|Home | About | Journals | Submit | Contact Us | Français|
Na+/H+ exchangers (NHEs) are ubiquitous proteins with a very wide array of physiological functions, which are summarized here with an emphasis on the most recent advances. Hypertension and organ ischemia are two disease states of paramount importance in which NHEs have been implicated. The involvement of NHEs in the pathophysiology of these disorders is incompletely understood. This paper reviews the principal findings and current hypotheses linking NHE dysfunction to hypertension and ischemia.
With the advent of large-scale sequencing projects and powerful in-silico analyses, we have come to know what is most likely the entire mammalian NHE gene family. Recent advances have detailed the roles of NHE proteins, exploring new functions such as anchoring, scaffolding and pH regulation of intracellular compartments. Studies of NHEs in disease models, though not conclusive to date, have contributed new evidence on the interplay of ion transporters and the delicate ion balances that may become disrupted.
This review provides the interested reader with a concise overview of NHE physiology, and addresses the implication of NHEs in the pathophysiology of hypertension and organ ischemia in light of the most recent literature.
Intrinsic to life is order. The strife for low entropy is a thermodynamically costly process and the maintenance of transmembrane ion gradients requires energy. Primary active transport involves direct coupling of ionic movement to the hydrolysis of high energy phosphate bonds. The transmembrane electrochemical potential difference thus created becomes the currency of energy for secondary active transport. The ability to exchange protons (H+) for sodium ions (Na+) across lipid bilayers by a secondary active mechanism is pervasive in all living organisms. Mammalian Na+/H+ exchangers (NHEs) catalyze the countertransport of Na+ and H+ across both plasma and organellar membranes.
Peter Mitchell first suggested the existence of cation/H+ exchange in the mitochondrial membrane in a pioneering monograph in 1961  and provided experimental evidence for such an exchange process [2,3]. To date, genes encoding NHEs have been cloned from the simplest prokaryotes to the most advanced multicellular eukaryotes [4••]. The human genome encodes nine NHE isoforms (SLC9A1-9 genes, NHE1-9 proteins) that have different tissue and sub cellular distributions, in addition to a putative Na+/H+ exchanger with restricted expression in spermatoza (SLC9A10 gene, sperm NHE protein). Given the ubiquity of this class of proteins (Table 1) and their extremely diverse functions (Fig. 1), it is not surprising that NHEs are implicated in multiple physiological and pathophysiological processes.
NHE1 was the first mammalian NHE identified, in 1989 , and is probably the most important isoform for intracellular pH, cell volume and intracellular sodium homeostasis. NHE1 has been shown to regulate cell cycle, proliferation, migration and adhesion, and to confer resistance to apoptosis [12,13•]. NHE1 mediates transepithelial transport in secretory parotid acinar cells  and may mediate ammonium reabsorption in the thick ascending limb of the loop of Henle. Apart from its ionexchange function, NHE1 interacts with the ezrin, radixin and moesin (ERM) family of actin-binding proteins, serving as a membrane anchor for the actin cytoskeleton  and as a scaffold for the assembly of signaling complexes [16•]. NHE1 may also modulate other transporters, such as apical membrane NHE3, possibly through actin cytoskeleton remodeling [17,18•,19•]. Further insight into NHE1 function is offered by two NHE1-defective mouse models (Table 2).
NHE2 is important for Na+ secretion in the parotid gland, and plays a role in maintaining parietal cell integrity in the stomach (Table 2). In the kidney, NHE2 functions in transepithelial NaHCO3 absorption in the distal nephron. At the macula densa NHE2 is the only luminal NHE, but its role in sodium sensing remains to be determined .
NHE3 differs from other isoforms in that it recycles between the plasma membrane (apical membrane of epithelia) and the endosomal compartment, and functions in both locations . NHE3 activity is regulated by alterations in its intrinsic activity and by trafficking between these two compartments [36-42]. NHE3 is responsible for a significant portion of renal and intestinal Na+ absorption (Table 2). The absorptive and secretory functions of luminal membrane NHE3 in the kidney are summarized in Fig. 1b. Intracellular NHE3 has been postulated to be important for endosomal acidification in the reabsorption of filtered proteins by receptor-mediated endocytosis .
NHE4 is involved in intracellular pH and volume regulation, especially in some cells lacking NHE1 (Table 1). NHE4 may mediate ammonium transport from the thick ascending limb lumen to the interstitium, but this theory has not been explored with the NHE4−/− mouse. Very little is known about the function of the neuron-specific isoform NHE5. The ubiquitously expressed NHE6, NHE7 and NHE9 likely serve as organellar pH regulators .
NHE8 was recently identified by in-silico cloning . NHE8 may recycle to the plasma membrane and be regulated by trafficking, similar to NHE3, but there are no available data as yet. In the kidney, NHE8 may functionally complement NHE3 (Fig. 2). Although NHE3−/− mice have no significant alteration in NHE8 protein expression [7•], the role of NHE8 in maintaining proximal sodium transport cannot be ruled out.
The sperm-specific NHE is a predicted hybrid protein expressed in spermatozoa, essential for sperm motility and male fertility . Its Na+/H+-exchange ability has not yet been demonstrated.
In spite of the impressive amount of knowledge that we have accumulated in the last years, including the completion of the Human Genome Project, the exact identity of the protein responsible for Peter Mitchell’s mitochondrial cation/H+ exchange has remained elusive [45,46].
The role of NHEs in hypertension is a question of pivotal importance. There is a massive body of literature on this topic that is not easily amenable to any attempts at a summary. Several broad statements can be made regarding NHEs and hypertension. The majority of findings in this vast literature are correlations of the hypertensive state with either direct measurements of NHE (activity, protein, transcript) in animal models, or surrogates of NHE (Na+/Li+ or Na+/H+ exchange in erythrocytes, lymphocytes, platelets) in humans. There is little or no reassurance that the surrogates actually reflect NHE and, even if they do, causality and significance are difficult to establish. Observations have been made in animal models of polygenic hypertension that demonstrate changes in NHE preceding the hypertensive state. These types of observation rule out NHE defects being results of hypertension but still do not distinguish the difference between causation and parallel phenomena.
The current models of possible involvement of NHEs in hypertension are centered mainly on two isoforms, NHE1 and NHE3 (Fig. 3). Transgenic overexpression of NHE1 leads to salt-sensitive hypertension in mice . Primary hypertension has been associated with increased NHE1 activity measured in both experimental animals and peripheral blood cells from some human subjects with primary hypertension . The NHE1 locus has been ruled out as a candidate in essential hypertension by genetic linkage analysis . When present, the defect is rather due to regulatory mechanisms that are, as yet, poorly defined.
One potential mechanism linking NHE1 to hypertension is overactivation of NHE1 in vascular smooth muscle cells (VSMCs), resulting in intracellular Na+ accumulation, slowing down or reversal of the Na+/Ca2+ exchanger (NCX), increased cytosolic Ca2+ concentration, and VSMC contraction. Chronically increased NHE1 activity may also drive abnormal VSMC growth and proliferation. There are several new arguments supporting the reversal of NCX exchange. The NCX inhibitor SEA0400 decreases systolic blood pressure in several rat models of salt-dependent hypertension [50•]. While NCX1−/− homozygosity is lethal, heterozygous NCX1+/− mice are viable and resistant to salt-dependent hypertension. Conversely, transgenic mice with VSMC-specific overexpression of NCX1 are more prone to high-salt-induced hypertension [50•]. However, there is no definitive evidence that NCX in VSMCs is slowed down or reversed as a result of NHE overactivity. There are alternative, non-mutually exclusive theories, such as the action of cardiotonic steroids (e.g. endogenous ouabain), which can cause intracellular Na+ accumulation by inhibition of the Na+/K+-ATPase.
Another major mechanism linking NHEs and hypertension could be renal alteration of sodium homeostasis and defective pressure natriuresis. NHE3−/− mice are hypotensive, even when small-intestine NHE3 is rescued by transgenic expression (Table 2). Conversely, overactivation of NHE3 could be a mechanism for hypertension. NHE3 has been studied in various rodent hypertension models (Table 3), and has been found to be implicated in some but not all the models. NHE3 regulatory proteins have also been studied in some models, with intriguing but inconclusive results . Several polymorphic variants of NHE3 did not associate with human primary hypertension in a recent case-control study .
Although there is no definitive evidence documenting an NHE defect in human hypertension, both NHE1 and NHE3 remain plausible candidates and more studies are needed to clarify their involvement, as well as the possible involvement of other NHE isoforms.
Ischemia/reperfusion injury is an extremely complex, incompletely understood phenomenon. A simplified working paradigm is that ischemia deprives energy required to maintain ionic gradients with reperfusion triggering an inflammatory response which further exacerbates injury. Mechanisms of ischemic injury are common to all solid organs, but there are specific characteristics for each.
One proposed model of ischemia/reperfusion-induced ionic imbalance poises NHE in an important role during both ischemia and reperfusion. NHE is activated by intracellular acidosis upon ischemia (Fig. 4a), but transport is repressed by the low pH of the interstitium. Restoration of flow during reperfusion (Fig. 4b) normalizes first extracellular pH, resulting in maximal kinetic stimulation of NHE. Restoration of intracellular pH by NHE comes at the price of rising intracellular Na+ concentration. Normally, the Na+/K+-ATPase would extrude excess Na+, but ATP depletion hampers this process. At high intracellular Na+ concentrations, Ca2+ extrusion via NCX is inhibited, and NCX can operate in a reverse mode, loading the cell with Ca2+ resulting in cellular injury by multiple mechanisms . While activation of NHE restores intracellular pH, the concomitant increase in intracellular Na+ can lead to Ca2+ overload and exacerbation of tissue injury, a phenomenon termed the pH paradox.
NHE1 is the predominant isoform expressed in cardiomyocytes, where it functions to rectify intracellular acidification. There is a steep relationship between pHi and NHE1 activity in cardiac cells, with pKi of about 7.4 . The relative resistance to cardiac ischemia/reperfusion injury in NHE1−/− mice is comparable with the effect of cariporide inhibition of NHE1 in wild-type mice , supporting the significance of NHE1 action in cardiac ischemia/reperfusion injury in the rodent model.
There are two major groups of NHE1 inhibitors: pyrazine derivatives (amiloride, dimethyl amiloride, ethylisopropyl amiloride) and benzoylguanidine derivatives (HOE-694, cariporide, eniporide) . Treatment with both pyrazine and benzoylguanidine derivatives has protective effects in cardiac ischemia/reperfusion injury in animal models – especially if used prior to, rather than during or after ischemia [65-69,70•].
As a warning remark, it has to be pointed out that although pharmacological inhibition is a powerful and valuable tool, its results have to be interpreted and translated to the molecular level with some caution. Amiloride has different inhibitory potencies on different NHEs, but also inhibits epithelial Na+ channels, acidsensing ion channels and Na+/Ca2+ exchangers [64,71]. Other NHE inhibitors have relatively higher specificities, but are still far from mechanistically inhibiting a single isoform . The same applies for NCX inhibitors .
In humans, the GUARDIAN  and ESCAMI  trials, using cariporide and eniporide respectively, failed to demonstrate a significant benefit in patients with various cardiac ischemic syndromes. However, in patients undergoing coronary artery bypass graft surgery, a subgroup of the GUARDIAN trial, cariporide offered a 25% relative risk reduction of all-cause death or myocardial infarction throughout the 6-month trial . The results of cariporide and eniporide in human trials are surprising and somewhat disappointing considering the wealth of animal data heralding the efficacy of these agents. One possible explanation could be the timing of the treatment. NHE inhibition has been mostly beneficial in animal models if applied before the onset of ischemia, and has only offered protection in humans undergoing coronary artery bypass graft. Another plausible explanation is that in humans with cardiac ischemia, the effect of NHE inhibitors may be masked by parallel events such as diseased vessels, which are inherently not present in animal models. Studies have started to address this issue by using senescent or atherosclerotic animals, but with inconclusive results as yet [76,77]. While it is difficult to provide treatment preceding ischemia in patients, inhibiting NHE exchange could be beneficial in improving the outcome of ischemia/reperfusion in humans when used in conjunction with other cardioprotective therapies.
Another potential target in the model depicted in Fig. 4 is NCX. In theory, inhibiting NCX should prevent intracellular accumulation of Ca2+ while still permitting the myocyte to defend against low pH. In isolated rabbit hearts cariporide infusion before ischemia reduced infarct size, but its infusion upon reperfusion failed to offer protection. In contrast, infusion of the NCX inhibitor KB-R7943 similarly reduced infarct size when infused before and after ischemia, suggesting that ischemia/reperfusion injury is more efficiently suppressed by blockade of NCX than NHE . Similarly, the NCX inhibitor SEA0400 had cardioprotective effects in canine and rodent cardiac ischemia/reperfusion models [79,80]. To date, human trials with NCX inhibitors are not available.
In neurons, pHi can affect neuronal excitability by ionchannel gating and, conversely, neuronal activity can alter pHi. Neuronal activity can vary enormously from one neuron to another; hence production of metabolic acids can vary greatly. Efficient acid-extrusion mechanisms are critical for homeostasis. Because of their high metabolic rate, neurons are susceptible to injury if exposed to excessive activity (seizure) or ischemia (stroke). pH regulation is likely to be crucial in both physiological and pathophysiological conditions. NHE1 plays an important role in the pH regulation of both neurons  and astrocytes [82•].
A salient characteristic of cerebral ischemia is the failure of ATP-dependent glutamate transporters to remove glutamate, the major excitatory neurotransmitter, from the synaptic cleft, leading to overactivation of glutamate receptors and a rise in cytosolic Ca2+ concentration – a phenomenon termed excitotoxicity . The fact that glutamate receptor antagonists have failed to offer neuroprotection in clinical trials  is likely due to the coexistence of multiple mechanisms contributing to ion imbalance in cerebral ischemia. These mechanisms may include opening of Ca2+- and Na+-permeable acidsensing ion channels (ASIC1a) [85,86], activation of the Na+/K+/Cl− cotransporter NKCC1 , excitotoxicity-induced proteolytic cleavage of NCX1 and NCX3 , as well as activation of NHE1 [81,82•].
The NHE1 inhibitor SM-20220 significantly reduced the extent of cerebral edema, Na+ content and infarcted area in a transient focal ischemia rat model . In primary culture of rat cortical neurons, SM-20220 was protective against glutamate-induced excitotoxicity [90•]. Treatment with cariporide before hypothermic circulatory arrest in pig dramatically improved early neurologic recovery . In primary cultured astrocytes from NHE1+/+ mice, oxygen and glucose deprivation led to a 5-fold increase in intracellular Na+ and cell swelling. In both NHE1−/− astrocytes and NHE1+/+ astrocytes treated with cariporide, the oxygen- and glucose-deprivation-induced Na+ rise was less than 2-fold, and swelling was significantly reduced [82•].
Five NHE isoforms (NHE1–4 and NHE8) are expressed in the plasma membrane of different renal cells (Fig. 2), rendering the understanding of their individual roles in response to ischemia/reperfusion injury more complex. Rats subjected to renal ischemia/reperfusion exhibited drastic reduction in kidney NHE3 mRNA, mild decrease in NHE2 and NHE4 mRNA, and mild increase in NHE1 mRNA [92,93]. At the protein level a different study found decreased NHE3, type II Na+/Pi cotransporter and Na+/K+-ATPase . A pathophysiologic interpretation of these findings could ascribe the observed renal Na+ wasting to loss of Na+ transporters due to ischemia damage [92-94]. An alternative and not exclusive view is that this is not a consequence of damage but rather an adaptive mechanism to cope with low ATP supply. If the primary insult is energy limitation, turning off NHE in the proximal tubule can achieve protective effects by blocking apical Na+ entry and thus reducing Na+/K+-ATPase ATP consumption. The increased distal delivery will repress glomerular filtration rate to avoid massive volume loss as per the ‘acute renal success’ theory of Thurau and Boylan . Reducing intracellular Na+ by blocking apical entry may also reduce Ca2+ loading by NCX. In one study, the NHE3 inhibitor S3226 improved renal function after experimental ischemia/reperfusion . The NCX inhibitor SEA0400 was beneficial in a rat renal ischemia/reperfusion model , and ischemia/reperfusion-induced renal dysfunction was attenuated in NCX1+/− heterozygous mice (having NCX1 protein expression in the kidney decreased to about half) compared to NCX1+/+ mice .
Vast amounts of published data have started to shed light on the elaborate works of the proteins capable of exchanging Na+ for H+ across lipid bilayers. Piecing together this massive body of literature is a formidable task, but even more challenging is the vastness of the areas that remain to be explored. From solving protein structures and explaining transport machineries in molecular detail, to establishing relevance for human disease and therapeutic intervention, Na+/H+-exchange research has yet to reach its full potential.
The authors are supported by the National Institutes of Health (R01-48481, P01-DK20543, M01-RR00633 to O.W.M.) and the National Kidney Foundation (to I.A.B.). We thank Dr Olivier Bonny for useful discussion and critical reading of the manuscript.
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest