In this chapter we review the transport of ions by the loop of Henle, distal convoluted tubule, the connecting tubule, and the collecting duct. We will place special emphasis on the cellular and molecular mechanisms responsible for Na + transport in these regions, as well as the factors that regulate Na + transport.
In this chapter we review the transport of ions by the loop of Henle, distal convoluted tubule, the connecting tubule, and the collecting duct. We will place special emphasis on the cellular and molecular mechanisms responsible for Na + transport in these regions, as well as the factors that regulate Na + transport.
The mammalian loop of Henle contains the descending thin limb, the ascending thin limb, and the thick ascending limb. The thin descending segment begins in the outer medulla after a gradual transition from the pars recta, and ends at the hairpin turn at the tip of Henle’s loop. The thin ascending limb begins at the tip of Henle’s loop and ends with its abrupt transition to the thick ascending limb. Loops of Henle that arise from superficial or midcortical nephrons may lack a thin ascending limb. In these short loops, thick limbs generally begin at, or slightly before, the hairpin turn (for a detailed discussion see Chapter 20 ).
The thick ascending limbs of Henle (TAL) of long looped nephrons begin at the boundary between the inner and outer medulla. The TAL of short looped nephrons does not extend as far into the medulla and may, in fact, be entirely cortical. The TAL extends up into the cortex, where it abuts the glomerulus of origin for that nephron and forms the macula densa part of the juxtaglomerular apparatus. The TAL is composed of two parts: a medullary portion and a cortical portion. The ratio of medullary to cortical TAL for a given nephron is a function of the depth of the glomerulus of the nephron such that superficial nephrons have primarily cortical thick limbs, while juxtamedullary nephrons possess primarily medullary thick limbs.
The distal nephron is divided into three segments: the distal convoluted tubule (DCT); the connecting tubule (CNT); and the collecting duct (CD). These segments are clearly delineated in the rabbit, but in many species the transition between segments is gradual. Therefore, distal tubule segments are most accurately defined by their respective cell types. The DCT begins about 50–100 µm beyond the macula densa, and is lined by a single type of cell: the DCT cell. Na + -K + -ATPase activity is particularly high in the basolateral membrane of this segment. The CNT forms a transition zone between the DCT and the cortical CD (CCD). In superficial nephrons, a single CNT drains the DCT into the collecting duct. The CNTs of deep nephrons, however, form arcades that ascend through the cortex draining several DCTs into a CCD. The CNT contains two types of cells: the CNT cell, exclusive to the CNT; and the intercalated cell, also found in the CD.
The collecting duct begins at or slightly before the confluence of two or more connecting tubules, and may be divided into three main parts: the CCD; the outer medullary collecting duct (OMCD); and the inner medullary collecting duct (IMCD). The CCD consists of at least three cell types: principal cells, responsible for Na + and K + transport; and two types of intercalated cells, responsible for H + and HCO 3 − transport. In the rabbit, principal cells account for 65–75% of cells in the CCD. The OMCD can be divided into two regions based on location: the outer stripe and the inner stripe. Approximately 80–90% of cells within the OMCD are principal cells. However, as will be discussed later, the functional properties of principal cells in the OMCD differ from those of the CCD. The IMCD extends from the junction between the inner and outer medulla to the tip of the papilla. The IMCD has been divided into three subsegments based on functional differences, including Na + transport. Intercalated cells account for about 10% of all cells in the initial portion of the IMCD, but are absent from the terminal IMCD.
Na + Transport in the Loop of Henle
The loop of Henle is responsible for absorbing 25 to 40% of the filtered sodium load. Moreover, the dissociation of salt and water absorption by the loop of Henle is ultimately responsible for the capacity of the kidney either to concentrate or to dilute the urine. The active absorption of NaCl in the water-impermeable TAL serves both to dilute the urine and supply the energy for the single effect of countercurrent multiplication.
Salt Transport by Thin Descending and Thin Ascending Segments
There is morphologic and functional evidence of interspecies and inter- and intranephron heterogeneity in the thin loop segments. The morphologic characteristics of the loop of Henle are covered in detail in Chapter 20 of this volume. Generally, loops of Henle can be divided into two groups: long loops and short loops. The thin descending limb of short loop nephrons is a simple, flat epithelium with few organelles and deep junctional complexes. The thin descending limbs of long loop nephrons are heterogeneous. The upper segment of these thin limbs has a larger diameter and thicker epithelium than short loops. The cells in this region have complicated basolateral interdigitations and apical microvilli, but shallow junctional complexes consist of a single junctional strand. These characteristics are most pronounced in rodents, while in rabbits and humans the upper portion of the thin descending limb has a simpler organization with less extensive interdigitation and deeper junctional complexes.
The lower portion of the thin descending limb consists of flat, noninterdigitating cells with a few apical microvilli and with junctional complexes of intermediate depth. There is little interspecies variability in this portion of the descending limb. The thin ascending limb, present only in long loop nephrons, consists of very flat cells connected by very shallow junctional complexes.
According to the passive models for urinary concentration (see Chapter 40 ), the thin descending limb should have very high water permeability such that the tubular fluid is concentrated by water abstraction rather than solute entry. In vitro microperfusion studies have confirmed that upper and lower portions of mammalian descending limbs are very permeable to water ( Table 34.1 ). The passive models also require the thin ascending limb to be rather impermeable to water, highly permeable to sodium chloride, and only modestly permeable to urea. As indicated in Table 34.1 , in vitro microperfusion studies of thin ascending limb segments have demonstrated that these requirements are, in fact, satisfied.
|P f (10 −3 cm/sec)||P Na (10 −5 cm/sec)||P Cl (10 −5 cm/sec)||P Na /P Cl||P urea (10 −5 cm/sec)|
|Hamster LDL u||403||45.6||4.2||–||4–6||1.5|
The permeability of thin descending limb segments to sodium and chloride has been measured in hamsters, rats, and rabbits. In hamsters and rats, the upper portion of long looped descending limbs has a higher sodium permeability and higher P Na /P Cl ratio than do descending limbs of short looped nephrons. In contrast, there is little difference in P Na /P Cl between long and short loop nephrons in rabbits. These results are consistent with the morphologic evidence of greater heterogeneity in rats and hamsters than in rabbits. The pathways for transepithelial movement of sodium and chloride in the descending limb are not defined.
The formation of dilute urine begins in the thin ascending limb of Henle. Fluid from the thin ascending limb is more dilute than fluid obtained from the descending limb at the same level. The decrease in osmolality is due primarily to a fall in the NaCl content of the luminal fluid. The mechanism for NaCl transport across the thin ascending limb epithelium is incompletely understood. Measurements of salt dilution potentials in microperfused thin ascending limb segments reveal them to be chloride selective, with P Cl /P Na ratios of 2.2 to 3.5 in rats and hamsters; segments perfused and bathed with symmetric solutions do not generate a spontaneous transepithelial voltage and do not show net transport of solute (reviewed in ). These observations have been interpreted to indicate that salt transport in vivo results from passive electrodiffusion rather than active transport. Although driven by passive electrochemical gradients, Cl − movement across the TAL proceeds through a transcellular, and regulated, pathway. 36 Cl − flux ratios and salt dilution voltages indicate that the Cl − pathway discriminates among anions and is saturable.
NaCl Absorption in Thick Ascending Limb
Rocha and Kokko and Burg and Green first demonstrated the salient characteristics of salt absorption in rabbit medullary and cortical TAL segments, respectively. First, net salt absorption resulted in a lumen-positive transepithelial voltage (Ve, mV), which could be abolished by furosemide, and in dilution of the luminal fluid. Second, the transport of Cl − under these circumstances occurred against both electrical and chemical gradients, and hence involved an active transport process. Third, both net chloride absorption and the transepithelial voltage depended on (Na + , K + )-ATPase activity, present in large amounts along the basolateral membrane of this segment. A final curious feature of the TAL is that this segment is a “hybrid epithelium” possessing a very low permeability to water, yet a high ionic conductance ( Table 34.2 ). The ionic conductance determined from the fluxes of 22 Na + and 36 Cl − is approximately 20–50 mS/cm 2 , and is cation selective (P Na /P Cl =2–6). This high electrical conductance is unusual among epithelia with low water permeabilities.
|VTe (mV)||Pf (µm/sec)||PNa (µm/sec)||PCl (µm/sec)||JCl (pEq/cm2/sec)|
Table 34.2 presents a summary of the important transport properties of the rabbit, mouse, and rat TAL which are relevant to the concentrating and diluting functions of this nephron segment. A low permeability to water is required for the TAL to function as a diluting segment. Hebert et al., and subsequently Hebert demonstrated that the apical cell membrane constitutes the major barrier to transcellular water flow in this segment.
Studies of the electrophysiologic ( Table 34.3 ) and biochemical properties of intact, isolated, perfused thick limb segments, and of apical and basolateral membranes of thick limb cells have provided insights into the specific transport mechanisms involved in salt absorption, and the origin of the lumen-positive transepithelial voltage in this nephron segment. A model for salt absorption by the TAL, which integrates the results of these studies, is shown in Figure 34.1 . Net Cl − absorption by the TAL is a secondary active transport process. Luminal Cl − entry into the cell is mediated by an electroneutral Na + -K + -2Cl − co-transport process driven by the favorable electrochemical gradient for sodium entry. More specifically, the net driving force for the entry of Cl − into the cell is determined by the sum of the chemical gradients for Na + , K + , and Cl − . Since the Na + gradient is maintained by the continuous operation of the basolateral membrane (Na + ,K + )-ATPase pump, apical entry of Cl − via the co-transporter ultimately depends on the operation of the basolateral (Na + ,K + )-ATPase. Accordingly, maneuvers that inhibit ATPase activity, such as removal of K + from or addition of ouabain to, peritubular solutions leads to dissipation of the Na + gradient and subsequent inhibition of apical membrane Cl − entry.
|Ve (mV)||Ge||Gc (mS/cm 2 )||Gs||Va (mV)||Vbl||Ra/Rb|
In contrast to the electroneutral entry of Cl − across the apical membrane, the majority of Cl − efflux across the basolateral membrane proceeds through conductive pathways. A favorable electrochemical gradient for Cl − efflux through dissipative pathways has been demonstrated by Greger et al. in the rabbit cTAL. In this study, an intracellular Cl − activity of 22 mM, measured using single Cl − -selective microelectrodes, was substantially above the equilibrium value (5 mM) predicted from the intracellular voltage. Intracellular Cl − is maintained at concentrations above electrochemical equilibrium by the continued entry of Cl − via the apical Na + -K + -2Cl − co-transporter. Blocking Cl − entry through this pathway with furosemide or substitution of extracellular Cl − by gluconate, caused the intracellular Cl − activity to fall to a value close to its equilibrium. In addition to electrodiffusive efflux of Cl − across the basolateral membrane, a component of electroneutral KCl co-transport has been proposed in some species, and the the K + -Cl − co-transporter, KCC4, has been shown to be present at the TAL basolateral membrane.
In addition, according to the model in Figure 34.1 , the potassium that enters TAL cells via the Na + -K + -Cl − co-transporter recycles, to a large extent, across the apical membrane by way of a K + conductive pathway. This apical K + recycling serves several purposes. First, it ensures a continued supply of luminal potassium in order to sustain Na + -K + -Cl − co-transport. Without recycling, the luminal K + concentration would fall rapidly as a consequence of K + entry via Na + -K + -Cl − co-transport, and would limit net NaCl absorption. Second, the apical membrane potassium current provides a pathway for net potassium secretion by the TAL, which is an active process, ultimately driven by the (Na + ,K + )-ATPase, proceeding in the face of a lumen-positive transepithelial potential. Third, under open circuit conditions, the transcellular and paracellular pathways form a current loop in which the currents traversing the two pathways are of equal size but opposite direction. The potassium current from cell to lumen polarizes the lumen and causes an equivalent current to flow from lumen to bath through the paracellular pathway. Since the paracellular pathway is cation-selective (P Na /P Cl =2–6), the majority of the current through the paracellular pathway is carried by sodium moving from the lumen to the interstitium. This paracellular absorption of sodium increases the efficiency of sodium transport by the TAL. With reference to Figure 34.1 , for each Na + transported through the cell – and requiring utilization of ATP – one Na + is transported through the paracellular pathway without any additional expenditure of energy. Finally, the apical K + current satisfies the continuity requirement imposed by a high degree of conductive Cl − efflux across basolateral membranes. A small component of sodium transport by the TAL (5–10%) is accounted for by sodium bicarbonate absorption. Sodium bicarbonate absorption is thought to be mediated by an apical membrane amiloride-sensitive Na + /H + exchanger and a basolateral membrane electrogenic 3Na + /(HCO 3 − ) co-transporter. The following sections will describe the individual components of the mechanism for TAL salt transport ( Figure 34.1 ) in greater detail.
Apical Na + -K + -Cl − Co-Transport
The predominant mode for Cl − entry into the TAL cell is via a Na + -K + -2Cl − co-transporter. A characteristic feature of this transporter is its sensitivity to inhibition by furosemide, bumetanide, and other 5-sulfamoylbenzoic acid derivatives. The first demonstration of dependence on luminal sodium for Cl − absorption was reported by Greger in isolated perfused rabbit cortical TAL segments. Moreover, a requirement for luminal potassium has been demonstrated for sodium and chloride absorption in both mouse medullary and rabbit cortical thick ascending segments perfused in vitro . As a result of these studies, it is now recognized that Cl − absorption in the TAL generally depends on the simultaneous presence of Na + and K + (or NH 4 ) in the lumen.
Measurement of isotope flux into TAL cells or membrane vesicles prepared from the inner stripe of outer medulla have delineated further the ionic requirements and stoichiometry of 1Na + -1K + -2Cl − co-transport. There is a general agreement that the co-transporter in the thick ascending limb conforms to this stoichiometry under most experimental conditions. However, under certain circumstances, which will be discussed below, apical membrane NaCl entry may be independent of luminal potassium. Early work suggested differences in apparent affinity constant for chloride along TAL which suggested axial heterogeneity in the properties of the Na + -K + -2Cl − co-transporter, because Greger, and Hus-Citharel and Morel examined the cortical TAL, while Koenig et al. and Burnham et al. prepared membranes from the medulla. An axial heterogeneity in the regulation of co-transporter activity as the thick limb ascends from the medulla into the cortex might be anticipated. This is explained by differences in affinity constant for ions of axially distributed along TAL of alternatively spliced variants of the Na + -K + -2Cl − co-transporter.
SLC12A1 gene encodes for the absorptive form of the co-transporter, referred to as BSC1 or NKCC2, located exclusively at the apical membrane of TAL, simultaneously cloned by Gamba et al. and Payne et al. Molecular biology of this gene is extensively discussed in Chapter 32 . Inactivating mutations in the SLC12A1 gene is the cause of Bartter syndrome type I. This syndrome, characterized by hypokalemia, metabolic alkalosis, hyperaldosteronism, and low blood pressure, results from a defect in salt absorption by the TAL, providing strong support for the conclusion that NKCC2 is responsible for apical Na + -K + -2Cl − co-transport in the thick ascending limb. The NKCC2 cDNA encodes a protein containing ~1100 amino acids and having a predicted molecular weight of 115–120 kDa. The observed molecular weight, however, is approximately 150 kDa due to extensive glycosylation. Hydropathy analysis of the amino acid sequence predicts a protein having 12 putative transmembrane spanning domains (TM). Six isoforms of NKCC2 have been identified that are results of alternative splicing in the 5′ and 3′ regions of the NKCC2 gene. Three 5′ splice products, termed A, B, and F, are expressed in all mammalian species. These variants differ only in a 96 bp region, which encodes the amino acids forming half of the second TM and interconnecting segment between TM2 and 3. The isoforms show differential expression within the thick ascending limb. Using RT-PCR, Yang et al. examined the distribution of NKCC2 isoforms in single nephron segments. The A isoform was found in both the cortical and medullary TAL, the B isoform was restricted to the cortical TAL, while the F isoform was present in the medullary, but not cortical, TAL and, to a lesser extent, in the outer medullary collecting duct. Altough the three variants mediate Na + -K + -2Cl − co-transport, they have different transport properties. The A and B isoforms have higher affinities for Na + , K + , and Cl − than the F isoform. The A isoform appears to have a greater transport capacity than the other isoforms (see Figure 32.3 of Chapter 32 ). Thus, the presence of the A and F isoform in the medullary thick ascending limb could account for the observed high rates of NaCl transport relative to the cortical segment ( Table 34.2 ), while the expression of the A and B isoforms in the cortical thick ascending limb may subserve the ability to achieve lower luminal NaCl concentrations in that segment. As discussed with detail in Chapter 32 , difference in affinity for ions among the NKCC2 variants B and F has been attributed to only six residues within these exons.
There is experimental evidence that the apical co-transporter in TAL might operate as a simple NaCl symporter under certain conditions. Eveloff and co-workers have reported potassium-independent, furosemide-sensitive NaCl transport in isolated rabbit TAL cells and membrane vesicles prepared from TAL cells. Potassium dependence of the co-transporter might be subject to physiologic regulation. Eveloff and Calamia have shown that under isotonic conditions the chloride-dependent, furosemide-sensitive component of sodium uptake is independent of potassium, but that under hypertonic conditions the sodium uptake becomes K + -dependent. Sun et al. have also reported that, in perfused mouse medullary TAL segments, basal sodium chloride transport was K + -independent, but that upon stimulation of salt transport by ADH NaCl transport became dependent on luminal K + . The NKCC2 variants at the carboxyl terminal domain help to explain this discrepancy. Alternative splicing at the 3′ end of NKCC2 produces additional isoforms with either long (C9) or short (C4) carboxy-termini. Under isotonic conditions, the truncated (C4) isoforms do not mediate Na + -K + -2Cl − co-transport. Under hypotonic conditions, however, the C4 isoforms mediate K + -independent NaCl co-transport, which is inhibited by cAMP and may account for the K + -independent NaCl transport noted in earlier studies. Of note, the C4 isoform inhibit the transport activity of the full-length (C9) isoforms when the two are co-expressed, and this inhibition is relieved by cAMP. The inhibition of C9 NKCC2 isoforms by C4 isoforms suggests a physical interaction between these proteins. Biochemical studies have established that NKCC2 exists as a dimer. Thus, it is possible that different combinations of NKCC2 isoforms within the dimer could produce transporters with a wide variety of functional properties. Moreover, the subunit composition of the dimers may be a point of physiologic regulation. Alternatively, the C4 isoform could alter membrane trafficking or stability of the C9 isoform. The C4 isoform resides predominantly in subapical vesicles rather than the cell membrane. Meade et al. found that co-expression of the C4 isoform reduced the abundance of C9 isoform in Xenopus oocyte membranes. The reduction in cell surface C9 isoform was reflected by a commensurate reduction in bumetanide-sensitive Rb uptake. The inhibitory effect of C4 on C9 cell surface localization could be prevented by cAMP or by disruption of microtubules, suggesting that C4 alters the trafficking of C9 in or out of the apical membrane.
Evidence for a model of two distinct binding sites for Cl − with differing affinities follows from studies of Forbush and Palfrey on 3 H-bumetanide binding to renal medullary membranes. In these studies, Na + , K + , and Cl − were all required for bumetanide to bind to canine renal medullary membranes. The K a values for sodium and potassium were 2 mM and 1 mM, respectively. The effect of Cl − concentration on bumetanide binding was biphasic. At low concentrations (<5 mM), Cl − enhanced 3 H-bumetanide-binding. These data are consistent with a model in which the binding of Cl − to a high-affinity site exposes a second lower-affinity site that may be occupied either by bumetanide or by the second Cl − . However, functional studies using chimeras constructed between the ubiquitously expressed Na + -K + -2Cl − co-transporter NKCC1 encoded by SLC12A2 gene and NKCC2, as well as in wild-type and mutant variants A, B, and F of NKCC2, revealed that changes in chloride affinity are not paralleled by changes in bumetanide affinity, and vice versa , questioning the competition of chloride and bumetanide for a single-binding site (see Chapter 32 on NaCl co-transporters).
The binding of ions to the co-transporter is thought to be ordered and cooperative. Thus, sodium binds to the co-transporter first, and promotes binding of a Cl − and then K + . Binding of K + to its site, in turn, promotes binding of the second Cl − to the co-transporter. Once fully occupied, the co-transporter–ion complex translocates to the internal surface of the cell membrane, where debinding occurs in the same order. The full reaction sequence of binding and debinding results in inward Na + /K + /2Cl − transport. Partial reactions permit K + /K + and Na + /Na + exchange.
Apical Potassium (K) Channels
Studies of isolated, perfused, thick limb segments have established that the predominant, and perhaps only, conductance across the apical membrane is to potassium. Blockade of the apical K conductance by luminal barium in mouse medullary thick ascending limb (mTAL) decreases the transepithelial electrical conductance (G e ) by roughly 50%, while increasing the apical-to-basolateral membrane resistance ratio (Ra/Rb) from 1.9 to 12.9. Moreover, changes in the luminal K + concentration produce essentially Nernstian changes in the apical membrane electrical potential. Electrophysiological studies demonstrated that the apical membrane K + conductance is sufficient to recycle the potassium uptake via the Na + -K + -2Cl − co-transporter.
The properties of the apical membrane K + conductance have been studied in plasma membrane vesicles prepared from outer renal medulla. Conductive K + fluxes in these vesicles can be measured by loading the vesicles with a high concentration of potassium, and then removing the potassium from the external solutions. Tracer amounts of 86 Rb + are then added to the extravesicular solutions to begin uptake. Under these conditions, the outwardly directed K + gradient creates an inside-negative diffusion potential that drives 86 Rb + uptake into the vesicles. Burnham et al. and Reeves et al. were thus able to demonstrate barium-sensitive 86 Rb + flux in membranes from rabbit outer medulla. The Rb + flux was conductive as judged by its inhibition via collapse of the intravesicular voltage or by measurement of intravesicular voltage using voltage-sensitive dyes. The K 0.5 value for barium inhibition of Rb + flux was 50–100 µM. The barium-sensitive Rb + flux was also dependent on the calcium activity within the vesicles. Moreover, in reconstituted proteoliposomes prepared from porcine outer medulla, the calcium dependence of the K + conductance was modulated by a high-affinity (K 0.5 =0.1 nM) calmodulin-binding site.
The patch-clamp technique has identified two types of apical K + channels in the TAL, a low-conductance (35 pS) K + channel and an intermediate-conductance (70 pS) K + channel in the apical membrane of the TAL. The 35 pS K + channel is observed in the apical membrane of rabbit, rat, and mouse. Moreover, exposure of the cytoplasmic surface of the patch to ATP inhibited channel activity. In addition to the 35 pS K + channel, Bleich et al. and Wang described an intermediate conductance (60–70 pS) K + channel in rat TAL segments. Like the low conductance K + channel, the 70 pS K + channel was inhibited by ATP and barium. In addition, the channel activity was voltage-dependent with depolarization increasing its activity; the channel could be blocked by quinine, verapamil, and diltiazem, and was inhibited at low cytosolic pH. The activities of both the intermediate and low conductance K + channels are increased by high dietary potassium. An overview regarding regulation of apical K + channels in the TAL is summarized in Figure 34.2 , and detailed information can also be found in the review articles.
An ATP-dependent K + channel was cloned from rat kidney by Ho et al. This channel, termed ROMK (Renal Outer Medullary K + channel), is the prototype for a large family of inward rectifying K + channels (K IR channels). K IR channels have two transmembrane spanning domains, intracellular N- and C-termini, and an extracellular helical domain between the two transmembrane domains. Structural studies of other K IR channels have established that the channels exist as heteromers of four K IR subunits. The extracellular helical domain appears to form the outer vestibule and selectivity filter of the channel, while the second transmembrane domain lines the pore cavity.
ROMK + channels have three alternatively spliced forms and they are ROMK2 (Kir1.1b), ROMK3 (Kir1.1c), and ROMK6 (Kir1.1d). The encoded ROMK proteins differ at the beginning of the N-terminus; ROMK2 has the shortest N-terminus and splicing adds either 19 or 26 amino acids for ROMK1 or ROMK3, respectively. Relative ROMK mRNA abundance measured by competitive PCR has shown that ROMK2 and ROMK3 are much more abundant than ROMK1 or ROMK6 in rat kidney. Moreover, single-nephron PCR analysis indicated that ROMK1 is expressed in the collecting duct, while both ROMK2 and ROMK3 are expressed in the medullary and cortical thick ascending limb. Immunolocalization of ROMK proteins, using antibodies that recognized all three isoforms, revealed apical membrane localization in the medullary and cortical thick ascending limb, the late distal tubule, the connecting tubule, and the cortical collecting duct. Oddly, the expression of ROMK in the thick limb was heterogeneous, with some cells lacking demonstrable staining. In contrast, the expression of the Na + -K + -2Cl − co-transporter was uniform within the TAL. This implies that apical K + recycling by the ROMK-positive cells may be sufficient to supply K + for Na + -K + -2Cl − transport in the ROMK-negative cells.
The 35 pS channel conductance of ROMK corresponds to the low conductance K + channel described by Wang et al. The current–voltage relations exhibit weak inward rectification, which is due to blocking of the channel by intracellular Mg 2+ or polyamines. The activity of this channel is modulated by intracellular pH and by intracellular ATP, through both phosphorylation-dependent and phosphorylation-independent pathways. The maintenance of ROMK + channel activity requires protein kinase A (PKA)-mediated phosphorylation and deletion of any two of three PKA-phosphorylation sites inactivates ROMK. Phosphorylation of ROMK affects both channel activity (P o ) and the number of functional channels in the plasma membrane. In addition, PKA-induced phosphorylation of ROMK enhances the sensitivity of the ROMK + channel to phosphatidylinositol phosphates (PIP2).
Nucleotides have both inhibitory and stimulatory effects on ROMK activity. At micromolar concentrations, ATP stimulates ROMK activity via PKA-mediated phosphorylation and production of PIP 2 . Millimolar concentrations of ATP inhibit ROMK activity. The inhibition by ATP can be relieved by increasing concentrations of ADP or by PIP 2 . Certain other members of the K IR family are also inhibited by ATP. The ATP sensitivity of these channels–for example, K IR 6.2–is endowed by the association of regulatory subunits. These regulatory subunits, such as the sulfonylurea receptor (SUR) and CFTR, belong to the ABC gene family which is characterized by nucleotide-binding domains. The sulfonylurea receptor, together with K IR 6.2, forms the glibenclamide-sensitive, ATP-sensitive K + channel that controls insulin secretion by pancreatic β-cells. The K IR /SUR channels are more sensitive to inhibition by ATP than those of heterologously expressed ROMK + channels. In addition, the native 35 pS K + channel, but not the heterologously expressed ROMK + channel, is inhibited by glibenclamide. These observations have prompted the search for regulatory subunits for the ROMK + channel. Both SUR2B and CFTR are expressed in the thick ascending limb and collecting duct, making them candidates for ROMK regulation. In transfection studies, co-expression of ROMK with either SUR2B or CFTR dramatically increases both the nucleotide dependence and glibenclamide sensitivity of ROMK. Immunoprecipitation studies confirmed that one of the ROMK isoforms, ROMK2, physically interacts with SUR2B. The role of CFTR in regulating the native ROMK + channels has been further suggested by the finding that PKA-induced regulation of ATP sensitivity of the native 35 pS K + channel is compromised in CFTR knockout mice. Figure 34.3 is a cartoon illustrating the current view regarding the composition of the native ATP-sensitive K + channel in the apical membrane of the TAL.
The electrophysiologic properties of the ROMK channels have strongly suggested that ROMK is the native 35 pS K + channel expressed in the thick ascending limb. In support of this view, no low-conductance K + channels were detected by patch-clamp analysis in ROMK knockout mice. Moreover, deleting the gene product encoding ROMK also abolished the expression of intermediate conductance K + channels in the thick ascending limb. It has been proposed that the intermediate conductance channel may be a heteromeric protein containing ROMK and other, as yet unidentified, subunits. The notion that one of the ROMK channels is the predominant channel responsible for apical K + recycling in the thick ascending limb is strongly supported by the finding of mutations in the ROMK gene in some families with Bartter syndrome. Thus, the presence of ROMK mutations as a cause of Bartter syndrome indicates the important role of ROMK in net salt absorption by the thick ascending limb.
Basolateral Potassium (K) Channels
Basolateral K + channels in the TAL play an important role in the regulation of transepithelial transport. They are responsible for generating basolateral membrane potential which is essential for Cl diffusion across the basolateral membrane in the TAL. Activation of basolateral K + channels in the TAL is expected to hyperpolarize the basolateral cell membrane, thereby augmenting the driving force for Cl exit across the basolateral membrane. In contrast, decreasing basolateral K + channel activity depolarizes the cell membrane potential, thereby diminishing the driving force for Cl − exit across the basolateral membrane. Consequently, inhibition of Cl − exit leads to an increase in intracellular Cl − concentration which decreases the activity of NKCC2, probably via inhibiting WNK3-SPAK.
The physiological importance of the basolateral K + channels in maintaining transepithelial membrane transport in the TAL and distal nephron is best demonstrated in SeSAME disease ( Se izures, S ensorineural deafness, A taxia, M ental retardation, and E lectrolyte imbalance). This disease is the result of defective gene product encoding KCNJ10, an inwardly-rectifying K + channel, which is also expressed in the basolateral membrane of TAL and distal nephron. The renal phenotypes of SeSAME disease are hypokalemia, metabolic alkalosis, and hypomagnesemia. Presumably, defective basolateral K + channels cause membrane potential depolarization, thereby decreasing Cl − diffusion across the basolateral membrane. Consequently, an increase in intracellular Cl − concentrations results in the inhibition of the apical Na entry through NKCC2, thereby suppressing Na absorption and diminishing the lumen-positive potential in the TAL. A decrease in the lumen-positive potential results in inhibition of magnesium absorption through the paracellular pathway in the TAL. Also, the inhibition of Na absorption in the TAL increases Na delivery to the distal nephron, accordingly stimulating Na absorption in the expanse of K + in the connecting tubule and causing K + -wasting. Therefore, an alteration in the basolateral K + channel activity in the TAL has a significant effect on transepithelial transport not only in the TAL, but also in the distal nephron segment.
Patch-clamp experiments demonstrated that an inwardly-rectifying 50 pS K + channel and a Na + and Cl − activated 140–180 pS K + channel are present in the basolateral membrane of the TAL. The 50 pS K + channel is highly-expressed, and may be the main K + channel in the basolateral membrane of the TAL. The regulation of the 50 pS K + channels has been intensively studied and several factors, including protein kinase A (PKA), 20-hydroxyeicosatetraenoic acid (20-HETE), and external Ca 2+ , have been identified to modulate the 50 pS K + channels ( Figure 34.2 ). Arachidonic acid inhibits the basolateral 50 pS K + channels in the TAL through the CYP-omega-hydroxylase-dependent metabolism, and 20-HETE mediates the effect of arachidonic acid on the K + channel. Also, raising the external Ca 2+ inhibited the basolateral 50 pS K + channels in the TAL by a PKC-dependent mechanism. Two lines of evidence suggest that the effect of the external Ca 2+ on the basolateral 50 pS K + channels was the result of stimulating the Ca 2+ -sensing receptor (CaSR) which is expressed in the TAL : (1) the inhibitory effect of raising the external Ca 2+ on the 50 pS K + channels was observed only in cell-attached patches, but not in excised patches; (2) the effect of raising external Ca 2+ on the 50 pS K + channel was absent in the presence of the CaSR antagonist. It is possible that the CaSR might directly interact with the basolateral 50 pS K + channels in the TAL. This speculation is supported by reports from several studies. First, the basolateral 50 pS K + channels in the TAL might represent heterotetramers made of Kir4.1/Kir5.1, because the basolateral K + channels in the distal tubules with similar biophysical properties to those of the 50 pS K + channel in the TAL are composed of Kir4.1 and Kir5.1. Second, immunostaining experiments demonstrated that Kir4.1 and Kir5.1 were expressed in the basolateral membrane of the TAL. Finally, the study using immunoprecipitation performed in the cells transiently transfected with Kir4.1 and the CaSR has demonstrated that Kir4.1 was associated with the CaSR. The regulation of the basolateral 50 pS K + channel by the external Ca 2+ should play a role in the modulation of transepithelial Na transport and concentrating ability in the TAL. Since the active reabsorption of NaCl − in the water-impermeable TAL is essential for urinary concentrating mechanism, inhibition of NaCl − reabsorption in the TAL should result in a decrease in concentrating ability. Indeed, it has been reported that hypercalcemia impairs urinary concentrating ability.
Because the activity of the basolateral K + channels is regulated by the external Ca 2+ /Mg 2+ concentrations at a physiologically relevant range (1 to 2 mM), the CaSR-mediated regulation of the membrane transport in the TAL is mainly through controlling the basolateral K + channel activity. Figure 34.4 is a cell scheme illustrating the role of basolateral K + channels in mediating the effect of stimulation of the CaSR on Na + and divalent cation transport in the TAL. Under physiological conditions, the basolateral K + channels maintain the cell membrane potential such that it could sustain a constant Cl − diffusion across the basolateral membrane. An increase in the external Ca 2+ (hypercalcemia) activates the CaSR, thereby inhibiting basolateral K + channels and decreasing the driving force for Cl − diffusion across the basolateral membrane. Inhibition of Cl − exit is expected to lead to hyperpolarization of transepithelial voltage (V te ). A less positive V te would diminish the reabsorption of Na + and divalent cations such as Ca 2+ and Mg 2+ . Indeed, it has been reported that an increase in plasma Ca 2+ /Mg 2+ level enhanced urinal Ca 2+ /Mg 2+ excretion.
Basolateral Membrane Cl − Channel and Transporter
Cl − Channel
Cl − exit across the basolateral membrane of TAL cells is largely conductive, proceeding down its electrochemical gradient through Cl − -selective channels in the basolateral membrane. The notion that basolateral Cl − transport is electrogenic first derived from observations that, in the mouse medullary TAL and rabbit cTAL, net Cl − absorption accounts for about 90% of the equivalent short-circuit current. Measurements of the basolateral membrane voltage by Greger and Schlatter confirmed that reductions in bath chloride concentration depolarized the basolateral membrane, while reductions in intracellular chloride concentration produced by blocking Cl − entry with furosemide hyperpolarized the basolateral membrane. Both sets of observations are consistent with the presence of a Cl − conductance in the basolateral membrane. Consistent with the view that basolateral Cl − transport is via chloride channels, a variety of compounds known to block Cl − channels also inhibit salt absorption in TAL segments. Wangemann et al. have catalogued the electrophysiologic effects, relative potencies, and structure–function relations of over 200 such compounds. The major effects of these agents, when present in the peritubular bathing solutions, are inhibition of transepithelial voltage, inhibition of the equivalent short circuit current, and hyperpolarization of the basolateral membrane.
Application of the patch-clamp technique to the TAL has established that Cl − channels are present in the basolateral membrane of TAL cells. Paulais and Teulon detected a 40 pS anion selective channel (P Cl /P Na =20) in the basolateral membrane of collagenase treated mouse cortical TAL segments. The I–V relations of the channel were linear in both the cell-attached and excised configurations. The open probability of the channel in the cell-attached state was voltage-dependent, increasing as the membrane was depolarized. In the excised patch configuration, the open probability was no longer voltage-dependent. Greger et al. described a Cl − channel in the basolateral membrane of rat TAL segments. This channel also has a conductance of about 40 pS, but rather than having a linear I–V relation, this channel exhibits outward rectification. The open probability increases with depolarization in both the cell-attached and -excised patch configuration. A low conductance Cl − channel (8–10 pS) having linear I–V relations has also been detected in the basolateral membrane of TAL cells. The activity of this channel is increased following incubation with cAMP-dependent protein kinase and ATP, and is inhibited by 20-HETE acid, a Cyp-ω-hydroxylase-dependent metabolite of arachidonic acid.
Evidence for Cl − channels in the TAL also comes from studies of Cl − flux in renal medullary membrane vesicles. Since TALs comprise approximately 70% of the volume in the inner stripe of outer medulla, vesicles prepared from this region should be predominantly derived from this segment. 36 Cl − flux into vesicles from rabbit outer medulla is electrogenic, cation-independent, inhibitable by chloride channel blockers, and has a low activation energy (E a =6.4 kcal/mole), characteristic of transport through a channel. Moreover, when vesicles from rabbit outer medulla were incorporated into planar lipid bilayers, chloride channel activity was demonstrated. These channels were anion selective (P Cl /P K =10) and had a single-channel conductance of 80–90 pS in 320 mM KCl solution. The I–V relations in symmetric solution were linear, and in asymmetric solutions the I–V relations conformed to the Goldman–Hodgkin–Katz equation. The open probability of this channel was voltage-dependent, increasing activity with depolarizing voltages. These channels are also seen in vesicles made from highly purified suspensions of mouse thick ascending limb.
Molecular cloning has demonstrated the basolateral Cl − channels are composed of ClC-K1 and ClC-K2, members of a large number of the ClC family of channels. ClC-K1 and ClC-K2 are expressed exclusively within the rat kidney. In the human, the two corresponding channels are denoted hCLC-Ka and hCLC-Kb, and are located contiguously on chromosome 1. Due to the high degree of sequence similarity between hCLC-Ka and hCLC-Kb, it is not certain which of the human channels correspond to CLC-K1 versus CLC-K2, although the distribution of ClC-K2 along the nephron most closely matches that of hClC-Kb. Zimniak and colleagues have cloned cDNA from rabbit renal outer medulla, named rbClC-Ka, which shares 80% homology to the rat CLC-K1 and CLC-K2. The distribution of rbClC-Ka along the nephron resembles that of CLC-K2 rather than ClC-K1. Several lines of evidence support the view that CLC-K2 (or the probable human homolog hCLC-Kb) is the channel that mediates chloride efflux across the basolateral membrane of the thick ascending limb. First, using polymerase chain reaction amplification of single tubule segments, the ClC-K2 and rbClC-Ka channels were shown to be expressed primarily in the thick ascending limb and the collecting duct. Second, immunohistochemical studies using an antibody against the rbClC-Ka channel revealed predominantly basolateral staining in the medullary and cortical thick ascending limb. Similar results were obtained by Vandewalle and colleagues using an antibody that recognized both CLC-K1 and CLC-K2. Finally, and most compelling, is the identification of mutations in hCLC-Kb in patients with neonatal Bartter syndrome. Thus, as was the case for the NKCC2 and ROMK proteins, linkage of ClC-Kb to Bartter syndrome establishes the importance of its gene product in transepithelial NaCl transport. Activating mutations of CLC-Kb have also been reported. Specifically, substitution of threonine by serine at position 481 of CLC-Kb results in a dramatic increase in Cl − currents, without a change in channel selectivity or cell surface expression. The T481S polymorphism is relatively common in the general population, particularly in African populations. Among Caucasians, the presence of the T481S polymorphism was associated with higher systolic and diastolic blood pressures, and a higher prevalence of hypertension. Thus, ClC-Kb is an attractive candidate gene for certain forms of essential hypertension, particularly salt-sensitive hypertension. Additional studies in different populations will be required to determine the significance of these ClC-Kb polymorphisms.
To form functional Cl − channels in thick ascending limb requires not only CLC-K2/hCLC-Kb, but also a subunit, named barttin. Lost function of mutation of barttin is responsible for a form of Bartter syndrome accompanied by sensorineural deafness. Barttin is believed to act, at least in part, by increasing the cell surface expression of ClC-Kb.
Some uncertainty remains regarding the role of basolateral electrochemical KCl symport in net Cl − efflux across that membrane. In the rabbit cortical TAL, Greger and Schlatter concluded that KCl symport accounted for about one-third of basolateral Cl − efflux. This conclusion is based on the following observations: an increase in the K + concentration or decrease in the Cl − concentration of the basolateral solution depolarized the basolateral membrane; bath barium depolarized the basolateral membrane and abolished the K + -induced changes in V b l; and barium had no discernible effect on the transepithelial resistance or fractional resistance of the basolateral membrane. The lack of an effect of barium on resistance persuaded the investigators to propose that a barium-sensitive KCl co-transporter was present. Alternatively, these data are compatible with parallel conductive pathways for K + and Cl − . The absence of a barium effect on transepithelial resistance could be due to an offsetting increase in basolateral Cl − conductance or to changes in basolateral membrane resistance below the experimental limits of detection. A cloned KCl co-transporter, KCC4, is present in the basolateral membrane of the thick ascending limb. However, its physiologic role in thick ascending limb function is not known.
Synchronous Na + /H + : Cl − /HCO 3 − Exchange
An additional form of apical membrane NaCl entry has been observed in the mouse cTAL. Friedman and Andreoli found that net Cl − absorption and the transepithelial voltage were doubled when CO 2 and HCO 3 − were added to the external solutions bathing cTAL segments. Since the (CO 2 HCO 3 − )-stimulated rate of NaCl absorption did not result in net CO 2 transport, and could be abolished by the lipophilic carbonic anhydrase inhibitor ethoxyzolamide or by the luminal addition of the anion exchange inhibitor SITS or DIDS, it was proposed that the apical membrane of the mouse cTAL contains parallel, near-synchronous Na + /H + :Cl − /HCO 3 − exchangers in addition to a Na + -K + -Cl − co-transporter. Subsequent studies have shown that, like the mouse mTAL, the apical membrane of the mouse cTAL contains a potassium conductance, and that both the (CO 2 HCO 3 − )-dependent and (CO 2 HCO 3 − )-independent components of NaCl absorption require luminal potassium. Thus, the cation exchange process may proceed as (Na + , K + )/2 H + . Both the (CO 2 HCO 3 − )-dependent and -independent components of NaCl absorption in mouse cTAL segments were equally susceptible to inhibition by luminal bumetanide (K i =5–8×10–7 M). Addition of CO 2 and HCO 3 − to the bathing solutions has no effect on net NaCl transport in either rabbit cTAL or mouse mTAL. Both the rat and mouse mTAL do contain Na + /H + exchangers in their apical membranes. However, in these segments, Na + /H exchange plays a role in net HCO 3 − transport and cell pH regulation, rather than transcellular NaCl absorption.
Bicarbonate and Ammonium Transport
Medullary and cortical TAL segments from the rat absorb bicarbonate and acidify the luminal fluid. The rates of bicarbonate absorption measured in in vitro perfused TAL segments account for most of the filtered bicarbonate that is reabsorbed by the loop of Henle in vivo . The rate of sodium bicarbonate absorption in the rat TAL is, however, only a small fraction (5–10%) of total sodium absorption by this segment. Thus, while bicarbonate transport by the TAL may play an important role in urinary acidification in some species, bicarbonate transport has little impact on net salt balance or free water excretion. There is considerable species variation in the rates of bicarbonate absorption by the TAL. No significant bicarbonate transport was detected in mouse and rabbit TAL. In the rabbit, this correlates with the absence of carbonic anhydrase activity in the TAL.
The mechanism of bicarbonate absorption by the rat TAL has been reviewed. Transcellular bicarbonate absorption results from proton secretion across the apical membrane and bicarbonate reabsorption across the basolateral membrane. The apical proton secretion occurs primarily by NaH exchange. Evidence for functional apical Na + /H + exchange has been presented for the rat and mouse TAL. In these segments, acidification of luminal fluid is sodium-dependent and amiloride-sensitive. Moreover, removal of luminal sodium or luminal amiloride results in cytoplasmic acidification. As noted above, the NHE3 isoform of the Na + /H + exchanger is expressed in the apical membrane of the thick ascending limb. NHE3 expression in the TAL is increased by chronic metabolic acidosis, providing a mechanism for enhanced HCO 3 − absorption in this setting. An apical membrane H-ATPase may also contribute to bicarbonate absorption in the TAL. Inhibition of the Na + -K + -Cl − co-transporter by furosemide stimulates bicarbonate absorption in the rat TAL. Thus, this transporter is not directly involved in bicarbonate absorption. Rather, the reduction in cell Na + activity that attends inhibition of Na + -K + -Cl − co-transport provides a greater driving force for apical Na + /H + exchange.
Krapf has demonstrated that base efflux across the basolateral membrane of perfused rat TAL segments occurs as Na + (HCO 3 − ) co-transport. This process is electrogenic (probable stoichiometry 1Na + :3HCO 3 − ), sodium-dependent, Cl − -independent, and SITS-sensitive. Na + (HCO 3 − ) co-transport has been demonstrated in mouse medullary TAL as well. In the mouse, the apical Na + /H + exchanger and basolateral Na + (HCO 3 − ) co-transporter play a role in cell pH regulation, rather than transcellular bicarbonate transport.
Origin of Transepithelial Voltage
The spontaneous, lumen-positive transepithelial potential that accompanies net NaCl absorption is, in principle, the sum of at least two terms: an electrogenic voltage arising from rheogenic cellular pathways; and a zero-current dilution voltage referable to salt accumulation in intercellular spaces during salt absorption. Given the cation selectivity of the paracellular pathway (P Na /P Cl ~2–6), an accumulation of Na + in the lateral intracellular space mediated by the (Na + , K + )-ATPase could create a lumen-positive diffusion potential across the junctional complex. As discussed earlier, the lumen-positive potential arising from rheogenic cellular transport serves to drive a proportion of net sodium reabsorption through the paracellular pathway. However, this paracellular sodium absorption will be diminished by the extent to which paracellular diffusion potential accounts for the spontaneous transepithelial voltage, V e . Thus, it is pertinent to consider the relative contributions of both electrogenic and diffusion potentials to the total transepithelial voltage.
Hebert and Andreoli assessed the possible contributions of a paracellular diffusion potential to V e in the mouse TAL. The conductance of the paracellular pathway, G s , was measured by blocking the transcellular conductance with 20 mM Ba 2+ in the lumen. The junctional complexes were then disrupted by the imposition of large osmotic gradients produced by the addition of urea to the luminal perfusate. Using the value of G s in the presence of 800 mM luminal urea as an estimate of the ionic conductance of the lateral interspace, exclusive of the junctional complex, Hebert and Andreoli calculated that, during net NaCl transport, the maximal rise in NaCl concentration in the lateral interspace was 10 mEq/liter, and that the resulting dilution potential was less than 1 mV. These results are consistent with the notion that virtually all of the transepithelial voltage is the result of rheogenic transcellular processes.
Specifically, because the apical membrane is exclusively conductive to potassium, and because the Na + K + 2Cl − co-transporter is electroneutral, the apical membrane voltage, V a , will approximate the K + equilibrium voltage, E K . The basolateral membrane voltage, V bl , in contrast, is a function of several conductive pathways; examples include K + and Cl − channels, Na + HCO 3 − symport, (Na + , K + )-ATPase. Of these, Cl − is likely the most important conductive species across the basolateral membrane. V bl , therefore, will be greater than E Cl , but less that E K , and hence less than V a ( Table 34.3 ). The lumen-positive V e then is the result of differing conductance characteristics of the apical and basolateral membranes. The exact value of V e will be determined by the relative magnitude of the basolateral K + and Cl − conductances, and the currents passing through the apical and basolateral membranes.
According to these arguments, the electrogenic nature of V e should allow for a significant fraction of net sodium absorption to proceed via the paracellular pathway. Indeed, for a constant stoichiometry of the Na + K + -2Cl − entry, the ratio of net Cl − absorption to paracellular sodium absorption should have a value of 2. The rate of net paracellular Na + absorption depends on a variety of factors, such as V e , G s , and P Na /P Cl , which vary considerably from tubule to tubule. When each of these variables was measured in the same tubule, however, the ratio of net Cl − absorption to paracellular Na + absorption was reasonably constant at 2.4±0.3. Thus, the stoichiometry of Na + K + -2Cl − entry may be constant under those experimental conditions, that is, ADH stimulated mTAL segments, and the variables G e , G s , and P Na /P Cl are related in a given tubule in such a way as to maintain the net Cl − to paracellular Na + ratio at 2.
Coupling of Substrate Utilization to Ion Transport
Transepithelial NaCl absorption accounts for the single largest expenditure of energy by TAL cells. Inhibition of the (Na + ,K + )-ATPase by ouabain reduces O 2 consumption by rabbit, rat, and mouse TAL segments by 50%. The effect of ouabain on O 2 consumption can be mimicked by either furosemide or by removal of Na + or Cl − from the bath solutions. This suggests that about one-half of the total cellular ATP supply is consumed by the basolateral membrane (Na + ,K + )-ATPase pump, and that this pump operates almost exclusively for the purpose of transepithelial NaCl transport.
TAL segments utilize a variety of metabolic substrates including D-glucose, D-mannose, butyrate, β-OH-butyrate, acetoacetate, lactate, acetate, and pyruvate to support the high ATP requirement imposed by salt transport. Uchida and Endou determined the ability of various substrates to maintain cellular ATP concentrations in microdissected mouse TAL segments. In mTAL segments, glucose, lactate, and β-hydroxybutyrate all conserved cellular ATP content equally well. In the cTAL, lactate and β-hydroxybutyrate increased cellular ATP more than did glucose. Wittner et al. reported that in the absence of exogenous substrate, NaCl transport fell by 73% in 10 minutes in rabbit cTAL. Chamberlin and Mandel assessed the ability of various endogenous and exogenous substrates to support oxidative metabolism in suspensions of rabbit mTAL segments. Unlike the findings in rabbit cTAL, the mTAL suspensions maintained 85% of their basal O 2 consumption in the absence of exogenous substrates. Inhibitors of glycolysis, fatty acid oxidation, and amino acid oxidation reduced O 2 consumption, indicating that glucose, fatty acids, and amino acids all serve as endogenous substrates capable of supporting oxidative metabolism when the availability of exogenous substrate is limited.
Therefore, the TAL, particularly the medullary portion, relies almost exclusively on oxidative metabolism as its source of energy. However, the mTAL resides in an oxygen-poor environment. The oxygen tension in the outer medulla is about 10 mmHg. Apparently, like the cTAL, the O 2 affinity of the mTAL is sufficiently high that oxidative metabolism is maintained even at low oxygen pressures. Nonetheless, the high oxygen requirement in an oxygen-poor environment might predispose the mTAL to injury if oxygen delivery falls or salt transport requirements increase. Brezis and co-workers have shown that the medullary TAL is, in fact, exquisitely susceptible to hypoxic injury. Maneuvers that diminish TAL salt transport, such as furosemide, exert a marked protective effect against hypoxic injury.
Regulation of Salt Absorption in TAL
The rate of salt absorption by the thick ascending limb is modulated by physical factors, such as luminal flow rate and the composition and osmolality of luminal and peritubular fluids, and hormones, such as vasopressin and glucagon, which exert their effects through interactions with specific cellular receptor proteins. The regulation of TAL salt transport is characterized by considerable interspecies variation, as well as intranephron heterogeneity. It is worth noting, again, that the cortical and medullary portions of the TAL have different transport properties ( Table 34.2 ) and subserve different functions. Salt absorption in both segments creates luminal fluid dilution and the potential for free water excretion; however, only salt absorption by the medullary TAL enriches the medullary interstitial osmolality and enhances urinary concentrating power. Given these differences, it is not surprising that the rates of transport in the medullary and cortical TAL are modulated quite differently. A good example of both species and intranephronal heterogeneity is the effect of antidiuretic hormone on TAL transport.
The ADH affects NaCl absorption in the ascending limb of Henle, and thereby regulates the countercurrent multiplication process. Support for this contention was provided by the demonstration, in some species, of an ADH-induced increase in adenylate cyclase activity and protein kinase activity in the medullary, but not cortical, portions of the TAL.
Table 34.4 summarizes the effects of ADH on salt transport in in vitro microperfused medullary TAL segments of mouse, rat, and rabbit. Hall and Varney established that ADH increased the lumen-positive transepithelial voltage and the net rate of tracer Cl − absorption in the mouse medullary TAL. Several laboratories have confirmed these effects of ADH on the transepithelial voltage and unidirectional Cl − flux in the mouse, and have demonstrated that the effect of ADH on NaCl transport was restricted to the medullary portion of this segment. Moreover, the maximal stimulation of the transepithelial voltage in the mouse medullary TAL occurred at hormone concentrations of approximately 20 pg/ml or 10 µU/ml, found during states of antidiuresis. A similar increase in transepithelial voltage can be obtained in the mouse medullary TAL, but not the cortical TAL, by the addition of cAMP analogs or forskolin, a nonhormonal activator of adenylate cyclase to the peritubular media. In contrast, ADH, even at peritubular concentrations of 250 µU/ml, has no effect on V e in the mouse cortical TAL. Taken together, these results indicate that the mouse medullary TAL and cortical TAL are functionally heterogeneous with respect to the effect of ADH on net NaCl absorption, and that stimulation of net NaCl absorption in the mouse medullary TAL by ADH is mediated by cAMP. The effect of ADH in TAL does not seem to be contant in all species. ADH causes a variable increase in cyclic AMP production in the rabbit TAL, but pharmacologic concentrations of ADH had no effects on either the transepithelial voltage or tracer Cl − efflux in this species. ADH or the V 2 selective analog dDAVP increased NaCl absorption from the TAL on homozygous Brattleboro (central diabetes insipidus) rats. In humans evidence is limited. Studies in pump-perfused human kidneys rejected for transplantation have shown ADH-sensitive adenylate cyclase activity in collecting ducts, but not in TAL segments. Since the ability of humans to elaborate concentrated urine is limited compared to the mouse, the failure of ADH to increase cAMP in the human TAL is consistent with Morel’s suggestion that the response of medullary TAL segments to ADH correlates directly with urinary concentrating ability.
|Species||ADH||V e (mV)||J NaCl (pmol/sec/cm 2 )|
Mechanisms of ADH Affect on TAL
The mechanism whereby ADH stimulates salt transport in the medullary TAL has been studied most extensively in the mouse. The effects of ADH in this segment are mediated predominantly through interaction of the hormone with the vasopressin V 2 receptor coupled to adenylate cyclase. Thus, ADH stimulates adenylate cyclase activity in the TAL, and cAMP analogs and forskolin mimic the effects of ADH on transport in this segment. Cholera toxin stimulates salt transport in the TAL, indicating that the V 2 receptor is coupled to the catalytic subunit of adenylate cyclase by a stimulatory guanine nucleotide-binding protein, G s . The catalytic subunit of adenylate cyclase may, in turn, be influenced by calmodulin. Takaichi and Kurokawa have shown that calmodulin inhibitors inhibited ADH-sensitive cAMP production in isolated mouse medullary TAL segments. Likewise, Ausiello and Hall showed that the addition of exogenous calmodulin stimulated ADH-sensitive cAMP production by LLC PK 1 cells.
Table 34.5 lists the effects of ADH on several electrophysiologic parameters of mouse medullary TAL segments perfused in vitro . The approximate doubling of the equivalent short-circuit current results from an increase in the transepithelial voltage and an increase of roughly 20% in the transepithelial conductance. Moreover, the ADH-induced increase in transepithelial conductance, G e , is referable to an increase in the barium-sensitive or transcellular conductance, Gc. ADH has no effect on the magnitude or permselectivity of the paracellular conductance, G s ( Table 34.5 ).
|ADH (μU/ml)||V e (mV)||G e (mS/cm 2 )||G c (mS/cm 2 )||G s (mS/cm 2 )||V a (mV)||V bl (mV)||R a /R b|
ADH increases both the apical and basolateral membrane conductance. The apical membrane of TAL cells is, as noted previously, predominantly K + conductive. In the presence of luminal furosemide to block K + uptake via the Na + -K + -2Cl − co-transporter, mouse TAL segments exhibit net K + secretion that is increased by ADH. Because the electrochemical driving force for K + secretion is presumably constant in the presence of furosemide, it was suggested that the increase in K + secretion was due to an increase in apical membrane K + conductance. In support of this notion, using membrane vesicles prepared from rabbit outer medulla, Reeves et al. demonstrated that in vitro exposure to cAMP-dependent protein kinase and ATP specifically enhanced barium-sensitive rates of 86 Rb + uptake. This finding has been confirmed in subsequent work that examined the effects of phosphorylation on the activity of ROMK channels. ROMK is phosphorylated by cAMP-dependent protein kinase at a number of sites, and this phosphorylation increases the activity of the channel.
The ADH-induced increase in transepithelial NaCl absorption can be blocked by luminal furosemide, and therefore represents an increase in apical Na + -K + -2Cl − co-transport activity. The mechanism whereby ADH increases apical co-transport activity is unclear. Sun et al. demonstrated that in isolated perfused mouse TAL segments, ADH changed the stoichiometry of NaCl entry from 1Na + :1Cl − to 1Na + :1 K + :2Cl − . This change in stoichiometry, accompanied by apical K + recycling, could create a lumen-positive potential to drive Na + reabsorption via the paracellular pathway. Phosphorylation of the cloned Na + -K + -2Cl − co-transporter NKCC2 by cAMP-dependent protein kinase does not appear to directly increase co-transporter activity. However, phosphorylation may influence co-transporter activity by modulating the interactions between NKCC2 isoforms. As noted above, an isoform with a truncated C-terminus, denoted C4, can inhibit the activity of the full-length NKCC2. However, the inhibition by the C4 isoform can be relieved by phosphorylation by cAMP-dependent protein kinase. Thus, ADH may increase apical Na + -K + -2Cl − entry by releasing NKCC2 from tonic inhibition by truncated isoforms. Of note, the short C4 variant is expressed in medullary TAL, but not in cortical TAL, providing a potential explanation for the lack of ADH effect on cortical TAL discussed above. Chronic exposure to ADH may increase apical Na + -K + -2Cl − entry by increasing the abundance of NKCC2.
The predominant portion of the ADH-induced increase in cellular conductance is accounted for by an increase in the basolateral membrane Cl − conductance. Two mechanisms have been suggested for the hormone-dependent increase in basolateral Cl − conductance. Schlatter and Greger proposed that the ADH-induced increase in intracellular cAMP results in a direct increase in chloride channel activity. Such a mechanism has been amply demonstrated in Cl − secreting epithelia, such as the trachea and intestine. In support of their proposal, Schlatter and Greger demonstrated that cAMP and ADH elicited a fall in the fractional resistance of the basolateral membrane in mouse medullary TAL segments, even when cell Cl − activity was kept at low levels by blocking apical Cl − entry with furosemide. Likewise, using patch-clamp analysis, Paulais and Teulon found that preincubation of mouse cortical TAL segments with forskolin or cAMP analogs increased the number of Cl − channels observed in basolateral membrane patches. It is now known that NKCC2 is sensitive to intracellular chloride concentration: chloride depletion activates the co-transporter by inducing phosphorylation of key amino terminal domain threonines of the co-transporter by a pathway involving the soluble kinases WNK3 and SPAK. Additionaly, elimination of TAL basolateral chloride channels in Bartter’s type III and IV reduces salt reabsorption by TAL, producing the disease. Thus, one mode of regulation of NKCC2 can be achieved by modulating intracellular chloride concentration in TAL.
Alternatively, ADH might enhance Cl − conductance indirectly by increasing apical membrane Cl − entry. According to this proposal, an ADH-dependent activation of apical membrane Na + -K + -2Cl − co-transporter and K + channels leads to an increase in intracellular Cl − activity, and a subsequent increase in basolateral Cl − conductance. In support of this view, Molony et al. found, also in mouse medullary TAL, that the ADH-dependent rise in cellular and basolateral conductance were much lower when Cl − entry was inhibited by furosemide than in control conditions. Studies of Cl − channels incorporated from basolaterally enriched renal medullary membrane vesicles into planar lipid bilayers have demonstrated several properties of these channels that may account for an intracellular Cl − concentration-dependent rise in basolateral Cl − conductance. First, the activity of the Cl − channels is increased with membrane depolarization evidently due to the basolateral membrane depolarization that follows ADH stimulation of TAL segments. Second, Cl − channels in the vesicles behave like Goldman rectifiers, so that increases in intracellular Cl − cause an increase in the outward single-channel conductance. Third, channel activity is dependent on the intracellular Cl − concentration, such that increases in Cl − over the range 2–50 mM result in large increases in the open time probability of these Cl − channels. In this respect, it should be recognized that the time-averaged conductance of a basolateral Cl − channel is given by the product g Cl P o , where g Cl is unit channel conductance and P o is open time probability.
Thus, an ADH-dependent increase in apical Cl − entry could stimulate basolateral Cl − conductance to a far greater extent than expected from simple Goldman rectification. In that connection, at least two laboratories have found, in mTAL segments, that reducing intracellular Cl − with luminal furosemide produced a three-fold reduction in basolateral Cl − conductance.
Modulation of ADH Effect on TAL Segments
Several factors, such as prostaglandins, peritubular calcium concentration, and peritubular osmolality, have been demonstrated to modulate the actions of ADH on NaCl absorption in the TAL (see the review articles by Hebert and Molony). In isolated mouse and rabbit TAL segments, increases in peritubular osmolality, induced by adding urea or impermeant solutes such as mannitol, rapidly and reversibly inhibit the ADH-stimulated rate of net Cl − absorption. Peritubular hypertonicity results in a prompt reduction in the transepithelial voltage and in the cellular conductance. Molony and Andreoli determined that hypertonicity inhibits the basolateral membrane chloride conductance. The antagonizing effect on ADH-induced stimulation of net Cl transport is not due to decreasing cAMP, because application of either ADH or cAMP is unable to reverse the hypertonicity-mediated effect. Thus, increasing the absolute magnitude of interstitial osmolality provides a negative feedback signal that can reduce ADH-dependent salt absorption by the mTAL.
PGE 2 also plays a role in modulating the effect of ADH on NaCl absorption in the mTAL (see Molony’s review article ). In the mouse mTAL, PGE 2 abolished the effect of ADH on transepithelial voltage and on net NaCl absorption, while it had no effect on NaCl salt absorption in the absence of ADH. Likewise, reported biochemical studies in the mTAL indicate that PGE 2 has no effect on cellular cAMP concentrations in the absence of ADH, and that PGE 2 markedly inhibits the ADH-dependent stimulation of cytosolic cAMP concentrations. The TAL, particularly in the macula densa, expresses COX-2. COX-2 expression in these cells may be coupled to renin secretion. Thus, COX-2 expression is increased by salt restriction, diuretics, and in Bartter syndrome all conditions characterized by hyperreninemia. In addition to PGE 2 , 20-HETE may also modulate the effect of cAMP on Na + transport in the TAL. McGiff and co-workers demonstrated that 20-HETE inhibits NaCl transport in the TAL by a mechanism involving inhibition of the apical K + channel and the Na + -K + -2Cl − co-transporter. Moreover, high 20-HETE concentration inhibits the effect of cAMP on the apical K + channels in the rat TAL.
Hypercalcemia often results in an ADH-resistant urinary concentrating defect, that is, nephrogenic diabetes insipidus. At least part of this concentrating defect results from the inhibition of ADH-stimulated cAMP production in the TAL by calcium. Takaichi and Kurokawa demonstrated that high ambient calcium inhibited cAMP production stimulated by forskolin, indicating that the inhibition probably involved the catalytic subunit of adenylate cyclase. Preincubation of tubule segments with pertussis toxin abolishes the effect of hypercalcemia on cAMP generation, indicating that the inhibition of cAMP generation is mediated through activation of G i . The effects of hypercalcemia on cAMP production are mediated by a G-protein-coupled CaSR present on the basolateral membrane of TAL cells. As discussed above, stimulation of CaSR also inhibits the apical 70 pS and basolateral 50 pS K channels. Consequently, activation of CaSR is expected to alter NaCl absorption, thereby affecting the concentrating mechanism.
Intracellular Mechanisms for NKCC2 modulation in TAL
The physiological pathways for NKCC2 regulation have begun to be uncovered. The rate of salt reabsorption by TAL depends on the amount of NKCC2 protein expressed in the apical membrane at any given time; that is, the protein half-life in the plasma membrane. As shown in Figure 34.5 , the amount of NKCC2 is the result of a dynamic process involving exocytosis and endocytosis mechanisms. Stimulators of NKCC2, such as cAMP (vasopressin, PTH, glucagon or β-adrenergic stimulation) promote exocytosis over endocytosis. Electron microscopy analysis of mouse TAL and biotinylation of rat TAL apical membrane demonstrated that only between 3–5% of total NKCC2 is present in the apical membrane in basal conditions. Vasopressin stimulates trafficking of NKCC2 towards the plasma membrane in Xenopus oocytes, in mouse TAL, and in rat TAL, by a process that is modulated by the vesicle-associated membrane proteins VAMP2 and VAMP3. Exocytosis of NKCC2 is a continuous dynamic process that is further activated by cAMP, via protein kinase A (PKA). Because PKA inhibition does not completely prevent cAMP-induced increase of NKCC2 presence in the apical membrane, it is possible that cAMP also inhibits endocytosis. Two amino acid residues of NKCC2 become phosphorylated by PKA, Ser126 and Ser874, but the exact consequence of this is not known. Endocytosis of NKCC2 occurs also constitutively in TAL. A fraction recycles toward the apical membrane and another fraction undergoes degradation. Inhibition of endocytosis by cholesterol depletion completely block NKCC2 retrieval from the membrane, increasing NKCC2 presence and chloride transport by TAL. Supporting this view, it has been shown that part of NKCC2 is associated with lipid rafts.