点击显示 收起
【摘要】 The aim of this study was to investigate ANG II-induced Ca 2+ signaling in freshly isolated afferent arterioles (AA) from two-kidney, one-clip hypertensive (2K1C) rats, which have an elevated plasma and renal ANG II level, and different perfusion pressure and vascular tone in the clipped and nonclipped kidney. The Ca 2+ responses in vessels from 2K1C and control rats were similar in all groups ( P 0.1). The intracellular Ca 2+ (Ca i 2+ ) response in the afferent arteriole after 10 -8 M ANG II stimulation was 0.57 ± 0.10, 0.50 ± 0.07, 0.48 ± 0.04, and 0.36 ± 0.05 in the control, sham, nonclipped, and clipped kidney, respectively. These data were consistent with the finding of unchanged AT 1a R mRNA levels in AAs from all groups. Although the absolute values were similar, the dose-response curves to ANG II were different. In the control, sham, and nonclipped kidney from 2K1C, the dose-response curve leveled off between 10 -8 and 10 -6 M ANG II. In the clipped kidney, the dose-response curve was linear, with a significantly increased response at 10 -6 M compared with 10 -8 M ANG II ( P < 0.05). Inhibition of cyclooxygenase-1 (COX-1) with indomethacin enhanced the ANG II response in the nonclipped ( 0.30 ± 0.09) and clipped ( 0.30 ± 0.09) kidneys from 2K1C ( P < 0.005), but not in control rats ( -0.02 ± 0.11, P 0.8). Conclusively, the ANG II-induced Ca i 2+ response was reduced by COX-1-derived prostaglandins in 2K1C, in contrast to control animals, where the COX-1 inhibition had no effect. COX-2 inhibition with NS-398 did not increase the ANG II-mediated Ca i 2+ response in any of the groups.
【关键词】 calcium signaling/physiology Fura indomethacin NS cyclooxygenase inhibitors receptors AT a R
TWO - KIDNEY, ONE - CLIP (2K1C) is a model of experimental renovascular hypertension and is established by increased activity of the renin-angiotensin system induced by a stenosis of the unilateral clipped kidney. The plasma renin activity and circulating ANG II levels are increased, and the systemic blood pressure (BP) is elevated after 3-4 wk. Removal of the clip in the early phase of the disease normalizes the systemic BP ( 15 ).
The renal content of renin is different in the two kidneys of 2K1C. The clipped kidney has a high level of renin, whereas the nonclipped kidney harbors small amounts of this enzyme ( 14 ). However, measurements performed on renal interstitial fluid show that ANG II levels are increased both in the clipped and nonclipped kidneys during the development of renovascular hypertension ( 21 ).
In the nonclipped kidney, the afferent arterioles are contracted due to the higher perfusion pressure induced by the high levels of circulating ANG II, while in the clipped kidney, the resistance vessels are dilated due to the pressure drop over the clip ( 14 ). Although the perfusion pressure behind the clip is similar to the normotensive kidney, an additional resistance seems to be interpolated between the clip and the afferent arterioles ( 27, 36 ). A support of this finding is reduced glomerular capillary pressure ( 36 ), dilated afferent arterioles, and lack of renal blood flow (RBF) autoregulation ( 14 ) in the clipped kidney. As a consequence, the vascular tone in afferent arterioles (AAs) is different in these two kidneys, and calcium signaling induced by ANG II may therefore be different. Because of current methodological limitations, the vessels had to be studied after removal from the animal. However, only properties retained through the isolation procedure that are visible ex vivo can eventually be observed in the isolated arterioles. Thus this limitation should be taken into account when evaluating data from the present study. The renal microvessels were isolated with the agarose infusion/enzyme treatment technique developed by Loutzenhiser and Loutzenhiser in 2000 ( 17 ). The technique permits the isolation of preglomerular vessel networks up to a length of 400 µm, containing an intact endothelium. The arterioles are not sieved, centrifuged, or shaken during the current isolation technique, and the vessels are filled with an elastic agarose gel that prevents structural deformation during contraction. We have earlier found this preparation well suited for measurements of changes in intracellular calcium ( 11 ). To our knowledge, Ca 2+ signaling in vascular smooth muscle cells from the clipped and nonclipped kidney in 2K1C has not been studied.
Studies from our laboratory recently showed decreased responses in RBF to renal ANG II injections in the nonclipped kidney of 2K1C compared with control animals, although the preglomerular AT 1a R levels ex vivo seem to be unchanged ( 2 ). The reason for this apparent dysregulation is unclear, and further investigations are needed. In the present study, we tested the hypothesis that ANG II-induced calcium signaling and AT 1a R regulation were different in the clipped and nonclipped kidneys of 2K1C, due to the differences in perfusion pressure in these two kidneys. The endogenous ANG II concentration in 2K1C is considerably higher than in normotensive rats and may affect ANG II-induced Ca 2+ signaling in 2K1C.
METHODS
Animals. Twenty-seven male Wistar rats aged 8 wk were obtained from Taconic M&B. Five animals were kept in each cage and fed ordinary rat chow containing 0.5% sodium, 0.6% potassium, 0.71% calcium, and 14.7% crude protein. The rats had free access to tap water. The experiments were performed under the approval of the Norwegian State Board for Biological Experiments with Living Animals.
Experimental groups. Thirteen rats were randomly assigned to the 2K1C hypertensive group, six rats were randomly assigned to the sham group (Ca i 2+ measurements only), and seven rats were used as the control group. After being clipped, the animals were followed by weekly measurements of systolic BP and body weight.
Induction of 2K1C hypertension. During pentobarbital sodium anesthesia (50-70 mg/kg), the right kidney was exposed through a lumbar incision. The right renal artery was clipped by placing of a rigid U-shaped silver clip with an internal opening of 0.25 mm. The animals usually develop hypertension after 2-3 wk. The animals in the sham group underwent the same surgical procedure as the 2K1C group, including exposure of the renal artery by removal of fat and connective tissues.
Measurements of systolic BP. BP was measured in conscious rats using the tail-cuff method (W+W Electronics, Ugo Bazile) after the animals were prewarmed in a warming cabinet for 10 min at 35°C.
Isolation of preglomerular renal vessels. One or two animals were randomly chosen each day and killed for Ca 2+ and receptor studies during a 2-wk period, starting 4 wk after clipping. Renal preglomerular vessels were isolated after pentobarbital sodium anesthesia (50-70 mg/kg). Ligatures were prepared 10 mm above the femoral bifurcation of the aorta, 2-3 mm posterior of the left renal artery, and 2-3 mm anterior of the right renal artery. The cannula was inserted into the aorta so the distal ligature could be tightened to avoid bleeding. The cannula was pushed upstream slightly above the middle ligature, and then the middle and proximal ligatures were tightened. Simultaneously, the kidneys were flushed with 5-10 ml warmed Roswell Park Memorial Institute solution (RPMI) without Ca 2+ (37°C) for 3-5 s, immediately followed by infusion of 3-4 ml Seaprep agarose solution (2%) in Ca 2+ -free RPMI (37°C) to create an elastic core of agarose inside the renal microvessels. About 100-µm-thick slices were cut from the cortex of the kidney with a Thomas slicer (Thomas Scientific) and incubated for 30 min at 32-33°C in 10 ml Ca 2+ -free RPMI with 5 µg collagenase (Sigma C5138, 246 U/ml) and 5 µg protease (Sigma P3417, 0.5 U/ml). Afferent fragments were picked with a pressure-controlled pipette (diameter = 100 µm) and transferred to acid-washed coverslips in a perfusion chamber. The microvessels usually attached strongly to the cover glass. The arterioles were loaded in 1.25 µmol/l fura 2-acetoxymethyl ester in RPMI at room temperature for 45 min. Thereafter, fura-2 was removed and the cells were incubated for 30 min (30°C) to ensure complete hydrolyzation of the fura-2 ester. The cells were kept at 30°C for up to 2 h before recording.
Superfusion of vessels in chamber. The perfusion chamber with a volume of 400 µl was gravity fed (2 ml/min) through a perfusion inline heater (Warner TC344-B) which maintained the temperature in the chamber at 36-37°C. The switching between perfusion fluids was done automatically with a Valvebank8 (AutoMate Scientific). The microvessels were initially superfused with normal RPMI for 150 s during the control period followed by superfusion of ANG II (10 -10, 10 -8, or 10 -6 M) for 150 s. Thereafter, the vessels were perfused with RPMI for another 150 s to wash out the hormone. Before superfusion was started, vessels were treated with indomethacin (10 -7 M) or NS-398 (10 -7 M) for 15 min (30°C) to study COX-1 and COX-2 inhibition, respectively. A concentration of 10 -8 M ANG II was chosen for stimulating the indomethacin and NS-398-treated vessels, as pilot experiments indicated that this concentration produced 50% of maximum Ca i 2+ response in the clipped kidney and would therefore be suitable to visualize a possible effect of COX inhibition in both the nonclipped and clipped kidneys.
Measurement of intracellular fura 2 ratio. The method for the fura ratio measurements and the identification of vessels have been described before ( 11 ). In short, the fura 2 ratio was measured using an inverted Olympus IX-70 with a x 40 UAPO objective. The cells were excited alternatively from a dual-excitation wavelength system (Delta-Ram) from Photon Technologies (PTI), with lights of 340- and 380-nm wavelengths After the signals passed through a barrier filter (510 nm), three fluorescence images per second were recorded by an IC-200 intensified CCD camera, resulting in a final time resolution of 1.6 Hz for the Ca i 2+ ratios. The ratio images were analyzed with ImageMaster 1.49 Software from PTI, and regions of interest (ROI) on the vessels were defined with the software to accurately collect the fura-2 fluorescence from the afferent segments. The ROIs were created at a minimum distance of 20 µm from the branching points between AA and ILA. Background fluorescence from a region adjacent to the sample was subtracted from the recordings. The vessels were continuously recorded from 150 s before ANG II addition to the perfusion bath, until 150 s after wash-out of the hormone. The initial response was defined as the maximum Ca i 2+ ratio during the 5 first s after stimulation. The sustained ratio was defined as the Ca i 2+ ratio 30 s after the stimulation. The initial ratio increase was calculated as the difference between initial and baseline Ca i 2+ ratio. The sustained ratio increase was calculated as the difference between sustained and baseline Ca i 2+ ratio.
Real-time PCR for measurements of AT 1a receptor mRNA on isolated AA. Quantification of AT 1a receptor mRNA was done by real-time PCR. Vessels were isolated as described above and resuspended in Cell to Signal lysis buffer from Ambion. Fifteen vessels from each animal were resuspended in 50 µl of buffer. The AT 1a R is transcribed by at least two different splicing variants both represented in the kidney. The protein encoding part, however, is encoded by one exon. To cover all different splicing variants, primers and probes were constructed against the protein encoding exon 3. To avoid amplifying genomic DNA, the samples were subjected to DNase Turbo (Ambion) treatment. The samples were added 40 U of RNasin (Promega) before DNase treatment. The DNase was removed by DNase inactivation agent (Ambion) before cDNA synthesis. First-strand cDNA was synthesized directly using chemicals from the Cells to Signal kit and primed by pd (N) 10 primers. Each cDNA synthesis was performed in a total volume of 20 µl. The amplification was done by a nested PCR. The outer primers for amplification of AT 1a R were selected for 208 bp (forward primer 5'-ACCGCTATGGAATACCGATG-3', reverse primer 5'-ccagccattagccagatgat-3'). The samples were then amplified by 12 PCR cycles. One microliter from this amplification was thereafter used for further AT 1a R mRNA quantification with real-time PCR. 18S rRNA was quantitated directly by real-time PCR amplification of cDNA without the use of nested PCR. All real-time PCR amplifications were done at the same time. Primers for real-time amplification of AT 1a R were selected for a 70-bp sequence, 5'-tcgctacctggccattgt-3' forward primer, 5'-aggtgactttggctaccagcat-3' reverse primer, and 5'-acccaatgaagtctcgcctccgc-3' Taq Man probe marked with 5'-FAM and 3'-TAMRA. The amplified AT 1a R cDNA was normalized against amplified 18S ribosomal RNA to compensate for any changes due to RNA degradation, reverse transcriptase efficiency or amplification success. The primers for 18S were made for a 68-bp fragment, 5'-agtccctgccctttgtacaca-3' forward primer, 5'-gatccgagggcctcactaaac-3' reverse primer, and 5'-cgcccgtcgctactaccgattgg-3' Taq Man probe marked with 5' Yakima Yellow and 3' TAMRA.
The amounts of AT 1a R and 18S cDNA were quantified using a standard curve for known quantities of AT 1a R and 18S DNA. The AT 1a R standard curve was made by amplifying a 208-bp region of the AT 1a R cDNA with the primers identical to the outer primers in the nested PCR. For the 18S standard curve, a 396-bp region of the rat 18S RNA cDNA was amplified using primers 5'-ttcagccaccgagattgagc-3' (forward) and 5'-cgcaggttcacctacggaaa-3' (reverse). The amplification products were thereafter cloned into pBAD TOPO TA vectors and transfected into top 10 Escherichia coli cells (Invitrogen). Plasmids containing the cloned material were purified from bacterial cultures using a Qiagen Plasmid Purification Midi kit. The purified plasmids were diluted to concentrations appropriate for the standard curve. The primer and probe constructions were done using Primer Express software from Applied Biosystems. The quantification was done on an ABI PRISM 7900 HT Sequence Detection System (Applied Biosystems) and with a qPCR Core Kit (Eurogentec). The primer concentrations were optimized before use in quantification. Forward primers for both AT 1a R and 18S were used in a final concentration of 0.3 µM. Reverse primers for both AT 1a R and 18S were used in a final concentration of 0.9 µM. All amplifications of both AT 1a R and 18S RNA were done using three parallel amplification reactions under standard ABI conditions using 19-µl reaction volumes.
Chemicals. All chemicals used in this experiment were from Sigma, except fura-2 acetoxymethyl which came from Molecular Probes. The RPMI media contained (in g/l) 7.65 NaCl, 0.40 KCl, 0.203 MgCl 2, 0.20 NaH 2 PO 4, 1.34 HEPES, 1 glucose, 0.11 Na Pyruvat, 0.35 CaHCO 3, 0.22 CaCl 2, RPMI vitamins (Sigma R7256), and amino acids (Sigma R7131).
Statistics. The data were presented as means ± SE. Calculation of unpaired two-sided t -test and ANOVA using Student-Newman-Keuls post hoc test were performed in SPSS 12.0. P values 0.05 were considered statistically significant.
RESULTS
The systolic BP increased continuously after clipping and became significantly different from the control group after 3 wk as shown in Fig. 1. Four to 6 wk after clipping, arteriolar fragments were isolated from one animal each day randomly chosen from the control or 2K1C groups. Systolic BP measurements were performed the same week as isolation of renal vessels was done and found to be 137 ± 1 mmHg in control rats ( n = 7) and 193 ± 10 ( n = 10) in the 2K1C group ( P < 0.01). Three animals that did not develop hypertension were discharged from the study.
Fig. 1. Increase in systolic blood pressure (BP) in control, sham-operated, and two-kidney, one-clip hypertensive (2K1C) rats during the development of renovascular hypertension. * P < 0.05 vs. control and sham-operated rats.
The baseline Ca i 2+ values were 0.72 ± 0.02, 0.68 ± 0.04, 0.75 ± 0.02, and 0.75 ± 0.02 in the controls, sham-operated controls, nonclipped, and clipped kidneys, respectively, and were not significantly different ( P 0.1).
In control animals, there was a dose-related calcium response to ANG II ( Fig. 2 ), which leveled off at high concentrations of ANG II. The initial fura-2 ratio response to 10 -10, 10 -8, and 10 -6 M ANG II was 0.16 ± 0.02 ( n = 5), 0.57 ± 0.09 ( n = 6), and 0.52 ± 0.07 ( n = 6), respectively. All sustained responses were smaller than the initial Ca 2+ responses ( P < 0.05), except at 10 -6 M ANG II ( P = 0.06).
Fig. 2. Dose responses to ANG II in afferent arterioles (AA) from control Wistar ( A ), sham-operated Wistar ( B ), clipped ( D ), and nonclipped ( C ) kidneys in 2K1C hypertension. Results from peak and sustained responses are shown. * P < 0.05 vs. 10 -8 M ANG II.
In sham-operated animals, there was a dose-related calcium response to ANG II ( Fig. 2 ), which was similar to the responses obtained in normal Wistar rats. The initial fura 2 ratio responses to 10 -10, 10 -8, and 10 -6 M ANG II were 0.11 ± 0.01 ( n = 6), 0.50 ± 0.07 ( n = 6), and 0.45 ± 0.11 ( n = 6), respectively, and were similar to data obtained in control rats without sham operation. All sustained responses were smaller than the initial Ca 2+ responses ( P < 0.01), except at 10 -6 M ANG II ( P 0.4).
The Ca i 2+ tracings in vessels from the nonclipped kidney in 2K1C showed a similar response pattern as seen in the control and sham-operated kidneys ( Fig. 2 ). The initial fura-2 ratio responses in AA to 10 -10, 10 -8, and 10 -6 M ANG II were 0.09 ± 0.01 ( n = 6), 0.48 ± 0.01 ( n = 8), and 0.45 ± 0.15 ( n = 6), respectively. These values were not significantly different from the corresponding data obtained in control animals ( P 0.8). The sustained responses were lower than the preceding initial Ca i 2+ response ( P < 0.05), except at 10 -6 M ANG II ( P = 0.07).
In contrast to the control, sham, and nonclipped kidney from 2K1C animals, the ANG II dose-response curve in AAs from the clipped kidney did not level off at high ligand concentration ( Fig. 2 ). The Ca i 2+ response at 10 -6 M ANG II was significantly higher than the Ca i 2+ response at 10 -8 M ( P < 0.05). The initial fura 2 ratio responses to 10 -10, 10 -8, and 10 -6 M ANG II were 0.08 ± 0.01 ( n = 8), 0.36 ± 0.05 ( n = 8), and 0.65 ± 0.13 ( n = 6), respectively. Even though the dose-response curve was linear, the absolute values were not significantly different from the corresponding data obtained in control animals or in AAs from the nonclipped kidney from 2K1C ( P 0.1). Similar to the other kidneys, the sustained response was lower than the preceding initial Ca i 2+ response ( P < 0.05). The sustained response to ANG II was numerically increased at 10 -6 M compared with 10 -8 M, but this difference did not reach significance ( P = 0.07).
The AT 1a R mRNA level in AA was measured with real-time RT-PCR. As seen in Fig. 3, the receptor mRNA from the AA was similar in the control ( n = 3), nonclipped ( n = 4), and clipped ( n = 4) kidneys ( P 0.3). Even with the use of sensitive real-time PCR, it was necessary to preamplify the AT 1a R mRNA with nested PCR before analysis. Each sample consisted of 15 independent AAs picked from one animal.
Fig. 3. AT 1a R mRNA/10 8 18s rRNA in AA from control and 2K1C hypertensive rats. Only the AT 1a R mRNA was amplified with nested RT-PCR. Fifteen AAs were harvested from each rat, and each value was averaged from 3 (control) or 4 (2K1C) animals.
In the control kidneys from Wistar rats, pretreatment with the nonspecific cyclooxygenase (COX) inhibitor indomethacin or the COX-2-selective inhibitor NS-398 did not enhance the ANG II Ca 2+ response in AA ( Fig. 4 ). Tracings of the Ca i 2+ ratio after ANG II 10 -8 M stimulation with and without indomethacin pretreatment are exemplified in Fig. 5. The ANG II-induced Ca i 2+ responses in AA after pretreatment with indomethacin and NS-398 were 0.57 ± 0.05 ( n = 6) and 0.39 ± 0.08 ( n = 5), respectively, and these responses were not different from the response in untreated vessels ( P 0.1). Similarly, the sustained response did not change after treatment with indomethacin or NS-398 ( P 0.4).
Fig. 4. Effect of cyclooxygenase inhibition on the ANG II (10 -8 M)-induced Ca i 2+ response in AA from control, clipped, and nonclipped kidneys in 2K1C hypertension. * P < 0.05 vs. untreated. ** P < 0.005 vs. untreated.
Fig. 5. Representative Ca i 2+ tracings during stimulation with 10 -8 M ANG II demonstrating the effect of indomethacin in AAs from control and 2K1C rats.
In the nonclipped kidney from 2K1C, pretreatment with indomethacin enhanced the initial Ca 2+ response to ANG II in AA compared with untreated vessels ( Figs. 4 and 5 ). The ANG II response after pretreatment with indomethacin and NS-398 in the AA was 0.78 ± 0.08 ( n = 6, P < 0.005 vs. control) and 0.46 ± 0.13 ( n = 5, P 0.9 vs. control), respectively. The sustained response was also increased after indomethacin treatment ( P < 0.005), but not after NS-398 treatment.
The clipped kidney showed an enhanced peak Ca 2+ ratio response to ANG II stimulation compared with untreated vessels after indomethacin treatment ( Figs. 4 and 5 ). No increase was seen using NS-398. The responses in indomethacin- and NS-398-treated AA were 0.66 ± 0.05 ( n = 7, P < 0.005 vs. control) and 0.44 ± 0.08 ( n = 8, P 0.4 vs. control), respectively. The sustained Ca 2+ response was also increased after indomethacin treatment ( P < 0.05), but not after NS-398 treatment. Although COX inhibition enhanced the ANG II response in both kidneys from 2K1C, the Ca i 2+ response was similar in indomethacin-treated AA from the control, nonclipped, and clipped kidneys ( P 0.1).
The kidney weight-to-body weight ratio in control rats (0.35 ± 0.01%, n = 10) was higher compared with the ratio for the clipped kidney (0.24 ± 0.02%, n = 10, P < 0.005). The kidney weight-to-body weight ratio for nonclipped kidneys in the 2K1C rats (0.43 ± 0.04%, n = 10) was increased compared with the ratio for control kidneys ( P = 0.05). The sham operation did not influence the kidney weight. These weight differences are typical for 2K1C animals with renovascular hypertension.
DISCUSSION
Our findings could not support the hypothesis that the Ca 2+ response to ANG II is altered in freshly isolated AAs from hypertensive 2K1C rats, compared with vessels from normal or sham-operated rats. The initial and sustained increases in Ca i 2+ ratio were similar, despite the different perfusion pressure and different ANG II levels these arterioles were exposed to in the animal.
Even though the absolute Ca i 2+ responses were similar, the shape of the dose-response curves was different. In vessels from the control, sham-operated, and nonclipped kidney, the dose-response curves leveled off at the highest concentration of ANG II. In vessels from the clipped kidney, a linear dose-response curve was seen, even at high concentration of ANG II. Inhibition of the COX-1 system enhanced the calcium responses in vessels from both kidneys from 2K1C, but not in arterioles from control rats. Inhibition with the selective COX 2 inhibitor NS-398 had no effect, neither in 2K1C nor in the control animals. The Ca i 2+ responses were consistent with the AT 1a R mRNA levels, which were similar in vessels from 2K1C and control animals. This novel information may be of importance in our understanding of the development and maintenance of hypertension in the 2K1C model. To our knowledge, investigation of calcium signaling in the afferent arteriole from 2K1C hypertensive rats has not been performed before.
Renovascular hypertension, usually induced after unilateral clipping of the renal artery, is followed by a rapid increase in systemic BP that stabilizes after 4-6 wk ( 16 ). The renin-angiotensin system (RAS) is upregulated in this period and continues to be high also into the chronic phase after 6 wk of unilateral stenosis ( 21 ). The hypertensive state is dependent of a functional AT 1a R, as knockout mice lacking this receptor are unable to develop renovascular hypertension ( 4 ). The renal tissue concentration of ANG II is even higher than in plasma ( 23 ), and an active role of the AT 1a R has been suggested in the development of 2K1C hypertension ( 5 ).
Earlier reports concerning the regulation of the AT 1a R have been conflicting, as downregulation ( 10 ), no change ( 8 ), or upregulation ( 19, 33 ) of the receptor have been reported to occur in 2K1C hypertension. This might be due to different methods for isolation of the renal tissue; measuring the ANG II receptor in whole tissue preparations of the renal cortex can produce arbitrary results, as the tubular and vascular AT 1a receptors often are differently regulated ( 6, 9 ).
The endogenous ANG II plasma level might be similar in the clipped and nonclipped kidney but is definitely lower in the control kidneys. Despite this, the AAs are dilated in the clipped kidney ( 16 ), suggesting that reduced perfusion pressure overrides the effect of ANG II. The nonclipped kidney has been shown to be nitric oxide (NO) dependent, probably more than the clipped kidney ( 38 ). According to Patzak et al. ( 25 ), NO release actively buffers the ANG II-induced Ca 2+ response in renal microvessels from rodents. Based on these data, we find it possible that NO might cause the leveling off of the dose-response curve seen at above physiological concentrations of ANG II in the present and earlier studies ( 13, 17 ), although receptor saturation may also explain this observation.
There is reason to believe that NO buffering is reduced in the clipped kidney. Although the pressure drop over the clip ( 50 mmHg) ( 14 ) reduces pressure to a normotensive level, dilated AAs ( 16 ) and reduced glomerular capillary pressure ( 36 ) indicate a lower microvascular perfusion pressure in the clipped kidney, and a site of resistance interpolated between the clip and the AAs has been suggested ( 26, 36 ). Data from our laboratory indicate a 50% reduction of flow in the clipped kidney ( 14 ), which has earlier been shown to decrease NO release due to lowered shear stress ( 32 ). In cell culture, shear stress increases endothelial nitric oxide synthase (eNOS) mRNA and protein ( 39 ). Reduced flow might therefore diminish the capacity of endothelial cells to release NO and possibly explain the linear dose-response curve to ANG II in the clipped kidney.
The isolated arterioles we used consist of both smooth muscle and endothelial cells, and there is reason to suggest that ANG II is able to stimulate both cell types in this preparation. Based on the structure of ANG II bound to the immunoglobulin Fab131 (Protein Data Bank file "3CK0", submitted by Pan YH and Amzel LM), we estimated the hydrodynamic radius of the ANG II molecule to be less than 1 nm. According to Pluen et al. ( 28 ), this implies that the diffusion coefficient for ANG II in 2% agarose is similar to that in water (10 5 µm 2 s -1 ), when assuming a gel-pore radius of 85 ± 9 nm by using the Ogston model ( 24 ). Also, the diameter of the AA opening ( 20 µm) was relatively large compared with the typical length of the isolated vessel fragments (100 µm). We expect ligand concentration to increase at approximately the same rate on the endothelial and muscle side of the agarose-infused vessels during ANG II stimulation, and that Ca i 2+ responses with the present experimental setup are the combined result of endothelial and muscle signaling pathways, which both modulate the contractile response in renal microvessels in vivo ( 22 ).
Although studies on preglomerular arterioles with the endothelium intact are a physiological approach to examine the integrated Ca i 2+ response in a vessel, it makes it more difficult to dissect the separated role of the two cytosolic compartments of these tissues. However, it is well established that the endothelium harbors important modulators of hormonal-induced vasoconstriction ( 40 ) mediated through several systems such as NO ( 20 ) or prostaglandins ( 22 ). Some of these endothelium-derived effects have been shown to be altered in 2K1C hypertension ( 3 ). Also, recent findings indicate that the relationship between AT 1 and AT 2 receptors might be of importance for ANG II-mediated vascular resistance ( 40 ), and in rabbits this effect has been shown to be endothelium dependent ( 1 ). We therefore decided that using intact vessels with interacting VSMC and endothelium was the best approach to study Ca i 2+ signaling in 2K1C.
In the present study, the agarose infusion method developed by Loutzenhiser and Loutzenhiser ( 17 ) was used to produce isolated AA from both the clipped and nonclipped kidneys. By optimizing the real-time PCR technique earlier used in our laboratory to analyze receptor mRNA in individual vascular segments ( 11 ), we were able to get reproducible readings of AT 1a receptor mRNA in relatively few afferent vessels isolated with the agarose infusion/enzyme digestion technique.
Contrary to our working hypothesis, the ANG II-induced Ca i 2+ response and the AT 1a R mRNA levels were similar between the 2K1C and control rats. We also found no difference between AAs isolated from the clipped and nonclipped kidney in 2K1C. These data were surprising because negative feedback, a classic regulation pattern of 7 trans -membrane receptors, would be expected to occur due to the greatly elevated levels of ANG II in the renal tissues of 2K1C animals ( 21 ). Supporting the present data, a recent experiment performed in our laboratory on VSMC isolated with the iron oxide method from the nonclipped kidney in 2K1C displayed also unchanged AT 1a R mRNA and protein levels in 2K1C ( 2 ). In the same study, the RBF response to ANG II was reduced in the nonclipped kidney, seemingly contradicting the present findings. However, the ANG II-mediated RBF response in vivo is the integrated result of many signaling processes. Also, no ANG II was added to the buffer to reflect the high ANG II level of the animal, recycling of AT 1a Rs after removal of the vessels from the animal might also have masked receptor internalization present in vivo. However, changes in the AT 1a receptor level have also been observed in vitro. AT 1a R changes were detected at both the mRNA and protein level in isolated vessels from normotensive rats on a low-salt or a high-salt diet ( 31 ). Our data indicate that AT 1a receptor regulation in 2K1C is different from that seen during physiological adaptation to changes in salt intake.
One of the major regulators of excitation stimuli are the COX-derived prostaglandins. Because the dose-response curve in the clipped kidney showed no sign of being buffered in the same manner as the nonclipped kidney, we wanted to investigate how the COX system modulated the ANG II-induced Ca i 2+ response. Several reports indicate that only COX-1 is present in preglomerular microvessels ( 7, 12 ); therefore, this isoform was the most likely target of the indomethacin treatment in the present study. The notion that COX-2 is absent from the vasculature was supported by the fact that the commonly used selective COX-2 inhibitor NS-398 did not enhance the Ca i 2+ response in any of the kidneys from the present study.
In control animals, indomethacin treatment had no effect on the ANG II-mediated Ca 2+ response ( Fig. 4 ), a finding that is different from earlier observations in our laboratory, where indomethacin treatment in control animals augmented the RBF response to ANG II injected into the renal artery ( 2 ). Observations of nonspecific COX inhibition in dogs with experimentally induced hypoperfusion indicate that the RBF response might be COX-2 dependent ( 37 ), and because this isoform is not present in isolated arterioles, it may also explain why vessels from normal rats do not respond to indomethacin treatment. However, it shall be added that in cultured renal VSMC from normal animals, an enhanced calcium response to ANG II was found after pretreatment with indomethacin ( 29 ). Cultured VSMC lack an endothelium and are kept under physiological conditions that are altered compared with those in the animal, and this might explain why freshly isolated vessels from control animals and cultured VSMC react differently to indomethacin treatment.
Contrary to the data from control animals, indomethacin treatment in 2K1C augmented the Ca i 2+ response to ANG II both in the clipped and nonclipped kidney ( Figs. 4 and 5 ). Several reports demonstrate that ANG II stimulates the turnover of arachidonic acid ( 30, 35 ) and stimulates the synthesis of prostaglandins ( 18, 34, 35 ). A possible explanation for the increased response in indomethacin-treated vessels might therefore be an increased effect of prostaglandins mediating vasodilation in 2K1C. The effect was seen in both the clipped kidney, where the afferent arterioles are dilated, and in the nonclipped kidney, where the afferent arterioles are constricted. Although the isolated AAs from the clipped and nonclipped kidneys were subjected to greatly different perfusion in vivo, they were identically affected by indomethacin treatment. It is therefore reasonable to believe that the prevailing ANG II levels in the control and 2K1C rats were more strongly influencing the COX-1 function in the isolated vessels. Although COX inhibition enhanced the Ca i 2+ response to ANG II in 2K1C, the responses after indomethacin treatment were similar in the control, nonclipped, and clipped kidneys.
In conclusion, the present results did not support our hypothesis, as the ANG II responses were similar in all kidneys. The findings were consistent with AT 1a receptor mRNA values, which were similar in the control, nonclipped, and clipped kidneys. The buffering of the ANG II response was, however, different; both kidneys from 2K1C exhibited a significantly higher COX-1-dependent buffering when stimulated with ANG II, compared with control animals. COX-2 inhibition did not increase the Ca i 2+ response in any of the groups. The clipped kidney in 2K1C had an altered dose-response curve to ANG II compared with the nonclipped kidney, a finding that may be due to different buffering mechanisms of ANG II.
GRANTS
This work was supported by grants from the Norwegian Council of Cardiovascular Disease and the Strategic Research Program at Haukeland University Hospital.
【参考文献】
Amiri F and Garcia R. Renal angiotensin II receptor regulation in two-kidney, one clip hypertensive rats: effect of ACE inhibition. Hypertension 30: 337-344, 1997.
Bivol LM, Vagnes OB, and Iversen BM. The renal vascular response to ANG II injection is reduced in the nonclipped kidney of two-kidney, one-clip hypertension. Am J Physiol Renal Physiol 289: F393-F400, 2005.
Callera GE, Varanda WA, and Bendhack LM. Impaired relaxation to acetylcholine in 2K-1C hypertensive rat aortas involves changes in membrane hyperpolarization instead of an abnormal contribution of endothelial factors. Gen Pharmacol 34: 379-389, 2000.
Cervenka L, Horacek V, Vaneckova I, Hubacek JA, Oliverio MI, Coffman TM, and Navar LG. Essential role of AT1A receptor in the development of 2K1C hypertension. Hypertension 40: 735-741, 2002.
Cervenka L, Wang CT, Mitchell KD, and Navar LG. Proximal tubular angiotensin II levels and renal functional responses to AT1 receptor blockade in nonclipped kidneys of Goldblatt hypertensive rats. Hypertension 33: 102-107, 1999.
Cheng HF, Becker BN, Burns KD, and Harris RC. Angiotensin II upregulates type-1 angiotensin II receptors in renal proximal tubule. J Clin Invest 95: 2012-2019, 1995.
Cheng HF and Harris RC. Cyclooxygenases, the kidney, and hypertension. Hypertension 43: 525-530, 2004.
Della Bruna R, Bernhard I, Gess B, Schricker K, and Kurtz A. Renin gene and angiotensin II AT1 receptor gene expression in the kidneys of normal and of two-kidney/one-clip rats. Pflügers Arch 430: 265-272, 1995.
Douglas JG and Hopfer U. Novel aspect of angiotensin receptors and signal transduction in the kidney. Annu Rev Physiol 56: 649-669, 1994.
Haefliger JA, Bergonzelli G, Waeber G, Aubert JF, Nussberger J, Gavras H, Nicod P, and Waeber B. Renin and angiotensin II receptor gene expression in kidneys of renal hypertensive rats. Hypertension 26: 733-737, 1995.
Hansen FH, Vagnes OB, and Iversen BM. Enhanced response to AVP in the interlobular artery from the spontaneously hypertensive rat. Am J Physiol Renal Physiol 288: F1023-F1031, 2005.
Harris RC, Zhang MZ, and Cheng HF. Cyclooxygenase-2 and the renal renin-angiotensin system. Acta Physiol Scand 181: 543-547, 2004.
Iversen BM and Arendshorst WJ. ANG II and vasopressin stimulate calcium entry in dispersed smooth muscle cells of preglomerular arterioles. Am J Physiol Renal Physiol 274: F498-F508, 1998.
Iversen BM, Heyeraas KJ, Sekse I, Andersen KJ, and Ofstad J. Autoregulation of renal blood flow in two-kidney, one-clip hypertensive rats. Am J Physiol Renal Fluid Electrolyte Physiol 251: F245-F250, 1986.
Iversen BM, Kvam FI, Matre K, and Ofstad J. Resetting of renal blood autoregulation during acute blood pressure reduction in hypertensive rats. Am J Physiol Regul Integr Comp Physiol 275: R343-R349, 1998.
Iversen BM, Morkrid L, and Ofstad J. Afferent arteriolar diameter in DOCA-salt and two-kidney one-clip hypertensive rats. Am J Physiol Renal Fluid Electrolyte Physiol 245: F755-F762, 1983.
Loutzenhiser K and Loutzenhiser R. Angiotensin II-induced Ca 2+ influx in renal afferent and efferent arterioles: differing roles of voltage-gated and store-operated Ca 2+ entry. Circ Res 87: 551-557, 2000.
McGiff JC and Quilley J. 20-HETE and the kidney: resolution of old problems and new beginnings. Am J Physiol Regul Integr Comp Physiol 277: R607-R623, 1999.
Modrall JG, Quinones MJ, Frankhouse JH, Hsueh WA, Weaver FA, and Kedes L. Upregulation of angiotensin II type 1 receptor gene expression in chronic renovascular hypertension. J Surg Res 59: 135-140, 1995.
Moncada S, Palmer RM, and Higgs EA. Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacol Rev 43: 109-142, 1991.
Navar LG, Harrison-Bernard LM, Nishiyama A, and Kobori H. Regulation of intrarenal angiotensin II in hypertension. Hypertension 39: 316-322, 2002.
Navar LG, Inscho EW, Majid SA, Imig JD, Harrison-Bernard LM, and Mitchell KD. Paracrine regulation of the renal microcirculation. Physiol Rev 76: 425-536, 1996.
Navar LG and Nishiyama A. Why are angiotensin concentrations so high in the kidney? Curr Opin Nephrol Hypertens 13: 107-115, 2004.
Ogston AG. The spaces in a uniform random suspension of fibers. Faraday Soc Trans 54: 1754-1757, 1958.
Patzak A, Lai EY, Mrowka R, Steege A, Persson PB, and Persson AE. AT1 receptors mediate angiotensin II-induced release of nitric oxide in afferent arterioles. Kidney Int 66: 1949-1958, 2004.
Ploth DW. Angiotensin-dependent renal mechanisms in two-kidney, one-clip renal vascular hypertension. Am J Physiol Renal Fluid Electrolyte Physiol 245: F131-F141, 1983.
Ploth DW, Schnermann J, Dahlheim H, Hermle M, and Schmidmeier E. Autoregulation and tubuloglomerular feedback in normotensive and hypertensive rats. Kidney Int 12: 253-267, 1977.
Pluen A, Netti PA, Jain RK, and Berk DA. Diffusion of macromolecules in agarose gels: comparison of linear and globular configurations. Biophys J 77: 542-552, 1999.
Purdy KE and Arendshorst WJ. Prostaglandins buffer ANG II-mediated increases in cytosolic calcium in preglomerular VSMC. Am J Physiol Renal Physiol 277: F850-F858, 1999.
Rao GN, Lassegue B, Alexander RW, and Griendling KK. Angiotensin II stimulates phosphorylation of high-molecular mass cytosolic phospholipase A2 in vascular smooth muscle cells. Biochem J 299: 197-201, 1994.
Ruan X, Wagner C, Chatziantoniou C, Kurtz A, and Arendshorst WJ. Regulation of angiotensin II receptor AT1 subtypes in renal afferent arterioles during chronic changes in sodium diet. J Clin Invest 99: 1072-1081, 1997.
Rubanyi GM, Romero JC, and Vanhoutte PM. Flow-induced release of endothelium-derived relaxing factor. Am J Physiol Heart Circ Physiol 250: H1145-H1149, 1986.
Sadjadi J, Puttaparthi K, Welborn MB III, Rogers TE, Moe O, Clagett GP, Turnage RH, Levi M, and Modrall JG. Upregulation of autocrine-paracrine renin-angiotensin systems in chronic renovascular hypertension. J Vasc Surg 36: 386-392, 2002.
Schlondorff D. Renal prostaglandin synthesis. Sites of production and specific actions of prostaglandins. Am J Med 81: 1-11, 1986.
Schlondorff D, DeCandido S, and Satriano JA. Angiotensin II stimulates phospholipases C and A2 in cultured rat mesangial cells. Am J Physiol Cell Physiol 253: C113-C120, 1987.
Schwietzer G and Gertz KH. Changes of hemodynamics and glomerular ultrafiltration in renal hypertension of rats. Kidney Int 15: 134-143, 1979.
Tokuyama H, Hayashi K, Matsuda H, Kubota E, Honda M, Okubo K, Takamatsu I, Ozawa Y, and Saruta T. Role of nitric oxide and prostaglandin E2 in acute renal hypoperfusion. Nephrologie 8: 65-71, 2003.
Turkstra E, Braam B, and Koomans HA. Impaired renal blood flow autoregulation in two-kidney, one-clip hypertensive rats is caused by enhanced activity of nitric oxide. J Am Soc Nephrol 11: 847-855, 2000.
Uematsu M, Ohara Y, Navas JP, Nishida K, Murphy TJ, Alexander RW, Nerem RM, and Harrison DG. Regulation of endothelial cell nitric oxide synthase mRNA expression by shear stress. Am J Physiol Cell Physiol 269: C1371-C1378, 1995.
Watanabe T, Barker TA, and Berk BC. Angiotensin II and the endothelium: diverse signals and effects. Hypertension 45: 163-169, 2005.
作者单位：Renal Research Group, Institute of Medicine, University of Bergen, and Haukeland University Hospital, Bergen, Norway