SLC1A5 (ASCT2) catalyzes an obligatory Na⁺-dependent antiport in which Na⁺ together with an extracellular neutral amino acid are exchanged with an intracellular neutral amino acid (Kanai and Hediger, 2004; Utsunomiya-Tate et al., 1996). Among the solute carrier family 1 (SLC1) sub-family, it is the only one that is competent to transport glutamine, and it serves as the primary transporter of glutamine in cancer cells (Fuchs and Bode, 2005). Inhibition of SLC1A5 glutamine transport by small molecules and shRNA-mediated SLC1A5 knockdown have been shown to decrease endometrial cancer cell growth (Marshall et al., 2017). Similarly, SLC1A5 inhibition and knockdown suppressed mTORC1 signaling and induced rapid cell death in breast cancer cells (van Geldermalsen et al., 2016). In addition, SLC1A5 ^-/- mice are deficient in T cell receptor-mediated activation of the metabolic kinase mTORC1 and as a result exhibit improved clinical scores in experimental autoimmune animal models of multiple sclerosis and colitis (Nakaya et al., 2014).

The structure and function of SLC1 transporters have been studied using bacterial aspartate transporter homologues Glt_PH from Pyrococcus horikoshii (Yernool et al., 2004; Boudker et al., 2007; Reyes et al., 2009; Georgieva et al., 2013; Akyuz et al., 2015; Ji et al., 2016), Glt_TK from Thermococcus kodakarensis (Jensen et al., 2013; Guskov et al., 2016), human glutamate transporter SLC1A3 (Canul-Tec et al., 2017) and glutamine transporter SLC1A5 (Garaeva et al., 2018; Garaeva et al., 2019). All these transporters form a trimer of independently functioning protomers that uses an elevator mechanism to carry amino acids across membranes (Akyuz et al., 2015; Erkens et al., 2013). Each protomer has a scaffold domain and a movable transport domain which slides and carries the solute. Glt_PH structures have been reported in both the outward- and the inward-facing states (Boudker et al., 2007; Reyes et al., 2009; Akyuz et al., 2015; Verdon et al., 2014; Verdon and Boudker, 2012). Crystal structures of unbound and substrate-bound Glt_TK have also been reported in the outward-facing conformation. The crystal structures of SLC1A3 were determined with competitive and allosteric inhibitors in the outward-facing conformation (Canul-Tec et al., 2017). Crystallographic studies of SLC1A3 required mutation of nearly 25% of its primary sequence suggesting the inherent technical challenges with this gene family (Canul-Tec et al., 2017). Recently, the cryo-EM structures of SLC1A5 in complex with glutamine or inhibitor were reported in the inward-facing state (Garaeva et al., 2018; Garaeva et al., 2019). To gain insights into structural features that permit substrate recognition in an outward-facing state, using an antibody fragment as a fiducial marker we have solved the cryo-EM structures of the unliganded transporter and its complex with glutamine at 3.5 and 3.8 Å resolution, respectively.

Our full-length SLC1A5 construct showed high expression and stability. (Figure 1—figure supplement 1). The presence of affinity tags used for purification did not affect the sodium-dependent glutamine uptake when HAP1 SLC1A5 knock-out cells were transiently transfected with full-length SLC1A5 but the observed transport activity was low, compared to the background. To assist structural determination, we generated its complex with a commercially available cKM4012 Fab fragment. It is reported that KM4012 antibody isolated through a cell-based screen inhibited glutamine-dependent cancer cell growth (Suzuki et al., 2017). The purpose of the Fab was to serve as a fiducial marker for particle alignment. Negative-stain electron microscopy revealed that one molecule of Fab was bound to each protomer of a SLC1A5 homotrimer (Figure 1—figure supplement 1). The three-fold symmetry was broken due to the dynamic nature of each Fab and protomer. Single-particle electron cryo-microscopy (cryo-EM) analysis without imposing any symmetry restraint resulted in a facile three-dimensional reconstruction at an overall resolution of 3.5 Å (Fourier shell correlation (FSC) = 0.143 criterion; Figure 1—figure supplement 2). The highly flexible Fab fragments were masked out to maximize the structural quality of SLC1A5 (Figure 1—figure supplements 2,3). The quality of the EM density map was sufficient to allow model building for residues 43–488 (Figure 1—figure supplement 4, Table 1). The limited resolution on the Fab region prevents a detailed and accurate analysis of the protein-fab interface.

The structure of SLC1A5 shows a homotrimer with an overall Glt_PH-like fold in an outward-facing conformation, Each SLC1A5 protomer contains two domains, a transport domain and a scaffold domain connected by the extracellular loop region 2 (ECL2a and ECL2b) (Figure 1a-c, Figure 1—figure supplement 5). The individual transport domain containing transmembrane helices TM3, TM6-TM8 and helical loops 1–2 (HP1-HP2) interacts exclusively with the scaffold domain from its own protomer. The scaffold domain including TM1-TM2 and TM4-TM5 is involved in inter-protomer interactions and forms a compact core. The ECL2 connects TM3 to TM5 via TM4 and bridges the two domains. ECL2a spans about 55 Å across a quarter of the transport domain, sequentially interacting with ECL3, TM8, TM7a, and HP2b on extracellular surface.

Figure 1 with 5 supplements see all

Download asset Open asset

Superposition with the inward-facing state demonstrates that the scaffold domain remains relatively rigid, while the transport domain moves toward the cytoplasm and the substrate binding site shifts by 19 Å toward the cytoplasmic face (Figure 2a-b) (Garaeva et al., 2018). The ECL3 possessing one α-helix with ~1.5 turns connects the TM6 to the TM5 of the scaffold domain in the outward-facing state (Figure 2a). In the inward-facing state (PDB: 6GCT), this 1.5 turn α-helix in ECL3 merges into TM6 which extends the length of TM6 by about 9.5 Å (Figure 2a,c) (Garaeva et al., 2018). As a result, the shortened ECL3 further pulls the N-terminus of TM6 close to the TM5 causing the movement of TM6 toward the cytoplasm with ~40° tilt compared to the outward-facing state (Figure 2a). Simultaneously, to compensate the movement of the transport domain and the conformational changes of TM6, the N-terminus of TM3 unwinds by approximately 1.5 turns and extends the TM2-3 loop in the cytoplasmic space (Figure 2b–c). The extension of TM2-3 loop increases the distance between TM2 and TM3, resulting in a ~ 20° tilt compared to the outward-facing state (Figure 2b). The structurally symmetrical TM3 and TM6 serve as a platform which holds the core region of the transport domain (Figure 2c). During the transition between outward- and inward-facing states, the changes in lengths of the TM3, TM2-3 loop, TM6 and ECL3 regions trigger the adjustment in the platform plane relative to the membrane plane, resulting in the movement of the transport domain coupled with a ~ 30° rotation (Figure 2a-c) (Garaeva et al., 2018). Notably, the well-defined ECL2a undergoes a large conformational rearrangement, swinging out by ~35° relative to the transport domain when compared with an inward-facing state of a cross-linked Glt_PH structure (Figure 2d) (Reyes et al., 2009). In the outward-facing state, the ECL2a encompasses one side of the transport domain and is positioned to expose both the ECL4 and ECL5. However, in the inward-facing state, the ECL2a forms a sharp turn and sits on top of the transport domain covering a narrow groove formed by the ECL4 and ECL5 to accommodate rigid body movement of the transport domain (Reyes et al., 2009; Verdon et al., 2014; Verdon and Boudker, 2012; Reyes et al., 2013).

Figure 2

Download asset Open asset

To understand the basis for substrate recognition by SLC1A5, we solved the cryo-EM structure of SLC1A5 in the presence of the substrate, L-Gln at 3.8 Å resolution (Figure 1—figure supplement 3). The density was sufficient to dock the L-Gln molecule (Figure 3a). The Gln binding pocket is formed by two oppositely oriented reentrant loops (HP1 and HP2), TM7 and TM8 of the transport domain (Figure 3a–b). The HP2 loop in the unliganded structure is in an open conformation providing a direct access to the binding site. Upon Gln binding, HP2 loop acts as a gatekeeper and comes close to the serine-rich HP1 loop to shield the substrate from the extracellular surface (Figure 3a). Furthermore, the substrate binding interaction is maintained between the two states of SLC1A5 (Figure 3c) (Garaeva et al., 2018). Both α-amino and α-carboxylate groups of the substrate appear to be anchored by the carbonyl backbone of Ser351, the amide backbone and the sidechain of Ser353 in the HP1 loop and can be further stabilized by sidechains of Asp464 and Asn471 in TM8 (Figure 3b). Importantly, the Nε2 of the substrate, L-Gln, is within hydrogen bond distance to the sidechain of Asp464 in TM8. A smaller substrate, L-Asn, could fit into the binding site to satisfy the interaction with Asp464, consistent with the comparable K_m values of human SLC1A5 for L-Asn and L-Gln determined in proteoliposomes (Scalise et al., 2014). The position of L-Gln in our SLC1A5 structure is similar to that of the L-Asp in the SLC1A3 crystal structure, suggesting a conserved-ligand binding pocket among SLC1 family members (Figure 3d) (Yernool et al., 2004; Canul-Tec et al., 2017).

Figure 3 with 4 supplements see all

Download asset Open asset

Cys467, a unique residue among the SLC1 family, is at a suitable distance to donate its proton to the carbonyl group of the Gln substrate (Figure 3b, Figure 1—figure supplement 5) and could thus contribute to substrate selectivity. Mutation of Cys467 to Ala in SLC1A5 showed significant increase in K_m for Gln (Scalise et al., 2018). The corresponding residue in all glutamate transporters of the SLC1 family is a conserved bulkier Arg residue (Figure 1—figure supplement 5). In SLC1A3, the Arg residue with its guanidinium group forms a hydrogen bond with the carboxylate group of the substrate Asp, and a favorable cation-π interaction with a Tyr residue (Figure 3d). As a result of the smaller amino acid at this location, SLC1A5 has an additional side cavity involving Phe393, Val436, Ala433, Val463 and Asp464, whereas no such cavity is present in SLC1A3 (Figure 3b,d). Besides the difference in the pocket size and shape, Arg in place of the Cys will lead to steric clashes with the Gln substrate in SLC1A5. The critical role of the Cys467 in SLC1A5 for substrate specificity was further supported by an equivalent mutation of threonine in SLC1A4 (Thr459) to a cysteine, resulting in alteration to L-Gln binding (Scopelliti et al., 2018) and inhibition by Cys modifying reagents in proteoliposome assays (Pingitore et al., 2013). Since the structure of SLC1A4 is unknown, we attempted to generate a model for SLC1A4 by replacing Cys with a threonine. Unfortunately, none of the three rotamers available for threonine can be satisfactorily accommodated because of steric clashes with backbone carbonyl atoms. This suggests that SLC1A4 must have a slightly different local conformation around the substrate binding site. Regardless of the local conformation that SLC1A4 adopts in this region, the side chain hydroxyl group of Thr459 (analogous to Cys467 in SLC1A5) is unlikely to form a productive hydrogen bonding interaction with the glutamine substrate and this could explain why the former transports only fewer amino acids. Ala390, a conserved residue in neutral amino acid transporters, SLC1A4 and SLC1A5, does not form any contact with the Gln substrate and could tolerate small neutral amino acid substrates (Figure 3b, Figure 1—figure supplement 5). In contrast, the structurally equivalent residue Thr in SLC1A3 forms a hydrogen bond with the carboxylate group of the substrate.

Stretches of residual EM densities were observed in a hydrophobic crevice near the domain interface in the presence or absence of glutamine (Figure 1a). Based on the shape of the density and the presence of cholesteryl hemisuccinate (CHS) during purification, we assigned this density to CHS. Using an unbiased lipidomics approach, CHS was detected as a major component in the sample used for cryo-EM. The identity of CHS was further confirmed by mass spectrometry using commercially available CHS as a reference (Figure 3—figure supplement 1).

The cryo-EM density supports a specific binding site for CHS in the outward-facing SLC1A5 conformation. Electron density suggests that CHS binds inside a hydrophobic pocket formed by TM3, TM4a, TM4c and TM7 from one protomer, as well as TM5 from the adjacent protomer (Figure 3—figure supplements 2,3). The putative lipid density was also observed in the inward-facing state of SLC1A5, located in a hydrophobic pocket formed by TM3, TM4a, TM4c, and HP2a from one protomer, and TM5 from the adjacent protomer (Garaeva et al., 2018). Docking a CHS molecule in this putative density shows that the corresponding location is similar to that of the outward-facing state despite different pocket shape between two conformational states (Figure 3—figure supplement 2). Remarkably, the specific CHS-binding site in SLC1A5 overlaps with the UCPH₁₀₁ allosteric binding site reported in SLC1A3 structure (Canul-Tec et al., 2017) (Figure 3—figure supplement 4). The highly lipophilic UCPH₁₀₁ binds in the outward-facing conformation and blocks the movement of the transport domain relative to the scaffold domain. Additional putative lipid density was reported previously in the inward-facing state of SLC1A5, located between the transport and scaffold domains within the same protomer (Garaeva et al., 2018). The corresponding density was not observed in the outward-facing state of SLC1A5 due to differences in the local environment between the transport and scaffold domains.

Human SLC1A5 is the first eukaryotic sodium-dependent neutral amino acid transporter for which both the inward- and outward-facing conformational states have been mapped by cryo-EM. Our cryo-EM structures of unbound and substrate-bound SLC1A5 revealed that HP2 is the extracellular gate that controls entry of the substrate (Boudker et al., 2007; Canul-Tec et al., 2017). As a gatekeeper, HP2 in the closed conformation is near the serine-rich HP1, shielding the substrate from the extracellular surface (Figure 3a). The structure of the unliganded form of SLC1A5 provides an example for the HP2 loop in an open conformation and serves as a useful starting model for designing competitive inhibitors such as TBOA (Boudker et al., 2007; Colas et al., 2015). The opening of the HP2 loop increases the accessible surface area, exposing a previously unidentified cryptic pocket which could accommodate a large inhibitor that could specifically target the open conformation. In contrast, the closed conformation of HP2 has a small binding site, limiting the size of the inhibitors (Colas et al., 2016; Singh et al., 2017).

The availability of the three-dimensional structure of SLC1A5 in its outward-facing conformation open several paths forward for the development of drugs. The structure could be used for example identification of chemical leads by virtual screening methods. Secondly, generating additional structures with lead molecules based on substrate-analogs could provide a strong structure-based drug design platform for medicinal chemistry optimization of compound potencies. One important aspect of drug discovery is the need to engineer selectivity of lead compounds for the specific target in a given family. This is especially true for modulators of SLC1A5 where dialing out off-target activities against other closely-related sub-family members is highly desirable. Historically, this has been an area of strength for structure-based design. Structures have also been incredibly useful for optimization of ADME properties of compounds which is a critical element in the overall drug discovery process. Lastly, we believe the most relevant conformation to target for SLC1A5 inhibitors is the outward-facing conformation that we have described in our current work as that might be readily accessible for ligands from extracellular space.

The human SLC1A5 represents a distinct SLC1 subfamily identifiable in part by the Cys and Ala residues for substrate specificity and the unique conformation of the two ECL2a and ECL2b loops allowing for structural flexibility at TM4b for the key trimer interaction. Furthermore, ECL2a encircles the transport domain in the outward-facing state, covering a large part of the interaction surface for the inward facing state. In the inward-facing state the ECL2a is repositioned, exposing the interaction surface for the scaffold domain (Figure 4) (Garaeva et al., 2018).

Figure 4

Download asset Open asset

The existence of a potential allosteric binding pocket observed in the outward- and inward-facing SLC1A5 structures could facilitate the design of selective inhibitors that can bind to a site that is remote from the conserved substrate-binding site. Such allosteric druggable pockets at the lipid-exposed surface near the intracellular portions of membrane proteins have recently been structurally elucidated in several proteins and this raises new opportunities for drug design (Figure 3—figure supplement 4). We believe that these findings provide new and exciting structural insights and opportunities for antagonizing glutamine metabolism at the transporter level. Our structures of human SLC1A5, in both unbound and substrate-bound forms, represent the first high resolution cryo-EM structures of SLC1 family captured in the outward-facing state. These structures are an early example of the value that cryo-EM may bring to structure-based drug design.

All the cryo-EM data were deposited to the Protein Data Bank (PDB ID: 6MP6, 6MPB) and the EMDB (EMD-9187, EMD-9188) for immediate release upon publication.

1. 1. Yu X
   2. Plotnikova O
   3. Bonin PD
   4. Subashi TA
   5. McLellan TJ
   6. Dumlao D
   7. Che Y
   8. Dong YY
   9. Carpenter EP
   10. West GM
   11. Qiu X
   12. Culp JS
   13. Han S

   (2019) Protein Data Bank

   ID 6MP6. SLC1A5_cKM4012.

   https://www.rcsb.org/structure/6MP6
2. 1. Yu X
   2. Plotnikova O
   3. Bonin PD
   4. Subashi TA
   5. McLellan TJ
   6. Dumlao D
   7. Che Y
   8. Dong YY
   9. Carpenter EP
   10. West GM
   11. Qiu X
   12. Culp JS
   13. Han S

   (2019) Protein Data Bank

   ID 6MPB. SLC1A5_cKM4012_L_Gln.

   https://www.rcsb.org/structure/6MPB
3. 1. Yu X
   2. Plotnikova O
   3. Bonin PD
   4. Subashi TA
   5. McLellan TJ
   6. Dumlao D
   7. Che Y
   8. Dong YY
   9. Carpenter EP
   10. West GM
   11. Qiu X
   12. Culp JS
   13. Han S

   (2019) Electron Microscopy Data Bank

   ID EMD-9187. SLC1A5_cKM4012.

   https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-9187
4. 1. Yu X
   2. Plotnikova O
   3. Bonin PD
   4. Subashi TA
   5. McLellan TJ
   6. Dumlao D
   7. Che Y
   8. Dong YY
   9. Carpenter EP
   10. West GM
   11. Qiu X
   12. Culp JS
   13. Han S

   (2019) Electron Microscopy Data Bank

   ID EMD-9188. SLC1A5_cKM4012_L_Gln.

   https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-9188

1. 1. Adams PD
   2. Afonine PV
   3. Bunkóczi G
   4. Chen VB
   5. Davis IW
   6. Echols N
   7. Headd JJ
   8. Hung LW
   9. Kapral GJ
   10. Grosse-Kunstleve RW
   11. McCoy AJ
   12. Moriarty NW
   13. Oeffner R
   14. Read RJ
   15. Richardson DC
   16. Richardson JS
   17. Terwilliger TC
   18. Zwart PH

   (2010) PHENIX: a comprehensive Python-based system for macromolecular structure solution

   Acta Crystallographica Section D Biological Crystallography 66:213–221.

   https://doi.org/10.1107/S0907444909052925

   • PubMed
   • Google Scholar
2. 1. Akyuz N
   2. Georgieva ER
   3. Zhou Z
   4. Stolzenberg S
   5. Cuendet MA
   6. Khelashvili G
   7. Altman RB
   8. Terry DS
   9. Freed JH
   10. Weinstein H
   11. Boudker O
   12. Blanchard SC

   (2015) Transport domain unlocking sets the uptake rate of an aspartate transporter

   Nature 518:68–73.

   https://doi.org/10.1038/nature14158

   • PubMed
   • Google Scholar
3. 1. Arnold K
   2. Bordoli L
   3. Kopp J
   4. Schwede T

   (2006) The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling

   Bioinformatics 22:195–201.

   https://doi.org/10.1093/bioinformatics/bti770

   • PubMed
   • Google Scholar
4. 1. Barad BA
   2. Echols N
   3. Wang RY
   4. Cheng Y
   5. DiMaio F
   6. Adams PD
   7. Fraser JS

   (2015) EMRinger: side chain-directed model and map validation for 3D cryo-electron microscopy

   Nature Methods 12:943–946.

   https://doi.org/10.1038/nmeth.3541

   • PubMed
   • Google Scholar
5. 1. Boudker O
   2. Ryan RM
   3. Yernool D
   4. Shimamoto K
   5. Gouaux E

   (2007) Coupling substrate and ion binding to extracellular gate of a sodium-dependent aspartate transporter

   Nature 445:387–393.

   https://doi.org/10.1038/nature05455

   • PubMed
   • Google Scholar
6. 1. Canul-Tec JC
   2. Assal R
   3. Cirri E
   4. Legrand P
   5. Brier S
   6. Chamot-Rooke J
   7. Reyes N

   (2017) Structure and allosteric inhibition of excitatory amino acid transporter 1

   Nature 544:446–451.

   https://doi.org/10.1038/nature22064

   • PubMed
   • Google Scholar
7. 1. Colas C
   2. Grewer C
   3. Otte NJ
   4. Gameiro A
   5. Albers T
   6. Singh K
   7. Shere H
   8. Bonomi M
   9. Holst J
   10. Schlessinger A

   (2015) Ligand discovery for the Alanine-Serine-Cysteine transporter (ASCT2, SLC1A5) from homology modeling and virtual screening

   PLOS Computational Biology 11:e1004477.

   https://doi.org/10.1371/journal.pcbi.1004477

   • PubMed
   • Google Scholar
8. 1. Colas C
   2. Ung PM
   3. Schlessinger A

   (2016) SLC transporters: structure, function, and drug discovery

   MedChemComm 7:1069–1081.

   https://doi.org/10.1039/C6MD00005C

   • PubMed
   • Google Scholar
9. 1. Emsley P
   2. Cowtan K

   (2004) Coot: model-building tools for molecular graphics

   Acta Crystallographica. Section D, Biological Crystallography 60:2126–2132.

   https://doi.org/10.1107/S0907444904019158

   • PubMed
   • Google Scholar
10. 1. Erkens GB
    2. Hänelt I
    3. Goudsmits JM
    4. Slotboom DJ
    5. van Oijen AM

    (2013) Unsynchronised subunit motion in single trimeric sodium-coupled aspartate transporters

    Nature 502:119–123.

    https://doi.org/10.1038/nature12538

    • PubMed
    • Google Scholar
11. 1. Fuchs BC
    2. Bode BP

    (2005) Amino acid transporters ASCT2 and LAT1 in Cancer: partners in crime?

    Seminars in Cancer Biology 15:254–266.

    https://doi.org/10.1016/j.semcancer.2005.04.005

    • PubMed
    • Google Scholar
12. 1. Gao Y
    2. Cao E
    3. Julius D
    4. Cheng Y

    (2016) TRPV1 structures in nanodiscs reveal mechanisms of ligand and lipid action

    Nature 534:347–351.

    https://doi.org/10.1038/nature17964

    • PubMed
    • Google Scholar
13. 1. Garaeva AA
    2. Oostergetel GT
    3. Gati C
    4. Guskov A
    5. Paulino C
    6. Slotboom DJ

    (2018) Cryo-EM structure of the human neutral amino acid transporter ASCT2

    Nature Structural & Molecular Biology 25:515–521.

    https://doi.org/10.1038/s41594-018-0076-y

    • PubMed
    • Google Scholar
14. 1. Garaeva AA
    2. Guskov A
    3. Slotboom DJ
    4. Paulino C

    (2019) A one-gate elevator mechanism for the human neutral amino acid transporter ASCT2

    Nature Communications 10:3427.

    https://doi.org/10.1038/s41467-019-11363-x

    • PubMed
    • Google Scholar
15. 1. Georgieva ER
    2. Borbat PP
    3. Ginter C
    4. Freed JH
    5. Boudker O

    (2013) Conformational ensemble of the sodium-coupled aspartate transporter

    Nature Structural & Molecular Biology 20:215–221.

    https://doi.org/10.1038/nsmb.2494

    • PubMed
    • Google Scholar
16. 1. Grant T
    2. Rohou A
    3. Grigorieff N

    (2018) cisTEM, user-friendly software for single-particle image processing

    eLife 7:e35383.

    https://doi.org/10.7554/eLife.35383

    • PubMed
    • Google Scholar
17. 1. Guskov A
    2. Jensen S
    3. Faustino I
    4. Marrink SJ
    5. Slotboom DJ

    (2016) Coupled binding mechanism of three sodium ions and aspartate in the glutamate transporter homologue GltTk

    Nature Communications 7:13420.

    https://doi.org/10.1038/ncomms13420

    • PubMed
    • Google Scholar
18. 1. Jensen S
    2. Guskov A
    3. Rempel S
    4. Hänelt I
    5. Slotboom DJ

    (2013) Crystal structure of a substrate-free aspartate transporter

    Nature Structural & Molecular Biology 20:1224–1226.

    https://doi.org/10.1038/nsmb.2663

    • PubMed
    • Google Scholar
19. 1. Ji Y
    2. Postis VL
    3. Wang Y
    4. Bartlam M
    5. Goldman A

    (2016) Transport mechanism of a glutamate transporter homologue GltPh

    Biochemical Society Transactions 44:898–904.

    https://doi.org/10.1042/BST20160055

    • PubMed
    • Google Scholar
20. 1. Kanai Y
    2. Hediger MA

    (2004) The glutamate/neutral amino acid transporter family SLC1: molecular, physiological and pharmacological aspects

    Pflügers Archiv European Journal of Physiology 447:469–479.

    https://doi.org/10.1007/s00424-003-1146-4

    • PubMed
    • Google Scholar
21. 1. Kimanius D
    2. Forsberg BO
    3. Scheres SH
    4. Lindahl E

    (2016) Accelerated cryo-EM structure determination with parallelisation using GPUs in RELION-2

    eLife 5:e18722.

    https://doi.org/10.7554/eLife.18722

    • PubMed
    • Google Scholar
22. 1. Kucukelbir A
    2. Sigworth FJ
    3. Tagare HD

    (2014) Quantifying the local resolution of cryo-EM density maps

    Nature Methods 11:63–65.

    https://doi.org/10.1038/nmeth.2727

    • PubMed
    • Google Scholar
23. 1. Lu J
    2. Byrne N
    3. Wang J
    4. Bricogne G
    5. Brown FK
    6. Chobanian HR
    7. Colletti SL
    8. Di Salvo J
    9. Thomas-Fowlkes B
    10. Guo Y
    11. Hall DL
    12. Hadix J
    13. Hastings NB
    14. Hermes JD
    15. Ho T
    16. Howard AD
    17. Josien H
    18. Kornienko M
    19. Lumb KJ
    20. Miller MW
    21. Patel SB
    22. Pio B
    23. Plummer CW
    24. Sherborne BS
    25. Sheth P
    26. Souza S
    27. Tummala S
    28. Vonrhein C
    29. Webb M
    30. Allen SJ
    31. Johnston JM
    32. Weinglass AB
    33. Sharma S
    34. Soisson SM

    (2017) Structural basis for the cooperative allosteric activation of the free fatty acid receptor GPR40

    Nature Structural & Molecular Biology 24:570–577.

    https://doi.org/10.1038/nsmb.3417

    • PubMed
    • Google Scholar
24. 1. Marshall AD
    2. van Geldermalsen M
    3. Otte NJ
    4. Lum T
    5. Vellozzi M
    6. Thoeng A
    7. Pang A
    8. Nagarajah R
    9. Zhang B
    10. Wang Q
    11. Anderson L
    12. Rasko JEJ
    13. Holst J

    (2017) ASCT2 regulates glutamine uptake and cell growth in endometrial carcinoma

    Oncogenesis 6:e367.

    https://doi.org/10.1038/oncsis.2017.70

    • PubMed
    • Google Scholar
25. 1. Nakaya M
    2. Xiao Y
    3. Zhou X
    4. Chang JH
    5. Chang M
    6. Cheng X
    7. Blonska M
    8. Lin X
    9. Sun SC

    (2014) Inflammatory T cell responses rely on amino acid transporter ASCT2 facilitation of glutamine uptake and mTORC1 kinase activation

    Immunity 40:692–705.

    https://doi.org/10.1016/j.immuni.2014.04.007

    • PubMed
    • Google Scholar
26. 1. Oppedisano F
    2. Pochini L
    3. Galluccio M
    4. Indiveri C

    (2007) The glutamine/amino acid transporter (ASCT2) reconstituted in liposomes: transport mechanism, regulation by ATP and characterization of the glutamine/glutamate antiport

    Biochimica Et Biophysica Acta (BBA) - Biomembranes 1768:291–298.

    https://doi.org/10.1016/j.bbamem.2006.09.002

    • PubMed
    • Google Scholar
27. 1. Penczek PA
    2. Grassucci RA
    3. Frank J

    (1994) The ribosome at improved resolution: new techniques for merging and orientation refinement in 3D cryo-electron microscopy of biological particles

    Ultramicroscopy 53:251–270.

    https://doi.org/10.1016/0304-3991(94)90038-8

    • PubMed
    • Google Scholar
28. 1. Pettersen EF
    2. Goddard TD
    3. Huang CC
    4. Couch GS
    5. Greenblatt DM
    6. Meng EC
    7. Ferrin TE

    (2004) UCSF chimera--a visualization system for exploratory research and analysis

    Journal of Computational Chemistry 25:1605–1612.

    https://doi.org/10.1002/jcc.20084

    • PubMed
    • Google Scholar
29. 1. Pingitore P
    2. Pochini L
    3. Scalise M
    4. Galluccio M
    5. Hedfalk K
    6. Indiveri C

    (2013) Large scale production of the active human ASCT2 (SLC1A5) transporter in Pichia pastoris--functional and kinetic asymmetry revealed in proteoliposomes

    Biochimica Et Biophysica Acta (BBA) - Biomembranes 1828:2238–2246.

    https://doi.org/10.1016/j.bbamem.2013.05.034

    • PubMed
    • Google Scholar
30. 1. Reyes N
    2. Ginter C
    3. Boudker O

    (2009) Transport mechanism of a bacterial homologue of glutamate transporters

    Nature 462:880–885.

    https://doi.org/10.1038/nature08616

    • PubMed
    • Google Scholar
31. 1. Reyes N
    2. Oh S
    3. Boudker O

    (2013) Binding thermodynamics of a glutamate transporter homolog

    Nature Structural & Molecular Biology 20:634–640.

    https://doi.org/10.1038/nsmb.2548

    • PubMed
    • Google Scholar
32. 1. Robertson N
    2. Rappas M
    3. Doré AS
    4. Brown J
    5. Bottegoni G
    6. Koglin M
    7. Cansfield J
    8. Jazayeri A
    9. Cooke RM
    10. Marshall FH

    (2018) Structure of the complement C5a receptor bound to the extra-helical antagonist NDT9513727

    Nature 553:111–114.

    https://doi.org/10.1038/nature25025

    • PubMed
    • Google Scholar
33. 1. Rosenthal PB
    2. Henderson R

    (2003) Optimal determination of particle orientation, absolute hand, and contrast loss in single-particle electron cryomicroscopy

    Journal of Molecular Biology 333:721–745.

    https://doi.org/10.1016/j.jmb.2003.07.013

    • PubMed
    • Google Scholar
34. 1. Scalise M
    2. Pochini L
    3. Panni S
    4. Pingitore P
    5. Hedfalk K
    6. Indiveri C

    (2014) Transport mechanism and regulatory properties of the human amino acid transporter ASCT2 (SLC1A5)

    Amino Acids 46:2463–2475.

    https://doi.org/10.1007/s00726-014-1808-x

    • PubMed
    • Google Scholar
35. 1. Scalise M
    2. Pochini L
    3. Console L
    4. Pappacoda G
    5. Pingitore P
    6. Hedfalk K
    7. Indiveri C

    (2018) Cys Site-Directed mutagenesis of the human SLC1A5 (ASCT2) Transporter: structure/Function relationships and crucial role of Cys467 for redox sensing and glutamine transport

    International Journal of Molecular Sciences 19:648.

    https://doi.org/10.3390/ijms19030648

    • PubMed
    • Google Scholar
36. 1. Scheres SH
    2. Chen S

    (2012) Prevention of overfitting in cryo-EM structure determination

    Nature Methods 9:853–854.

    https://doi.org/10.1038/nmeth.2115

    • PubMed
    • Google Scholar
37. 1. Schrodinger LLC

    (2015)

    The PyMOL Molecular Graphics System

    The PyMOL Molecular Graphics System.

    • Google Scholar
38. 1. Scopelliti AJ
    2. Font J
    3. Vandenberg RJ
    4. Boudker O
    5. Ryan RM

    (2018) Structural characterisation reveals insights into substrate recognition by the glutamine transporter ASCT2/SLC1A5

    Nature Communications 9:38.

    https://doi.org/10.1038/s41467-017-02444-w

    • PubMed
    • Google Scholar
39. Patent

    1. Shiraishi N
    2. Furuya A
    3. Toki H
    4. Ando H
    5. Suzuki M
    6. Kubota T

    (2012) Anti-system asc amino acid transporter 2 (ASCT2) (US-8268592-B2)

    ANTIBODYKYOWA Hakko Kirin Co Ltd. United States.

    https://app.dimensions.ai/details/patent/WO-2010008075-A1
40. 1. Singh K
    2. Tanui R
    3. Gameiro A
    4. Eisenberg G
    5. Colas C
    6. Schlessinger A
    7. Grewer C

    (2017) Structure activity relationships of benzylproline-derived inhibitors of the glutamine transporter ASCT2

    Bioorganic & Medicinal Chemistry Letters 27:398–402.

    https://doi.org/10.1016/j.bmcl.2016.12.063

    • PubMed
    • Google Scholar
41. 1. Suzuki M
    2. Toki H
    3. Furuya A
    4. Ando H

    (2017) Establishment of monoclonal antibodies against cell surface domains of ASCT2/SLC1A5 and their inhibition of glutamine-dependent tumor cell growth

    Biochemical and Biophysical Research Communications 482:651–657.

    https://doi.org/10.1016/j.bbrc.2016.11.089

    • PubMed
    • Google Scholar
42. 1. Utsunomiya-Tate N
    2. Endou H
    3. Kanai Y

    (1996) Cloning and functional characterization of a system ASC-like na+-dependent neutral amino acid transporter

    Journal of Biological Chemistry 271:14883–14890.

    https://doi.org/10.1074/jbc.271.25.14883

    • PubMed
    • Google Scholar
43. 1. van Geldermalsen M
    2. Wang Q
    3. Nagarajah R
    4. Marshall AD
    5. Thoeng A
    6. Gao D
    7. Ritchie W
    8. Feng Y
    9. Bailey CG
    10. Deng N
    11. Harvey K
    12. Beith JM
    13. Selinger CI
    14. O'Toole SA
    15. Rasko JE
    16. Holst J

    (2016) ASCT2/SLC1A5 controls glutamine uptake and tumour growth in triple-negative basal-like breast Cancer

    Oncogene 35:3201–3208.

    https://doi.org/10.1038/onc.2015.381

    • PubMed
    • Google Scholar
44. 1. Verdon G
    2. Oh S
    3. Serio RN
    4. Boudker O

    (2014) Coupled ion binding and structural transitions along the transport cycle of glutamate transporters

    eLife 3:e02283.

    https://doi.org/10.7554/eLife.02283

    • PubMed
    • Google Scholar
45. 1. Verdon G
    2. Boudker O

    (2012) Crystal structure of an asymmetric trimer of a bacterial glutamate transporter homolog

    Nature Structural & Molecular Biology 19:355–357.

    https://doi.org/10.1038/nsmb.2233

    • PubMed
    • Google Scholar
46. 1. Yernool D
    2. Boudker O
    3. Jin Y
    4. Gouaux E

    (2004) Structure of a glutamate transporter homologue from Pyrococcus horikoshii

    Nature 431:811–818.

    https://doi.org/10.1038/nature03018

    • PubMed
    • Google Scholar
47. 1. Zhang K

    (2016) Gctf: real-time CTF determination and correction

    Journal of Structural Biology 193:1–12.

    https://doi.org/10.1016/j.jsb.2015.11.003

    • PubMed
    • Google Scholar
48. 1. Zhang H
    2. Qiao A
    3. Yang D
    4. Yang L
    5. Dai A
    6. de Graaf C
    7. Reedtz-Runge S
    8. Dharmarajan V
    9. Zhang H
    10. Han GW
    11. Grant TD
    12. Sierra RG
    13. Weierstall U
    14. Nelson G
    15. Liu W
    16. Wu Y
    17. Ma L
    18. Cai X
    19. Lin G
    20. Wu X
    21. Geng Z
    22. Dong Y
    23. Song G
    24. Griffin PR
    25. Lau J
    26. Cherezov V
    27. Yang H
    28. Hanson MA
    29. Stevens RC
    30. Zhao Q
    31. Jiang H
    32. Wang MW
    33. Wu B

    (2017) Structure of the full-length glucagon class B G-protein-coupled receptor

    Nature 546:259–264.

    https://doi.org/10.1038/nature22363

    • PubMed
    • Google Scholar
49. 1. Zheng SQ
    2. Palovcak E
    3. Armache JP
    4. Verba KA
    5. Cheng Y
    6. Agard DA

    (2017) MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy

    Nature Methods 14:331–332.

    https://doi.org/10.1038/nmeth.4193

    • PubMed
    • Google Scholar

Thank you for submitting your article "Structural basis for the transport mechanism of the human glutamine transporter SLC1A5 (ASCT2)" for consideration by eLife. Your article has been reviewed by Olga Boudker as the Senior Editor, a Reviewing Editor and three peer reviewers. The following individuals involved in review of your submission have agreed to reveal their identity: Robert John Vandenberg (Reviewer #3); Joel Meyerson (Reviewer #4).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

The neutral amino acid transporter ASCT2 is a member of the SLC1 family of transporters and a target of cancer therapeutics. The manuscript by Yu et al. presents the determination of a structure of ASCT2 in an outward-facing conformation, complementing a previously published structure of an inward facing structure of ASCT2, and providing details of the substrate binding site, which should prove invaluable for the design of drugs to inhibit the transporter.

The reviewers appreciate the high quality of the structural work and general importance of the structure, but note that it requires a more direct connection to physiology, e.g. the differences between exchangers and transporters, and/or to drug design. To focus the framing of the article we therefore strongly suggest that you rewrite the article as a Short Report, focus on the substrate and putative CHS binding sites, and exclude the HDX data as explained below.

Essential revisions:

The following are the points that should be considered when improving the physiological context of the structure in the revised manuscript:

1) Based on the structures, is it clear why ASCT2 and ASCT1 have different substrate specificity?

2) From the drug design perspective please provide more specific comments on the potential applications of the structure for the development of drugs.

3) It is not clear whether the protein imaged without L-Gln was in high Na⁺ concentration. If yes, then the structure likely represents a Na⁺-bound transporter, which should be evident from the arrangement of the Na⁺-binding residues. If not, or if the affinity for Na⁺ is low, then one would expect to find disrupted Na⁺ binding sites, which again should be visible in the structure.

4) ASCTs are obligate Na⁺-dependent exchangers, while EAATs are concentrative pumps. Could it be rationalized from the structures, why ASCT2 cannot return to the outward-facing states without the bound amino acid and Na⁺ ions?

5) Mutations have been made in GltPh mimicking ASCT2 (including R to C mutation), but when the structure was solved with Gln, the HP2 loop was still open. Can it be rationalized from the current structure what is the difference between the binding pockets and/or HP2 between ACST2 and GltPh that allows larger substrates (even when R is mutated to C)?

6) The manuscript methodically indexes many structural features including protomer interactions, helix orientations, key amino acid positions, glycosylation sites, glutamine and CHS binding sites, etc. However, this style is primarily descriptive, so if the authors intend to retain these details, they should to be enhanced with a clearer narrative explaining the significance of the structural findings. Below are three examples.

The authors point out that "based on sequence alignment, both SLC1A4 and SLC1A5 have the longest ECL2a region compared to other SLC1 family members including prokaryotic homologues." What is the significance of this?

Similarly, the authors write: "Importantly, the ECL2b was reported to play a critical role in determining the receptor properties of SLC1A5 for retroviruses." Can the structure help rationalize this previous report?

The authors write: "Similarly, Arg376 of the transport domain has a potential to form a salt bridge with the Glu264 of scaffold domain in the outward-facing structure (Figure 3B-C). Both Arg376 and Glu264 are highly conserved among the SLC1 family (Figure 1—figure supplement 6)." What are the implications of these observations?

In addition, please address the following issues in the manuscript and/or in responses to the reviewers:

1) Is it known if the Fabs used to aid particle alignment influence transport?

2) The characterization of the transport properties of ASCT2 are based on a single concentration of substrate, which generates a signal to noise ratio of 2. This level of analysis does not instill sufficient confidence in the functional properties of the transporter. It would be preferable to have a measure of substrate affinity, and also demonstration of the exchange properties of the transporter, which would reflect previous studies measuring the functional properties of ASCT2.

3) How significant are the deviations form 3-fold symmetry between protomers? From the superpositions it looks like the deviations are truly minute. Is higher resolution obtained if 3C symmetry is imposed during 3D alignment and refinement? If yes, improved resolution may help analyzing the status of Na⁺ binding sites.

4) Have the authors attempted to perform symmetry expansion to detect whether there is any heterogeneity in the protomers? Such heterogeneity might be masked even if C1 is used for processing.

5) The HDX-MS data is analyzed only very superficially, and as it stands, the level of detail provided does not allow a thorough analysis of the data. On the face of it, these data also do not appear to add significant understanding to the physiological mechanism nor the drug targeting of ASCT2. We recommend removing this section and publishing elsewhere.

[Editors' note: further revisions were requested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Cryo-EM structures of the human glutamine transporter SLC1A5 (ASCT2) in the outward-facing conformation" for further consideration at eLife. Your revised article has been favorably evaluated by Olga Boudker as the Senior Editor, and a Reviewing Editor.

The manuscript has been improved but there is one remaining issue that needs to be addressed before acceptance, as outlined below:

- Since it appears that little uptake activity is seen (over the background) in either cells or proteoliposomes, please move Figure 1A to the supplementary information and state in the main text explicitly that the observed transport activity was low, compared to the background. Please also use color to distinguish the points in that figure.

Summary:

The neutral amino acid transporter ASCT2 is a member of the SLC1 family of transporters and a target of cancer therapeutics. The manuscript by Yu et al. presents the determination of a structure of ASCT2 in an outward-facing conformation, complementing a previously published structure of an inward facing structure of ASCT2, and providing details of the substrate binding site, which should prove invaluable for the design of drugs to inhibit the transporter.

The reviewers appreciate the high quality of the structural work and general importance of the structure, but note that it requires a more direct connection to physiology, e.g. the differences between exchangers and transporters, and/or to drug design. To focus the framing of the article we therefore strongly suggest that you rewrite the article as a Short Report, focus on the substrate and putative CHS binding sites, and exclude the HDX data as explained below.

We would like to thank the reviewers for their valuable time and careful and critical evaluation of our manuscript. We have addressed each of their concerns to the best of our ability as described below. We have rewritten the article as a more concise Short Report that focuses on the substrate and CHS binding sites and we have removed the HDX data as requested by the editor.

Essential revisions:

The following are the points that should be considered when improving the physiological context of the structure in the revised manuscript:

1) Based on the structures, is it clear why ASCT2 and ASCT1 have different substrate specificity?

Careful comparison of the sequences of ASCT1 and ASCT2 around the substrate binding site show that the most significant difference is at position 467 – ASCT1 has a threonine at this position while ASCT2 has a cysteine. The side chain sulfhydryl group of Cys 467 in ASCT2 appears to form a hydrogen bond with the side chain of glutamine in our structure. Since the structure of ASCT1 is unknown, we attempted to generate a model for ASCT1 by replacing Cys with a threonine. Unfortunately, none of the three rotamers available for threonine can be satisfactorily accommodated because of steric clashes with backbone carbonyl atoms. This suggests that ASCT1 must have a slightly different local conformation around the substrate binding site. Regardless of the local conformation that ASCT1 adopts in this region, the side chain hydroxyl group of Thr459 (analogous to Cys467 in ASCT2) is unlikely to form a productive hydrogen bonding interaction with the glutamine substrate and this could explain why the former transports only fewer amino acids. We have incorporated this description into our revised manuscript.

2) From the drug design perspective please provide more specific comments on the potential applications of the structure for the development of drugs.

SLC1A5 is a high value target in the pharmaceutical industry for the development of drugs for immunomodulatory drugs. The availability of the three-dimensional structure of SLC1A5 in its outward facing conformation open several paths forward for the development of drugs. (1) The structure could be used for identification of chemical leads by virtual screening methods. (2) Secondly, generating additional structures with lead molecules based on substrate-analogs could provide a strong structure-based drug design platform for medicinal chemistry optimization of compound potencies. (3) One important aspect of drug discovery is the need to engineer selectivity of lead compounds for the specific target in a given subfamily. This is especially true for modulators of SLC1A5 where dialing out off-target activities against other closely-related sub-family members is highly desirable. Historically, this has been an area of strength for structure-based design. (4) Structures have also been incredibly useful for optimization of ADME properties of compounds which is a critical element in the overall drug discovery process. (5) We believe the most relevant conformation to target for SLC1A5 inhibitors is the outward-facing conformation that we have described in our current work as that is the most relevant conformation for ligands from extracellular space. We have revised the manuscript to incorporate the above section.

3) It is not clear whether the protein imaged without L-Gln was in high Na⁺ concentration. If yes, then the structure likely represents a Na⁺-bound transporter, which should be evident from the arrangement of the Na⁺-binding residues. If not, or if the affinity for Na⁺ is low, then one would expect to find disrupted Na⁺ binding sites, which again should be visible in the structure.

We agree this is an important question to address. All protein samples in this work were imaged in presence of 100 mM NaCl. Thus, we do believe our apo structure represents Na-bound state of the transporter. We have revised the Materials and methods section to clearly state this. Given the moderate resolution of the structure and the lack of density for sodium and nearby side chains, we are unable to make a categorical statement regarding this. Future structures of ASCTs at different sodium concentrations and at higher resolution could help to clarify the precise role sodium plays in conformational switching.

4) ASCTs are obligate Na⁺-dependent exchangers, while EAATs are concentrative pumps. Could it be rationalized from the structures, why ASCT2 cannot return to the outward-facing states without the bound amino acid and Na⁺ ions?

This is also a very interesting question, but difficult to assess due to the fact that sodium ion is not visible in our structure and we do not have a structure in its absence. It is highly likely that significant conformational changes accompanies binding of sodium ions. Based on our structures in the presence and absence of glutamine, we do know that the presence of substrate does not impede the movement of HP2 loop. It is possible that higher resolution structures with and without both Na⁺ and Gln bound may shed light on this mechanism.

5) Mutations have been made in GltPh mimicking ASCT2 (including R to C mutation), but when the structure was solved with Gln, the HP2 loop was still open. Can it be rationalized from the current structure what is the difference between the binding pockets and/or HP2 between ACST2 and GltPh that allows larger substrates (even when R is mutated to C)?

We believe the editor is referring to the work of Scopelliti et al., 2018, which describes the structures of GltPH (R397C) bound to serine, cysteine or benzylCys. However, there is no Gln-bound structure of GltPh (WT or R397C mutant) that has been reported. In the Cys-bound GltPh-R397C structure, the HP2 loop adopts a “closed” conformation. Similarly, the HP2 loop is closed in one of the protomers in the Ser-bound GltPh structure, However, electron density suggests open conformations for the other two protomers but they were not modeled due to limited resolution. Therefore, it would be difficult to assess/compare structures of ASCT2 and GltPh with bound glutamine or comment on the precise conformation of the HP2 loop.

6) The manuscript methodically indexes many structural features including protomer interactions, helix orientations, key amino acid positions, glycosylation sites, glutamine and CHS binding sites, etc. However, this style is primarily descriptive, so if the authors intend to retain these details, they should to be enhanced with a clearer narrative explaining the significance of the structural findings. Below are three examples.

We thank reviewers for pointing out the description of structural features. Some of them have been streamlined or removed to fit the “short report” style and focus on the significance of the structural findings.

The authors point out that "based on sequence alignment, both SLC1A4 and SLC1A5 have the longest ECL2a region compared to other SLC1 family members including prokaryotic homologues." What is the significance of this?

As there is no significance, we have removed the sentence.

Similarly, the authors write: "Importantly, the ECL2b was reported to play a critical role in determining the receptor properties of SLC1A5 for retroviruses." Can the structure help rationalize this previous report?

As the structure doesn’t rationalize, we have removed.

The authors write: "Similarly, Arg376 of the transport domain has a potential to form a salt bridge with the Glu264 of scaffold domain in the outward-facing structure (Figure 3B-C). Both Arg376 and Glu264 are highly conserved among the SLC1 family (Figure 1—figure supplement 6)." What are the implications of these observations?

In the revised “Short Report” version, we removed the discussion.

In addition, please address the following issues in the manuscript and/or in responses to the reviewers:

1) Is it known if the Fabs used to aid particle alignment influence transport?

The biochemical role of the KM4012 for transport activity is currently unknown. It is reported that KM4012 antibody was isolated through a cell-based screen and inhibited glutamine-dependent cancer cell growth (Suzuki et al). The antibodies suppressed glutamine-dependent growth of WiDr colorectal cancer cells, which require glutamine in the culture medium to grow.

In this revised version we added a sentence “It is reported that KM4012 antibody isolated through a cell based screen inhibited glutamine-dependent cancer cell growth. (Suzuki et al)” for further clarification.

2) The characterization of the transport properties of ASCT2 are based on a single concentration of substrate, which generates a signal to noise ratio of 2. This level of analysis does not instill sufficient confidence in the functional properties of the transporter. It would be preferable to have a measure of substrate affinity, and also demonstration of the exchange properties of the transporter, which would reflect previous studies measuring the functional properties of ASCT2.

Thank you for pointing out the weak signal to noise ratio. The level of glutamine transport measured with our SLC1A5 PL assay system was ~6 times lower than the level reported by Pingatore et al. and by Graraeva et al. While Pingatore et al. reported that SLC1A5 in their proteoliposomes (PL) was 100% in the correct orientation (i.e. an orientation needed to measure glutamine uptake PL) only 20-25% of the SLC1A5 in our PL was in that same correct orientation (determined by the availability of an affinity tag in relation to PL surface (data not shown). This could be due to differences in the expression systems that were used by the two different labs. Both Pingatore et al. and Graraeva et al. used a yeast expression system, but our expression system used the mammalian HEK293 cell line. These differences could account for weaker signal we observed in our assay.

In lieu of further PL experiments, we demonstrate that when expressed in cells, the affinity-tagged SLC1A5 used in this study produced a level of glutamine uptake that was comparable to untagged WT SLC1A5, thus suggesting that the presence of the affinity-tags did not alter the biological activity of SLC1A5. To make it more evident that the glutamine transport activity of our tag-version of SLC1A5 is comparable to the wild type protein, we tested a broad range of glutamine concentration and both wild type and the tag version of SLC1A5 showed similar activity. This is depicted in the revised Figure 1A.

We have also explored biophysical studies using SPR but were unsuccessful due to technical challenges with the detergent solubilized protein reagent.

3) How significant are the deviations form 3-fold symmetry between protomers? From the superpositions it looks like the deviations are truly minute. Is higher resolution obtained if 3C symmetry is imposed during 3D alignment and refinement? If yes, improved resolution may help analyzing the status of Na⁺ binding sites.

From the Supplementary Figure 7 (initial submission) on the superposition of each protomer, the results revealed that three protomers from the outward-facing SLC1A5 trimers share similar global fold but are not positioned symmetrically in space. Even though the deviations are truly minute, the fab-bound SLC1A5 trimers adopt the pseudo C3 symmetry and imposing C3 operation during the refinement did not improve the resolution further. The reason could be the presence of the Fab disrupts the low-resolution pseudosymmetry. We further experimented with Fab signals subtraction followed by refinements with/without C3 symmetry operation and the efforts did not improve. the resolution for the SLC1A5-cKM4012 (Fab) structure or the SLC1A5-cKM4012 (Fab)_Gln complex. In the revised “Short Report” version, we removed this figure.

4) Have the authors attempted to perform symmetry expansion to detect whether there is any heterogeneity in the protomers? Such heterogeneity might be masked even if C1 is used for processing.

Since the structures were reconstructed without any symmetry operations, no symmetry expansion was performed. From the 3D classification and the final reconstruction results, only the outward-facing state was observed. Minor heterogeneities were observed at the loop regions among the protomers.

5) The HDX-MS data is analyzed only very superficially, and as it stands, the level of detail provided does not allow a thorough analysis of the data. On the face of it, these data also do not appear to add significant understanding to the physiological mechanism nor the drug targeting of ASCT2. We recommend removing this section and publishing elsewhere.

We have removed the HDX section.

[Editors' note: further revisions were requested prior to acceptance, as described below.]

The manuscript has been improved but there is one remaining issue that needs to be addressed before acceptance, as outlined below:

- Since it appears that little uptake activity is seen (over the background) in either cells or proteoliposomes, please move Figure 1A to the supplementary information and state in the main text explicitly that the observed transport activity was low, compared to the background. Please also use color to distinguish the points in that figure.

We appreciate the referee’s concern and comment. We have moved “Figure 1A” to “Figure 1—figure supplement 1A”. The figure was revised in color to distinguish the points. We have also explicitly stated in the main text to read “Our full-length SLC1A5 construct showed high expression and stability. (Figure 1—figure supplement 1). The presence of affinity tags used for purification did not affect the sodium-dependent glutamine uptake when HAP1 SLC1A5 knock-out cells were transiently transfected with full-length SLC1A5 but the observed transport activity was low, compared to the background.” in the revised manuscript.

Author details

1. Xiaodi Yu

   Medicine Design, Pfizer Inc, Groton, United States

   Present address

   SMPS, Janssen Research and Development, Spring House, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   Contributed equally with

   Olga Plotnikova

   Competing interests

   is affiliated with Pfizer Inc. The author has no other competing interests to declare.

2. Olga Plotnikova

   Medicine Design, Pfizer Inc, Groton, United States

   Contribution

   Resources, Data curation, Formal analysis, Validation, Visualization, Methodology, Writing—original draft

   Contributed equally with

   Xiaodi Yu

   Competing interests

   is affiliated with Pfizer Inc. The author has no other competing interests to declare.

3. Paul D Bonin

   Medicine Design, Pfizer Inc, Groton, United States

   Contribution

   Resources, Data curation, Formal analysis, Validation, Methodology, Writing—original draft

   Competing interests

   is affiliated with Pfizer Inc. The author has no other competing interests to declare.

4. Timothy A Subashi

   Medicine Design, Pfizer Inc, Groton, United States

   Contribution

   Resources, Data curation, Formal analysis, Methodology

   Competing interests

   is affiliated with Pfizer Inc. The author has no other competing interests to declare.

5. Thomas J McLellan

   Medicine Design, Pfizer Inc, Groton, United States

   Contribution

   Resources, Methodology

   Competing interests

   is affiliated with Pfizer Inc. The author has no other competing interests to declare.

6. Darren Dumlao

   Medicine Design, Pfizer Inc, Groton, United States

   Contribution

   Resources, Validation, Methodology

   Competing interests

   is affiliated with Pfizer Inc. The author has no other competing interests to declare.

7. Ye Che

   Medicine Design, Pfizer Inc, Groton, United States

   Contribution

   Resources, Formal analysis, Validation, Methodology

   Competing interests

   is affiliated with Pfizer Inc. The author has no other competing interests to declare.

8. Yin Yao Dong

   Structural Genomics Consortium, University of Oxford, Oxford, United Kingdom

   Present address

   Neurosciences Group, Nuffield Department of Clinical Neuroscience, Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, United Kingdom

   Contribution

   Resources, Formal analysis, Validation, Methodology

   Competing interests

   No competing interests declared

9. Elisabeth P Carpenter

   Structural Genomics Consortium, University of Oxford, Oxford, United Kingdom

   Contribution

   Resources, Methodology

   Competing interests

   No competing interests declared

10. Graham M West

    Medicine Design, Pfizer Inc, Groton, United States

    Contribution

    Resources, Formal analysis, Methodology, Writing—original draft

    Competing interests

    is affiliated with Pfizer Inc. The author has no other competing interests to declare.

11. Xiayang Qiu

    Medicine Design, Pfizer Inc, Groton, United States

    Contribution

    Conceptualization, Resources, Formal analysis, Validation, Methodology, Writing—original draft

    Competing interests

    is affiliated with Pfizer Inc. The author has no other competing interests to declare.

12. Jeffrey S Culp

    Medicine Design, Pfizer Inc, Groton, United States

    Contribution

    Conceptualization, Supervision, Validation, Investigation, Methodology, Writing—original draft

    Competing interests

    is affiliated with Pfizer Inc. The author has no other competing interests to declare.

13. Seungil Han

    Medicine Design, Pfizer Inc, Groton, United States

    Contribution

    Conceptualization, Data curation, Supervision, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

    For correspondence

    seungil.han@pfizer.com

    Competing interests

    is affiliated with Pfizer Inc. The author has no other competing interests to declare.

    "This ORCID iD identifies the author of this article:" 0000-0002-1070-3880

• 5,127

  Page views

• 824

  Downloads

• 37

  Citations

Article citation count generated by polling the highest count across the following sources: Scopus, Crossref, PubMed Central.