SLC Transporters as Therapeutic Targets: Emerging Opportunities

Members of the SLC transporter families are important in human physiology

For example, amino acids, which are needed for protein synthesis, are absorbed and reabsorbed by transporters of the SLC1, SLC3, SLC6, SLC7, SLC25 and SLC36 families, many members of which are expressed in the intestine and kidney. In the intestine, amino acids are absorbed from the lumen into the body; in the kidney, amino acids that are filtered out of the bloodstream by the glomerulus are reabsorbed in the proximal tubule by many of the same transporters that are involved in intestinal absorption¹². Members of the SLC2, SLC5 and SLC50 families are required for intestinal absorption and renal reabsorption of glucose and for glucose uptake into neurons, erythrocytes, hepatocytes and other cell types^{13, 14}. Metals often serve as essential cofactors for important enzymes, but toxicities may occur when they are present at excess concentrations. Zinc transporters of the SLC30 and SLC39 families and iron transporters of the SLC11 and SLC40 families regulate zinc and iron levels in the body, respectively. Similarly, water-soluble vitamins are essential for various processes but require transporters for cellular uptake; for instance, SLC19 family members transport folate and thiamine, SLC46 family members transport folate and the SLC52 family transports riboflavin. In the brain, neurotransmitters released into the synapse are taken back into presynaptic neurons through SLC1 and SLC6 transporters. Interestingly, the roles of many transporters in human physiology have been discovered through studies of Mendelian disease (Table 1; see Supplementary information S1 (table)).

Mendelian diseases associated with SLC transporters

Given the key physiological roles of SLC transporters, it is not surprising that defects in a single transporter can result in a serious disease. Mutations in 20% of the genes encoding known SLC transporters in humans have been associated with Mendelian disease (see the Online Mendelian Inheritance in Man (OMIM) database in Databases), and it is likely that mutations in more SLC transporter genes will be found to be causal for some of the remaining half of Mendelian diseases with no known cause. The defective transporters or transporter deficiencies that cause these diseases result in a wide range of symptoms that affect almost all organ systems; although some of these diseases are considered to be benign, others cause serious illness and death. Table 1 and Supplementary information S1, S2 (tables) summarize the information that is known about each transporter and the Mendelian disease associated with its dysfunction or lack of expression. Figure 1 illustrates the potential mechanisms through which mutations in genes encoding SLC transporters can result in reduced function. Figure 2 shows the transporters that are implicated in Mendelian diseases, organized by the general class of substrates they transport.

Common diseases associated with SLC transporters

SLC transporter gene polymorphisms that are associated with common disease have generally been identified through the genotyping of candidate genes, or from GWASs (see Catalog of Published Genome-wide Association Studies in Further information). Examples of transporter genes for which polymorphisms have been associated with human traits or disease include the following: SLC22A4 and SLC22A5, which are associated with inflammatory bowel disease^{15, 16, 17}; SLC2A9, SLC22A11 and SLC22A12, which are associated with gout and uric acid levels^{18, 19, 20}; SLCO1B1 and SLCO1B3, which are associated with high bilirubin levels^{21, 22, 23, 24}; and SLC24A5 and SLC45A2, which are associated with skin colour^{25, 26}. Many of the polymorphisms implicated in human disease are non-synonymous — that is, they cause amino acid substitutions in the encoded transporter that reduce transporter function or expression levels. For example, non-synonymous variants of the uric acid transporters encoded by SLC2A9, SLC22A11 and SLC22A12 show reduced uric acid transport in vitro. A reduced-function variant will reabsorb less uric acid from the urine. As a result, hypouricaemia can occur, thus reducing the risk of gout.

Polymorphisms in the genes encoding the hepatic transporters OATP1B1 (SLCO1B1) and OATP1B3 (SLCO1B3) provide a second example. These transporters are responsible for the uptake of conjugated bilirubin — the product of haem metabolic catabolism — into hepatocytes. From there, it is excreted into the bile. Common polymorphisms in the genes of transporters that are involved in the hepatic bilirubin elimination pathway may lead to hyperbilirubinaemia and jaundice. Moreover, the complete loss of SLCO1B1 and SLCO1B3 results in a rare disease known as Rotor syndrome²⁷ (Box 3). This example, wherein a common reduced-function gene polymorphism is associated with a less serious phenotype, whereas a rare, more severe defect in the same gene causes a Mendelian disease, is not uncommon in human genetic studies.

There have been several transporters discovered through GWASs that were not previously known to be important in human disease, including the following examples: SLC30A8, which encodes a pancreatic β-cell-specific zinc transporter, a genetic polymorphism of which has been associated with diabetes^{28, 29}; SLC14A1, which encodes a urea transporter, variants of which have been associated with bladder cancer³⁰; SLC4A7, which encodes a bicarbonate transporter, for which certain polymorphisms have been associated with abnormal blood pressure^{31, 32} or with breast cancer³³; and the gene encoding a nucleoside-sugar transporter that also transports thiamine, SLC35F3, some polymorphisms of which have been associated with increased blood pressure and predicted to cause disturbances in cardiac and autonomic function³⁴.

Variants in SLC30A8 have been associated with type 2 diabetes in several GWASs and across different ethnic groups. A non-synonymous variant in SLC30A8, with a high allele frequency (>25% in Caucasians and Asians), p.Arg325Trp (rs13266634), was identified through a GWAS²⁸. This risk allele, p.Arg325Trp, is associated with reduced susceptibility to type 2 diabetes and higher zinc transport activity. It was subsequently demonstrated in Slc30a8-knockout mice that the transporter plays a part in transporting zinc ions from the cytoplasm into insulin granules³⁵, and further studies suggested that zinc may have an important role as an endogenous regulator of insulin homeostasis^{36, 37}. Furthermore, although Slc30a8-deficient mice exhibit phenotypic variability, they generally show lower blood insulin levels and increased insulin clearance³⁸. Interestingly, a recent paper by Flannick et al.³⁹ demonstrated that carriers of loss-of-function and rare missense variants in SLC30A8 were at reduced risk for type 2 diabetes. The researchers performed sequencing or genotyping on more than 150,000 individuals and demonstrated that loss-of-function missense SLC30A8 variants that led to truncation of the transporter protein actually protected individuals from type 2 diabetes. Overall, these human genetic studies suggest that SLC30A8 could be a potential target for the treatment and/or prevention of type 2 diabetes.

Inhibition of transporter function

Most current drugs — as well as novel drugs in clinical trials — that modulate SLC transporters do so by inhibiting transporter activity, and examples of these are described in more detail in the subsequent sections. For diseases in which decreased transporter activity leads to a potentially beneficial effect, high-throughput screening (HTS) of large compound libraries using cell lines that overexpress the transporter of interest can be used to discover lead inhibitor molecules^{40, 41}.

Many SLC transporters may function in both influx and efflux of their substrates, and the net direction of flux across a cellular membrane is dependent on the substrate gradient (or, in the case of secondary or tertiary transporters, the gradient of the co- or counter-transported ions). Because of their simplicity, influx assays are frequently used in screening studies. In particular, cells that are grown on solid support are typically used to screen for inhibitors that prevent the uptake or influx of a fluorescent substrate probe^{42, 43, 44}. Even for transporters that require a co- or counter-transported ion, the assays may be set up in the influx mode for convenience. For example, a high concentration of extracellular sodium may be used to drive the influx of a fluorescent probe, or the cells may be pre-loaded with a counterion to drive fluorescent probe influx. For transporters that are primarily efflux transporters, such as the ABC transporters bile salt export pump (BSEP; also known as ABCB11), canalicular multidrug resistance protein (cMRP; also known as ABCC2) and breast cancer resistance protein (BCRP; also known as ABCG2), a vesicular-transport assay is frequently used. Instead of adherent cells, membrane fractions that contain inside-out vesicles are prepared from cells that overexpress the efflux transporter^{45, 46, 47, 48}.

A fluorescent or radiolabelled substrate probe that can be quantitatively measured is added to each well in the presence and absence of each test compound to determine the transporter-inhibiting activity of the compounds. The concentration of the substrate probe and uptake time that produce the best Z′ assay sensitivity factor is selected; a Z′ > 0.8 is ideal for HTS, although Z′ > 0.5 is also acceptable⁴⁹. Such methods have been used to screen thousands of compounds in HTS campaigns. However, HTS also suffers from limitations. One commonly observed issue relates to assay interference, whereby aggregation of a test compound that causes protein sequestration or nonspecific inhibition results in a false-positive hit; this problem affects 1.7–1.9% of total compounds in compound libraries^{50, 51, 52}. Another problem arises from the interference of fluorescent test compounds or contaminating particulate matter with the measurement of the fluorescent substrate probe. A ‘pre-read’ control measurement after the addition of the test compound but before substrate addition can prevent this problem. Use of an orange or red fluorophore as the substrate also helps to avoid this interference, as fluorescence and spurious light emissions from test compounds have been found to be more likely to occur at shorter wavelengths⁵⁰.

The structure of identified potential inhibitors can be modified to produce more-potent or more-selective molecules. Pharmacophore modelling and quantitative structure–activity relationship (QSAR) modelling may be used to engineer improved compounds with more-desirable properties^{53, 54}. Novel inhibitors for many transporters have been identified through these methods, including inhibitors of apical sodium-dependent bile acid transporter (ASBT; also known as SLC10A2)^{55, 56}, excitatory amino acid transporter 3 (EAAT3; also known as SLC1A1)⁵⁷, dopamine transporter (DAT; also known as SLC6A3), serotonin transporter (SERT; also known as SLC6A4), noradrenaline transporter (NET; also known as SLC6A2)^{58, 59, 60} and vesicular glutamate transporter 2 (VGLUT2; also known as SLC17A6)⁶¹. For example, the O-spiroketal C-arylglucoside scaffold — which forms the basis of tofogliflozin (Apleway/Deberza; Chugai Pharmaceutical/Sanofi/Kowa, a selective sodium–glucose cotransporter 2 (SGLT2; also known as SLC5A2) inhibitor recently approved in Japan for the treatment of diabetes — was found by pharmacophore modelling of SGLT2 inhibitors and a search of the Cambridge Structural Database⁶² (see Databases).

Homology modelling and docking have been widely used to identify and optimize lead compounds. As the crystal structures of SLC transporters from various species become increasingly available^{3, 63}, homology models of human orthologues are being constructed that can be used in molecular docking studies to design more-selective transporter inhibitors. These methods have been successfully used to identify binding sites, to accurately predict the inhibitory actions of already-prescribed drugs on SLC transporters and to identify potent transporter inhibitors for treating diseases^{5, 11, 64}. For example, high-throughput docking of the glutamate-binding site on the Pyrococcus horikoshii glutamate transporter homologue was used to identify a potent inhibitor of EAAT2 (also known as SLC1A2)⁶⁵, and the crystal structure of the same concentrative nucleoside transporter from Vibrio cholerae was used to identify a selective anticancer drug^{66, 67}. The crystal structures of bacterial homologues of neurotransmitter transporters bound to antidepressant drugs were recently elucidated, enabling researchers to identify the amino acid residues that comprise the binding sites of these transporters^{68, 69, 70}. Such information may be implemented to design and synthesize more-potent serotonin- and/or noradrenaline-reuptake inhibitors⁷¹.

Enhancement of transporter function

For most Mendelian diseases in which causal variants of the genes encoding SLCs result in loss of transporter function, compounds that enhance the function of the affected transporter are needed; these compounds are also of interest in the treatment of common diseases. To our knowledge, no current drugs were originally developed specifically to activate SLC transporters. Riluzole, a drug used to treat amyotrophic lateral sclerosis, inhibits glutamate release, thereby preventing glutamate-induced activation of sodium channels on postsynaptic neurons⁷², reducing excitotoxicity. However, riluzole was subsequently found to enhance the transport activity of EAAT1 (also known as SLC1A3), which takes up extracellular glutamate from the synapse and thus reduces glutamate levels in the synaptic cleft^{73, 74}. Another study showed that riluzole stimulates glutamate uptake by increasing excitatory amino acid carrier 1 (EAAC1; also known as SLC1A1) levels in astroglial cells⁷⁵, although the exact mechanism for this effect is not completely understood.

There is interest in identifying additional targets to increase glutamate clearance. Similar to EAAT1, EAAT2 (the principal glutamate transporter in astrocytes) reduces glutamate levels in the synaptic cleft, thereby attenuating excitotoxicity. Some examples of neurodegenerative diseases that could benefit from upregulated EAAT2 function include Alzheimer disease, amyotrophic lateral sclerosis and Parkinson disease — all of which have been associated with decreased EAAT2 protein expression levels^{76, 77} — as well as epilepsy, stroke and neurotrauma⁷⁸. Therefore, developing a drug that enhances EAAT2 activity represents a promising approach in central nervous system (CNS) drug development⁷⁹. Two approaches are currently being used to identify enhancers of EAAT2. First, an HTS assay that used a reporter gene containing the SLC1A2 promoter upstream of the luciferase gene found that several currently prescribed drugs, such as ceftriaxone, increased luciferase transcript levels⁸⁰. Second, an astrocyte-based enzyme-linked immunosorbent assay has been developed to screen large compound libraries for molecules that can induce translation of silenced SLC1A2 transcripts, as the expression of EAAT2 protein is highly regulated at the translational level by extracellular factors⁸¹.

These and other screening technologies — for example, high-throughput fluorescence assays — could be used to identify compounds that enhance SLC transporter activity. The development of drugs that enhance the activity of cystic fibrosis transmembrane conductance regulator (CFTR; also known as ABCC7), the ABC transporter that is defective in cystic fibrosis, could provide a model for the identification of small molecules that may rescue mutated SLC transporters. Small-molecule therapies for modulating CFTR fall into two categories: potentiators and correctors. Various assays have been developed to identify potentiators and correctors, such as ivacaftor, the first US-regulator-approved drug to target a specific CFTR mutant, CFTR-G551D. Ivacaftor acts as a potentiator; it increases the activity of CFTR-G551D that is already present on the plasma membrane, by increasing the probability of the open state of the CFTR channel⁸². By contrast, correctors enhance the trafficking of mutant proteins such as the most common cystic-fibrosis-causing mutant protein, CFTR-F508del, to the plasma membrane. Early clinical results suggest that the combination of a corrector and a potentiator results in enhanced efficacy in the treatment of individuals with cystic fibrosis who harbour the CFTR-F508del mutation⁸³.

Circumventing transporters

Another potential therapeutic strategy for treating rare diseases caused by mutant SLC transporters is to develop substrates that circumvent transporters. SLC transporter gene mutations that lead to Mendelian disease often also lead to subnormal levels of an essential nutrient, such as thiamine or carnitine, in specific tissues. Therapies that circumvent the use of the mutant transporter could be developed. For example, creation of a hydrolysable hydrophobic derivative of an essential nutrient that is normally hydrophilic may result in enhanced permeability of the nutrient. Thus, the hydrophobic derivative would enter the cell (without the aid of a transporter) where it would undergo hydrolysis and release the nutrient, thus enhancing the availability of the nutrient in the cell to treat, for instance, thiamine deficiency⁸⁴.

However, a potential problem in the development of diffusible analogues that circumvent transporters may be the low cellular availability of the analogue. For example, many transporters are secondary-active and therefore concentrate their substrates intracellularly, whereas diffusible analogues may not accumulate to the same extent. Furthermore, hydrophobic diffusible analogues may be substrates for efflux pumps of the ABC transporter families, such as MDR1, that preferentially handle hydrophobic substrates, and may therefore exhibit poor cellular availability. Finally, diffusible analogues may distribute widely into body tissues rather than selectively targeting the tissues of greatest need (for example, in the case of carnitine deficiency, the liver).

Gene therapy

Gene therapy, in principle, may be a strategy for the treatment of diseases caused by mutant transporter proteins. Several animal studies have provided proof of concept for this strategy. For example, Yiu et al.⁸⁵ infused an adenoviral vector containing the gene encoding human glucose-6-phosphate transporter (SLC37A4) into Slc37a4-deficient mice (which recapitulate the human glycogen-storage disease type Ib that is attributable to mutations in SLC37A4). The gene therapy restored SLC37A4 expression in various tissues (including the liver), normalized serum glucose and serum lipid profiles, and reduced glycogen deposition in the liver. Although historically fraught with issues, gene therapy may become a viable strategy for the treatment of Mendelian disorders that involve defective membrane transporters as more-advanced methods are applied.

SLC12 transporters as targets of diuretic drugs

Diuretics are important drugs used to treat hypertension and heart failure. The discovery of the first diuretic, chlorothiazide (which belongs to the thiazide class), in 1957, was followed by the discoveries of the loop diuretics bumetanide and furosemide. Subsequently, Wiley and Cooper⁹³ observed that furosemide inhibits the influx and efflux of sodium and potassium in human red blood cells and concluded that the drug inhibits the cotransport of sodium and potassium. In addition, using isolated perfused rabbit kidney tubules, Burg et al.⁹⁴ described the inhibitory effect of furosemide on active chloride transport in the thick ascending limbs of the loop of Henle, which results in a decreased net absorption of sodium chloride and a decreased electrical potential (that is, a more positive electrical potential) in tubule cells. Sodium–chloride cotransport was recognized in the 1970s and has since been studied in a wide variety of animal cells and tissues.

The role of a sodium–chloride cotransporter is to maintain and regulate cell volumes and ion gradients; thus, diuretics that inhibit such cotransporters in the renal tubules decrease extracellular fluid volume and regulate electrolyte levels in the body⁹⁵. The functional expression of the sodium–potassium–chloride cotransporter in HEK293 kidney cells was first revealed by Xu et al.⁹⁶, who used radiolabelled rubidium as a tracer for potassium movement and showed that cells transfected with the sodium–potassium–chloride cotransporter had a greater influx of ⁸⁶Rb, which was inhibited by bumetanide, a loop diuretic. Subsequently, it was established that diuretics inhibit the sodium–potassium–chloride cotransporter NKCC2 (also known as SLC12A1) to reduce sodium reabsorption in the thick ascending limb of the loop of Henle. The reduction in sodium reabsorption results in increased sodium levels in the tubule fluid and a corresponding loss of body water, thus increasing urine production and decreasing blood volume.

Interestingly, mutations in the gene encoding NKCC2 (SLC12A1), which is expressed in the kidney, were found to cause Bartter syndrome, an inherited hypokalaemic alkalosis featuring hypercalciuria and severe blood volume depletion⁹⁷. Moreover, mutations in SLC12A3, which encodes the sodium–chloride cotransporter TSC (normally expressed in the renal distal tubule) were found to cause Gitelman syndrome, in which patients present with hypokalaemic alkalosis and abnormalities in electrolyte homeostasis⁹⁸. In contrast to NKCC2, which is sensitive to loop diuretics such as bumetanide but insensitive to thiazides, TSC is sensitive to the thiazides chlorothiazide and hydrochlorothiazide but is insensitive to loop diuretics.

At high doses, thiazides are used to relieve systemic and pulmonary oedema due to chronic heart failure, and at lower doses to reduce blood pressure. One of the major adverse effects of thiazides and loop diuretics is hypokalaemia, which is especially dangerous in patients with severe cardiovascular disease. Furthermore, high doses of thiazides can also cause hyperuricaemia. One possible explanation for this effect is that diuretics may inhibit renal transporters such as OAT1 (SLC22A6), OAT3 (also known as SLC22A8), sodium–phosphate cotransporter 1 (NPT1; also known as SLC17A1) and NPT4 (SLC17A3), which are all involved in uric acid secretion, and thus increase serum uric acid levels^{99, 100}.

Urea transporters of the SLC14 family also play an important role in the recycling of urea and in concentrating urine. As a result, several research groups are screening for small-molecule inhibitors to target urea transporters, particularly UTB (also known as SLC14A1) and UTA (also known as SLC14A2). Such inhibitors cause an increase in urine output due to decreased urea concentrations in the urine. Perhaps in the near future, a new class of diuretics that inhibit UTB and UTA transporters will be developed.

SLC6 transporters as neuropsychiatric drug targets

Drugs for treating depression can be classified according to their presumed targets, which may be transporters. For example, the monoamine-reuptake inhibitors — which include the serotonin-selective reuptake inhibitors (SSRIs) and noradrenaline-reuptake inhibitors (NRIs) — inhibit neurotransmitter transporters. Over the past several decades, the development of novel antidepressants has moved from ‘discovery by chance’ to single-target strategies, and then to multiple-target strategies.

Several transporters in the SLC6 family have important roles in the uptake of monoamines in the synapses of the CNS. These include NET (SLC6A2), DAT (SLC6A3) and SERT (SLC6A4), which are transporters for noradrenaline, dopamine and serotonin (also known as 5-hydroxytryptamine), respectively, although they all exhibit overlapping substrate specificity. Inhibition of these transporters by drugs reduces the clearance of monoamine neurotransmitters from the synapse, thus increasing their dwell time in the synaptic cleft. The resulting increased concentrations of monoamines in the synaptic cleft enhance receptor occupancy, leading to increased activation of ligand-gated ion channels and modulation of G-protein-coupled receptor signalling¹⁰¹. However, the downstream mechanisms by which such drugs ultimately exert antidepressant effects (which can take weeks to become apparent) are not yet clear¹⁰².

The first-generation tricyclic antidepressants (TCAs) inhibit serotonin and noradrenaline reuptake, but also interact with other targets in the CNS. In the 1960s, several investigators showed that TCAs such as imipramine competitively inhibit serotonin and noradrenaline uptake in several tissue types, including platelet-rich plasma, brain slices and synaptosomes. TCAs were used as the primary treatment for depression until the US approval of the first SSRI, fluoxetine, in 1987. Fluoxetine was found to inhibit serotonin uptake into rat brain synaptosomes, without inhibiting noradrenaline uptake into rat hearts^{103, 104}.

Since the discovery of fluoxetine and other SSRIs, other classes of compounds that affect neurotransmitter reuptake have been developed. For example, serotonin–noradrenaline reuptake inhibitors (SNRIs), such as venlafaxine, are a newer class of antidepressants that selectively target two transporters, NET and SERT. The use of SSRIs and SNRIs has also expanded to other neuropsychiatric disorders, including anorexia nervosa, bulimia nervosa, obsessive-compulsive disorder, panic disorders and anxiety disorders, as well as for relief of menopausal symptoms.

Another therapeutic target for the treatment of neuropsychiatric disorders is vesicular monoamine transporter 2 (VMAT2; also known as SLC18A2), which transports monoamines from the cellular cytosol into synaptic vesicles. An inhibitor of VMAT2, tetrabenazine, is currently US-approved for the treatment of hyperkinetic disorders associated with Huntington disorder¹⁰⁵. Other VMAT2 inhibitors are in clinical development, including NBI-98854 (developed by Neurocrine Biosciences), which is in Phase III trials for dyskinesia (ClinicalTrials.gov identifier: NCT02274558; note that all trials cited in this article are from this database (see Databases)) and a natural product, lobeline (developed by Yaupon Therapeutics), which is in Phase II trials for the treatment of attention-deficit hyperactivity disorder in adults (NCT00664703). VMAT2 inhibitors may also be developed for the treatment of psychostimulant abuse and addiction¹⁰⁶.

Glucose transporter inhibitors

The SGLT2 (SLC5A2) transporter, which has a low affinity and high capacity for glucose, plays a major part in renal glucose reabsorption, whereas the high-affinity, low-capacity SGLT1 (also known as SLC5A1) transporter has a major role in glucose absorption in the small intestine¹⁰⁷. By mimicking the effects of a loss-of-function mutation in SGLT2 (Table 1), SGLT2 inhibitors enhance renal glucose excretion and consequently lower plasma glucose levels.

Many companies have been developing drugs that target SGLT1 and SGLT2 for type 2 diabetes, in which patients exhibit hyperglycaemia owing to insulin resistance. In such programmes, compound libraries were screened for lead compounds that inhibit glucose uptake in cell lines that stably expressed human SGLT1 or SGLT2 (Refs 108,109). Once the lead compounds were identified, their in vivo effects were determined in murine and other animal models to confirm that the compound could increase urinary glucose excretion by inhibiting renal glucose reabsorption. Importantly, oral glucose tolerance tests in diabetic rats revealed that these animals were responsive to treatment with SGLT2 inhibitors¹¹⁰. Moreover, patients with type 2 diabetes who were treated with the SGLT2 inhibitor empagliflozin showed improved β-cell function and insulin sensitivity. Chronic dosing with empagliflozin for 4 weeks in these patients led to a significant decline in both glycated haemoglobin and fasting glucose levels¹¹¹.

Canagliflozin (Invokana; Mitsubishi Tanabe Pharma/Johnson & Johnson) was the first SGLT2 inhibitor to be approved by the US Food and Drug Administration (FDA) in 2013, and was followed by dapagliflozin (Farxiga/Forxiga; Bristol-Myers Squibb/AstraZeneca) and empagliflozin (Jardiance; Boehringer Ingelheim/Eli Lilly and Company) in 2014. All three drugs are being prescribed as second-line therapy for patients whose diabetes is inadequately controlled by a single antidiabetic agent¹¹². Tofogliflozin and ipragliflozin (Suglat; Astellas/Kotobuki) have both been approved for the treatment of type 2 diabetes in Japan, and both have active Phase IV trials ongoing (tofogliflozin: NCT02201004 and NCT02284269; ipragliflozin: NCT02291874 and NCT02175784). Other SGLT1 and/or SGLT2 inhibitors are being studied in early- and late-phase clinical studies either alone or in combination with metformin (which suppresses glucose production by the liver)^{113, 114, 115}. These SGLT inhibitors include the following: remogliflozin (also known as GSK189075; developed by GlaxoSmithKline), which is in Phase II trials (NCT00495469); GSK1614235 (GlaxoSmithKline), in Phase I trials (NCT01607385); sotaglifozin (Lexicon Pharmaceuticals), in Phase III trials (NCT02384941); and ertugliflozin (Merck &Co./Pfizer), in Phase III trials (NCT01986881, NCT01999218, NCT01958671, NCT01986855, NCT02036515 and NCT02226003). In addition to having notable effects on reducing plasma glucose levels and increasing urinary glucose excretion, SGLT2 inhibitors have also been associated with weight loss and lowering of blood pressure owing to their osmotic diuretic effects¹¹⁶.

Uric acid transporter inhibitors

Gout is the most common inflammatory arthritic disease, and has been increasing in prevalence in the United States and in other countries, partly as a consequence of the obesity epidemic¹¹⁷. Gout is caused by the accumulation of monosodium urate monohydrate crystals in the joints and soft tissue, as a result of hyperuricaemia. Most current treatments for gout generally target the inflammation that occurs during an acute gout attack or reduce uric acid production by inhibiting the enzymes involved (for example, allopurinol, which inhibits xanthine oxidase). However, many patients fail to respond to these treatments or suffer from serious side effects (for instance, allopurinol can cause drug hypersensitivity reactions)¹¹⁸.

Recent advances in the understanding of uric acid transporters, through GWASs and other studies, have opened the door for the development of novel compounds that target these transporters in the kidney as a way to block reuptake of uric acid and thereby promote elimination of uric acid in the urine. The primary uric acid reabsorption transporters include URAT1 (SLC22A12) and glucose transporter type 9 (GLUT9; also known as SLC2A9), which are expressed in the proximal tubule, although other transporters have also been identified that transport urate or are associated with gout^{118, 119}. Although probenecid, which has been used for decades to treat gout, blocks uric acid reabsorption through organic anion transporters, several novel compounds are currently in development for this indication that are more selective, and most target URAT1. Lesinurad, a URAT1 inhibitor developed by Ardea Biosciences/AstraZeneca was recently approved by the European Medicines Agency on the basis of results from two trials testing the drug for the treatment of hyperuricaemia in combination with xanthine oxidase inhibitors in patients with gout (NCT01510158 and NCT01493531). Another URAT1 inhibitor, RDEA3170, is also being developed by Ardea Biosciences/AstraZeneca and has completed Phase II trials (NCT01927198)¹²⁰. Meanwhile, Pfizer recently licensed KUX-1151 from Kissei Pharmaceutical; this dual inhibitor of xanthine oxidase and of URAT1 is currently in a Phase II trial in Japan (NCT02190786). As URAT1 inhibitors selectively inhibit URAT1 and reduce uric acid reabsorption in the kidney, their effects mimic those observed in patients with URAT1 mutations that lead to renal hypouricaemia type 1.

Glycine transporter inhibitors

In the CNS, glycine has important roles in neuronal inhibition and excitation. It functions as an inhibitory neurotransmitter by activating ionotropic glycine receptors, enabling an influx of chloride ions and hyperpolarization of the postsynaptic membrane. Glycine also binds to excitatory N-methyl-D-aspartate (NMDA) receptors to enable receptor activation by glutamate¹²¹. The glycine transporters GLYT1 (also known as SLC6A9) and GLYT2 (also known as SLC6A5) on neurons, astrocytes and glial cells regulate levels of extracellular glycine in the brain, and thereby regulate NMDA receptor activity. According to the glutamate hypothesis of schizophrenia, symptoms of the disease are caused by deficient glutamatergic (NMDA) signalling, and therefore GLYT1 inhibitors have been developed to inhibit glycine reuptake and increase levels of glycine at NMDA receptors to enhance NMDA signalling in this disorder.

Bitopertin (also known as RO4917838), a highly selective and potent GLYT1 inhibitor developed by Roche, is furthest along in development¹²². However, several Phase III studies of bitopertin in schizophrenia (NCT01235559, NCT01235585, NCT01192906 and NCT01192880) failed to reach their primary end point, leaving bitopertin with an uncertain future. Nevertheless, there is a Phase II trial currently ongoing using bitopertin with SSRIs in patients with obsessive-compulsive disorder (NCT01674361). The GLYT1 inhibitor PF-04958242 is still listed in Pfizer’s Phase I pipeline and has completed multiple Phase I studies (NCT01159483, NCT02228395, NCT02332798, NCT01511510 and NCT01518894), but the development of several other investigational GLYT1 inhibitors for the treatment of schizophrenia, including compounds from Pfizer, GlaxoSmithKline, Amgen, Merck, Johnson & Johnson and Sanofi, appears to have stopped (see Evaluate Pharma article 21 Jan 2014 in Further information). However, there is evidence to suggest that GLYT2 inhibitors reduce pain in animal models of chronic pain, and thus there is probably work ongoing in the development of partial GLYT2 inhibitors that can be used for pain relief without serious side effects¹²³. One such compound, VVZ-149, a dual inhibitor of GLYT2 and 5-hydroxytryptamine receptor 2A, is under development by Vivozon and completed a Phase I study in 2014 (NCT01905410).

Bile acid transporter inhibitors

Bile acids are the principal components of bile and their primary function is to solubilize fat and other dietary nutrients, although they can also function as signalling molecules¹²⁴. Bile acids are synthesized in the liver from cholesterol through the action of the rate-limiting enzyme cytochrome P450 7A1. Bile acid uptake and efflux transporters in the intestine and liver play key roles in the highly efficient enterohepatic recycling pathway, through which more than 90% of the total pool of bile acids are reabsorbed from the intestine and transported back to the liver via the portal circulation¹²⁵. ASBT (SLC10A2) mediates the reabsorption of bile acids from the lumen of the intestine¹²⁶. Inhibition of bile acid reabsorption in the intestine would result in a loss of bile acids and increased conversion of cholesterol to bile acid in the liver, thereby lowering circulating cholesterol levels and potentially ameliorating or preventing the progression of cardiovascular disease. Thus, not surprisingly, inhibiting ASBT has received a substantial amount of attention as a potential therapeutic strategy for lowering cholesterol¹²⁷, and several pharmaceutical companies are developing potent and specific inhibitors of ASBT as therapeutic agents with this aim¹²⁶. Interestingly, the rare genetic disorder primary bile acid malabsorption syndrome is caused by loss-of-function mutations in ASBT¹²⁸. Individuals with this disorder suffer from chronic diarrhoea and fat-soluble vitamin deficiency, and importantly, exhibit lower plasma cholesterol levels¹²⁸.

Given the potency and efficacy of statins, as well as newer agents such as ezetimibe (Zetia; Merck & Co.) that target cholesterol absorption in the intestine, the extent of clinical benefit of ASBT inhibition (relative to the potential side effects) remains to be clarified. However, there now appears to be more-compelling evidence for targeting ASBT for cholestatic liver diseases such as primary biliary cirrhosis, for which treatment options have been limited. LUM001, an ASBT inhibitor developed by Lumena Pharmaceuticals (now Shire), is currently in Phase II clinical trials for the treatment of cholestatic liver diseases, including primary biliary cirrhosis¹²⁹ (NCT02057718, NCT01904058, NCT02057692, NCT02117713 and NCT02061540). Another ASBT inhibitor developed by Shire, SHP626, is currently recruiting participants for Phase I trials for the treatment of nonalcoholic steatohepatitis in overweight adults (NCT02287779). Albiero Pharma is also currently developing ASBT inhibitors of its own — elobixibat has just completed a Phase III trial for chronic idiopathic constipation (NCT01895543). A4250 (Albireo) received orphan drug designation from the FDA in 2012 for the treatment of progressive familial intrahepatic cholestasis and primary biliary cirrhosis, and is currently in Phase II trials in patients with primary biliary cirrhosis (NCT02360852). It is likely that there are other pharmaceutical companies pursuing ASBT as a therapeutic target for cholestatic liver disease, as many have done so for reducing cholesterol.

Because bile acids promote the secretion of intestinal incretin peptides (such as glucagon-like peptide 1) that in turn stimulate pancreatic insulin secretion, inhibition of ASBT may also be a viable strategy for the treatment of type 2 diabetes¹³⁰. GlaxoSmithKline recently completed two Phase II studies using a combination of GSK2330672 plus metformin to treat diabetes (NCT02202161 and NCT01929863). Furthermore, targeting ASBT has also been proposed as a prodrug strategy for compounds that have poor oral bioavailability, as coupling an active drug molecule with a bile acid moiety would increase the absorption of the drug in the intestine¹³¹.

Nutrient transporter inhibitors

Tumour cells have an increased demand for nutrients owing to their rapid proliferation and growth. However, many amino acids and carbohydrates are too polar to diffuse across cell membranes, and thus require transporter proteins for cellular uptake. In theory, tumour growth can be controlled by starving the cells of the metabolic precursors needed for sustained growth. Such an approach can be used to selectively kill tumour cells, which are generally locked into a state of rapid growth by oncogenes and are therefore more sensitive to starvation than are normal cells. Moreover, nutrient transporters have been found to be overexpressed in a wide variety of cancers¹³². To date, no anticancer drug has been developed to specifically inhibit a nutrient transporter with the explicit goal of controlling cell growth. However, inhibition of key nutrient transporters, such as the glucose transporters of the SLC2 and SLC5 families, amino acid transporters of the SLC7 family and lactate transporters of the SLC16 family, may be a reasonable approach for discovering new anticancer drugs¹³³.

As glucose is the primary energy source in most cells, it is unsurprising that glucose transporters such as GLUT1 (SLC2A1) are often overexpressed in tumour cells in a variety of tissues^{134, 135}. Indeed, [¹⁸F]-fluoro-2-deoxy-D-glucose (¹⁸F-FDG) positron emission tomography (PET) imaging, which is used to detect and stage tumours, is dependent on the rate of cellular glucose uptake. However, increased glycolysis is associated with an increase in the generation of lactate, which cells need to remove to maintain cellular pH levels. There is often also an upregulation of lactate transporters (such as monocarboxylate transporter 1 (MCT1; also known as SLC16A1))^{136, 137}, or transporters important in pH regulation (such as sodium–hydrogen exchanger 1 (NHE1; also known as SLC9A1))^{138, 139} in cancer cells.

Various research groups have identified MCT1 inhibitors. For example, the MCT1 inhibitor AZD3965, developed by Cancer Research UK, is currently in Phase I clinical trials for patients with advanced solid tumours or lymphoma (NCT01791595). Similarly, amino acid transporters (such as L-type amino acid transporter 1 (LAT1; also known as SLC7A5)^{140, 141}, ASC amino acid transporter 2 (ASCT2; also known as SLC1A5)¹⁴² and cystine–glutamate exchange transporter (XCT; also known as SLC7A11)¹⁴³) are also overexpressed in a variety of tumours and may be potential anticancer targets. Sulfasalazine, which was initially developed as an anti-inflammatory drug to treat rheumatoid arthritis, was subsequently found to inhibit XCT¹⁴⁴ and has anticancer effects in various cancer xenograft models¹⁴³. As our understanding of the importance of transporters in tumour cell metabolism increases, it is highly likely that the development of anticancer drugs that target different nutrient transporters will continue.