Research Article

Structural basis for leucine sensing by the Sestrin2-mTORC1 pathway

See allHide authors and affiliations

Science  01 Jan 2016:
Vol. 351, Issue 6268, pp. 53-58
DOI: 10.1126/science.aad2087

From sensing leucine to metabolic control

The mTORC1 protein kinase complex plays central roles in regulating cell growth and metabolism and is implicated in common human diseases such as diabetes and cancer. The level of the amino acid leucine tells an organism a lot about its physiological state, including how much food is available, how much insulin is going to be needed, and whether new muscle mass can be made (see the Perspective by Buel and Blenis). Wolfson et al. identified a biochemical sensor of leucine, Sestrin2, which connects the concentration of leucine to the control of organismal metabolism and growth. When leucine bound to Sestrin2, it was released from a complex with the mTORC1 regulatory factor GATOR2, activating the mTORC1 complex. Saxton et al. describe the crystal structure of Sestrin2 and show how it specifically detects leucine. Aylett et al. determined the structure of human mTORC1 by cryoelectron microscopy and the crystal structure of a regulatory subunit, Raptor. The results reveal the structural basis for the function and intricate regulation of this important enzyme, which is also a strategic drug target.

Science, this issue p. 43, p. 48, p. 53; see also p. 25


Eukaryotic cells coordinate growth with the availability of nutrients through the mechanistic target of rapamycin complex 1 (mTORC1), a master growth regulator. Leucine is of particular importance and activates mTORC1 via the Rag guanosine triphosphatases and their regulators GATOR1 and GATOR2. Sestrin2 interacts with GATOR2 and is a leucine sensor. Here we present the 2.7 angstrom crystal structure of Sestrin2 in complex with leucine. Leucine binds through a single pocket that coordinates its charged functional groups and confers specificity for the hydrophobic side chain. A loop encloses leucine and forms a lid-latch mechanism required for binding. A structure-guided mutation in Sestrin2 that decreases its affinity for leucine leads to a concomitant increase in the leucine concentration required for mTORC1 activation in cells. These results provide a structural mechanism of amino acid sensing by the mTORC1 pathway.

The mechanistic target of rapamycin complex 1 (mTORC1) protein kinase is a major growth regulator that coordinates cell anabolism and catabolism with the availability of key nutrients such as amino acids (13). Among the amino acids, leucine is of particular interest because of its ability to promote important physiological phenomena, including muscle growth and satiety (46), in large part through the activation of mTORC1 (7, 8). However, the biochemical mechanism of leucine sensing by the mTORC1 pathway has remained elusive.

Whereas growth factors, energy, and other inputs signal to mTORC1 primarily through the tuberous sclerosis complex–Rheb axis (911), amino acids act by regulating the nucleotide state of the heterodimeric Rag guanosine triphosphatases (GTPases) and promoting the localization of mTORC1 to the lysosome, its site of activation (1214). Lysosomal amino acids including arginine are thought to signal to the Rags through a lysosomal membrane–associated complex consisting of the v-ATPase (vacuolar-type H+-dependent adenosine triphosphatase) (15), the Ragulator complex (16), and the putative arginine sensor SLC38A9 (17, 18). Cytosolic leucine, however, signals to the Rags through a distinct pathway consisting of GATOR1, which is the GTPase-activating protein for RagA and RagB, and a pentameric protein complex of unknown function called GATOR2 (19, 20).

Proteomic studies have identified the Sestrins as GATOR2-interacting proteins that inhibit mTORC1 only in the absence of amino acids (21, 22). Subsequent in vitro studies demonstrated that the Sestrin2-GATOR2 interaction is sensitive specifically to leucine, which binds Sestrin2 with a dissociation constant of ~20 μM. Human embryonic kidney (HEK)–293T cells expressing a Sestrin2 mutant that cannot bind leucine fail to activate mTORC1 in response to leucine, suggesting that Sestrin2 is the primary leucine sensor for the mTORC1 pathway in these cells (20). However, Sestrin2 shares no sequence similarity with known amino acid–binding domains, raising the question of how this protein can detect leucine and signal its presence to mTORC1.

Here we present the structure of human Sestrin2 in complex with leucine, revealing in atomic detail the mechanism of leucine sensing by the mTORC1 pathway.

Structure of leucine-bound Sestrin2

To understand how Sestrin2 detects leucine, we expressed and purified full-length human Sestrin2 from Escherichia coli and verified binding to leucine in vitro by differential scanning fluorimetry (fig. S1) (23). Although we were unable to obtain crystals of Sestrin2 alone, incubation of the protein with leucine allowed the formation of crystals containing leucine-bound Sestrin2 that diffracted to 2.7 Å resolution. We solved the structure using single-wavelength anomalous dispersion with selenomethionine-derivatized protein and refined the model against the native data to final working and free residuals, Rwork and Rfree, of 19.6% and 22.3%, respectively (table S1). Sestrin2 crystallized in acubic space group containing five copies per asymmetric unit.

Sestrin2 is a 55-kD, monomeric, all α-helical, globular protein that contains distinct N-terminal (NTD, residues 66 to 220) and C-terminal (CTD, residues 339 to 480) domains connected by a partially disordered, partially helical linker domain (residues 221 to 338) (Fig. 1A). The N-terminal 65 residues of the protein appear disordered and were not observed in our structure. Electron density map analysis revealed the presence of a single leucine molecule bound to Sestrin2 in the CTD (Fig. 2A).

Fig. 1 Structure of leucine-bound Sestrin2.

(A) Two views of human Sestrin2 are shown as ribbon diagrams, with the NTD, linker domain, and CTD colored in blue, gray, and teal, respectively. The bound leucine molecule is shown in orange. Disordered residues not present in the crystal structure (1 to 65, 242 to 255, 272 to 280, 296 to 309) are shown as dashed lines. (B) Structural superposition of the Sestrin2 NTD (blue, residues 66 to 220) and CTD (teal, residues 339 to 480). (C) Structural superposition of the Sestrin2 NTD (blue) and CTD (teal) with a R. eutropha AhpD dimer (pink; PDB accession code: 2PRR). (D) Immunoprecipitation of N- and C-terminal fragments of Sestrin2. HEK-293T cells transiently transfected with FLAG-metap2, FLAG-Sestrin2 full length (FL), FLAG-Sestrin2-NTD (1 to 220), FLAG-Sestrin2-CTD+L (CTD plus linker, 220 to 480), or both Flag-Sestrin2-NTD and HA-Sestrin2-CTD+L (HA, hemagglutinin epitope) were starved for amino acids for 50 min. FLAG immunoprecipitates were prepared from cell lysates. Immunoprecipitates and lysates from one representative experiment were analyzed by immunoblotting for the indicated proteins. WDR24 and Mios were used as representative GATOR2 components. (E) [3H]leucine binding assay using N- and C-terminal fragments of Sestrin2. FLAG immunoprecipitates prepared from HEK-293T cells transiently expressing the indicated proteins were used as described in the methods (supplementary materials). Unlabeled leucine was used as a competitor where indicated. Values are means ± SDs for three technical replicates from one representative experiment. Two-tailed t tests were used for comparisons between two groups.

Fig. 2 Recognition of leucine by Sestrin2.

(A) Close-up view of the leucine-binding pocket in Sestrin2, focusing on the bound leucine (shown in orange) together with its 2Fo–Fc electron density map, which was calculated and contoured at 1.5σ from an omit map lacking leucine and all pocket residues. Predicted hydrogen bonds or salt bridges are shown as black dashed lines. Helix numbers are labeled as in Fig. 1A. (B) Surface representation of leucine-bound Sestrin2, focusing on the leucine-binding pocket. The bound leucine is represented as a stick model (orange). Residues 373 to 387 are omitted to allow visibility of the pocket. Residue Glu451, which contacts the amine of leucine, is shown in red; Arg390, which contacts the carboxyl of leucine, is shown in blue. The domains of Sestrin2 are colored as in Fig. 1A. (C) Binding of the E451Q, R390A, and W444E mutants of Sestrin2 to leucine. HA immunoprecipitates prepared from HEK-293T cells transiently expressing the indicated HA-tagged proteins were used in binding assays with [3H]leucine. Binding was analyzed as in Fig. 1E. (D) Effect of leucine on the interactions of Sestrin2 E451Q, R390A, or W444L with GATOR2. FLAG immunoprecipitates were prepared from cells stably expressing FLAG-WDR24 and transiently expressing the indicated HA-tagged Sestrin2 constructs. The immunoprecipitates were treated with the indicated concentrations of leucine and analyzed by immunoblotting for the indicated proteins. (E) Multiple sequence alignment of Sestrin2 homologs from various organisms. The positions of residues contacting leucine are indicated with orange dots. Positions are colored white to blue according to increasing sequence identity.

The NTD and CTD of Sestrin2 appear to be structurally similar and superpose well, with a root mean square deviation (RMSD) of ~3.0 Å over 55 aligned Cα positions, despite a low sequence identity of 10.9% (Fig. 1B). Furthermore, the two domains make extensive contacts with each other, primarily through the two core hydrophobic helices N9 and C7, burying 1872 Å2 of surface area (Fig. 1A).

A small region in the N terminus of Sestrin2 has weak sequence similarity to the bacterial alkylhydroperoxidase AhpD (24). Analysis of our structure with the National Center for Biotechnology Information’s Vector Alignment Search Tool [VAST (25)] showed that Sestrin2 shares a common fold with the carboxymucolactone decarboxylase (CMD) protein family, consisting of bacterial γ-CMD as well as AhpD (Pfam database identification number: PF02627). Despite low sequence similarity, Sestrin2 strongly resembles an AhpD homodimer, with each half of Sestrin2 matching a single AhpD molecule (Fig. 1C and fig. S2A). The NTD and CTD both superpose well with Ralstonia eutropha AhpD, with RMSDs of ~2.0 Å over 129 and 101 Cα positions, respectively. Thus, Sestrin2 structurally resembles an intramolecular homodimer of two CMD-like domains, despite extensive divergence in the primary sequence.

To test the importance of the intramolecular contacts between the two domains of Sestrin2, we expressed the FLAG-tagged N- and C-terminal halves either alone or together as separate polypeptides and performed coimmunoprecipitation analysis. Although neither domain alone bound GATOR2, the separated halves, when expressed together, bound strongly both to each other and to GATOR2 (Fig. 1D). Similarly, although neither half of Sestrin2 alone bound to leucine, the coexpressed halves did bind leucine (Fig. 1E). Therefore, the NTD and CTD of Sestrin2 interact stably with each other and are both required for the interactions with GATOR2, as well as with leucine.

In addition to its role as a leucine sensor and GATOR2 inhibitor, Sestrin2 has been reported to have peroxiredoxin reductase activity, based in large part on its weak sequence similarity to bacterial AhpD, which does have this activity (24). However, other groups have failed to reproduce this finding (26). The active site of AhpD contains two cysteines, both of which are required for its catalytic activity (26, 27). Superposing our Sestrin2 structure with AhpD confirms previous reports (24, 26) that only one of these active site residues is present in the N-terminal half of Sestrin2 (Cys125), whereas both are absent from the C-terminal half (fig. S2B). This suggests that Sestrin2 either does not reduce peroxiredoxins or does so through an entirely different mechanism than does AhpD.

Sestrin2 is also reported to inhibit the mTORC1 pathway by directly acting as a guanine nucleotide dissociation inhibitor (GDI) for RagA and RagB through a motif consisting of Arg419, Lys422, and Lys426 (28). However, in our structure, two of these three residues are buried (Lys422 and Lys426; fig. S2C), and Sestrin2 shows no structural similarity to known GDI proteins.

Recognition of leucine by Sestrin2

Sestrin2 binds leucine through a single pocket formed at the intersection of helices C2, C3, and C7 in the CTD. Charged residues Glu451 and Arg390 form two sides of the pocket and anchor leucine in place through salt bridges with the free amine and carboxyl groups, respectively (Fig. 2, A and B). In addition, helix L1 in the linker domain packs against the side of the pocket via residue Leu261 (fig. S3A). This is consistent with mutagenesis studies that identified Glu451 and Leu261 as critical for leucine binding (20). Meanwhile, the isopropyl side chain of the bound leucine points down toward the hydrophobic base of the pocket, forming extensive van der Waals contacts with residues Leu389, Trp444, and Phe447 (Fig. 2, A and B). The depth and overall hydrophobicity of this pocket floor make it well suited to accommodate leucine (Fig. 2B).

To test the importance of these protein-ligand interactions, we generated a series of Sestrin2 leucine-pocket mutants. Disrupting the electrostatic coordination of the free amine by switching a single oxygen atom in Glu451 to nitrogen [Glu451→Gln451(E451Q)] resulted in a complete loss of leucine binding, as did eliminating the interaction between Arg390 and the free carboxyl (R390A; Fig. 2C). In addition, although leucine readily triggered dissociation of the wild-type Sestrin2-GATOR2 complex, both the E451Q and R390A mutants remained constitutively bound to GATOR2, even in the presence of leucine (Fig. 2D). The hydrophobic integrity of the pocket floor is also critical: The insertion of a single charged residue (W444E) was sufficient to abolish any detectable interaction with leucine (Fig. 2, C and D). Consistent with an essential role for these residues in leucine sensing, a multiple sequence alignment of Sestrin homologs showed that both Glu451 and Arg390 are strictly conserved in Sestrin proteins across phylogenetically diverse organisms, as is the hydrophobic nature of the pocket floor (Fig. 2E).

These results provide a molecular explanation for how the Sestrin2-mTORC1 pathway specifically detects leucine and not other amino acids. Although Glu451 and Arg390 probably interact with any amino acid containing free amine and carboxyl groups, the hydrophobic base of the pocket excludes all charged and polar amino acids. Furthermore, large hydrophobic residues such as phenylalanine will clash with Trp444 in the pocket floor, whereas small aliphatic amino acids such as alanine or valine will fail to make favorable van der Waals contacts. Thus, only leucine and the structurally similar amino acids isoleucine and methionine interact appreciably. This is consistent with the finding that only leucine and, to a much lesser extent, isoleucine and methionine disrupt the Sestrin2-GATOR2 interaction in vitro (20).

The corresponding region in the NTD of Sestrin2 is filled by protein side chains and cannot accommodate leucine (fig. S3B). However, the positions of key residues including Trp444 and Glu451 are conserved in the NTD pocket (Trp189 and Glu193), and a leucine side chain contributed by Leu107 occupies the same position as the bound leucine in the CTD (fig S3B).

A lid-latch mechanism is required for leucine binding by Sestrin2

In addition to contacting the charged sides and hydrophobic base of the pocket, a “lid” formed by a loop connecting helices C2 and C3 encloses the top of the leucine, so that it is completely buried within the structure (Fig. 2A). Three highly conserved threonine residues (Thr374, Thr377, and Thr386) are positioned directly above the leucine and help coordinate the free amine and carboxyl groups, locking the ligand in place (Figs. 2E and 3A). The side chain hydroxyl groups of Thr374 and Thr386 make hydrogen bond contacts with the carboxyl group of leucine, whereas the free amine donates a hydrogen bond to the backbone carbonyl of Thr377 (Figs. 2A and 3A).

Fig. 3 A lid-latch mechanism is required for leucine binding by Sestrin2.

(A) Top-down view of the leucine-bound pocket, focusing on the lid residues Thr374, Thr377, and Thr386, which form hydrogen bonds with the amine and carboxyl groups of leucine (indicated by black dashed lines). Leucine is represented as a stick model (orange). Helix numbers are labeled as in Fig. 1A. (B) Orthogonal view of the leucine-binding pocket, focusing on the latch formed by the predicted hydrogen bond between Tyr375 and His86, which locks the lid in place over the bound leucine (orange). Helix numbers are labeled as in Fig. 1A. (C) Binding of Sestrin2 T374A, T386A, Y375F, and H86A mutants to leucine. Binding assays were performed and immunoprecipitates analyzed as in Fig. 2C. (D) Effect of leucine on the interactions of Sestrin2 T386A, Y375F, or H86A with GATOR2 in vitro. FLAG immunoprecipitates were prepared from cells stably expressing FLAG-WDR24 and transiently expressing the indicated HA-tagged Sestrin2 constructs and were analyzed as in Fig. 2D.

To analyze the importance of these lid interactions for leucine detection by Sestrin2, we generated mutants that we predicted would eliminate the critical contacts between the lid and leucine. Mutation of either Thr374 or Thr386 (T374A or T386A) abolished the interaction with leucine and resulted in a constitutive interaction with GATOR2 (Fig. 3, C and D), demonstrating a crucial role for the lid in leucine binding.

Although our in vitro binding data demonstrated a requirement for both the N- and C-terminal halves of Sestrin2 (Fig. 1D), the structural model shows that the bound leucine only makes direct contacts with residues in the CTD (Fig. 2A). Further structural analysis, however, revealed that the lid residue Tyr375 forms a tight hydrogen bond with the N-terminal residue His86, located in a loop between helices N2 and N3 adjacent to the leucine-binding pocket (Fig. 3B). This interaction appears to form a “latch,” which locks the lid in place over the bound leucine. This interdomain contact appears to be critical for the Sestrin2-leucine interaction, given that specifically eliminating this hydrogen-bonded latch with either a Y375F or H86A mutation abolished leucine binding (Fig. 3, C and D). Both Tyr375 and His86 are also highly conserved in Sestrin proteins across organisms (Fig. 2E and fig. S4). The requirement for His86 to maintain the latch interaction probably explains why the NTD of the protein is also essential for the interaction with leucine.

Altering the leucine sensitivity of the mTORC1 pathway in cells

One prediction for a bona fide cellular leucine sensor is that its affinity for leucine should in part determine the sensitivity of the mTORC1 pathway to leucine. We tested this hypothesis directly by generating a mutant of Sestrin2 with a lower affinity for leucine. We predicted that deepening the hydrophobic base of the pocket by mutating Trp444 to Leu (W444L) would reduce the van der Waals contacts with the bound leucine side chain, thereby weakening but not eliminating the interaction (Fig. 4A). Consistent with this, the Sestrin2 W444L mutant bound one-sixth to one-eighth the amount of leucine that bound to wild-type Sestrin2 (Fig. 4B). Furthermore, the addition of ~10 to 15 times more leucine was required to fully dissociate the W444L mutant from GATOR2, as compared with wild-type Sestrin2 (Fig. 4C).

Fig. 4 Altering the leucine sensitivity of the mTORC1 pathway in cells.

(A) Close-up view of Sestrin2-bound leucine (orange) and the pocket floor residues Phe447 F477 and W444, with the W444L mutant (red) overlaid onto the wild-type protein (teal). Both residues are represented as stick models. Numbers indicate the distance from leucine to residue 444 in Sestrin2 wild type and Sestrin2 W444L. (B) Leucine binding by the Sestrin2 W444L mutant. Binding assays were performed and immunoprecipitates were analyzed as in Fig. 2C. (C) Higher concentrations of leucine are required to dissociate Sestrin2 W444L from GATOR2 compared with Sestrin2 wild type. FLAG immunoprecipitates were prepared from cells stably expressing FLAG-WDR24 and transiently expressing the indicated HA-tagged Sestrin2 constructs and were analyzed as in Fig. 2D. (D) Sensitivity of the mTORC1 pathway to leucine in Sestrin1, -2, and -3 triple-knock out (TKO) cells expressing Sestrin2 wild type or W444L. HEK-293T cells generated with the CRISPR/Cas9 system expressing the indicated proteins via lentiviral transduction. Cells were starved of leucine for 50 min and then restimulated with the indicated amount of leucine for 10 min. Cell lysates from one representative experiment were prepared and analyzed via immunoblotting.

To test the effect of this mutation on mTORC1 signaling in cells, we used a HEK-293T cell line in which Sestrin1, -2, and -3 were knocked out by the CRISPR-Cas9 system (Sestrin TKO cells). The mTORC1 signaling in these cells is fully resistant to leucine deprivation, and reintroduction of wild-type Sestrin2 restores normal signaling, with half-maximal mTORC1 activity occurring on addition of ~20 to 50 μM leucine (Fig. 4D) (20). Expression of Sestrin2 W444L in the Sestrin TKO lines, however, shifted the dose response of mTORC1 to leucine, so that addition of ~250 to 500 μM was required to achieve half-maximal activation of the pathway (Fig. 4D). Thus, the affinity of Sestrin2 for leucine is a major determinant of the sensitivity of the mTORC1 pathway to leucine in human cells.

Although the overall hydrophobicity of the pocket floor is well conserved, the specific residues present at the W444 and F447 positions vary across organisms, and some organisms, including Drosophila, carry the corresponding W444L mutation (Fig. 2E). These differences may alter the shape and depth of the leucine-binding pocket, leading to different affinities or specificities for leucine in different organisms. This may represent an evolutionary adaptation to enable efficient sensing of leucine concentrations that are physiologically relevant in these organisms.

Characterizing the GATOR2 binding site on Sestrin2

To better understand how leucine binding triggers dissociation of Sestrin2 from GATOR2, we sought to structurally characterize the GATOR2 binding interface of Sestrin2. Mutagenesis studies identified residue S190 in the NTD as required for GATOR2 binding (20); however, this site is distal to the leucine-binding pocket. Mapping electrostatic potential onto the solvent-exposed surface of Sestrin2 revealed a region in close proximity to the leucine-binding site that contains the highly conserved charged residues Asp406 and Asp407 (Fig. 5A and fig. S5, A and B). Mutating these residues to alanine (DD406-7AA) completely eliminated GATOR2 binding without affecting leucine binding (Fig. 5B and fig. S5C), suggesting that this region is required for the Sestrin2-GATOR2 interaction. Therefore, Sestrin2 may make multiple contacts with GATOR2 through both the NTD and CTD (Figs. 5C and 1B), consistent with both halves of Sestrin2 being required for GATOR2 binding (Fig. 1D).

Fig. 5 Identification of the GATOR2 binding site and model of leucine sensing by Sestrin2.

(A) View highlighting the conserved surface aspartates Asp406 and Asp407 (D406 and D407) and their positions relative to the bound leucine (orange). (B) Coimmunoprecipitation of GATOR2 with Sestrin2 wild type or Sestrin2 DD406-7AA. FLAG immunoprecipitates were prepared from cells stably expressing FLAG-WDR24 and transiently expressing the indicated HA-tagged Sestrin2 constructs and were analyzed as in Fig. 2D. (C) Surface view of Sestrin2, highlighting the GATOR2 binding sites (red) and their positions relative to the leucine-binding pocket (orange). Domains are colored as in Fig. 1A. (D) Model of leucine sensing by Sestrin2. Binding of leucine (orange) causes closing of the lid-latch, resulting in a conformational change that alters the position of the GATOR2 binding site in the CTD. This leads to dissociation of Sestrin2 from GATOR2, enabling GATOR2 to activate the mTORC1 pathway.


Our results provide a structural model of leucine sensing by the Sestrin2-mTORC1 pathway and shed light on the mechanism through which mTORC1 couples cell growth to leucine availability. The structure shows that Sestrin2 contains an evolutionarily unique leucine-binding pocket consisting of a hydrophobic floor that determines specificity for the side chain of leucine, with adjacent glutamate and arginine residues that coordinate the free amine and carboxyl groups, respectively. An additional lid-latch mechanism helps lock the ligand in place and is required for binding.

Our structure also reveals a highly conserved GATOR2 binding site in close proximity to the leucine-binding pocket, suggesting possible mechanisms for how leucine binding can cause dissociation of Sestrin2 from GATOR2. The key residues for the GATOR2 interaction, Asp406 and Asp407, are located on a loop separated from the lid of the pocket by the 15-residue helix C3 (Fig. 5A). It is therefore conceivable that a conformational change in the lid, corresponding to leucine binding or release, could transmit a conformational change to the GATOR2 binding site via movement of helix C3 (Fig. 5D). Alternatively, a segment of the partially disordered linker domain, which contacts the leucine-binding pocket via Leu261 in helix L1 (fig. S3A), is also in close proximity to the GATOR2 binding site in our structure (Fig. 5C). Therefore, changes in the leucine-binding state of Sestrin2 could potentially alter the position of the linker domain, thereby affecting the availability of the GATOR2 binding site.

Despite these insights, several important questions remain. Fully understanding how leucine binding causes dissociation of Sestrin2 from GATOR2 will probably require ascertaining the structure of either apo-Sestrin2 or the Sestrin2-GATOR2 complex. Furthermore, understanding the exact mechanism by which Sestrin2 inhibits the mTORC1 pathway awaits the elucidation of the molecular function of GATOR2.

Finally, as a critical regulator of cell growth, mTORC1 is misregulated in various human diseases, including cancer and diabetes, as well as inaging (1, 29). By revealing the mechanism through which a natural small molecule regulates this pathway, our results may enable the identification of compounds to pharmacologically target the nutrient-sensing pathway upstream of mTORC1 in vivo.

Supplementary Materials

Materials and Methods

Figs. S1 to S6

Table S1

References (3044)


ACKNOWLEDGMENTS: D.M.S. is a founder, a member of the Scientific Advisory Board, a paid consultant, and a shareholder of Navitor Pharmaceuticals, which is targeting for therapeutic benefit the amino acid sensing pathway upstream of mTORC1. We thank all members of the Sabatini and Schwartz laboratories for helpful insights. We also thank Cell Signaling Technologies for providing many antibodies. This work is based on research conducted at the Northeastern Collaborative Access Team beamlines, which are funded by the National Institute of General Medical Sciences from the National Institutes of Health (P41 GM103403). The Pilatus 6M detector on 24-ID-C beam line is funded by a NIH-ORIP HEI grant (S10 RR029205). This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under contract no. DE-AC02-06CH11357. This work was supported in part by the NIH Predoctoral Training Grant T32GM007287. This work has also been supported by grants from NIH (R01CA103866 and AI47389) and the U.S. Department of Defense (W81XWH-07- 0448) to D.M.S. Fellowship support was provided by NIH to R.L.W. (awards T32 GM007753 and F30 CA189333), L.C. (F31 CA180271), and T.W. (F31 CA189437). T.W. is also supported by an award from the MIT Whitaker Health Sciences Fund. M.E.P. is supported by the Sally Gordon Fellowship of the Damon Runyon Cancer Research Foundation (award DRG-112-12) and a Department of Defense Breast Cancer Research Program Postdoctoral Fellowship(award BC120208). D.M.S. is an investigator of the Howard Hughes Medical Institute. Coordinates and structure factors for the x-ray crystal structures of Sestrin2 have been deposited in the Protein Data Bank (PDB) with accession code 5DJ4.

Stay Connected to Science

Navigate This Article