Increasing evidence suggests that the ATP/P2 receptor system acts in an autocrine or paracrine fashion to affect various aspects of renal function. P2 receptors have been identified in most renal vessels and nephron segments; ATP is released from renal epithelial cells; and enzymes responsible for ATP breakdown are expressed in the vasculature and tubules. Stimulation of P2 receptors in the afferent arterioles by ATP released from renal nerve endings or from adjacent macula densa cells induces vasoconstriction and contributes to the control of renal haemodynamics. In the tubule, there is evidence for a variety of P2-mediated effects: inhibition of proximal tubular reabsorption; inhibition of Na + ,K + ,2Cl − cotransporter activity (through increased nitric oxide synthesis) in the thick ascending limb of the loop of Henle; inhibition of magnesium reabsorption in the distal tubule; and modulation of sodium and water reabsorption in the collecting duct. Finally, P2 receptors, particularly P2X subunits, appear to play an important role in renal pathology, specifically in cyst formation in polycystic kidney disease and in some forms of renal injury and inflammation.
kidney, purinergic, purinoceptor, ATP, nitric oxide, nephron, ectonucleotidase, cyst, inflammation
Historically, the control of renal vascular and tubular function has been attributed solely to neural and endocrine regulation. However, in addition to these extrinsic factors, it is now recognized that several complex humoral control systems exist within the kidney that act in an autocrine and/or paracrine manner. One of these is the extracellular nucleotide/P2 receptor system.
Although physiological actions of extracellular adenine nucleotides were reported as long ago as 1929, it was not until many years later (1972) that the importance of ATP as a transmitter for non-adrenergic, non-cholinergic neurones of the autonomic nervous system was proposed by Geoffrey Burnstock. Since then it has become apparent that the function of extracellular nucleotides is not confined to neurones: rather, they are ubiquitous autocrine/paracrine agents regulating diverse physiological processes in almost every tissue in the body. Information on their roles in the kidneys has only really begun to emerge in the last decade.
Extracellular nucleotides exert their effects by binding to and activating cell surface located receptors; P2 receptors. These are subdivided into P2X receptors, of which seven mammalian subunits have been cloned (P2X 1–7 ), and P2Y receptors, of which eight mammalian subtypes are currently recognized, P2Y 1, 2, 4, 6, and P2Y 11–14 .
P2X receptor subunits are proteins with two transmembrane-spanning regions, the N- and C-termini being within the cell. Three P2X subunits assemble to form a P2X receptor ion channel that, when activated, is permeable to small cations (Na + , K + , Ca 2+ ). Each of the seven P2X subunits can make homomeric ion channels, and can also form heteromeric assemblies involving more than one type of subunit. Until recently it had been thought that P2X 7 subunits could only make homomeric assemblies, but a P2X 4/7 heteromer has now been described. As well as a non-selective ion channel, the P2X 7 receptor can form a larger membrane pore, and initiate cell death by necrosis or apoptosis.
The principal natural ligand for all P2X subunits is ATP. The P2X 1 subunit is the most sensitive (requiring sub-micromolar concentrations of ATP); P2X 2–6 subunits require micromolar concentrations, while the P2X 7 subunit is easily the least sensitive, requiring almost millimolar concentrations.
P2Y receptors are G-protein-coupled receptors with seven transmembrane-spanning regions; the C-terminus is inside the cell and the N-terminus extracellular. In rodents, ATP is probably the principal natural ligand for P2Y 2, 4, and P2Y 11 subtypes and, at sufficiently high dose and/or receptor density, can activate P2Y 1, 12, and P2Y 13 subtypes. However, the natural ligand for P2Y 1, 12, and P2Y 13 subtypes is ADP. Although P2Y 6 receptors can also be activated by ADP, UDP is much more potent. In rodents, UTP activates P2Y 2 and P2Y 4 subtypes with similar potency to ATP, an observation often used in physiological studies as an initial pointer to receptor identity. Human P2Y 4 receptors, however, are activated primarily by UTP (50-fold more potent than ATP ). This is a particularly striking example of species differences, serving to highlight the need for caution before extrapolating from findings in one species to another. The P2Y 14 receptor is exceptional in that its natural ligand is UDP-glucose; although originally believed to be unaffected by unglycosylated purine- or pyrimidine-based nucleotides, it is now known that UDP is also a full agonist at rat and human P2Y 14 receptors.
P2Y receptors are coupled to either G q or G i signaling proteins. P2Y 1, 2, 4, 6, and P2Y 11 subtypes are coupled to G q /G 11 , resulting in PLC-β activation and increased [Ca 2+ ] i , while P2Y 12–14 are coupled to G i /G o , resulting in adenylyl cyclase inhibition and reduced cAMP levels. The P2Y 11 subtype is unusual, in that it can couple to both G q and G s , resulting in both PLC-β and adenylyl cyclase activation, causing increased cAMP levels.
Heterodimeric Receptors and Dinucleotide Receptors
A further layer of complexity has been added to the picture with the finding that adenosine A1 receptors can be co-expressed with P2Y 1 or P2Y 2 receptors (and possibly other P2Y subtypes) as a discrete receptor type, at least in non-renal cells. The chimeric nature of such receptors is reflected in their mixed pharmacological and signaling properties. The possible functional significance of these heterodimeric P1/P2Y receptors with regard to the kidneys is currently unknown.
Finally, a number of dinucleotides, in which the 5′-carbon positions of two nucleosides are linked by a polyphosphate chain, occur naturally in the body. These dinucleotides can be symmetrical (e.g., Ap 4 A, where two adenosine moieties are linked by a chain of four phosphates) or asymmetrical (e.g., Up 4 A, where a uridine moiety and an adenosine moiety are similarly linked). Dinucleotides can have both vascular and tubular effects within the kidneys ( vide infra ), but the receptors responsible are unknown; evidence for dinucleotide-specific receptors has been provided in other tissues, but several P2Y receptors (P2Y 1, 2, 4, and P2Y 6 ) and P2X receptors (P2X 1–5 ) are known to be dinucleotide-sensitive.
It is likely that both adenine-based and uracil-based nucleotides are released from most cells in the body (including renal cells); moreover, ecto-enzymes that metabolize nucleotides, either inactivating them or converting them to molecular forms that can stimulate different P2 receptor subtypes, are ubiquitous ( vide infra ). Figure 18.1 shows the molecular structures of some of the principal nucleotides involved, and Figure 18.2 provides a simplified overview of nucleotide release, degradation, and purinoceptor (i.e., P1 (adenosine) receptor and P2 receptor) activation.
Synthetic Agonists and Antagonists of P2 Receptors
An ever-increasing range of synthetic nucleotide analogs and non-nucleotide agonists is being developed in an attempt to find agents that, unlike naturally occurring nucleotides, are not subject to degradation by ectonucleotidases, and can act as selective agonists for given receptor subtypes. Such exclusivity is rarely achieved, although substantial progress is now being made. Unfortunately, many of the initial observations on P2 receptor stimulation and renal function were made at a time when information on the selectivity of agonists was incomplete, and the agonists used were often more promiscuous than was appreciated, giving rise to misleading interpretations. Thus, although 2 meSADP, for example, has been used as an agonist for P2Y 1 receptors, it also activates P2Y 12 and P2Y 13 subtypes; the same applies to 2 meSATP which, additionally, can stimulate a number of P2X receptors, while ATPγS, originally used as a P2Y 2 and/or P2Y 4 agonist, is now known as a broad-spectrum agonist, being effective in a range of P2Y and P2X receptors. Another ATP analog, 2′3′- O -(4-benzoylbenzoyl)ATP (BzATP), has often been used as a “selective” P2X 7 agonist, given that it is more potent than ATP at this receptor subunit, but it is also effective at P2X 1, 3 and P2Y 5 subunits, so it is in reality only a non-selective P2X agonist. Furthermore, BzATP has been shown to act as an antagonist at P2Y 4 receptors.
As our knowledge of truly selective P2 agonists expands, future investigations should provide more precise information about the purinoceptor subtype(s) involved in a given physiological response. That knowledge, however, is still limited. The N -methanocarba-ADP derivative MRS2365 is selective for P2Y 1 receptors; MRS2698 and INS365 (Up 4 U or “diquafosol”) are selective P2Y 2 agonists; and UDPβS, INS48823, and MRS2693 are selective P2Y 6 agonists. At the time of writing, a selective agonist for P2Y 4 receptors has not been identified. Similarly, a truly selective agonist for any of the P2X subunits is still lacking.
As with agonists, nucleotide receptor-selective antagonists are something of a rarity. Probably the compound most commonly used to inhibit P2 receptors is suramin, although it also affects a variety of other cellular processes. In sufficient concentration, suramin antagonizes practically every P2 receptor subtype, be it P2Y or P2X. The same comment applies to PPADS (pyridoxal-5-phosphate-6-azophenyl 2′,4′-disulphonic acid) and, to a lesser extent, reactive blue 2 (RB-2). However, a clutch of selective and potent antagonists is now available. Thus, the ADP derivatives MRS2179, MRS2279, and MRS2500 are selective P2Y 1 antagonists; AR-C126313 and AR-C118925 are selective P2Y 2 antagonists; MRS2578 is a selective P2Y 6 antagonist; INS49266, INS50589, and AZD6140 are selective P2Y 12 antagonists; and MRS2211 is a selective P2Y 13 antagonist.
For P2X subunits, the list is shorter. Ip 5 I is a selective P2X 1 antagonist, and A-740003 and A-438079 are selective P2X 7 antagonists. Trinitrophenyl-ATP (TNP-ATP) “selectively” antagonizes P2X 1–5 subunits without affecting P2Y receptors.
Assignment of Physiological Responses to Specific P2 Receptor Subtypes
The plasma membranes of any renal cell, be it vascular or tubular, can contain a variety of P2 receptor subtypes. Moreover, epithelial cells can have different (as well as the same) subtypes on their apical and basolateral membranes. This raises the question of how to attribute a given functional response to a particular subtype. A number of approaches can be used. First, it is useful to identify immunologically the subtypes present in the region of interest (although this, of course, depends on the availability of suitable antibodies) and, if possible, to localize the receptor to apical and/or basolateral membrane. In some cases, instead of the immunohistochemical approach, determination of mRNA has been used, although this obviously does not guarantee the presence of the receptor protein itself. Second, it is possible to try to mimic the effect of the naturally occurring nucleotide using “selective” agonists and antagonists. However, as indicated above, only a few of these are truly selective (although the situation is improving). Consequently, it is usually necessary to compare the individual responses to a variety of agonists (both natural and synthetic) to provide a pharmacological profile from which tentative conclusions can be drawn, but even then their effects will depend not only on agonist/antagonist concentration, but also on the number and distribution of receptor subtypes. Moreover, naturally occurring agonists are degraded by ectonucleotidases, making it difficult to control their absolute concentrations at the receptor site. A further limitation is the use of intracellular Ca 2+ transients to assess responses to direct application of agonists, since these are not invariably associated with recognizable functional changes.
A completely different, and superficially more attractive, approach is to use “knockout” mice in which the gene encoding the receptor of interest has been deleted. However, this is not without its own potential problems. Life-long, global deletion of a receptor subtype that performs a vital function is likely to lead to compensatory changes in several organ systems. The P2 receptor profile within the kidney may then change in order to restore overall excretion rates, which could then lead to misleading conclusions about the role of the receptor. The Cre-loxP system adds a degree of refinement to the gene-targeting approach, permitting tissue- or cell-type-specific deletion. Nevertheless, compensatory changes in up- or downstream nephron segments cannot be excluded, and this approach is further complicated by incomplete (knockdown rather than knockout) and off-target deletion. Furthermore, genetically-engineered deletions have so far been restricted to mice, where P2Y 2 receptors seem to predominate in the renal tubule. There are important differences in the distributions of receptor subtypes between mice and rats – and presumably between mice and other species. Thus, for the foreseeable future it seems that we will need to continue to rely on a combination of approaches; as yet, there is no “silver bullet” when it comes to defining P2 receptor function.
Finally, to obviate the need for working with complex renal tubules, many investigators have made use of simpler systems: immortalized cell lines originally derived from renal-like tissue (e.g., Madin–Darby canine kidney (MDCK) cells). Unfortunately, these cell lines often express membrane proteins that differ from those found in native tissue. Consequently, in this chapter we will avoid deductions based solely on observations concerning P2 receptors in non-native renal tissue.
P2 Receptors and Renal Function
The Renal Vasculature
Figure 18.3 summarizes current knowledge about the distribution of P2 receptors in renal vascular and tubular structures. P2 receptors are expressed widely in the renal vasculature, in the glomerulus, and in the extraglomerular mesangium. Immunohistochemical and Western analyses indicate that P2X 1 receptors are present in the vascular smooth muscle of the rat renal artery, arcuate and interlobular arteries, and the afferent arteriole, but not in the efferent arteriole. Functional approaches have confirmed the expression of a P2X 1 -like receptor in the afferent arteriole. P2X 2 subunits have been immunolocalized in the smooth muscle of larger arteries and veins within the kidney, and molecular evidence has recently been provided for P2X 4 subunits, at least in arcuate and interlobular arteries. Of the P2Y receptors, P2Y 1 has an extensive distribution, being expressed in the endothelium of the large arteries, and both afferent and efferent arterioles.
Most information concerning P2 receptor expression in the glomerulus comes from cell culture systems. On the basis of mRNA detection and/or agonist profiling, P2Y 1,2,4, and P2Y 6 subtypes and P2X 2,3,4,5, and P2Y 7 subunits have been identified in glomerular mesangial cells ; P2Y 1,2 and P2Y 6 subtypes in podocytes ; and P2Y 1 and P2Y 2 subtypes in glomerular endothelial cells. Studies performed on RNA extracted from pools of intact glomeruli from rats found messages encoding P2Y 1,2,4 and P2Y 6 subtypes ; the expression of other P2Y receptors was not assessed. Immunohistochemical analysis and measurements of agonist-induced phosphoinositide production confirmed the presence P2Y 1 and P2Y 2 subtypes in the rat glomerulus. On the basis of co-localization with cell-specific markers, P2Y 1 receptors were localized in mesangial cells and P2Y 2 receptors in podocytes ; expression of P2Y 4 and P2Y 6 receptor protein could not be confirmed, either functionally or immunologically. Of the P2X subtypes, only a low and variable expression of P2X 7 immunoreactivity was found in the rat glomerulus.
Infusion of ATP into the renal artery has long been known to alter renal vascular resistance, although the nature and magnitude of the response are dependent upon species, basal vascular tone, and to some extent the experimental approach. The larger renal arteries serve principally as conductance vessels, and renal vascular resistance (and therefore renal blood flow) is regulated primarily through pressure-dependent vasoactivity of the preglomerular arterioles and, to a lesser extent, the small interlobular arteries. The responsiveness to ATP of the arcuate and interlobular arteries and the glomerular arterioles has been evaluated in rats using the isolated perfused kidney preparation. The preglomerular arteries were relatively insensitive to ATP, with micromolar concentrations evoking transient vasoconstriction ( Figure 18.4 ). In contrast, the afferent arteriole underwent sustained contraction at concentrations in the submicromolar range, whereas the efferent arteriole was unresponsive to extracellular ATP. Thus, in the isolated perfused rat kidney, intrarenal administration of ATP is normally vasoconstrictive. This vasoconstriction can be potentiated by inhibition of nitric oxide (NO) synthesis. However, when baseline renal vascular resistance is high, ATP induces vasodilatation , due to P2Y-mediated production of NO. Thus, P2 receptor “tone” can influence renal vascular resistance, with P2Y/NO-mediated vasodilatation exerting a counterbalancing influence on P2X 1 -mediated vasoconstriction. The dominant receptor pool, as well as the source and local concentration of extracellular nucleotide, will therefore influence the net physiological response to extracellular nucleotides. ATP released from renal nerve terminals, for example, will act directly on the vascular smooth muscle, promoting P2X 1 -mediated vasoconstriction. Conversely, release of ATP in the vicinity of the endothelial P2Y receptors would be expected to promote NO synthesis and vasodilatation.
The majority of vascular beds stabilize blood flow in the face of fluctuating blood pressure. This autoregulation of blood flow is an intrinsic property of the vasculature, and in the kidney it is highly efficient: over the physiological range, renal blood flow can be effectively independent of blood pressure. Whole-kidney autoregulation is governed through the combined influence of at least two mechanisms, tubuloglomerular feedback (TGF) and the intrinsic myogenic response of the vascular smooth muscle; these regulatory systems have different, but overlapping, operational frequencies. Computational analysis of the dynamic frequencies of the two systems indicates a degree of interaction; constriction of the terminal afferent arteriole by TGF increases pressure in the upstream vasculature and the myogenic response is enhanced.
Myogenic Responses to Altered Perfusion Pressure
Of the two major components, only the intrinsic myogenic response to altered perfusion pressure is both necessary and sufficient for full, whole kidney autoregulation. The myogenic response operates along the preglomerular vascular tree, reacting to increased transmural pressure by channel-mediated calcium influx and reflex vasoconstriction of the vascular smooth muscle. The exact signaling mechanisms are not defined, but local release of ATP is implicated. In the afferent arteriole, pressure-mediated vasoconstriction is markedly blunted by PPADS or suramin, or by the saturation and subsequent desensitization of the P2 receptor system. The central role of the P2 system is further underscored by experiments in P2X 1 -deficient mice, in which pressure-induced reductions in afferent arteriole diameter are abolished ( Figure 18.5 ). Similarly, pharmacological or pathological maneuvers that impair P2X 1 receptor signaling significantly attenuate whole kidney autoregulation of blood flow both in vivo and in vitro . Furthermore, mice with a targeted deletion of the ectonucleotidase NTPDase1 (thereby prolonging the half-life of extracellular ATP – vide infra ) exhibit enhanced pressure-induced vasocontriction in the mesenteric artery, consistent with a key role for local nucleotide signaling in the general myogenic response.
Tubuloglomerular Feedback and the Juxtaglomerular Apparatus
Tubuloglomerular feedback (TGF) is a dynamic process whereby changes in the concentration of NaCl in the fluid emerging from the loop of Henle elicit inverse changes in the glomerular filtration rate of the nephron of origin. TGF is mediated by the juxtaglomerular apparatus (JGA), which includes a sensor, the macula densa, and an effector, the granulated cells of the afferent arteriole; other components of the JGA (e.g., mesangial cells) also play a role.
Bell and colleagues demonstrated the release of ATP across the basolateral membrane of the macula densa plaque in response to altered luminal NaCl concentration within the physiological range, and the concentration of ATP in the cortical interstitium was shown to respond appropriately to inhibition or activation of TGF in vivo . This compelling evidence suggests that ATP is the primary signaling molecule for TGF, the effects of which might well be modulated by other paracrine agents produced in the macula densa cell, such as nNOS- (NOS1-) derived NO and COX2-derived prostaglandin E 2 . Gene-targeting experiments, however, suggest that ATP may not be the ultimate signal through which TGF causes constriction of the afferent arteriole; hydrolysis of ATP to adenosine appears to be critical. In vivo TGF responses are attenuated in mice lacking either adenosine A 1 receptors or ecto-5′-nucleotidase, the enzyme catalyzing the final stage of the degradation of ATP to adenosine. This proposition is supported by a recent in vivo study in which the TGF response in mice (as assessed by changes in stop-flow pressure in the proximal tubule) was unaffected during intravenous infusion of PPADS or suramin.
It would be rash, however, to conclude that the P2 receptor system has no role in TGF. Desensitization of this system inhibits TGF in rats, and it is notable that in vivo manipulations of TGF affect the interstitial concentration of ATP, but not of adenosine. Furthermore, an anatomical consideration argues strongly for involvement of the P2 receptor system in the TGF response: the ATP released from macula densa cells cannot directly activate P2 receptors in the afferent arteriole, being physically separated in most species by the extraglomerular mesangium. An intact mesangium is required for TGF responses, and Peti-Peterdi has demonstrated that TGF activation causes a wave of increased cytosolic calcium to pass through the mesangium, to the granulated cells of the afferent arteriole and into the glomerular podocytes. Propagation of this calcium wave was abolished by suramin, but not by adenosine receptor antagonism. The P2 receptor response was later shown to be dependent on gap junctional coupling, being inhibited by specific antagonists against connexins 37 and 40 ( vide infra for discussion of connexins).
The basolateral membrane of macula densa cells expresses a P2Y 2 -like receptor, the function of which is not yet known. It is possible that this provides a negative feedback loop for TGF signaling or a mechanism through which ATP release can be coupled to production.
Glomerular and Medullary Microcirculation
Infusions of nucleotide analogs into the renal artery exert powerful effects on regional blood flow, which can be measured by laser-Doppler flow probes inserted into specific regions of the kidney. In the rabbit, ATP evokes a biphasic response, with vasoconstriction of the medullary blood flow being followed by hyperemia. On the basis of relative agonist potency, the vasoconstriction was attributed to P2X 1 receptors; the secondary vasodilatation, which was independent of NO, was partially mediated by adenosine receptors. In the rat, the net effect of ATP is influenced by sodium status. In sodium-restricted rats, ATP increased medullary blood flow in a nitric oxide-dependent manner. In rats fed a high-salt diet, ATP caused vasoconstriction in the outer medulla, without affecting inner medullary flow. The authors’ speculated that the inner medullary vasodilatation reflected an effect of nucleotides on vasa recta pericytes. However, preliminary data from Peppiatt-Wildman’s laboratory, obtained in slices of rat kidney, suggest that P2 receptor activation promotes vasoconstriction in this setting.
The renin–angiotensin system is influenced by many factors, the final pathways of which converge at the level of altered [Ca 2+ ] i in the granular cell; renin secretion is inversely related to [Ca 2+ ] i . The combined use of receptor-selective agonists and antagonists has demonstrated that A 1 receptors exert a tonic inhibitory effect on renin secretion at the level of the granular cell. However, the adenosine receptor system is not vital for the control of renin secretion, since A 1 receptor knockout mice are able to raise their secretion appropriately in response to a low-salt diet, a regimen that increases two-fold the sensitivity to luminal NaCl of ATP release by the macula densa.
The role of the P2 receptor system in the regulation of renin release is not entirely clear, and is to some extent contradictory. Purinoceptor signaling is a prerequisite for synchronization of the intercellular calcium wave that controls renin secretion in the JGA, and infusion of ATP into the isolated perfused rat kidney causes profound inhibition of renin secretion. On the other hand, activation of an ADP-selective receptor, thought to be P2Y 1 , was shown to stimulate renin secretion in rat renal cortical slices via a nitric oxide-dependent mechanism. Resolution of this issue awaits further investigation.
Immunohistochemical studies have identified apical expression of P2Y 1 and P2X 5 receptors in the S3 segment of the rat pars recta, and basolateral expression of P2Y 4 and P2X 6 receptors in the proximal convoluted tubule (PCT); low-level expression of P2X 4 protein was also seen in the PCT, although the membrane domain was not identified. Western blot analysis has additionally shown the presence of P2Y 1 receptors in brush-border membrane vesicles from the S2 segment of rat PCT (see Figure 18.3 ). Messenger RNA expression has been assessed for only four P2 receptor subtypes: P2Y 1, 2, 4 and P2Y 6 are all present in rat proximal tubule. In terms of Ca 2+ transients following application of P2 receptor agonists of varying selectivity, supportive evidence has been provided for apical P2Y 1 -like receptors in an immortalized cell line with a proximal phenotype, and for basolateral P2Y 1 receptors in native rat PCT ; Bailey and colleagues also reported that basolateral UDP was effective in increasing [Ca 2+ ] i , corroborating the presence of P2Y 6 receptors. Finally, ATP and UTP were equally effective in stimulating Ca 2+ transients when applied to rat or rabbit basolateral membranes, implying mediation by P2Y 2 or P2Y 4 receptors; the immunohistochemical evidence in rats favors P2Y 4 receptors.
Using a stationary microperfusion technique in rat PCT in vivo , Bailey showed that addition of adenosine nucleotides to the lumen inhibited bicarbonate reabsorption. ADP was more effective than ATP, implicating P2Y 1 receptors; this was supported by the observation that the P2Y 1 agonist 2 meSADP also had a potent inhibitory effect, which was blocked by the P2Y 1 -selective antagonist MRS2179. (When the tubule was perfused with MRS2179 alone, a small increase in bicarbonate reabsorption was seen, suggesting a tonic inhibitory effect of endogenous nucleotides acting via P2Y 1 receptors.) The P2Y 1 -mediated effect on bicarbonate reabsorption involved inhibition of the Na + /H + exchanger NHE3, since it was not additive to that of EIPA. The effect was blocked by either U73122 or H89, indicating involvement of phospholipase C and protein kinase A. In apparent contrast to these findings from intraluminal perfusions, Diaz-Silvester et al. found that addition of ATP to peritubular capillaries perfused in vivo caused an increase in transepithelial bicarbonate reabsorption in rat PCT. Conceivably, given the presence of ectonucleotidases in peritubular capillaries and the peritubular space ( vide infra ), degradation of ATP through to the nucleoside adenosine (which stimulates proximal tubular bicarbonate reabsorption ) could not be ruled out. However, increasing the viscosity of the peritubular perfusate also stimulated bicarbonate reabsorption, and this effect was blocked by peritubular suramin, suggesting P2 receptor mediation. (Shear stress was proposed as the activating factor.) Interestingly, the increase in bicarbonate reabsorption induced by ATP or by raised viscosity could be blocked by a nitric oxide synthase inhibitor.
In a preliminary study of membrane transporters in the tubules of P2Y 2 receptor knockout mice, Listhrop et al. reported increased expression of NaPT2 protein in the proximal tubule (but no change in NHE3 abundance). In line with this, ATP has been shown to inhibit phosphate uptake (and mRNA for NaPT2) in primary cultures of rabbit PCTs. Interestingly, in the same preparation, ATP stimulates sodium-glucose co-transport by increasing both SGLT1 and SGLT2 protein expression.
A renal clearance study in rats, using lithium clearance as an index of end-proximal tubular fluid delivery, reported remarkable effects of the naturally occurring diadenosine polyphosphate Ap 4 A. When infused intravenously, Ap 4 A increased lithium clearance almost two-fold, despite a fall in GFR, indicating a profound reduction in fractional proximal tubular reabsorption. Although a fascinating observation, it is debatable whether intravenous infusion of relatively high-dose exogenous nucleotide provides physiologically useful information about normal autocrine/paracrine control by endogenous agents. It is also difficult to know which P2 receptor(s) is/are involved, since Ap 4 A can stimulate a number of subtypes, including P2Y 1 and P2Y 4 receptors, which are both expressed in the rat proximal tubule (P2Y 1 apically, P2Y 4 basolaterally); intravenous delivery of the agonist does not allow differentiation between these possibilities.
In addition to effects on proximal tubular transport, both adenine-based and uracil-based nucleotides can stimulate gluconeogenesis, an important metabolic function of this nephron segment. Diadenosine polyphosphates also have this effect. As these experiments were performed using tubule suspensions or isolated tubules, the agonists will presumably have gained access to both apical and basolateral membranes; moreover, ectonucleotidase-mediated metabolism of the nucleotides is a possibility, hindering identification of the receptor subtype(s) responsible. However, ATP and UTP were equipotent in stimulating gluconeogenesis, implicating P2Y 2 or P2Y 4 receptors. Although these authors plumped for P2Y 2 mediation, the fact that P2Y 2 receptors have not been found in rat proximal tubules, whereas P2Y 4 receptors have ( vide supra ), makes a basolateral P2Y 4 -mediated effect more likely.
Loop of Henle
The pars recta (also called the thick descending limb of Henle), has been dealt with in the preceding section. In the rat thin descending limb there is some immunohistochemical evidence for P2X 4 and P2X 6 receptors (membrane domain not stated ), and indications, from measurements of [Ca 2+ ] i transients during superfusion of isolated segments with various agonists, of a basolateral pyrimidine receptor, although no P2Y 2 or P2Y 4 protein has yet been identified. Messenger RNA is expressed for P2Y 1 and P2Y 6 , but again immunohistochemical evidence of receptor protein is lacking. In the rat thin ascending limb, similar evidence for a basolateral pyrimidine receptor is in this case accompanied by immunohistochemical confirmation of (intracellular) P2Y 2 receptor protein; low-level P2X 4 and P2X 6 protein expression has also been reported. Hardly surprisingly, given the paucity of information regarding normal transport processes in these nephron segments, the functional significance of P2 receptors in the thin limbs is unknown.
Consideration of P2 receptors in the thick ascending limb (TAL) of the loop of Henle must take account of species differences between rat and mouse. In the rat TAL, binding sites for ATPγS are present on the basolateral membrane ; this agonist stimulates several P2Y and most P2X subtypes. Immunohistochemistry has identified P2Y 2 (intracellularly), P2X 4 , and P2X 6 (membrane domain not stated) receptor proteins. In addition, mRNA is expressed for P2Y 1,2,4, and P2Y 6 subtypes. It has been reported that rat TAL segments are poorly responsive to basolateral application of nucleotides, at least in terms of Ca 2+ transients, which is in marked contrast to the situation in mice, where basolateral ATP and UTP were each found to evoke large Ca 2+ transients, consistent with activation of P2Y 2 receptors. A major role for P2Y 2 receptors in mice was supported by Jensen et al. who showed, in mouse medullary TAL (mTAL) perfused in vitro , that luminal application of ATP or UTP caused almost identical increases in Ca 2+ i and, importantly, that these increases were absent in P2Y 2 knockout mice. In the same study, however, significantly different results were obtained when the nucleotides were applied basolaterally: both ATP and UTP caused an initial peak in [Ca 2+ ] i followed by a sustained plateau, but whereas both phases were virtually abolished in mTAL from P2Y 2 knockout mice treated with UTP, the plateau phase in mTAL from P2Y 2 knockout mice treated with ATP persisted, suggesting the presence of an additional basolateral P2 receptor. Since the plateau phase was dependent on extracellular Ca 2+ , the authors proposed a Ca 2+ -permeable P2X receptor.
Some information, albeit circumstantial, on the effects of P2Y 2 receptors on transport processes in the TAL has come from a comprehensive study of the renal phenotype of P2Y 2 knockout mice by Vallon’s group. These animals were shown to exhibit increased expression of the apical Na + K + 2Cl − co-transporter (NKCC-2) in mTAL, associated with an increased natriuretic response to furosemide. The obvious implication is that nucleotide activation of P2Y 2 receptors inhibits NaCl transport in mouse TAL.
During the last decade, in a series of meticulously controlled experiments using cell suspensions or nephron segments perfused in vitro , Garvin’s group has begun to piece together evidence for a functional role of ATP on TAL function in the rat. Using suspensions of rat mTAL, Silva et al. showed that ATP increased intracellular NO production in a concentration-dependent manner, and that the response was significantly inhibited by suramin. Although the EC 50 value for the NO response to ATP was high, at 37 µM, prevention of ATP hydrolysis by administration of the ectonucleotidase inhibitor ARL67156 reduced the EC 50 to 0.8 µM. On the basis that the ATP analog βγmeATP caused an increase in NO production, it was argued that the response was mediated primarily by P2X receptors, although it was noted that UTP also had a weak effect. A recent study from the same group has provided some insight into the signaling cascade involved in ATP-stimulated NO production. Confirmation that endothelial nitric oxide synthase (eNOS; or NOS3) is the enzyme responsible came from the finding that ATP was unable to stimulate NO production in TAL cells from NOS3 knockout mice, whereas a normal response was seen in wild-type mice given NOS1- or NOS2-selective inhibitors. The PI 3 kinase inhibitor LY294002 caused a major reduction in the response to ATP, and a similar reduction was seen in the presence of an Akt-selective inhibitor. This finding, together with the observation that ATP stimulated Akt1 (serine threonine kinase; also called protein kinase B) phosphorylation, whereas phosphorylation of Akt2 and Akt3 was either unchanged or reduced, led the authors to conclude that ATP increases NOS3-derived NO via activation of Akt1. A possible functional link between the recently observed effect of increased flow on nucleotide release in the TAL ; ( vide infra ) and the production of NO can be drawn from an earlier study by Ortiz et al., in which increasing flow (in the physiological range) in isolated perfused TALs from rat caused markedly increased NO production (which was all but blocked by the NOS inhibitor L-NAME). Notably, increasing luminal flow caused a redistribution of NOS3 within the TAL cells, with translocation towards the apical membrane. It was already known that NO (and presumably therefore ATP) can reduce TAL transport by inhibiting NKCC-2 activity and (to a lesser extent) Na + /H + exchange, but a recent study from this group explored the possibility that ATP might have a primary effect in reducing basolateral Na + K + -ATPase activity, and thereby Na + extrusion from the TAL cell. Using rat mTAL suspensions, it was found that ATP reduced oxygen consumption in a dose-dependent manner and that this was blocked by suramin, but not by the adenosine receptor antagonist theophylline; it was also blocked by the NOS inhibitor L-NAME. The “2P2X-selective” agonist βγmeATP also reduced oxygen consumption concentration-dependently, while the “P2X-selective” antagonist NF023 blocked ATPs action. (However, as with nucleotide-stimulated NO production, it was found that UTP had a (weak) inhibitory effect on oxygen consumption, suggesting some P2Y involvement.) When NKCC-2 and Na + /H + exchange were blocked with a combination of furosemide and dimethyl amiloride, oxygen consumption fell, and was no longer affected by ATP; while the Na + ionophore nystatin increased oxygen consumption to a similar extent in TALs treated with ATP or vehicle alone. These experiments provide powerful confirmatory evidence that ATP, by increasing NO production, can inhibit Na + transport in the rat TAL, not by inhibiting basolateral Na + K + -ATPase activity, but principally by reducing apical Na + entry, particularly via the Na + K + 2Cl − co-transporter. Whilst this series of in vitro findings is strongly suggestive of a physiological role for nucleotides in autocrine/paracrine control of TAL function (a putative schema is shown in Figure 18.6 ), a full assessment awaits a comprehensive investigation of electrolyte transport in the loop of Henle in vivo .
Little is known about P2 receptor distribution in this nephron segment (which, here, we arbitrarily define as distal convoluted tubule (DCT) plus connecting tubule (CNT); the final segment of the properly defined distal tubule – the initial collecting tubule – will be included under “Cortical Collecting Duct”). Immunohistochemical studies have identified P2X 4 and P2X 6 receptors on the basolateral membrane in rat distal tubule (although it is not clear which region of the distal tubule was involved, as no markers of cell types were used in this study), and basolateral application of ATP to microdissected rat DCTs resulted in (weak) Ca 2+ transients, but no corresponding investigations have been made in other species or in CNT. Furthermore, no direct studies of distal tubular transport function have been made in native tissue, either in vitro or in vivo ; the only information we have from whole-animal studies is that thiazide-sensitive sodium excretion is unaffected in P2Y 2 knockout mice, corresponding with a lack of change in Na + -Cl − co-transporter (NCC) abundance. Consequently, our knowledge of the role of P2 receptors in these nephron segments is fragmentary and largely restricted to findings from studies of primary cultures of native cells or immortalized distal or “distal-like” cell lines.
Considerable evidence exists for the expression of a number of P2 receptor subtypes and for a range of P2-mediated actions in Xenopus A6 cells and Madin–Darby canine kidney (MDCK) cells, both widely used as “distal-like” cell lines. However, as indicated above, for reasons of physiological relevance it will not be considered here, as such cell lines often express membrane proteins different from those found in native tissue. Similar considerations may apply to immortalized cell lines derived from DCT: activation of apical receptors, characterized pharmacologically as P2Y 2 subtype, in immortalized rabbit DCT was shown to increase apical chloride conductance ; while activation of receptors (membrane domain not stated) characterized pharmacologically as P2X, rather than P2Y, in immortalized mouse DCT was shown to inhibit magnesium reabsorption. Finally, cultured cells from rabbit CNTs responded to extracellular ATP with an increase in [Ca 2+ ] i and inhibition of sodium and calcium absorption, although these inhibitory effects were not dependent on the Ca 2+ transient. Either apical or basolateral application of ATP was effective and, when ATP was added to both compartments, the inhibitory effects were additive. On the basis of pharmacological profiling, P2Y 2 receptors were implicated.
It is difficult to build a coherent picture from these disparate findings. As with the loop of Henle, a comprehensive in vivo assessment of distal tubular function is required.
A large array of P2 receptor subtypes has been reported in rat collecting duct (CD) ( Figure 18.3 ). Immunohistochemistry has indicated the expression of P2Y 2 , P2Y 4 , P2Y 6 , P2Y 11 , P2Y 12 , and P2Y 13 subtypes, and P2X 1 (sodium-restricted rats only, intercalated cells only), P2X 2 , P2X 4 , P2X 5 , and P2X 6 subunits with, in some cases, differential expression in the different subsegments of the CD. With respect to membrane localization in principal cells, P2Y 4,6, and P2Y 11 were reported to be exclusive to the apical membrane, whereas P2Y 2 and P2X 4 and P2X 6 were found in both apical and basolateral membranes; staining for P2Y 12 and P2Y 13 subtypes and P2X 2 and P2X 5 subunits was designated “intracellular”. In the mouse, immunohistochemistry has localized P2X 1 and P2X 4 subunits to the apical membrane of medullary CD cells.
Expression of P2 receptor mRNA in the rat kidney broadly agrees with the immunohistochemical findings – at least in the one study in which both methodologies were used. Messenger RNA has been identified for P2Y 1,2,4, and P2Y 6 subtypes, and P2X 4 subunits in cortical (CCD) and outer medullary CD (OMCD), and for P2Y 1,2,4 and P2Y 6 subtypes in inner medullary CD (IMCD). Additionally, mRNA for P2X 1 and P2X 6 receptors has been reported in CCD and OMCD following dietary sodium restriction. Messenger RNA levels for P2Y 11,12,13, and P2Y 14 receptors have not yet been investigated. Studies using mice have so far focused on P2X receptors, and have identified mRNA for P2X 1,4,5,6, and P2X 7 subunits in CCD and OMCD, suggesting a species difference. As far as the human kidney is concerned, the only published information we have comes from a heroic study by Charbardès-Garonne and co-workers, in which the transcriptome from human kidneys was characterized using serial analysis of gene expression (SAGE). They found that, of tags for 258 genes conferring transport properties, the only P2X receptor detected in significant amounts in the CD was P2X 4 .
A complex picture is beginning to emerge concerning the role of P2 receptors in the CD. A combination of approaches has demonstrated that extracellular nucleotides, acting from both apical and basolateral sides, can have significant effects on water and electrolyte handling in this important nephron segment – the final site of regulation of urinary output.
In the mid-1990s it was shown that activation of basolateral P2 receptors in rabbit CCD and rat IMCD, perfused in vitro , reversibly inhibited vasopressin-stimulated osmotic water permeability. On the basis that UTP and ATP were equipotent, whereas other nucleotides were without effect, the inhibition found in the rat was attributed to basolateral P2Y 2 receptors ; this P2 receptor-mediated inhibition has been found to be PKC-dependent, and to result from decreased intracellular cAMP and increased PGE 2 levels. The inhibitory action (at least in IMCD) appeared to be mediated only by basolateral receptors, since luminal application of ATP was without effect in this nephron segment. Enhanced expression of P2Y 2 mRNA, and of the receptor protein itself, in the inner medulla of hydrated versus dehydrated rats, has provided additional evidence for a regulatory role for P2Y 2 receptors in modulating CD water reabsorption, and this view is supported by the observation that chronic vasopressin V 2 receptor stimulation with dDAVP reduces inner medullary P2Y 2 mRNA and protein expression. Gene deletion studies further substantiate a role for the P2Y 2 receptor in ATP-evoked inhibition of AVP-stimulated osmotic water permeability in the CD. Under basal conditions, P2Y 2 −/− mice concentrated their urine to a greater degree, and their renal medullary aquaporin-2 (AQP2) abundance was significantly higher, compared with values in wild-type mice, despite almost identical plasma vasopressin levels, and following chronic dDAVP treatment, inner medullary AQP2 expression was increased to a markedly greater extent in the P2Y 2 knockout animals. In summary, the overall picture is that P2Y 2 receptor activation inhibits vasopressin-stimulated, AQP2-mediated water transport in the CD, and this results from decreased intracellular cAMP and increased intracellular PGE 2 ; the latter, in turn, can reduce cAMP levels and effect the retrieval of AQP2 from the apical membrane.
A recent study from our laboratory, albeit in a cultured, immortalized mouse CCD cell line (mpkCCDc14), has provided additional evidence that P2 receptor activation may exert its inhibitory effect on water transport via altered AQP2 trafficking. Application of dDAVP to the basolateral membrane for four days resulted in marked AQP2 immunofluorescence in the apical membrane, but when ATP or ATPγS was then added to the medium, either apically or basolaterally, the AQP2 was internalized. Treatment with dDAVP induced gene expression of P2X 1 in the apical domain, and led to translocation of P2X 2 and P2Y 2 to the apical and basolateral membranes, respectively. When these three subtypes were co-expressed with AQP2 in Xenopus oocytes, their activation reduced cell membrane AQP2 abundance and consequently reduced water permeability. These findings suggest that: (1) in addition to basolateral P2Y 2 receptors, apically located P2 receptors can contribute to the downregulation of AQP2-stimulated water transport; (2) altered trafficking of AQP2 is involved; and (3) vasopressin itself can increase membrane abundance of P2 receptors (c.f. ). However, it must be stressed that the observations were confined to in vitro systems using a mouse CD-derived cell line and Xenopus oocytes. As such, they must be viewed with caution.
Although early studies in MDCK cells reported that ATP activates K + channels, evidence in native CD suggests that K + secretion by principal cells is inhibited by nucleotides. A patch-clamp investigation of split-open mouse CCDs (allowing access to the apical membrane) demonstrated that ATP reversibly inhibits the activity of the small-conductance K + (SK; also called ROMK) channels, which are believed to mediate most potassium secretion in the distal nephron. On the basis of equipotency of ATP and UTP, and the absence of effect of αβmeATP and 2 meSATP, it was concluded that apical P2Y 2 receptors were responsible. That P2Y 2 knockout mice maintain a robust potassium excretion despite mild hypokalemia is consistent with this hypothesis. It is worth noting that the inhibitory effect of ATP on SK channel activity observed by Lu and colleagues could be blocked by the NOS inhibitor L-NAME, which implies that the NO-dependency of nucleotide-induced physiological actions may not be confined to the TAL ( vide supra ).
K + secretion in the distal nephron is generally enhanced when tubular flow rates are increased. This flow-induced increase in K + secretion is now thought to be mediated not by SK channels, but by large-conductance, maxi-K (big K; BK) channels. Activation of BK channels is through increased [Ca 2+ ] i , and there is good evidence for a causal link between increased tubular flow rate, increased tubular nucleotide secretion, and increased [Ca 2+ ] i ( vide infra ). Although highly speculative at this stage, one implication of these various observations is that nucleotides may have conflicting effects on SK and BK channels.
Studies into the effects of extracellular nucleotides on CD Na + transport have generally used sensitivity to amiloride as a basis for identifying ENaC-mediated transport; although occasionally sensitivity to benzamil – a more selective inhibitor of ENaC – has been employed. Koster and colleagues were the first to report that benzamil-sensitive transcellular Na + transport is inhibited by nucleotide activation of P2 receptors. Using primary cultures of rabbit CD (and CNT) cells grown to confluence, they demonstrated that apically or basolaterally applied ATP inhibited benzamil-sensitive short circuit current (SCC; used as an index of Na + transport) across cell monolayers; the mechanism involved activation of PKC and/or PLC. The P2 receptor responsible for this inhibition was equally sensitive to ATP and UTP, but was insensitive to ADP. On this basis, the inhibition of ENaC was attributed to activation of P2Y 2 receptors. Subsequent studies using the mouse M1 cell line reported similar findings, except that the mechanism did not involve PKC. Another CD cell model, the mouse mIMCD-K2, responded similarly to apical (but not basolateral) nucleotides, although in this case, on the basis of pharmacological profiling and mRNA expression, P2X receptors (P2X 3 and P2X 4 ) as well as P2Y receptors (P2Y 1 and P2Y 2 ) were thought to be responsible. A more recent study, using the mouse IMCD-3 cell line, provided an exception to the “rule” of nucleotide-induced inhibition of CD transport: apical application of ATP induced an increase in SSC, although sensitivity to amiloride or benzamil was not tested. Since this effect could be reproduced by the P2X agonist BzATP, it was inferred that the receptors responsible were P2X 1 and/or P2X 4 , as located in native medullary collecting duct ( vide supra ), although UTP was able to increase SCC to some extent, implying a contribution from P2Y receptors.
Investigations of nucleotide effects on CD sodium transport are not limited to cell cultures: nucleotide-induced inhibition of Na + reabsorption in the CD has also been reported in native tissue. In mouse CCD perfused in vitro , ATP and UTP, applied either luminally or basolaterally, caused an increase in [Ca 2+ ] i (and subsequent activation of PKC), and inhibition of amiloride-sensitive SCC, an effect attributed to P2Y 2 receptor activation. Subsequent single-channel patch-clamp experiments, using both rat and mouse CCD, showed that activation of apical P2 receptors with ATP decreased ENaC open probability (P o ), via PLC-dependent breakdown of PIP 2 . Although not tested in rats, it was found that in mice UTP was able to reduce P o to the same extent as ATP. Further pharmacological profiling led to the conclusion that P2Y 2 receptors were responsible (at least in mice). The fact that in P2Y 2 −/− mice the effect of ATP on ENaC P o was severely blunted provided strong support for this conclusion, although a residual effect was still evident, suggesting partial involvement of other P2 receptors. A follow-up study from the same laboratory showed that increased dietary sodium causes a lowering of ENaC P o in wild-type mice, but not in P2Y 2 knockout mice, implying a central role for P2Y 2 receptors in the CD response to changes in sodium intake. More recently, the same group has extended this conclusion to implicate P2Y 2 receptors in the phenomenon of aldosterone escape – the restoration of normal sodium excretion rates in the face of chronically raised mineralocorticoid levels. Hitherto, aldosterone escape had been thought to rely on compensatory changes in more proximal segments of the nephron delivering an increased sodium load to the aldosterone-sensitive sites. However, Stockand and colleagues found that whereas wild-type mice on a high-sodium diet excreted appropriately high levels of sodium in the face of three days of deoxycorticosterone acetate treatment, owing partly to reduced ENaC activity, in P2Y 2 −/− mice the reduction in ENaC activity was much less pronounced, and the natriuresis correspondingly compromised.
At the time of writing, only one study of the effect of nucleotides on CD Na + reabsorption in vivo has been published. Late distal tubules of rats were microperfused with artificial tubular fluid containing 22 Na, the urinary recovery of which was recorded. In animals fed a low-sodium diet (to upregulate ENaC activity), addition of ATPγS to the luminal perfusate was found to inhibit CD 22 Na reabsorption. Despite firm evidence from in vitro studies in mice for P2Y 2 mediation, “selective” P2Y 2 /P2Y 4 agonists were ineffective in vivo in rats, and a P2X heteromer-mediated effect was suggested. A recent patch-clamp investigation of split-open rat CCD (allowing access to the apical membrane) has provided evidence that both apical P2X and P2Y receptors can affect ENaC activity. Activation of P2Y receptors, molecularly and pharmacologically characterized as P2Y 2 and/or P2Y 4 subtypes, inhibited ENaC activity by a PLC-dependent mechanism. Notably, activation of P2X receptors, characterized as P2X 4 and/or P2X 4/6 receptors, either inhibited or potentiated ENaC activity, depending on the luminal concentration of sodium. When luminal sodium was 145 mM (which is the concentration typically used in this type of in vitro experiment), P2X 4 and/or P2X 4/6 activation with 2 meSATP inhibited ENaC activity, whereas when luminal sodium concentration was 50 mM (which mimics the normal sodium concentration of fluid entering the CD in vivo ), P2X 4 and/or P2X 4/6 activation with 2 meSATP potentiated ENaC activity ( Figure 18.7 ). These findings led us to propose that P2X 4 and/or P2X 4/6 receptors might act as apically expressed sodium sensors for the local regulation of ENaC activity in the rat CD. The situation is almost certainly more complex, since preliminary data from our laboratory, using the same electrophysiological techniques, have demonstrated that P2X 4 and P2X 4/6 , and P2Y 2 and P2Y 4 receptor-mediated regulation of ENaC is also dependent on nucleotide concentration, duration of exposure to nucleotide, and tubular pH: higher nucleotide concentrations and prolonged exposure favor P2Y-mediated inhibition, while reducing tubular fluid acidity favors P2X-mediated increases in ENaC activity. Staying with the P2X 4 theme, a causal link has been established between P2X activation, the apical insertion of ENaC, and enhancement of sodium transport, albeit in a “distal-like” cell line; activation of a basolateral P2X 4 -like receptor in Xenopus A6 cells alters cell shape by a rearrangement of the cytoskeleton, which results in increased Na + transport brought about by the unruffling of the apical membrane and insertion of ENaC.