Re: Amino Acid Transporters(아미노산 수송체) as Disease Modifiers and Drug Targets

[문형철]

beyond reason

아미노산은 세포와 기관에서 에너지 대사, 신경전달물질, 많은 물질의 전구체로 작용하여 다양한 기능을 함. 

아미노산 수송체는 이러한 다양한 기능을 하는데 중요한 역할을 함.

아미노산 수송체의 억제는 당뇨, 신경질환, 암 등에서 치료전략으로 사용됨. 

Amino Acid Transporters as Disease Modifiers and Drug Targets

Stefan Bröer

First Published March 20, 2018 Review Article Find in PubMed

Abstract

Amino acids perform a variety of functions in cells and organisms, particularly in the synthesis of proteins, as energy metabolites, neurotransmitters, and precursors for many other molecules. Amino acid transport plays a key role in all these functions.

Inhibition of amino acid transport is pursued as a therapeutic strategy in several areas, such as diabetes and related metabolic disorders, neurological disorders, cancer, and stem cell biology. The role of amino acid transporters in these disorders and processes is well established, but the implementation of amino acid transporters as drug targets is still in its infancy.

This is at least in part due to the underdeveloped pharmacology of this group of membrane proteins. Recent advances in structural biology, membrane protein expression‎, and inhibitor screening methodology will see an increased number of improved and selective inhibitors of amino acid transporters that can serve as tool compounds for further studies.

Keywords solute carrier, diabetes, cancer, drug development

Introduction

Amino acids are key metabolites that are essential for cells and organisms. They form the building blocks for the synthesis of proteins, they are used as energy metabolites, and they are or contribute to the biosynthesis of signaling molecules, such as peptide hormones and neurotransmitters. They also contribute to the synthesis of many other molecules, such as nucleotides, creatine, polyamines, glucosamines, and related carbohydrates, and are the principal generators of C1 compounds.1

In view of this diversity of roles, it is not surprising that amino acid homeostasis is highly regulated and that all cells maintain a pool of the 20 proteinogenic amino acids at any given time.2 Mammalian cells can synthesize nonessential amino acids (alanine, arginine, asparagine, aspartate, cysteine, glutamate, glutamine, glycine, proline, and serine), but need to acquire essential amino acids from nutrition (histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine). Cellular protein, in particular muscle protein, is the storage site for amino acids in organisms. Most cells can survive several hours without amino acids, inducing autophagy to degrade cellular protein for the synthesis of essential proteins.3 Protein degradation is combined with efficient mechanisms to recycle the generated amino acids; thus, net loss of amino acids is limited.

Transporters play a key role in the acquisition of amino acids by the whole organism from the nutrition and in the acquisition of amino acids by individual cells from blood plasma.2,4,5

They are several aspects of human physiology and pathophysiology where amino acid transporters are considered disease modifiers and drug targets (Fig. 1 and Table 1).5–9 These are

• - Metabolic disorders, such as diabetes and obesity

• - Neurological disorders

• - Cancer

• - Embryo development and stem cells

Figure 1. Current concepts of amino acid transporters as drug targets. Reducing uptake of amino acids in the intestine causes protein restriction and can improve metabolic health. Uptake into cells can be blocked to increase signaling, particularly when amino acids serve as neurotransmitters or neuromodulators. Blocking uptake of essential amino acids can be used to reduce the growth of malignant cells. In vitro amino acid supplementation and transporter modulation can be used to optimize conditions for embryo transfer and implantation.

Table 1. Amino Acid Transporters Covered in This Review and Their Function.

View larger version

Three approaches are generally used to influence amino acid homeostasis (Fig. 1). First is the blockage of transporters to induce amino acid restriction or starvation. This is used as a possible treatment to reduce the growth of cancer cells,10–12 but also to induce protein restriction as a novel attempt to treat diabetes.13 Second is the blocking of transporters to increase the concentration of amino acids in the interstitial space. This is often used to treat neurological disorders, where amino acids such as glutamate and glycine act as neurotransmitters.7,8,14 Third is the use of transporters to selectively provide amino acids to certain cells. This approach is used less frequently but is particularly interesting with regard to embryo development and stem cell biology.15–22 The latter could be complemented by compounds to induce transporter function or expression‎, but this has not been translated to date.

Principles of Transporter Inhibition

Biomembrane transporters, including amino acid transporters, are attractive drug targets for several reasons.

First, they are located at the cell surface and, similar to receptors, are easily accessible by drugs from the bloodstream.23 Notable exceptions are transporters in the brain, which are shielded by the blood–brain barrier against most drugs,24–26 and transporters in organelles, such as lysosomes or synaptic vesicles.

Second, as part of their function, amino acid transporters have deep substrate binding pockets, which often reach halfway through the membrane (Fig. 2A). The presence of a substrate binding site with a vestibule allows manifold interactions and bond formation to stabilize binding of an inhibitor.27 In contrast to enzymes, where transition-state analogues are considered to be the best inhibitors,28 ground-state analogues are preferred as transport inhibitors.29 The ground state of a transporter for drug binding is the outside open conformation.30 In contrast to the majority of enzymes where the substrate adopts a transition state through noncovalent interactions, transporters form a transition state around the substrate (induced fit), while the substrate remains in its original conformation.31,32 The transition state is called the occluded conformation, in which the transporter protein fully encloses the substrate (Fig. 2B).33 Thus, only compounds that are very similar to the substrate itself fit into this cavity. Accordingly, transport inhibitors typically contain a substrate-like fragment to which a large fragment is coupled, often containing aromatic rings. Examples of this design are shown in Figure 3. The rationale behind this design is the wedge effect exerted by the large aromatic fragment, which blocks the transition of the transporter to the occluded conformation. The substrate-like fragment ensures recognition and specificity; the wedge fragment provides additional interactions with the transporter to increase affinity and at the same time blocks adoption of the transition state. This design is well suited to a medicinal chemistry approach.34 However, transport inhibitors based on substrate analogues have two potential problems. First, inhibitors may have problems binding to the transporter in vivo in the presence of high concentrations of substrates. The combined concentration of amino acid substrates of a specific transporter in blood serum can easily reach 1 mM, against which a competitive inhibitor has to find its target. Second, substrate-like inhibitors may show lack of isoform specificity. For instance, dl-threo-β-benzoylaspartate (TBOA) inhibits all glutamate transporter isoforms (Fig. 3).35 Allosteric inhibitors, by contrast, are not displaced by substrates and can be isoform specific. For instance, UCPH101 (Fig. 4A) is a specific inhibitor of the glutamate transporter EAAT1, due to its binding between the scaffold and transporter domain of EAAT1 (Fig. 4B), which has a less conserved sequence than the substrate binding site.36 Allosteric inhibitors are more likely to arise from high-throughput screening (HTS) or specialized screens,37 but can be further modified through medicinal chemistry. The betain-γ-amino butyric acid (GABA) transporter BGT1, for instance, can be specifically inhibited by N-(1-benzyl-4-piperidinyl)-2,4-dichlorobenzamide (Fig. 4A), which binds to the vestibule of the transporter without penetrating the binding site.38 Allosteric transporter inhibitors have also been identified by screening cyclic peptide libraries.39

Figure 2. Structure of amino acid transporters and their binding sites. (A) The outward-open conformation of the leucine transporter from Aquifex aeolicus (LeuTAa) is shown (PDB ID 3TT1). The cavity that leads to the substrate binding site is visible in the center of the molecule. (B) The occluded state of LeuTAa (PDB ID 2A65) is shown from the side and sliced to reveal the substrate binding site, which encloses the substrate (magenta) tightly. The vestibule is visible at the top.

Figure 3. Examples of transport inhibitors based on substrate analogues. Three amino acid transport inhibitors are shown. The substrate-like part of each inhibitor is highlighted in magenta. Bulky side chains (black) prevent the occlusion of the substrate by the transporter. See references 12, 129, and 181 for the depicted inhibitors.

Figure 4. Allosteric inhibitors of amino acid transporters. (A) Structures of allosteric transport inhibitors are unrelated to the substrate. Top: UCPH101, a specific EAAT1 inhibitor. Bottom: The BGT-1 inhibitor BPDBA is thought to bind an allosteric secondary binding site observed in some transporters of the SLC6 family. (B) The binding site of the EAAT1 inhibitor (UCPH101) was resolved by x-ray crystallography and is located at a distance from the substrate binding site.182 The substrate is shown in the center of the transporter, while the inhibitor is located between the translocation and scaffold domain of the transporter (lower right). The inhibitor glues the two domains together, thereby inhibiting transport.

Methods to Identify Transporter Inhibitors

Medicinal chemistry is one of the major tools to design transport inhibitors, usually starting with one of the substrates of the transporter or a close analogue. The reader is referred to a comprehensive review of the medicinal chemistry of α-amino acids for further information.34 To identify substrate-unrelated starting points for medicinal chemistry, HTS is increasingly used in academic environments to identify transport inhibitors. The most frequently used techniques for transporters are briefly described in the following.

Accumulation of Radiolabeled Substrate

Although radiolabeled compounds are less frequently used in HTS environments due to regulatory reasons, inhibition of radiolabeled substrate uptake is still the most robust and quantitative assay to study transport inhibitors. Moreover, it is the gold standard to identify false-positive hits from other screening methods described below and to characterize the mode of inhibition. Typically, a cell line stably expressing the transporter of choice is washed with isosmotic buffer, and subsequently incubated with radiolabeled substrate for 5–30 min in the presence or absence of inhibitors. After washing the cells to remove extracellular radiolabel, the amount of radioactivity is measured in cell extracts. This method has been used to screen for inhibitors of glycine transporter (GlyT1)40 and neutral amino acid transporter asc-1.41 The method can also be used in conjunction with reconstituted protein in proteoliposomes, in which case filtration is used to separate the proteoliposomes from radiolabeled compounds after incubation.42

Scintillation Proximity Assay

In this assay, the radioactive substrate of a transporter is brought into proximity with a scintillant. Weak β-particles are strongly quenched in water; thus, only radioactive compounds in close proximity to the scintillant generate light flashes, which can be recorded with a normal scintillation counter. Several variations of this assay are available, depending on whether cell lines, membrane preparations, or reconstituted transporters are used. Cells can be seeded out onto specific multiwell plates with a scintillating base.43 Uptake of the radiolabeled substrate into the cells brings the radioactivity into proximity of the scintillant generating light emissions, which are recorded. In a variation of this procedure, suspended cells are brought into contact with polylysine-coated scintillation proximity assay (SPA) beads, also resulting in an uptake-dependent scintillation signal.41 For membrane fragments or reconstituted protein, beads are used that are coated with polylysine to bind membrane lipids. Purified transporters in detergent micelles have also been characterized by this method using copper chelate–coated beads to immobilize a tyrosine transporter through its His tag.44

Membrane Potential–Sensitive Dyes

These can only be applied to electrogenic transporters and require significant transport activity. Voltage-sensitive dyes have been used for many years to study transport and ion channels in membrane vesicles and adherent cells.45 The dyes are membrane permeable and therefore can be washed out. The assay has been improved by the addition of specific dyes to quench signals occurring outside the cell (U.S. patent 6,420,183, EPO patent 0 906 572) and is available commercially as FLIPR Membrane Potential Assay. A cell expressing the selected transporter is required, although expression‎ can be endogenous or heterologous. Endogenous transporters may interfere with the signal generated by heterologous expressed transporters. The assay has been successfully used to screen for inhibitors of several amino acid transporters, such as the neutral amino acid transporter SLC6A19,46 glycine transporters GlyT147 and GlyT2,48 and glutamate transporters EAAT1–3.49 In the case of SLC6A19, several factors were important to optimize the signal. First, heterologous expression‎ of both SLC6A19 and its traffic subunit TMEM27 was required for significant expression‎. CHO cells were used as a host cell line because it had no measurable Na+-dependent uptake for isoleucine, one of the preferred substrates of SLC6A19. While a tailored platform exists for the FLIPR assay, it can be implemented on any suitable fluorescent plate reader.

Ion-Sensitive Dyes

Three different sodium dyes have been eval‎uated in prostate cancer cell lines, namely, benzofuran isophthalate (SBFI), CoroNa Green (Corona), and Asante NaTRIUM Green-2 (ANG-2).50 While calibration was robust in the three chosen cell lines, the response of each cell line to changes in the extracellular Na+ concentration was quite different. Inhibition of the Na+/K+-ATPase by Ouabain caused a significant increase of intracellular Na+ that was readily detected by SBFI and ANG-2, but not by Corona. Interestingly, the addition of veratridine to open Na+ channels did not result in a detectable increase of the intracellular Na+ concentration. Although principally suitable for HTS, the author is not aware of any study that has used Na+-sensitive dyes to screen for inhibitors of Na+ cotransporters. The pH-sensitive dye BCECF has been used in Caco-2 cells to monitor H+-peptide cotransport via the peptide transporter SLC15A1.51 BCECF is a ratiometric dye and can be preloaded as an AM ester. The ratiometric mode of measurement is useful when BCECF is leaking out of the cells. BCECF has not been used in combination with amino acid transporters.

Biosensors

Glutamine transport has been analyzed in living cells using a genetically encoded biosensor, for instance, in Cos-7 cells transiently expressing the glutamine transporter SLC1A5. The biosensor is based on the Escherichia coli glutamine binding protein, glnH.52 Förster resonance energy transfer (FRET) occurs between monomeric teal fluorescent protein (mTFP) 1 and the yellow fluorescent protein venus. FRET-based arginine biosensors have been generated based on the arginine repressor/activator ahrC gene from Bacillus subtilis53 and through protein engineering of bacterial amino acid binding proteins.54 A biosensor for methionine and branched-chain amino acids (BCAAs) has been developed using the leucine-responsive protein Lrp of Corynebacterium glutamicum.55

Mass Spectrometry

Amino acid uptake via a transporter can be monitored by mass spectrometry (MS) either detecting accumulation of the natural amino acid or using stable isotopes. Classical techniques, such as gas chromatography–mass spectrometry (GC-MS) or liquid chromatography–mass spectrometry (LC-MS), are less suitable for large-scale inhibitor screening due to the slow chromatography step. With the development of high-resolution mass spectrometers, complex mixtures of metabolites from cell extracts can be analyzed in a single step or following simple extraction. Typically, electrospray ionization or matrix-assisted laser desorption ionization are used in combination with a quadrupole time-of-flight (Q-TOF) mass spectrometer to analyze liquid samples directly or after extraction using solid-phase extraction cartridges. This methodology has been used to identify novel inhibitors of P-glycoprotein.56 The development of echo acoustic liquid handling by Labcyte Inc. (San Jose, CA) will provide further improvement of sample preparation and injection for MS.57

Docking Studies

Computational screening has become more feasible due to the increasing number of available transporter structures.58 Several obstacles need to be overcome for computational approaches to become competitive to HTS. First, very few mammalian amino acid transporter structures have been resolved, particularly in the outside open conformation, which is the most relevant structure for inhibitor identification.36,59 As a result, most docking studies rely on homology models.46,60,61 Second, docking studies are still relatively slow. Even with generous computation time, only small libraries (tens of thousands of compounds) can be screened. Larger screens make use of a prescreen using a pharmacophore similarity screen.62 This allows reducing very large libraries to smaller compound libraries, but at the same time decreases the likelihood of finding novel structures. Third, docking studies have a strong bias toward larger molecules, because an increase of interaction surface automatically increases the docking score. Larger molecules often do not comply with Lipinski’s rule of five, thereby reducing their suitability as starting points for medicinal chemistry.63 Moreover, molecular dynamics simulations are required to test accessibility of the compounds to the binding site.64 Thus far, only low-affinity compounds have been identified through computational screens for amino acid transport inhibitors, which require significant modification to improve binding.46,60,61

Methods to Characterize Inhibitor Action in Detail

The methods described above are frequently used to screen for inhibitors. Further characterization of the mechanistic action of inhibitors can require additional methods. As outlined above, noncompetitive inhibitors can provide better specificity and resistance against substrate competition. Electrophysiological techniques, such as voltage-clamp recordings, are frequently used to monitor electrogenic transporters, but on occasion can also be used to monitor transport-associated currents in electroneutral transporters. Electrophysiological techniques can be applied to transfected cells and Xenopus laevis oocytes expressing recombinant transporter mRNA. Rapid perfusion techniques and the use of caged compounds that can be released upon light exposure are particularly valuable to analyze binding events during the transport cycle. Moreover, electrophysiology can be combined with fluorescent reporter groups that are attached to mobile parts of the transporter structure. An extensive review of these methods has been compiled by Grewer et al.65 In the case of transporters, electrophysiological methods report currents from a large number of transporters, particularly in oocytes. For reasons that are not fully elucidated, transporters expressed in X. laevis oocytes are often refractory to inhibition by hydrophobic compounds, thus resulting in IC50 values that are significantly higher than in cultured cells.46,66,67 If purified protein is available, proteoliposomes can be used to study transporter inhibition. Proteoliposomes are obtained by detergent removal from liposome–detergent–protein mixtures.42,68 Due to the curvature of the vesicles, incorporation of protein is often asymmetric, allowing the design of efflux or influx experiments.

More recently, single-molecule techniques have been developed to follow transport processes in individual molecules using FRET.69 Typically, these require the introduction of cysteine residues into cys-less or cys-reduced versions of the native transporter, which are then modified with donor and acceptor fluorophores. These methods can be used to study transport inhibitors. Inhibitors can provide valuable insight for these studies because they can lock transporters in a particular state, which is recognizable due to the movement of fluorophores or the lack thereof.69

Role of Transporters in Nutritional Disorders, Such as Diabetes and Obesity

Several studies have reported an association between obesity and polymorphisms in the gene encoding the general amino acid transporter SLC6A14.70–72 The contribution to obesity of this transporter appears to be weak, with allele frequencies showing only a few percent differences in normal and obese individuals. The result suggests only a small influence of SLC6A14 on the obesity phenotype in general. A mechanism has not been established but could be related to amino acids contributing to appetite regulation in the pituitary gland.73 Alternatively, it could provide tryptophan for serotonin production, a neurotransmitter involved in appetite control. Another scenario relates to the function of SLC6A14 in fetal development.74 Reduced amino acid provision to the fetus could result in epigenetic programming, which in turn could foster inappropriate eating behavior later in life.

SLC6A19 is the major transporter for neutral amino acids in the small intestine and in the proximal tubule of the kidney.75 Elevated levels of BCAAs have been detected in individuals with type 2 diabetes and are thought to be strong predictors of future onset of type 2 diabetes.76–79 Mechanistically elevated levels of BCAAs are thought to cause insulin resistance over the long term by a variety of mechanisms.80,81 Leucine is an important regulator of mTORC1, which through p70S6 kinase can cause insulin resistance by phosphorylation of insulin receptor substrate. Probably more relevant is the effect of BCAA metabolites on insulin resistance.76,77,79 Carnitine esters of branched-chain fatty acids have been identified as possible causes of insulin resistance79 and branched-chain keto acids that are generated after transamination of BCAAs appears to reduce GLUT4-mediated uptake of glucose into muscle after insulin stimulation.82 On a broader scale, low-protein diets promote general metabolic health and possibly longevity,83 and so does methionine restriction.84 More importantly, gastric bypass surgery has dramatic effects on insulin sensitivity within days of surgery, long before significant weight loss occurs.85 At least some of these effects are thought to arise from reduced protein uptake. Together, this suggests that reducing amino acid absorption could be an avenue to treat type 2 diabetes and obesity. SLC6A19 is the major transporter for BCAAs and methionine in the intestine, and mice lacking this transporter show a host of responses that are similar to those observed after gastric bypass surgery.13 The main responses were improved glycemic control, reduced plasma insulin, increased secretion of GLP-1 and GIP, increased production of FGF-21, reduced plasma fatty acids and liver triglycerides, browning of subcutaneous adipose tissue, and increased ketone body production. FGF-21 could be an important factor of this metabolic phenotype causing ketone body production from plasma and liver fatty acids, and browning of adipose tissue. Humans with mutations in SLC6A19 have Hartnup disorder, a largely benign syndrome that is sometimes associated with photosensitive skin rash.86,87 Very rarely has cerebellar ataxia been reported in young individuals with Hartnup disorder, which may be coincidental or caused by poor nutrition. Thus, it appears likely that inhibition of the transporter is generally safe in mature individuals with overnutrition.

Through the use of computational approaches and screening assays based on membrane potential–sensitive dyes, benztropine was identified as an inhibitor of SLC6A19.46 Through the use of SLC6A19 containing proteoliposomes, nimesulide was identified as another inhibitor of SLC6A19.88 Both compounds have other molecular targets, and the IC50 values are such that they can only serve as research tool compounds.

SLC7A5 is the major transporter for BCAAs in nonepithelial cells.89 It is an amino acid exchanger that can use intracellular glutamine and other amino acids as an exchange substrate to import essential amino acids.2,90 Transport experiments performed in X. laevis oocytes suggest that glutamine is a rather poor substrate of SLC7A5; however, studies in cell lines are consistent with a significant role of glutamine as an exchange substrate, due to its abundance in the cytosol.91 The transporter is expressed in β cells, and leucine uptake via SLC7A5 is important for activation of mTORC192 and also stimulates the release of insulin as an allosteric activator of glutamate dehydrogenase.93 Inhibition of SLC7A5 using BCH reduces mTORC1 activation and insulin release. However, BCH has also been reported to stimulate insulin release through the same mechanism as leucine.94

Amino Acid Transporters as Disease Modifiers and Drug Targets - Stefan Bröer, 2018

Amino acids play an important role in the modulation of insulin release from pancreatic β cells,94,95 and transporters are involved in this process.92 It is thought that amino acids stimulate insulin secretion through ATP production inside mitochondria. Glutamine and glutamate, in particular, are readily metabolized inside mitochondria to produce ATP. Leucine, in addition, acts as an allosteric inducer of glutamate dehydrogenase, which provides intermediates for the tricarboxylic acid (TCA) cycle derived from glutamine. As a result, a combination of leucine and glutamine potently increases insulin secretion.94 Glutamine can enter β cells via the glutamine transporter SLC38A3,96 while leucine is transported by SLC7A5 (see above). Expression‎ of glutamate transporters is too low to support significant metabolism.97

Glutamine uptake via SLC38A5 supports α-cell proliferation. Glucagon is produced in α cells of the pancreas acting, among other cells, on hepatocytes to induce gluconeogenesis. Glutamine and alanine are the main precursors in the early hours of fasting. In diabetes, elevated levels of fasting glucose are observed, suggesting an overstimulation of gluconeogenesis. However, attempts to reduce glucagon signaling cause hyperglucagonemia and α-cell hyperplasia.98 Mice with a liver-specific lack of the glucagon receptor show a 3- to 7-fold increased proliferation of α cells, suggesting a factor that is released by hepatocytes to induce α-cell proliferation.99 Systematic metabolomic analysis revealed that l-glutamine was the hepatocyte factor, acting through mTOR signaling and the FoxP transcription factor.21,22 Expression‎ of SLC38A5 was upregulated in α cells from glucagon receptor knockout mice, while it was barely detectable in wild-type mice. Blocking glucagon signaling in the liver, by contrast, reduced the expression‎ of amino acid transporters SLC38A3, SLC38A4, and SLC7A2, which mediate the import of amino acids for gluconeogenesis.22

As an overall strategy, it is conceivable that insulin secretion and hepatic glucose output can be modulated by blocking amino acid transporters. This could be combined with transport inhibitors that cause methionine restriction.

Role of Amino Acid Transporters in Neurological Disorders

GABA, glutamate, glycine, d-serine, and glutamine are neuroactive amino acids, the modulation of which can be exploited to treat several neurological pathologies.

Glycine transporters GlyT1 (SLC6A9) and GlyT2 (SLC6A5) are the major transporters for the neurotransmitter glycine in the brain. Pain is a normal response to injury or inflammation and is mediated by nociceptive neurons.8,14 Acute adaptive changes to these circuits also lead to hypersensitivity that serves a protective function during healing. However, nerve injury in the peripheral or central nervous system can lead to chronic pain or hypersensitivity that persists beyond the healing process. Glycine and GABA are the major inhibitory neurotransmitters in the nervous system and are released to dampen excitatory neurotransmission, which is used to transmit signals to the brain, including pain. Physiologically, neurotransmitters are rapidly removed from the synaptic cleft to terminate the signal. It is thought that chronic pain is caused by a reduction of glycinergic inhibition in the spinal cord, where glycine transporters are particularly abundant.8 A broad variety of glycine transport inhibitors have been generated, for both GlyT1 and GlyT2.14 Inhibitors based on sarcosine (N-methyl-glycine) are selective for GlyT1. However, GlyT1 is widely expressed in astrocytes of the nervous system, and its inhibition enhances excitatory neurotransmission by increasing extracellular glycine as a co-agonist at NMDA receptors.9 GlyT2 inhibitors ALX1393 and ORG25543 were shown in a mouse model of neuropathic pain to have antiallodynia effects,100 that is, preventing an innocuous stimulus from becoming painful.

Schizophrenia is a devastating neuropsychiatric disorder characterized by positive symptoms (i.e., symptoms that healthy people do not experience, such as delusions and hallucinations), negative symptoms (i.e., deficits compared with healthy individuals, such as lack of emotion and motivation), and cognitive deficits.9,101 The cause of this disease remains unclear, but reduced activity of NMDA receptors may be involved in the negative symptoms. NMDA receptors are allosterically modulated by d-serine and glycine, and their extracellular levels are controlled by specific transporters. In the case of glycine, GlyT1 is thought to control glycine levels in the synaptic cleft.9 The main transporter for d-serine is the neutral amino acid transporter asc-1;102 smaller contributions are mediated by amino acid transporters SLC1A4 and SLC1A5.103 In both cases, blocking the uptake of glycine or d-serine would increase the concentration of these allosteric modulators in the synaptic cleft, thereby enhancing NMDA receptor activity. Surprisingly, inhibitors of asc-1 decreased extracellular d-serine levels, suggesting that this transporter is rather involved in the efflux of d-serine from astrocytes.104,105 While modulation of d-serine levels in schizophrenia requires more research, GlyT1 inhibitor bitopertin has been used in clinical trials (up to clinical trial phase 3).9 Unfortunately, the results were inconclusive and no separation between placebo and bitopertin-treated patient groups was observed.106 Bitopertin was well tolerated, overcoming concerns that complete inhibition of GlyT1 could result in severe motor and respiratory deficits, as observed in GlyT1 knockout mice.107

Despite years of intensive research on glutamate transporters in the brain, a promising pharmacological approach has yet to emerge. Glutamate transporters are essential for the removal of the excitatory neurotransmitter glutamate from the synaptic cleft. Modulation of glutamatergic neurotransmission is involved in many neurological and psychiatric diseases.7 It is thought that elevated levels of glutamate are neurotoxic due to overactivation of glutamate receptors, which in turn causes excessive calcium influx into neurons, resulting in cell damage and death. As a result, pharmacological approaches using glutamate transporters rely on compounds that enhance glutamate transport rather than inhibiting it. In a broad screen, Rothstein et al. identified β-lactam antibiotics as compounds that increase transcription of the gene encoding glutamate transporter EAAT2 (SLC1A2).108 Activators of EAAT2 translation were identified by using an enzyme-linked immunosorbent assay (ELISA) in combination with antibodies that bind to the C-terminus of EAAT2.109 In the used cell line, EAAT2 expression‎ was driven by a cytomegalovirus (CMV) promoter to increase the likelihood of identifying translational activators. In amyotrophic lateral sclerosis (ALS), motor neurons gradually degenerate, resulting in muscle atrophy, weakness, and eventually respiratory failure and death. The etiology of neurodegeneration in ALS is not fully understood, but glutamate excitotoxicity may contribute to the disease progression.110 A clinical trial was conducted to eval‎uate whether enhanced expression‎ of EAAT2 by ceftriaxone would delay the functional decline in patients with ALS.110 It appears that ceftriaxone showed some beneficial effects in earlier stages of the disease, but no differences were observed in overall survival. The translational activator 3-([(2-methylphenyl)methyl]thio)-6-(2-pyridinyl)-pyridazine was able to protect neurons from excitotoxicity and delayed functional decline in an animal model of ALS.111

Role of Amino Acid Transporters in Cancer

Alteration of metabolism is almost universally observed in cancer cell lines.112 These adaptations are governed by the needs of an anabolic cell to generate sufficient biomass to produce new cells. Protein forms the majority of new cell mass, and cancer cells therefore show enhanced rates of amino acid uptake. In addition, amino acids are used for a wide variety of metabolic purposes in cancer cells. One of the earliest approaches to combat cancer relied on antifolates, which are essential to provide one-carbon compounds to metabolic pathways.113 One-carbon compounds are particularly important for the synthesis of nucleobases. 5-Fluorouracil is an inhibitor of thymidylate synthase, which converts nucleotide dUMP to dTMP for DNA biosynthesis. In the process, N5,N10-methylene tetrahydrofolate is converted to dihydrofolate. Dihydrofolate is recycled in two steps involving reduction and recharging of a C1 compound, which is derived from serine-to-glycine conversion by serine hydroxymethyl transferase. Thus, serine is an important metabolite for cancer cells, which can be generated from the glycolytic intermediate 3-phosphoglycerate. Serine depletion causes growth arrest in several cancer cell lines through its role in C1 metabolism, but in addition, serine depletion increases respiration and production of reactive oxygen species, by reducing the allosteric activation of pyruvate kinase PKM2.114 Equally important in cancer cells is the amino acid glutamine, which contributes to the biosynthesis of biomolecules in several ways. First, it provides nitrogen from its amino group for nucleobase biosynthesis, Second, it is used to synthesize aspartate (also required for nucleobase biosynthesis) and pyruvate in a pathway called glutaminolysis, a linearized version of the TCA cycle used in rapidly proliferating cells.115,116 Through this pathway, glutamine can sustain anaplerosis of the TCA cycle, which provides citrate, for example, for fatty acid biosynthesis. Third, glutamine is essential for the hexosamine pathway that produces aminated carbohydrates for the purposes of glycosylation, the amino group being derived from glutamine.117 Fourth, glutamine is an important precursor of glutathione biosynthesis.118 Differentiated cells often produce glutamine to dispose of ammonia from amino acid metabolism, while cancer cells become glutamine dependent. Many cancer cells cease to grow when glutamine is removed from the media.119 The glutamine dependence is a consequence of the multiple uses of glutamine in the pathways mentioned above. Consequently, glutamine transport has been considered a target for cancer therapy, but the approach is not straightforward due to the multitude of glutamine transporters in mammalian cells.6,10,120,121 Mammalian cells, in general, can only synthesize the nonessential amino acids, while essential amino acids must be acquired from nutrition or blood supply in the case of cancer cells.

SLC1A5 is an amino acid exchanger for small neutral amino acids, including serine, threonine, and glutamine.122 It is upregulated in many cancer cells and functionally the dominant glutamine transporter in most cancer cell lines.122 Silencing of SLC1A5 reduces cell growth in some cancer cell lines123–126 but not in others.121,126 Related to this, SLC1A5 silencing reduces mTORC1 activity in some cell lines90,123 but not in others.121 Pharmacological studies have been hampered by the poor selectivity of initially published inhibitors and by the fact that most of them are competitive binders and are easily displaced by amino acids in media.60,121,127,128 Potentially more powerful and selective inhibitors have been published more recently (Fig. 3).129 The mechanism by which blockage of SLC1A5 reduces tumor growth is still unclear. As an exchanger, SLC1A5 cannot mediate net uptake of amino acids, but it can quickly restore depleted amino acids at the expense of abundant amino acids.121 It appears likely that in cell lines where blocking SLC1A5 reduces mTORC1 activity, reduced growth is observed, while in cells where mTORC1 signaling is unaffected, growth persists. Silencing of SLC1A5 induces amino acid restriction and activation of the GCN2/ATF4 pathway, which in turn upregulates other amino acid transporters.121,126 The variability in the response of cancer cell lines to blockage of SLC1A5 is also illustrated by the antitumor efficacy eval‎uation of a monoclonal antibody against SLC1A5.130 Due to the lack of potent inhibitors, clinical trials or pharmacological experiments in animals have not been reported. SLC1A5 knockout mice are viable and fertile,131 suggesting that blocking of ASCT2 in tumor cells is unlikely to affect differentiated cells. Immune cell populations were found to be normal even when proliferating intensively.131

SLC7A5 is the major transporter for many essential amino acids in cancer cells.11,122,132 Like SLC1A5, it is an amino acid exchanger, which can mediate the uptake of essential amino acids (BCAAs, aromatic amino acids) at the expense of nonessential amino acids. Moreover, it is overexpressed in many cancer cell lines,11,12,122,133 and high expression‎ is often correlated with poor clinical prognosis.134 Genetic deletion of LAT1 in human colon adenocarcinoma cells significantly reduced cell proliferation, the formation of tumor spheroids, and the volume of tumor xenografts.135 Similar to the deletion of SLC1A5, deletion of SLC7A5 induced an amino acid stress response, as evidenced by phosphorylation of eIF2α and GCN2 and transcription of ATF4. This has also been observed in other cancer cell lines.136 It is well established that SLC7A5 forms a covalent heteromeric complex with CD98, which is required for trafficking to the plasma membrane,137,138 but is not involved in the transport process itself.139 However, genetic deletion of CD98 in human colon adenocarcinoma cells had a much milder phenotype than deletion of SLC7A5, although transport activity was reduced by 90%.135 Blocking the residual transport activity with the SLC7A5 inhibitor JPH203 completely abolished growth of the CD98 knockout cells. Reduced growth was accompanied by reduced mTORC1 activity. Inhibitors for SLC7A5 have been developed and reduce tumor growth in vivo12,140 or in vitro.141–143 No human clinical trials have been reported. A potential concern is the role of SLC7A5 as the major transporter for essential amino acids in the blood–brain barrier.144 Global knockout of SLC7A5 and CD98 in mice has been reported as being embryonically lethal.145,146

SLC7A11 encodes a glutamate/cystine exchanger, which plays an important role in providing cysteine for the synthesis of glutathione in a variety of cell types.147 Triple-negative breast cancer cell lines that were highly glutamine dependent also showed elevated expression‎ of SLC7A11, cystine uptake, and glutamate secretion. SLC7A11 expression‎ is responsible for the methionine dependency of cancer cell growth observed in some breast cancer cell lines.148 In these cells, cystine uptake reduces the use of methionine as a precursor for cysteine. SLC7A11 is highly regulated in cancer cells, being downregulated by oncogenic PI3K148 and also by overexpression‎ of mutated p53.149 Downregulation of SLC7A11 renders cells more susceptible to cell death caused by reactive oxygen species, such as ferroptosis.149 The anti-inflammatory pro-drug sulfasalazine, an inhibitor of SLC7A11, reduced tumor size in xenografts.118 While sulfasalazine is not used as a drug to treat cancer due to its chemical instability, it provides proof of concept that targeting xCT could be a therapeutic strategy in certain subgroups of cancer, particularly in combination with anticancer drugs that increase oxidative stress.150

SLC6A14 is a Na+- and Cl–-dependent amino acid transporter, which accepts all amino acids, except aspartate and glutamate. The transporter is expressed in selected tissues, such as lung, trachea, salivary gland, pituitary, distal ileum, and colon. Due to its tissue-specific expression‎, it is not generally upregulated in cancer but is found in colon cancer,151 cervical cancer,152 estrogen receptor–positive breast cancer,153 and pancreatic cancer.120 The tumor-promoting role of SLC6A14 has been documented in estrogen receptor–positive breast cancer using α-methyl-tryptophan, a small-molecule blocker of the transporter.154 SLC6A14 knockout mice, when crossed with the polyoma middle T oncoprotein mouse breast cancer model, showed delays in the development of mammary tumors.155 More advanced inhibitors are required to eval‎uate this target in cancer treatment.

SLC43A1 is a Na+-independent transporter for large neutral amino acids.156 In contrast to SLC7A5, which operates as an amino acid exchanger, SLC43A1 mediates facilitative diffusion of its substrates.157 The gene is overexpressed in prostate cancer, together with SLC7A5. Silencing of SLC43A1 reduced mTORC1 signaling and cell growth in LNCaP cells, but less so in PC-3 cells, which appear to rely more on SLC7A5 activity.136 In prostate cancer cells, SLC7A5 and SLC43A1 were regulated in a coordinated fashion. High expression‎ of SLC43A1 is accompanied by low expression‎ of SLC7A5 and vice versa. The expression‎ of SLC43A1 is regulated by androgen receptor binding, while SLC7A5 is regulated by an amino acid response element that increases transcription after amino acid depletion.136

SLC38A1 and SLC38A2 are Na+-dependent transporters for small and medium-sized neutral amino acids. When sufficient amino acids are present, SLC38A2 is not found at the plasma membrane, but its expression‎ increases manifold after amino acid depletion.121,158 The expression‎ is tightly regulated at the transcriptional level,159 mRNA level,160 and protein level.161,162 Importantly, SLC38A1 and SLC38A2 are involved in providing glutamine for glutaminolysis.121 The pharmacology of SLC38A1 and A2 is ill-developed. The proline analogue N-methylamino-isobutyric acid is frequently used as an SLC38A1/2 inhibitor and reduces growth of osteosarcoma cells.121 Silencing of SLC38A1 also reduced the growth of colon cancer cell lines and pancreas cancer cell lines.163,164

These results suggest that amino acid transporters have a significant potential in cancer therapy, but the field lacks, with a few exceptions, highly specific inhibitors that could be used in animal models and clinical trials.

Role of Amino Acid Transporters in Inflammation and Immune Disorders

Arginine is the precursor for the generation of nitric oxide. The inducible cationic amino acid transporter 2 (SLC7A2, splice form CAT2B) provides arginine for nitric oxide synthase, particularly in macrophages.165 Nitric oxide is thought to contribute to the pathology of inflammatory bowel disease and Crohn’s disease. Provision of arginine has shown beneficial effects in animal models of colonic inflammation.166 The beneficial effects were lost in mice lacking SLC7A2.167 Consistently, SLC7A2 knockout mice develop chronic inflammation in the lungs, potentially because of reduced NO production in macrophages.168

SLC7A5 is the major transporter for BCAAs in activated lymphocytes.145 A global knockout of SLC7A5 is embryonically lethal, but a CD4-driven cre-lox SLC7A5 knockout results in normal animal and normal immune cell number in unchallenged animals.145 However, immunological activation of T cells was largely abolished in SLC7A5 knockout cells. Deletion of SLC7A5 not only prevented activation of mTORC1, but also interfered with metabolic reprogramming, such as upregulation of glucose and glutamine transport.145

SLC1A5 is an amino acid exchanger that is used to harmonize amino acid levels within rapidly proliferating cells. As a result, it is highly expressed in cancer cells, but also in rapidly proliferating immune cell populations. Nakaya et al.169 reported that SLC1A5 knockout animals had attenuated inflammatory T-cell responses. Particularly, interferon-γ-producing T-helper cells were significantly reduced. Masle-Farquhar et al., by contrast, reported normal B- and T-cell development and normal B-cell function, suggesting a limited effect of this transporter on immune function.131

SLC16A10 is the major transporter for aromatic amino acids in many cell types.170 The transporter mediates facilitated diffusion of aromatic amino acids and may therefore serve as a route for uptake (e.g., in the liver, where metabolism of aromatic amino acids occurs) or efflux (e.g., in epithelial cells, where aromatic amino acids are released across the basolateral membrane). NOD (nonobese diabetic) mice are a widely used model of type 1 diabetes, an autoimmune disorder, which results in the destruction of β cells by lymphocytes infiltrating the islets of the pancreas.171 NOD mice spontaneously develop type 1 diabetes at an elevated rate, with an onset at an age of more than 100 days. Transposon-mediated inactivation of SLC16A10 increased the susceptibility to T1D in NOD mice.172 The transporter is highly expressed in macrophages, which play an important role in antigen presentation and the release of pro-inflammatory cytokines. The mechanism by which inactivation of SLC16A10 increases the incidence of T1D remains to be elucidated.

Role of Amino Acid Transporters in Embryo Development and Stem Cells

In vitro fertilization is an important process in research (mouse embryos), animal husbandry, and human reproduction. Significant efforts have been made to optimizie media for embryo development in vitro. While embryos develop at physiological osmolarity in vivo, in vitro development beyond the two-cell stage requires reduced osmolarity. The two-cell block can, however, be overcome in normal osmolarity by the addition of osmolytes, such as glycine, betaine, proline, β-alanine, and hypotaurine,19 which are accumulated by amino acid transporters. Amino acid transporters are highly regulated during early embryo development.16,173 For instance, GLYT1 is essential for the regulation of oocyte size and is activated as the oocyte undergoes meiosis,19 transporting the osmolyte glycine. The osmolyte betaine is transported by SLC6A20, which is upregulated in one- and two-cell embryos.20 Cationic amino acids are recruited through SLC7A9, while neutral amino acids are taken up by SLC7A5.174 The sodium neutral amino acid transporter SLC38A2 is expressed in pluripotent stem cells, and its activity is required for cell differentiation by accumulating l-proline.175 Competitive inhibition of SLC38A2 by high concentrations of glycine and alanine in the oviductal fluid may prevent development before implantation in the uterus.16 Future research in this area is projected to focus on modulation of epigenetic signatures,176 which could be achieved by media supplements, but also by activation or inhibition of amino acid transporters. Embryonic stem cells, for instance, acquire epigenetic changes associated with the formation of early primitive ectoderm-like cells, which are induced by l-proline.175,177

Amino acid supplementation and modulation of amino acid transport could be used to optimize the growth and differentiation of stem cells and to optimize conditions for in vitro fertilized oocytes.

Conclusion and Future Outlook

Amino acid transporters are often found at the beginning of metabolic pathways and influence a wide variety of biological processes, in particular, cell proliferation, but also cell signaling and nutrient homeostasis. Thus far, very few amino acid transport inhibitors have found their way into clinical trials. However, this is largely the result of the limited development of high-affinity and selective inhibitors. With improvements of tools to express membrane transporters and HTS being increasingly available in academic environments, this is likely to change in the near future. Solute carriers are considered an underdeveloped area of drug development.178 In addition to current approaches presented in this review, several strategies are emerging as new opportunities for targeting amino acid transport. One such strategy is augmentation (Fig. 5). Whole-genome sequencing is increasingly used to identify mutations, including those affecting amino acid transporters associated with disease phenotypes. These mutations are most likely inactivating transporters, and as a result, transport activators rather than inhibitors will be required. For instance, mutation in the putative monocarboxylate transporter Slc16a11 was identified as being associated with the development of type 2 diabetes in Mexican populations.179 This provides evidence that an activator of Slc16a11 could be used to treat or prevent type 2 diabetes.180 Transport activators can be identified at three levels, namely, transcriptional, translational, and allosteric. Transcriptional and translational activators have been successfully identified for amino acid transporters, as outlined above.108,109 Another strategy is combination (Fig. 6). Amino acid transport is a tempting target for tumor chemotherapy, but a better understanding of amino acid transporter redundancy and plasticity will be required. Tumor cells often escape inhibition of amino acid transporters through upregulation of alternative transporters via the GCN2/ATF4 pathway.121,136 This pathway reduces conventional Cap-dependent translation and, at the same time, activates transcripts with complex secondary structures.2 As outlined above, translational activators have been identified that appear to weaken complex secondary structures occurring in mRNA molecules.109 It appears feasible to identify compounds that stabilize secondary structures, thereby inhibiting translation. Combining amino acid transporter inhibition with inhibition of rescue mechanisms could open new chemotherapeutic approaches.

Figure 5. Future strategies to modulate metabolic diseases. Lack of transporter activity is associated with metabolic diseases. Enhancing the activity of the transporter protects the cell against metabolic stress.

Figure 6. Future strategies combining two stress factors to reduce cell growth. A significant problem using amino acid transporters as drug targets is redundancy. Blocking the adaptive stress response or inducing an independent stress can overcome metabolic adaptation.

Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research in the laboratory of the author is supported by grants of the National Health and Medical Research Council ID 1128442 and 1105857 and the Australian Research Council DP180101702.

ORCID iD
Stefan Bröer 

https://orcid.org/0000-0002-8040-1634

References

1.	Wu, G. Amino Acids: Biochemistry and Nutrition; CRC Press: Boca Raton, FL, 2013. Google Scholar | Crossref
2.	Broer, S., Broer, A. Amino Acid Homeostasis and Signalling in Mammalian Cells and Organisms. Biochem. J. 2017, 474, 1935–1963. Google Scholar | Crossref
3.	Kuma, A., Mizushima, N. Physiological Role of Autophagy as an Intracellular Recycling System: With an Emphasis on Nutrient Metabolism. Semin. Cell Dev. Biol. 2010, 21, 683–690. Google Scholar | Crossref
4.	Taylor, P. M. Role of Amino Acid Transporters in Amino Acid Sensing. Am. J. Clin. Nutr. 2014, 99, 223S–230S. Google Scholar | Crossref
5.	Broer, S., Palacin, M. The Role of Amino Acid Transporters in Inherited and Acquired Diseases. Biochem. J. 2011, 436, 193–211. Google Scholar | Crossref