The aim of the study is to evaluate the impact of nitric oxide (NO) produced by endothelial NO synthase (eNOS) and neuronal NOS (nNOS) on the angiotensin II response in afferent arterioles (Af). Dose responses were assessed for angiotensin II in microperfused Af of mice homozygous for disruption of the eNOS gene [eNOS(−/−)], or nNOS gene [nNOS(−/−)], and their wild-type controls, eNOS(+/+) and nNOS(+/+). Angiotensin II at 10−8 and 10−6 mol/l reduced the lumen to 69% and 68% in eNOS(+/+), and to 59% and 50% in nNOS(+/+). NG-nitro-l-arginine methyl ester (l-NAME) did not change basal arteriolar diameters, but augmented angiotensin II contraction, reducing diameters to 23% and 13% in eNOS(+/+), and 7% and 10% in nNOS(+/+) at 10−8 and 10−6 mol/l. The response to angiotensin II was enhanced in nNOS(−/−) mice (41% and 25% at 10−8 and 10−6 mol/l) and even more enhanced in eNOS(−/−) mice (12% and 9%) compared with nNOS(+/+) and eNOS(+/+). l-NAME led to complete constriction of Af in these groups. Media-to-lumen ratios of Af did not differ between controls and gene-deficient mice. mRNA expression of angiotensin II receptor types 1A and 1B and type 2 also did not differ. The results reveal that angiotensin II-induced release of NO from both eNOS and nNOS significantly contributes to the control of Af. Results also suggest that eNOS-derived NO is of greater importance than nNOS-derived NO in this isolated arteriolar preparation.
- renin-angiotensin system
- eNOS knockout
- nNOS knockout
- NG-nitro-l-arginine methyl ester
- renal hemodynamics
- juxtaglomerular apparatus
in several studies, a significant influence of nitric oxide (NO) on glomerular hemodynamics has been shown, whereby NO acts as a vasodilator (4, 9, 17). There are two constitutive NO synthase (NOS) isoforms expressed in the juxtaglomerular apparatus, the endothelial NOS (eNOS) and the neuronal isoform (nNOS) in the macula densa cells (2, 38). The NO production of both synthases can be stimulated by many factors, whereby physical stimuli, such as shear stress, play an important role for endothelial NO release. The effect of various vasoactive substances is mediated by NO derived from the endothelium (12). NO produced by nNOS in macula densa cells seems to be intertwined with the tubuloglomerular feedback (TGF) (21, 30, 37). In addition to a tonic production of NO, there is evidence for an activation of NOS by angiotensin II (17, 22, 27, 28). Although these results suggest a significant role of NO derived from eNOS, as well as nNOS in the control of arteriolar tone, the differential contribution of eNOS and nNOS to adjustment of basal tone and to angiotensin II-induced NO release is not clear. The availability of mice homozygous for the disruption of either the eNOS gene [eNOS(−/−)] (36) or the nNOS gene [nNOS(−/−)] (16) allows study of the effects of chronically disabled endothelial NO production and NO production by nNOS selectively. In the present study, we examine the influence of chronic and acute NO production on the angiotensin II response of isolated, perfused afferent arterioles (Af), using eNOS(−/−), nNOS(−/−) mice, and their controls.
In eNOS(−/−) mice, blood pressure is increased by about 20 mmHg (36, 40). This can cause structural changes of the walls of the arteries and arterioles and may also be associated with changes in angiotensin receptor expression (47). A recent study described a correlation between NO efficiency and reduced angiotensin type 2 receptor expression (48). Furthermore, Guo et al. (14) demonstrated that NO stimulates endothelial cell growth, but inhibits smooth muscle cell proliferation. To test whether such possible effects on the arteriolar wall and angiotensin II receptor expression may influence angiotensin II-induced contractions in Af in our mouse models, we analyzed the media-to-lumen ratio of Af and quantified angiotensin II receptor mRNA in the kidneys.
Solutions and Drugs
Physiological saline solution (PSS) was used with the following composition (in mmol/l): 115 NaCl, 25 NaHCO3, 2.5 K2HPO4, 1.3 CaCl2, 1.2 MgSO4, and 5.5 glucose. The K+ solution consisted of 100 mmol/l KCl, whereby 95 mmol/l NaCl were substituted by KCl. DMEM (1,000 ml) was supplied with 100 mg streptomycin, and 100,000 units penicillin. The bicarbonate buffered solution was equilibrated with 5% CO2 in air. The pH was adjusted to 7.4 after the addition of BSA. The concentration of BSA in DMEM (preparation solution) and in PSS (bath solution) was 0.1%. The perfusate of the afferent arteriole consisted of PSS with a BSA concentration of 1%. BSA was obtained from SERVA Electrophoresis, angiotensin II, DMEM from Sigma-Aldrich, and NG-nitro-l-arginine methyl ester (l-NAME) from Alexis Biochemicals. 7-Nitroindazole (7-NI), streptomycin, and penicillin were from Sigma-Aldrich.
Animals and Microperfusion Procedure
The following mice strains were used in this study: C57BL/ 6J-Nos3tm1Unc (36), C57BL/6J B6, 129S-Nos1tm1Plh (16), and B6129F2/J wild-type mice, all from The Jackson Laboratory. The eNOS(−/−) were back crossed with the C57BL/6J B6 [eNOS(+/+)] and kept as inbreeding of homozygous animals [eNOS(−/−)]. The nNOS(−/−) mice were obtained by breeding of heterozygous (+/-) mice, obtained from crossing 129S-Nos1tm1Plh and B6129F2/J wild-type mice. Homozygous (+/+) mice of this breeding served as controls [nNOS(+/+)]. The breeding was continuously monitored by assessing the genetic status of the animals via polymerase chain reaction. The genetic testing using the tip of the mouse tail was done according to the protocols of The Jackson Laboratory. Animals were fed with standard mouse chow and allowed free access to tap water. All animal procedures conformed to the guidelines for care and handling of animals established by the US Department of Health and Public Services and published by the National Institutes of Health. The local authority, LAGetSi Berlin, approved the experimental protocol.
After the animal was killed by craniocervical dislocation, the kidneys were immediately removed and sliced along the corticomedullary axis. The Af were prepared at 4°C in DMEM, which was enriched with 0.1% albumin. The dissection procedure of the Af was the same as described before (26). Special care was taken to preserve the macula densa and adjacent parts of the tubule to prevent variability in the influence of the macula densa on arterioles. The arteriole with its intact glomerulus was transferred into a thermoregulated chamber (volume 1.5 ml; VETEC) on a stage of an inverted microscope (Axiovert 100; Carl Zeiss). The perfusion system allowed movement and adjustment of concentric holding and perfusion pipettes (Luigs and Neumann). Pipettes were produced with the help of a self-manufactured apparatus using custom glass tubes (Drummond Scientific). The holding pipette, into which the proximal end of the arteriole was aspirated, had an aperture of roughly 26 μm at the tip and a constriction of about 20 μm after customizing. The inner perfusion pipette with a diameter of 5 μm was advanced into the lumen of the arteriole. It was connected to a reservoir containing the perfusion solution and to a manometer. Af of mice were perfused with pressures between 60 and 80 mmHg (unless otherwise noted) in the pressure head, which just opened the vessels. Although eNOS(−/−) mice develop a mild hypertension (36, 46), perfusion pressure was the same for all series. It was shown that the response to vasoactive substances was not affected by the perfusion pressure (30 or 90 mmHg) in Af (51). Therefore, the perfusion pressure should not be an important determinant of the arteriolar response. To test for arteriolar remodeling and differences in angiotensin receptor expression, we determined the media-to-lumen ratio as well as expression of AT1A, AT1B, and AT2 in eNOS(−/−) and nNOS(−/−) mice and their respective controls. Elevated or reduced systemic blood pressure has not been described for nNOS(−/−) mice (3).
If perfusion was not achieved within 120 min after the mouse was killed, the experiment was terminated. After perfusion started in the heated chamber (37°C), 20 min were allowed for adaptation. Only arterioles with a remaining basal tone were used. Hypoxic or otherwise injured vessels were readily identified by pronounced vasodilatation and failed complete constriction to K+-rich solution. This was used to test the viability of the arterioles at the beginning of the experiment. After the test, a recovery of 10 min was allowed. Basal values (control) for the luminal diameter were obtained at the end of this period. In all series, the last 10 s of the treatment period were used for statistical analysis of steady-state responses.
Each experiment of all series was carried out using a separate dissected Af. Only one Af was used per animal.
Experimental Protocols in Perfused Arterioles
Functional test for macula densa cells.
A proper function of macula densa cells is important for the interpretation of the results in the preparation of the juxtaglomerular apparatus used in this study. It was shown that angiotensin II induces NO release from macula densa cells (22). Therefore, we studied basal as well as angiotensin II-induced NO release of the macula densa by use of the nNOS-specific NOS-inhibitor 7-NI as a functional test of macula densa in our preparation. Perfusion pressure in the pressure head was 100 mmHg in these series. 7-NI (10−5 mol/l) was administered for 15 min followed by a cumulative application of angiotensin II (10−12 to 10−6 mol/l) together with 7-NI in C57BL/6J B6 mice (n = 6). In a control series, we measured the dose response for angiotensin II (10−12–10−6 mol/l) without 7-NI (n = 6).
Influence of nonselective inhibition of NO synthesis on arteriolar tone and angiotensin II response in wild-type controls.
In Af from eNOS(+/+) and nNOS(+/+) mice, the cumulative dose response to angiotensin II was determined at concentrations ranging from 10−12 to 10−6 mol/l. Af diameters were measured before and during the respective angiotensin II dose. Each treatment period was 2 min. Further experiments in Af from eNOS(+/+) and nNOS(+/+) mice were performed to determine how NO modulates the arteriolar reactivity to angiotensin II. The bath solution was enriched with l-NAME at a concentration of 10−4 mol/l and, 10 min later the dose-response to angiotensin II was determined in the presence of l-NAME.
Influence of NO derived from eNOS and nNOS on angiotensin II responses.
To test the differential effect of total, chronic lack of NO production by either eNOS or nNOS on arteriolar reactivity, Af from eNOS(−/−) and nNOS(−/−) mice were used. The reactivity of Af was determined by adding angiotensin II in concentrations of 10−12 to 10−6 mol/l. To assess the effect of NO derived from remaining NOS in eNOS(−/−) or nNOS(−/−) mice, further experiments in Af from eNOS(−/−) and nNOS(−/−) mice were performed; l-NAME (10−4 mol/l) was added to the bath solution, and the dose-response curve for angiotensin II was determined.
Experiments were recorded on SVHS video tapes (video recorder AG-MD 830; Panasonic). The magnification results from an objective (×40, Carl Zeiss, Germany) and projection (×1) on a 0.3″ chip digital camera (CB-3803S, GKB). Video sequences were digitized using a frame-grabber card (UDT 55-LC-EZ-50; Data Translation). The vessel diameters were determined using customized software (Dr. H. Siegmund, Institute of Physiology, Universitätsmedizin Berlin Charité, Berlin, Germany). The equipment allowed a resolution of 0.2 μm of the vessel structures. The luminal diameter of the arterioles was determined during the last 10 s of the treatment and control periods.
Whole kidneys were fixed in buffered formaldehyde (pH 7.4, 4%), dehydrated, and embedded into paraffin. From the paraffin-embedded tissues, 2-μm sections were cut with a Leica microtome (Leica Microsystems, Wetzlar, Germany) and collected onto capillary gap microscope slides (7.5 μm) for use with TechMate-500 immuno-stainer (Biotec Solutions, Santa Barbara, CA). Staining with hematoxylin and eosin was performed by standard procedures. Sections were also stained with labeled antibodies against smooth muscle actin (DAKO, Glostrup, Denmark). For further analysis, the slides were mounted on an inverted microscope (Axiovert 35; Carl Zeiss, Germany) and the sections were scanned for Af using an immersion objective with a magnification ×100 and oculars ×12.5. Stained Af were identified by their position within the cortex and their situation in relation to the glomeruli and to the interlobular arteries. The images were digitized using a projection (×1) on a 0.3″ chip digital color camera (model CC-8703; GKB) and a frame-grabber card (all in wonder Radeon, ATI Technologies, Markham, ON, Canada). The digitized pictures had a format of 480 × 640 pixels with a calibration of 8.7 pixels/μm. The vessel diameters were determined using customized software (Dr. R. Mrowka, Institute of Vegetative Physiology, Universitätsmedizin Berlin Charité, Berlin, Germany). Stained Af were identified by their position within the cortex and their situation in relation to the glomeruli and to the interlobular arteries. The areas of the media (stained smooth muscle cells) and of the lumen were determined. All vessels are seldom perfectly cross-sectioned in histological slides. Therefore, the part of the arteriole with the smallest media diameter was used for the computer-aided analysis. Vessels with nonregular shape, which did not allow the measurement of the outline and inline diameters of the stained media of the vessels, were discarded. In the present study, immersion-fixed tissue was used because infusion fixation in mice showed higher variability of the size of the arterioles in our hands. Immersion-fixed tissue is acceptable for measuring medial thickness (35). Since samples of both groups were treated with the same method, comparisons are also valid. Slides were analyzed in a blinded manner.
Isolation of preglomerular vessels for receptor expression analysis.
Af of mice were isolated using a modified iron oxide-sieving technique according to Chaudhari and Kirschenbaum (8). Basic modification consisted in the perfusion of the kidneys, which was performed via cannulation of the aorta, and the use of smaller sizes for the needles used for separation of the tissue, as well as of smaller pores of the sieves (100 and 80 μm). The isolated vessels were transferred to a tube with 1,000 μl Trizol and immediately stored at −80°C.
Analysis of angiotensin receptor mRNA expression.
RNA was isolated using Trizol reagent and reverse transcribed using Superscript and random hexamers (Invitrogen). Quantitative PCR analysis was performed using a GeneAmp 5700 (Applied Biosystems). SYBR Green was used for the fluorescent detection of DNA generated during the PCR. The PCR reaction was performed in a total volume of 25 μl with 0.4 pmol/μl of each primer, and ×2 SYBR Green master mix (Applied Biosystems); 2 μl cDNA corresponding to 10 ng RNA was used as template. Primers, which bridge at least one intron, were designed for PCR amplification on the base of published sequences for mouse AT1A (NM_177322): AT1afw, 5′-GAT TGG TAT AAA ATG GCT GG-3′; AT1arev, 5′-TCT GGG TTG AGT TGG TCT CA-3′; AT1B (NM_175086): AT1bfw 5′-CAC TGT AGA TGG GGA GCA GCC AA-3′; AT1brev 5′-GGG AGT AGG GAT CAT GAC AA-3′; and AT2 (NM_007429): AT2fw 5′-GCT TAC TTC AGC CTG CAT TT-3′; AT2rev 5′-GGA CTC ATT GGT GCC AGT TG-3′. The expression levels of angiotensin II receptor mRNA were analyzed according to the ΔCt-method. The reference gene was β-actin.
Assessment of luminal diameter of isolated perfused arterioles.
Diameter of arterioles was obtained from the mean of five measurements during the control situation and at the end of the treatment periods. The Brunner test for nonparametric analysis of longitudinal data (6) was used to test for dose-dependent changes in the arteriolar diameter for differences between two groups and to check for differences in the dose-dependent changes in diameter between the groups. The Mann-Whitney U-test was applied for the comparison of independent measurements and was used for comparison of diameters of the different groups in the steady state during the control situation and for comparison of mRNA expression. Data are presented as mean ± SE.
Determination of the media-to-lumen ratio.
Sections from 10 animals of each group [eNOS(−/−), nNOS(−/−)], and their controls were scanned for Af. In each group, 230 Af were obtained. Outer and luminal diameters of Af were measured three times at representative sites of the vessels. The thickness of the media was calculated, and the areas for the media, as well as for the lumen, and their ratio were computed. The mean of all these parameters was obtained. To compare media-to-lumen ratio of corresponding groups, we calculated histograms. Classes were 0<“1”≤1, 1<“2”≤2… 9<“10”≤10 and spanned over the same data range in all groups. Differences in the histograms were checked using Brandt-Snedecor's χ2-test for comparison of arbitrary distributions (33). The Mann-Whitney U-test was applied to test the null hypothesis regarding diameters and ratios of knockout mice and their controls. The confidence level of P was set to 0.05 for all statistical tests.
Functional Test for Macula Densa Cells
The specific nNOS inhibitor 7-NI at 10−5 mol/l did not change the arteriolar lumen during the control situation within 15 min of application. However, the angiotensin II response was significantly stronger in 7-NI-treated (n = 6) Af compared with the nontreated arterioles (n = 6; Fig. 1).
Influence of nonselective inhibition of NO synthesis on arteriolar tone and angiotensin II response in wild-type controls.
Without l-NAME, angiotensin II reduced Af diameter of eNOS(+/+) dose dependently to 69.4 ± 8.0% and 67.6 ± 6.5% of control diameter at angiotensin II concentrations of 10−8 and 10−6 mol/l, respectively (Fig. 2; right, n = 8). l-NAME at a concentration of 10−4 mol/l, which is known to completely block NOS isoforms, did not change the arteriolar lumen during the control situation within 10 min of application in eNOS(+/+) (Fig. 2, left; n = 7). However, the angiotensin II response was increased; luminal diameters were reduced to 23.1 ± 8.6% and 12.7 ± 6.2% of control at 10−8 and 10−6 mol/l angiotensin II, respectively.
Without l-NAME, angiotensin II reduced Af diameter of nNOS(+/+) mice to 58.5 ± 6.0% and 50.1 ± 5.2% at 10−8 and 10−6 mol/l, respectively (Fig. 2, right; n = 8). l-NAME did not change arteriolar diameters significantly in nNOS(+/+) (Fig. 2 left, n = 7), but increased the angiotensin II-induced contractions (7.4 ± 4.9% and 9.9 ± 6.5% at 10−8 and 10−6 mol/l, respectively). The angiotensin II response in nNOS(+/+) during l-NAME showed a steeper course than in eNOS(+/+) during l-NAME, suggesting higher angiotensin II sensitivity under these conditions.
Influence of NO derived from eNOS and nNOS on angiotensin II responses.
Arteriolar diameters during control conditions did not differ between eNOS(−/−) mice (n = 7) and their respective controls [eNOS(+/+), n = 8; Fig. 3. left]. However, angiotensin II reduced arteriolar diameters much more in eNOS(−/−) than in the control group: at 10−8 and 10−6 mol/l, angiotensin II reduced diameters to 11.9 ± 8.1% and 8.7 ± 5.8% in the eNOS(−/−) group and to 69.4 ± 8.0% and 67.6 ± 6.5%, respectively, in eNOS(+/+) (Fig. 3 right). l-NAME (10−4 mol/l) did not reduce diameters significantly in eNOS(−/−) during the control situation. There was no difference in luminal diameters between the l-NAME-treated (n = 10) and untreated group (Fig. 3, left). With l-NAME, angiotensin II almost totally constricted Af of eNOS(−/−) at 10−8 to 10−6 mol/l (Fig. 3). Because angiotensin II had greatly reduced Af diameter of eNOS(−/−) even without l-NAME, the difference between these groups did not reach statistical significance.
Arteriolar diameters during control conditions did not differ between nNOS(−/−) mice (n = 7) and their respective controls [nNOS(+/+), n = 8; Fig. 4, left]. Angiotensin II reduced arteriolar diameters much more in nNOS(−/−) than in the control group, at 10−8 and 10−6 mol/l. Angiotensin II reduced diameters to 40.1 ± 8.4% and 24.8 ± 9.4% in nNOS(−/−), and 58.5 ± 6.0% and 50.1 ± 5.2% in control mice, respectively (Fig. 4, right). This angiotensin II response was weaker in nNOS(−/−) compared with eNOS(−/−). l-NAME did not reduce basal diameters significantly in nNOS(−/−). There was no difference in luminal diameters between the l-NAME-treated (n = 6) and untreated group (Fig. 4, left). With l-NAME, the angiotensin II response in nNOS(−/−) was markedly increased; Af completely constricted at 10−9 to 10−7 mol/l angiotensin II.
Histograms of media-to-lumen ratios of Af from both eNOS(−/−) and eNOS(+/+) showed skewed distributions with peak values at a ratio of four (Fig. 5). Distributions of media/lumen ratios did not differ between these two groups. Also, in nNOS(−/−) and nNOS(+/+) mice, the ratios showed skewed distribution and did not differ between both groups. For each group, n = 230 Af were obtained from 10 mice.
Angiotensin II receptor mRNA expression.
Expression of mRNA for AT1A, AT1B, and AT2 was determined by quantitative PCR analysis in preglomerular vessels for eNOS(−/−) (n = 10), eNOS(+/+) (n = 11), nNOS(−/−) (n = 11), and nNOS(+/+) (n = 10). Expressions of all receptor types did not significantly differ between the NOS-deficient mice and their respective wild-type controls (Fig. 6).
The study shows that NO derived from both the endothelial and the neuronal isoforms of NOS contribute to counteracting angiotensin II constrictor effects in Af. However, the data suggest a greater role for endothelium-derived NO than for nNOS-derived NO.
It is known that angiotensin II can induce NO release in Af of mice (17, 27, 28). The stimulation of NOS by angiotensin II is mediated by AT1 receptors in the juxtaglomerular apparatus (22, 27). The present results indicate that angiotensin II-induced NO release plays a major role in Af; whereas l-NAME per se did not reduce luminal diameters significantly, it markedly enhanced angiotensin II-induced contractions. This activation of NOS by angiotensin II is most likely specific, since contractions induced by other vasoconstrictors, such as norepinephrine are not strengthened by l-NAME (27).
It is well known that in vivo administration of l-NAME results in significant increase in total renal vascular resistance. Because Afs markedly contribute to renal vascular resistance, it may appear surprising that l-NAME per se at 10−4 mol/l (a concentration assumed to completely inhibit NOS) did not significantly decrease Af diameter in the present study, although there was a trend to smaller diameters in wild-type mice. In earlier studies in isolated rabbit Af by Ito and colleagues, 10−4 mol/l l-NAME decreased basal Af diameter significantly by 18% (17) or 34% (19). Also, studying rabbit Af at 10−4 mol/l l-NAME, Wang et al. (49) reported a 12% decrease, whereas Uhrenholt et al. (44) did not observe significant constrictions in rabbits. In rats, the same dose resulted in Af diameter reductions ranging from 15 to 30% in juxtaglomerular nephron preparations (5, 11, 29), whereas no significant reductions were observed in hydronephrotic kidney preparations (43). Studying isolated mouse Af, Hansen et al. (15) found a 25% diameter reduction by 10−4 mol/l l-NAME in wild-type mice of the 129J/C57BL6 genetic background. Albeit the wild-type mice of our present study are of similar background, we did not find a significant diameter reduction, thus corroborating earlier results of ours in C57BL6 mice (28). However, in another study, we observed a significant 17% diameter reduction in Af of wild-type mice of the NMRI background (26), demonstrating that our preparation is suitable for detecting possible l-NAME effects on basal tone of Af. Taken together, the rather inconsistent findings by various groups may not easily be ascribed to differences in species or strain or differences in preparations. Thus, the role of NO for basal tone of Af remains to be defined.
The finding of a recent study (28) that angiotensin II responses in Af of eNOS(−/−) mice are similar compared with that of l-NAME-pretreated wild-type mice raised the hypothesis that endothelial-derived NO plays a prominent role in counteracting angiotensin II effects. Results of the present study confirm this hypothesis. First, angiotensin II-induced contractions were clearly stronger in eNOS(−/−) mice than in nNOS(−/−) animals, while responses did not differ significantly between the respective control wild types. Second, inhibition by l-NAME of the remaining NOS activity in nNOS(−/−) mice resulted in markedly stronger angiotensin II-induced contractions, indicating a significant NO production by eNOS in nNOS(−/−) mice. In eNOS(−/−) mice, l-NAME had a markedly weaker effect; the difference between l-NAMEtreated and untreated Af did not reach statistical significance. Angiotensin II-induced contractions in eNOS(−/−) mice were already almost complete without l-NAME; thus, the remaining NO production by nNOS appears to play a minor role.
In the kidney, nNOS mRNA expression is mainly located in macula densa cells (23). NO released by theses cells modulates the TGF. It has been shown that NO release increases with activation of this feedback control and may limit afferent arteriolar contraction (for reviews, see Refs. 18, 20, and 25). In a preparation of isolated juxtaglomerular apparatus in which Af and the macula densa were simultaneously perfused, acute inhibition of nNOS by 7-NI potentiated the TGF response, whereas the response was basically unchanged in nNOS(−/−) mice (31). However, our data from nNOS(−/−) mice indicate that chronic lack of NO production by nNOS modulates angiotensin II-induced contractions of Af, and may, therefore, also reset the TGF. This is in line with results in nNOS(−/−) mice obtained by measurements of stop-flow pressure (45) where resetting of the TGF was observed. In the present study, the macula densa was not perfused and therefore faces the relatively high NaCl concentration of the bath solution. Stimulation of macula densa nNOS can be expected. Treatment with the specific nNOS inhibitor 7-NI did not change basal diameters, but angiotensin II-induced constriction of Af was enhanced. This observation indicates a role of angiotensin II for NO release in macula densa cells. NO liberation by angiotensin II via AT1 receptors has also been described in the perfused macula densa of rabbits (22).
The angiotensin II response of mice Af observed in the present as well as in previous studies (27, 28) is clearly smaller than in rabbit Af (19, 50). In rabbits, angiotensin II concentrations of 10−8 mol/l reduced the luminal diameter of Af to ∼20% in one study (50) and to 40% in an other (19), while in wild-type mice of the present study a reduction to 60–70% was reached by this angiotensin II concentration. In rats, Yuan et al. (51) observed maximal contractions of Af at 10−7 mol/l angiotensin II; at this concentration, the arteriole nearly completely constricted. In NMRI mice, angiotensin II responses are slightly stronger than in wild-type mice of the present study: 10−8 mol/l angiotensin II reduced arteriolar diameters to 55% in NMRI mice (26). It seems that responses to angiotensin II vary more between species than between mouse strains. The reasons for species differences are not clear. One may speculate that the smaller angiotensin II response in mice may rely on a stronger activation of the NO system in this species compared with rats and rabbits.
Gene knockout models as used in the present study are not only characterized by the innate lack of the respective gene's product, but also by mechanisms of adaptation or compensation. For instance, it has been described that NOS isoforms can partly replace one another functionally (7, 24, 39). Thus, it is conceivable that nNOS and inducible NOS (iNOS) are more active in eNOS(−/−) mice, and, vice versa, eNOS and iNOS are more active in nNOS(−/−) mice. The present results do not support the notion of a significant functional replacement of NOS isoforms because angiotensin II responses were clearly enhanced in both eNOS(−/−) and nNOS(−/−) mice. NO derived from iNOS may not play a significant role in the present study, at least in wild-type controls. As shown in several studies, transcriptional activity for iNOS is normally low but readily enhanced by appropriate stimuli in mesangial cells, vascular endothelium in the renal cortex, and in the macula densa (10, 13, 34, 41). Without stimulation, the effects of iNOS-derived NO are small in the renal cortex (1).
Long-term changes in the NO concentration may alter the structure of vessels. For instance, the NO-donor CAS-1609 induces proliferation of endothelial cells but inhibits the proliferation of smooth muscle cells as induced by the platelet-derived growth factor-BB (14). Pharmacological studies suggest that eNOS plays a role in the regulation of vascular smooth muscle remodeling in response to blood flow changes (42). In another study, external carotid artery ligation resulted in a paradoxical increase in arterial wall thickness and a hyperplastic response in eNOS knockout mice compared with wild-type mice (32). These studies suggest that long-term changes in NO levels can alter arterial wall structure, and may, thus, be the reason behind the altered angiotensin II-responses of Af observed in eNOS(−/−) and nNOS(−/−) mice. Therefore, we determined media/lumen ratios of Af and found no difference between NOS deficient mice and their respective wild-type controls. The seeming discrepancy between our finding and those of others may be related to the different experimental models, e.g., the use of small resistance vessels in the present study and larger conductive vessels (carotid arteries) by Rudic et al. (32) and Tronc et al. (42). The lack of differences in the Af media-to-lumen ratio between knockout mice and their wild-types controls indicates that higher sensitivity for angiotensin II in both knockout strains is not due to structural changes of Af in our investigation.
Furthermore, the stronger response to angiotensin II in NOS-deficient mice compared with their wild-type controls is most probably not due to a differential upregulation or downregulation of angiotensin II receptors. Analysis of the mRNA-expression of AT1A, AT1B, and AT2 receptors allows assessing long-term influences on angiotensin receptor expression. We did not find significant differences in mRNA of these receptors between Af obtained in NOS-deficient mice compared with wild-type mice. Receptor expression at the protein level could not been determined, since receptor protein concentration is below the detection limit in the isolated microvessels.
In summary, this study shows a significant contribution of NO production by both eNOS and nNOS in the control of Af. This is concluded from greater angiotensin II responses in mice lacking eNOS or nNOS. Furthermore, angiotensin II stimulates both isoforms of NOS. The endothelial production of NO is more important than that by nNOS in this model of isolated, perfused Af.
The excellent technical assistance of Uta Stangenberg, Jeannette Werner, and Christiane Wehner is gratefully acknowledged.
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. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society