INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

ACUTE HYPERTENSION provokes a rapid decrease of NaCl reabsorption in proximal tubules and a subsequent increase of proximal tubular flow rate and distal NaCl delivery in Sprague-Dawley (SD) rats (10, 11). The increased tubular flow and reduced NaCl reabsorption in proximal tubules contribute to the increased mass flow of pressure natriuresis, although more distal natriuretic adjustments are also undoubtedly involved (17). The inhibition of NaCl reabsorption in proximal tubules also provides the error signal required by tubuloglomerular feedback to autoregulate the renal blood flow in response to a sustained increase of arterial pressure (10). Pressure natriuresis is known to occur even when renal blood flow and glomerular filtration rate are held constant by autoregulation. Thus the process that initiates pressure natriuresis has a central role in renal autoregulation, but the mechanism of the inhibition has not been identified. Using a subcellular fractionation membrane strategy and immunoblotting, Zhang et al. (44) observed that Na⁺/H⁺ exchanger (NHE) isoform NHE3 was redistributed from renal cortex brush-border membranes into putative intracellular membranes and that Na⁺-K⁺-ATPase activity was decreased in acute hypertension. Urine output and clearance of endogenous lithium were increased by threefold when acute hypertension was induced (44). Since Na⁺ is reabsorbed in the proximal tubule via apical NHE, removal of apical NHE from brush border will result in inhibition of Na⁺ reabsorption. Similarly, deactivation of basolateral Na⁺-K⁺-ATPase will reduce the Na⁺ gradient across the apical membrane, which drives Na⁺ reabsorption. Therefore the response of both transporters can account for inhibition Na⁺ reabsorption and natriuresis observed in acute hypertension (10). However, the subcellular membrane fractionation approach provides limited spatial resolution of this redistribution process. A recent study in immunolocalization of apical NHE3 with high-resolution optics indicated that there are putative intracellular stores of NHE3 beneath the brush border (7). The observation of immunocytochemistry was substantiated with immunoelectron microscopy in the same report. Thus the redistribution of NHE3 induced by acute hypertension might be detectable by immunocytochemistry, if an appropriate detector was used such as a point-scanning, laser-confocal microscope. The lateral resolution of confocal fluorescence microscopy is known to be improved by 30% compared with epifluorescence microscopy, and the out-of-focus fluorescence is excluded from the detector (29).

Spontaneous variations of mammalian blood pressure fluctuate over a wide range of amplitude and frequencies (21, 30). Acute changes of 30-50 mmHg in arterial pressure occur intermittently in vivo. If the redistribution of NHE protein in proximal tubules is physiologically relevant in regulating urinary Na⁺ excretion, then the increase of arterial pressure over a longer time scale in experimental hypertension should also provoke the redistribution. Alternatively, the redistribution process might be impaired in experimental hypertension (4, 35, 36). These issues were addressed in the present study by coupling confocal fluorescence microscopy with immunocytochemistry. The aims of the present study were to acquire the spatial information of NHE3 redistribution induced by acute hypertension in proximal tubules and to determine whether the apical NHE redistribution can be detected in spontaneously hypertensive rats (SHR) and in two-kidney, one-clip (2K-1C) Goldblatt hypertensive rats (GH).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animal preparations. Four groups of male rats were used: normotensive SD controls (250-300 g); 5-wk- and 12-wk-old SHR; SD rats in which hypertension was induced by placing a constricting clip (0.25 mm diam) around the right renal artery, leaving the other kidney untouched (2K-1C Goldblatt hypertension, GH rats); and sham-operated SD controls. All rats were purchased from Harlan Farms. The rats had free access to food and tap water before the experiments. Anesthesia was induced by placing each rat in a chamber containing 5% halothane administered in 25% oxygen-75% nitrogen through a Fluotec Mark-3 vaporizer. A tracheostomy was performed, and the rats were placed on a servo-controlled heated operating table, which maintained the body temperature at 37°C. The tracheostomy tube was connected to a small animal respirator (Harvard model 683) adjusted to maintain arterial blood pH between 7.35 and 7.45 with a mixture of 25% oxygen-75% nitrogen. The final concentration of halothane needed to maintain sufficient anesthesia was around 1%.

Tissue preparation for immunofluorescence. Kidneys from rats with or without induced acute hypertension were perfusion fixed with a mixture of fresh 4% paraformaldehyde and 4% sucrose in PBS by retrograde perfusion via a cannula inserted into the descending aorta distal to the renal artery. The kidneys were then removed, and blocks of tissue were cut into 4-mm cubes. Blocks of tissue were postfixed overnight in the same fixative at 4°C, washed with 0.1 M aqueous NH₄Cl solution for 30 min, and cryoprotected by incubation in 2.3 M sucrose in PBS for 1 h. The tissue blocks were next snap-frozen in freezing medium with liquid nitrogen-cooled isopentane. Cryosections (7 µm) were cut and transferred to Fisher Superfrost Plus-charged glass slides and air dried.

Indirect immunofluorescence. Sections were first incubated with 1% SDS in PBS for 5 min for antigen retrieval as suggested by Brown et al. (8). After removal of SDS by washing in PBS, the sections were blocked with a mixture of 20% donkey serum in PBS and unconjugated donkey anti-rabbit IgG antibody (1:10) for 20 min and then incubated with primary antibodies after washing (31). After 2-h incubation, the sections were washed three times with fresh PBS and incubated with the appropriate secondary antibodies for 60 min. All incubations were carried out in a moistened chamber at room temperature. The sections were then rinsed with fresh PBS three times and mounted with Gel-Mount (Fisher). For control, frozen sections were incubated with the anti-NHE3 antisera preabsorbed for 1 h with the corresponding immunogen (glutathione-S-transferase/NHE3, 200 µg/ml).

Image acquisition. Sections were examined with a Zeiss Axiovert 100TV inverted microscope coupled to an MRC-1000 confocal scanning unit (Bio-Rad) equipped with a krypton-argon laser, three photomultiplier detectors, and one transmitted light detector. All images were acquired with Zeiss plan-apochromat objectives (×20, NA 0.75; 63×, NA 1.4). All images were collected at ~300 µm from renal capsule and in the immediate vicinity of a glomerulus, where cross sections of S1 and S2 tubules are abundant. High-resolution images were acquired with the ×63 objective at a zoom factor set to 3.6 in the confocal-scanning unit. A zoom factor of 3.6 is the upper limit of zooming to improve the resolution without violating the criteria imposed by the sampling theory in collecting confocal images (38). The calculation was based on an array size of 768 × 512 pixels in the detector and a laser line of 647 nm in excitation. Autofluorescence images collected with FITC filter (excitation 488 nm, emission 522/35 nm) were used to indicate the position of brush border in proximal tubules.

Antibodies. Antiserum against NHE3 (Ab1381, 1:50) was raised in rabbit against the fusion protein of glutathione-S-transferase and the last 85 amino acids of NHE3, which is a sequence unique to this isoform (22, 37). Monoclonal anti-' and -adaptin antibody (1:50) was purchased from Sigma Chemical to label clathrin-coated pits. Polyclonal anti-dipeptidyl peptidase IV (DPP IV) antibody (1:100) was kindly provided by Dr. M. Farquhar and was used as marker of brush-border plasma protein. Polyclonal anti-human Tamm-Horsfall protein antibody was used to identify thick ascending limb in serial sections (1:400; Biomedical Technologies). Primary antibodies were visualized with either donkey indodicarbocyanine (CY5)-conjugated or Lissamine rhodamine (LRSC)-conjugated secondary antibodies (1:400; Jackson Immunoresearch Laboratory). All secondary antibodies used were affinity purified for multiple labeling. BSA (0.2%) and Triton X-100 (0.1%) were included in PBS for antibody dilution.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Cryosections harvested from six SD rats were used to determine the distribution of NHE3 in the proximal tubules. The general distribution of NHE3 in the renal cortex was shown in Fig. 1A (acquired with a ×20 objective). NHE3 was localized in the apical surface of proximal tubules (2, 6, 7) and thick ascending limbs in renal cortex (2, 7). The signal of immunofluorescence was stronger in thick ascending limb than in proximal tubules as reported previously (2). The immunofluorescence signal on the brush border was significantly reduced after preincubation of the antibody with the corresponding fusion protein (1:50) used to raise the antibody (Fig. 1B), but not if only the glutathione-S-transferase was incubated with the antisera (data not shown). These observations confirm the previously observed specificity of the antisera in membrane immunoblotting (22). The specific distribution of NHE3 on the brush border in proximal tubules is shown in Fig. 2, A and B, (acquired with a ×63 objective). Figure 2B is a high-resolution image from a proximal tubule in Fig. 2A. Figure 2C is the immunofluorescence of adaptin of the same tubule imaged with double labeling. Adaptin, which labels clathrin-coated vesicles, was localized at the base of microvilli as expected (5, 9). Figure 2D was the confocal autofluorescence image of the same tubule to indicate the position of brush border. Proximal tubule is known for its strong autofluorescence when excited with the 488-nm line of an argon laser (39).

View larger version (72K): [in this window] [in a new window]   Fig. 1.   Immunofluorescence of Na+/H+ exchanges isoform NHE3 in renal cortex of Sprague-Dawley rat. Cryosections were stained with anti-NHE3 serum (A) or anti-NHE3 serum preabsorbed with NHE3 fusion protein (B). Arrows point to the renal capsule. TAL, thick ascending limb. NHE3 was visualized with CY5-conjugated secondary antibody. Bar in A is 100 µm.

View larger version (138K): [in this window] [in a new window]   Fig. 2.   Effects of 20-min acute hypertension in distribution of NHE3 in Sprague-Dawley rat proximal tubules. Immunofluorescence of NHE3 in control (A) and after 20 min of acute hypertension (E). B and F: high-resolution images of proximal tubules in A and E. C and G: immunofluorescence of adaptin from the same tubules. D and H: corresponding confocal autofluorescence image (excitation 488 nm, emission 522/35 nm). Shadow in D and H indicates the position of the brush border. White arrows point to the brush border. Traced arrows point to the base of brush border. NHE3 was visualized with Lissamine rhodamine-conjugated secondary antibody. Adaptin was visualized with indodicarbocyanine (CY5)-conjugated secondary antibody. TAL, thick ascending limb. Bars in C and G are 10 µm.

The distribution of NHE3 was determined from another six Sprague-Dawley rats that were subjected to induced acute hypertension for 20 min before the kidney was harvested. Increasing the peripheral resistance by ligating the abdominal aorta and mesenteric artery significantly (P < 0.05) increased the mean blood pressure from 96 ± 3 (n = 12) to 140 ± 5 mmHg (n = 6) in SD rats. After 20 min of induced acute hypertension, NHE3 was found to redistribute from the brush border into the base of microvilli (Fig. 2E) compared with the control. The redistribution was found in both convoluted and straight segments of proximal tubules. These observations were consistent with the results of the membrane fractionation study, which indicated that 30-40% of NHE3 was redistributed from the brush border into putative intracellular plasma membrane in acute hypertension (44). Figure 2F is the high-resolution image from a proximal tubule in Fig. 2E. Figure 2G is the immunofluorescence of adaptin from the same proximal tubule to indicate the position of the base of brush border, where the adaptin and NHE3 were colocalized. These observations were reproducible in all six rats tested.

To test whether other membrane-bound proteins in proximal tubule brush border were redistributed in the same pattern as NHE3 in acute hypertension, the localization of DPP IV in brush border was determined in control and pressurized kidneys of SD rats. DPP IV was found on the brush border as expected in all control kidneys (Fig. 3, A and B). DPP IV was also distributed evenly on the brush border of pressurized kidneys. No translocation of DPP IV was found in all pressurized kidneys (Fig. 3, D and E). This observation suggested that the redistribution of NHE3 detected by immunocytochemistry was unique to this protein.

View larger version (157K): [in this window] [in a new window]   Fig. 3.   Immunolocalization of dipeptidyl peptidase IV of rat kidney in control (A) and after 20 min of acute hypertension (D). B and E: high-resolution images of proximal tubules in A and D. C and F: autofluorescence images of the corresponding proximal tubules, which indicate the position of brush border (arrows). Immunofluorescence of dipeptidyl peptidase IV was visualized with a CY5-conjugated secondary antibody. Bars in A and D are 40 µm.

Since it is known that pressure natriuresis is blunted in SHR, we tested the hypothesis that the redistribution process of apical NHE induced by acute hypertension is impaired in SHR. Cryosections harvested from five adult SHR (12 wk old) without induced acute hypertension were investigated. Surprisingly, NHE3 was not evenly distributed on the brush border of proximal tubule but was found mostly close to the base of microvilli in adult SHR (Fig. 4, A and B). The same pattern of distribution was found in all SHR investigated. Procedures used to induce hypertension significantly increased the mean blood pressure of adult SHR from 134 ± 5 (n = 12) to 161 ± 6 mmHg (n = 7) ( P < 0.05). However, no redistribution of NHE3 was detected after 20 min of acute hypertension (data not shown). Seven adult SHR rats were used to confirm this observation.

View larger version (119K): [in this window] [in a new window]   Fig. 4.   Immunolocalization of NHE3 in proximal tubules of adult SHR without induced acute hypertension. A: immunofluorescence of NHE3. B: high-resolution image of proximal tubules in A. C: immunofluorescence of adaptin from the same tubules. D: corresponding autofluorescence image. Symbols (arrows) are the same as in Fig. 2. Bar in A is 40 µm.

To decide whether these observations are due to variations among strains or a consequence of the chronic increase of arterial pressure, the distributions of apical NHE3 in proximal tubules in young SHR (5-wk-old, prehypertensive) and 2K-1C GH rats with or without acute hypertension were next investigated. Cryosections from six young SHR were harvested without exposure to acute hypertension. Isoform NHE3 was found to localize on the brush border in young SHR as in SD (Fig. 5, A-D). Acute hypertension was then induced in five young SHR for 20 min before harvesting of the kidneys. Procedures to induce acute hypertension in young SHR significantly (P < 0.05) increased the mean arterial pressure from 104 ± 2 (n = 11) to 130 ± 4 mmHg (n = 5). The magnitude of the induced hypertension was similar in young (26 mmHg) and adult SHR (27 mmHg) despite the difference in the baseline. The acute increase of arterial pressure provoked a redistribution of NHE3 from brush border to the base of microvilli (Fig. 5, E-H) as seen in SD. The same observation was repeatedly seen in all five pressurized young SHR. These observations suggested that the distribution of NHE3 and the response to acute change of arterial pressure in young SHR are not different than in normal SD rats. The altered distribution of NHE3 observed in adult SHR could be the result of chronic increase of arterial pressure.

View larger version (144K): [in this window] [in a new window]   Fig. 5.   Effects of 20-min acute hypertension in distribution of NHE3 in 5-wk-old SHR proximal tubules. Immunofluorescence of NHE3 in control (A) and after 20 min of acute hypertension (E). B and F: high-resolution images of proximal tubules in A and D. C and G: immunofluorescence of adaptin. D and H: corresponding autofluorescence images. Arrows are the same as in Fig. 2. Bars in A and E are 40 µm.

After 2 wk of renal artery clipping, the GH rats became moderately hypertensive (mean arterial pressure 131 ± 6 mmHg, n = 6) as reported previously (41). The kidneys from three of the six GH rats were harvested without induced acute hypertension, whereas acute hypertension was induced for 20 min in the other three GH rats before their kidneys were harvested for cryosections. Significant hypertrophy of the nonclipped kidney compared with the clipped one was found in all six GH rats. The wet weights of the clipped and nonclipped kidney were 1.09 ± 0.06 and 1.49 ± 0.08 g (n = 6, P < 0.05), respectively. In sham-operated SD rats, NHE3 was found on the brush border as expected (data not shown), and the arterial pressure was normal (mean arterial pressure 98 ± 4 mmHg, n = 2). In the clipped kidneys that were not exposed to the hypertension because of the arterial clips, NHE3 was found on the brush border (Fig. 6, E-H). However, NHE3 was concentrated at the base of microvilli in all nonclipped kidneys (Fig. 6, A-D). The distribution of NHE3 in the nonclipped kidney was the same as observed in Sprague-Dawley rats subjected to acute hypertension. Procedures used to increase peripheral resistance significantly (P < 0.05) elevated the mean arterial pressure from 131 ± 6 (n = 6) to 168 ± 10 mmHg (n = 3) in GH. No further translocation of apical NHE was detected in either the nonclipped or clipped kidney (data not shown). These observations from adult SHR and GH rats suggested that the chronic increase of arterial pressure in experimental hypertension can trigger the redistribution of apical NHE in proximal tubules, regardless of whether the hypertension is genetic or renovascular in origin.

View larger version (139K): [in this window] [in a new window]   Fig. 6.   Immunolocalization of NHE3 in proximal tubules of nonclipped kidney (A-D) and clipped kidney (E-H) of Goldblatt hypertensive rat. Immunofluorescence of NHE3 in nonclipped kidney (A) and clipped kidney (E). B and F: high-resolution images of proximal tubules in A and E. C and G: immunofluorescence of adaptin. D and H: corresponding confocal autofluorescence images. Arrows are the same as in Fig. 2. Bars in C and G are 10 µm.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Subcellular membrane fractionation relies on density gradient and membrane markers to separate and identify different membrane populations. Only limited information about the spatial distribution of a particular marker (e.g., NHE3) within the intact tissue can be recovered. In the present study, we utilized the improved resolution of laser-scanning confocal fluorescence microscopy (29) and indirect immunofluorescence staining to detect the redistribution of NHE induced by changes in arterial pressure. The specificity of antibodies used, Ab1381, has been established in a previous study by immunoblotting membranes from NHE-deficient PS120 cells that had been transfected with either NHE1, NHE2, or NHE3 and from rat jejunal and kidney cortex brush border (22). Anti-NHE3 antibody Ab1381 recognized NHE3 as a 85-kDa protein in rat intestinal brush-border membranes. In the present study, preincubating the antibodies with the fusion protein used to raise the antibodies diminished the signal in immunostaining. The blocking effect was due to the NHE fusion protein, but not the glutathione-S-transferase, because preincubation of the antibodies with glutathione-S-transferase alone did not diminish the immunofluorescence signal.

NHE3 was detected in the proximal tubule brush border in SD rats as reported by others (2, 7). That thick ascending limb had stronger immunofluorescence of NHE3 than proximal tubules was previously observed in rat kidneys (2). Similar observations were also found in the present study. Although other NHE isoforms have been suggested to be present on the brush border, NHE3 is the primary NHE isoform responsible for the apical Na⁺/H⁺ exchange activities (40). Twenty minutes of induced acute hypertension provoked the redistribution of NHE3 from brush border into the base of microvilli as a ring behind the brush border in proximal tubules, which might be equivalent to the redistribution of NHE3 from brush-border vesicles to heavier membrane vesicles observed in the subcellular membrane fractionation study (44). Using a nonobstructive optical method to measure proximal tubular flow rate, tubular fluid reabsorption was still inhibited after 30 min of induced hypertension (11). The redistribution of NHE3 found after 20 min of acute hypertension was consistent with the time course of changes in proximal tubular flow.

Double labeling was employed to colocalize clathrin-coated vesicles and NHE3. Clathrin-coated pits were used as markers of intermicrovillar clefts in brush border to indicate the relative translocation of NHE3 (5, 9). The adaptin antibody used in this study can recognize clathrin-coated membrane at plasma membrane and in the Golgi regions (1). The NHE3 seemed to be redistributed from the brush border to the same intracellular region that contains clathrin-coated pits in the high-resolution images. However, it was unlikely that NHE3 was internalized into clathrin-coated pits. Beimesderfer et al. (7) have recently demonstrated that NHE3 was only found in a population of endosomes in the subapical region of proximal tubules but not in clathrin-coated pits.

Immunolocalization of DPP IV in proximal tubules of control and pressurized kidneys showed that there was no detectable redistribution of this brush-border protein. Redistribution of DPP IV was found in fractionated membrane but was less conspicuous than that of NHE3 (44). Confocal imaging might not have sufficient spatial resolution to detect the translocation of DPP IV. These observations suggested that spatial redistribution of NHE3 induced by acute hypertension was unique to this protein and that there were different intracellular stores for NHE3 and DPP IV.

To test whether there is NHE redistribution when SHR are challenged by acute hypertension, the localization of NHE3 in proximal tubules with or without induced acute hypertension was compared in adult SHR. NHE3 was found to concentrate on the base of microvilli even before the challenge of acute hypertension. No further redistribution of NHE3 could be detected in the pressurized kidneys. The finding that most NHE3 immunofluorescence signal was found on the base of microvilli in SHR raised the issue of whether the localization of NHE3 in brush border is genetically predisposed or is the consequence of chronic increase of arterial pressure in spontaneous hypertension. The observations that NHE3 was distributed uniformly on the brush border of young SHR (5-wk-old, prehypertensive) and that acute hypertension could trigger the redistribution of NHE3 in young SHR similar to that of SD rats support the latter hypothesis. The difference in the localization of NHE3 in 5-wk and 12-wk SHR predicts a difference in sodium handling between these two age groups. This is consistent with the observation that sodium retention occurs in young SHR but not adult SHR (4). Apical NHE3 activity in isolated proximal tubules and brush-border vesicles of young SHR is known to be enhanced compared with that of age-matched Wistar-Kyoto rats (WKY) (14, 19, 23). No literature is available to compare the apical NHE3 activity of SHR at different ages. There is also no consensus of what causes the enhanced NHE3 activity of proximal tubules in young SHR. NHE3 expression was found to increase in 8-wk-old SHR compared with age-matched WKY (23). However, no difference in the amount of NHE3 mRNA was found in 4-wk- and 7-wk-old SHR compared with the age-matched WKY (19). Interestingly, the difference of Na⁺/H⁺ exchange activity between SHR and WKY was diminished after 10 wk of age (14). It is consistent with our hypothesis that chronic increase of arterial pressure in SHR triggers apical NHE3 internalization. NHE3 internalization might reduce/normalize the Na⁺/H⁺ exchange activity in SHR with respect to age-matched WKY.

To test the hypothesis that chronic increase of arterial pressure can provoke the NHE redistribution, the distribution of NHE3 was determined in 2K-1C GH rats. The difference in NHE distributions between the nonclipped and clipped kidneys is consistent with the hypothesis that a chronic increase of arterial pressure triggers the redistribution, because the clipped kidneys are not exposed to the increased arterial pressure in this model of hypertension. The different distribution of apical NHE3 in clipped and nonclipped kidneys is also consistent with the observation that the clipped kidney tends to reabsorb more Na⁺ (27). Although it is not known whether acute and chronic hypertension might share the same signaling pathway to initiate the redistribution process, the end points of the redistribution processes are likely the same. The observations to support this conclusion were that acute hypertension did not induce further translocation of NHE3 in adult SHR and GH rats and that NHE3 was redistributed to the same region where clathrin-coated pits were localized in chronic and acute hypertension.

The proximal tubule has been known as a locus of pressure natriuresis for years (10, 25), but the mechanisms of inhibition of sodium transport have not been fully identified. One theory is that papillary blood flow is not autoregulated as efficiently as cortical blood flow as the renal perfusion is increased (12). Increased papillary blood flow following elevation of renal perfusion pressure is posited to elevate hydrostatic pressure within the vasa recta capillaries and renal interstitium, and the latter is supposed to inhibit tubular reabsorption (24). It remains controversial, however, that autoregulation of renal medullar blood flow is less efficient than cortical blood flow (13, 28), because the studies purporting to show lack of medullary autoregulation were performed in volume-expanded rats (34). Volume expansion inhibits autoregulation in all parts of the renal circulation (3). Nor has a mechanism been established by which increased renal interstitial pressure could inhibit tubular reabsorption (24, 25). Direct increase of renal interstitial pressure by intrarenal volume expansion suggested that the inhibition of reabsorption might be due to an increase of paracellular flux in proximal tubules (18, 26). Recently, Zhang et al. (44) reported that proximal tubule natriuresis provoked by acute hypertension was associated with the internalization of apical NHE and inhibition of the activity of basolateral Na⁺ pumps, which is the first report to indicate that transcellular sodium transport is also inhibited during pressure natriuresis.

The signaling pathway through which the increased renal perfusion pressure triggers the redistribution of NHE3 is not yet known. It is unlikely that a local increase in hydraulic pressure is the triggering signal, because the kidney begins to autoregulate quickly (43). Single-nephron blood flow is known to autoregulate well when renal perfusion pressure is acutely changed by increasing peripheral resistance (42). There is also no difference in the mean proximal tubular pressure measured from normal SD rats, SHR, and GH rats (41). Change in renal interstitial hydrostatic pressure is also unlikely to be the signal, because the apical surface is directly exposed to tubular pressure. It was well established that proximal tubular pressure oscillates spontaneously in normotensive rats and fluctuates in hypertensive rats with a larger amplitude than the variations of renal interstitial hydrostatic pressure induced by changing renal perfusion pressure (16, 18, 41). A number of hormones have been shown to modulate proximal Na⁺ reabsorption through their effects on the luminal Na⁺/H⁺ exchanger; these include parathyroid hormone (PTH), angiotensin II, and dopamine (15, 20). The only agent known to induce translocation and covalent modification of NHE in proximal tubules is PTH (20). NHE is first internalized from apical membrane and then deactivated (20). Although PTH is not related to pressure natriuresis directly, PTH does inhibit Na⁺ pump in proximal tubule through a cytochrome P-450 pathway (32). The identification of the signaling pathway for NHE redistribution induced by acute hypertension might shed light on the cellular mechanism of pressure natriuresis in proximal tubules.

In summary, the present study detected the redistribution of apical NHE3 from the proximal tubule brush border to the base of microvilli when blood pressure was acutely increased. Similar redistribution was also detected in young SHR. In adult SHR and the nonclipped kidney of GH rats, the NHE3 was found on the base of the microvilli, and induced acute hypertension did not provoke further translocation. We conclude that redistribution of apical NHE3 was sensitive to the arterial blood pressure.