Immunohistochemically Detected Protein Nitration Indicates Sites of Renal Nitric Oxide Release in Goldblatt Hypertension
Abstract In the kidney, nitric oxide (NO) from the macula densa (MD) is considered an integral modulator of the tubulovascular message system, whereas endothelium-derived NO is a major vasorelaxing factor. The goal of the present study was to determine extracellular pathways of NO in rats with renovascular two-kidney, one clip Goldblatt hypertension (2K1C). To localize NO in the tissue, immunohistochemical detection of NO-dependent tyrosine nitration was performed using a monoclonal antibody against nitrotyrosine. Nitration of phenolic compounds such as tyrosine results from the reaction with peroxynitrite (ONOO−) formed by NO and molecular oxygen or superoxide and may therefore be used as a footprint for local release of NO. Significant nitrotyrosine immunoreactivity was detected in the extraglomerular mesangium (EGM) of the stenotic kidney in 2K1C rats, whereas in the nonclipped contralateral kidney and in control animals no signal was detected at this site. Positive staining of the EGM was paralleled by enhanced NADPH diaphorase (NADPH-d) staining of the adjacent MD, signifying increased type I nitric oxide synthase (NOS) activity in the stenotic kidney. In contrast, in the cortical vasculature selectively enhanced nitrotyrosine immunoreactivity was detected in the arteriolar wall of the nonclipped contralateral kidney, and endothelial NADPH-d signal, indicating NOS Type III activity, was enhanced in parallel. Our results suggest that in MD, stimulation of NOS in the stenotic Goldblatt kidney induces the release of NO into the EGM. From there an NO-dependent intermediate stimulus may reach the glomerular vasculature. Footprints of NO-dependent effects in the vascular smooth muscle layer of the non-clipped contralateral kidney indicate a marked vasodilatory response that may have been caused by enhanced shear stress and/or angiotensin II levels.
Nitric oxide is an important mediator in cell-to-cell communication. Constitutive, calcium/calmodulin–dependent isoforms of NOS, which synthesize NO from l-arginine, play a significant role in the regulation of renal and glomerular hemodynamics and tubular function.1 2 3 The presence of the NOS I in MD cells has been reported in a variety of mammalian species.4 5 6 Synthesis of NO in MD cells may be involved in the mediation or modulation of the tubulovascular signal transfer from the MD, ie, tubuloglomerular feedback response and/or MD-dependent regulation of afferent arteriolar renin production.7 8 9 10 Renal vascular NO synthesis is accomplished by the NOS III, which has been localized in cortical and medullary endothelia.4 Endothelial NOS provides a vasodilator tone that antagonizes vasoconstrictors such as angiotensin II,2 and intrarenal inhibition of NOS produces an increase in renal vascular resistance and a decrease in renal plasma flow.2 3 11 Recent work by Beckman and associates12 has demonstrated that the release of NO into the tissue can be visualized histochemically. This detection was based on the notion that NO reacts at near-diffusion–limited rates with molecular oxygen or, at a more rapid rate, superoxide anion (O2·−) to form the powerful oxidant and nitrating agent peroxynitrite (ONOO−).13 ONOO− readily nitrates phenolic compounds such as tyrosine residues in proteins by adding a nitro group to the ortho position of tyrosine, thereby leaving a footprint that may be detected immunohistochemically using antibody against nitrotyrosine. With this technique, NO-related effects have been demonstrated in the human atherosclerotic process,12 acute lung injury,14 and other conditions13 in humans.
The present study was undertaken to demonstrate the release of and to localize possible pathways for NO at the juxtaglomerular apparatus and in the renal vasculature. The 2K1C Goldblatt model of renovascular hypertension has been chosen for investigation, since previous studies had revealed differential stimulation of both constitutive NOS isoforms.7
Male Sprague-Dawley rats were obtained from the local animal facility. All animals were perfusion-fixed at the end of the experiments. The experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the German Law on the Protection of Animals.
To achieve renovascular hypertension by a unilateral stenosis of the renal artery, a total of 14 rats initially weighing 150±10.6 g (mean±SD) were used. Animals were divided into two groups, ie, a “clip” group that received a renal stenosis during 40 days (n=7) and an age-matched, sham-operated control group (n=7). For surgery, animals were anesthetized with ether, and then the right renal artery was isolated through flank incision; a 0.2-mm silver clip was placed in the animals designated for clipping. In sham-operated animals the right renal artery was partially exposed by removal of surrounding fat and connective tissue. Systolic arterial pressure was measured under light ether anesthesia using the tail-cuff method. Blood pressure values varied between 105 and 140 mm Hg in the sham-operated control group and 190 and 260 mm Hg in the clip group at the end of the experiment. All animals received standard laboratory chow and tap water ad libitum. Final body weights were 374.4±22.3 g for the control group and 325.4±28.3 g for the clip group.
For histochemical evaluation, kidneys were in vivo perfusion-fixed. Rats were anesthetized by pentobarbital sodium (40 mg/kg IP), and perfusion was performed for 90 seconds by cannulation of the abdominal aorta at a pressure of 220 mm Hg with 3% freshly prepared paraformaldehyde in PBS (pH 7.4). At reduced pressure (60 mm Hg), perfusion with the fixative was continued for another 4 minutes, followed by perfusion with a sucrose-PBS solution adjusted to 800 mOsm/kg H2O to prevent the formation of freezing artifacts. After perfusion, kidneys were removed. Pieces of part of the kidneys were routinely processed for paraffin embedding, and part of them immediately shock-frozen in liquid nitrogen–cooled isopentane.
For histochemical demonstration of NADPH-d activity, 5-μm-thick cryostat sections were incubated in 0.1 mol/L phosphate buffer containing nitroblue tetrazolium (NBT), β-NADPH, and Triton X-100.4 In control animals, the reaction was stopped when the MD signal was clearly distinguishable and background staining had not yet appeared; thereby the optimal exposure time was set at 25 minutes (incubation at 37°C). No reaction product was found when NADPH was replaced by NADH.
For detection of nitrotyrosine, a mouse monoclonal antibody directed against nitrotyrosine (gift from J. Beckman, Birmingham, Ala) was used on paraffin sections. After deparaffination, sections were incubated for 30 minutes with 5% normal rabbit serum diluted in phosphate buffer used as blocking medium, followed by incubation with specific antibody (1:10 in PBS) for 12 to 18 hours at 4°C. After sections were rinsed in PBS, bound antibody was detected by the PAP method with an anti-mouse PAP complex and diaminobenzidine and H2O2 for color development. Control experiments were made by blocking the specific antibody with 10 mmol/L nitrotyrosine as well as by omitting the first antibody and using nonimmune serum instead.
Quantification of Histochemical Signal
For histochemical evaluation, two sections from each animal were viewed by two independent persons in a blinded fashion. For semiquantitative evaluation of the degree of nitration of protein tyrosine within the extraglomerular mesangium, the fraction of identifiable glomerular vascular poles with a positive staining was established. For the semiquantitative evaluation of NADPH-d enzyme activity, the amount of MD cells with positive NADPH-d reaction within a standard area was established as previously shown.7
Chemical Reagents, Kits, and Instruments
β-NADPH, NADH, and diaminobenzidine were from Sigma Chemical Co. Biotin-streptavidin-peroxidase complex was from Amersham, anti-mouse PAP complex was from Dianova, and nitroblue tetrazolium was from Biomol. Slides were viewed in a Polyvar II light microscope (Reichert) equipped with interference contrast optics.
The long-term unilateral stenosis of the renal artery during 40 days resulted in a marked elevation of mean systolic blood pressure (235.1±35.6 mm Hg at the end of the experiment). Similar to results from a previous study performed by our group,7 NADPH-d reaction intensity as expressed by the mean number of NADPH-d–positive MD cells per tissue area of 100 glomeruli was roughly 2.4-fold increased in the MD of the clipped compared with the nonclipped kidney (P<.005). Enhanced NADPH-d staining in the MD of the stenotic kidney as shown in Fig 1⇓ indicates an increase in NOS I activity and probably also reflects an increase in the amount of enzyme, since enhanced NOS I mRNA levels in the MD of the stenotic kidney had been detected previously.7 In the nonclipped contralateral kidneys, NADPH-d staining in the MD was markedly reduced or missing (Fig 1⇓). Sham-operated control kidneys showed intermediate levels of NADPH-d staining at the MD in both kidneys (Fig 2⇓).
With the use of anti-nitrotyrosine antibody, extensive immunoreactivity was detected in the EGM of the stenotic kidneys adjacent to the NADPH-d–positive MD, whereas in nonclipped contralateral kidneys as well as in kidneys from sham-operated control kidneys, no such staining was observed (Figs 1⇑ and 2⇑). Cells of the MD were not stained by anti-nitrotyrosine antibody in all specimens investigated. Semiquantitative evaluation revealed that approximately two thirds of the total of identified vascular poles of the stenotic kidneys were clearly positive for nitrotyrosine immunostaining. The proportion of stained EGMs varied considerably between individuals (values ranging between 39% and 84%), and in some specimens evaluation was impaired by nonspecific background staining. The most intensive nitrotyrosine staining was found in those animals that showed the highest blood pressure values, but a significant correlation between these two parameters could not be established. Control incubations for the histochemical reaction, which comprised omission of the second antibody or coincubation of specific antibody with 10 mmol/L nitrotyrosine, respectively, were lacking EGM staining (Fig 1⇑).
In the sham-operated control kidneys, the endothelium of cortical renal arterioles was moderately positive, with NADPH-d staining that has previously been identified as endothelial NOS by colocalization with anti–NOS III antibody.4 Under renovascular hypertension, contrary to juxtaglomerular staining, endothelial NADPH-d staining in the clipped kidneys was often weaker than in control kidneys, whereas in the nonclipped contralateral kidney there was a marked increase in staining (Fig 3⇓). In parallel, a strong signal was encountered in the arteriolar wall and periarteriolar connective tissue of interlobular and glomerular afferent arterioles of the nonclipped kidneys when anti-nitrotyrosine antibody staining was applied (Fig 3⇓). Other portions of the renal vasculature showed no such staining. Arterioles were identified using interference contrast microscopy; the small arterioles have only a single muscle cell layer, and the afferent can be distinguished from the efferent arteriole by the drastic difference in number of endothelial cells protruding into the lumen. Structural resolution of nitrotyrosine staining was not quite sufficient to permit a distinction between endothelial and vascular smooth muscle signals, but it seemed rather that both structures were positive. No arteriolar nitrotyrosine staining was re- corded in sham-operated control kidneys. As in the EGM, controls for specificity of the nitrotyrosine antibody staining resulted in a total absence of signal in the arteriolar wall and surrounding tissue (Fig 3⇓).
The present results demonstrate that long-term unilateral stenosis of the renal artery in the 2K1C Goldblatt model induces differentially localized nitration of protein tyrosines as revealed by the immunoreactivity for nitrotyrosine. On the clipped side, immunoreactive nitrotyrosine was exclusively present in the EGM, whereas on the nonclipped contralateral side, only the arteriolar walls of small preglomerular cortical arterioles were labeled. In both cases, the immunohistochemical signal for nitrotyrosine was spatially associated with a strongly enhanced histochemical signal for NADPH-d staining. The sites of enhanced diaphorase reaction have previously been identified to contain constitutive-type isoforms of NOS.4 7 These observations suggest local synthesis of NO and its paracellular diffusion to the adjacent tissue.
Increased tyrosine nitration has been detected in diseases associated with oxidative stress, ie, in atherosclerotic lesions of human vasculature,12 in infants with acute lung injury,14 and in the aorta15 and lung16 of rats injected with bacterial endotoxin. These results suggested that large amounts of NO originating from the induction of cytokine-dependent inducible type II NOS reacted with locally generated superoxide to form ONOO−, which in turn can hydroxylate and nitrate phenols such as tyrosine.12 17
In the present study, however, NO release was not likely to be derived from NOS II because no histological indication of an inflammatory process in either kidney from the 2K1C group was encountered in or near sites of nitrotyrosine immunoreactivity. A “constitutive” presence of inducible NOS expression in interlobular arteries or glomeruli, as proposed by Mohaupt and associates,18 could contribute to vascular or juxtaglomerular NO release, but we could not confirm these findings using in situ hybridization (unpublished results from our laboratory), and the transcription rate for inducible NOS may therefore be small. More likely, as shown in a previous histochemical investigation that identified the local distribution of NOS isoforms,4 release of NO was based on the activation of endothelial NOS III and epithelial NOS I, respectively, that is to say, from sources that normally produce NO to tonically act within physiological ranges.19 20 Under the specific conditions of chronic 2K1C-Goldblatt hypertension, enhanced stimulation of constitutive NOS isoforms may have led to the release of particularly high levels of NO comparable to the amount of NO released under inflammatory conditions.
This stimulation may have been effective not only at the level of calcium/calmodulin–dependent activation of NOS but also may have been reinforced by means of increased de novo synthesis of an NOS enzyme. Augmented NOS I mRNA levels were detected in the MD of the clipped kidney in response to a chronic decrease of the transport rate at the MD,7 which suggests additional formation of NOS up to amounts that may equal induction of type II isoform. Likewise, an augmentation of endothelial NOS under chronic stimuli such as increased shear stress has been reported.21 22 Apart from new enzyme synthesis, altered supply with substrate and/or cofactors and changes in degradation of the enzyme may well be involved in the enhancement of constitutive NOS. It should be noted in this respect that a principal difference in the amount of NO produced per quantity of a given isoform of NOS has not been demonstrated so far.
Immunoreactivity for nitrotyrosine, as observed in the present study, may be based on tyrosine nitration by NO alone forming 4-nitrosophenol, a reaction that is assumed not to be very effective since it implies the slow reaction of NO with oxygen to form nitrogen dioxide.13 Alternatively, ONOO−, which is readily formed by the reaction of NO and superoxide anion (O2·−), was suggested to be the more likely compound for inducing nitration of tyrosine residues at a rapid rate.13 In the present model, superoxide, which is required for this reaction, may have been generated by flavoprotein-containing oxidoreductases such as NADH oxidase or NADPH oxidase, which are present in the endothelium23 and may also be localized in the EGM and vascular wall; these enzymes were described as major sources of vascular O2·− release.23 24 Elevated angiotensin II levels in this model may have contributed to enhanced O2·− release, since it has recently been demonstrated that angiotensin II can increase both NADH and NADPH oxidase activity in cultured vascular smooth muscle.25 Since in sham-operated controls no protein nitration was found in the EGM despite significant NOS activity, it must be questioned whether increased NOS activity alone has led to protein nitration in the stenotic kidney. Previous semiquantitative histochemical evaluation of NOS I mRNA expression and NADPH-d in the MD under similar conditions has revealed significant increases in NOS I mRNA but less marked changes in NOS activity in stenotic compared with sham-operated control kidneys.7 It may therefore be considered that other factors, such as local superoxide formation, could be selectively active in the clipped kidney with an impact on protein nitration in the EGM. However, since the previous quantification analysis referred to the number of histochemically labeled MD cells rather than to the intensity of label of the individual MD cells directly adjacent to the EGM,7 the output of NO in the stenotic kidney may be higher than can be estimated from the presently available evidence.
When the particular microanatomy of the EGM is considered, which is a closed, avascular compartment in which little washout of intercellular compounds may occur compared with the “regular” renal cortical interstitium, local concentrations of NO might reach relatively high levels that in turn may have caused the observed intensive immunoreactivity for nitrotyrosine. The absence of nitrotyrosine immunoreactivity in MD cells proper may be related to a lack or low quantity of O2·− formation in the MD—contrary to the EGM—thus preventing intraepithelial ONOO− formation.
The abundant presence of G6PD, as previously observed in MD and EGM,26 and unpublished observations from our laboratory may be functionally related to the high local output of NO. G6PD is known to play a prominent role in the defense against oxidative stress27 and may therefore serve as a local scavenger for NO. Conditions that provoke an increase in juxtaglomerular NOS I activity/transcription also induce an increased activity of G6PD.7 26 Thus, upregulation of G6PD in the stenotic kidney could actually decrease the effect of NO on protein nitration in the EGM. However, since G6PD may as well function to supply NOS with its cosubstrate NADPH, an enhancing effect is equally well possible.
Nitrotyrosine immunoreactivity in the vascular walls of the nonclipped contralateral kidney suggests an enhanced tyrosine nitration caused by endothelial release of NO in the cortical vasculature. One of the principal factors that determine endothelial NO synthesis is shear stress. Since it can be anticipated that blood flow in the contralateral kidney was markedly increased, activation of an NO-dependent vasorelaxing response is likely to occur in the present model. In addition, others have shown that NO-induced vasodilatation is an important regulatory response to maintain contralateral renal perfusion despite elevated angiotensin in 2K1C hypertension.2
Apart from the use of nitrotyrosine immunoreactivity as a footprint for the local release of NO, possible functional or pathological roles of tyrosine nitration on protein function are poorly elucidated. It has been suggested that via the inhibition of tyrosine phosphorylation, cellular regulation and signal transduction may be altered,28 and general protein conformation and function may be changed by nitration. It is unknown whether such effects play a role in the current model.
In conclusion, the documentation of nitrotyrosine immunoreactivity in parallel with enhanced staining for NOS suggests the local formation of NO. The notion that NO may be released into the EGM in response to a stimulation of MD NOS I may help to direct future work toward the role of NO in tubulovascular signaling. NO with its broad impact on biological functions may primarily exert an effect in the EGM that has yet to be defined; from there a secondary stimulus could be transmitted to the glomerular vasculature, which is in intimate contact with the EGM. Vascular effects of NO in the nonclipped contralateral kidney may document reactive changes in response to altered blood flow and/or to stimulation of the renin-angiotensin system.
Selected Abbreviations and Acronyms
|NO(S)||=||nitric oxide (synthase)|
|NOS I, II, III||=||constitutive type I, II, or III isoform of NOS|
|2K1C||=||two-kidney, one clip|
This work was supported by the Deutsche Forschungs-gemeinschaft (Ba700/10-1).
- Received August 1, 1996.
- Revision received August 29, 1996.
- Accepted February 27, 1997.
Ito S, Ren S. Evidence for the role of nitric oxide in macula densa control of glomerular hemodynamics. J Clin Invest. 1993;2:1093-1098.
Sigmon DH, Beierwaltes WH. Renal nitric oxide and angiotensin II interaction in renovascular hypertension. Hypertension. 1993;22:237-242.
Bachmann S, Bosse HM, Mundel P. Topography of nitric oxide synthesis by localizing constitutive NO synthases in mammalian kidney. Am J Physiol. 1995;268:F885-F898.
Thorup C, Persson AEG. Inhibition of locally produced nitric oxide resets tubuloglomerular feedback mechanism. Am J Physiol. 1994;267:F606-F611.
Bosse HM, Boehm R, Resch S, Bachmann S. Parallel regulation of constitutive NO synthase and renin at the JGA of rat kidney under various stimuli. Am J Physiol. 1995;269:F793-F805.
Schricker K, Kurtz A. Liberators of NO exert a dual effect on renin secretion from isolated mouse renal juxtaglomerular cells. Am J Physiol. 1993;265:F180-F186.
Singh IJ, Grams M, Wang WH, Yang T, Killen P, Smart A, Schnermann J, Briggs JP. Coordinate regulation of nitric oxide synthase, renin, and angiotensinogen mRNA by dietary salt. Am J Physiol. 1996;270:F1027-F1037.
Wilcox CS, Welch WJ, Murad F, Gross SS, Taylor G, Levi R, Schmidt HHW. Nitric oxide synthase in macula densa regulates glomerular capillary pressure. Proc Natl Acad Sci U S A. 1992;89:11993-11997.
Baylis CB, Mitruka B, Deng A. Chronic blockade of nitric oxide synthesis in the rat produces systemic hypertension and glomerular damage. J Clin Invest. 1992;90:278-281.
Crow JP, Beckman JS. Reactions between nitric oxide, superoxide, and peroxynitrite: footprints of peroxynitrite in vivo. In: Ignarro L, Murad F, eds. Advances in Pharmacology: Nitric Oxide, Vol 34. San Diego, Calif: Academic Press; 1995:17-43.
Haddad IY, Pataki G, Hu P, Galliani C, Beckman JS, Matalon S. Quantitation of nitrotyrosine levels in lung sections of patients and animals with acute lung injury. J Clin Invest. 1994;94:2407-2413.
Wizemann TM, Gardner CR, Laskin JD, Quinones S, Durham SK, Goller NL, Ohnishi ST, Laskin DL. Production of nitric oxide and peroxynitrite in the lung during acute endotoxemia. J Leukoc Biol. 1994;56:759-768.
Xie QW, Nathan C. The high-output nitric oxide pathway: role and regulation. J Leukocyte Biol. 1994;56:576-582.
Nishida K, Harrison DG, Navas JP, Fisher AA, Dockery SP, Uematsu M, Nerem RM, Alexander RW, Murphy TJ. Molecular cloning and characterization of the constitutive bovine aortic endothelial cell nitric oxide synthase. J Clin Invest. 1992;90:2092-2096.
Förstermann U, Kleinert H. Nitric oxide synthase: expression and expressional control of three isoforms. Arch Pharmacol. 1995;352:351-364.
Münzel T, Sayegh H, Freeman BA, Tarpey MM, Harrison DG. Evidence for enhanced vascular superoxide anion production in nitrate tolerance. J Clin Invest. 1995;95:187-194.
Griendling KK, Mineri CA, Ollerenshaw JD, Alexander RW. Angiotensin stimulates NADH and NADPH activity in cultured vascular smooth muscle cells. Circ Res. 1994;74:1141-1156.
Hess R, Gross F. Glucose-6-phosphate dehydrogenase and renin in kidneys of hypertensive or adrenalectomized rats. Am J Physiol. 1959;197:869-872.
Martin BL, Wu D, Jakes S, Graves DJ. Chemical influences on the specificity of tyrosine phosphorylation. J Biol Chem. 1990;265:7108-7111.