Volume 149, Issue 2, Pages 410-424 (April 2012) Leucyl-tRNA Synthetase Is an Intracellular Leucine Sensor for the mTORC1-Signaling Pathway  Jung Min Han, Seung Jae Jeong, Min Chul Park, Gyuyoup Kim, Nam Hoon Kwon, Hoi Kyoung Kim, Sang Hoon Ha, Sung Ho Ryu, Sunghoon Kim  Cell  Volume 149, Issue 2, Pages 410-424 (April 2012) DOI: 10.1016/j.cell.2012.02.044 Copyright © 2012 Elsevier Inc. Terms and Conditions

Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 1 LRS Is an mTOR-Associated Protein (A) 293T cells were starved for leucine for 1 hr and restimulated with 0.8 mM leucine for 10 min, cell lysates were immunoprecipitated with anti-mTOR antibody, and coprecipitated LRS and Raptor were determined by immunoblotting. Goat IgG and anti-mTOR antibody plus blocking epitope peptide were used as negative controls. (B) 293T cells were transfected with control plasmid (EV), Myc-LRS, or MRS. Cell lysates were immunoprecipitated with anti-Myc antibody, and the coprecipitated mTOR and Raptor were determined by immunoblotting. (C) Colocalization of LRS with mTOR in HeLa cells. Cells were reacted with anti-LRS, anti-MRS, anti-IRS, and anti-mTOR antibodies and visualized with Alexa 488-conjugated and Alexa 594-conjugated secondary antibodies, respectively. (D) Quantification of the colocalization in (C) was performed by using the colocalization function of ImageJ. The error bars represent mean ± standard deviation (SD). (E–H) Colocalization of LRS with Raptor in HeLa cells. Cells were starved for leucine for 1 hr and restimulated with 0.8 mM leucine for 10 min. Cells were reacted with anti-LRS and anti-Raptor antibodies (E) or anti-IRS and anti-Raptor antibodies (G) and visualized with Alexa 488-conjugated and Alexa 594-conjugated secondary antibodies, respectively. Each labeling (green, red, and blue) as well as the merge images are shown. Colocalization was also visualized by using the ImageJ colocalization finder plugin (white color). Quantification of the colocalization between LRS (p = 0.0005) (F) or IRS (p = 0.61) (H) and Raptor was performed by using the colocalization function of ImageJ. The index of colocalization corresponds to the mean ± SD of the overlap coefficient (R)∗100 obtained for more than 10 cells for each colabeling. See also Figures S1 and S2. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 2 The Effect of LRS on Activation and Lysosomal Localization of mTORC1 (A) 293T cells were transfected with six kinds of LRS siRNA for 48 hr, and amino acid-dependent S6K phosphorylation was determined by immunoblotting. (p < 0.0001). (B) 293T cells were transfected with control, mTOR, LRS, IRS, MRS, or VRS siRNA for 48 hr, and amino acid-dependent S6K phosphorylation was determined by immunoblotting. (C) 293T cells were transfected with control, LRS, IRS, MRS, or VRS siRNA for 48 hr, and leucine-dependent S6K phosphorylation was determined by immunoblotting. (D) 293T cells were transfected with control, LRS, or IRS siRNA for 48 hr, starved for isoleucine for 1 hr, and restimulated with 0.8 mM isoleucine for 10 min, and then isoleucine-dependent S6K phosphorylation was determined by immunoblotting. (E) 293T cells were transfected with control or LRS siRNA for 48 hr, and cells were starved for amino acids for 1 hr and restimulated with amino acids for 5 min. Lysosomal proteins were immunoblotted with anti-mTOR, anti-Raptor, anti-LRS, and anti-LAMP2 antibodies. 10% FBS means normal cell culture condition and was used as control. (F) 293T cells were transfected with control or LRS siRNA for 48 hr, and cells were starved for leucine for 1 hr and restimulated with leucine for 10 min. Cells were fractionated with a lysosome isolation kit (Sigma-Aldrich). Lysosomal proteins were immunoblotted with anti-mTOR, anti-Raptor, anti-LRS, and anti-LAMP2 antibodies. (G) HeLa cells were transfected with control or LRS siRNA for 48 hr. Cells were starved for leucine for 1 hr and restimulated with leucine for 10 min, then were reacted with anti-Raptor and anti-LAMP2 antibodies and visualized with Alexa 488-conjugated and alexa 594-conjugated secondary antibodies, respectively. Colocalization of the two proteins results in a yellow color. Colocalized pixels were also visualized by using the ImageJ colocalization finder plugin (white). Pixels over a fixed threshold where a green and red fluorescence were depicted with a ratio 1/1 are shown in white on the merge image. (H) Quantification of the colocalization between Raptor and LAMP2 proteins was performed by using the colocalization function of ImageJ. The index of colocalization corresponds to the mean ± SD of the overlap coefficient (R)∗100 obtained for more than 10 cells for each colabeling. The ratio between green and red signals is comprised between 0.8 and 1.2. See also Table S1 and Figure S3. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 3 Direct Interaction of LRS with RagD GTPase (A) Purified GST-LRS was incubated with protein extracts from 293T cells transfected with HA-RagA, RagB, RagC, RagD, Rheb1, GβL, Raptor, or mTOR, and the coprecipitation of HA-tagged proteins was determined by immunoblotting with anti-HA antibody. Inputs are the amount of 10% protein extract used. (B) 293T cells were transfected with the indicated cDNAs in expression vectors. Cell lysates were prepared, and cell lysates and HA-tagged immunoprecipitates were analyzed by immunoblotting with anti-Myc or anti-HA antibodies. WCL, whole-cell lysate. (C) After cotransfection of HA-RagD with Myc-LRS, IRS, or MRS, cell lysates were immunoprecipitated with anti-HA antibody, and the coprecipitated Myc-tagged protein was determined by immunoblotting with anti-Myc antibody. (D) 293T cells were transfected with the indicated cDNAs in expression vectors. Cell lysates were prepared, and cell lysates and Myc-tagged immunoprecipitates were analyzed by immunoblotting with anti-FLAG, anti-Myc, or anti-HA antibodies. (E) 293T cells were transfected with control or Myc-RagD/HA-RagB. Cell lysates were immunoprecipitated with anti-Myc antibody, and the Myc-tagged immunoprecipitates were analyzed by immunoblotting with anti-HA, anti-LRS, or anti-Raptor antibodies. (F) 293T cells were transfected with HA-RagB and Myc-RagD for 24 hr and then starved for leucine for 1 hr and restimulated with 0.8 mM leucine for the indicated times. Cell lysates were immunoprecipitated with anti-Myc antibody, and the coprecipitates were analyzed with anti-LRS, anti-Raptor, anti-Myc, or anti-HA antibodies. (G) Each of the functional domains of RagD GTPase was expressed as a GST fusion protein. Purified GST-RagD proteins were incubated with Myc-LRS, and the coprecipitation of Myc-LRS was determined by immunoblotting with anti-Myc antibody. (H) After cotransfection of FLAG-LRS with HA-RagB, and Myc-WT or mutated RagD, cell lysates were immunoprecipitated with anti-Myc antibody, and the coprecipitated LRS and RagB were determined by immunoblotting with anti-FLAG and anti-HA antibodies. (I) 293T cells were transfected with the indicated cDNAs in expression vectors. Cell lysates were prepared, and cell lysates were analyzed by immunoblotting with anti-p-S6K, anti-S6K, anti-HA, anti-Myc, or anti-tubulin antibodies. See also Figures S4 and S5. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 4 LRS Forms a Molecular Complex with RagD and Raptor in an Amino Acid-Dependent Manner (A) Amino acid-stimulated interaction of LRS with RagD and Raptor. 293T cells were starved for amino acids for 1 hr and restimulated with amino acids for 5 min. Cell lysates were immunoprecipitated with anti-Raptor antibody, and the coprecipitated LRS and RagD were determined by immunoblotting with anti-LRS and anti-RagD antibodies. (B) 293T cells were transfected with the indicated cDNAs in expression vectors. Cells were starved for leucine for 1 hr and restimulated with leucine for 10 min. Cell lysates and Myc-tagged immunoprecipitates were analyzed by immunoblotting with anti-FLAG and anti-Myc antibodies. WCL, whole-cell lysates. (C) 293T cells were transfected with the indicated cDNAs in expression vectors. Cells were starved for amino acids for 1 hr and restimulated with amino acids for 5 min. Cell lysates and HA-tagged immunoprecipitates were analyzed by immunoblotting with anti-Myc, anti-FLAG, and anti-HA antibodies. (D) LRS is necessary for the complex formation of RagD with Raptor. 293T cells were transfected with control or LRS siRNAs for 48 hr. Cells were starved for amino acids for 1 hr and restimulated with amino acids for 5 min. Cell lysates were immunoprecipitated with anti-Raptor antibody, and the precipitates were analyzed by immunoblotting with anti-LRS and anti-RagD antibodies. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 5 LRS Functions as a Leucine Receptor for mTORC1 Signaling (A) Primary sequence alignment of an N-terminal region of several species LRSs. The class 1a conserved HIGH motif, which is important to ATP binding, is boxed in gray. Conserved Phe and Tyr are colored in cyan. (B) Leucylations by LRS WT and F50A/Y52A mutant were carried out with 4 μM tRNALeu and 50 nM enzymes. (C) 293T cells were transfected with LRS WT or F50A/Y52A mutant for 24 hr and then starved for leucine for 1 hr and restimulated with leucine for 5 min. Leucine-dependent S6K phosphorylation was determined by immunoblotting. (D) After cotransfection of HA-RagD/Myc-RagB with Myc-WT or mutated LRS, cell lysates were immunoprecipitated with anti-HA antibody, and the coprecipitated LRS was determined by immunoblotting with anti-Myc antibody. (E) 293T cells were transfected with the indicated cDNAs in expression vectors. Cell lysates were immunoprecipitated with anti-HA antibody, and the coprecipitated LRS and Raptor were determined by immunoblotting with anti-Myc antibody. See also Table S2 and Figures S6 and S7. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 6 Interaction of LRS with RagD Depends on the Nucleotide-Binding State of RagD (A and B) Effects of expressing the indicated proteins on the phosphorylation of S6K in response to starvation and stimulation with (A) amino acids or (B) leucine. Cell lysates were prepared from 293T cells starved for 1 hr of (A) amino acids or (B) leucine and then stimulated with amino acids or leucine for 5 min. (C) Purified GST or GST-LRS protein was incubated with HA-RagD-transfected cell lysates in the presence of GDPβS or GTPγS. The coprecipitated RagD was determined by immunoblotting with anti-HA antibody. (D) Purified GST or GST-LRS protein was incubated with Myc-RagD WT, S77L (GDP), or Q121L (GTP) transfected cell lysates. The coprecipitated RagD was determined by immunoblotting with anti-Myc antibody. (E) After cotransfection of FLAG-LRS with Myc-WT or mutated RagD, cell lysates were immunoprecipitated with anti-Myc antibody, and the coprecipitated LRS and RagD were determined by immunoblotting with anti-FLAG and anti-Myc antibodies. (F) HeLa cells were transfected with Myc-RagD WT, Q121L, or S77L for 24 hr. Cells were starved for leucine for 1 hr and restimulated with leucine for 10 min. Cells were reacted with anti-LRS and anti-Myc antibodies and visualized with Alexa 488-conjugated and Alexa 594-conjugated secondary antibodies, respectively. Colocalization of the two proteins results in a yellow color. Colocalized pixels were also visualized by using the ImageJ colocalization finder plugin (white). Pixels over a fixed threshold where a green and red fluorescence were depicted with a ratio 1/1 are shown in white on the merge image. (G) Quantification of the colocalization between LRS and Myc-RagD proteins was performed by using the colocalization function of ImageJ. The index of colocalization corresponds to the mean ± SD of the overlap coefficient (R)∗100 obtained for more than 10 cells for each colabeling. (H) 293T cells were transfected with the indicated cDNAs in expression vectors. Cell lysates were prepared, and cell lysates and Myc-tagged immunoprecipitates were analyzed by immunoblotting with anti-FLAG or anti-Myc antibodies. (I) 293T cells were transfected with the indicated cDNAs in expression vectors. Cell lysates were prepared, and cell lysates and Myc-tagged immunoprecipitates were analyzed by immunoblotting with anti-FLAG, anti-HA, or anti-Myc antibodies. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure 7 LRS Acts as a GAP for RagD (A and B) Myc-RagD WT was transfected into 293T cells. After 24 hr, the cells were labeled with 100 μCi/ml 32P-orthophosphate for 8 hr, starved for amino acids (A) or leucine (B) for 1 hr, and then restimulated with amino acids (A) or leucine (B) for 10 min. Myc-RagD was immunoprecipitated, and the bound nucleotides were eluted and analyzed by TLC. GDP%, GDP/(GDP + GTP) × 100. (C) The indicated amounts of the His-LRS (LRS-C; 759–1176 aa) fragment were incubated with 0.15 μM RagD for 20 min at 37°C. Error bars represent mean ± SD (n = 3). (D) His-LRS-C (0.3 μM) was incubated with RagD for the indicated times. The error bars represent mean ± SD (n = 3). (E) After cotransfection of HA-RagD WT with Myc-LRS F50A/Y52 mutant or WT, cells were labeled with 100 μCi/ml 32P-orthophosphate for 8 hr, starved for leucine for 1 hr, and then restimulated with leucine for 10 min. Myc-RagD was immunoprecipitated, and the bound nucleotides were eluted and analyzed by TLC. (F) 293T cells were transfected with control or LRS siRNAs for 48 hr. Cells were labeled with 100 μCi/ml 32P-orthophosphate for 8 hr, starved for leucine for 1 hr, and then restimulated with leucine for 10 min. Myc-RagD was immunoprecipitated and the bound nucleotides were eluted and analyzed by TLC. (G) Sequence alignment of putative GAP motif of LRS with several species Arf-GAPs. Conserved residues are black. h, hydrophobic; s, Gly or Ala; x, any residue. hs, Homo sapiens; rn, Rattus norvegicus; dm, Drosophila melanogaster; sc, Saccharomyces cerevisiae; ss, Sus scrofa. (H) Effects of LRS WT and mutants on in vitro GTP hydrolysis of RagD. Purified WT, H844A, or R845A LRS-C was incubated with RagD for 20 min at 37°C. The error bars represent mean ± SD (n = 3). (I) 293T cells were transfected with LRS WT or GAP mutants (H844A, R845A) for 24 hr and then starved for leucine for 1 hr and restimulated with leucine for 10 min. Leucine-dependent S6K phosphorylation was determined by immunoblotting. (J) His-LRS full-length or LRS-C (0.3 μM) was incubated with purified RagC or RagD (0.15 μM) for 30 min. ARD1, which is a known Arf-GAP, was used as a control. The error bars represent mean ± SD (n = 3). Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure S1 Subcellular Localization of LRS, Related to Figure 1 (A) Subcellular fractionation of LRS. Each fraction was subjected to immunoblotting with anti-LRS, IRS, MRS, and mTOR antibodies. LaminA and LAMP2 were used as nucleus and endomembrane markers, respectively. Nuc, nucleus; PM, plasma membrane; EM, endomembrane; Cyt, cytosol. (B) Immunofluorescence staining of LRS in HeLa cells. HeLa cells were reacted with anti-LRS, anti-calnexin (ER marker), anti-GM130 (Golgi marker), anti-LAMP2 (lysosome marker), or anti-EEA1 (endosome marker) antibodies and visualized with Alexa 488-conjugated and Alexa 594-conjugated secondary antibodies, respectively. Colocalized pixels were also visualized with white color. (C) Quantification of colocalization between LRS and the indicated proteins was performed by using the colocalization function of ImageJ. The index of colocalization corresponds to the mean ± SD of the overlap coefficient (R)∗100 obtained for more than 10 cells for each colabeling. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure S2 Lysosomal Localization of LRS, Related to Figure 1 (A) Lysosomal localization of LRS. 293T cells were starved for amino acids for 1 hr and restimulated with amino acids for 5 min, then fractionated with lysosome isolation kit. Lysosomal proteins were immunoblotted with anti-mTOR, anti-Raptor, anti-LRS, and anti-LAMP2 antibodies. (B) 293T cells were starved for leucine for 1 hr and restimulated with 0.8 mM leucine for 10 min. Lysosomal proteins were immunoblotted with anti-mTOR, anti-Raptor, anti-LRS, and anti-LAMP2 antibodies. (C) 293T cells were transfected with EGFP control or EGFP-LRS expression vector for 36 hr and then stained with LysoTracker Red DND-99 (Molecular Probes) for 30 min. Cells were starved for leucine for 50 min and restimulated with 0.8 mM leucine for 10 min. Figures of leucine starvation (0 min, 50 min) and leucine restimulation (5 min, 10 min) are shown. Colocalized pixels were also visualized with Nikon imaging software NIS-element AR 64-bit version 3.00 (yellow color). (D) Quantification of colocalization between LRS and LysoTracker Red DND-99 was performed with Nikon imaging software NIS-element AR 64-bit version 3.00. Lysosomal localization (RFU/Cell) corresponds to the mean ± SD of the ROI/cell obtained for more than 5 cells for each colabeling. (E) HeLa cells were starved for amino acids for 1 hr and restimulated with amino acids for 10 min and then were permeabilized with 25 μg/ml digitonin for 10 min on ice to remove cytosolic proteins. Cells were fixed and then reacted with anti-LRS and anti-LAMP2 antibodies and visualized with Alexa 488-conjugated and Alexa 594-conjugated secondary antibodies, respectively. The images show LRS (green) and LAMP2 (red). The images in the red box were enlarged (right). (F) Colocalization of LRS with Raptor in HeLa cells. Cells were reacted with anti-LRS, anti-MRS, anti-IRS, and anti-Raptor antibodies and visualized with Alexa 488-conjugated and Alexa 594-conjugated secondary antibodies, respectively. (G) The quantification of the colocalization in (D) was performed by using the colocalization function of ImageJ. The index of colocalization corresponds to the mean of the overlap coefficient (R)∗100 obtained for more than 10 cells for each colabeling. The ratio between green and red signals is ranged between 0.8 and 1.2. The data represent mean ± SD. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure S3 The Effect of LRS on Cell Size and Autophagy, Related to Figure 2 (A) Cell size distributions of cells transfected with control, LRS, IRS, VRS, or MRS siRNA. (B) Cell size distributions (FSC of G1 cells) from (A) were quantified (n = 3, p < 0.0001). (C) Cell size distributions of cells transfected with control, LRS, IRS, VRS, or MRS siRNA. Cells were starved for amino acids for 1 hr and then restimulated with vehicle (open square) or amino acids (closed square) for 4 hr. Cell size distributions (FSC of G1 cells) were quantified (n = 3, p < 0.0001). (D) Cell size distributions of cells transfected with control or LRS siRNA. After 24 hr transfection, cells were treated with vehicle or rapamycin (20 nM) for 24 hr. Cell size distributions (FSC of G1 cells) were quantified (n = 3, p < 0.0001). The data represent mean ± SD. (E) 293T cells were transfected with the indicated siRNAs for 48 hr and starved for leucine for 2 hr. Cell lysates were prepared, and LC3-I and -II were determined by immunoblotting with anti-LC3 antibody. Autophagy induction is indicated by the ratio of LC3-II/LC3-I. (F) After cotransfection of EGFP-LC3 with the indicated siRNAs, Cells were starved for leucine and serum for 2 hr. Accumulation of EGFP-LC3 in puncta was monitored. (G) Quantitative analysis of EGFP-LC3 puncta from (H) was performed by using ImageJ. At least eight cells were analyzed per sample. LRS siRNA- and mTOR siRNA-transfected cells showed statistically significant increase of LC3 puncta per cell compared to control siRNA-transfected cells (n = 8, p = 0.0005 and p < 0.0001, respectively). The data represent mean ± SD. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure S4 The Effect of RagD on Leucine-Induced S6K Phosphorylation, Related to Figure 3 (A) Each of the C-terminal fragments of LRS was expressed as GST fusion protein. Purified GST-LRS proteins were incubated with HA-RagD-transfected cell lysates, and the coprecipitation of HA-RagD was determined by immunoblotting with anti-HA antibody. (B) After cotransfection of HA-RagD with Myc-WT or mutated LRS, cell lysates were immunoprecipitated with anti-HA antibody, and the coprecipitated LRS was determined by immunoblotting with anti-Myc antibody. (C) HeLa cells were transfected with Myc-LRS WT or N969A/K970A mutant for 24 hr. Cells were starved for leucine for 1 hr and restimulated with leucine for 10 min. Cells were reacted with anti-Myc antibody and visualized with Alexa 488-conjugated secondary antibody. (D) Leucylations by LRS WT and N969A/K970A mutant were carried out by using 4 μM tRNALeu and 50 nM enzymes. The data represent mean ± SD. (E) 293T cells were transfected with LRS WT or N969A/K970A mutant for 24 hr and then starved for leucine for 1 hr and restimulated with leucine for 10 min. Leucine-dependent S6K phosphorylation was determined by immunoblotting. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure S5 LRS Is an Upstream Regulator of the RagB/RagD Heterodimer, Related to Figure 3 (A) RT-PCR results of RagC and RagD in 293T cells. (B) 293T cells were transfected with control, RagC, or RagD siRNA for 48 hr, starved for leucine for 1 hr, and then restimulated with leucine for 10 min. Leucine-dependent S6K phosphorylation was determined by immunoblotting. (C) After cotransfection of Myc-LRS WT with control, RagC, or RagD siRNA into 293T cells, cells were starved for leucine for 1 hr and then restimulated with leucine for 10 min. Leucine-dependent S6K phosphorylation was determined by immunoblotting. (D) 293T cells were transfected with the indicated cDNAs (RagB Q99L/RagD S77L) and siRNAs for 48 hr and leucine-dependent S6K phosphorylation was determined by immunoblotting. (E) 293T cells were transfected with the indicated cDNAs (RagB Q99L/RagC S75L) and siRNAs for 48 hr, and leucine-dependent S6K phosphorylation was determined by immunoblotting. (F) After cotransfection of HA-RagB with Myc-RagC or -RagD, cell lysates were immunoprecipitated with anti-Myc antibody and the coprecipitated RagB and the Regulator complex component, p18, were determined by immunoblotting with anti-HA or Myc antibody. (G) 293T cells were transfected with the indicated cDNAs for 24 hr, and leucine-dependent S6K phosphorylation was determined by immunoblotting. (H) 293T cells were transfected with the indicated cDNAs or siRNAs for 48 hr, and cells were starved for leucine for 1 hr and restimulated with leucine for 10 min. Cells were fractionated with lysosome isolation kit (Sigma-Aldrich). Lysosomal proteins were immunoblotted with anti-mTOR, anti-Raptor, anti-LRS, and anti-LAMP2 antibodies. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure S6 The Effect of LRS Knockdown on Leucine- or Leucine Analog-Stimulated S6K Phosphorylation, Related to Figure 5 (A) The effect of leucine analogs on leucine-stimulated S6K phosphorylation. 293T cells were starved for 1 hr of leucine and preincubated with either 0.8 or 8 mM leucinol or leucinamide. After 5 min, 0.8 mM leucine was added. After 5 min incubation, cells were harvested, and S6K phosphorylation was determined by immunoblotting. (B) HeLa cells were starved for 1 hr of leucine and preincubated with either 0.8 or 8 mM leucinol or leucinamide. After 5 min, 0.8 mM leucine was added. After 5 min incubation, cells were harvested, and S6K phosphorylation was determined by immunoblotting. (C) 293T cells were transfected with control or LRS siRNAs for 48 hr, and leucine- or leucine analog-stimulated S6K phosphorylation was determined by immunoblotting. Concentration of L-leucine and leucinamide was 0.8 mM, and concentration of D-leucine, norleucine, and leucinol was 8 mM. (D) HeLa cells were transfected with control or LRS siRNAs for 48 hr and leucine- or leucine analog-stimulated S6K phosphorylation was determined by immunoblotting. Concentration of L-leucine and leucinamide was 0.8 mM, and concentration of D-leucine, norleucine, and leucinol was 8 mM. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions

Figure S7 LRS Is Involved in mTORC1 Activation in a tRNA Charging-Independent Manner, Related to Figure 5 (A) The effect of tRNA on in vitro LRS-RagD binding. 293T cell lysates were incubated with purified GST or GST-fused LRS in the presence of the combinations of leucine (0.1 mM), ATP (0.1 mM), and tRNALeu (25 μg). The precipitated RagD was determined by immunoblotting with anti-RagD antibody. (B) Primary sequence alignment of several species leucyl-tRNA synthetases. The class 1a conserved KMSKS motif, which is important to tRNA binding, is boxed in black. (C) Leucylation and ATP-PPi exchange activities by LRS K716A/K719A mutant were carried out. (D) Effect of K716A/K719A mutant on RagD binding. 293T cells were transfected with Myc-tagged LRS WT or mutant, and HA-tagged RagD for 24 hr. Cell lysates were immunoprecipitated with anti-HA antibody, and the coprecipitated LRS and RagD were determined by immunoblotting with anti-Myc and anti-HA antibodies. (E) 293T cells were transfected with the indicated cDNAs for 24 hr, and leucine-dependent S6K phosphorylation was determined by immunoblotting. (F) tSH1 temperature-sensitive mutant CHO cells were transfected with Myc-RagD WT for 24 hr at 34°C and then incubated at 34°C or 39.5°C. Cells were starved for leucine for 1 hr and then restimulated with leucine for 10 min. Cell lysates were immunoprecipitated with anti-Myc antibody, and the coprecipitated LRS was determined by immunoblotting with anti-LRS antibody. Cell 2012 149, 410-424DOI: (10.1016/j.cell.2012.02.044) Copyright © 2012 Elsevier Inc. Terms and Conditions