Editor's evaluation: Conserved allosteric inhibition mechanism in SLC1 transporters

Abstract

Article Figures and data Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Excitatory amino acid transporter 1 (EAAT1) is a glutamate transporter belonging to the SLC1 family of solute carriers. It plays a key role in the regulation of the extracellular glutamate concentration in the mammalian brain. The structure of EAAT1 was determined in complex with UCPH-101, apotent, non-competitive inhibitor of EAAT1. Alanine serine cysteine transporter 2 (ASCT2) is a neutral amino acid transporter, which regulates pools of amino acids such as glutamine between intracellular and extracellular compartments . ASCT2 also belongs to the SLC1 family and shares 58% sequence similarity with EAAT1. However, allosteric modulation of ASCT2 via non-competitive inhibitors is unknown. Here, we explore the UCPH-101 inhibitory mechanisms of EAAT1 and ASCT2 by using rapid kinetic experiments. Our results show that UCPH-101 slows substrate translocation rather than substrate or Na+ binding, confirming a non-competitive inhibitory mechanism, but only partially inhibits wild-type ASCT2. Guided by computational modeling using ligand docking and molecular dynamics simulations, we selected two residues involved in UCPH-101/EAAT1 interaction, which were mutated in ASCT2 (F136Y, I237M, F136Y/I237M) in the corresponding positions. We show that in the F136Y/I237M double-mutant transporter, 100% of the inhibitory effect of UCPH-101 could be restored, and the apparent affinity was increased (Ki = 4.3 μM), much closer to the EAAT1 value of 0.6 μM. Finally, we identify a novel non-competitive ASCT2 inhibitor, through virtual screening and experimental testing against the allosteric site, further supporting its localization. Together, these data indicate that the mechanism of allosteric modulation is conserved between EAAT1 and ASCT2. Due to the difference in binding site residues between ASCT2 and EAAT1, these results raise the possibility that more potent, and potentially selective ASCT2 allosteric inhibitors can be designed . Editor's evaluation The goal of this study is to identify allosteric modulators of an SLC-1 amino acid transporter, ASCT2, which has been implicated in cancer progression. By combining computational and docking methods with functional measurements, this study provides convincing evidence for a conserved allosteric SLC-1 inhibition mechanism. The findings are important to the fields of transporter mechanism and SLC-1 pharmacology. https://doi.org/10.7554/eLife.83464.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Excitatory amino acid transporters (EAATs) from the solute carrier 1 (SLC1) family are an important class of membrane proteins, which are strongly expressed in the mammalian central nervous system (CNS). EAATs are responsible for the transport of glutamate across neuronal and astrocytic membranes in the CNS. Glutamate transporters are secondary active transporters, which utilize energy from the sodium concentration gradient between the extra- and intracellular sides of the membrane to take up glutamate into cells against a large concentration gradient (Wadiche et al., 1995; Zerangue and Kavanaugh, 1996a). In addition, H+ is co-transported and K+ is counter-transported (Zerangue and Kavanaugh, 1996a). This glutamate uptake is an important process, preventing glutamate concentrations from reaching neurotoxic levels in the extracellular space (Zerangue and Kavanaugh, 1996a; Tanaka et al., 1997). The SLC1 family contains five glutamate transporters (EAATs1–5), as well as two neutral amino acid transporters, the alanine serine cysteine transporters alanine serine cysteine transporter 1 and 2 (ASCT1 and -2) (Arriza et al., 1993; Utsunomiya-Tate et al., 1996; Zerangue and Kavanaugh, 1996b). ASCT2 has been implicated in rapidly growing cancer cells as a supply mechanism for glutamine, and inhibition of the transporter was shown to reduce cell proliferation and tumor size (Bröer et al., 1999; Son et al., 2013; Wahi and Holst, 2019; Kanai et al., 2013). Therefore, ASCT2 is emerging as a promising drug target. Structurally, SLC1 family members are assembled from three identical subunits (Canul-Tec et al., 2017; Garaeva et al., 2019; Yu et al., 2019), each contain eight transmembrane domains (TM1–8). TMs 1, 2, 4, and 5 form the trimerization domain, and TMs 3, 6, 7, and 8, together with hairpin loops HP1 and HP2, form the transport domain, which is based on an inverted repeat structure (Crisman et al., 2009; Reyes et al., 2009). Glutamate transporter inhibitors were widely studied for pharmacological purposes. Several types of competitive inhibitors were developed based on an aspartate scaffold, for example, TBOA (DL-threo-beta-benzyloxyaspartate) (Shimamoto et al., 1998) and TFB-TBOA ((3S)-3-[[3-[[4-(trifluoromethyl)benzoyl]amino]phenyl]methoxy]-L-aspartic acid) (Shimamoto et al., 2004). In addition, non-competitive inhibitors were identified such as UCPH-101 and -102 (Erichsen et al., 2010; Abrahamsen et al., 2013). Characterization with electrophysiological methods suggested UCPH-101 was specific for EAAT1 rather than EAATs 2–5. The inhibition of EAAT1 by UCPH-101 was not affected by glutamate or competitive inhibitors, such as TBOA (Shimamoto et al., 1998). Subsequently, the binding site of UCPH-101 was identified, first by site-directed mutagenesis, and then in a UCPH-101-bound crystal structure (Canul-Tec et al., 2017). UCPH-101 was found to bind to a hydrophobic region between the trimerization and transport domains (Canul-Tec et al., 2017). In addition to UCPH-101, other allosteric modulators, GT949 and derivatives, were first introduced by Falucci et al. (Falcucci et al., 2019; Kortagere et al., 2018). GT949 is a positive allosteric modulator in EAAT2. However, it operates as an inhibitor in other EAAT subtypes. After incubation of GT949, glutamate uptake activity and vmax increase in EAAT2, without altering the substrate affinity. The report also suggested that the allosteric binding site is located between the transport and the trimerization domains (Falcucci et al., 2019; Kortagere et al., 2018). For ASCTs, inhibitor pharmacology is less well studied. Currently, a number of competitive inhibitors have been identified, on the basis of structural similarity with EAAT inhibitors, as well as docking to homology models (Singh et al., 2017; Ndaru et al., 2019). While these newly discovered substrate binding site inhibitors show improved potencies and selectivities (Garibsingh et al., 2021), because of their high similarity to amino acid substrates they (i) likely have poor bioavailability and (ii) need to compete with the amino acid-rich media; thus, they are unlikely to be useful as future anti-cancer drugs via inhibition of ASCT2-mediated transport. It is therefore critical to identify specific, allosteric, non-amino acid-like inhibitors for ASCT2. EAAT1 and ASCT2 share 57.6% sequence similarity and 38.5% sequence identity (Figure 1—figure supplement 1); however, it is not known if the allosteric binding site of EAAT1 is also found in ASCTs, and whether small molecules can modulate ASCT2 non-competitively via this potential site. Structural analysis of the allosteric sites of EAAT1 and ASCT2 suggests that two non-conserved residues in EAAT1 (e.g., F136, and I237) significantly contribute to the UCPH-101 interaction, while other binding site residues that are less critical for binding are conserved between EAAT1 and ASCT2 (Figure 1—figure supplement 1). In this report, we first compare and detail the effects of UCPH-101 on the SLC1 family members EAAT1 and ASCT2, including analysis of rapid chemical kinetic experiments. While UCPH-101 was found to be a partial and low affinity inhibitor of wild-type ASCT2, engineering of a double-mutant ASCT2 transporter based on computational docking analysis on the basis of the EAAT1 binding site resulted in a largely recovered UCPH-101 effect. In addition, we describe the identification of a non-competitive inhibitor that is not related to UCPH-101 from a virtual screen. Finally, we discuss the importance of our results to the understanding of allosteric inhibition in SLC1 transporters as well as their relevance to the identification of future, non-competitive modulators for ASCTs, including anti-cancer drugs. Results UCPH-101 has been previously reported to be a potent non-competitive glutamate transporter inhibitor with high specificity for EAAT1 rather than the other SLC1 glutamate transporter subtypes. The EAAT1 structure shows UCPH-101 (Figure 1A) deeply buried between the transport domain and the trimerization domain interface (Figure 1B and C). The hydrophobic pocket is located between TM3, TM7, and TM4 (Figure 1C). The residues that interact with UCPH-101 include G120 (TM3), V373 (TM7a), M231 (TM4c), Y127 (TM3), F369 (TM7), and F235 (TM4c) (Figure 1D). This binding pocket is at a distance of over 15 Å from the substrate binding site (Figure 1B), suggesting that UCPH-101 may not preclude substrate binding (Canul-Tec et al., 2017). Figure 1 with 1 supplement see all Download asset Open asset Structure of the excitatory amino acid transporter 1 (EAAT1) UCPH-101-bound state. (A) UCPH-101 structure and substituents used for constraints in docking calculations. (B) EAAT1 structure (PDB id: 5MJU) in complex with the competitive inhibitor TFB-TBOA ((3S)-3-[[3-[[4-(trifluoromethyl)benzoyl]amino]phenyl]methoxy]-L-aspartic acid) (cyan sticks) in the substrate binding site and the allosteric inhibitor UCPH-101 (purple sticks) in the allosteric site. The trimerization and transport domain are highlighted in light green and orange, respectively. (C) Illustration of the UCPH-101 binding site at the domain interface. The electrostatic surface calculated using the Adaptive Poisson-Boltzmann Solver (APBS) plugin in Pymol (default parameters) in the absence of UCPH-101 is highlighted. (D) EAAT1 transmembrane (TM) helices and amino acid residues in proximity to the UCPH-101 binding site. A comparison of the UCPH-101 binding pocket between EAAT1 and ASCT2 (Figures 1C and 2B and C) reveals similarities and differences in the overall shape and physical chemical properties of the allosteric binding site of EAAT1 and the equivalent region in ASCT2 (Figures 1D and 2C). Several residues are conserved between these proteins maintaining the overall shape of the site. For example, L104, V227, F369 in EAAT1 correspond to L112, V235, and F377 in ASCT2. Conversely, some residues are not fully conserved. Most notably, Y127 and M271 correspond to F136 and I237 in ASCT2, affecting the shape and polarity of the binding site (Figure 1—figure supplement 1 and Figure 2B and C). We hypothesized that these differences as well as other changes will lead to differences in inhibitor specificity of the corresponding allosteric sites. Figure 2 Download asset Open asset Alanine serine cysteine transporter 2 (ASCT2) structures and predicted allosteric binding site. (A) ASCT2 structure (PDB id 7BCS) with the competitive inhibitor L-cis-PBE (cyan sticks). The trimerization and transport domains are highlighted in dark green and dark orange, respectively. (B) Illustration of the putative allosteric binding site (UCPH-101 binding site in excitatory amino acid transporter 1 [EAAT1]) at the domain interface. The electrostatic surface of the binding site is highlighted. (C) Close-up view of the putative ASCT2 allosteric binding site; UCPH-101 coordinates (purple sticks) were derived from the EAAT1 structure in the corresponding site. Indeed, molecular docking of UCPH-101 to the putative allosteric site of ASCT2 fails using multiple approaches including both rigid and flexible docking, as well as constraints on different interactions and residues (Materials and methods). However, when F136 and I273 are remodeled to tyrosine and methionine respectively, UCPH-101 can be successfully docked into ASCT2 with a pose similar to that seen in the EAAT1 structures (see below). This result added further support to our hypothesis that Y127 and M271 are key residues impacting binding of UCPH-101 in EAAT1 and explains why inhibition is only partial in wild-type ASCT2 (see below). Next, our goal was to experimentally test predictions from docking with respect to interaction of UCPH-101 with the SLC1 family member ASCT2, contrasting them with results from EAAT1. To validate our experimental system and to compare inhibitor effects on pre-steady-state currents, which had not been investigated in the past, we first briefly summarize UCPH-101 effects on EAAT1. UCPH-101 is a slow-binding, non-competitive inhibitor of EAAT1, which blocks translocation rather than other partial reactions of the transport cycle As reported previously (Erichsen et al., 2010; Abrahamsen et al., 2013; Jensen et al., 2009), glutamate-induced EAAT1 anion current was inhibited in a concentration-dependent manner by UCPH-101 (Figure 3A and B). At low [UCPH-101] (2 μM) (Figure 3A, red trace) onset of inhibition was slow with complete inhibition reached only after several seconds. In agreement with the expectation of a non-competitive inhibition mechanism, the apparent Ki was essentially independent of the glutamate concentration (Figure 3B and C). In contrast, if UCPH-101 would show a competitive inhibition pattern, an increase of apparent Ki with the glutamate concentration would be expected, according to Equation 2. (1) Ki(S)=Ki(0)+[S]Ki(0)/Km(substrate) Figure 3 with 1 supplement see all Download asset Open asset UCPH-101 is a high-affinity, non-competitive inhibitor of excitatory amino acid transporter 1 (EAAT1) anion current. (A) Typical whole-cell current recordings in the absence (black) and presence of 2 μM (red) and 100 μM (blue) UCPH-101. [Glutamate] was 100 μM. Experiments were performed using 140 mM sodium methanesulfonate (NaMes) in the extracellular buffer, and 130 mM KSCN intracellularly (forward transport conditions). (B) Dose-response curves to determine the apparent Ki for UCPH-101 at increasing glutamate concentration of 100 μM (black, n=16), 200 μM (red, n=19), and 300 μM (blue, n=13). (C) Glutamate concentration dependence of the apparent Ki for UCPH-101 suggests non-competitive inhibition mechanism. All experiments were performed at 0 mV membrane potential. Figure 3—source data 1 The source data contains original current traces for Figure 3A, and the original data for panels (B) and (C). https://cdn.elifesciences.org/articles/83464/elife-83464-fig3-data1-v3.xlsx Download elife-83464-fig3-data1-v3.xlsx In Equation 1, Ki(0) and Ki(S) are the Ki values (inhibition constants) in absence and presence of glutamate. Km(substrate) is the apparent Michaelis-Menten constant for glutamate, and [S] is the substrate concentration. Furthermore, we determined the voltage dependence of the UCPH-101 inhibitory effect. As expected, anion currents were fully inhibited at all membrane potentials (Figure 3—figure supplement 1). To obtain a high time resolution for determining the effect of UCPH-101 on transport kinetics, we applied glutamate rapidly to EAAT1 using a piezo-based solution exchange device. This system has a 5–10 ms time resolution when applied to whole cells, and also allows the rapid removal of substrate, providing information on the recovery of current after glutamate removal. However, due to the millisecond time resolution, it will limit determination of very early intermediates in the transport cycle, which are present in a sub-millisecond time domain, but the time resolution is high enough for determining glutamate transporter turnover rate, and how it is affected by the presence of inhibitor. EAAT1 anion currents (Figure 4A) and transport currents (Figure 4E) show rapid rise of the current after glutamate application within the time resolution afforded by solution exchange (5–10 ms). The rise of the current is followed by rapid decay of a transient current component to the steady state, as demonstrated previously (Grewer et al., 2000). This transient phase of the current was previously assigned to steps associated with the glutamate translocation reaction (Wang et al., 2018). After pre-incubation and in the continuous presence of UCPH-101, both transient and steady-state current amplitude were reduced in a dose-dependent manner, as expected (Figure 4B and F, currents were fully eliminated at 40 μM UCPH-101), but at concentrations close to the apparent Ki, the decay of the current was also slowed (Figure 4D, τ=11 ± 1.3, 16±2.1, 20±2.1 ms, at 0, 0.2, and 2 μM UCPH-101, respectively). The reduction of the decay rate of the transient current indicates a slowing of the translocation reaction in the presence of UCPH-101. In contrast to competitive inhibitors, UCPH-101 application in the absence of glutamate did not induce any currents, as expected for non-competitive inhibitors, which are not known to be able to block the glutamate transporter leak anion conductance (Figure 4C and G). Figure 4 with 1 supplement see all Download asset Open asset UCPH-101 has only minor effect on recovery kinetics of excitatory amino acid transporter 1 (EAAT1) current after glutamate removal. (A) Anion currents in response to two pulses of rapid glutamate application (1 mM), with varying inter-pulse interval (pulse protocol shown at the top) under forward transport conditions. The intracellular solution contained 130 mM KSCN, the extracellular solution contained 140 mM sodium methanesulfonate (NaMes). (B) Similar experiment as in (A), but in the presence of 0.2 μM UCPH-101 (pre-incubated for 5 min, see open bar for timing of solution exchange, top of the figure). (C) Application of UCPH-101 alone did not induce any currents. (D) Time constants for the fast and slow phase of the current recovery, for the two exponential components. (E–G) Experiments similar to (A–C), but for the transport component of the current (the permeant intracellular anion, SCN-, was replaced with the non-permeant Mes- anion). (H) Recovery of the transient current in the presence and absence of 0.2 μM UCPH-101. The membrane potential was 0 mV in all experiments. Figure 4—source data 1 Current traces for Figure 3A,B,C,E,F and G, and the original data for panels (D) and (H). https://cdn.elifesciences.org/articles/83464/elife-83464-fig4-data1-v3.xlsx Download elife-83464-fig4-data1-v3.xlsx Upon removal of glutamate, recovery of the transient current upon glutamate application in a second pulse was relatively slow, with a time constant of 95±4 ms, most likely reflecting the turnover time of the transporter, which has to recycle its binding sites to outward-facing, after passing through a whole transport cycle. At 0.2 μM UCPH-101 recovery of the transient current after glutamate removal occurred with a time constant of 115±11 ms, as quantified in Figure 4F and H. Therefore, it appears that UCPH-101 does not significantly slow the recovery rate of the transient current. The most likely explanation is that, at low concentrations (<1 μM) UCPH-101 binding to EAAT1 and unbinding are very slow, with time constant of 10 s for binding, and 100 s for dissociation (Abrahamsen et al., 2013). Thus, within the time course of the paired-pulse experiment (1.4 s), UCPH-101-bound and -unbound transporter populations do no interconvert, and the time course of current recovery, thus, reflects that of the UCPH-101-free state. Finally, we tested whether UCPH-101 binding affects how Na+ interacts with the apo (glutamate-free) form of the transporter. This question was raised from the crystal structures of EAAT1 in the presence of UCPH-101 (Canul-Tec et al., 2017). The structure shows Na+ bound only at the Na2 position (Canul-Tec et al., 2017). However, two more Na+ binding sites exist, Na1 and Na3 (Canul-Tec et al., 2017; Reyes et al., 2009; Boudker et al., 2007; Yernool et al., 2004; Verdon and Boudker, 2012; Guskov et al., 2016; Wang et al., 2019). As reported previously, Na+ binding to a glutamate-free form of EAAT2 and -3 is electrogenic (Grewer et al., 2000; Watzke et al., 2001; Grewer et al., 2012), generating transient currents due to Na+ movement into its binding site upon voltage jumps to more negative membrane potentials. Consistently, step changes of the membrane potential from –100 to +60 mV resulted in transient currents (Figure 4—figure supplement 1), which decayed with a time constant in the 0.5 ms range (140 mM Na+), in the absence of glutamate. Non-specific currents were subtracted by applying 200 μM TBOA. In the presence of UCPH-101, the charge movement was still present (Figure 4—figure supplement 1B), and the midpoint potential (a measure of Na+ affinity) was not significantly changed (Figure 4—figure supplement 1C). These results are consistent with the interpretation that Na+ binding (most likely to the Na1 and Na3 sites) is not inhibited by bound UCPH-101 and occurs with similar affinity as in the absence of the inhibitor. UCPH-101 is a partial inhibitor of ASCT2 In the next paragraphs, we describe electrophysiological measurements on ASCT2 in the presence of UCPH-101, to test whether the inhibition mechanism is conserved between the EAAT1 and ASCT2 members of the SLC1 family. To corroborate the residues directly contributing to the UCPH-101-bound state in EAAT1 suggested by docking, we used molecular dynamics (MD) simulations, to investigate the stability of UCPH-101 in its EAAT1 binding pocket (Materials and methods). From the EAAT1 structure bound to the substrate aspartate and UCPH-101 in the allosteric site (5LLM), we generated a model of EAAT1 inserted into a lipid bilayer in a water box. After equilibration, UCPH-101 remained stable in the original binding pocket in six 100 ns simulations (four representative ones shown in Figure 5D). We selected two conserved residues (Y127, M271; Figure 5A and Figure 1—figure supplement 1) to estimate distance changes to bound UCPH-101 together with RMSD calculations. Interestingly, the trajectories show that the position of UCPH-101 is stable relative to Y127 and M271 in 100 ns simulations, and these two residues stay in close contact with the UCPH-101 pyran ring oxygen and the amino group nitrogen (Figure 5D). It should be noted that UCPH-101 has two stereo-centers. For the simulations, we used the stereo-configuration analogous to the one from the EAAT1 UCPH-101-bound structure. Figure 5 with 1 supplement see all Download asset Open asset EAAT1, ASCT2WT, and ASCT2F136Y/I237M residues contribute to UCPH-101 stability in the binding site. (A) Original state of the EAAT1-UCPH-101 complex from structure 5LLM (Canul-Tec et al., 2017). Y127 and M271 (EAAT1 sequence number) contribute to the EAAT1-UCPH-101 interaction. The binding state before (green) and after (gray) 100 ns molecular dynamics (MD) simulation are shown in (B). (C) Original state of the docked ASCT2WT-UCPH-101 complex (7BCS) with modeled side chains. Trajectories for four independent simulations runs for EAAT1, ASCT2WT, and ASCT2F136Y/I237M (one UCPH-101 molecule at each subunit interface) were analyzed and are shown from (D) to (F) (distance calculation [red, black (UCPH-101), green (amino acid substrate)] and RMSD distance [blue]). For distance calculations, in EAAT1, we selected atoms Y127(CA) and M271(CA) for EAAT1 and (O) and (N) for UCPH-101 (Materials and methods) as reference atoms. In ASCT2WT, we selected atoms F136(CZ) and I237(CD) for ASCT2WT and (N1) and (O2) for UCPH-101. In ASCT2F136Y/I237M distance calculation, we selected atoms Y136(OH) and M237(SD) from the transporter, and (N1) and (O2) from UCPH-101 as reference. Figure 5—source data 1 Trajectory and RMSD data for panels (D-F). https://cdn.elifesciences.org/articles/83464/elife-83464-fig5-data1-v3.xlsx Download elife-83464-fig5-data1-v3.xlsx To investigate the stability of inhibitors in the predicted ASCT2 allosteric binding pocket, we also performed MD simulations on UCPH-101-bound ASCT2WT and ASCT2F136Y/I237M. Residues F136Y and I237M (Figure 5C, Figure 5—figure supplement 1) were selected to evaluate distance changes to UCPH-101 during equilibration. The distance of UCPH-101 is relatively stable with respect to Y136 and M237 in 300–500 ns simulation trajectories (Figure 5E, Figure 5—figure supplement 1), while a slight, initial increase in the RMSD indicates relaxation of the structure compared to the initial, docked starting state. Two full dissociation events into the lipid bilayer were observed in six simulations within 300 ns for ASCT wild-type (Figure 5E), consistent with the predicted slow binding/dissociation kinetics, while no dissociation was observed in the double-mutant transporter (Figure 5F). The distance between selected atoms of UCPH-101 and ASCT2 were maintained within 6 Å in the double-mutant transporter, whereas in the wild-type these distances were 8–11 Å (Figure 5E), indicating that these interactions are less favorable in the wild-type transporter. Finally, as a control, we measured the distance between bound amino acid substrate and a residue in the binding site throughout the simulation. This distance remained stable throughout the simulations, until amino acid substrate dissociation occurred (dissociation is expected due to the low substrate affinity), illustrating convergence of our simulations and increasing the confidence in our approach. To test the predictions from docking calculations and MD simulations, we next experimentally investigated the interaction of UCPH-101 with ASCT2WT, and transporters with mutations to the analogous positions in ASCT2, using electrophysiological characterization of anion current, which is activated during substrate exchange. Anion conductance is an indirect measure of ASCT2 transport activity, but was previously shown to correlate with transport activity (direct amino acid transport measurements are shown below) (Grewer and Grabsch, 2004). Typical anion currents, using the highly permeant anion SCN-, were inwardly directed, due to SCN- outflow through the ASCT2 anion conductance (Figure 6A). As expected, the anion currents increased with increasing serine concentration, with an apparent Km of 280±40 μM. When serine was co-applied together with UCPH-101, the current was inhibited, but not to 100%, even at saturating [UCPH-101] (Figure 6B). In addition, inhibition required much higher UCPH-101 concentrations than in EAAT1. The UCPH-101 dose-response relationship, at a constant serine concentration of 100 μM (Figure 6C) demonstrates maximum inhibition of about 45% at 200 μM UCPH-101, with an apparent Ki value of 77±20 μM (Figure 6C and D). Although full saturation could not be reached due to UCPH-101 not being fully soluble at a concentration of 500 μM, the fit to the dose-response curve indicates partial inhibition even at saturating concentration (Figure 6C, red line). Maximum inhibition was not dependent on the serine concentration, as expected (Figure 6D). Furthermore, the apparent Ki was only weakly dependent on [serine] (Figure 7D). From these results, we presume that UCPH-101 operates as a weak, non-competitive, partial inhibitor for the wild-type ASCT2 transporter. Figure 6 with 1 supplement see all Download asset Open asset Alanine serine cysteine transporter 2 (ASCT2) amino acid-induced anion current is partially inhibited by UCPH-101. (A) Typical serine-induced ASCT2 anion current increases with increasing serine concentrations. The extracellular solution contained 140 mM sodium methanesulfonate (NaMes), with 130 mM NaSCN, and 10 mM Ser incorporated into the whole-cell recording electrode. The application time for serine is indicated by the gray bar. The apparent affinity for Ser were calculated as Km = 280 ± 40 μM. (B) Same experiment as in (A) at 100 μM Ser, but in the presence of UCPH-101 at three concentrations. (C) UCPH-101 dose-response curve at 100 μM Ser (n=15). (D) Relative current at saturating [UCPH-101], shown at varying Ser concentrations. (E) Uptake of 14C serine in hASCT2-expressing HEK293 cells in the absence and presence of 200 μM UCPH-101. Competitive ASCT2 inhibitor L-cis-BPE was used at a saturating concentration to determine specific uptake by ASCT2. Figure 6—source data 1 Original current traces for Figure 6A and B, and the original data for panels (C-E). https://cdn.elifesciences.org/articles/83464/elife-83464-fig6-data1-v3.xlsx Download elife-83464-fig6-data1-v3.xlsx Figure 7 Download asset Open asset The F136Y/I237M ASCT2 double mutation restores complete inhibition of anion current by UCPH-101. (A) Predicted binding pose of UCPH-101 (cyan sticks) to the F136Y/I237M-double-mutant ASCT2 transporter (the trimerization and transport domains are highlighted in dark green and dark orange, respectively) in overlay with EAAT1 (white). (B) Typical whole-cell current recordings for ASCT2F136Y/I237M at 100 μM serine. (C) Dose-response curves for ASCT2WT (black), and the F136 (red) and F136Y/I237M (blue) mutant transporters. Blocking effects were measured at a serine concentration of 100 μM and at 0 mV transmembrane potential. (D) UCPH-101 apparent inhibition constant (Ki) plotted as a function of the serine concentration. (E) Comparison of steady-state current at 100 μM Ser and 200 μM UCPH-101, wild-type ASCT2 currents without UCPH-101 were set as reference to 1, error bars represent ± SD. Figure 7—source data 1 Current traces for Figure 7B, and the original data for panels (C-E). https://cdn.elifesciences.org/articles/83464/elife-83464-fig7-data1-v3.xlsx Download elife-83464-fig7-data1-v3.xlsx As a direct measure of ami