SUPPLEMENTARY INFORMATION

Transcription

1 doi:.38/nature234 Note 1. LINC00961 encodes a lysosomal Type I transmembrane polypeptide In order to determine if the polypeptide encoded by LINC00961 was an integral membrane protein or simply associated with the membrane, cell homogenates were treated with Na 2 CO 3 or NaOH. As shown in Extended Data Fig. 2b, this treatment disrupted the interaction of the membrane associated GM130, while treatment did not affect the localization of the integral membrane protein calnexin (Extended Data Fig. 2b and 2c). Consistent with the transmembrane domain identified by SMART 1, our -tagged LINC00961 encoded polypeptide remained in the membrane fraction upon treatment, indicating that the polypeptide penetrates the membrane via this putative transmembrane domain (Extended Data Fig. 2b and 2c). To further characterize how this polypeptide is integrated with cellular membranes, membrane fractions were treated with proteinase K. This treatment degrades proteins extending into the cytosol, while proteins extending into the luminal space are protected from enzymatic activity. The treatment of the membrane fraction with proteinase K in the absence of Triton X- resulted in the partial digestion of calnexin, while the bands for GM130 and -tagged LINC00961 encoded polypeptide completely disappeared (Extended Data Fig. 2d), indicating that the C terminus of the polypeptide molecule is exposed to the cytosol. Thus, as is the case for Type I transmembrane proteins, the LINC00961 encoded polypeptide harbors a luminal N terminus with a cytoplasmic C terminus. Although the LINC00961 encoded polypeptide localizes to the lysosome it does not appear to have any known lysosomal sorting signals (e.g. di-leucine (DXXLL or [DE]XXXL[LI]) and/or tyrosine motifs (YXXØ) 2, which are normally found at the C terminus of such transmembrane proteins 2. Thus, in order to better characterize the region responsible for lysosomal localization, we generated a series of chimeric proteins using the human CD8 receptor. CD8 is a human T-cell surface marker, a type I transmembrane protein, and is commonly used for chimeric fusions as it is well established as a neutral reporter 3. We generated chimeric fusion proteins with either the transmembrane domain or the C terminus of the LINC00961 encoded polypeptide (Extended Data Fig. 2g). As shown, the transmembrane region or C terminal region of 1

2 RESEARCH SUPPLEMENTARY INFORMATION the CD8 reporter was replaced with the ORF1 transmembrane region or C-terminal region, similar to previous CD8 fusion experiments 3. Overexpression of the of full length CD8 protein (CD8(WT)) in HeLa cells was found to predominantly localize to the plasma membrane (Extended Data Fig. 2h), which was also the case for the CD8 chimeric protein containing the LINC00961 polypeptide transmembrane domain (CD8(N)-ORF1(TM)-CD8(C)) (Extended Data Fig. 2i). The localization of this fusion protein was mutually exclusive with the lysosomal marker LAMP1, demonstrating that this region was not responsible for lysosomal targeting. However, a CD8 chimeric fusion in which the CD8 cytosolic domain was replaced with the C terminal cytoplasmic region of the LINC00961 polypeptide (CD8(N and TM)-ORF1(C)) demonstrated an exquisite localization to intracellular organelles that co-localized with LAMP1 (Extended Data Fig. 2j). Thus, while it is possible that the N terminal region may also contribute to localization of the LINC00961 encoded polypeptide, our data demonstrate that the C terminus is sufficient for lysosomal localization. 2

3 RESEARCH Note 2. The LINC00961 polypeptide does not impact v-atpase localization, assembly or lysosomal functions As HEK 293T do not express LINC00961, we generated cell lines that were stably transduced with empty vector control (Mock), wild type LINC00961 (WT) and LINC00961 with both ATG initiation codons mutated (ΔATG1+2). These cell lines were subsequently utilized to evaluate the affect of the LINC00961 encoded polypeptide on v- ATPase localization and assembly at steady state, and were used to examine lysosomal function upon polypeptide overexpression as measured by lysosomal ph, cathepsin activities (i.e. hydrolase function) and lysosomal morphology. Sucrose gradient fractionation was used to evaluate v-atpase localization and to determine if the LINC00961 polypeptide overexpression could affect this. The sucrose gradient fractionation process enabled us to purify light, medium and heavy membrane fractions, which correspond to a variety of subpopulations of vesicles within the cell. Our data established that the v-atpase complex is enriched in medium membrane fractions (Extended Data Fig. 3a), which corresponds to late endosome/lysosome and endoplasmic reticulum localizations as demonstrated by enrichment of Lamp2 and calnexin in these subcellular fractions (Extended Data Fig. 3a). Importantly, overexpression of the LINC00961 polypeptide had no impact on the localization of any of the v-atpase subunits analyzed. In order for the v-atpase complex to activate its proton pump activity, it is required that the V0 and V1 multi-protein domains associate with one another at the surface of cellular membranes 4. The V0 domain is an integral membrane protein complex that translocates protons, while the V1 domain protein complex transiently associates with the V0 domain and can hydrolyze ATP, and which together are required for proton pump activity 4. In order to determine if at steady state v-atpase assembly was altered upon expression of the LINC00961 polypeptide, we analyzed the levels of the V1 component ATP6V1A in cytoplasmic and membrane fractions, relative to ATP6V0D1, a component or the V0 domain (Extended Data Fig. 3b). In order to measure relative assembly of this complex at steady state, we calculated the ratio of ATP6V1A (V1 marker) to ATP6V0D1 (V0 marker) in membrane pellets, and found to be no difference between Mock, WT or ΔATG1+2 transduced cell lines (Extended Data Fig. 3c). In 3

4 RESEARCH SUPPLEMENTARY INFORMATION contrast, chloroquine is known to promote v-atpase assembly 5, and was used as a positive control for increased assembly in our studies (Extended Data Fig. 3b and 3c). Thus, these data suggest that v-atpase assembly may be unaffected by the LINC00961 polypeptide. As the LINC00961 polypeptide expression appeared to have no affect on v-atpase localization or assembly, we next sought to understand if overexpression could alter lysosomal characteristics and functions associated with v-atpase activity. As illustrated in Extended Data Fig. 3d and 3e, flow cytometry analysis using Lysosensor DND-189 can be utilized to monitor lysosomal acidification. The mean fluorescence intensity (MFI) of Mock or WT overexpressing cells was not altered (Extended Data Fig. 3e), while the v-atpase inhibitor bafilomycin A1 (BafA1) clearly decreased lysosomal acidification (Extended Data Fig. 3d). Similarly, hydrolase function was evaluated by Magic Red TM Cathepsin Assay Kits, which were used to quantify cathepsin L and K activity, and no difference was observed upon LINC00961 overexpression (Extended Data Fig. 3f and 3g). Finally, lysosomal morphology was analyzed by measuring lysosomal diameter upon stable expression of WT LINC While BafA1 clearly showed a significant increase in lysosomal diameter in HEK 293T cells (Extended Data Fig. 3h and 3i), LINC00961 expression had no effect (Extended Data Fig. 3j and 3k). Thus, these data appear to indicate that the LINC00961 polypeptide functions at the lysosome to regulate a process other than v-atpase localization, assembly and lysosomal function at steady state. 4

5 RESEARCH Note 3. Alternative activation of mtor or MAPK is not impacted by LINC00961 While our data established that the LINC00961 encoded polypeptide inhibits mtorc1 activation in response to amino acid signaling, we also examined if this polypeptide could influence activation of mtor by alternative means. Insulin stimulation acts to activate mtorc1 through the canonical PKB/AKT signaling pathway. However, overexpression of LINC00961 (WT) failed to impact mtorc1 activation mediated by insulin stimulation after 24 hours of serum starvation, as evaluated by phosphorylation of, and 4EBP (Extended Data Fig. 5a and 5b). In addition, recent reports reveal Sestrin2 as a cytosolic leucine sensor and CASTOR proteins as a cytosolic arginine sensor, and SLC38A9 as a putative lysosomal arginine sensor for mtorc1 signaling 6-9 (Extended Data Fig. 4a). Therefore, we also examined the effect of leucine-only depletion followed by leucine stimulation, or arginine-only depletion followed by arginine stimulation on mtorc1 activation in the presence of LINC00961 (WT) overexpression. However, we failed to see any affect on mtorc1 signaling in these conditions (Extended Data Fig. 5c-e). Furthermore, overexpression of LINC00961 (WT) showed no significant effects on phosphorylation of AKT at S473, a substrate for mtorc2, in response to insulin stimulation, or phosphorylation of ERK1/2 in response to EGF stimulation (Extended Data Fig. 5f-h). 5

6 RESEARCH SUPPLEMENTARY INFORMATION Note 4. Whole Genome Sequencing Analysis of CRISPR/Cas9 engineered mouse Off-target mutations in Cas9-modified mice have been reported to be low (Lyer et. al., Nature Methods), and for our sgrnas there were few predicted off-target sites (Supplementary Table 2), with the majority of predicted targets having at least 4 mismatched nucleotides (Supplementary Table 2). However, in order to fully address this point, we carried out whole genome sequencing (WGS) on three independent Spar KO mice, which had been backcrossed to the C57BL/6J line for at least 4 generations. No significant alterations were observed in our mice, and none of the alterations observed corresponded with predicted off-targets, indicative of normal and spontaneous variations rather than off-target activity of the Cas9 nickase (Supplementary Table 2). However, there did appear to be an enrichment of SNP nucleotides commonly associated with wild-type genetic background of the LG/J and SM/J strains of mice, rather than the C57BL/6J background of our mice, within the Sfi1 gene (Supplementary Table 2). Although these SNPs are normally found in the wild-type strains, we performed PCR amplification of the region and subsequent Sanger sequencing, and confirmed the presence of two of these observed SNP alleles in the genomes of the three mice subjected to WGS, as well as in five additional wild type and five Spar KO mice (Extended Data Fig. 7f and 7g). Thus, we are confident that the phenotype described in our manuscript is as a direct result of Spar deletion, rather than some other artifact resulting from non-specific CRISPR/Cas9 activity. 6

7 RESEARCH Note 5. The role of rapamycin in muscle regeneration In order to confirm that mtorc1 signaling is required for efficient regeneration of muscle, we treated wild type mice with rapamycin (2 mg/kg/day) to block mtorc1 activation prior to the administration of a single dose of cardiotoxin (CTX) by intramuscular injection in the tibialis anterior (TA) muscle (Extended Data Fig. 8a). Following injury, TA muscles were collected at 4 days post injury to monitor mtorc1 activation by western blot analysis (Extended Data Fig. 8b and 8c), and collected at 14 days post injury to monitor the muscle regeneration by histological analysis (Extended Data Fig. 8d-f). Rapamycin treatment greatly suppressed mtorc1 activation and muscle regeneration in this setting. Importantly, treatment with rapamycin had no impact on total body weight over this time, and therefore does not account for the decrase in TA muscle weight in response to CTX and rapamycin treatment (Extended Data Fig. 8g). Thus, these data establish a clear role for mtorc1 in muscle regeneration. 7

8 RESEARCH SUPPLEMENTARY INFORMATION Note 6. SPAR inactivation promotes muscle regenerative capacity As shown in Fig. 4d, treatment of TA muscle with CTX results in a strong activation of mtorc1 that is enhanced by the absence of Spar. Importantly, the activation of mtorc1 in the TA muscle of both Spar +/+ and Spar -/- mice is completely inhibited by the deprivation of leucine. We carefully evaluated TA muscle weight as a measure of regeneration in this tissue according to the schematic outlined in Extended Data Fig. 8i. The body weight of mice was taken at 7 days post-ctx administration, and while no difference in weight was observed between Spar +/+ or Spar -/- cohorts, leucine deprivation did result in a decrease in total body weight for both Spar +/+ or Spar -/- mice (Extended Data Fig. 8j). As mtor plays a role in growth and metabolism, we evaluated food intake over the 7-day period post-ctx administration, and found no significant difference in food intake relative to body weight for Spar +/+ and Spar -/- mice fed on a control or leucine-free diet (Extended Data Fig. 8k). As might be expected in the case of mice fed a leucine-free diet, both Spar +/+ and Spar -/- mice showed a decrease in control lateral uninjured TA muscle (TA(-)) weight, corresponding to the decrease observed for total body weight (Extended Data Fig. 8l). However, when normalized for total body weight, no difference was observed between control or leucine-fed Spar +/+ or Spar -/- mice (Extended Data Fig. 8m). Importantly, in the CTXinduced injured TA muscle (TA(+)) Spar -/- mice on the control diet showed a significant increase in both absolute weight (Extended Data Fig. 8n) and normalized weight (Extended Data Fig. 8o) at 7 days post-ctx administration relative to Spar +/+ mice, as a result of increased mtorc1 activity. In contrast, both Spar +/+ and Spar -/- mice fed a leucine free diet showed no difference in TA(+) muscle weights, indicative of a complete block in mtorc1 activity, as shown in Fig. 4d. 8

9 RESEARCH References 1 Letunic, I., Doerks, T. & Bork, P. SMART: recent updates, new developments and status in. Nucleic acids research, doi:.93/nar/gku949 (14). 2 Braulke, T. & Bonifacino, J. S. Sorting of lysosomal proteins. Biochim Biophys Acta 1793, , doi:.16/j.bbamcr (09). 3 Ihrke, G., Gray, S. R. & Luzio, J. P. Endolyn is a mucin-like type I membrane protein targeted to lysosomes by its cytoplasmic tail. Biochem J 345 Pt 2, (00). 4 McGuire, C., Cotter, K., Stransky, L. & Forgac, M. Regulation of V-ATPase assembly and function of V-ATPases in tumor cell invasiveness. Biochim Biophys Acta 1857, , doi:.16/j.bbabio (16). 5 Stransky, L. A. & Forgac, M. Amino Acid Availability Modulates Vacuolar H+- ATPase Assembly. J Biol Chem 290, , doi:.74/jbc.m (). 6 Chantranupong, L. et al. The CASTOR Proteins Are Arginine Sensors for the mtorc1 Pathway. Cell 165, 3-164, doi:.16/j.cell (16). 7 Jung, J., Genau, H. M. & Behrends, C. Amino Acid-Dependent mtorc1 Regulation by the Lysosomal Membrane Protein SLC38A9. Mol Cell Biol 35, , doi:.1128/mcb.001- (). 8 Rebsamen, M. et al. SLC38A9 is a component of the lysosomal amino acid sensing machinery that controls mtorc1. Nature 519, , doi:.38/nature147 (). 9 Wolfson, R. L. et al. Sestrin2 is a leucine sensor for the mtorc1 pathway. Science 351, 43-48, doi:.1126/science.aab2674 (16). 9

10 RESEARCH SUPPLEMENTARY INFORMATION Supplementary Figure 1. Original immunoblot images Fig. 1b Fig. 1f Fig. 1g ORF1 ORF1 Fig. 1c (Short exp.) 1 LAMP (Long exp.) Fig. 2a Fig. 2b Fig. 2c 2 1 ATP6V0A ATP6V0A2 ORF1 HA 2 1 ATP6V0A1 2 1 ATP6V0A2 RagC

11 RESEARCH Supplementary Figure 1. Original immunoblot images (continued) Fig. 2d Fig. 2f p- p- p-4ebp p-4ebp 4EBP 4EBP Fig. 3c Fig. 3e Fig. 3g Atp6V1A p- p

12 RESEARCH SUPPLEMENTARY INFORMATION Supplementary Figure 1. Original immunoblot images (continued) Fig. 4a Fig. 4d Extended Data Fig. 1k Extended Data Fig. 2a Spar Lamp1 Fig. 4c 2 1 (Short exp.) (Long exp.) p- ΔATG2 ΔATG1 Calnexin 2 1 p- Extended Data Fig. 2b Extended Data Fig. 2d Extended Data Fig. 2l GM ATP6V0A1 1 Calnexin 1 Calnexin 2 1 ATP6V0A2 1 GM ORF1 (Antibody No.2) 12

13 RESEARCH Supplementary Figure 1. Original immunoblot images (continued) Extended Data Fig. 3a ATP6V0A1 ATP6V0A2 ATP6V0D1 1 Extended Data Fig. 3b ATP6V1A ATP6V0D1 LAMP2 1 Extended Data Fig. 4c p- Extended Data Fig. 4d ATP6V1A PC3 LAMP2 Calnexin 1 2 HeLa EEA1 1 Extended Data Fig. 4e Extended Data Fig. 4g Extended Data Fig. 4i p- p- p- p-4ebp 4EBP 13

14 RESEARCH SUPPLEMENTARY INFORMATION Supplementary Figure 1. Original immunoblot images (continued) Extended Data Fig. 5a Extended Data Fig. 5c Extended Data Fig. 5f p-akt (S473) Akt p- p- p-erk1/2 ERK1/2 p-4ebp 4EBP Extended Data Fig. 6c Extended Data Fig. 6d Extended Data Fig. 6e p- p- 14

15 RESEARCH Supplementary Figure 1. Original immunoblot images (continued) Extended Data Fig. 6f Extended Data Fig. 6i Extended Data Fig. 7j Extended Data Fig. 7k Spar Lamp1 Gapdh 2 1 Atp6v0a1 Atp6v0a Atp6v0a1 Atp6v0a2 Atp6v1a Atp6v1a Atp6v0d1 Lamp1 1 Spar Atp6v0d1 1 Lamp1 Extended Data Fig. 8b Extended Data Fig. 9g Extended Data Fig. 9o Pax7 Myog p- p- GAPDH