We have previously demonstrated that mouse proximal tubules in vitro respond to changes in luminal flow with proportional changes in Na+ absorption (Du Z, Duan Y, Yan Q, Weinstein AM, Weinbaum S, and Wang T. Proc Natl Acad Sci USA 101: 13068–13073, 2004). It was hypothesized that brush-border microvilli function as a sensor to detect and amplify luminal hydrodynamic forces and transmit them to the actin cytoskeleton. In the present study we examine whether 1) flow-dependent HCO3− transport is proportional to flow-dependent variations in microvillous torque (bending moment); 2) both luminal membrane Na+/H+ exchange (NHE3) and H+-ATPase activity are modulated by axial flow; and 3) paracellular permeabilities contribute to the flux perturbations. HCO3− absorption is examined by microperfusion of mouse S2 proximal tubules in vitro, with varying perfusion rates, and in the presence of the Na/H-exchange inhibitor EIPA, the H+-ATPase inhibitor bafilomycin, and the actin cytoskeleton inhibitor cytochalasin D. Paracellular permeability changes are assessed with measurements of epithelial HCO3− permeability and transepithelial potential difference (PD). It is found that 1) an increase in perfusion rate enhances HCO3− absorption and microvillous torque, and the fractional changes of each are nearly identical; 2) inhibition of NHE3 by EIPA, or H+-ATPase by bafilomycin, produced only partial inhibition of flow-stimulated bicarbonate transport; 3) disruption of the actin cytoskeleton by cytochalasin D blocked the increment of HCO3− absorption by high flow; and 4) HCO3− permeability and transepithelial PD are not modulated by flow. We conclude that flow-dependent modulation of proximal tubule HCO3− reabsorption is due to changes in both NHE3 and H+-ATPase activity within the luminal cell membrane and this requires an intact actin cytoskeleton. Paracellular permeability changes do not contribute to this flow dependence. Perfusion-absorption balance in the proximal tubule is a direct effect of flow-induced torque on brush-border microvilli to regulate luminal cell membrane transporter activity.
- kidney proximal tubule
- glomerulotubular balance
- flow-dependent transport
- sodium-hydrogen exchange
na+/h+ exchange and h+-atpase are the major apical H+ extrusion mechanisms mediating HCO3− reabsorption in proximal tubules. A numbers of investigators have studied flow-dependent HCO3− transport in proximal tubules of rat kidney using in vivo microperfusion and demonstrated that HCO3− absorption is regulated by variations in tubular flow (2, 8, 25). The underlying mechanism for flow-dependent changes in HCO3− reabsorption was studied by measuring cellular pH recovery after an ammonium pulse in perfused proximal tubules of rat kidney in vivo (29). With increases in luminal flow rate, the pH recovery mediated by Na+/H+ exchange was enhanced, suggesting that these increases in axial flow velocity recruit new transporters into the luminal membrane (29). However, whether the apical H+ extrusion mechanisms are regulated by axial flow in mouse proximal tubules in vitro has not been examined. In part, the present work is motivated by the prediction of a detailed model of acid/base transport across the brush-border microvilli, that unstirred layer effects would have only trivial impact on NHE function (22). We have previously investigated Na+ transport activity in response to changing axial flow in mouse proximal tubules in vitro using both experimental and mathematical models (11). It was found that theoretically predicted variations in microvillous torque (bending moment) produced nearly identical fractional changes in Na+ reabsorption, provided flow-induced changes in tubule diameter were taken into account (11). In addition, flow-dependent Na+ reabsorption is diminished in NHE3 knockout mice and abolished by nontoxic disruption of the actin cytoskeleton (11). These data supported our hypothesis that the brush-border microvilli serve a mechanosensory function in which fluid dynamic torque is transmitted to the actin cytoskeleton and modulates Na+ absorption in kidney proximal tubules as first proposed in Guo et al. (17).
In this study, we pursue the investigation of flow-dependent transport in vitro by measuring both Na+ and HCO3− absorption, calculating the microvillous torque under different perfusion rates in isolated mouse proximal tubules. The mechanisms underling the altered HCO3− transport are defined by examining the effect of the Na+/H+ exchange inhibitor EIPA, the H+-ATPase inhibitor bafilomycin, and the actin cytoskeleton inhibitor cytochalasin D on flow-dependent Na+ and HCO3− transport. Paracellular permeabilities are assessed by measuring the HCO3− permeability and by measuring the transepithelial electrical potential. We also examine the interaction of flow-dependent effects with the impact of peritubular protein to modulate Na+ and HCO3− absorption. The experimental data summarized below further support the hypothesis that in kidney proximal tubules, brush-border microvilli function as a flow sensor in regulating luminal cell membrane activity of a number of transporters.
MATERIALS AND METHODS
EIPA, bafilomycin, and cytochalasin D were purchased from Sigma (St. Louis, MO). Control and experimental groups were studied under identical experimental conditions. The experiments were conducted under animal protocol 2004–10522 approved by the Institutional Animal Care and Use Committee.
Proximal convoluted tubules (S2 segments) were perfused in vitro using conventional methods (12). Briefly, kidneys were removed and cut into coronal slices from either C57/B6J mice purchased from Jackson Laboratory or NHE3 knockout mice obtained from Dr. Gary Shull's laboratory at the University of Cincinnati (34). The mice were anesthetized by an intraperitoneal injection of 50 mg/kg body wt of pentobarbital sodium. The ages of animals were matched in all experimental groups. Individual tubules were dissected in cooled (4°C) Hanks' solution containing (in mM) 137 NaCl, 5 KCl, 0.8 MgSO4, 0.33 Na2HPO4, 1 MgCl2, 10 Tris, 0.25 CaCl2, 2 glutamine, and 2 l-lactic acid. Proximal tubules (S2) were perfused with an ultrafiltrate-like solution at the perfusion rate of 5, 10, 15, 20, and 25 nl/min, respectively. The perfusion rates were adjusted by changing the gravity of the reservoir connected to the perfusion pipette and measured by constant-bore glass capillary tubes. The solution for luminal perfusion contained (in mM) 125 NaCl, 22 NaHCO3, 1 CaCl2, 1.2 MgSO4, 2 glutamine, 2 lactic acid, 10.5 glucose, 5 KCl, and 1.2 phosphoric acid. The bath medium consisted of (in mM) 101 NaCl, 22 NaHCO3, 1 CaCl2, 1.2 MgSO4, 2 glutamine, 2 lactic acid, 10.5 glucose, 5 KCl, 1.2 phosphoric acid, and 32.5 HEPES as well as 5 g/dl albumin. All solutions were bubbled with 95% O2-5% CO2 and had a pH of 7.4. The osmolalities of the bath and perfusate were adjusted to 300 mosmol/kgH2O by the addition of either H2O or NaCl. The extensively dialyzed [methoxy-3H]inulin was added to the perfusate at a concentration of 30 μCi/ml as a volume marker. Proximal tubules (S2) isolated from mouse kidney were perfused at 37–38°C in a 1.2-ml temperature-controlled chamber. Bath fluid was continuously changed at a rate of 0.5 ml/min to maintain the constancy of pH and bath osmolality during the experiment (12). The first period of collection began after an equilibration time of 30–60 min. For each experimental period, four timed collections of tubular fluid were made.
The volume of the perfusate and collected samples was measured in a fixed volume collection pipette, and [3H]inulin concentrations in those samples were determined in a liquid scintillation counter (LS5801, Beckman). The rate of net fluid reabsorption (Jv) was calculated according to the [3H]inulin concentration changes between the original and collected fluid. The HCO3− concentration in the perfusate and collected tubular fluid was measured using the microcalorimetric method (Picapnotherm) as described previously (36). Samples were stored under oil, and the volumes obtained in 15-nl aliquots were compared with NaHCO3− standards. The rate of HCO3− absorption (JHCO3) was calculated according to the HCO3− concentration changes between the original and collected fluid. The Jv and JHCO3 are expressed per minute per millimeter of proximal tubule. Tubular inner diameters were measured from the center of the tubule under the different perfusion rates.
HCO3− permeability was measured by the methods described (1, 10, 27). Briefly, proximal tubules were perfused with HCO3−-free solution in the lumen and 22 mM HCO3− solution in the bath. Acetazolamide (10−4 M) was added to both lumen and bath solutions. Net HCO3− transport was evaluated from changes in tracer inulin concentrations and total CO2 concentrations measured by microcalorimetry. The permeability of HCO3− in proximal tubules was calculated according to the following equation (1) where PHCO3 is HCO3− permeability, JHCO3 is passive HCO3− flux from bath to lumen, Cp is the measured HCO3− concentration in the bath, and CL represents the bicarbonate concentration in the collected fluid.
The transepithelial potential differences (PDs) were measured with a microelectrode inserted into a port connected to the luminal solution through an agar bridge. The PDs were measured at low (5.25 ± 0.25 nl/min) and high (18.50 ± 0.96 nl/min) perfusion rates.
Because increases in the tubule flow rate also induce significant changes in tubule diameter, and hence the hydrodynamic drag forces acting on the microvilli, a theoretical model was developed by Du et al. (11) to relate changes in microvilli torque, T, to changes in flow rate, Q, and tubule radius, R. This relationship is given by the following equation (2) where R is the inner tubule radius with brush border, r is the index to the reference value, L is the length of the microvilli (L =2.5 μm), δ is the thickness of the hydrodynamic interaction layer of the microvilli tip (δ ≈ 150 nm), μ is fluid viscosity, and Q is the flow rate in the tubule.
In applying Eq. 2, we calculate the ratio of the torque, T, at any flow rate, Q, to a reference torque, Tr, at a reference flow rate, Qr. By using this ratio, one does not need to count the number of microvilli per unit tubule length because this unknown geometric parameter cancels out of the equation.
Data are presented as means ± SE. Student's t-test was used to compare control and experimental groups. The difference between the mean values of an experimental group and a control group was considered significant at P < 0.05.
We consider, first, whether changes in tubular flow rate alter net HCO3− absorption in vitro. JHCO3 was measured in isolated mouse proximal tubules at perfusion rates of 5, 10, 15, 20, and 25 nl/min, respectively. Figure 1A displays the rate of HCO3− absorption in response to the changing perfusion rate. HCO3− absorption rates are proportionally increased with increased perfusion rates (R2 = 0.99), and overall JHCO3 was doubled when the perfusion rate went from 5 to 25 nl/min. This is the first demonstration that HCO3− absorption is regulated by axial flow velocity in mouse proximal tubules. It is important to note that a fivefold increase in flow rate produces only a twofold increase in JHCO3. This nonidentical scaling can be explained using Eq. 2 if one also considers the changes in tubule diameter with flow rate, which are shown in Fig. 1B. Such diameter changes would reduce the microvillus bending moment in response to flow because they reduce the fluid shear stress at the tips of the microvilli. We then asked whether the changes in luminal flow rate altered HCO3− permeability, which may impact on HCO3− absorption. HCO3− permeabilities were measured at a high and low luminal perfusion rate. We found no significant changes in HCO3− permeability by increasing the flow rate. Figure 1C shows that the HCO3− permeabilities were 0.99 ± 0.08 and 1.29 ± 0.28 × 10−5 cm2/s (n = 8, P > 0.05) at a low and high flow rate, respectively. These results indicate that flow-altered HCO3− transport is not due to passive paracellular permeability changes. To further examine the paracellular contribution to flow-dependent HCO3− transport, we measured the transepithelial PD. No significant changes were found in PD between low and high flow rates. The PD was −2.06 ± 0.25 and −2.45 ± 0.22 mV at low (5.25 ± 0.25 nl/min) and high (18.50 ± 0.96 nl/min) perfusion rates, respectively (n = 6, P > 0.05).
It must be acknowledged that any increase in luminal flow must result in an increase in luminal HCO3−, which, by itself, would tend to enhance proximal HCO3− absorption. To test this possibility, we compared the changes in luminal HCO3− concentration and HCO3− transport as a function of tubular flow. Data are summarized in Fig. 2 showing the changes in HCO3− concentration and JHCO3 at perfusion rates of 10, 15, 20, and 25 nl/min compared with the reference flow rate of 5 nl/min. The changes in HCO3− concentration in the collected fluid are very small compared with the rate changes of HCO3− absorption. In particular, when the flow rate is higher than 15 nl/min, the changes in luminal HCO3− concentration are the same (5 mM), but the significant increment of transport still exists. In short, changes in luminal HCO3− concentration due to flow cannot account for the flow dependence of HCO3− absorption.
We next assessed the correlation of microvillus torque with HCO3− absorption. Figure 3A shows the effects of flow rate on the total microvillous torque, and Fig. 3B shows the relationship between changes in torque and changes in HCO3− absorption. In both panels, the reference values used for torque, Tr, and HCO3− reabsorption (JHCO3)r were both measured at the perfusion rate of 5 nl/min. Our results in Fig. 3A show that the changes in torque calculated using Eq. 2, while proportional to the variation of flow rate, do not scale directly with the flow rate (5-fold increase in flow produces only a 2.2-fold increase in torque) because of tubule diameter changes. In contrast, when these diameter changes are taken into account in Eq. 2, the fractional changes in torque and HCO3− transport are nearly identical. These results support the hypothesis that the alteration of torque on brush-border microvilli by flow is the signal regulating HCO3− transport in the proximal tubule when the filtration rate is changed.
Because NHE3 is the major transporter responsible for reabsorption of filtered Na+ and HCO3− in the proximal tubule (37, 38), we examined flow-dependent Na+ and HCO3− absorption in NHE3 knockout mice. In addition to the reduction in baseline Jv at a low flow rate, the increments in flow-stimulated Na+ and HCO3− absorption were reduced 57 (0.34 vs. 0.80 nl·min−1·mm−1) and 35% (45.1 vs. 69.2 pmol·min−1·mm−1), respectively, in NHE3-null mice compared with wild-type (Tables 1 and 2; Fig. 4). The mechanism of this stimulation of JHCO3 by flow in the NHE3-null mouse is likely due to altering H-ATPase activity when flow rate is increased. To test this hypothesis, we perfused tubules from wild-type mice and examined the effects of EIPA, bafilomycin, and EIPA plus bafilomycin on flow-dependent HCO3− transport. Data are summarized in Tables 1 and 2 and Fig. 5. JHCO3 increased 69.2 pmol·min−1·mm−1 (from 68.8 to 137.9 pmol·min−1·mm−1) when flow rate increased from 4 to 20 nl/min under control conditions. Using similar perfusion rates, EIPA inhibited 40% (ΔJHCO3 = 41.7, P > 0.05), and bafilomycin inhibited 43% (ΔJHCO3 = 39.2, P < 0.05) of the flow-stimulated JHCO3 increase. The increment of JHCO3 due to enhanced flow rate was reduced by 97% (ΔJHCO3 = 1.99, P < 0.0001) when both EIPA and bafilomycin were added to the luminal perfusate, indicating that both NHE3 and H-ATPase activity are regulated by tubular flow.
Previously, we found that the flow-dependent Na+ transport was blunted substantially by an actin cytoskeleton inhibitor, cytochalasin D, when applied to the lumen (11). In the present work, we examined HCO3− absorption in the absence and presence of 3 μM cytochalasin D in the luminal perfusate. Cytochalasin D did not change the baseline JHCO3 at a low flow rate but significantly reduced the flow-stimulated increase in JHCO3. These results indicated that 3 μM cytochalasin D was not toxic to cells but inhibits actin cytoskeletal signaling, which is critical to the flow-stimulated HCO3− transport in the proximal tubule.
It has been suggested that the mechanisms for glomerulotubular balance include both peritubular capillary effects and luminal factors. With respect to peritubular factors, any increase in filtration fraction must result in an increased peritubular protein concentration, and increased peritubular capillary protein concentration enhances proximal reabsorption. We examined whether changes in protein concentration in the bath alter flow-dependent Na+ and HCO3− transport. As shown in Tables 3 and 4, and in Fig. 6, increasing albumin concentration from 2.5 to 5 g/dl in the bath solutions enhanced baseline Jv and JHCO3 but did not influence transport activity in response to a change in flow rate. The curves of change in Jv and in JHCO3 by flow are parallel, indicating that the changes induced by basolateral protein concentration are additive to the sensory function of the microvilli.
The experiments discussed in this paper used an in vitro preparation to assess the determinants of glomerulotubular balance, namely, modulation of overall Na+ reabsorption in proportion to the glomerular filtration rate (14). The mechanisms underlying balanced tubular reabsorption include both peritubular capillary effects (13) and luminal factors (40). In the rat kidney in vivo, there is a direct correlation between the single-nephron filtration fraction and proximal tubule Na+ reabsorption (24, 33). Because a lower filtration fraction reduces oncotic pressure within peritubular capillaries, Lewy and Windhager (24) surmised that this would lead to reduced capillary uptake of fluid from the renal interstitium and lateral intercellular space, and, hence, elevated interspace pressure. This, in turn, would result in a backflux of Na+ already transported into the lateral interspace, returning across the tight junction into the lumen (41). The impact of peritubular protein concentration on proximal Na+ reabsorption has been documented by a number of workers, perhaps most directly in measurements of proximal Na+ reabsorption during simultaneous perfusion of peritubular capillaries (9, 16). The mechanism underlying this protein effect is less well understood. While evidence supports renal interstitial pressure as a mediator, peritubular protein is also capable of modulating Na+ reabsorption by rabbit proximal tubules in vitro, where there is no elevation in peritubular hydrostatic pressure (20). In particular, paracellular permeability changes have been absent in rabbit tubules perfused in vitro (4, 6, 23). A specific effect of peritubular protein on cellular transport pathways has not been identified.
Mathematical modeling of a proximal nephrovascular unit (glomerulus plus peritubular capillary plus proximal tubule) has indicated that a prerequisite for robust glomerulotubular balance under a variety of conditions is that both peritubular protein (filtration fraction) and luminal fluid flow must modulate proximal Na+ reabsorption and that these effects must be additive (39). The assertion derives from the fact that glomerular filtration is the product of the filtration fraction and glomerular plasma flow; either can vary without any change in the other. The impact of the luminal flow rate has been termed “perfusion-absorption balance” (40) and has been demonstrated in rat microperfusion studies (3, 18, 28, 31). In addition to the effect on Na+ and volume reabsorption, luminal flow has been found to influence the transport of glucose (21), HCO3− (2, 8, 25), and Cl− (15, 42). One of the best illustrations of this phenomenon is the in vivo microperfusion data of Chan et al. (8), in which a threefold increase in luminal perfusion rate (with trivial changes in luminal HCO3 concentration) produced a doubling of the rate of HCO3− reabsorption. In the rat, axial flow has been shown to modulate the density of the luminal membrane Na+/H+ exchanger (NHE3). Preisig (29) loaded rat proximal tubule cells with the pH indicator BCECF and examined recovery from an acute acid load in vivo (ammonium pulse). With increases in luminal flow rate, the pH recovery mediated by Na+/H+ exchange was enhanced. Maddox et al. (26) subjected rats to acute changes in vascular volume to obtain hydropenic, euvolemic, and volume-expanded groups, with respective grouping according to a decreased, normal, and increased glomerular filtration rate. When brush-border membrane vesicles were prepared from each of these groups, and Na+/H+ kinetic parameters were assessed, it was found that the Vmax determinations stratified in parallel with the glomerular filtration rate.
An in vitro model suitable for study of perfusion-absorption balance has become available only recently. Early in the development of microperfusion of rabbit tubules, Burg and Orloff (7) found no significant effect of axial flow on volume reabsorption, and that observation remained unchallenged. It was only when we examined the effect of axial flow on mouse proximal tubules that flow-dependent volume reabsorption was demonstrated (11). In that study, it was noted that fractional changes in flow were more than double the fractional changes in volume reabsorption. However, when the variation in luminal diameter at different perfusion rates was taken into account, and the estimated change in microvillous torque was compared with volume reabsorption, the proportions were nearly identical. A formula for microvillous torque was derived (Eq. 2), which displayed a direct dependence on axial flow rate and fluid viscosity, along with an inverse dependence on tubule cross-sectional area. Du et al. (11) also demonstrated that increasing tubule fluid viscosity by the addition of dextran-80 to increase microvillous torque led to an increase in Na+ reabsorption, without any change in luminal perfusion rate. In addition, the data of Burg and Orloff (7) included measurements of luminal diameter at each flow rate, and when these were reconsidered (11), it was apparent that stretch in the rabbit tubules had blunted the impact of increased flow on microvillous torque. In the study of Du et al. (11), however, individual solute fluxes underlying the change in volume reabsorption were not identified. In view of its importance in overall Na+ transport, it was speculated that axial flow increased transport via NHE3. Consistent with this view, the effect of flow on volume transport in tubules from NHE3-null mice was diminished. Nevertheless, these tubules still displayed a flow dependence on volume reabsorption (∼50% that of control tubules), and this suggested a broader impact than just NHE3 density.
The present work begins the dissection of the components of flow-dependent proximal reabsorption. What was shown is that the impact of luminal flow on HCO3− reabsorption is comparable to its effect on volume reabsorption, that the change in reabsorptive flux is not a consequence of altered luminal HCO3− concentration, and that this is blunted by a nontoxic dose of cytochalasin. As with the effect on overall Na+ reabsorption, flow-dependent HCO3− reabsorption scales nearly identically with microvillous torque. Our theoretical model for the torque (Eq. 2) shows that the leading term is proportional to flow, Q, and inversely proportional to tubule diameter and that this explains why a fivefold increase in flow in Fig. 1A produces only a 2.2-fold increase in HCO3− reabsorption. With respect to the components of HCO3− reabsorption, it was found that total HCO3− flux was inhibited by about one-third with luminal EIPA, and the effect of bafilomycin was similar. Together, the two inhibitors abolished ∼80% of HCO3− reabsorption, suggesting some compensatory HCO3− flux when only a single inhibitor was used. The potency of each inhibitor is comparable to prior observations from in vivo microperfusion in the mouse (38). What is new here is that both components of proximal proton secretion appear to be flow dependent, and there seems to be little difference in the relative magnitude of the flow effect on each transporter. Although the absolute value of HCO3− reabsorption is reduced in NHE3-null mice, the relative effect of flow on this flux is identical to that in control tubules. Although Cl− reabsorption was not measured in these experiments, an indirect argument suggests that its flux is also flow dependent. When luminal perfusion was increased from 4 to 20 nl/min (Table 1), the increment of volume reabsorption was 0.80 nl·min−1·mm−1, and this corresponds to an increase in Na+ reabsorption by 118 pmol·min−1·mm−1. Under comparable conditions, the HCO3− flux increased by 69 pmol·min−1·mm−1, suggesting a flow-dependent increase in Cl− flux of 49 pmol·min−1·mm−1.
In this study, there was no effect of luminal flow on overall epithelial HCO3− permeability or on transepithelial PD, and thus no reason to suspect an effect of axial flow on paracellular properties. The HCO3− permeability measurement, 1.0–1.3 × 10−5 cm2/s, is the first determination of this parameter for mouse proximal tubules. This value is about half the value that Holmberg et al. (19) found for rabbit proximal tubule HCO3− permeability (2.2 × 10−5 cm2/s), and those workers also saw no effect of perfusion rate on permeability. In that study, proximal tubule Cl− permeability was ∼2.5 times the HCO3− permeability. The findings may also be compared with those in rat proximal tubules, in which HCO3− permeability is 6.7 × 10−5 cm2/s, and with Cl− permeability about threefold greater (35). A lower value for rat proximal tubule HCO3 permeability, 1.8 × 10−5 cm2/s, was obtained by Chan et al. (9), and this permeability was also independent of luminal perfusion rate. In the rat, proximal tubule Cl− permeability has been examined, both as a function of luminal flow and of luminal diameter, and over a broad range, the impact of either variable is minor (30). With regard to the present work, if one estimates mouse proximal tubule Cl− permeability as 2.5 times its HCO3− permeability, and assumes an ambient Cl− concentration of 100 mM, then for a tubule with a diameter of 16 μm, a 1-mV perturbation of the transepithelial PD will change paracellular Cl− flux by ∼3 pmol·min−1·mm−1. This is an inconsequential flux compared with the flow-dependent increase in Na+ reabsorption, or even the estimated change in Cl− reabsorption. Thus it is unlikely that flow-dependent transepithelial PD changes modulate Cl− reabsorption. It must be acknowledged that none of the mouse tubule reflection coefficients are known. For the rat in vivo, there is general uniformity in the finding of a low value, σCl ≅ 0.5 (35), although for the rabbit in vitro, there is controversy, with a range of estimates from 0.75 to 1.0 (32). With respect to flow-dependent Cl− reabsorption in the mouse, a reflection coefficient of 0.75 and a flow-dependent Jv increment of 0.80 nl·min−1·mm−1 would predict an increase in convective flux of Cl− of ∼25 pmol·min−1·mm−1, again significantly smaller than the estimated value. In sum, it seems likely that there is a flow-dependent increase in transcellular Cl− reabsorption in the mouse, but the arguments are indirect.
In this study, increasing peritubular protein from 2.5 to 5.0 g/dl increased volume reabsorption by 50% from 0.8 to 1.2 nl·min−1·mm−1 (at a luminal perfusion of ∼15 nl/min). This is comparable to the finding in rabbit tubules perfused in vitro, in which Imai and Kokko (20) increased bath protein from 6.4 to 12.5 g/dl and observed a 40% increase in Jv from 1.0 to 1.4 nl·min−1·mm−1. A similar effect of peritubular protein has also been observed in the rat, in which both tubules and capillaries were perfused. Green et al. (16) increased peritubular protein from 2.5 to 10 g/dl and noted a 70% increase in Jv from 2.0 to 3.4 nl·min−1·mm−1. Chan et al. (9) increased peritubular protein from 8 to 15 g/dl and observed a 32% increase in Jv fro m 2.9 to 3.8 nl·min−1·mm−1. In the present work, the increase in peritubular protein also increased HCO3− reabsorption by ∼30 pmol·min−1·mm−1, regardless of the luminal perfusion rate. Because of the flow dependence of HCO3− reabsorption, this absolute increase in JHCO3 corresponded to an 80% change at the lowest perfusion rate and a 30% change at the highest flow. In perfusing rabbit proximal convoluted tubules, Berry and Cogan (5) removed protein from the peritubular bath and measured the change in reabsorption of volume, HCO3−, and glucose. They observed a 37% decrease in volume reabsorption, from 1.4 to 0.9 nl·min−1·mm−1, but no change in fluxes of HCO3− or glucose. They concluded that the impact of peritubular protein was restricted to NaCl flux. In a subsequent rat study by Chan et al. (9), the findings were different. They found that superfusion of peritubular capillaries with protein-free Ringer decreased JHCO3 by 14%, from 148 to 127 pmol·min−1·mm−1; increasing peritubular protein from 8 to 15 g/dl increased JHCO3 by 16%, from 140 to 163 pmol·min−1·mm−1. Our results are more compatible with the findings of Chan et al., but there is no apparent explanation for the difference from the experiments in rabbit tubules. It should be noted that in these experiments, the protein-induced increase in Jv corresponds to a Na+ flux of 59 pmol·min−1·mm−1. The fact that the increment in HCO3− flux is about half this value suggests that protein has also increased Cl− reabsorption.
Perhaps the most important aspect of this study is the demonstration that mouse proximal tubules in vitro can be used to study both luminal and peritubular factors that mediate glomerulotubular balance. The specific findings are that luminal flow modulates reabsorptive HCO3− flux in proportion to the estimated microvillous torque, and this has an impact on both NHE3 and the H+-ATPase in a proportional manner. Given the magnitude of the flow-dependent change in overall Na+ reabsorption, it is also likely that flow impacts transcellular Cl− flux as well. We have also documented the effect of peritubular protein to increase proximal HCO3− reabsorption, and this is an unambiguous cellular effect of peritubular protein. This observation derives its significance from the focus on the tight junction, embodied in the classic backflux hypothesis. The present study is incomplete in a variety of aspects: the spectrum of luminal transporters that are modulated by flow remains to be identified; beyond the implication of the actin cytoskeleton, the cellular mediators of the flow effect are unknown; the luminal transporters that are modulated by peritubular protein remain to be identified; and the cellular mediators of the peritubular effect are also unknown. From the data shown here, it appears that the luminal and peritubular effects are additive (Fig. 6). If this observation is robust, it will need to be understood in relation to the convergence of luminal and peritubular signaling systems. Finally, as the signal strength of luminal and peritubular effects is better defined, it will be important to examine both effects in a model nephrovascular unit to see how their superimposition yields glomerulotubular balance.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants RO1-DK-62289–01 (to T. Wang and S. Weinbaum) and RO1-DK-29857 (to A. M. Weinstein).
We thank Dr. Gary Shull (University of Cincinnati) for providing NHE3 mutant mice.
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 © 2006 the American Physiological Society