| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Cellular and Molecular Physiology, Yale University School of Medicine, New Haven, Connecticut
Submitted 22 December 2006 ; accepted in final form 7 March 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
kidney; angiotensin; acid-base; out-of-equilibrium CO2/HCO3– solutions; fluid reabsorption
Part of the response to respiratory acidosis is mediated by the renal proximal tubule (PT), which reabsorbs a near-isosmotic fluid that represents about two-thirds of the fluid and 
80% of the HCO3– filtered by the glomerulus. The PT cell actively secretes H+ into the tubule lumen (2, 7, 71) and uses this H+ to titrate filtered HCO3– in the lumen to CO2 and H2O, catalyzed by apical carbonic anhydrase IV (11, 72, 74). The newly formed CO2 and H2O then diffuse into the PT cells by crossing the apical membrane of PT, where carbonic anhydrase II (72, 74) catalyzes the regeneration of H+ and HCO3–. The cell extrudes the H+ across the apical membrane via Na/H exchangers (5, 6, 49, 54, 69, 77) and H+ pumps (29) and exports the HCO3– across the basolateral membrane, mainly via the electrogenic Na/HCO3 cotransporter (8, 19, 64, 65).
In previous work on rabbit S2 proximal tubules, in which we used out-of-equilibrium (OOE) CO2/HCO3– solutions to alter [CO2], [HCO3–], or [H+] one at a time (84, 85), we demonstrated that, at least in regard to acute acid-base disturbances, H+ secretion responds not to changes in basolateral pH, but to changes in basolateral [CO2] and [HCO3–] (90). In addition to acid-base disturbances, another powerful modulator of PT acid-base transport is angiotensin II (ANG II). Many investigators have shown that ANG II, added to the apical or basolateral solution, has a biphasic concentration-dependent effect on the rates of fluid reabsorption (JV) and HCO3– reabsorption (JHCO3) by the PT, increasing them at low concentrations and decreasing them at high concentrations (13, 18, 22, 28, 31, 43, 70, 75, 80). Our previous work shows that the stimulatory effect of low-dose ANG II (applied to either the apical or basolateral side) on HCO3– reabsorption is greatest at low levels of basolateral CO2 and falls as CO2 approaches 20%. These results are consistent with the idea that the signal-transduction pathways for ANG II and basolateral CO2 intersect or perhaps even merge to produce similar end effects (88).
In 1984, based on differences in the affinity of 125I-labeled ANG II for binding to rat liver membranes, Gunther (30) characterized two classes of ANG II receptor subtypes, a high-affinity "A1" and a low-affinity "A2." The cDNA encoding the AT1 receptor, a G protein-coupled receptor that appears to mediate the major physiological cardiovascular actions of ANG II, was first cloned and sequenced from rat vascular smooth muscle cells (50) and bovine adrenal gland (66). The cDNA encoding the AT2 receptor, whose deduced amino acid sequence is only one-third identical to that of AT1, was first cloned from a rat fetus (48) and rat pheochromocytoma cells (37). The AT1 receptors produce the familiar cardiovascular effects of ANG II (e.g., renal fluid reabsorption, aldosterone secretion, and vasoconstriction) as well as cell growth and proliferation. AT2 receptors appear to be antiproliferative and promote differentiation or apoptosis (20).
Rodents have two distinct rat AT1 receptor genes, the original AT1A as well as AT1B (25, 35, 67), which are 96% identical at the amino acid level (35). These variants have a distinctive tissue distribution. The AT1A subtype is expressed mainly in liver, lung, vascular smooth muscle, and kidney. The AT1B subtype is expressed predominantly in adrenal and anterior pituitary (20). The AT1A receptor, expressed in the basolateral and the apical membranes of the PT, appears to mediate the biphasic effects of ANG II on JHCO3 (87) and NBCe1 activity (32) in mice. AT1B may also contribute to the stimulatory effect of ANG II on NBCe1 (32).
In the present study, we examine the importance of apical ANG II receptors in the stimulation of HCO3– reabsorption by basolateral (BL) CO2 induced in renal proximal tubules. Our approach was to perfuse isolated, S2 PTs from the rabbit or PTs from the mouse and to use OOE solutions to vary basolateral CO2 concentration ([CO2]BL) from 0 to 20% while keeping basolateral HCO3– concentration ([HCO3–]BL) and pH (pHBL) fixed near their physiological values. We found that an active apical AT1A receptor is necessary for baseline HCO3– reabsorption (i.e., at 5% CO2) and for the tubule to transduce the CO2 signal to an increase in JHCO3.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
All of experiments were carried out in 1) "pathogen-free" female rabbits (New Zealand White, Elite, Covance, Denver, PA) or 2) male and female AT1A knockout (KO) and wild-type (WT) mice (The Jackson Laboratory, Bar Harbor, ME). The mice were littermates, the progeny of parents that were heterozygous for the AT1A gene (+/–). We determined the genotype of the mice as described previously (34), and worked with either +/+ or –/– mice.
According to procedures that were approved by the Yale Animal Care and Use Committee, we obtained and perfused rabbit or mouse proximal tubules. In the case of rabbit tubules, our approach was similar to that using methods that were similar to those originally described by Burg et al. (12) and later modified by Baum et al. (4) and also by our laboratory (52, 85). Briefly, we euthanized a rabbit (1.4–2.0 kg) with a single intravenous overdose of pentobarbital sodium (40 mg/kg), surgically exposed and removed the left kidney, obtained coronal sections, incubated these sections in Hanks’ solution, and then microdissected the slice to obtain individual midcortical S2 segments, 1.5–1.7 mm in length (85).
In the case of mouse proximal tubules, we euthanized an AT1A KO mouse or WT mouse (37 ± 1 days, means ± SE) with a single intraperitoneal overdose (20 mg) of pentobarbital sodium. An incision in the abdominal wall exposed the abdominal aorta and left kidney. We perfused the abdominal aorta at 2.5 ml/min with 4°C modified Hanks’ solution (solution 1 in Table 1), immediately cut the left renal vein, and then continued to perfuse for 1–2 min. We rapidly removed the kidney, prepared coronal slices, incubated these in Hanks’ solution, and then microdissected the slice to obtain individual segments of cortical proximal tubules 0.6–0.8 mm in length.
|
55 nl; summarized in Fig. 1B of Ref. 85).
|
We perfused the basolateral side of the tubule (i.e., "bath") at 7 ml/min, with a solution at 37°C.
Experimental Protocol and Solutions
The experimental protocol was similar to that of previous studies (88–90) except for the addition of inhibitors to the luminal and basolateral solutions. These inhibitors were saralasin (A2275, Sigma-Aldrich, St. Louis, MO), candesartan (kind gift of AstraZeneca, Mölndal, Sweden), and PD123319 (P186, Sigma-Aldrich). Table 1 lists the compositions of the solutions, which were identical to those used in the aforementioned studies. Solution 2 was always the luminal perfusate and contained dialyzed [3H]methoxyinulin (PerkinElmer Life Sciences, Boston, MA) as a volume marker. Solution 3, which contained 2% albumin, flowed through the bath during a 20- to 30-min warm-up period at 37°C. After this warm-up period, we collected luminal fluid during two periods. In the first collection period, we always used solution 4 to perfuse the bath. In the second collection period, we perfused the bath with solutions 5 or 6. These were OOE CO2/HCO3– solutions in which we varied [CO2]BL at fixed levels of [HCO3–]BL and pHBL (either solution 5or6) by rapidly mixing streams of two dissimilar solutions (i.e., mixing solutions 5A and 5B to yield solution 5, and mixing solutions 6A and 6B to yield solution 6). All solutions had osmolalities of 300 ± 2 mosmol/kgH2O.
Measurement of JHCO3 and JV
Our approach for measuring JHCO3 (pmol·min–1·mm–1 tubule length) and JV (nl·min–1·mm–1) was similar to that used by McKinney and Burg (46). We determined total CO2 in aliquots of the perfusate and collected fluid using a WPI "NanoFlo" device (World Precision Instruments, Sarasota, FL) and our own reagents, as described elsewhere (89). As described previously (90), the mean JHCO3 (or JV) values that we report for either rabbits or mice in the first of two collection periods, the first collection period being always that with the "control" condition (i.e., equilibrated 5% CO2/22 mM HCO3–/pH 7.40), are simple averages (i.e., not normalized). The mean values in the second of two collection periods, the second collection period being always the one with the "experimental" condition, were normalized as follows. For each tubule, we divided the JHCO3 (or JV) value obtained during the second collection period by the comparable value obtained during the first collection period for that individual tubule. The result is a pair of ratios (one for JHCO3 and one for JV) for the second/first collection periods for an individual tubule. For each tubule, we then multiplied the second/first ratio for JHCO3 (or JV) by the unnormalized mean JHCO3 (or JV) value that we obtained during the first collection periods in all experiments following the identical protocol. We have used this identical approach in three previous studies (88–90).
Data Analysis
For comparisons of two means, we performed two-tailed, unpaired t-tests using the Analysis Toolpack of Microsoft Excel. For comparisons of more than two means, we performed a one-way ANOVA and Dunnett's multiple comparison using KaleidaGraph (version 4, Synergy Software). Results are given as means ± SE, with the number of tubules (n) from which it was calculated.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
In all rabbit PT experiments reported elsewhere (88, 89), as well as those in the present study, we randomly perfused the tubules either orthograde (i.e., perfusate flowing from the direction of the S1 to the direction of the S3 segment) or retrograde (S3
S1), and recorded the orientation. Table 2 summarizes the effect of tubule orientation on the JHCO3 and JV data, accumulated over a 4-yr period, and shows that the direction of luminal flow does not have a significant effect on either JHCO3 or JV.
|
The open circles in Fig. 1 are our control (i.e., drug-free) data on rabbit S2 proximal tubules, showing the effect of raising [CO2]BL or bath [CO2] from 0 to 20% at a fixed [HCO3–]BL of 22 mM and pHBL of 7.4. These data are from our earlier studies (89, 90), with the 5% data being augmented by 6 points from the present study.
To examine the role of apical and basolateral ANG II receptors in the basolateral CO2 dependence of JHCO3 and JV, our first approach was to perfuse the PT lumen with solution 2containing 10–8 M saralasin, a peptidic antagonist (14) of both AT1 and AT2 ANG II receptors. The IC50 values are 0.43 nM for AT1 receptors (50) and 0.09 nM for AT2 receptors (3). The luminal saralasin was present throughout the entire experiment. During the first of two collection periods, we perfused the bath of rabbit S2 proximal tubules to an equilibrated CO2/HCO3– solution, where [CO2]BL = 5%, [HCO3–]BL = 22 mM, and pHBL = 7.4 (solution 4). During the second collection period, the bath contained an OOE solution to lower [CO2]BL to 0% (solution 5) or to raise [CO2]BL to 20% (solution 6) while fixing [HCO3–]BL at 22 mM and pHBL at 7.4. As shown by the filled circles in Fig. 1A, luminal saralasin largely blunted the effect on JHCO3 of raising [CO2]BL.
To assess our JHCO3 data quantitatively, we applied a one-way ANOVA to all of the data at 0, 5, and 20% CO2 in Fig. 1A as well as in Figs. 2A and 3A. The overall P value for the three JHCO3 groups (0, 5, and 20% CO2) was 0.89 for [CO2]BL = 0% and <0.0001 for [CO2]BL = 5% and [CO2]BL = 20%.
|
|
), we used Dunnett's multiple comparison, which indicates that luminal 10–8 M saralasin produces a statistically significant decrease in JHCO3 from 54 ± 3 to 35 ± 5 pmol·min–1·mm–1 at a [CO2]BL of 5% (P = 0.0015) and from 83 ± 2 to 36 ± 7 pmol·min–1·mm–1 at a [CO2]BL of 20% (P < 0.0001). However, saralasin had no significant effect on JHCO3 at a [CO2]BL of 0% (28 ± 7 vs. 33 ± 2 pmol·min–1·mm–1, P = 0.79).These data indicate that an apical ANG II receptor is required for baseline HCO3– reabsorption as well as for the transduction of the basolateral CO2 signal in PT cells.
In a second series of experiments, we perfused the PT lumen throughout the experiment with solution 2, but with no addition of saralasin. During the first collection period, the bath contained solution 4 without saralasin (Fig. 1, 
). During the second collection period, the bath contained solution 6 with 10–8 M saralasin (Fig. 1, 
).
To assess the effects of basolateral 10–8 M saralasin on the basolateral CO2 dependence of JHCO3 at a [CO2]BL of 20% (Fig. 1A, ), we again used Dunnett's multiple comparison, which indicates that basolateral 10–8 M saralasin does not produce a statistically significant change in JHCO3 (86 ± 7 vs. 83 ± 2 pmol·min–1·mm–1, P = 0.96).
Figure 1B summarizes the JV data (, ). To assess our JV data quantitatively, we again applied a one-way ANOVA to all of the data at 0, 5, and 20% CO2 in Figs. 1B, 2B, and 3B. The overall P value for three JV groups (0, 5, and 20% CO2) was 0.15 for [CO2]BL = 0%, 0.09 for [CO2]BL = 5%, and 0.03 for [CO2]BL = 20%. Dunnett's multiple comparison shows that luminal 10–8 M saralasin had no statistically significant effect at 0% CO2 (P = 0.44), 5% CO2 (P = 0.08), or 20% CO2 (P = 0.46) on the JV of rabbit S2 proximal tubules. Similarly, basolateral 10–8 M saralasin had no statistically significant effect on JV at 20% CO2 (P = 0.11).
Effects of Luminal Candesartan on Basolateral CO2 Dependence of JHCO3 and JV
Our saralasin results show that an apical but not a basolateral ANG II receptor is required for baseline HCO3– reabsorption as well as for the transduction of a basolateral CO2 signal to a change in JHCO3. In the following series of experiments, we examine the role of apical AT1 receptors. Throughout the experiment, we perfused the PT lumen with solution 2 containing 10–8 M candesartan, a specific nonpeptide AT1 receptor inhibitor with an IC50 of 0.9 nM (26). During the first collection period, we exposed the basolateral surface of rabbit S2 proximal tubules to an equilibrated CO2/HCO3– solution: [CO2]BL = 5%, [HCO3–]BL = 22 mM, and pHBL = 7.4 (solution 4). During the second collection period, the bath contained an OOE solution to lower [CO2]BL to 0% (solution 5) or to raise [CO2]BL to 20% (solution 6) while fixing [HCO3–]BL at 22 mM and pHBL at 7.4. As shown by the filled diamonds in Fig. 2A, luminal candesartan lowers baseline JHCO3 and eliminates the effect on JHCO3 of raising [CO2]BL. The control data () in Fig. 2A are the same as in Fig. 1A.
Dunnett's multiple comparison indicates that luminal 10–8 M candesartan produces a statistically significant decrease in JHCO3 from 54 ± 3 to 31 ± 3 pmol·min–1·mm–1 at a [CO2]BL of 5% (P = 0.0002) and from 83 ± 2 to 31 ± 5 pmol·min–1·mm–1 at a [CO2]BL of 20% (P < 0.0001). However, candesartan had no significant effect on JHCO3 at a [CO2]BL of 0% (31 ± 4 vs. 33 ± 2 pmol·min–1·mm–1, P = 0.99). These data indicate that an apical AT1 receptor is required for baseline HCO3– reabsorption and for the transduction of the basolateral CO2 signal in PT cells.
Figure 2B summarizes the JV data (
). Dunnett's multiple comparison shows that luminal 10–8 M candesartan had no statistically significant effect at 0% CO2 (P = 0.07), 5% CO2 (P = 0.99), or 20% CO2 (P = 0.43) on the JV of rabbit S2 proximal tubules. The control data () in Fig. 2B are the same as in Fig. 1B.
Effects of Luminal PD123319 on Basolateral CO2 Dependence of JHCO3 and JV
Our observation that saralasin or candesartan eliminates the proximal tubule's JHCO3 response to changes in [CO2]BL suggests that an apical AT1 receptor is at least permissive for a component of baseline HCO3– reabsorption and for the CO2 signal-transduction pathway. In the following series of experiments, we examine a potential role of an apical AT2 receptor. We perfused the lumen of the PT throughout the experiment with solution 2 containing 10–7 M PD123319, which is a specific AT2 receptor inhibitor with an IC50 of 1.7 nM (38). During the first collection period, we exposed the basolateral surface of rabbit S2 proximal tubules to an equilibrated CO2/HCO3– solution-[CO2]BL = 5%, [HCO3–]BL = 22 mM, and pHBL = 7.4 (solution 4). During the second collection period, the bath contained an OOE solutions to lower [CO2]BL to 0% (solution 5) or to raise [CO2]BL to 20% (solution 6) while fixing [HCO3–]BL at 22 mM and pHBL at 7.4. The filled squares in Fig. 3A show that PD123319 has no effect on the JHCO3 response to alterations in [CO2]BL. The control data () in Fig. 3A are the same as in Figs. 1A and 2A.
Dunnett's multiple comparison indicates that luminal 10–7 M PD123319 does not produce a significant effect on JHCO3 at a [CO2]BL of 0% (33 ± 2 vs. 31 ± 4 pmol·min–1·mm–1, P = 0.98), a [CO2]BL of 5% (54 ± 3 vs. 59 ± 5 pmol·min–1·mm–1, P = 0.70), or a [CO2]BL of 20% (83 ± 2 vs. 88 ± 7 pmol·min–1·mm–1, P = 0.82). These data indicate that an apical AT2 receptor is not required for baseline HCO3– reabsorption or for the transduction of the basolateral CO2 signal in proximal PT cells.
Figure 3B summarizes the JV data (
). Dunnett's multiple comparison shows that luminal 10–7 M PD123319 had no statistically significant effect at 0% CO2 (P = 0.23), 5% CO2 (P = 0.99), or 20% CO2 (P = 0.99) on the JV of rabbit S2 PT. The control data () in Fig. 3B are the same as in Figs. 1B and 2B.
Effects of Knocking Out AT1A Receptor on Basolateral CO2 Dependence of JHCO3 and JV in Mouse PT
The preceding data on rabbit proximal tubules indicate that an apical AT1 receptor plays a very important role in response of proximal tubules to change [CO2]BL. In our final series of experiments, we randomly used male or female AT1A-null mice to examine the role of side AT1A receptor on the basolateral CO2 dependence ofJHCO3 and JV. In control experiments, we used WT mice (AT1A +/+) with same genetic background as the AT1A KO mice (AT1A –/–). Throughout all experiments, we perfused the PT lumen with solution 2. The basolateral protocol was identical to the one we used in the rabbit PT experiments. That is, during the first collection period, we exposed the basolateral surface of to an equilibrated CO2/HCO3– solution (solution 4). During the second collection period, the bath contained an OOE solution with a [CO2]BL of 0% (solution 5) or 20% (solution 6) at a fixed [HCO3–]BL of 22 mM and pHBL of 7.4.
We found that the JHCO3 response of WT mouse PTs to changes in [CO2]BL (Fig. 4A, 
) is very similar to what we had found in rabbit PTs (see in Figs. 1A, 2A, and 3A as well as Ref. 90). On the other hand, the response of the AT1A-null PTs (Fig. 4, 
) was very similar to that of rabbit PTs perfused with candesartan. The major difference was that, compared with the effect of candesartan in the rabbit experiments, knocking out AT1A in mice shifted the JHCO3 vs. [CO2]BL curve somewhat upward. In the rabbit experiments, candesartan uniformly lowered JHCO3 to values that were indistinguishable from the value observed in the absence of candesartan at 0% CO2, whereas in the mouse experiments, knocking out AT1A uniformly stabilized JHCO3 at a value about midway between 0% CO2 and 5% CO2 JHCO3 values observed in WT mice at 5% CO2.
|
Figure 4B summarizes the JV data. Knocking out AT1A did not significantly change JV at a [CO2]BL of 5% (P = 0.539) or at a [CO2]BL of 20% (P = 0.558). However, at a [CO2]BL of 0%, PTs from AT1A KO mice had a significantly higher JV (from 0.58 ± 0.10 to 0.80 ± 0.04 nl·min–1·mm–1, P = 0.041).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
With the discovery that the central cilium senses flow (58–60), it occurred to us that the cilia in a living PT might have become biased to respond differently to flow in the orthograde vs. retrograde directions. However, Table 2 shows that tubule orientation has no effect on either JHCO3 or JV.
Endogenous Angiotensin Production by PT Cells
Previous work. ANG II is a powerful hormone that modulates several cardiovascular functions via angiotensin receptors in the membranes of target cells. Aside from the ANG II in the blood plasma that is generated by the classic renin/angiotensin-converting enzyme (ACE) pathway, ANG II can arise locally in several tissues, including the brain (23, 27, 57); testis (42, 55), epididymis (40–42, 82, 86), and other elements of the male reproductive tract; adrenal gland (1, 56); heart (24, 36) and arterial wall (36, 39, 51); as well as the kidney (9, 16, 33, 73). Indeed, the PT appears to have all the molecular machinery necessary to generate its own luminal ANG II, including angiotensinogen (33, 63, 79, 83), renin or renin-like activity (47, 83), and ACE (15, 44). In addition, kininase activity, which can convert ANG I to ANG II, is present in the PT (44). Moreover, levels of ANG I and ANG II can be much higher in the lumen of the PT than in the blood plasma (17, 53).
Endogenous ANG II appears to promote fluid reabsorption in the PT. Quan and Baum (61) found that perfusing the lumen of rat PTs with either an AT1 antagonist or an ACE inhibitor (10–4 M enalaprilat) decreases JV. Later, Baum et al. (4) showed that perfusing the lumen of rabbit PTs with an AT1 antagonist also reduces JV and that perfusing with either an AT1 antagonist or high levels of an ACE inhibitor reduces JHCO3.
Our JHCO3 data. Our present studies confirm that endogenous, luminal ANG II plays a key role in HCO3– reabsorption by the PT. We found that the indiscriminant blockade of apical angiotensin receptors by saralasin (Fig. 1A) as well as the specific blockade of apical AT1 receptors by candesartan (Fig. 2A) substantially reduce JHCO3 at a [CO2]BL of 5%. Moreover, these same treatments eliminate the response to changes in [CO2]BL. However, indiscriminant blockade of basolateral ANG II receptors by saralasin (Fig. 1A), or specific blockade of apical AT2 receptors by PD123319 (Fig. 3A), was without effect. Our inhibitor data are consistent with the hypothesis that 1) a major component of PT baseline HCO3– reabsorption and 2) all of the ability of the PT to modulate HCO3– reabsorption in response to changes in [CO2]BL require that endogenous ANG II in the tubule lumen bind to an apical AT1 receptor. However, our data indicate that neither apical AT2 receptors nor basolateral ANG II receptors play a substantial role in HCO3– reabsorption, at least under the conditions of our experiments. If our hypothesis concerning the role of apical AT1 receptors is true, then blockade of ACE or ACE-like activity in the PT cell should produce the same effect as blockade of apical AT1 receptors, as we have indeed found in preliminary experiments (Zhou Y and Boron WF, unpublished observations).
Our JV data. We observed no significant effects of indiscriminant blockade of apical or basolateral ANG II receptors with saralasin (Fig. 1B) or of selective blockade of AT1 receptors with candesartan (Fig. 2B) or of AT2 receptors with PD123319 (Fig. 3B).
Flux of other solutes. The open circles connected by the solid line in Fig. 5A are a replot of the control JHCO3 rabbit PT data from Figs. 1A, 2A, and 3A. Similarly, the filled blue circles connected by the solid line in Fig. 5C are a replot of the saralasin JHCO3 data from Fig. 1A, and the filled diamonds connected by the solid line in Fig. 5C are a replot of the candesartan JHCO3 data from Fig. 2A. From these three sets of JHCO3 data and the corresponding JV data, we computed the reabsorption rate of solutes other than NaHCO3 (JOther) using the approach described in the figure legend. In Fig. 5, A and C, these calculated JOther values are represented by symbols connected by dashed lines. As observed previously (90) for rabbit PTs, Fig. 5A shows that, in the absence of inhibitors, JOther changes inversely with JHCO3 which is why JV is relatively stable in the B panels of Fig. 1, Fig. 2, and Fig. 3. As summarized in Fig. 5C, saralasin and candesartan cause JOther, compared with their corresponding control values in Fig. 5A, to rise significantly, except for saralasin at 0% CO2, which is why these inhibitors decrease JHCO3 (Figs. 1A and 2A) but not JV (Figs. 1B and 2B).
|
Response to exogenous ANG II. Given the secretion of endogenous ANG II by the PT, one question that arises is how the tubule should respond to the luminal addition of exogenous ANG II. We (88) as well as others (43, 80) have observed that adding a low dose of ANG II to the lumen increases JHCO3. On the other hand, adding higher doses of ANG II to the lumen either has no effect or reduces JHCO3 (80, 88). Presumably, the actual luminal ANG II concentration ([ANG II]) in these studies is the sum of secreted and added ANG II, Although it is possible that the very addition of exogenous ANG II modifies the secretion of endogenous ANG II. Clearly, understanding how added, luminal ANG II affects the PT will require either measuring the actual luminal [ANG II] or blocking the secretion of endogenous ANG II.
Effects of Knocking Out the AT1A Receptor
Previous work. In 1995, Ito et al. (34) and Sugaya et al. (76) independently disrupted the AT1A gene in mice. Both groups observed that the AT1A-null mice are hypotensive. In 2002, Horita et al. (32) used measurements of intracellular pH (pHi) in collapsed PTs from AT1A-null vs. WT mice to show that the AT1A receptors are required for the biphasic response of the electrogenic Na/HCO3 cotransporter to low- vs. high-dose ANG II. Inhibitor data indicated that AT2 receptors are not involved in this response. In 2003, Zheng et al. (87) used a stop-flow microperfusion technique and luminal pH measurements to examine the effect of this knockout on the response of the isolated, perfused mouse PT to exogenous ANG II, added to the lumen. Although these authors did not explicitly comment on the effect of the knockout on baseline JHCO3, examination of their Figs. 1 and 2 suggests that knocking out the AT1A receptor did not substantially alter baseline JHCO3. However, Zheng et al. clearly demonstrated that luminal ANG II fails to elicit biphasic effects on JHCO3 in tubules from AT1A-null mice. These two sets of data are consistent with the hypothesis that the apical AT1A receptor is the only ANG II receptor necessary for the JHCO3 response to exogenous luminal ANG II.
Our JHCO3 data. The results from the present study demonstrate that, in PTs from WT mice, changes in [CO2]BL produce nearly the same fractional changes in JHCO3 as we had previously observed in rabbit S2 PTs in the absence of inhibitors (90). Moreover, we find that 1) the AT1A receptor is critical for baseline JHCO3. That is, we found that tubules from AT1A-null mice had a somewhat lower JHCO3 than tubules from WT mice at 5% CO2 (Fig. 4A). Although the data of Zheng et al. (87) do not reveal such an effect, those authors did not explicitly compare baseline JHCO3 values in WT vs. AT1A-null mice, and they used a different technical approach (stop-flow luminal pH measurements in microperfused PTs) to infer effects on JHCO3. 2) The AT1A receptor is critical for the JHCO3 response to changes in [CO2]BL (Fig. 4A, 0 vs. 5% and 20 vs. 5% CO2).
Both of the above effects presumably involve endogenous ANG II. It is interesting that, although the acute blockade of ANG II receptors reduced JHCO3 at 5% CO2 (filled symbols in Figs. 1A and 2A) to a value indistinguishable from that observed in the absence of inhibitors at 0% CO2 (open symbols in Figs. 1A and 2A), the knockout of the AT1A receptor only produced a partial reduction of JHCO3 at 5% CO2 (Fig. 4A). In other words, the AT1A-null tubules seem to have upshifted the JHCO3 "pedestal" (i.e., JHCO3 at 0% CO2), above which changes in [CO2]BL are able to modulate JHCO3. Perhaps the upshifting of this pedestal reflects a compensation by the animal to minimize the tendency toward metabolic acidosis. In the signal-transduction cascade that links the as-yet-unidentified basolateral CO2 sensor to the acid-base transporters, this compensation would have to act downstream from the AT1A receptor. For example, it would be interesting to determine whether AT1A-null tubules have increased expression of NHE3 and/or NBCe1.
Flux of other solutes. The open triangles connected by the solid line in Fig. 6A are a replot of the WT JHCO3 data from Fig. 4A. In Fig. 6A, the open black triangles connected by black dashed lines are the JOther values that we computed, as in Fig. 5, A and C, assuming that the PT resorbate had an osmolality of 300 mosmol/kgH2O. Note that the computed JOther values at 5 and 20% CO2 are negative, which would correspond to the secretion (rather than reabsorption) of other solutes. Similarly, the computed [Other] values at 5 and 20% CO2 are negative; the absolute values of these values would correspond to the concentrations in the secreted fluid. However, we presume that such a secretion is unlikely and that the negative JOther and [Other] values are instead the result of our inappropriately assuming that the osmolality of the PT resorbate is 300 mosmol/kgH2O.
|
167 mM at 5% CO2 (Fig. 6B) requires that the osmolality of the absorbate be at least 335 mosmol/kgH2O. Thus, although investigators typically assume that the PT reabsorbs an isosmotic fluid, the mouse PT, at least under the conditions of our experiments, must be reabsorbing a slightly hypertonic fluid. The data of others are consistent with this observation. In free-flow micropuncture studies, Vallon et al. (78) observed that late-PT fluid has an osmolality that is 12 mosmol/kgH2O lower than that of the blood plasma. Based on their data and that of Schnermann et al. (68), we calculate that the resorbate of WT mice is typically 334 mosmol/kgH2O.1
Our computed [NaHCO3] at 20% CO2 is 224 mM (Fig. 6B), which requires that the osmolality of the absorbate be at least 447 mosmol/kgH2O. Thus, with HCO3– reabsorption nearly maximally stimulated, the osmotic water permeability (Pf) of the mouse PT is far lower than that required for reabsorbing an isosmotic fluid. Note that, in rabbit PTs, we calculate that, even at 20% CO2, the resorbate is approximately isosmotic NaHCO3 (see Fig. 5B). Thus, in rabbit tubules, the Pf is able to keep up with maximally stimulated NaHCO3 reabsorption. That is, compared with mouse PTs, rabbit PTs must have a lower ratio of (maximal JHCO3)/Pf.
The values represented by the red symbols in Fig. 6, A and B, follow from the assumption that JOther, and thus [Other], is always 
0. This analysis thus assumes that, at [CO2]BL values of 5 and 20%, the PT transports only NaHCO3 across the epithelium; that is, that JOther = 0 and [Other] = 0. If JOther were 0, then the actual osmolality of the reabsorbate would be even greater than the values indicated in Fig. 6B.
The filled triangles connected by the solid lines in Fig. 6C are a replot the AT1A-null JHCO3 data from Fig. 4A. The filled triangles connected by the dashed lines are the JOther values that we computed, assuming that the PT resorbate had an osmolality of 300 mosmol/kgH2O. In a one-way ANOVA of the JOther data of AT1A-null mice in Fig. 6C, the overall P value was 0.62. Dunnett's multiple comparison showed that the JOther value at 5% CO2 was not significantly different from either the value at 0% CO2 (P = 0.88) or 20% CO2 (P = 0.52). Thus, in PTs from AT1A-null mice, changes in [CO2]BL not only failed to alter JHCO3, they also failed to alter JOther, which is why JV is relatively stable in Fig. 4B.
We used a two-tailed unpaired t-test to compare each of the JOther values for AT1A-null mice in Fig. 6C with the corresponding JOther data for WT mice, represented by the filled symbols in Fig. 6A (i.e., assuming that the reabsorbate had an osmolality of 300 mosmol/kgH2O). The filled characters in Fig. 6C show that the JOther values for AT1A-null tubules at 5 and 20% CO2 are significantly higher than the corresponding values for WT tubules. We also compared the AT1A-null JOther data in Fig. 6C with the WT JOther data represented by the red symbols in Fig. 6A (assuming that JOther is always 0). The red characters in Fig. 6C show that the JOther value for AT1A-null tubules at 5% is the only one that is significantly higher than the corresponding values for WT tubules.
We used a two-tailed unpaired t-test to compare each of the [NaHCO3] and [Other] values for AT1A-null PTs in Fig. 6D with the corresponding wild-type data in Fig. 6B. The analysis shows that the calculated resorbate [NaHCO3] of AT1A-null PTs is less than that of the wild-type PTs at 5 and 20% CO2. The results for [Other] are identical to those summarized above for JOther.
Thus, at 5 and 20% CO2, knocking out AT1A receptors causes reciprocal changes in JOther and JHCO3 and also causes reciprocal changes in [Other] and [NaHCO3].
Conclusions. Our data, as well as the data of others, suggest that the PT cell has the ability to generate luminal angiotensin II, which then binds to apical AT1A receptors. Three nonexclusive possibilities for the mechanism of this endogenous angiotensin production are that the PT secretes 1) angiotensinogen, which is cleaved by apical/luminal renin-like activity to produce the inactive ANG I, which is then cleaved to active ANG II by apical/luminal ACE-like activity; 2) ANG I, which is then converted to ANG II; or 3) ANG II directly. If option 1 were true, then blocking apical/luminal renin-like activity should reduce baseline JHCO3 and eliminate the response to changes in [CO2]BL. If option 2 were true, then blocking apical/luminal ACE-like activity should reduce baseline JHCO3 and similarly eliminate [CO2]BL sensitivity. If option 3 were true, then blocking apical/luminal renin- or ACE-like activity should have no effect on JHCO3.
Our JHCO3 data with basolaterally applied saralasin suggest that the PT either does not secrete endogenous ANG II into the basolateral solution or, if it does, that the ANG II is washed away before having an effect. Another possibility is that the PT secretes angiotensinogen or ANG I across the basolateral membrane, but that the enzymes are not present to convert these substances to ANG II.
Although the endogenously produced ANG II is not important for baseline JV (i.e., at [CO2]BL = 5%), it is critical for baseline JHCO3. To our knowledge, the present study is the first to examine JV in AT1A-null mice. In rabbit tubules, we found that acutely blocking apical receptors with either saralasin or candesartan reduces JHCO3 to the pedestal value observed at 0% CO2 in the absence of inhibitors. Chronically eliminating AT1A receptors in AT1a-null mice produces a smaller reduction in baseline JHCO3. The present paper is also the first to demonstrate that the interaction of endogenously produced ANG II with AT1A receptors is at least permissive for the modulation of HCO3– reabsorption by renal proximal tubules in response to changes in [CO2]BL.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
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.
1 Given an osmolality of 309 mosmol/kgH2O for late-PT fluid and a transtubular osmotic gradient of 12 mosmol/kgH2O (78), we calculate a plasma osmolality of 321 mosmol/kgH2O. Given a single-nephron GFR of 9.6 nl/min, a collection rate of 4.9 nl/min, and therefore an absorption rate of 4.7 nl/min (68), we calculate that a typical PT reabsorbed osmolytes at the rate of 1,568 posmol/min. Thus the osmolality of the reabsorbate would be (1,568 posmol/min)/(4.7 nl/min) = 334 mosmol/kgH2O. 
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
0.05, **P 