| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RENAL HEMODYNAMICS AND CARDIORENAL INTEGRATION
1Institute of Vegetative Physiology, University Hospital Charité, Humboldt-University of Berlin, Berlin, Germany; and 2Department of Medical Cell Biology, Division of Physiology, University of Uppsala, Uppsala, Sweden
Submitted 21 May 2007 ; accepted in final form 20 September 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
cytosolic calcium; p38 mitogen-activated protein kinase; tubuloglomerular feedback; kidney
Ado is a candidate for mediation of the tubuloglomerular feedback in the kidney. Assumed involvement in the tubuloglomerular feedback bases on Ado's ability to constrict afferent arterioles and has further support by observations of no tubuloglomerular feedback in A1AR-deficient mice (3, 30). ANG II acts as a modulating factor for the tubuloglomerular feedback. ANG II type 1 receptor inhibition, lack of these receptors, or inhibition of the angiotensin-converting enzyme reduces the tubuloglomerular feedback (for review, see Ref. 27). On the other hand, Ado influences the ANG II sensitivity of the renal microvasculature acting on A1AR and A2AR (18), suggesting an interplay of Ado and ANG II in the control of renal vascular resistance (9, 11, 23, 33).
In a recent study, our group (17) showed a restoration of desensitized ANG II-induced contractions by Ado in isolated, perfused afferent arterioles. Ado treatment in between successive ANG II applications restored contraction up to the initial level of the ANG II application. Ado restored the ANG II response of afferent arterioles by a receptor-independent mechanism, which is caused by enhanced calcium sensitivity in these vessels.
In the present study, we tested the hypothesis of a prolonged action of Ado on the ANG II-induced constriction in afferent arterioles. We show that transient Ado treatment induces a significant amplification of the ANG II response of afferent arterioles for as long as 30 min after end of treatment. This effect of Ado is not mediated by Ado receptors, depends on Ado transportation into the cell, and includes p38 MAPK activation. Such a prolonged influence of ANG II-induced contractions in arterioles by the transient extracellular increase of Ado may help control renal filtration via tubuloglomerular feedback.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Mice of the C57BL/6 strain (male, body mass between 22 and 28 g; Scanbur BK, Sollentuna, Sweden) were included. A1AR-deficient mice (A1–/–) and their controls (A1+/+) were included to test the effect of chronic lack of A1AR. These mice were female with a body mass between 23 and 29 g. A1–/– mice were generated as described by Johansson et al. (15). A1–/– and A1+/+ mice were siblings from mating between A1+/– mice. Animals were fed with standard mouse chow and allowed free access to tap water.
All procedures conformed to the Guide for Care and Use of Laboratory Animals prepared by the Institute for Laboratory Animal Research. The local ethics committee for Uppsala University approved the procedures for this study.
Dissection and Perfusion of Afferent Arterioles
Dissection and perfusion procedures have been described before (24). In brief, outer cortical afferent arterioles of mice were dissected at 4°C in albumin-enriched DMEM (0.1%). Arterioles with their glomeruli were perfused in a thermoregulated chamber (37°C) fixed on the stage attached to an inverted microscope (Nikon, Badhoevedorp, The Netherlands), using a perfusion system that allows the adjustment of outer holding and inner perfusion pipettes (Vestavia Scientific, Vestavia Hills, AL). The perfusion pipette, with a diameter at the tip of exactly 5 µm, was connected to a reservoir containing the perfusion solution. A second holding pipette was used to maintain the attached glomerular tuft. The pressure in the pressure head was 100 mmHg, which corresponds to physiological pressure and flow (
50 nl/min) in the connected afferent arteriole (24). Criteria for using an arteriole included a satisfactory remaining basal tone and no vasodilatation. Both criteria were tested by rapidly increasing the perfusion pressure and assessing the change in the luminal diameter, which corresponds to transient constriction. A further criterion was a fast and complete constriction in response to KCl (100 mmol/l) solution.
Measurement of Isotonic Contraction
The experiments were recorded by a video system, off-line digitized, and analyzed as described before (24). Luminal diameters of the arterioles were measured to estimate the effect of vasoactive substances. Because we did not see systematic differences in arteriolar responses to agonists, except for the part connected to the holding and perfusion pipette, the site with the strongest contraction was chosen for the measurement. In all series, the last 10 s of a control or treatment period were used for statistical analysis of steady-state responses.
Calcium Imaging
Afferent arterioles were isolated, perfused, and loaded with fura 2-AM (in DMSO 10–5 mol/l) in the bath solution for 45 min. Loading was facilitated with Pluronic F-127 (end concentrations of DMSO and Pluronic F-127 were <0.1%). Fluorescence was measured with the digital imaging system QC-900 (Applied Imaging, Sunderland, UK). The arterioles were excited alternately at 340 and 380 nm, and the emission was measured at 510 nm. The 340-to-380-nm emission ratio was used to determine the cytosolic calcium concentration after calibration with a fura 2 calcium imaging calibration kit (F-6774) in vitro according to the protocol of Molecular Probes.
Isolation of Preglomerular Vessels and Phosphorylation Studies
Preglomerular vessels (including mainly interlobular arteries and afferent arterioles) of mice were isolated by a modified iron oxide-sieving technique according to Chaudhari and Kirschenbaum (6). Modifications were 1) the method to perfuse the kidneys, which was performed via cannulation of the aorta, and 2) the use of smaller sizes for the needles needed for separation of the tissue, as well as 3) use of sieves with smaller pores (100 and 80 µm). Isolated, nonperfused preglomerular vessels were treated with physiological salt solution, to simulate Ado application (handling control, n = 5), with Ado (10–11 to 10–4 mol/l, cumulatively, 2 min each concentration; n = 5), or with Ado followed by 10-min washout and application of ANG II (10–8 mol/l, 2 min; n = 10), norepinephrine (NE) (10–5 mol/l, 2 min; n = 10), or endothelin-1 (ET-1) (10–8 mol/l, 2 min; n = 10). Vessels were shock frozen exactly 2 min after the ANG II application in 10% TCA-aceton and kept at –80°C.
Quantification of 20-kDa Myosin Regulatory Light Chain and p38 MAPK Phosphorylation
Cellular extracts from preglomerular vessels were prepared as described previously (17). Protein extracts were separated by SDS-PAGE and transferred to Hybond-P membranes. A 20-kDa myosin regulatory light-chain (MLC20) phosphorylation was detected with a pS19/pS20-specific anti-MLC antibody (R1535P; Acris Antibodies). Membranes were stripped for 5 min with distilled water, for 5–15 min with 0.2 M NaOH, and for 5 min with distilled water and reprobed with an anti-phosphorylated p38 MAPK antibody (506119; Calbiochem). Detection of relative smooth muscle-specific 
-actin levels using an anti--actin antibody (AB5694; Acris Antibodies) served as loading control.
Solutions
DMEM-Ham's F-12 with 10 mmol/l HEPES (Invitrogen, Lidingö, Sweden) was used for dissection, bath, and perfusion. The pH was adjusted to 7.4 after addition of BSA. The concentration of BSA was 0.1% in dissection and bath solutions and 1% in the perfusion solution. BSA was obtained from SERVA Electrophoresis (Heidelberg, Germany) and DMEM from Sigma-Aldrich (Munich, Germany). The potassium solution had the following composition (in mmol/l): 20 NaCl, 95 KCl, 25 NaHCO3, 2.5 K2HPO4, 1.3 CaCl2, 1.2 MgSO4, and 5.5 glucose and was equilibrated with 5% CO2 in air.
Pharmacological Agents
The following drugs were used: ANG II, Ado, NE, ET-1, N6-cyclopentyladenosine (CPA), 8-(p-sulfophenyl)theophylline (8-SPT), 8-cyclopentyltheophylline (CPT), 4-[2-7-amino-2-(2-furyl)(1,2,4)triazolo(2,3-a) (1,3,5)triazin-5-ylaminoethyl]phenol (ZM-241385), nitrobenzylthio inosine (NBTI), and SB-203580 (Sigma-Aldrich, Stockholm, Sweden).
Statistics
ANOVA for repeated measurements (nonparametric Brunner test) was used to test time-dependent changes in the arteriolar diameter and to check for differences between the groups (SAS system). Post hoc comparisons were performed with Tukey's test. Wilcoxon's tests were applied for treatment effects on basal diameter, for comparison of control diameters between groups, and for luminescence signals of Western blots. Data are presented as means ± SE. The confidence level P was set to 0.05.
Experimental Protocols for Contraction Measurement and Imaging
Effect of Ado on ANG II-induced constriction. Ado was applied in cumulative concentrations, each for 2 min, from 10–11 to 10–4 mol/l. After washout for 10 (n = 5) or 30 min (n = 5), ANG II was applied cumulatively from 10–12 to 10–6 mol/l (each dose 2 min). For manipulation control, physiological solution (DMEM + albumin) was applied instead of Ado followed by 10-min washout and measurement of an ANG II concentration-response curve in another series of experiments (n = 6). The control curve for the ANG II concentration response was measured by application of ANG II in cumulative concentrations from 10–12 to 10–6 mol/l in separate vessels (n = 10).
Specificity of the Ado effect. Ado was applied in cumulative doses from 10–11 to 10–4 mol/l. After washout for 10 min, the NE concentration response (10–9 to 10–5 mol/l; n = 5) or the ET-1 concentration response (10–12 to 10–7; n = 6) was measured. For control, the concentration response curves were measured for both substances but without Ado pretreatment in separate arterioles (NE: n = 5, ET-1: n = 8).
Role of Ado receptors. The selective A1AR inhibitor CPT alone (10–5 mol/l; n = 5), CPT together with the A2AAR inhibitor ZM-241385 (10–7 mol/l; n = 6), or the nonselective Ado receptor inhibitor 8-SPT (5 x 10–5 mol/l; n = 6) was applied simultaneously with Ado. Inhibitors were also present during the 10-min washout period. In another series, inosine (10–11 to 10–4 mol/l; n = 6), which acts on A3AR, was applied instead of Ado followed by 10-min washout. In all of these series, ANG II was administered cumulatively from 10–12 to 10–6 mol/l. In addition, the effect of A1AR stimulation was tested by application of the selective A1AR agonist CPA (10–11 to 10–6 mol/l; n = 6), followed by 10-min washout. The efficacies of the receptor agonist and inhibitors have been shown in mouse arterioles. CPA induced a constrictor response, and CPT inhibited the constrictor effect of Ado effectively, whereas inhibition of A2AAR with ZM-241385 (all drugs with same concentration as in the present study) induced concentration-dependent constrictor responses to Ado (18). The ability of 8-SPT to block Ado receptors in the concentration used here was also shown in mouse arterioles (18).
Contribution of cytosolic calcium. To test whether the enhancement of the ANG II response of the arterioles was calcium dependent, cytosolic calcium transients were measured by load with fura 2. Transients were obtained for ANG II concentrations of 10–10 and 10–8 mol/l with (n = 9 and n = 6, respectively) and without Ado pretreatment (n = 7 and n = 8, respectively).
Role of Ado transport through the membrane. Because Ado receptors participated only to a small amount in mediation of the Ado effect, it was tested whether transport of Ado through the cell membrane by an equilibrative transporter plays a role. Arterioles were treated with NBTI (3 x 10–7 mol/l) during extracellular Ado application. After washout, the ANG II concentration-response curve was measured (n = 5). To exclude a nonspecific effect of NBTI on arteriolar constriction, arterioles were treated with NBTI alone, and the ANG II concentration-response curve was measured after 10-min washout (n = 5).
Intracellular action of Ado. To test the contribution of p38 MAPK to the Ado effect, the selective inhibitor of SB-203580 enzyme (10–5 mol/l) was applied 10 min before and during the Ado treatment; SB-203580 and Ado were then washed out (10 min), and the ANG II concentration response was measured (n = 5). All drugs were added to the bath solution.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Ado induced modest changes in arteriolar diameter during cumulative application, which agrees with results of a recent study (18). There was no significant change in the diameter at the end of the cumulative application, namely at 10–4 mol/l Ado, except for a slight tendency to greater diameters. Washout of Ado reestablished diameters to control levels. The ANG II response of afferent arterioles was significantly stronger after Ado pretreatment than the ANG II concentration curve without Ado pretreatment, especially at low ANG II concentrations (ANOVA, P < 0.05; Fig. 1). Enhancement of the ANG II response was similar for washout times of 10 and 30 min (Fig. 1). Time of control experiments revealed no effect of the procedure on the ANG II concentration response of arterioles (Fig. 1).
|
The concentration responses to NE and ET-1 were not influenced by Ado treatment (Figs. 2 and 3, respectively).
|
|
Using selective and nonselective inhibitors of Ado receptors, we tested the possible contribution of these to the Ado effect. CPT, a selective A1AR inhibitor, did not influence luminal diameters in the control situation. It did not prevent the augmented ANG II response when applied simultaneously with Ado (ANOVA, P < 0.05; Fig. 4). Combination of CPT and ZM-241385, a selective A2AAR inhibitor, also had no influence on arteriolar diameters and did not prevent increased arteriolar constriction on lower ANG II concentrations; i.e., the ANG II response was enhanced compared with the control ANG II response in the low concentration range. For higher ANG II concentrations, the constriction corresponded to that of the control ANG II response (Fig. 4). Inosine, an A3AR agonist, did not influence the ANG II response (Fig. 4). The nonselective inhibition of Ado receptors with 8-SPT did not affect the Ado-induced enhancement of the ANG II response (Fig. 4). This substance had no effect on arteriolar diameters in the control situation.
|
|
|
ANG II application of 10–10 and 10–8 mol/l increased cytosolic calcium concentrations by 21.4 ± 4.9% and 90.7 ± 22.3%, respectively. Calcium transients did not significantly change after Ado pretreatment (33.6 ± 11.3% and 86.4 ± 15.2% for 10–10 and 10–8 mol/l ANG II, respectively; Figs. 7 and 8).
|
|
Blockade of Ado transport into the cell using NBTI during extracellular Ado application significantly inhibited the Ado effect on the ANG II response (ANOVA, P < 0.05; Fig. 9). NBTI itself did not change arteriolar diameters in the control situation (not shown) and did not change the ANG II response significantly (Fig. 9).
|
Selective inhibition of p38 MAPK with SB-203580 prevented the enhancing effect of Ado on the ANG II-induced constriction in afferent arterioles (Fig. 10). The response to ANG II did not differ from the control curve for ANG II (see Fig. 1).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In addition to the well-known Ado-ANG II interactions in the renal vasculature mediated by their receptors, it has been shown that Ado can also restore contraction of ANG II-desensitized afferent arterioles. This latter effect takes place via a nonreceptor-mediated increase of the calcium sensitivity (17). Here, we show a clear long-lasting increase of the ANG II response after Ado treatment, which is equally receptor independent. The enhanced contractile response to Ado is reflected by an increase in the MLC20 phosphorylation.
Both receptor-dependent and -independent effects of Ado on the ANG II response may contribute to the tubuloglomerular feedback in the kidney. The tubuloglomerular feedback mechanism links distal tubular chloride load and filtration rate of the same glomerulus. Larger sodium chloride load results in an increased release of Ado or ATP (and subsequent increased interstitial generation of Ado from ATP) from macula densa cells (4, 28). Ado then constricts arterioles by acting on A1AR (10). Data of the present study suggest that a transient elevation of Ado concentration sensitizes the response to ANG II dramatically for at least 30 min. This could result in a prolonged change of glomerular hemodynamics and consequently of the filtration rate, since the tone of the afferent arteriole is increased for a given ANG II concentration in the blood and kidney, respectively. However, the effect of transient elevations of Ado concentrations on efferent arterioles has not yet been investigated, limiting conclusions regarding the renal resistance and filtration rate.
Interactions of Ado and ANG II in the renal vasculature are well known. In general, constrictor effects of ANG II are enhanced by Ado and vice versa (9, 11, 13, 18, 33). A1AR, A2AR, and AT1 are involved in this interaction (11, 18). In contrast to these findings of a receptor-mediated, specific interplay of both substances, the enhancement of the ANG II response in the present study did not, or only to a small degree, involve Ado receptors. Neither selective inhibition of A1AR, A1AR, and A2AAR together nor general blockade of Ado receptors prevented the Ado effect on the ANG II response of arterioles. Also, genetically induced lack of A1AR did not influence the increased ANG II response after Ado. Furthermore, A3AR was also shown not to be involved, since the specific agonist inosine did not affect ANG II responses. CPA pretreatment did not mimic the Ado effect, supporting the conclusion that A1ARs are not involved. Rather, the ANG II response was attenuated at 10–8 and 10–7 mol/l. This observation suggests that A1AR activation can have a long-term influence on the contractility. This effect might be neutralized by the simultaneous activation of A2AR during Ado treatment.
Remarkably, the cytosolic calcium transients on ANG II after Ado treatment do not correspond to the enhanced ANG II response but behaved like control arterioles. This finding suggests a calcium-sensitizing mechanism enhancing the ANG II-induced constriction. In a previous study, Lai et al. (17) found resensitization of ANG II responses of afferent arterioles to repeated ANG II applications induced by Ado treatment between the applications. The resensitization of the contraction was not because of receptor-related mechanisms. Rather, the calcium sensitivity was increased and went along with increased phosphorylation of the regulatory unit of the MLC. Inhibition of p38 MAPK prevented the effect of Ado in their study. We show here that the selective p38 MAPK inhibitor SB-203580 also prevented the long-term Ado effect on ANG II-induced constrictions. It was shown that SB-203580 does not influence the ANG II-induced constriction of afferent arterioles not treated with Ado (17). Furthermore, the phosphorylation of p38 MAPK determined by Western blotting was enhanced in Ado-treated preglomerular vessels after ANG II application. This observation hints at a contribution of this kinase in mediating Ado effects on ANG II-induced contractions. p38 MAPK is involved in cell responses to several stimuli, which act as stressors (for review, see Ref. 25). Ado, which concentration increases in hypoxic and ischemic tissues, activates p38 MAPK, as shown in the present study in arterioles and in other studies in myocardial tissue (1, 14). Apart from a receptor-mediated activation (29), the present study suggests that Ado can activate p38 MAPK via intracellular action. The p38 MAPK kinase may in turn activate a signaling pathway resulting in an increased phosphorylation of MLC20 after ANG II application. Ado and ANG II seem to act collectively in the activation of p38 MAPK, which stimulates MAPK-activated protein kinase-2, resulting in the phosphorylation of heat shock protein 27 (20). This induces inhibition of the MLC phosphatase and an increase of MLC20 phosphorylation.
Remarkably, the phosphorylation of p38 MAPK was reduced in vessels treated with Ado and then subjected to ET-1 or NE. This observation was not expected, since Ado, NE, and ET-1 increase the phosphorylation of p38 MAPK in vascular smooth muscle cells (19, 22). Although the combination of Ado and ANG II, with the latter also activating p38 MAPK (31), results in increased p38 MAPK phosphorylation, Ado combined with NE and ET-1 decreases it compared with vessels not treated with Ado. This suggests different intracellular pathways of p38 MAPK activation for NE or ET-1 compared with ANG II. It has been shown that ANG II uses redox-sensitive mechanisms (20), whereas the NE-induced activation is dependent on the calcium influx. The response to ET-1 was dependent on a herbimycin A-sensitive tyrosine kinase and calcium influx in a study in small arteries (22).
Ado treatment did not influence the phosphorylation of MLC20 in combination with NE or ET-1 in the preglomerular vessels. This agrees with the lack of influence of Ado on NE- or ET-1-induced constrictions of the afferent arterioles. One reason may be the decreased p38 MAPK phosphorylation observed for this combination of drugs.
Data indicate that the long-term Ado effect requires a transportation of Ado through the cell membrane into the cytosol. The transport and thus the enhancement of the ANG II response are inhibited by NBTI, which blocks an important part of the Ado transporters. These transporters normally equilibrate the Ado concentrations inside and outside of the cell; i.e., the direction of transport depends on the concentration gradient (32). The efficacy of NBTI to inhibit the Ado transport has been shown in freshly isolated porcine coronary smooth muscle cells and in cultured human coronary smooth muscle cells (7). Concentrations of 3 x 10–7 mol/l NBTI, as used in the present study, reduced the Ado uptake to 20% of the value without NBTI in the cited study. Furthermore, NBTI significantly reduced the Ado uptake in the mouse cardiomyocyte cell line HL-1 (5) and in the porcine coronary smooth muscle (26).
In the present study, Ado was applied in concentrations from 10–11 to 10–4 mol/l, each for 2 min. To estimate whether there is a net inward Ado flow, the tissue concentrations of Ado must be known. Measurements of intra- and extracellular Ado concentrations in different models produced variable results. In vitro studies with the microdialysis technique showed values of 2 x 10–7 mol/l in the interstitium of the rat renal cortex (2). Tissue concentration of Ado in the rat kidney was measured to be 4 x 10–6 mol/kg kidney wet wt in another study (16). Investigations of the NBTI-sensitive nucleoside transporter in freshly isolated smooth muscle cells of porcine coronary arteries also provided estimates on Ado concentrations. The Km values of these transporter were shown to be 4–5 x 10–6 mol/l (7). The Ado concentrations were comparably higher at the end of the cumulative application of the present study. Therefore, an inward transportation and transient increase of the intracellular Ado concentration must be assumed.
The long-term influence of transient increases of extracellular Ado concentrations on ANG II response in afferent arterioles seems to be specific for ANG II, since the responses on NE and ET-1 were not affected by Ado treatment. This observation suggests specificity of the mechanisms involved in the process of enhancement of the ANG II response. Because receptors do not play an important role, interactions at the receptor level can be excluded. The influence of Ado on the ANG II-induced contraction most likely includes ANG II-specific intracellular signaling pathways, via protein kinase cascades, enhancing the calcium sensitivity of the contractile machinery in the arterioles.
In summary, the present study suggests a specific, long-term or "memory effect" of transient extracellular Ado on the ANG II-mediated constriction in smooth muscle cells of afferent arterioles. Ado contributes to mediation of the tubuloglomerular feedback; i.e., it constricts afferent arterioles and reduces the filtration rate in response to increased sodium chloride load in the distal tubule. An enhanced reactivity of afferent arterioles to ANG II for a longer time after a transient increase of Ado may maintain reduced filtration rate. The Ado effect critically depends on transportation of Ado into the cell, does not require the action on Ado receptors, and includes the activation of p38 MAPK.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
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.
* A. Patzak and E. Y. Lai contributed equally to this work. 
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
and $, Significant differences for single ANG II concentrations 10 and 30 min after washout, respectively, compared with the control ANG II response (P < 0.05).
