower concentrations and inhibited proximal tubular reabsorption at higher concentrations. Consistent with an inhibitory effect of NO on proximal tubular fluid reabsorption, a split-drop micropuncture study in rat kidneys by Eitle et al. (13), showed that SNP at the concentration of 0.1 mM added to either the luminal perfusate or the peritubular perfusate decreased proximal tubular fluid reabsorption. SNP over the range of 1 µM to 1 mM dose dependently stimulated cGMP accumulation in proximal tubule suspensions. In contrast to Wang's studies, no effect on basal proximal tubular fluid reabsorption was observed with 1 µM SNP.

Altogether, these in vivo studies generally support the notion that NO decreases proximal tubular transport. Compared with systemic drug administration and measurements, microperfusion and micropuncture studies are designed to reduce the systemic interference and better reflect changes in specific tubular segments. However, the diffusibility and metabolic kinetics of NO as well as NOS inhibitors in tubular tissues and renal interstitial fluid are not well delineated. Therefore, it is unclear whether NO or NOS inhibitors administered into the tubular lumen would be confined to the lumen and the surrounding tubular cells. It remains to be determined whether these studies reflect cross-talk of NO with other regulatory mechanisms of proximal tubular transport and/or direct effects of NO on the transport machinery.

Interaction with other mechanisms regulating proximal tubular sodium reabsorption. Evidence exists that NO can interact with other regulatory mechanisms of proximal tubular transport, including driving forces, ANG II, and renal nerves. Renal interstitial hydrostatic pressure is an important determinant of the driving force for tubular transport (18, 30). Systemic administration of NOS inhibitors increases renal perfusion pressure and renal interstitial hydrostatic pressure accompanied by natriuresis (20). When renal interstitial hydrostatic pressure is prevented from increasing by renal decapsulation, L-NMMA-induced natriuresis is abolished (20). In a study by Nakamura et al. (52), pretreatment of rats with 5 µg · kg¹ · min¹ L-NAME was shown to abolish the increase in renal interstitial hydrostatic pressure and FE_Li caused by increases in renal perfusion pressure, suggesting a role for NO in the transmission of renal perfusion pressure to renal interstitium.

Interaction with ANG II may also be an important factor in determining the effect of NO on proximal tubular transport, although the mode of interaction remains obscure. ANG II by itself has a dose-dependent biphasic effect on proximal tubular reabsorption (21), being stimulatory at picomolar concentrations and inhibitory at nanomolar concentrations. In a micropuncture study by De Nicola et al. (11), it was shown that L-NMMA markedly decreased the absolute and fractional proximal tubular reabsorption, which was prevented by DUP-753 (losartan), an ANG II type 1 (AT₁) receptor antagonist. On the basis of these data, the authors speculated that basal NO masked the inhibitory effect of ANG II and allowed the expression of stimulatory effect of ANG II. However, DuP-753 alone also decreased the absolute and fractional proximal tubular reabsorption (11), suggesting that endogenous ANG II stimulated proximal tubular reabsorption through AT₁ receptors. Therefore, it is not clear why the absolute and fractional proximal tubular reabsorption remained unchanged in the presence of both L-NMMA and DUP-753. In contrast, in the split-drop micropuncture study by Eitle et al. (13), SNP (1 µM or 0.1 mM) abolished the stimulatory effect of luminal ANG II (1 nM) on proximal tubular fluid reabsorption.

Regulation of proximal tubular reabsorption involves renal adrenergic nerve activity (30). A study by Gabbai et al. (15) showed that intravenous infusion of L-NMMA (30 mg · kg¹ · h¹) decreased fractional proximal reabsorption in Munich-Wistar rats. When kidneys were denervated ~1 wk before the experiment, the decrease in fractional proximal reabsorption by L-NMMA was prevented (15). However, closer inspection of the data reveals that L-NMMA similarly decreased the fractional proximal reabsorption from 43 to 35% in sham-operated rats and from 43 to 36% in denervated rats, although the difference was statistically significant in sham-operated rats, but not in denervated rats. In a study by Thomson and Vallon (62) using similar experimental design, denervation was again shown to abolish the decrease in fractional proximal reabsorption caused by L-NMMA. The effect of denervation was reversed by systemic infusion of the ₂-agonist B-HT-933. Moreover, in innervated rats, infusion of the ₂-antagonist yohimbine abolished the inhibitory effect of L-NMMA on the fractional proximal reabsorption. With the use of acute renal denervation, however, Khraibi (28) showed that, in Wistar-Kyoto rats, the natriuretic effect of L-NMMA was not dependent on renal innervation, whereas it was attenuated by acute renal denervation in Okamoto spontaneously hypertensive rats.

Effects on the machinery for proximal tubular sodium reabsorption. With regard to direct effects of NO on the proximal tubular transport machinery, several studies have been performed in in vitro proximal tubules or proximal tubule cell cultures to identify the targets of NO and the mechanism. Proximal tubular transport is primarily driven by Na⁺-K⁺-ATPase located on the basolateral membrane. A study by McKee et al. (46) demonstrated that endogenous or exogenous NO inhibited Na⁺-K⁺-ATPase activity in rat renal medulla slices. This study did not examine which tubular segments were the source of Na⁺-K⁺-ATPase and NO involved in the observed inhibition. In a study performed in MCT cells, a mouse proximal tubular cell line, Guzman et al. (19) showed that both LPS-interferon--stimulated endogenous NO and NO donors SNP (0.4 mM) or SIN-1 (30 µM) inhibited Na⁺-K⁺-ATPase by ~30%. This effect of NO was prevented by superoxide dismutase and partially mimicked by a cGMP analog but not affected by alteration of Na⁺ entry into the cell. On the basis of these data, it was suggested that peroxynitrite played a role in this inhibition. The inhibitory effect of NO on proximal tubular Na⁺-K⁺-ATPase was confirmed by studies performed in OK cells, another proximal tubular cell line (40). This inhibition reflects a reduction of Na⁺-K⁺-ATPase molecular activity rather than changes in number of functional enzyme units on cell surface as measured by ouabain binding. This is in contrast with effect of NO on the Na⁺-K⁺-ATPase in a medullary thick ascending limb cell line that appeared to involve inhibition of transcription of ₁-subunit of this enzyme (32). The inhibition of Na⁺-K⁺-ATPase in OK cells by NO was reproduced by a cGMP analog and blunted by a soluble guanylate cyclase inhibitor, suggesting a role for cGMP in mediating this inhibition. Furthermore, NO was shown to activate protein kinase C- (PKC-) in OK cells and the inhibition of Na⁺-K⁺-ATPase by NO in OK cells was abolished by PKC inhibitors, but not by a cGMP-dependent protein kinase inhibitor (38). These data indicate that the inhibition of Na⁺-K⁺-ATPase by NO in proximal tubular OK cells involves activation of PKC-, a mechanism shared by several other hormones such as dopamine (10). It remains to be determined how NO activates PKC- in OK cells and how cGMP and PKC- interact with each other in mediating the inhibition of Na⁺-K⁺-ATPase by NO. Interestingly, NO did not affect the Na⁺-K⁺-ATPase activity in LLC-PK₁ cells, a proximal tubular cell line of pig origin, suggesting the presence of heterogeneity of regulation of proximal tubular Na⁺-K⁺-ATPase by NO (38). In addition to Na⁺-K⁺-ATPase, Na⁺/H⁺ exchangers that mediate the majority of Na⁺ entry through the apical membrane of proximal tubular epithelia have also been shown to be affected by NO. In cultured rabbit proximal tubular cells, NO donors inhibited Na⁺/H⁺ exchanger activity by ~30%, which was reproduced by a cGMP analog and partially prevented by a soluble guanylate cyclase inhibitor (56).

Effects of NO on Na⁺-K⁺-ATPase and Na⁺/H⁺ exchangers demonstrated in these studies are consistent with an inhibitory role of NO in the regulation of proximal tubular reabsorption. It is important to note that doses of the NO donor SNP used in these studies were in the range of 0.4-1 mM (19, 40, 56), which are close to doses shown to inhibit proximal tubular reabsorption in vivo (13, 66) but substantially higher than the dose shown to stimulate proximal tubular reabsorption in vivo (66). Although several in vivo microperfusion and micropuncture studies suggest a stimulatory role of basal NO on proximal reabsorption (11, 66), no study has demonstrated stimulatory effects of NO on the machinery for proximal sodium reabsorption.

Effects on proximal tubular paracellular transport. In addition to a transcellular pathway, proximal tubules possess an unusually leaky paracellular pathway (17). Although the exact quantity is not yet clear, taking the literature as a whole, about one-third of the total fluid reabsorption in the proximal tubule occurs through the paracellular pathway (17). In a study using OK cell sheets cultured on permeable supports, whose paracellular permeability is close to that of the in vivo proximal tubule (41), NO was shown to increase the paracellular permeability of OK cells (39). This effect of NO appeared to involve reduction of cellular ATP content. The dose of the NO donor SNP required to elicit this effect, although conceivably available in proximal tubules, was rather high (2 mM for 24 h), suggesting that this effect of NO might be relevant under circumstances with prolonged overproduction of NO. As described in PRODUCTION OF NO BY THE PROXIMAL TUBULE, NO production in proximal tubules could be substantially and rapidly enhanced under situations such as hypoxia/reoxygenation (70). The overproduction of NO in kidneys persists at 24 h after ischemia-reperfusion injury (9, 53). In a study by Kwon et al. (33), it was shown that the paracellular permeability of proximal tubules was enhanced in human postischemic renal allografts with sustained acute renal failure. It would be interesting to determine if overproduction of NO plays a role in this pathological change.

An integrative view of the role of proximal tubular effects of NO in the regulation of fluid and electrolyte homeostasis. Reported effects of NO on the proximal tubular transport are summarized in Fig. 1. Despite the apparent controversy engendered by the natriuretic effects of L-NMMA, the effect of NO on the proximal tubular transport machinery in vitro (19, 38, 40, 56) and the effect of NO donors (0.1-1 mM SNP) on the proximal tubular fluid reabsorption in vivo (13, 66) indicate that NO functions as an inhibitor for the proximal tubular fluid and sodium reabsorption. This is in concert with inhibitory effects of NO on fluid and sodium reabsorption in other tubular segments (16, 32, 59). In this sense, NO is a natriuretic agent. This is, in principle, consistent with the prominent role of NO in maintaining vascular tone and preventing increase in blood pressure. However, the final effect of NO on proximal tubular sodium reabsorption and its role in the overall fluid and electrolyte homeostasis may vary under different circumstances. This is mainly caused by the complex effect of NO on various targets, including hemodynamics, the renin-angiotensin system, and the tubular system. For example, increases in arterial blood pressure and renal interstitial hydrostatic pressure caused by systemic inhibition of NO synthesis lead to decreases in tubular reabsorption of fluid and sodium (20, 52). In this case, inhibition of NO synthesis is natriuretic. Apparently, the natriuresis seen under this situation would favor the return of the abnormally increased arterial blood pressure to normal. Therefore, the role of NO in the regulation of proximal tubular transport can only be understood when it is integrated with other aspects of NO functions and when it is properly incorporated into the overall regulation of fluid and electrolyte homeostasis.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effects of nitric oxide (NO) on proximal tubular transport. , stimulatory; , inhibitory.

	ROLE OF NO IN THE INJURY OF PROXIMAL TUBULES

TOP ABSTRACT INTRODUCTION PRODUCTION OF NO BY... ROLE OF NO IN... ROLE OF NO IN... PERSPECTIVES SUMMARY REFERENCES

NO belongs to a class of relatively weak free radicals. It can react with a variety of substances, including oxygen, superoxide anion, thiol groups, heme proteins, peroxyl radicals, iron-sulfur complexes, and nonthiol amino acid residues (22). Some products of these reactions, such as peroxynitrite and hydroxyl radical, are cytotoxic. Some of these reactions also have deleterious effects. Consistently, NO has been shown to be involved in the pathophysiology of many diseases, including kidney diseases (31). On the other hand, other functions of NO, such as maintaining vascular tone and antiplatelet aggregation, may have beneficial effects under certain circumstances.

Several studies related to the toxic effect of NO on proximal tubules were performed in the context of acute renal failure. As mentioned above, either through iNOS or constitutive NOS, NO production by proximal tubules is enhanced by hypoxia. Increases in lactate dehydrogenase (LDH) release from isolated rat proximal tubules caused by 15 min of hypoxia followed by 35 min of reoxygenation were prevented by L-NAME and enhanced by L-arginine, the substrate for NO synthesis (70). Addition of SNP further enhanced hypoxia/reoxygenation injury, which was prevented by the NO scavenger hemoglobin. These data indicated a pivotal role of NO in the hypoxic injury of proximal tubules. In contrast to a deleterious role of NO, inhibition of NOS in vivo with NOS inhibitors was shown to aggravate renal dysfunction in renal ischemia or radiocontrast or cyclosporin nephrotoxicity (1, 6, 53, 58), suggesting a beneficial role of NO in these models of acute renal failure. One explanation for this discrepancy was the nonspecificity of NOS inhibitors that might compromise the potential protective effect of eNOS-derived NO. With the use of isoform-specific oligodeoxynucleotides, Noiri et al. (53) showed that in vivo targeting of iNOS protected rat kidney against ischemia. Both the elevation of NO production and iNOS induction in rat kidneys caused by ischemia were largely blunted by targeting of iNOS with oligodeoxynucleotides. In the meantime, normal renal function and renal tubular histology were preserved. It is interesting to note that oligodeoxynucleotides targeted specifically to vascular smooth muscle type of iNOS did not protect kidneys against ischemia, but rather exacerbated renal ischemic injury. This suggests that perhaps only macrophage-type iNOS-derived NO is involved in renal ischemic injury. In agreement with Noiri's studies, Chiao et al. (9) also demonstrated that iNOS proteins in outer medulla were increased after bilateral renal ischemia and that -melanocyte-stimulating hormone, a potent anti-inflammatory agent, protected against renal ischemic injury concomitant with inhibition of iNOS induction. These studies indicated that iNOS-derived NO was responsible for NO-mediated renal toxicity in acute renal failure. This is in agreement with an in vitro study showing that proximal tubules isolated from mice lacking iNOS, but not eNOS or nNOS, were resistant to hypoxic injury (43). On the other hand, several studies suggested that constitutive forms of NOS in proximal tubular cells were activated by hypoxia (12, 24). The role of this activation of constitutive NOS is not clear. In the study by Noiri et al. (53), increment of eNOS proteins in kidney tissues was minimal after ischemia.

NO has also been shown to participate in the cytotoxic effect of lipid A (63, 64), HgCl₂ (68), and iron (8) on proximal tubule cells. Treatment with iron increased NO₂/NO₃ accumulation in the media of rat proximal tubular cells by about twofold. However, the relatively selective iNOS inhibitor aminoguanidine, which completely abolished iron-induced NO production, only partially blunted iron-induced LDH release (8). This suggests that NO only plays a partial role in the iron toxicity in proximal tubules.

As mentioned in an earlier section, the proximal tubule is also rich in arginase activity that degrades arginine to form urea and ornithine (37). Ornithine is the precursor for the synthesis of polyamines that are required for cell proliferation and L-proline that is a substrate for collagen synthesis (26, 51). NO, on the other hand, is known to cause cytostasis (55, 60). In a study performed in a mouse proximal tubule cell line, exogenous NO or cytokine-induced endogenous NO production was shown to inhibit both the rate-limiting enzyme for polyamine synthesis and the uptake of polyamines (57). The interaction between the two pathways of arginine metabolism may be of significance in the regulation of cell proliferation.

	PERSPECTIVES

TOP ABSTRACT INTRODUCTION PRODUCTION OF NO BY... ROLE OF NO IN... ROLE OF NO IN... PERSPECTIVES SUMMARY REFERENCES

Although substantial progress in the understanding of the production and effects of NO by and on the proximal tubule has been made, it is evident that many inconsistencies and gaps remain. Studies aimed to dissect the effect of NO on the proximal tubule at cellular and molecular levels should be helpful in further defining the role of NO in this tubular segment and its mechanisms. Equally important is the integration of the direct effect of NO on the proximal tubule and the wide variety of other effects of NO. Both lines of studies are critical in enhancing our understanding of NO biology and proximal tubule physiology as well as in developing effective interventions in case of abnormal proximal tubular transport or proximal tubular injury. For example, it would be very interesting to determine whether NO production in the proximal tubule is altered by sodium loading or volume expansion and whether the regulation of proximal tubular transport by NO plays any role in the response to sodium loading or volume expansion. Novel approaches need to be developed to address these questions. For instance, approaches allowing in vivo measurement of proximal tubule-specific NO production and those allowing proximal tubule selective manipulation of NO production, such as proximal tubule-targeted disruption or delivery of NOS, would be helpful.

	SUMMARY

TOP ABSTRACT INTRODUCTION PRODUCTION OF NO BY... ROLE OF NO IN... ROLE OF NO IN... PERSPECTIVES SUMMARY REFERENCES

Although whether the proximal tubule produces NO under basal conditions is still controversial, evidence suggests that the proximal tubule is constantly exposed to NO. Moreover, the proximal tubule is able to produce large quantities of NO upon a variety of stimulation. NO plays an important role in regulating proximal tubular reabsorption of fluid, sodium, bicarbonate, and phosphate. NO regulates Na⁺-K⁺-ATPase, Na⁺/H⁺ exchangers, and paracellular permeability of proximal tubular cells, which may contribute to its effect on proximal tubular transport. The final effect of NO on proximal tubular reabsorption appears to depend on the concentration of NO and involve interaction with other regulatory mechanisms. Enhanced production of NO, perhaps macrophage-type inducible NO synthase-dependent production of NO, participates in hypoxic/ischemic proximal tubular injury. In conclusion, NO plays an indispensable and fundamental role in both physiology and pathophysiology of the proximal tubule.