The physiological role of mammalian solute carrier (SLC) proteins is to mediate transmembrane movement of electrolytes, nutrients, micronutrients, vitamins, and endogenous metabolites from one cellular compartment to another. Many transporters in the small intestine, kidney, and solid tumors are H+-coupled, driven by local H+-electrochemical gradients, and transport numerous drugs. These transporters include PepT1 and PepT2 (SLC15A1/2), PCFT (SLC46A1), PAT1 (SLC36A1), OAT10 (SLC22A13), OATP2B1 (SLCO2B1), MCT1 (SLC16A1), and MATE1 and MATE2-K (SLC47A1/2).
The selective movement of ions, nutrients, metabolites, and other organic compounds across biological membranes is a fundamental physiological process of all living cells. The Human Genome Organisation has categorized over 360 genes into 48 solute carrier (SLC) families due to the ability (or predicted ability) of their related proteins to act as membrane transporters (40). Many of these transporters are secondary active transport systems that utilize transmembrane ion gradients as an energy currency to drive uptake or efflux of solutes. Although originally thought to be limited to lower organisms, there is now substantial evidence that certain solute carriers in animal cells are driven by H+-electrochemical gradients (40, 108).
H+-electrochemical gradients (“chemical” pH gradients plus electrical gradients) exist across the membranes of subcellular organelles such as lysosomes and secretory vesicles due to H+-ATPase activity that maintains low intra-organelle pH. Many SLC transporters utilize these gradients to mediate transmembrane solute flux (for example, members of the SLC18 family load synaptic vesicles with neurotransmitters; Table 1) (25). Similarly, at the inner mitochondrial membrane, the electron transport chain maintains an inward H+-electrochemical gradient that drives solute transport via members of the SLC25 family (Table 1) (79). At the mammalian plasma membrane, although an inside negative membrane potential is maintained, transmembrane pH gradients are usually minimized by the action of pH homeostatic mechanisms that transport acid/base equivalents such as the ubiquitous Na+/H+ exchanger NHE1 (SLC9A1) and a number of HCO3− transporters. Consequently, interstitial pH changes, and any resulting pH gradients across the plasma membrane, are generally considered to be pathophysiological (associated with events such as acidosis/alkalosis, hypoxia, or malignant cell metabolism). Indeed, maintenance of a constant bulk extracellular pH is a cornerstone of homeostasis. However, distinct, localized, extracellular pH domains and H+-electrochemical gradients do exist at the plasma membrane of certain cell types such as the brush-border (apical) membranes of the small intestine and renal proximal tubule under physiological conditions (see below). These H+-electrochemical gradients promote (re)absorption and secretion of solutes via many different H+-dependent carriers.
As well as the essential physiological roles played by membrane transporters, it is evident that many also function as integral determinants of drug disposition. Clinically useful drugs are often structural analogs of endogenous solutes and, as a result, can hijack related transporters. In addition to mediating the uptake and efflux of therapeutic compounds in individual cells, transporters expressed in epithelial and endothelial tissues (e.g., small intestine, kidney, liver, choroid plexus, blood-brain barrier) can influence systemic and CNS exposure by determining absorption and clearance. Knowledge of transporter expression and function is now being integrated into rational drug design to predict and improve oral bioavailability and cellular targeting and to limit toxicity. Here, we summarize recent advances made in our understanding of how transporters, driven by H+-electrochemical gradients, are involved in drug uptake and efflux. We will focus on two tissues where physiological H+-electrochemical gradients exist: the small intestine and renal proximal tubule. We also explore the potential of H+-coupled transporters as routes for anti-cancer drug delivery (and as drug targets per se) in solid cancers, where pathophysiological pH gradients are often established through altered metabolism.
H+-Coupled Transport in the Small Intestine
A H+-electrochemical gradient exists across the brush-border membrane of the small intestine that can act as a driving force for the absorption of many nutrients and drugs (108). The acid microclimate is an area of low pH found adjacent to the luminal surface of the small intestinal wall (21, 60, 94). The jejunal microclimate has been recorded in the pH range of 6.1–6.8 (for review, see Ref. 108). From the 1960s onward, the characterization of transport mechanisms has resulted in an ever-growing list of solute carriers that utilize the acid microclimate as a driving force for nutrient uptake (108). The transporters highlighted below have diverse physiological substrates (di/tripeptides, folate, amino acids, and organic anions) but can also transport many groups of orally delivered drugs (FIGURE 1A).
The Proton-Coupled Di/Tripeptide Transporter PepT1 (SLC15A1)
The absorption of intact di- and tripeptides in the mammalian small intestine has been studied intensively since the 1950s to 1960s, yet the H+-coupled nature of the transporter involved was not elucidated until the 1980s (FIGURE 1A) (32, 65). Brush-border membrane vesicle preparations from rodent small intestine were used to show that uptake of dipeptides such as Gly-Pro was stimulated by lowering extravesicular pH to pH of 5.5–6.0 (33). Furthermore, studies using a protonophore (to dissipate H+ gradients) and valinomycin (to generate an inside-negative membrane potential) demonstrated that the pH dependence of dipeptide uptake was due to a requirement for a transmembrane H+-electrochemical gradient rather than to low external pH per se (33). The coupling of H+ flux to dipeptide uptake was confirmed using a pH-sensitive dye to measure intracellular acidification of intestinal epithelial cells on exposure to apical dipeptides (111). Rabbit and human PepT1 cDNAs were subsequently isolated and, when expressed in Xenopus laevis oocytes, were found to induce H+-coupled, rheogenic transport of di- and tripeptides (28, 56). PepT1 is found along the length of the small intestine (30) and is localized to the brush-border membrane (123) where it acts as a relatively high-capacity, low-affinity (Km ∼0.5–10 mM) transporter with broad substrate specificity. Since PepT1 can transport most of the naturally occurring di-/tripeptides tested so far (including zwitterionic, cationic, and anionic peptides), it has the potential to transport thousands of different compounds and is likely to be responsible for the absorption of a significant proportion of dietary protein.
Recently, attention has been focused on identifying the extent to which PepT1 can be exploited as a route for hydrophilic drug absorption in the intestine (15, 84). A diverse range of peptide-like drugs are substrates for PepT1 (FIGURE 1A), including the photodynamic therapy agent 5-aminolevulinic acid, used orally in cancer treatment (24), and certain β-lactam antibiotics. The relative affinity of different β-lactams for PepT1 correlates well with their relative oral bioavailability in humans (15, 84). Angiotensin-converting enzyme (ACE) inhibitors, used in hypertension, also interact with PepT1. However, the degree to which PepT1 mediates intestinal absorption of ACE inhibitors is still being debated (15). PepT1 is also being targeted as a route for improving oral bioavailability of poorly absorbed drugs through synthesis of pro-drug substrates (FIGURE 1A). For example, the oral absorption (and PepT1 transport) of both the Parkinson's drug l-DOPA and the anti-viral acyclovir are dramatically improved when conjugated to the amino acids phenylalanine and valine, respectively (11, 31, 99). Therefore, our understanding of the structural requirements for compounds to be substrates for PepT1 (84) is influencing drug design and ultimately improving drug delivery.
The Proton-Coupled Folate Transporter PCFT (SLC46A1)
The stimulation of intestinal folate absorption by decreasing extracellular pH was recorded as early as the 1970s (96), making it one of the first pH-dependent transport systems to be described in mammals. Nevertheless, the molecular identity of the intestinal H+-coupled folate transporter was not elucidated until 2006 (82). Goldman and colleagues identified a gene that encodes a high-affinity, rheogenic, H+-coupled folate transporter (named PCFT for Proton-Coupled Folate Transporter) (FIGURE 1A) (82). PCFT has a high affinity for both the oxidized folic acid and the naturally occurring reduced 5-methyltetrahydrofolate [Km ∼1.5–3 μM at pH 6.5 (82–83)]. PCFT is highly expressed in the duodenum and proximal jejunum, the principal site of folate absorption, and is localized at the apical membrane (45, 83, 92). Mutations in PCFT cause the rare autosomal recessive disorder hereditary folate malabsorption (82).
In addition to dietary folates, PCFT can transport a number of antifolate analogs, such as methotrexate, used extensively in cancer therapy and as an anti-rheumatic (FIGURE 1A) (82). The uptake of methotrexate in proximal small intestinal brush-border membrane vesicles is maximal at pH 5.0 (91). Similarly, in PCFT-expressing oocytes, methotrexate transport is negligible at extracellular pH 7.4, but at pH 6.5 (around the pH in the jejunum), substantial, high-affinity (Km ∼7 μM) H+-coupled methotrexate transport is measurable (45, 83). Therefore, PCFT function is likely to be a key determinant of antifolate intestinal absorption after oral dosage. Sulfasalazine, used orally in the treatment of rheumatoid arthritis and ulcerative colitis, decreases intestinal folic acid uptake (although it is not clear whether it is a substrate or inhibitor of PCFT) (45). Therefore, PCFT may act as a site of drug-drug interaction whereby oral sulfasalazine results in malabsorption of co-administered antifolates.
The Proton-Coupled Amino Acid Transporter PAT1 (SLC36A1)
A H+-coupled amino acid transporter found at the brush-border membrane of the intestinal epithelium was characterized extensively in the 1990s and named System PAT (107, 113, 114). The related gene was isolated and named firstly LYAAT1 (Lysosomal Amino Acid Transporter 1) (86) and subsequently PAT1 (Proton-coupled Amino acid Transporter 1) (13, 19). PAT1 protein is localized at the brush-border membrane in human and rat small intestine (3) and represents the rat imino acid carrier described in the 1960s (for review, see Ref. 107). PAT1 is a high-capacity, low-affinity (Km ∼0.5–10 mM) transporter of small neutral amino/imino acids and naturally occurring analogs including glycine, alanine, proline, hydroxy-proline, the neurotransmitter GABA, and the osmolytes taurine and betaine (Table 1) (4, 12, 19, 86, 110, 114, 115). PAT1 can transport many d-amino acids (e.g., d-cysteine and d-serine), often in preference to the l form (12, 19, 115). Tryptophan analogs, including serotonin, act as non-transported PAT1 inhibitors (68). PAT1 transports zwitterionic amino acids in a 1:1 stoichiometry with a H+, making this transporter rheogenic and membrane potential sensitive (13, 19). However, when expressed in Xenopus laevis oocytes, PAT1 can also mediate electroneutral H+/monocarboxylate transport, suggesting that acetate and butyrate are also natural substrates (29).
The relatively high capacity of PAT1 makes it an attractive target for oral delivery of hydrophilic, amino acid-like therapeutics. To date, PAT1 has been identified as the principal brush-border membrane transporter for d-serine (used in schizophrenia), the antibiotic d-cycloserine, the anti-hyperglycemic β-guanidinopropionic acid, and a number of therapeutic GABA analogs such as the anti-seizure drug vigabatrin (FIGURE 1A) (1, 54, 67, 109, 110, 115). Recently, PAT1 (along with PepT1) has been identified as a transporter of 5-aminolevulinic acid, an orally delivered drug used widely in photodynamic therapy and imaging of cancer (5).
The Organic Anion Transporters OATP2B1 (SLCO2B1), MCT1 (SLC16A1), and OAT10 (SLC22A13)
The rapid absorption of certain organic anions was one of the key observations that led to the proposal of an acid microclimate in the small intestine (41). At the time, the acid microclimate was thought to influence simple diffusion of organic anions via “pH partition.” However, it is now known that uptake of organic anions (both endogenous and exogenous) in the small intestine can occur via a number of different pH-dependent transporter systems, although the molecular identity of the transporter proteins is still being determined.
The organic anion transporting polypeptide OATP2B1 (formally OATP-B) is localized at the apical membrane of the human small intestinal epithelium and intestinal Caco-2 cell monolayers (53, 87). When expressed heterologously, OATP2B1 mediates high-affinity (Km 1–100 μM), pH-dependent, electroneutral uptake of endogenous weak acids including the vitamin nicotinate, sulphate-conjugated steroids, and bile acids (39, 75). OATP2B1 can also transport numerous orally delivered drugs such as the cholesterol-lowering statins, the anti-histamine fexofenadine, and the anti-diabetic glibenclamide (FIGURE 1A) (39, 75, 88, 100). Interestingly, the substrate specificity of OATP2B1 is much broader at acidic pH than neutral pH (75). The mechanism by which OATP function is enhanced following a decrease in extracellular pH remains unclear. Evidence is available to support either organic anion influx by H+-cotransport (a low extracellular pH-stimulated increase in Vmax in OATP2B1-transfected HEK293 cells, and an overshoot in brush-border membrane vesicles) (75, 87) or HCO3− exchange (in OATP2B1-CHO cells) (55), which, again, would be stimulated by lowering extracellular pH. Stieger and colleagues (55) also identified an increase in substrate affinity at pH 6.5 compared with pH 8.0 in OATP2B1-expressing cells. The pH sensitivity of several OATPs (of the SLCO family) was substrate specific, perhaps due to the existence of distinct substrate binding domains (55). Further investigation may be required to clarify how extracellular pH alters OATP function in situ and the role of H+ influx and HCO3− (OH−) efflux gradients. Determining the degree to which OATP2B1, or other OATPs, contributes to anionic drug uptake in the small intestine is complicated by the overlap in substrate specificity with two other pH-dependent/H+-coupled transporters, the monocarboxylate transporter MCT1 and the recently identified organic anion transporter OAT10 (9, 26, 66, 108).
MCT1 is a fairly low-affinity (Km ∼1–10 mM), electroneutral, H+-coupled transporter of organic anions including short-chain fatty acids (e.g., lactate, pyruvate, butyrate), nicotinate, and xenobiotics (some also transported by OATP2B1; FIGURE 1A) including salicylate and statins (16, 26, 35, 101). The function and substrate specificity of MCT1 and other SLC16 transporters has been reviewed thoroughly elsewhere (26, 66). The cellular localization of MCT1 in the intestine is controversial, with both apical and basolateral immunolocalization being identified in crypt and/or villus cells in various different species (Refs. 35, 37, 101; for review, see Ref. 108). Due to the overlap in function, substrate specificity, and, potentially, location, further work is required to separate the contribution of MCT1 and OATPs to intestinal anionic drug absorption.
OAT10 is a high-affinity nicotinate transporter (Km 22 μM at pH 5.0), which likely corresponds to the H+-coupled transporter described previously at the brush-border membrane of intestinal epithelial Caco-2 cells and in human brush-border membrane vesicles (9, 73). Nicotinate uptake via OAT10-expressing oocytes is stimulated by reducing extracellular pH, suggesting that OAT10 may act as a H+-coupled nicotinate transporter. However, since many of the OAT members of the SLC22 family function as anion exchangers, it is possible that OAT10 may function as a pH-dependent exchanger in situ in the manner proposed for OATP2B1 (see above). Competition and trans-stimulation studies identified that OAT10 may also transport urate, lactate, estrone sulphate, and xenobiotics, including salicylate and furosemide (FIGURE 1A), but with significantly lower affinity than nicotinate (9).
Conclusions From Intestinal Studies
The H+-electrochemical gradient found at the apical membrane of the small intestinal epithelium drives the uptake of diverse drug groups via several H+-coupled/pH-dependent transporters. It is likely that other intestinal H+-coupled transporters (Table 1) will prove useful targets for improving oral delivery of therapeutic compounds. The importance of maintaining the H+-electrochemical gradient for optimal nutrient and drug absorption is exemplified by certain disease states where the acid microclimate pH becomes more alkaline (e.g., celiac disease or enterotoxin exposure), resulting in reduced absorption of H+-coupled transporter substrates such as folate and salicylate (59, 61). At the level of the brush-border membrane, the absorptive capacity of both PepT1 and PAT1 are dependent on the coexpression and function of the Na+/H+ exchanger NHE3 (SLC9A3) (FIGURE 2) (3, 7, 8, 50, 112). This functional cooperativity seems to require NHE3 to maintain intracellular pH, and thus the transmembrane H+-electrochemical gradient, during H+-coupled nutrient and drug transport. Pharmacological or (patho)physiological inhibition of NHE3-mediated H+ efflux will indirectly reduce nutrient and drug uptake via both PepT1 and PAT1 (FIGURE 2) (6–8, 51, 112). Whether such regulation affects the pharmacokinetics of therapeutic substrates and, as such, produces inter-/intrapatient variability in drug handling remains to be identified.
H+-Coupled Transport in the Proximal Tubule
The lumen of the proximal tubule is progressively acidified by proton secretion from renal epithelial cells. This acidification is due primarily to the apical Na+/H+ exchanger NHE3 (71, 90). The luminal pH, measured in situ with microelectrodes in the mid to late proximal tubule, is ∼6.7–6.8 (121). Thus, like in the small intestine, an inward H+-electrochemical gradient exists across the renal brush-border membrane that is available to drive both nutrient and drug reabsorption via H+-coupled symporters and organic cation secretion through H+-antiporters.
The Proton-Coupled Di-/Tripeptide Transporter PepT2 (SLC15A2)
The kidney has considerable capacity to reabsorb filtered di-/tripeptides as well as those produced by luminal peptidases in the proximal tubule. In 1995, a di-/tripeptide transporter, related to PepT1, was isolated from human kidney and named PepT2 (58). PepT2 is a rheogenic, H+-coupled transporter with essentially identical substrate specificity but generally higher affinity (Km ∼50–1,000 μM) to PepT1 (Table 1) (14). Both low- and high-affinity peptide transporters have been described in renal brush-border membrane vesicles (56), and both proteins are found at the brush-border membrane of the proximal tubule, although in sequential sections (PepT1 in S1, PepT2 in S2-S3 of rat) (93). Several investigations have shown PepT2 to be the dominant di-/tripeptide transporter in the kidney. For example, in vivo microperfusion studies identified the principal site of dipeptide reabsorption as the middle/late proximal tubule (95), which correlates with the location of PepT2. Similarly, PepT2 knockout mouse studies identified PepT2 as being responsible for ∼65–86% of dipeptide reabsorption (77, 85).
Like PepT1, PepT2 can transport many clinical and experimental therapeutic compounds (FIGURE 1B) (15, 84). Therefore, PepT2 is likely to have a considerable impact on the systemic exposure to such compounds through determining the degree of renal clearance. Direct evidence for a role of PepT2 in drug handling comes from knockout mouse studies where plasma levels of both the β-lactam antibiotic cefadroxil and the anti-cancer drug 5-aminolevulinic acid are reduced compared with wild-type animals due to an increase in renal clearance (47).
Since a number of techniques have identified PepT2 as the predominant renal di-/tripeptide transporter, the role of renal PepT1 in peptide and peptidomimetic reabsorption is unclear. As well as being measured in brush-border membrane vesicles, PepT1-like function has been measured in a rat renal lysosomal membrane vesicle preparation (133). It is possible that brush-border PepT1 is endocytosed in parallel with glomerular proteins and mediates di-/tripeptide efflux from acidic endosomes after protein degradation. Further investigation is required to identify the role of PepT1 in both physiology and drug handling in the human kidney.
The Organic Cation Transporters OCTN1 (SLC22A4), MATE1 (SLC47A1), and MATE2-K (SLC47A2)
Renal secretion plays a critical role in limiting the potential toxic effects of xenobiotic agents derived from diet, environment, and therapy. A key step in renal elimination of many organic cations is efflux across the brush-border membrane of the proximal tubule via an electoneutral cation/H+ exchanger (42, 128). An early candidate for this transporter was the novel organic cation transporter OCTN1, which, when expressed heterologously, mediates organic cation uptake or efflux in the presence of an opposing pH gradient (98, 102). However, the functional characteristics of the endogenous cation/H+ antiporter are somewhat different from exogenously expressed OCTN1 (for review, see Ref. 128). A recent report using OCTN1 knockout mice has rather indicated a role for OCTN1 in mediating reabsorption of the dietary antioxidant ergothioneine (49).
There is now convincing evidence that the multi-drug and toxic compound extrusion proteins MATE1 and/or MATE2-K are the molecular equivalents of renal cation/H+ antiporter function (Table 1). MATE1 was isolated due to its weak homology with the bacterial MATE proteins (78), followed by the related protein MATE2-K (63). [Note that the protein originally identified as MATE2 in mouse (78) is an alternative splice variant of MATE2-K (63, 105).] Immunohistochemistry identifies both MATE1 and MATE2-K at the brush-border membrane of the proximal tubule (63, 78). Heterologous expression of MATE1 or MATE2-K produces uptake and efflux of prototypical organic cation transporter substrates such as tetraethylammonium (TEA) (63, 78). The affinity for TEA uptake via MATE1 (Km 220 μM) and MATE2-K (Km 830 μM) is comparable to that for pH-dependent TEA uptake in rat renal brush-border membrane vesicles (Km 210 μM at intravesicular pH 6.0, extravesicular pH 7.5) (62, 63, 78). Although most experiments have only measured cation influx, it is assumed that the physiological function of renal MATE transporters will be in the efflux direction due to the prevailing inward H+-electrochemical gradient in the proximal tubule.
The MATE proteins act as multi-specific transporters that can transport various therapeutic agents including cimetidine (H2 blocker), procainamide (anti-arrhythmic), and metformin (anti-diabetic) (Km ∼0.1–4 mM) (FIGURE 1B) (63, 103). Both transporters can also accept anionic estrone sulphate and acyclovir (anti-viral), and, in addition, MATE1 transports the zwitterions of cephalexin and cephradine (antibiotics) (103). MATE1 knockout mice display reduced renal clearance of drug substrates such as metformin (116). The ability of MATE2-K to discriminate between certain anti-cancer platinum drugs is hypothesized to be directly related to the relative nephrotoxicity of these drugs, whereby MATE transport prevents accumulation within the kidney (105).
Conclusions From Renal Studies
The H+-electrochemical gradient in the renal proximal tubule will influence both drug reabsorption via H+-cotransporters (e.g., PepT2) and secretion via H+-antiporters (MATEs). Many other H+-coupled transporters are expressed in specific sections of the proximal tubule including the amino acid/drug transporter PAT1 (see above) (70, 107), the related PAT2 (SLC36A2) (Table 1) (120), and the nicotinate transporter OAT10 (see above) (9). These transporters, and others, may significantly influence the systemic exposure to certain drug substrates.
H+-Coupled Transporters in Tumor Cells
Tumor cells often undergo a metabolic switch (the Warburg effect) to become highly glycolytic. These cells must extrude large quantities of lactate and protons produced by glycolysis, to maintain intracellular pH and thus viability. Therefore, the extracellular pH of many solid tumors is acidic (by an average of ∼0.5 pH units) compared with normal tissue (36, 126). The resultant H+-electrochemical gradient not only favors passive weak acid uptake by pH partition but also acts as a driving force for H+-coupled solute transport at the cancer cell plasma membrane. In addition, since a decrease in external pH may increase affinity and broaden substrate specificity of pH-dependent transport systems [such as PCFT and OATPs (55, 75, 82)], selective uptake into cancer cells is possible (even though tissue distribution of the transporter may be widespread). H+-coupled transporters that are overexpressed in certain cancers (and have the potential to be utilized in targeting anti-cancer therapy) include PCFT, PepT1, and PepT2 (FIGURE 3).
The level of antifolate uptake into tumor cells is a key determinant of cancer sensitivity or resistance to these drugs (64, 130). A good deal of research on antifolate transport has focused on the role of the apparently ubiquitous reduced folate transporter RFT (SLC19A1) and, to a lesser extent, uptake via (very high affinity) folate receptor-mediated endocytosis (64, 130). However, there is substantial evidence that the H+-coupled transporter PCFT is expressed in many tumors and influences antifolate uptake (FIGURE 3). For example, in a study of 32 human cancer cell lines, 29 showed significant uptake of methotrexate at pH 5.5, which is suggestive of PCFT activity (129). PCFT also has a high affinity for the new-generation antifolate pemetrexed (Km ∼2 μM at pH 6.5) (125, 132). Recently, several novel pyrrolo[2,3-d]pyrimidine thienoyl antifolates have been developed with selectivity for PCFT (and folate receptors) over RFT (124). Such compounds may improve targeting to PCFT-expressing tumors and thus reduce potential toxicity by avoiding transport via the ubiquitous RFT.
Aberrant H+-coupled di-/tripeptide transport has been characterized in cancer cell lines originating from stomach, pancreas, and biliary duct, and in fibrosarcoma cells (38, 46, 52, 74). PepT1 and PepT2 mRNA expression has also been identified in many other human cancer cell lines (69), and PepT1 mRNA is overexpressed in colorectal cancer biopsies compared with normal colon (5). In addition, PET scanning has shown accumulation of the dipeptide 11C-glycylsarcosine in tumor xenografts of certain human cancer cells, suggesting such labeled substrates could have utility in imaging tumors (69). Overexpression of PepT1 and PepT2 may be useful for targeting a number of experimental and clinical anti-cancer substrates to tumor cells, including the photodynamic therapy and imaging agent 5-aminolevulinic acid (5, 24) and the aminopeptidase inhibitor bestatin (74) (FIGURE 3). In mice, oral bestatin accumulates in xenografts of PepT1-transfected HeLa cells and inhibits the growth of these tumors (74). Pro-drugs of floxuridine and cytarabine may also be transported by PepT1 (97, 117) (FIGURE 1A).
Other pH-dependent transporters that could prove useful in mediating uptake of anti-cancer drugs into tumor cells include the H+-coupled amino acid transporter PAT1 and the pH-dependent OATPs (FIGURE 3). PAT1 can transport 5-aminolevulinic acid (5) and the cancer cell-growth inhibitor l-cycloserine (3). Several OATPs are expressed in various cancer types where they could influence uptake of numerous drug substrates (39). For example, OATP1B3 (SLCO1B3, formally LST-2) is expressed in several gastrointestinal cancers and influences sensitivity of cancer cell lines to methotrexate (2) (FIGURE 3).
H+-Coupled Transporters as Anti-Cancer Drug Targets
Transporters not only influence drug uptake into tumor cells but may even be considered as potential therapeutic targets in their own right. Transporters identified as crucial to cancer cell viability (due to their roles in nutrient delivery or tumor metabolism), could be targeted with specific inhibitors as a novel type of anti-cancer therapy. For example, the overexpression of PepT1 or 2 might be exploited therapeutically by using specific inhibitors as a means to starve tumors of amino nitrogen. Several high-affinity inhibitors have been characterized (15). The H+-coupled monocarboxylate transporters, particularly MCT1 and MCT4 (SLC16A3), are overexpressed in certain cancer cells and mediate lactate flux (26, 34, 66). Inhibitors of lactate transport are effective at inducing acidosis-associated apoptosis in vitro (27), which could translate into novel therapies based on altered cancer cell pH homeostasis. Similarly, a decrease in Na+/H+ exchanger NHE1 activity inhibits tumor growth (104). It is tempting to speculate that tumor cells may favor overexpression of H+-coupled transporters, in particular, to meet their nutritional and metabolic requirements to gain a competitive advantage in the acidic microenvironment. Exploiting altered pH homeostasis of cancer cells through targeting of H+-coupled transporters may prove an exciting avenue for development of novel therapies.
Localized pH Gradients in Other Tissues
Intriguingly, the expression of H+-symporters and H+-antiporters is not limited to tissues and subcellular compartments where maintained pH gradients have been measured directly. Rather, their expression is relatively widespread. pH-dependent transporters are expressed at the plasma membrane of, for example, choroid plexus (e.g., PepT2 and PCFT) (76, 127, 131), blood-brain barrier (e.g., MCT1, OATP1A2 [SLCO1A2], and OATP2B1) (26, 39, 119), and liver (e.g., PCFT, MATE1, and multiple OATPs) (39, 43, 78) where their function will influence the CNS and systemic exposure to therapeutics and other xenobiotic substrates. In addition, the relative function of such transporters in all cells will influence cell-specific drug accumulation and thus both drug efficacy and side-effects.
Heterologous expression studies show that many H+-coupled transporters can function in the absence of a pH gradient (albeit with reduced capacity, affinity, and/or restricted substrate specificity). Solute flux via rheogenic transporters can be driven by membrane potential alone, whereas electroneutral transporters may rely solely on prevailing substrate concentration gradients. However, it is also likely that highly localized (microenvironment) transmembrane pH gradients are found in many tissues under normal physiological conditions. For example, efflux of acid produced by normal cell metabolism is associated with decreased cell surface pH (measured with microelectrodes) in both individual neurons and muscle (22, 106). Similarly, distinct alkaline and acid extracellular transients have been characterized in mammalian brain following neural activity (20, 23). Such changes in extracellular pH may act as a type of paracrine signal within the brain. Furthermore, concurrent but opposite changes in neuronal and glial cell intracellular pH have been identified with activity, suggesting that cell-specific, temporal changes in transmembrane H+-electrochemical gradients may occur in brain tissues (20, 23). In liver, a proton diffusion barrier that results in a localized acidic compartment in the perisinusoidal space has been hypothesized (44). Such local gradients will control the level of H+-coupled transporter activity. In other tissues, H+-coupled transporters function in a cooperative manner with pH regulatory mechanisms, such as Na+/H+ exchangers, to maintain localized H+-electrochemical gradients. Experiments in the small intestinal epithelium have identified a functional interaction of both PepT1 and PAT1 with NHE3, which is required for optimal nutrient uptake (FIGURE 2) (3, 107, 108, 112). NHE3 extrudes H+ ions back across the apical membrane and in doing so maintains intracellular pH and the transmembrane H+-electrochemical gradient for further H+-coupled nutrient and drug uptake. A similar relationship between a Na+/H+ exchanger and PepT2 has been described in astrocytes (122). In the choroid plexus, although PCFT is localized to the basolateral membrane, a model has been proposed whereby the highly pH-dependent PCFT is endocytosed alongside folate bound to the folate receptor FRα. PCFT subsequently exports folate from the acidified endosome (the H+-electrochemical gradient being created by H+-ATPase activity) into the cytosol (127, 131). Evidence for a similar coupling between folate receptors, PCFT, and H+-ATPase activity has been found in placental cells (83) and hypothesized for retinal and renal cells (118, 130).
Therefore, understanding how H+-coupled transporters function in any particular cell to optimize solute and drug movement requires knowledge of not only the local pH microenvironment (both intra- and extracellular) but also of how this environment and transmembrane gradients are modulated in both time and space. pH gradients are dynamic, are highly localized, and can influence, and be influenced by, numerous acid/base transporters. The significance of transporter trafficking between compartments with differing ionic microenvironments is an area that requires further investigation.
There is now convincing evidence that H+-coupled solute carriers play unique and varied roles in physiology, being responsible for the uptake and efflux of a diverse range of endogenous substrates (Table 1). These transporters can be hijacked by certain therapeutic agents (FIGURE 1) and thus can influence drug handling at the level of absorption, clearance, and tissue targeting. The function of some H+-coupled transporters has been thoroughly characterized (e.g., PepT1 and 2), and this information is now being integrated into the drug design stage of therapeutic development. However, with the elucidation of the human genome, it is evident that there are far more solute carrier gene products than can be accounted for by our current knowledge of endogenous transporter function. Understanding how these transporters contribute to physiology, pathophysiology, and drug disposition will depend on detailed investigations into their tissue and subcellular expression, functional characteristics, substrate specificity, and regulation. It is becoming apparent that an understanding of how transporter function is altered by microenvironment and transmembrane electrochemical gradients is of equal importance. The transporters reviewed here can exhibit differences not only in substrate affinity and capacity but also specificity depending on the prevailing pH and transmembrane H+-electrochemical gradients. This is also observed with other putative H+-coupled transporters. For example, the equilibrative nucleoside transporter ENT4 (SLC29A4) was identified originally as mediating organic cation transport (when measurements were made at pH 7.4) but was shown subsequently to have a high affinity for adenosine when measurements were made at acidic extracellular pH (10). The purpose of this review is to raise awareness of the extent of H+ (or base equivalent) coupling within the SLC family of mammalian transport systems. We have chosen a few specific examples to emphasize key points that reflect our own personal interests in epithelial physiology. However, there are many other important H+-coupled transport systems that have not been covered here in detail but which are key drug targets such as NHE1 in ischemic reperfusion injury (48), MCT1 in immunosuppression (72), members of the vesicular neurotransmitter transporter (SLC17, SLC18, and SLC32) (18), and excitatory amino acid transporter (SLC1) (17) families to name but a few examples. We apologize to the readers that each transport system did not receive an equivalent level of attention, but space constraints prevented detailed exploration of every transport system listed in Table 1. In addition, there are other H+-coupled transporters that are not included within the SLC families such as the H+/Cl− antiporters ClC-4 and ClC-5, which are members of the CLC gene family (80, 89).
Overall, the functional dependence on local pH gradients suggests that transporters may function differently when expressed in different tissues or subcellular locations, and this differential function could ultimately be exploited for targeting drugs to specific tissues. Clearly, physiological, pathophysiological [as observed in cancer (see above), hypoxia, and acidosis/alkalosis], or pharmacologically induced changes in pH will have a significant effect on H+-coupled transporter function and, therefore, nutrient and drug handling by the body as a whole.
This work was supported by the Wellcome Trust (grant no. 078640/Z/05/Z).
No conflicts of interest, financial or otherwise, are declared by the author(s).
- ©2010 Int. Union Physiol. Sci./Am. Physiol. Soc.