(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



Folliculin promotes substrate-selective mTORC1 activity by activating RagC to recruit TFE3

['Kristina Li', 'Cardiovascular Institute', 'Perelman School Of Medicine', 'University Of Pennsylvania', 'Pennsylvania', 'United States Of America', 'Department Of Bioengineering', 'Shogo Wada', 'Bridget S. Gosis', 'Chelsea Thorsheim']

Date: 2022-04

Abstract Mechanistic target of rapamycin complex I (mTORC1) is central to cellular metabolic regulation. mTORC1 phosphorylates a myriad of substrates, but how different substrate specificity is conferred on mTORC1 by different conditions remains poorly defined. Here, we show how loss of the mTORC1 regulator folliculin (FLCN) renders mTORC1 specifically incompetent to phosphorylate TFE3, a master regulator of lysosome biogenesis, without affecting phosphorylation of other canonical mTORC1 substrates, such as S6 kinase. FLCN is a GTPase-activating protein (GAP) for RagC, a component of the mTORC1 amino acid (AA) sensing pathway, and we show that active RagC is necessary and sufficient to recruit TFE3 onto the lysosomal surface, allowing subsequent phosphorylation of TFE3 by mTORC1. Active mutants of RagC, but not of RagA, rescue both phosphorylation and lysosomal recruitment of TFE3 in the absence of FLCN. These data thus advance the paradigm that mTORC1 substrate specificity is in part conferred by direct recruitment of substrates to the subcellular compartments where mTORC1 resides and identify potential targets for specific modulation of specific branches of the mTOR pathway.

Citation: Li K, Wada S, Gosis BS, Thorsheim C, Loose P, Arany Z (2022) Folliculin promotes substrate-selective mTORC1 activity by activating RagC to recruit TFE3. PLoS Biol 20(3): e3001594. https://doi.org/10.1371/journal.pbio.3001594 Academic Editor: Anne Simonsen, Institute of Basic Medical Sciences, NORWAY Received: February 18, 2021; Accepted: March 7, 2022; Published: March 31, 2022 Copyright: © 2022 Li et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: All relevant data are within the paper and its Supporting Information files. Funding: SW was supported by a postdoctoral fellowship from the American Diabetes Association, BG was supported by the National Institutes of Health (NIH) (F30) and the Blavatnik Family Foundation, and ZA was supported by the NIH (R01 DK107667). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Abbreviations: AA, amino acid; BHD, Birt–Hogg–Dubé; dFBS, dialyzed FBS; DMEM, Dulbecco’s Modified Eagle Medium; FLCN, folliculin; GAP, GTPase-activating protein; gRNA, guide RNA; LOH, loss of heterozygosity; mTORC1, mechanistic target of rapamycin complex I; RCC, renal cell carcinoma; WT, wild type

Introduction The ability of a cell to sense and respond to the intracellular and extracellular environment is vital for it to maintain metabolic homeostasis. Doing so is also fundamentally necessary for the cell to align its metabolic programming to ongoing cellular physiological needs. A major component of sensory integration occurs at the mechanistic target of rapamycin complex I (mTORC1) kinase complex [1–5]. This multisubunit complex integrates numerous inputs, including signals from growth factors, ambient levels of various amino acids (AAs), the cellular energy state, and hypoxia and DNA damage. In turn, it regulates multiple metabolic programs, for example, promoting anabolic processes such as lipid and protein synthesis, while inhibiting catabolic processes such as autophagy and lysosome biogenesis [1–5]. The mTORC1 complex, nucleated around the adaptor protein Raptor, is recruited to the lysosome membrane upon AA sufficiency and then activated by Rheb in response to growth factors, achieved by relieving the repression of Rheb by the TSC complex [1–5]. AA sensing by mTORC1 is complex, including sensing of leucine by Sestrin and sensing of arginine by SLC38A9. In response to these integrated inputs, mTOR phosphorylates a myriad of targets, including p70S6K and 4EBP1 to promote protein translation and ribosome biogenesis, ULK1 to suppress autophagy, Lipin1 to promote lipid synthesis, and the TFE3/B transcription factors to suppress lysosome biogenesis [1–5]. The mTORC1 pathway is thus often depicted as monolithic, acting as a single on/off switch that senses dozens of upstream informational inputs and integrates them into the single response of phosphorylating its multiple targets [1–5]. However, such a monochromatic model of central control of cellular homeostasis is highly unlikely to be accurate. We have recently identified a substrate-specific branch of mTORC1 signaling, providing the first example of specific regulation of different branches of mTORC1 signaling [6,7], subsequently also reported by the Zoncu and Ballabio groups [8,9]. In this pathway, the protein folliculin (FLCN) regulates mTORC1-mediated phosphorylation of only TFE3/B, while not affecting phosphorylation of other canonical substrates such as S6K and 4EBP1. Thus, deletion of FLCN completely abrogates phosphorylation of TFE3, releasing it from 14-3-3 binding and cytoplasmic sequestration and allowing its nuclear translocation to drive genes of lysosome and mitochondria biogenesis. In contrast, deletion of FLCN does not disable phosphorylation of canonical substrates like S6K and 4EBP1 [6,7]. Understanding how, mechanistically, FLCN confers this substrate specificity onto the mTORC1 complex is thus of significant interest. FLCN is a GTPase-activating protein (GAP) and thus stimulator of the small G-proteins RagC and D, which are active in their GDP-bound state [10]. RagC and D heterodimerize with RagA or B to incorporate into the mTORC1 complex and positively regulate mTORC1 activity. Structures elucidated by cryoEM reveal FLCN to bind directly to RagC/D [8,11], confirming earlier coprecipitation studies [12], and prior work has indicated that RagC binds to TFE3 [13]. We thus hypothesized here that the mechanism by which FLCN modulates only the TFE3/B arm of mTORC1 signaling is by activating RagC to recruit TFE3 to the mTORC1 complex, i.e., achieving substrate specificity via specific recruitment of substrate to the complex. While the work that we report here was being finalized, the Ballabio group reported overlapping findings with TFEB [9].

Discussion The mechanisms by which mTORC1 integrates upstream signals and transmits them downstream has been extensively and elegantly characterized [1]. However, how such a complex integrator of multiple inputs achieves specificity in its outputs has received little attention. We first demonstrated clearly that one branch of mTORC1 output could be independently regulated from another, i.e., we showed that loss of FLCN, a RagC/D GAP, abrogated mTORC1-mediated phosphorylation of TFE3 while having no impact on canonical phosphorylation of S6K and 4EBP [6,7,19]. The impact in vivo of such selective regulation in different cell types included beiging of adipocytes and chronic activation of monocytes. Lacking from these studies, however, was a clear mechanistic understanding of how FLCN confers substrate specificity on mTORC1. We elucidate here this mechanism of substrate specificity, whereby FLCN activates RagC to its GDP-bound form via its GAP activity; activated RagC then physically recruits TFE3 to lysosome surface, thereby promoting its phosphorylation by mTORC1; phosphorylated TFE3 is then bound to 14-3-3 and sequestered in the cytoplasm, thus suppressing TFE3 target pathway activation in the nucleus. Similar findings were recently reported for the regulation of TFEB [9]. Importantly, these events occur independently of Rheb and RagA-mediated regulation of canonical phosphorylation of S6K and 4EBP. Disruption of GATOR1, a GAP of RagA/B, renders mTORC1 insensitive to AA withdrawal, maintaining TFE3 as well as canonical substrates phosphorylated even in the absence of AAs. Concomitant loss of FLCN selectively blunted TFE3 phosphorylation while phosphorylation status of canonical substrates remains insensitive to AA withdrawal, further supporting the separable branches of mTORC1. Heterozygous loss-of-function germline mutations in FLCN cause Birt–Hogg–Dubé (BHD) syndrome, which is marked by chronic development of lung cysts, abundant benign dermal hamartoma-like tumors, and a high incidence of renal cell carcinoma (RCC) [20]. Both the dermal tumors and RCC are characterized by high canonical mTORC1 activity and yet occur in the context of loss of heterozygosity (LOH), i.e., loss of FLCN-mediated activation of mTORC1. The existence of the substrate-specific mechanism described here helps to explain this seeming paradox: loss of FLCN unleashes TFE3 to the nucleus, but has no direct impact on canonical mTORC1 signaling. Moreover, as we have shown before [7], an indirect positive feedback loop explains how in some cell types canonical mTORC1 activity in fact increases in the absence of FLCN: Nuclear TFE3 strongly induces gene expression of RagD [17], which can drive canonical mTORC1 phosphorylation of S6K even in the absence of FLCN [7]. The mechanistic separation of mTORC1 signaling into FLCN-independent (canonical) and FLCN-dependent (noncanonical) arms thus explains the apparent paradoxical development of tumors with high mTORC1 activity in BHD patients. We note evidence of 2 reciprocal feedback loops between these 2 arms of mTORC1 signaling. On the one hand, inactivation of FLCN can lead to RagD-mediated activation of canonical S6K phosphorylation, as described above. Conversely, we also note that constitutive activation of canonical signaling, achieved via deletion of Tsc2, leads to reciprocal partial suppression of TFE3 phosphorylation (Fig 1). This observation is consistent with a previous study, in which unbiased genetic screens revealed TSC to act upstream of FLCN and TFE3 in the regulation of exit from pluripotency in embryonic stem cells [21]. The mechanism for this second feedback loop remains unclear. TFE3 is member of a small family of bHLH-ZIP-type transcription factor that includes TFEB, TFEC, and MITF [22]. Interestingly, TFE3 translocations and gene duplications (i.e., gain-of-function variants) are a relatively common cause of kidney cancer, associated with high mTORC1 activity, thus mimicking the effects of FLCN deletion in BHD syndrome [23]. TFEB and MITF mutations have also been noted in kidney cancers, albeit more rarely. Genetic deletion of Flcn in the kidney in mice yields severe polycystic disease, but not frank cancer, indicating that additional genetic hits are likely required to develop cancer. Ballabio’s group recently showed that codeletion of Tfeb rescues the polycystic phenotype of kidney-specific Flcn deletion [9]. In the same study, the authors show similar effects of RagC on TFEB as we show here on TFE3. There is thus likely a fair amount of similarity between TFE3 and TFEB pathways. The fact that deletion of either Tfeb or Tfe3 abrogates the effect of Flcn deletion suggests that TFE3 and TFEB may heterodimerize, although such interaction has not been reported to date. Alternatively, TFE3 and TFEB perform different functions in different tissues, as suggested by, for example, the lethality of whole-body deletion of Tfeb, while Tfe3 knockout mice are viable, with little baseline phenotype [24,25]. In summary, we elucidate here the mechanistic basis by which FLCN confers substrate specificity upon the mTORC1 complex: FLCN activates RagC to physically recruit TFE3 to the mTORC1 complex, promoting TFE3 phosphorylation while having little impact on canonical substrates such as S6K. Our work, combined with similar work with TFEB [9], mechanistically exposes the first clear example of parsing of mTORC1 signaling.

Materials and methods Cell culture Mouse C2C12 myoblasts were cultured in Gibco Dulbecco’s Modified Eagle Medium (DMEM) with high glucose and GlutaMAX (Invitrogen 10569010) supplemented with 10% FBS and 1% penicillin streptomycin (Invitrogen 15140122). Cells were incubated in 37°C and 5% CO2. DMEM media was changed every 2 days and split with trypsin (Invitrogen 25200056) when cells reached 95% confluence. Nutrient withdrawal and restimulation experiment For AA withdrawal experiments, cells were washed with sterile PBS and AA-free DMEM with 10% dFBS was placed on cells for specified times. For restimulation experiments, AA free media was replaced with complete media (DMEM with 10% FBS). Antibodies Phospho-TFE3 (Ser320) antibody was a gift from Dr. Rosa Puertollano and previously described [13]. Other antibodies used are as follows: total TFE3 (Cell Signaling Technology, 14779), phospho-p70S6K (Thr389) (Cell Signaling Technology, 9234), HA tag (Cell Signaling Technology, 2367), LAMP2 (Abcam, ab13524), FLCN (Abcam, ab124885), total p70S6K (Cell Signaling Technology, 2708), beta-actin (Cell Signaling Technology, 4970), and 14-3-3 (Cell Signaling Technology, 8312) Gene deletion by CRISPR/Cas-9 system lentiCRISPR version 2 was a gift from Feng Zhang (Department of Biological Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA) (Addgene, plasmid 52961). The guide RNAs (gRNAs) were designed using the Optimized CRISPR Design website (http://crispr.mit.edu) from Zhang Lab. Mouse C2C12 myoblast cells were infected with lentivirus encoding for the Cas-9/gRNA, selected using puromycin, and validated using western blot before being used as populations. The gRNA sequence was as follows (the pam sequence is excluded): mouse nontarget control (5′-ATTGTTCGACCGTC TACGGG-3′), mouse flcn (5′-TCCGTGCAGAAGAGCGTGCG-3′), mouse tsc2 (5′-TTGATGCAATGTATTCGTCA-3′), mouse depdc5 (5′-GACAAGTTTGTAGACCTTTG-3′), and mouse RagC (5′-GGACTTCGGCTACGGCGTGG-3′). Lenti and retro virus production Lenti transfer plasmid, psPAX2 (Addgene, plasmid 12260), and pMD2.G (Addgene, plasmid 12259) (both gifts from Didier Trono, School of Life Sciences, EcolePolytechnique Federale de Lausanne, Lausanne, Switzerland) were cotransfected onto HEK293T cells using Lipofectamine 3000 (Thermo Fisher Scientific). Plasmids were removed after 18 hours on HEK293T cells, and media was replenished. After 48 hours, conditioned media was collected and passed through a low protein binding 0.45-μm syringe filter to remove cell debris. Mouse C2C12 myoblasts were plated on 6-well multiwell plates and infected via spinfection method. Polybrene was added to the virus containing media (final concentration of 8 μg/mL) and added on top of the C2C12 cells. Each 6-well plate was centrifuged for 90 minutes at 1,000 g (spinfection), and media was replaced. Stable cells were selected 24 hours postspinfection with the appropriate antibiotic. Western blot Cell culture samples were lysed with RIPA buffer with a proteinase inhibitor (Complete miniproteinase inhibitor cocktail, Roche) and a phosphatase inhibitor (PhosSTOP, Roche). Samples were sonicated and spun down to remove lipid and insoluble debris. A BCA protein assay kit (Thermo Fisher Scientific) was used to quantify and normalize protein concentrations. The same amount of protein (10 to 20 μg) was loaded on to a 4% to 20% gradient Tris-glycine polyacrylamide gel (Bio-Rad) and electrophoresed (SDS-PAGE). Samples were transferred to PVDF membrane (MilliporeSigma) and blocked with 5% milk for 1 hour and incubated with primary antibody overnight. The following day, membranes were washed with TBS-T and incubated in appropriate HRP-conjugated secondary antibody for 60 minutes. Images were taken using the ImageQuant LAS 4000 (GE Healthcare Life Sciences). Immunoprecipitation Cells were washed with cold PBS and lysed with IP Lysis Buffer (Thermo Fisher Scientific) with proteinase inhibitor (Complete miniproteinase inhibitor cocktail, Roche) and a phosphatase inhibitor (PhosSTOP, Roche). Samples were sonicated and centrifuged. Supernatant was added to Pierce Anti-HA Magnetic Beads (Thermo Fisher Scientific) and incubated at room temperature for 30 minutes with end-over-end mixing. Samples were washed 3 times with TBS-T, reconstituted in 1xSDS RIPA, and boiled at 95°C for 5 minutes. HA beads were removed. Immunocytochemistry Cells were grown on glass coverslips precoated with collagen type I. The following day, cells were fixed with 4% paraformaldehyde (Thermo Fisher Scientific) for 15 minutes. Cells were washed with PBS and incubated in blocking buffer (1XPBS/5% Normal Goat Serum/0.3% Triton X-100) for 60 minutes at room temperature. Primary antibodies were diluted as indicated on respective datasheets in blocking buffer and incubated overnight on cells at 4°C. Cells were washed with PBS and incubated in appropriate Alexa Flour (Thermo Fisher Scientific) secondary antibodies diluted in blocking buffer for 1 hour at room temperature in the dark. After washing with PBS, coverslips were mounted with Prolong Gold Antifade Reagent with DAPI (Cell Signaling Technology). Images were captured using the Zeiss LSM 710 confocal microscope, and image analysis was done using ImageJ.

[END]

[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001594

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.