Adenosine triphosphate (ATP) and norepinephrine (NE) interact in the control of blood flow in the kidney. A combined effect of NE and ATP has not been previously investigated at the level of the afferent arteriole (Af). We studied the effects of ATP on the contractile response of the Af to NE. Vascular reactivity to ATP, NE, and their combination was investigated in isolated perfused Af from mice. The roles of α-adrenoceptors and P2-ATP-receptors were investigated by use of specific agonists and antagonists. Cytosolic calcium was measured using the fluorescent calcium dye fura-2. ATP in concentrations from 10−12 to 10−4 mol/l induced transient contractions. NE constricted the Af in a dose-dependent manner and induced significant contractions at > 10−7 mol/l. Treatment with ATP (10−8 and 10−6 mol/l) increased the NE response. Diameters were reduced by 20% already at 10−11 mol/l NE during ATP treatment of 10−6 mol/l. ATP increased the calcium response to NE significantly at 10−8 and 10−7mol/l NE. The P2-type ATP receptor blocker pyridoxal-phosphate-6-azophenyl-2′,4′-disulfonic acid (PPADS) (10−5 mol/l) abolished the sensitization of the NE response by ATP. The α1-blocker prazosin (10−7 mol/l) inhibited the ATP effect, as did the α2-blocker yohimbine (10−7 mol/l). Neither the phenylephrine- nor clonidine-induced concentration response curves was affected by ATP in the bath solution. Costimulation with ATP enhances the response of the Af to NE. This effect is mediated by increased cytosolic calcium. The enhancing effect involves P2-type ATP receptors and both α1- and α2-adrenoceptors.
- cytosolic calcium
- renal circulation
afferent arterioles are pivotal to the regulation of renal blood flow and glomerular filtration rate because of their significant contribution to total renal resistance and their key position in control systems. The tone of afferent arterioles is determined by the action of the sympathetic nervous system and several endocrine and paracrine factors (16) as well as by autoregulation (1). NE is the primary peripheral neurotransmitter of the sympathetic nervous system. In the kidney, norepinephrine (NE) has been shown to be important for the regulation of the afferent arteriolar tone through its α1-receptor (5). Activation of α2-receptor has little effect in itself but has been shown to augment the response to ANG II (18). Adrenergic β-receptors are of lesser importance in the renal vasculature (4). There are two cotransmitters of NE, ATP, and neuropeptide-Y that contribute to the effect of sympathetic nerves on renal vessels (25). ATP itself affects renal vessels acting on P2X-receptors causing the afferent arterioles to constrict (14–16). Recent results indicate that the interaction of ATP and NE may be important in the regulation of vascular tone at the afferent arteriole and its regulation of renal function. In the isolated perfused kidney, inhibition of ATP P2-receptors decrease the pressor response of renal nerve stimulation (34). ATP has also been shown to modulate the different vascular responses to sympathetic nerve stimulation and to exogenous NE in the kidneys (31). In addition to its corelease from sympathetic nerves, ATP may be released from macula densa cells (3, 19) and play a role in mediating the tubuloglomerular feedback (TGF) in the kidney (16, 28). However, the effect of combined ATP and NE stimulation has not been investigated at the level of the afferent arteriole.
The aim of the present study was to investigate how ATP interacts with NE in controlling the afferent arteriolar tone and to study which receptors are involved in regulating such an effect.
MATERIALS AND METHODS
Male mice of the C57 black 6J strain (Scanbur BK, Solna, Sweden) were used in these experiments. The mice weighed between 25 and 30 g. Animals were fed standard pelleted food (Scanbur BK) and had free access to tap water. The experiments were ethically approved by the board of animal experiments at the county court of Uppsala.
Dissection and perfusion.
Dissection and perfusion were performed as previously published (26). In short, the animals were killed by cervical dislocation, and the kidneys were removed and immediately placed in ice-cold DMEM with 0.1% albumin. The kidneys were sliced in thin transversal sections, and the afferent arterioles were dissected by using sharpened forceps (no. 5; Dumont) under a stereomicroscope (Leica Microsystems, Wetzlar, Germany).
Arterioles from the outer cortex with attached glomerulus were selected and transferred to a thermoregulated chamber on the stage of an inverted microscope (Noran, Middleton, WI). Glass pipettes (Drummond Scientific, Broomall, PA) mounted in a perfusion system (Vestavia Scientific, Vestavia Hills, AL) were used to fix and perfuse the arteriole and glomerulus (Fig. 1A). The holding pipette (OD, 2.13 mm; ID, 1.63 mm) had an aperture of roughly 26 μm at the tip and a constriction of about 20 μm after customizing. The proximal end of the arteriole was aspirated into this pipette. The inner, perfusion pipette (OD, 1.19 mm; ID, 1.02 mm) with a diameter of the tip of 5 μm was advanced into the lumen of the arteriole. This pipette was connected to a reservoir containing the perfusion solution and to a manometer. The arterioles were perfused at 37°C with a pressure of 100 mmHg in the pressure head, which we have previously shown produces a perfusion pressure within the autoregulatory range and with a physiological perfusion flow rate (26). Arterioles that exhibited a basal tone and clear lumen (Fig. 1B) were tested for viability by exchanging the bath solution to a 100 mmol/l KCl after which the arterioles were allowed to recover for 10 min. Only arterioles showing a strong contraction to the KCl challenge, as shown in Fig. 1C, were used.
Measurement of arteriolar diameter.
These experiments were recorded on super video home system videotapes (video recorder Panasonic NV-HS830; Matsushita Audio Video, Lueneburg, Germany). The end-magnification results were from a Nikon ×60/1.2 water-immersion objective lens and projection (×1) on a 0.3″ chip digital camera (model VCAM110; Phytec Technology Holding, Mainz, Germany). Video sequences were digitized by using a frame-grabber card (pciGrabber-4plus; Phytec Technology Holding). The vessel diameters were determined by using customized software (from Dr. H. Siegmund, Johannes-Müller-Institute of Physiology, Humboldt-University of Berlin, Germany). The equipment allowed a resolution of 0.11 μm of the vessel structure. The luminal diameter of the afferent arteriole was measured at one point in five consecutive pictures with 1-s intervals. Each vessel was used for one experiment only. Control diameter was obtained after the recovery from the KCl test, shortly before starting the experimental protocol.
Measurement of intracellular calcium.
In separate experiments, afferent arterioles were dissected and perfused as described above and incubated with fura-2 AM (VWR International, Stockholm, Sweden) at 10−5 mol/l for 45 min. The arterioles were excited alternately at 340 and 380 nm and the emission was measured at 510 nm with 3-s collection periods for each excitation. The ratio between emissions in successive time periods (340/380 nm) was used to determine intracellular calcium concentration against a calibrated standard (model F-6774 fura-2 calcium imaging calibration kit; Molecular Probes) in the same microscope. Fluorescence was detected using an Applied Imaging Quanticell-900 (VisiTech International, Sunderland, UK). Three regions of interest were selected on each vessel, one covering the whole width of the vessel and one from each side of the vessel wall as shown in Fig. 2A. The tracings from these areas generally were very similar. Results from the three areas were averaged for each vessel. The fura-2 ratio was measured for 2 min after application of each incremental dose of NE. The calcium value for the baseline was acquired before application of agonist, the peak value was selected manually, and the plateau value was acquired from the collection periods ∼105 s after the application of agonist, corresponding to the measurement time for the sustained contraction.
Diameter time-response curves.
The time response of the afferent arterioles was investigated for ATP at a concentration of 10−6 mol/l and NE at 10−4 mol/l. Each experiment was 3 min long, and measurements were taken every 6 s. After 1 min, the bath solution was exchanged to a solution with the above-indicated concentrations of ATP or of NE. The curves were adjusted so that the first strong contraction after superfusion with agonist was superimposed for the individual curves, thereby giving a more accurate time illustration of the time response (Figs. 3A and 5A ).
Diameter dose-response curves.
A normal dose-response curve for NE was obtained by exchanging the bath solution at 2-min intervals in steps of 10× dilution from 10−12 to 10−4 mol/l. The vessel diameter was measured at the end of each period as a measure of sustained vessel diameter. The ATP dose-response curve was obtained in a similar manner.
ATP effect on NE diameter dose response.
To study the effect of ATP on the dose-response curve for NE, the bath solution was exchanged to solutions containing different concentrations of ATP (10−10, 10−8, and 10−6 mol/l). Thereafter, the ATP concentration was kept constant during the measurement of the dose-response reaction of the afferent arteriole to NE from 10−12 to 10−4 mol/l.
Use of inhibitors.
All antagonists were bought from Sigma (St. Louis, MO). When the effects of specific receptor blockers on the interaction between NE and ATP were investigated, the receptor antagonists were added together with ATP and a NE dose-response curve was recorded as described above with constant concentrations of ATP and the antagonist. The receptor antagonists used were the α1-adrenoceptor blocker prazosin at a concentration of 10−7 mol/l, the α2-adrenoceptor blocker yohimbine at a concentration of 10−7 mol/l, and the ATP-receptor blocker PPADS at a concentration of 10−5 mol/l as has been previously shown to be sufficient in vascular preparations from the mouse (10, 27).
Diameter dose-response curves for specific agonists.
To test whether either α1- or α2-adrenoceptor activation is sufficient for the effects of ATP on the NE dose response, we recorded additional dose-response curves for the α1-agonist phenylephrine and the α2-agonist clonidine (both from Sigma) with and without ATP in the bath solution as described for NE above.
The curves were analyzed with two-way ANOVA to test for contraction over time and difference between curves. Tukey's honestly significant difference test was used as post hoc test between curves, and paired t-tests were used to test for significant calcium increase compared with baseline. P < 0.05 was considered significant. Statistics were calculated using R 2.1.0 (32).
ATP elicited contraction and calcium transients.
Bolus application of 10−6 mol/l ATP caused a strong (77 ± 13%, P < 0.05) but transient contraction in the afferent arteriole. The maximum contraction was reached during the first measurement interval after superfusion with agonist. The arterioles reopen to the control diameter after the bolus application (Fig. 3A). The transient contraction was accompanied by a small (from 185 ± 7 nM to 227 ± 11.6 nM, P < 0.05) increase in intracellular calcium exhibiting a flat time course without clear peak or any sustained response (exemplified in Fig. 3B with averages in 3C).
NE-elicited contraction and calcium transients.
NE (10−4 mol/l) elicited a complete and stable contraction in all arterioles. The arterioles contracted within one to two measurement intervals (6–12 s) by 99 ± 1% (P < 0.05) and remained fully contracted during the whole time of measurement (Fig. 5A). The calcium response to 10−4 mol/l NE showed a peak increase from 171 ± 24 nM to 334 ± 83 nM (P < 0.05) and a sustained plateau at 226 ± 29 nM (P < 0.05) for the full 2 min of measurement as shown in Fig. 5, B and C (Fig. 2, B–G, shows corresponding original fluorescence images).
Sustained contractions elicited by ATP and NE.
The sustained response to successive applications of ATP from 10−12 to 10−4 mol/l showed no diameter change compared with the baseline at any concentration (Fig. 4, A and B). NE alone elicited no transient or sustained vascular reaction below 10−7 mol/l in any vessel and showed a 44 ± 14% contraction around 10−6 mol/l from a basal diameter of 10 ± 0.9 μm. At 10−5 and 10−4 mol/l, NE elicited a near-complete contraction of the afferent arteriole (95 ± 2% and 96 ± 2%, respectively) (Fig. 6, ▪).
ATP effect on NE-elicited contraction.
The combination of ATP and NE partially increased the sensitivity of the afferent arteriole to NE. ATP concentrations of 10−8 and 10−6 mol/l in bath solution caused a reaction to NE at successively lower concentrations (Fig. 6, A and B). ATP (10−10 mol/l) did not affect the NE dose-response curve (P > 0.2 by two-way ANOVA) having a basal diameter of 9.1 ± 0.6 μm and showing a 64 ± 21% contraction at 10−6 mol/l NE and 98 ± 1% and 96 ± 2% at 10−4 and 10−5 mol/l NE. The contraction caused by 10−6 mol/l and 10−8 mol/l ATP in combination with NE was significant already at 10−11 mol/l NE (21 ± 5%) and 10−8 mol/l NE (22 ± 6%), respectively, with baselines at 6.3 ± 0.9 μm and 8.4 ± 0.7 μm. The dose-response curves to NE in combination with 10−8 mol/l and 10−6 mol/l ATP show a bimodal course. The sensitized contraction at low concentrations decreased the resting diameter of the afferent arteriole with ∼20%. A second phase of contraction was evident at 10−6 mol/l NE and above. This resulted in a 92 ± 5% at 10−4 mol/l NE in both groups (Fig. 6, A and B).
ATP effect on NE elicited calcium transients.
Cytosolic calcium transients elicited by successive applications of NE between 10−12 mol/l and 10−4 mol/l showed no change at concentrations lower than 10−7 mol/l. At 10−7 mol/l NE a small calcium increase (from 168 ± 20 nM to 191 ± 22 nM) was detectable but without sustained calcium increase. Higher concentrations of NE induced a strong calcium increase with a clear peak/plateau time course with a calcium increase at 10−6 mol/l NE of 93 ± 27 nM or 52% and a plateau phase at 58 ± 20 nM above the baseline at 177 nM (Fig. 7A). With 10−6 mol/l ATP in the bath solution the calcium increase is stronger than for NE alone (from 168 ± 9 nM to 192 ± 8 nM at 10−8 mol/l NE and from 166 ± 4 nM to 204 ± 6 nM at 10−7 mol/l NE (Fig. 7B, P < 0.05 for both). The response at 10−6 mol/l NE is similar to that of NE alone (Fig. 7B).
Blocking the P2-type ATP receptors with 10−5 mol/l PPADS during performance of a NE dose-response curve with 10−6 mol/l ATP in the bath solution gives a curve that is similar to that of NE alone (Fig. 8, A and B) and significantly different from the NE dose-response curve with 10−6 mol/l ATP (P < 0.05 by two-way ANOVA). The α1-adrenoceptor blocker prazosin completely blocked the effect of ATP at low concentrations of NE and shifted the contraction of the afferent arteriole to the right (P < 0.05 by two-way ANOVA) (Fig. 8, C and D). The baseline diameter of the prazosin group was larger than that of the group treated with NE and 10−6 mol/l ATP. However, when comparing the baseline diameters of all groups, there are no significant differences between groups and we believe the differences shown are illustrative of the variation in basal afferent arteriolar diameters. The α2-adrenoceptor yohimbine was also able to block the ATP effect (P < 0.05 compared with NE with 10−6 mol/l ATP) without changing the contraction caused by NE (P > 0.6 compared with NE alone by two-way ANOVA) (Fig. 8, E and F).
Specific α-adrenergic agonists.
The α1-agonist phenylephrine elicited a reaction similar to the endogenous agonist NE (P > 0.5 by ANOVA, n = 6) showing a 31 ± 8.5% contraction at 10−6 mol/l and a near-full contraction at higher concentrations (70 ± 2.7% and 85 ± 2.2% from a 11.0 ± 1.1 μm baseline). However, the dose-response curve of phenylephrine showed no change under the influence of 10−6 mol/l ATP (P > 0.4 by ANOVA, n = 5) with a 43 ± 13% contraction at 10−6 mol/l and 82 ± 6% and 90 ± 3% at the higher concentrations with the baseline at 9.8 ± 0.5 μm.
In response to the α2-agonist clonidine the afferent arterioles contracted at 10−5 and 10−4 mol/l (14 ± 5% and 18 ± 5% from a 8.4 ± 1.1-μm baseline, respectively; P < 0.05; n = 6). No further contraction was seen when clonidine was combined with ATP (P > 0.4 by two-way ANOVA, n = 6) showing a 11 ± 4% contraction at 10−5mol/l and 12 ± 4% at 10−4 mol/l from the 10.0 ± 0.6-μm baseline (P < 0.05).
The main finding presented in this paper is that ATP increases the response of the afferent arteriole to low concentrations of NE through a pathway depending on P2-type ATP receptors and both the α1- and α2-type adrenoceptors.
The finding of a strong contractile response to 10−6 mol/l ATP (Fig. 3A) is consistent with earlier findings (16). On the other hand, we have not been able to show any sustained contraction to increasing concentrations of ATP. This is in contrast to work by Inscho et al. (16) and others (36) where ATP is shown to elicit a sustained contraction on the afferent arteriole in the juxtamedullary nephron preparation at 10−6 mol/l, causing about 17% constriction. That we do not see this contraction in the isolated afferent arteriole may reflect differences both in species and in the preparations. While we see a small calcium increase after application of 10−6 ATP (Fig. 3B), it is not sufficient to cause a sustained contraction (Fig. 3A). These results indicate that the afferent arteriole from the mouse may have a different pharmacological profile than that from the rat. Both P2X and P2Y receptors have been found in the intrarenal arteries, and both receptor types are involved in regulating the pressor response to renal nerve stimulation (34). In the rat, ATP exerts its paracrine or neurotransmitter function through P2X receptors on the interlobular arteries and afferent arterioles (15, 36). In contrast, in the rabbit, P2X inhibition has been shown not to effect the renal blood flow response to renal nerve stimulation (9).
The 10−4 mol/l NE produced a quickly established, stable contraction (Fig. 5A), which is in agreement with earlier results from our laboratory showing no weakening of calcium transients with successive applications of NE (20). In addition, it is in agreement with the earliest results in the isolated perfused afferent arteriole of rabbits showing sustained contraction over a 3-min measurement period (6) and with studies in mouse arterioles (13). The response has slightly lower sensitivity than that seen in the rabbit (6, 7), while the rat exhibits a more gradual contraction from 10−9 to 10−5 mol/l (35). The strong calcium increase upon stimulation with 10−4 mol/l NE with a clear peak/plateau time course indicates that the response to NE is significantly mediated by changes in the cytosolic calcium. This observation is consistent with earlier findings (20, 21).
ATP concentrations of 10−8 and 10−6 mol/l sensitized the afferent arteriole to low concentrations of NE in a dose-dependent manner (Fig. 6). A sensitizing effect of ATP on the afferent arteriole to NE stimulation can also be concluded from results from renal nerve stimulation of the isolated perfused kidney during P2-receptor inhibition (34). This suggests an enhancing effect of ATP on NE-induced contractions of glomerular arterioles that with a 20% decrease in diameter would cause a 240% increase in resistance and may cause as much as a 60% decrease of renal blood flow if the average is extrapolated to all nephrons (29). This may provide a mechanism for the resetting of TGF, given that ATP is released from macula densa cells upon TGF activation (3) and the fact that sympathetic stimulation has been shown to reset TGF (12, 22). It may further indicate that ATP may act to enhance not only the effect of direct sympathetic activity by also adrenergic stimulation at extra junctional sites, which has been proposed as an important signaling pathway given that the range of physiological responses shown for sympathetic activity does not clearly coincides with anatomical findings (2).
Baseline calcium measurements are comparable to what we have previously shown to be normal for the distal afferent arteriole with attached glomerulus (11, 21). Although the dose response of calcium to increasing concentrations of NE was statistically increased by costimulation with 10−6 mol/l ATP at 10−8 and 10−7 mol/l NE, the numerical difference was small (Fig. 7, A and B) compared with the contractile response seen at lower concentrations (Fig. 5). This suggests that the increased sustained contraction is not only induced by calcium-dependent signaling, but may also depend on greater sensitivity of the contractile apparatus to calcium as it has been shown for the adenosine ANG II interaction (23). It has been shown that calcium sensitivity may increase through activation of Rho kinase, protein kinase C, and p38 mitogen activated protein kinase leading to increased phosphorylation of the myosin light chain (23, 30, 33).
Our results indicate that activation of a P2-type ATP receptor is necessary for the increased response to NE seen in the afferent arteriole of the mouse (Fig. 8A). Both the α2-blocker yohimbine and the α1-receptor blocker prazosin were able to block the sensitization (Fig. 8, B and C). This may indicate a general dependence on α-receptors. However, one has to take into consideration that prazosin is not completely selective in the mouse (17, 24).
The rightward shift of the dose-response curve of NE in response to prazosin (Fig. 8B) indicates a considerable role of the α1-receptor in mediating NE-induced contractions. This is consistent with findings by other investigators (4, 7). The minimal effect of α2-inhibition on the contraction caused by NE (Fig. 7C) suggests a minor contribution of α2-receptors to the contraction caused by NE alone. The finding is also in agreement with earlier results (4, 7). There are, however, results showing that both α1- and α2-receptors significantly contribute the effect of sympathetic stimulation on renal blood flow (8). This may be interpreted as consistent with the present finding of a dual dependence on α1- and α2-receptors for the effect of ATP costimulation on afferent arteriolar sensitivity to NE.
The idea that both α1- and α2-adrenoceptors are important for the interaction of ATP with NE is further supported by the inability of ATP to change the dose-response curves of the selective agonists phenylephrine and clonidine.
In conclusion, we have shown that the enhancing effect of ATP on sympathetic regulation of renal blood flow may be explained by mechanisms intrinsic to the afferent arteriole. Furthermore, the sensitized response to NE by costimulation with ATP is dependent on both α1- and α2-adrenoceptors and P2-type ATP receptors.
Perspectives and Significance
The present work may serve to highlight the importance of interactions between different transmission systems in the kidney. An observation of interest is the relatively small changes in calcium signaling seen during the augmented contractions, indicating a significant increase in calcium sensitivity or the activation of calcium-independent contraction. This is an area that warrants deeper investigation to completely understand the mechanisms. Furthermore, the interaction of ATP and NE signaling may provide a mechanism for the interaction between the TGF and sympathetic activity reported earlier.
↵* M Hultström and E. Y. Lai contributed equally to this study.
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 © 2007 the American Physiological Society