H + transport in urinary epithelia

Vacuolar-type proton pumps in insect epithelia

Active transepithelial cation transport in insects was initially discovered in Malpighian tubules, and was subsequently also found in other epithelia such as salivary glands, labial glands, midgut and sensory sensilla. Today it appears to be established that the cation pump is a two-component system of a H(+)-transporting V-ATPase and a cation/nH(+) antiporter. After tracing the discovery of the V-ATPase as the energizer of K(+)/nH(+) antiport in the larval midgut of the tobacco hornworm Manduca sexta we show that research on the tobacco hornworm V-ATPase delivered important findings that emerged to be of general significance for our knowledge of V-ATPases, which are ubiquitous and highly conserved proton pumps. We then discuss the V-ATPase in Malpighian tubules of the fruitfly Drosophila melanogaster where the potential of post-genomic biology has been impressively illustrated. Finally we review an integrated physiological approach in Malpighian tubules of the yellow fever mosquito Aedes aegypti which shows that the V-ATPase delivers the energy for both transcellular and paracellular ion transport.

The V-type H+ ATPase: molecular structure and function, physiological roles and regulation

It was nearly 30 years before the V-type H+ ATPase was admitted to the small circle of bona fide transport ATPases alongside F-type and P-type ATPases. The V-type H+ ATPase is an ATP-driven enzyme that transforms the energy of ATP hydrolysis to electrochemical potential differences of protons across diverse biological membranes via the primary active transport of H+. In turn, the transmembrane electrochemical potential of H+ is used to drive a variety of (i) secondary active transport systems via H+-dependent symporters and antiporters and (ii) channel-mediated transport systems. For example, expression of Cl- channels or transporters next to the V-type H+ ATPase in vacuoles of plants and fungi and in lysosomes of animals brings about the acidification of the endosomal compartment, and the expression of the H+/neurotransmitter antiporter next to the V-type H+ ATPase concentrates neurotransmitters in synaptic vesicles. First found in association with endosomal membranes, the V-type H+ ATPase is now also found in increasing examples of plasma membranes where the proton pump energizes transport across cell membranes and entire epithelia. The molecular details reveal up to 14 protein subunits arranged in (i) a cytoplasmic V1 complex, which mediates the hydrolysis of ATP, and (ii) a membrane-embedded V0 complex, which translocates H+ across the membrane. Clever experiments have revealed the V-type H+ ATPase as a molecular motor akin to F-type ATPases. The hydrolysis of ATP turns a rotor consisting largely of one copy of subunits D and F of the V1 complex and a ring of six or more copies of subunit c of the V0 complex. The rotation of the ring is thought to deliver H+ from the cytoplasmic to the endosomal or extracellular side of the membrane, probably via channels formed by subunit a. The reversible dissociation of V1 and V0 complexes is one mechanism of physiological regulation that appears to be widely conserved from yeast to animal cells. Other mechanisms, such as subunit-subunit interactions or interactions of the V-type H+ ATPase with other proteins that serve physiological regulation, remain to be explored. Some diseases can now be attributed to genetic alterations of specific subunits of the V-type H+ ATPase.

Glucose activates H+-ATPase in kidney epithelial cells

The vacuolar H+-ATPase (V-ATPase) acidifies compartments of the vacuolar system of eukaryotic cells. In renal epithelial cells, it resides on the plasma membrane and is essential for bicarbonate transport and acid-base homeostasis. The factors that regulate the H+-ATPase remain largely unknown. The present study examines the effect of glucose on H+-ATPase activity in the pig kidney epithelial cell line LLC-PK1. Cellular pH was measured by performing ratiometric fluorescence microscopy using the pH-sensitive indicator BCECF-AM. Intracellular acidification was induced with NH3/NH4+prepulse, and rates of intracellular pH (pHi) recovery (after in situ calibration) were determined by the slopes of linear regression lines during the first 3 min of recovery. The solutions contained 1 μM ethylisopropylamiloride and were K+free to eliminate Na+/H+exchange and H+-K+-ATPase activity. After NH3/NH4+-induced acidification, LLC-PK1cells had a significant pHirecovery rate that was inhibited entirely by 100 nM of the V-ATPase inhibitor concanamycin A. Acute removal of glucose from medium markedly reduced V-ATPase-dependent pHirecovery activity. Readdition of glucose induced concentration-dependent reactivation of V-ATPase pHirecovery activity within 2 min. Glucose replacement produced no significant change in cell ATP or ADP content. H+-ATPase activity was completely inhibited by the glycolytic inhibitor 2-deoxy-d-glucose (20 mM) but only partially inhibited by the mitochondrial electron transport inhibitor antimycin A (20 μM). The phosphatidylinositol 3-kinase (PI3K) inhibitor wortmannin (500 nM) abolished glucose activation of V-ATPase, and activity was restored after wortmannin removal. Glucose activates V-ATPase activity in kidney epithelial cells through the glycolytic pathway by a signaling pathway that requires PI3K activity. These findings represent an entirely new physiological effect of glucose, linking it to cellular proton secretion and vacuolar acidification.

pH-stat experiments in proximal renal tubules

The pH-stat technique has been used to measure H+ fluxes in gastric mucosa and urinary bladder "in vitro" while keeping mucosal pH constant. We now report application of this method in renal tubules. We perfused proximal tubules with double-barreled micropipettes, blocked luminal fluid columns with oil and used a double-barreled Sb/reference microelectrode to measure pH, and Sb or 1 N HC1-filled microelectrodes to inject OH- or H+ ions into the tubule lumen. By varying current injection, pH was kept constant at adjustable levels by an electronic clamping circuit. We could thus obtain ratios of current (nA) to pH change (apparent H(+)-ion conductance). These ratios were reduced after luminal 10(-4) M acetazolamide, during injection of OH-, but they increased during injection of H+. The point-like injection source causes pH to fall off with distance from the injecting electrode tip even in oil-blocked segments. Therefore, a method analogous to cable analysis was used to obtain H+ fluxes per cm2 epithelium. The relation between JH+ and pH gradient showed saturation kinetics of H fluxes, both during OH- and H+ injection. This kinetic behavior is compatible with inhibition of JH by luminal H+. It is also compatible with dependence on Na+ and H+ gradients of a saturable Na/H exchanger. H(+)-ion back-flux into the tubule lumen also showed saturation kinetics. This suggests that H+ flow is mediated by a membrane component, most likely the Na(+)-H+ exchanger.

Regulation of net bicarbonate transport in rabbit cortical collecting tubule by peritubular pH, carbon dioxide tension, and bicarbonate concentration

The effects of changes in peritubular pH, carbon dioxide tension (PCO2), and HCO3- concentration on net HCO3- transport was examined in in vitro perfused cortical collecting tubules (CCTs) from unpretreated New Zealand white rabbits. Lowering peritubular HCO3- concentration and pH by reciprocal replacement of HCO3- with Cl-, significantly stimulated net HCO3- absorption. Lowering peritubular HCO3- concentration and pH, by substitution of HCO3- with gluconate, while keeping Cl- concentration constant, also stimulated net HCO3- absorption. Raising peritubular HCO3- concentration and pH, by reciprocal replacement of Cl- with HCO3-, inhibited net HCO3- absorption (or stimulated net HCO3- secretion). When the tubule was cooled, raising peritubular HCO3- concentration had no effect on net HCO3- transport, suggesting these results are not due to the passive flux of HCO3- down its concentration gradient. The effect of changes in ambient PCO2 on net HCO3- transport were also studied. Increasing the ambient PCO2 from 40 mmHg to either 80 or 120 mmHg, allowing pH to fall, had no effect on net HCO3- transport. Similarly, lowering ambient PCO2 to 14 mmHg had no effect on net HCO3- transport. Simultaneously increasing peritubular HCO3- concentration and PCO2, without accompanying changes in peritubular pH, i.e., isohydric changes, stimulated net HCO3- secretion to the same degree as nonisohydric increases in peritubular HCO3- concentration. Likewise, isohydric lowering of peritubular HCO3- concentration and PCO2 stimulated net HCO3- absorption. We conclude that: acute changes in peritubular HCO3- concentration regulate acidification in the CCT and these effects are mediated by a transcellular process; acute changes in ambient PCO2 within the physiologic range have no effect on HCO3- transport in the in vitro perfused CCT; and acute in vitro regulation of CCT acidification is independent of peritubular pH.

Cell-mediated extracellular acidification and bone resorption: evidence for a low pH in resorbing lacunae and localization of a 100-kD lysosomal membrane protein at the osteoclast ruffled border

The extracellular compartment where bone resorption occurs, between the osteoclast and bone matrix, is shown in this report to be actively acidified. The weak base acridine orange accumulates within this compartment but dissipates after incubation with ammonium chloride. Upon removal of ammonium chloride, the cells are able to rapidly reacidify this compartment. The highly convoluted plasma membrane of the osteoclast facing this acidic compartment (ruffled border) is shown to contain a 100-kD integral membrane protein otherwise present in limiting membranes of lysosomes and other related acidified organelles (Reggio, H., D. Bainton, E. Harms, E. Coudrier, and D. Louvard, 1984, J. Cell Biol., 99:1511-1526; Tougard, C., D. Louvard, R. Picart, and A. Tixier-Vidal, 1985, J. Cell Biol. 100:786-793). Antibodies recognizing this 100-kD lysosomal membrane protein cross-react with a proton-pump ATPase from pig gastric mucosae (Reggio, H., D. Bainton, E. Harms, E. Coudrier, and D. Louvard, 1984, J. Cell Biol., 99:1511-1526), therefore raising the possibility that it plays a role in the acidification of both intracellular organelles and extracellular compartments. Lysosomal enzymes are also directionally secreted by the osteoclast into the acidified extracellular compartment which can therefore be considered as the functional equivalent of a secondary lysosome with a low pH, acid hydrolases, the substrate, and a limiting membrane containing the 100-kD antigen.

Proton transport and cell function

The past five years have witnessed an explosion of information on the many and varied roles of H+ transport in cell function. H+ transport is involved in three broad areas of cell function: (a) maintenance and alteration of intracellular pH for initiation of specific cellular events, (b) generation of pH gradients in localized regions of the cell, including gradients involved in energy transduction, and (c) transepithelial ion transport. These processes each involve one or more of several H+ translocating mechanisms. The first section of this review will discuss these H+ translocating mechanisms and the second part will deal with the cellular functions controlled by H+ transport.

Golgi membranes contain an electrogenic H+ pump in parallel to a chloride conductance

Rat liver Golgi vesicles were isolated by differential and density gradient centrifugation. A fraction enriched in galactosyl transferase and depleted in plasma membrane, mitochondrial, endoplasmic reticulum, and lysosomal markers was found to contain an ATP-dependent H+ pump. This proton pump was not inhibited by oligomycin but was sensitive to N-ethyl maleimide, which distinguishes it from the F0-F1 ATPase of mitochondria. GTP did not induce transport, unlike the lysosomal H+ pump. The pump was not dependent on the presence of potassium nor was it inhibited by vanadate, two of the characteristics of the gastric H+ ATPase. Addition of ATP generated a membrane potential that drove chloride uptake into the vesicles, suggesting that Golgi membranes contain a chloride conductance in parallel to an electrogenic proton pump. These results demonstrate that Golgi vesicles can form a pH difference and a membrane potential through the action of an electrogenic proton translocating ATPase.

Anion dependence of rabbit medullary collecting duct acidification

Rabbit medullary collecting duct (MCD) acidification has been demonstrated to occur by means of a sodium-independent, aldosterone-stimulated mechanism. We have examined the anionic dependence of this process by means of the isolated perfused tubule technique. Total replacement of perfusate chloride with gluconate enhanced tubular bicarbonate reabsorption (JHCO3), from a basal rate of 10.7 +/- 1.0 pmol X mm-1 X min-1 to a rate of 15.01 +/- 1.0 pmol X mm-1 X min-1. Removal of bath chloride, with and without removal of perfusate chloride completely abolished acidification. Bath, but not luminal 4-acetamido-4' isothiocyano-2,2'-disulfonic stilbene provoked a marked decrease in JHCO3 from 10.1 +/- 1.2 pmol X mm-1 X min-1 to 2.3 +/- 0.3 pmol X mm-1 X min-1. Measurement of chloride reabsorptive rate (JCl) revealed colinearity between JHCO3 (9.18 +/- 0.9 pmol X mm-1 X min-1) and JCl (9.75 +/- 1.18 pmol X mm-1 X min-1). We propose a model of mammalian distal nephron acidification in which (a) cellular base exit is effected by means of a basolateral membrane Cl-base exchanger and (b) net electroneutrality of electrogenic proton secretion is maintained by the parallel movement of an anionic species, functionally chloride.

Intracellular pH regulation in the renal proximal tubule of the salamander. Na-H exchange

Using pH-sensitive microelectrodes to measure intracellular pH (pHi) in isolated, perfused proximal tubules of the tiger salamander Ambystoma tigrinum, we have found that when cells are acid-loaded by pretreatment with NH+4 in a nominally HCO3--free Ringer, pHi spontaneously recovers with an exponential time course. This pHi recovery, which is indicative of active (i.e., uphill) transport, is blocked by removal of Na+ from both the luminal and basolateral (i.e., bath) solutions. Re-addition of Na+ to either the lumen or the bath results in a full pHi recovery, but at a lower-than-normal rate; the maximal rate is achieved only with Na+ in both solutions. The diuretic amiloride reversibly inhibits the pHi recovery when present on either the luminal or basolateral sides, and has its maximal effect when present in both solutions. The pHi recovery is insensitive to stilbene derivatives and to Cl- removal. A transient rise of intracellular Na+ activity accompanies the pHi recovery; there is no change of intracellular Cl- activity. These data suggest that these proximal tubule cells have Na-H exchangers in both the luminal and basolateral membranes.

Role of luminal buffers in renal tubular acidification

The acidification of kinetics of artificial solutions containing buffers of different permeancy were studied in rat proximal tubules by means of stationary microperfusion techniques. Luminal pH changes were measured by antimony microelectrodes and used to calculate net rates of acidification and the approach to steady-state pH levels. For most buffer species, tracer efflux out of the lumen was compared with changes in buffer concentration as derived from calculations based on the Henderson Hasselbalch equation. Steady-state luminal pH was similar for most buffer systems studied. However, secretory hydrogen ion fluxes into the lumen were significantly higher for permeant than for less permeant buffers. The most likely explanation is that permeant buffers behave as "open" systems maintaining constant low diffusible acid levels in the lumen, whereas impermeant buffers behave as "closed" systems in which non-ionized acid levels are maintained at higher levels. A behavior consistent with this thesis was directly demonstrated for glycodiazine and, to a lesser degree, for DMO. In contrast, phosphate and creatinine behave like buffers in a "closed" system. Characteristics of proximal tubular acidification, of buffer reabsorption, and the effect thereupon of carbonic anhydrase inhibitors are satisfactorily explained by an essential role of (1) hydrogen ion secretion, (2) pK differences, and (3) different permeance of the non-ionized buffer species. However, specific transport mechanisms may, in addition, also contribute to differences in transepithelial buffer movement.

Effects of sugars, amino acids and inhibitors on electrolyte transport across hen colon at different sodium chloride intakes

1. The isolated colonic mucosa from adult hens has been studied in vitro under short-circuit conditions. Colonic mucosa was prepared from hens fed either a NaCl-poor or a NaCl-rich diet. 2. The unidirectional transmucosal fluxes of chloride, K and Na were measured. For the high NaCl colon the fluxes of lysine were also measured. The effects of galactose, glucose, leucine, lysine, amiloride (10(-4) M) and acetazolamide (5 X 10(-4) M) on the short-circuit current (Isc) were examined. 3. The Isc of the high NaCl colon was stimulated by galactose, glucose, leucine and lysine in a manner which suggests that these substances were transported by the epithelium in co-transport with Na. The Isc was in most cases insensitive to amiloride (mucosal side), but moderately stimulated by acetazolamide (serosal side). The Isc was closely matched by the sum of the net transmucosal fluxes of chloride, lysine, K and Na. The effects of galactose, leucine and lysine on the Isc were maintained only when glucose was present on the serosal side of the preparations. 4. The Isc of the low NaCl colon depended on the presence of glucose on the serosal (or mucosal) side. It was not stimulated by galactose or by lysine, but leucine did stimulate when added to the mucosal solution. Adding amiloride on the mucosal side invariably eliminated or, in most cases, led to an inversion of the Isc. Acetazolamide markedly stimulated the Isc; and, when added after amiloride, acetazolamide caused the Isc to become positive again. The data on Isc and net fluxes of chloride, lysine, K and Na are consistent with the presence of a process of bicarbonate absorption or H ion secretion. 5. There is a net secretion of K across both the high and the low Na colon, which is highest in the former although both Isc and net absorption of Na are highest in the latter. 6. The results from both types of colon suggest that the transport of chloride differs with the functional status of tissues.

Diuretics stimulate H+ secretion in turtle urinary bladder

The effect of various diuretics on H+ secretion was studied in the isolated short-circuited urinary bladder of the turtle. Mucosal (urinary) chlorothiazide stimulated H+ secretion promptly, from 1.33 +/- 0.24 to 3.03 +/- 0.25 mueq/h (P less than 0.001). The effect was rapidly reversible upon washout of the drug, H+ returning to control levels, 1.37 +/- 0.26 mueq/h (P less than 0.001). Similar effects were observed with mucosal hydrochlorothiazide and mucosal ethacrynic acid/cysteine. Stimulation of H+ secretion occurred in the presence or the absence of exogenous CO2, in the presence or absence of mucosal Na+ and during inhibition of Na+ transport by ouabain. There was no stimulation of H+ secretion by uncomplexed ethacrynic acid or by mucosal furosemide. The nondiuretic sulfonamide, sulfasoxizole, and the nonsulfonamide buffer, borate, had no effect on H+ SECRETION. These observations indicate that the stimulatory effect of diuretics on H+ secretion is not related to active sodium transport, transepithelial electrical potential, or the buffering capacity of the drugs. Since the transepithelial pH gradient at which active H+ secretion was abolished was identical for chlorothiazide-treated tissues (2.68 pH U) as for control tissues (2.65 pH U, NS), the data suggest that the protonmotive force of the H+ pump was unaffected by the diuretic. This observation, plus the rapid onset and reversibility of the drugs, is consistent with an effect on the mucosal membrane to increase H+ conductance (K). The findings raise the possibility that direct enhancement of renal H+ secretion may play a role in the metabolic alkalosis induced by some diuretics.