The GATOR2 complex maintains lysosomal-autophagic function by inhibiting the protein degradation of MiT/TFEs

SUMMARY

Lysosomes are central to metabolic homeostasis. The microphthalmia bHLH-LZ transcription factors (MiT/TFEs) family members MITF, TFEB and TFE3 promote the transcription of lysosomal and autophagic genes and are often deregulated in cancer. Here we show that the GATOR2 complex, an activator of the metabolic regulator TORC1, maintains lysosomal function by protecting MiT/TFEs from proteasomal degradation independent of TORC1, GATOR1 and the RAG GTPase. We determine that in GATOR2 knockout HeLa cells, members of the MiT/TFEs family are ubiquitylated by a trio of E3 ligases and degraded, resulting in lysosome dysfunction. Additionally, we demonstrate that GATOR2 protects MiT/TFE proteins in pancreatic ductal adenocarcinoma and Xp11 translocation renal cell carcinoma, two cancers that are driven by MiT/TFEs hyperactivation. In summary, we find that the GATOR2 complex has independent roles in TORC1 regulation and MiT/TFE protein protection and thus is central to coordinating cellular metabolism with control of the lysosomal-autophagic system.

Graphical Abstract

eTOC Blurb

MiT/TFEs transcription factors promote autophagy and lysosomal function. Yang et al. show that the GATOR2 complex inhibits the proteasomal degradation MiT/TFEs proteins independent of both canonical and noncanonical TORC1 signaling. Thus, the GATOR2 complex independently regulates nutrient-dependent TORC1 activity and the MiT/TFE-dependent maintenance of the lysosomal-autophagic system.

INTRODUCTION

The deregulation of lysosomal functions contributes to a wide range of diseases including cancer, neurogenerative diseases and lysosomal storage disorders¹. The microphthalmia bHLH-LZ transcription factors (MiT/TFEs) family members MITF, TFEB and TFE3, regulate lysosome function by promoting the transcription of numerous lysosomal and autophagic genes^2–5. The tight regulation of MiT/TFEs activity allows cells to maintain metabolic homeostasis in various cellular environments by stimulating lysosome and autophagosome biogenesis during periods of stress^6–9. Loss of MiT/TFEs results in lysosomal-autophagic defects and dysfunction^7,10. Notably, MiT/TFEs are frequently deregulated in cancer^11,12, including in the metabolic hyperactive KRas-driven pancreatic duct adenocarcinoma (PDAC)¹³ and renal cell carcinoma¹⁴. Thus, MiT/TFEs represent attractive targets for cancer therapy.

The Gap Activity Towards Rags 2 (GATOR2) complex, activates TORC1 by opposing the activity of the TORC1 inhibitor GATOR1^15–18. Computational and structural analysis indicates that multiple components of the GATOR2 complex have structural features characteristic of coatomer proteins and membrane tethering complexes^19–22. Consistent with the structural similarity to membrane coat proteins GATOR2 is localized to lysosomes, the site of TORC1 regulation^18,23. Recent studies have shown that TORC1 targets two sets of proteins for phosphorylation using overlapping components of the TORC1-GATOR signaling pathway. In canonical TORC1 signaling, activated Rag GTPase (RagA-GTP) recruits substrates that contain a TOR signaling (TOS) motif site, including S6K, 4EBP1 and ULK1, to the lysosome where they are phosphorylated by TORC1 in a process dependent on the small GTPase Rheb²⁴. In a second non-canonical pathway, the Rag GTPase recruits MiT/TFE transcription factors to the lysosome where they are phosphorylated by TORC1 in a Rheb-independent manner, resulting in their cytosolic retention and/or degradation^25,26. Unlike canonical TORC1 signaling the recruitment of MiTF is dependent on RagC-GDP and thus requires the RagC GAP Folliculin^26–28.

Here we demonstrate that the GATOR2 complex regulates lysosomal function by preventing the proteasomal degradation of MiT/TFE transcription factors independent of both the GATOR1-TORC1 axis and non-canonical TORC1 signaling. We find that GATOR2 promotes the malignant behavior of pancreatic ductal adenocarcinoma cells by protecting MiT/TFEs from proteolytic degradation. Finally, we show that loss of GATOR2 reduces the levels of an oncogenic TFE3 fusion protein in cells derived from an Xp11 translocation renal cell carcinoma. Thus, the GATOR2 complex regulates two pathways critical to the control of cellular metabolism, TORC1 signaling and the MiT/TFE-dependent maintenance of the lysosomal-autophagic system.

RESULTS

GATOR2 maintains lysosomal function independent of TORC1 activity

Lysosomes act as a hub for TORC1 activation²⁹. The GATOR2 complex, which localizes to the surface of lysosomes and consists of 5 highly conserved proteins (MIOS, WDR24, WDR59, SEH1 and SEC13), activates TORC1 by opposing the function of the TORC1 inhibitor GATOR1 (Figure 1A)^{15,17,18,23,30}. Our previous study in HeLa cells and Drosophila determined that the GATOR2 subunit WDR24 is required for lysosomal cargo degradation²³. We wanted to determine if the GATOR2 complex regulates lysosome function independent of both GATOR1 and TORC1 activity in mammalian cells. As shown in Figure 1C and 1D, knockdowns of the GATOR1 component NPRL3 rescued TORC1 activity in WDR24-KO cells as indicated by increased phosphorylation of the TORC1 target S6K. However, lysosomal pH and cargo protein LC3 and p62 degradation remained defective (Figure 1B, 1C and 1E). To confirm that GATOR2 maintains lysosomal function independent of GATOR1, we generated double knockout cells of WDR24 and the GATOR1 component DEPDC5 (WDR24-KO/DEPDC5-KO). Previous work has shown that deleting any of the three subunits of GATOR1 is sufficient to disable its function^16,31–34. We determined that p62 accumulates in WDR24/DEPDC5 double knockout (dKO) cells (Figure 1O and 1R), but not in DEPDC5-KO cells. Additionally, we used DQ-BSA green, a self-quenched fluorescence probe, to measure lysosomal degradative function³⁵. We found that cells lacking WDR24 have a decreased level of the green fluorescence signal, indicating lysosomal dysfunction, independent of the presence of DEPDC5 (Figure S3C and D). These data demonstrate that the GATOR2 component WDR24 is required to maintain the degradative function of lysosomes independent of GATOR1 in mammalian cells. Thus, we have defined a conserved function of the GATOR2 complex in the maintenance of lysosomal homeostasis that is independent of the GATOR1-TORC1 signaling axis.

Figure 1. The GATOR2 complex promotes lysosome function and the expression of lysosomal-autophagic genes.

(A) Diagram of the GATOR2-GATOR1-TORC1 signaling axis.

(B) WDR24-KO HeLa cells have increased lysosomal pH that is not rescued by siRNA knockdowns of NRPL3. *: P < 0.05. ns: no significance. Error bars represent standard deviation.

(C) Immunoblot demonstrating that WDR24-KO HeLa cells accumulate p62 and LC3. siRNA knockdowns of NRPL3 in WDR24-KO cells increased TORC1 activity but failed to rescue the degradation defects.

(D-E) Quantification of relative P-S6K levels (D), and relative LC3B and p62 levels (E) from (C). Error bars represent standard deviation.

(F) Immunoblot demonstrating WDR24-KO cells have decreased levels of several V-ATPase subunits including ATP6V1B2, ATP6V1D and ATP6V0D1.

(G-I) Quantification of ATP6V1B2 (G), ATP6V1D (H) and ATP6V0D1 (I) from (F). Error bars represent standard deviation.

(J) Immunoblot demonstrating Cathepsin D protein levels are decreased in WDR24-KO cells.

(K) Quantification of precursor and mature Cathepsin D from (J). Error bars represent standard deviation.

(I) Immunoblot demonstrating a decrease in endogenous LAMP1 and LAMP2 protein levels in WDR24-KO cells.

(M-N) Quantification of LAMP1 (M) and LAMP2 (N) from (L). Error bars represent standard deviation.

(O) Immunoblot demonstrating that WDR24/DEPDC5 double knockout cells have decreased levels of lysosomal proteins.

(P-R) Quantification of relative protein levels of V-ATPases subunits (P), lysosomal proteins and Cathepsin D (Q), p62 (R) from (O). Error bars represent standard deviation.

(S) WDR24-KO HeLa cells have decreased ATP6V1B2, ATP6V1D, ATP6V0D1, Cathepsin D, LAMP1 and LAMP2 transcript levels. qRT-PCR values were from three independent experiments. RPS18 and RPL18A were used as negative controls. Error bars represent standard deviation.

GATOR2 promotes the expression of lysosomal and autophagic genes

To better understand the role of the GATOR2 complex in lysosomal function we examine the levels of multiple lysosomal proteins in WDR24-KO cells. The lysosomal V-ATPase complex pumps protons from the cytoplasm into the lysosomal lumen to maintain an acidic lysosomal pH³⁶, which is essential for lysosomal enzyme activity. As shown in Figure 1F–1I, in WDR24-KO cells the protein levels of V-ATPase subunits ATP6V1B2, ATP6V1D, and ATP6V0D1, were significantly lower than in wildtype (WT) cells. Importantly, the higher lysosomal pH and decreased V-ATPase subunit protein level in WDR24-KO cells were rescued by overexpression of an HA-tagged WDR24 protein (Figure S1A–1C). In addition to components of the V-ATPase, protein levels of the lysosomal enzyme Cathepsin D, as well as the lysosomal structural proteins LAMP1 and LAMP2, were significantly lower in WDR24-KO cells relative to wildtype (Figure 1J–1N). Furthermore, knockdowns or knockout of GATOR1 subunits Nprl3 and DEPDC5 respectively, did not rescue low lysosomal protein levels in WDR24-KO cells (Figure 1F–1Q). These data indicate that GATOR2 is required to maintain the levels of numerous lysosomal proteins independent of GATOR1. Notably, qRT-PCR experiments revealed that the deletion of WDR24 resulted in decreased transcript levels for all six genes. (ATP6V1B2, ATP6V1D, ATP6V0D1, Cathepsin D, LAMP1 and LAMP2) (Figure 1S).

A common feature of many genes in the lysosomal autophagic pathway, including the six genes examined above, is the presence of a coordinated lysosomal expression and regulation (CLEAR) element³⁷. CLEAR elements are targets of the MiTF/TFEs family of transcription factors and drive the expression of a network of genes required for lysosomal-autophagic function^2,3. We used qRT-PCR to determine the transcript levels of 50 additional genes that contain the CLEAR motif in WT versus WDR24-KO cells. Notably, the transcript levels of all 50 CLEAR element-containing genes were decreased in WDR24-KO cells relative to wild-type (Figure S2, primers for qRT-PCR please see Table S1). In summary, we have determined that the GATOR2 complex has a TORC1-independent function that promotes the expression of a wide range of CLEAR motif-bearing lysosomal-autophagic genes.

GATOR2 maintains protein levels of MiT/TFEs

Because the transcriptions of genes with CLEAR elements were downregulated in WDR24-KO cells, we hypothesized that MiT/TFEs protein behavior is deregulated in GATOR2 mutant cells, resulting in a downregulation of lysosomal genes and function (Figure 1 and Figure S2). TFEB, MITF and TFE3 are members of the MiT/TFE family of transcription factors that have both unique and overlapping targets in the lysosomal autophagic pathway^9,11. Endogenous TFEB immunostaining indicates that TFEB is transferred into the nucleus in WDR24-KO cells, as a result of a decreased TORC1 activity, however the overall signal intensity of TFEB appeared significantly lower than in WT cells (Figure S3A). In nutrient replete conditions, TFEB is recruited to lysosomes by the Rag GTPase where it is phosphorylated on multiple Serines by TORC1, including S211, which prevents its translocation to the nucleus^38–40. We found that in fed WT HeLa cells, TFEB colocalized with the lysosomal marker LAMP1 (Figure S3B). However, Western blots revealed dramatically reduced levels of both endogenous TFEB and TORC1 phosphorylated TFEB S211 in WDR24-KO cells relative to wildtype HeLa cells (Figure 2A and 2B). As predicted from our results in Figure 1B and 1C, restoring TORC1’s canonical kinase activity by knocking down NPRL3 did not increase endogenous TFEB levels. Moreover, overexpression of TFEB by transfecting a CMV promoter controlled GFP-TFEB plasmid failed to rescue TFEB protein levels in WDR24-KO cells (Figure 2C and 2D). Finally, we determined that the low levels of TFEB, MITF and TFE3 observed in WDR24 single knockouts were not rescued in WDR24-KO/DEPDC5-KO double knockout cells. (Figure 2L and 2O). In summary, the GATOR2 subunit WDR24 is required to maintain TFEB protein levels in HeLa cells independent of GATOR1.

Figure 2. The GATOR2 complex maintains protein levels of MiT/TFEs.

(A) Immunoblot showing decreased levels of endogenous TFEB and P-TFEB S211 in WDR24-KO HeLa cells. Short: 1 min exposure. Long: 5 min exposure.

(B) Quantification of TFEB and P-TFEB S211 from (A). Error bars represent standard deviation.

(C) Overexpression of TFEB did not rescue low TFEB protein levels.

(D) Quantification of GFP-TFEB and GFP-P-TFEB S211 from (C). Error bars represent standard deviation.

(E) Immunoblot showing decreased levels of endogenous TFEB, TFE3 and MITF in WDR24-KO and MIOS-KO, but not in WDR59-KO cells.

(F) Quantification of TFEB, MITF, TFE3 and P-S6K from (E). Error bars represent standard deviation.

(G) Immunoblot demonstrating a downregulation of lysosomal function in WDR24-KO, MIOS-KO cells, but not in WDR59-KO cells.

(H-I) Quantification of ATP6V1B2, ATP6V1D (H) and p62 (I) from (G). Error bars represent standard deviation.

(J) Immunoblot demonstrating knockdowns of Seh1 caused a decrease in TFEB protein levels.

(K) Quantification of P-S6K and TFEB levels from (J). Error bars represent standard deviation.

(L) Immunoblot demonstrating WDR24/DEPDC5 double knockout (dKO) cells have decreased protein levels of MiT/TFEs.

(M) Immunoblot demonstrating WDR24/RAGA dKO cells have decreased protein levels of MiT/TFEs.

(N) Immunoblot demonstrating WDR24/RAGC dKO cells have decreased protein levels of MiT/TFEs.

(O-Q) Quantifications of MiT/TFEs and P-TFEB S211 protein levels from WDR24/DEPDC5 dKO (O), WDR24/RAGA dKO (P) and WDR24/RAGC dKO (Q) cells. Error bars represent standard deviation.

Next, we examined if other members of the GATOR2 complex, including MIOS, Seh1 and WDR59, are required to maintain MiT/TFE protein levels. Notably, we observed a similar pattern of low MiT/TFEs protein levels in MIOS-KO and Seh1-knockdown cells (Figure 2E, 2F, 2J and 2K) as well as lower levels of several V-ATPase subunits and defects in p62 degradation (Figure 2G–2I). However, in WDR59-KO cells, MiT/TFEs protein levels were comparable to WT. Moreover, WDR59-KO cells had normal lysosomal function (Figure 2E–2I). These data are consistent with recent evidence indicating that WDR59 regulates the interaction between GATOR1 and GATOR2 but is not an obligate component of GATOR2 in all cell types⁴¹. Taken together our data indicate that the GATOR2 complex components, MIOS, Seh1, and WDR24 are required for the maintenance of MiT/TFEs protein levels.

The RAG GTPase is an important regulator of MiT/TFEs^{28,38–40,42}. In nutrient replete conditions, RAGA and RAGC, recruit MiT/TFEs to lysosomes where they are phosphorylated by TORC1’s non-canonical kinase activity^28,38–40, resulting in their sequestration in the cytoplasm and/or degradation by the proteasome²⁵. Thus, phosphorylation by TORC1 restricts MiT/TFE activity in nutrient replete conditions. As previously reported, P-TFEB S211 levels are largely decreased in RAGA and RAGC-KO cells, while total protein levels of MiT/TFEs increased relative to wildtype (Figure 2M, 2N, 2P and 2Q). If GATOR2 controls MiT/TFE stability through the Rag GTPase, then double knockouts of GATOR2 and RAGA or RAGC will have increased MiT/TFE levels, similar to those observed in RAGA and RAGC single KO cells. However, we find that WDR24/RAGA and WDR24/RAGC double knockout (dKOs) cells had low levels of MiT/TFE proteins similar to those observed in WDR24 single KO cells (Figure 2M, 2N, 2P and 2Q). From these data, we conclude that the GATOR2 subunit WDR24 regulates MiT/TFEs protein levels independent of the RAG GTPase.

We previously reported that in wdr24, mio and seh1, mutants of Drosophila, the female germline contains enlarged Atg8 positive structures and has diminished autophagic flux, phenotypes associated with lysosomal storage disorders²³. MITF is the only Drosophila ortholog of mammalian MiT/TFE^10,43. We first used germline clonal analysis and found that in RagC and BHD (FLCN in mammals) mutant clones, MITF was imported into the nucleus (Figure S4A and B), indicating that the inhibitory regulation of MITF by the FLCN-RagC axis²⁸ is conserved in Drosophila⁴⁴. However, we find that similar to HeLa cells, MITF protein levels in wdr24 mutant adult ovaries were significantly lower than WT (Figure S4C and D). Immunofluorescence staining results also confirm a significantly decreased MITF signal in the female germline of wdr24 Drosophila mutants (Figure S4E and F). The lower levels of MITF were accompanied by a decrease in Drosophila V-ATPases subunit Vha68 (Figure S4C and D). Interestingly, the levels of MITF protein in wdr24 mutant adult heads, which do not exhibit lysosomal defects, were comparable to WT (Figure S4C and D). Thus, in Drosophila the requirement for GATOR2 components to promote the accumulation of MITF is tissue specific. These data suggest that GATOR2 regulates the level of the MITF protein in both Drosophila and mammals.

GATOR2 inhibits proteasomal degradation of MiT/TFEs

The decreased levels of MiT/TFEs proteins in GATOR2 mutant cells could be the result of either decreased gene transcription or translation, or increased protein degradation. qRT-PCR data did not reveal a dramatic decrease in MITF or TFE3 transcript levels in WDR24-KO cells (Figure 3A), while the transcript level of TFEB decreased by approximately 50%. This result suggests that WDR24 may regulate TFEB’s protein level in part by affecting its transcription. However, a 50% reduction in mRNA level cannot fully explain the dramatic reduction of TFEB protein in WDR24-KO cells (Figure 2A). Also, our observation that the levels of TFEB protein overexpressed from the CMV promoter, is dramatically lower in WDR24-KO cells compared to wildtype cells (Figure 2C), suggests GATOR2 controls TFEB protein levels through pathways other than transcription. Therefore, we next examined if MiT/TFEs undergo protein degradation in GATOR2 mutant cells. As shown in Figure 3B–3D, MiT/TFEs were ubiquitylated in WDR24-KO cells, and the levels of MiT/FTE proteins increased after treatment with the proteasomal inhibitor MG132 (Figure 3E). These data suggest that in GATOR2 mutants, MiT/TFEs are degraded through ubiquitin-mediated proteasomal degradation. In summary, we find that GATOR2 subunits inhibit the degradation of MiT/TFE proteins.

Figure 3. MiT/TFEs are degraded by proteasomes in cells without a functional GATOR2.

(A) qRT-PCR demonstrating that transcript levels of TFEB, MITF and TFE3 are not downregulated in WDR24-KO HeLa cells. Error bars represent standard deviation.

(B-D) Immunoblot showing that MiT/TFEs are ubiquitylated in WDR24-KO cells.

(E) Quantification of GFP-TFEB, GFP-MITF and GFP-TFE3 from (B-D). Error bars represent standard deviation.

(F-H) Immunoblot showing that MiT/TFEs are ubiquitylated by K48 ubiquitin in WDR24-KO cells.

(I) Protein sequence of TFEB, red K represents Lysines being ubiquitylated in WDR24-KO cells.

(J) Immunoblot showing that TFEB4KR is resistant to ubiquitylation in WDR24-KO cells.

(K) Live cell imaging showing that TFEB4KR proteins response to amino acids availability and enter the nucleases under amino acids starvation conditions in WT HeLa cells. Scale bar: 10 μm.

(L) Immunoblot demonstrating the TFEB4KR mutant was resistant to degradation. The overexpression of TFEB4KR rescued lysosomal function.

(M-N) Quantification of GFP-TFEB, ATP6V1B2, ATP6V1D (M), p62 and LC3B (N) from (L). Error bars represent standard deviation.

Proteins can be K48 or K63 ubiquitylated, resulting in proteasomal degradation or functional regulation respectively⁴⁵. In WDR24-KO cells, MiT/TFEs are degraded by the proteasome, and thus we predicted they would be K48 ubiquitylated. To test this hypothesis, we transfected cells with a HA-tagged K48 ubiquitin and performed an immunoprecipitation using an antibody against HA to pull down proteins associated with the K48 ubiquitin in WT and WDR24-KO cells. We found that MiT/TFEs were K48-ubiquitylated in WDR24-KO HeLa cells (Figure 3F–3H). We we confirmed this result by transfecting the cells with a ubiquitin K48R mutant, which blocks the polyubiquitination process. As expected, we observe a reduction in levels of K48-ubiquitylated MiT/TFEs when using the K48R mutant. In summary, MiT/TFEs are K48 ubiquitylated in WDR24-KO cells.

Finally, using UbPred, a software that predicts protein ubiquitylation sites⁴⁶, we identify four Lysines (K219, 347, 430, 431) in TFEB that are ubiquitylated in WDR24-KO cells (Figure 3I). We find that when these Lysines were converted to Arginine using in vitro mutagenesis, the mutant form of TFEB (TFEB4KR: Lysine to Arginine), was resistant to ubiquitylation (Figure 3J) and targeted to the nucleus independent of nutritional status (Figure 3K). Notably, we observed a higher protein level of TFEB4KR relative to wildtype TFEB in WDR24-KO cells, suggesting the mutant was resistant to degradation. Moreover, unlike what is observed with wildtype TFEB, the expression of TFEB4KR rescued WDR24-KO lysosomal defects (Figure 3L–3N). Finally, protein alignment analysis of TFEB, MITF and TFE3 reveals that two lysine sites, K219 and K347, are conserved among MiT/TFEs (Figure S5). In summary, our ubiquitylation analysis of TFEB confirms that the lysosomal defects in WDR24-KO cells are due to the increased proteasomal degradation of MiT/TFEs.

A trio of E3 ubiquitin ligases ubiquitylates MiT/TFEs in GATOR2 mutant cells

To identify E3 ubiquitin ligases that interact with GATOR2 subunits, we search the large-scale protein interaction dataset BioGrid⁴⁷ and chose several E3 ubiquitin ligases including HERC2, SKP2 and HECW2 that were reported to interact with GATOR2 subunits. Specifically, HERC2 has a reported interaction with WDR24 and MIOS, HECW2 with MIOS and Seh1, and SKP2 with WDR59^48–50. We perform siRNA knockdowns of the above E3 ligases in WT and WDR24-KO HeLa cells and then assayed for the recovery of TFEB levels. Notably, knockdowns of HERC2 resulted in a mild recovery of TFEB levels in the WDR24-KO HeLa cells (Figure 4A and B). Previous work has shown that HERC2 has both a functional and physical interaction with a second E3 ligase, UBE3A (E6AP)⁵¹. We find that knockdowns of UBE3A also increased TFEB protein levels in WDR24-KO cells (Figure 4A and B). Finally, we determine that depletions of STUB1, an E3 ligase which was previously shown to act in P-TFEB S211 degradation⁵², also increase TFEB levels in WDR24-KO cells (Figure 4A and B). Interestingly, knockdowns any of the individual E3 ligases resulted in a partial increase of TFEB protein levels, whereas a triple knockdown of HERC2, UBE3A and STUB1 led to a full restoration of TFEB comparable to that observed in WT (Figure 4C and D) suggesting that all three E3 ligases target TFEB for degradation.

Figure 4. A trio of E3 ligases mediates K48-ubiquitylation on MiT/TFEs.

(A) Immunoblot demonstrating single knockdowns of the individual E3 ligases, HERC2, UBE3A and STUB1 slightly increase TFEB level in WDR24-KO cells.

(B) Quantification of TFEB from (A). Error bars represent standard deviation.

(C) Immunoblot demonstrating a triple knockdown of the E3 ligases HERC2, UBE3, and STUB1 fully restored TFEB protein levels and lysosomal function in WDR24-KO cells.

(D-E) Quantification of TFEB, LAMP1, ATP6V1B2 (D) and p62 (E) from (C). Error bars represent standard deviation.

(F-K) Immunoblot demonstrating single knockdowns of the individual E3 ligases exhibit decreased K48-ubiquitylation on TFEB, MITF and TFE3 in WDR24-KO cells. GFP-TFEB, GFP-MITF and GFP-TFE3 were overexpressed in cells and anti-GFP antibody conjugated beads were used to IP individual GFP tagged MiT/TFEs. An antibody against K48-polyubiquitin was used to detect ubiquitylation level on the MiT/TFEs. Representative immunoblot of TFEB (F), MITF (H) and TFE3 (J) and quantification of the K48-ubiquitin levels in TFEB (G), MITF(I) and TFE3 (K). Data in bar graphs were collected from three independent experiments. Bar height indicates mean, error bars represent standard deviation.

(L-N) Immunoblot showing that TFEB (L), TFE3 (M) and MITF (N) are ubiquitylated by UBE3A, STUB1 and HERC2 in vitro.

MiT/TFEs are ubiquitylated by K48 ubiquitin in WDR24-KO cells (Figure 3F–H). To investigate how the three E3 ligases regulate the K48 ubiquitylation in MiT/TFEs, we use an antibody that specifically recognized K48 ubiquitin and found that the degree of K48-ubiquitin on TFEB, MITF and TFE3 in WDR24-KO cells decreased after individual knockdowns of the three E3 ligases, HERC2, UBE3A and STUB1 (Figure 4F–K). To examine if MiT/TFEs are direct targets of the E3 ligases, we performed in vitro ubiquitylation assays. We determine that when added to MiT/TFEs and other components of the ubiquitylation system, UBE3A, HERC2 or STUB1 can individually ubiquitylate MiT/TFEs (Figure 4L–4N). Moreover, the levels of ubiquitylation increased when all three E3 ligases were included in the reaction (Figure 4L–N), suggesting that UBE3A, STUB1 and HERC2 may cooperate to promote MiT/TFEs ubiquitylation. In summary, these data confirm that a trio of E3 ligases target MiT/TFEs for ubiquitin-mediated proteasomal degradation in GATOR2 mutant cells.

Finally, we perform co-immunoprecipitations (Co-IPs) to determine if the GATOR2 subunits WDR24, MIOS and WDR59 associate with the E3 ligases HERC2, UBE3A and STUB1. We find that under both endogenous (Figure 5A and B), WDR24 and MIOS co-immunoprecipitate with all three ligases. These data are consistent with our earlier result that WDR59 is not required for MiT/TFE stability (Figure 2E). Notably, the interactions between the E3 ligases and the GATOR2 subunits required the presence of both MIOS and WDR24, with WDR24 failing to co-IP the three E3 ligases in MIOS-KO cells and vice versa (Figure 5A, B). In addition, we performed an endogenous Co-IP to examine the interactions between UBE3A, HERC2 and STUB1 with MiT/TFEs. As expected, MiT/TFEs associate with the E3 ligases only in WDR24-KO cells (Figure 5C–E), further suggesting that GATOR2 inhibits the E3 ligases from accessing MiT/TFEs. Taken together our data demonstrate that the E3 ligases, HERC2, UBE3A and STUB1, promote the K48-ubiquitylation and destruction of MiT/TFEs in cells lacking a functional GATOR2 complex.

Figure 5. Physical interactions between the E3 ligases, GATOR2 and MiT/TFEs.

(A) Endogenous Co-IP demonstrating that WDR24 associates with HERC2, UBE3A and STUB1 in WT HeLa cells, but not in MIOS-KO cells.

(B) Endogenous Co-IP demonstrating that MIOS associates with HERC2, UBE3A and STUB1 in WT HeLa cells, but not in WDR24-KO cells.

(C-E) Endogenous Co-IP demonstrating that the three E3 ligases HERC2, UBE3A and STUB1 physically interact with TFEB (C), TFE3 (D) and MITF (E) in WDR24-KO cells, but not in WT cells.

The GATOR2-dependent MiT/TFE protection mechanism maintains PDAC cell malignancy

MiT/TFE proteins are central to the metabolic reprogramming that drives pancreatic cancer¹³. In PDAC cells with a KRas hyperactive mutation, MiT/TFEs become refractory to the TORC1 inhibitory regulation that confines them to the cytoplasm¹³. The increased nuclear import of MiT/TFEs in turn drives the expression of lysosomal genes that enhance the catabolic functions essential for pancreatic cancer growth and invasion. We hypothesize that the GATOR2 complex contributes to the aggressive malignancy of PDAC cells by promoting TORC1 activity and maintaining the levels of MiT/TFEs. To test this model, we use two PDAC cell lines, Panc-1 and Panc 03.27, with KRas G12D and G12V hyperactive mutations respectively⁵³. Consistent with phenotypes observed in GATOR2 knockout HeLa cells, siRNA knockdowns of WDR24 in both Panc-1 and Panc 03.27 cells, resulted in increased lysosomal pH (Figure 6A), decreased protein levels of V-ATPases subunits and MiT/TFEs (Figure 6B, 6C and 6D), and defects in lysosomal cargo degradation (Figure 6B, 6E and 6F), Additionally, qRT-PCR experiments revealed that knockdowns of WDR24 resulted in decreased transcript levels of a wide range of lysosomal-autophagic genes in both PDAC cells (Figure S6B–G and S6I–N), but did not impact the levels of MiT/TFEs transcripts (Figure S6A and S6H), similar to what we find in HeLa cells (Figure S2). Notably, restoration of TORC1 activity in WDR24-NPRL3 co-depletion PDAC cells did not rescue low MiT/TFEs protein levels or the lysosomal phenotypes. These data indicate that, as is observed in HeLa cells, in PDAC cells regulation of MiT/TFE protein levels in GATOR2 mutants is decoupled from TORC1 activity. In addition, the GATOR2 subunit WDR59 is not required to maintain protein levels of TFEB and TFE3 in PDAC cells (Figure S7), consistent with HeLa cells (Figure 2E). Importantly, the overexpression of the TFEB4KR, in WDR24-KD PDAC cells rescued the defective lysosome phenotype (Figure 6G–6K), as indicated by the increased levels of V-ATPases subunits (Figure 6H and 6I), p62 degradation (Figure 6J and 6K) and transcription of lysosomal-autophagic genes (Figure S6). In summary, we confirm that PDAC cells share the same GATOR2-dependent MiT/TFEs protection system as HeLa cells which are required for lysosomal function.

Figure 6. The GATOR2 complex contributes to the malignancy of PDAC cells.

(A) siRNA knockdowns of WDR24 caused increased lysosomal pH in Panc-1 and Panc 03.27 cells. ****: P < 0.0001. ns: no significance. Error bars represent standard deviation.

(B) Immunoblot demonstrating knockdowns of WDR24 cause decreased protein levels of MiT/TFEs and lysosomal function in Panc-1 and Panc 03.27 cells.

(C-F) Quantification of P-S6K, TFEB, TFE3, MITF, ATP6V1B2, ATP6V1D from Panc-1 (C) and Panc 03.27 cells (D), p62 from Panc-1 (E) and Panc 03.27 cells (F) from (B). Error bars represent standard deviation.

(G) Immunoblot demonstrating the overexpression of the TFEB4KR mutant restored lysosomal function in Panc-1 and Panc 03.27 cells.

(H-K) Quantification of GFP-TFEB, ATP6V1B2, ATP6V1D in Panc-1 (H), Panc 03.27 cells (I) and p62 in Panc-1 (J), Panc 03.27 (K) from (G). Error bars represent standard deviation.

(L) Cell proliferation assay demonstrating knockdowns of WDR24 decrease cell proliferation of Panc 1-cells and Panc 03.27 cells (M). Error bars represent standard deviation.

(N) Cell invasion assay demonstrating knockdowns of WDR24 caused a decreased ability of invasion of Panc-1 cells and Panc 03.27 cells (O). Error bars represent standard deviation.

(P) Immunoblot demonstrating WDR24 is required for maintaining EMT in PDAC cells.

(Q-R) Quantification of E-Cadherin, ZEB1 and Slug in Panc-1 cells (Q) and Panc 03.27 (R) from (P). Error bars represent standard deviation.

We next examine how GATOR2 affects PDAC cell malignancy. As shown in Figure 6L and 6M, proliferation rates of Panc-1 and Panc 03.27 cells were significantly lower after knockdowns of WDR24. Restoring TORC1 activity by knocking down NPRL3 failed to fully rescue cell proliferation. However, the overexpression of the TFEB4KR mutant, combined with siNPRL3 treatment, rescued the cell proliferation defect caused by the knockdown of WDR24 levels. In cancer cell metastasis and invasion require the robust function of lysosomal enzymes⁵⁴. Similar to what we find for cell proliferation, we determine that the GATOR2 component WDR24 promotes the invasive behavior of PDACs independent of TORC1 regulation. (Figure 6N and 6O). Taken together, our results suggest that the GATOR2 complex contributes to the invasive behavior and increased proliferation of PDAC cells, at least in part, by protecting MiT/TFEs from proteolytic degradation.

The epithelial–mesenchymal transition (EMT) is a process by which cells gain migratory and invasive properties and is a hallmark of metastatic potential in cancer cells⁵⁵. We use antibodies against the EMT marker: E-Cadherin, ZEB1 and Slug⁵⁶ to see if the GATOR2 subunit WDR24 is required in the EMT of PDAC cells. Surprisingly, we find siRNA knockdowns of WDR24 in both Panc-1 and Panc 03.27 cells increased protein levels of E-Cadherin, while decreasing levels of ZEB1 and Slug (Figure 6P–6R), suggesting that those PDAC cells lost EMT hallmarks upon the loss of WDR24 expression. Interestingly, contrary to our results a previous study suggests that TFEB and TFE3 promote the expression of E-Cadherin⁵⁷. Thus, it remains to be determined whether GATOR2 directly affects the expression of E-Cadherin or acts via its downstream TORC1 and MiT/TFEs signals. In summary, we demonstrate that the GATOR2 complex advances the metastatic potential of PDAC cells by promoting TORC1 activity, inhibiting MiT/TFEs degradation and maintaining the EMT process.

GATOR2 maintains MiT/TFEs protein levels in renal cell carcinoma cells

Deregulation of MiT/TFEs contributes to the development of renal cell carcinoma^11,12,14. To complement our studies, we examined two patient-derived kidney cancer cell lines that disrupt the regulation of MiT/TFE activity by blocking different aspects of non-canonical TORC1 signaling. The UOK 257 cell line was derived from a human renal carcinoma of an individual with Birt-Hogg-Dubé (BHD) syndrome resulting from a pathogenic variant in the FLCN gene⁵⁸. Folliculin (FLCN), the product of the FLCN gene, regulates lysosome function by promoting the TORC1-dependent phosphorylation and cytoplasmic sequestration of TFE3 and TFEB through the regulation of the Rag GTPase^26–28. In the absence of FLCN, TFE3 and TFEB are not recruited to the lysosome but instead enters the nucleus and are active. The UOK 124 cell line, which was originally derived from a 21 y.o. female with an aggressive, advanced RCC, was found to be characterized by an Xp11.2 translocation⁵⁹ resulting in a genomic rearrangement of TFE3 which produced an oncogenic TFE3 fusion protein, PRCC-TFE3⁶⁰. Xp11.2 tRCCs represent an aggressive type of kidney cancer resulting from various genomic rearrangements of TFE3. The fusion TFE3 is resistant to TORC1-dependent degradation and sequestration and thus is constitutively active in the nucleus²⁵. By using siRNA against WDR24, we find that protein levels of TFEB, TFE3 and MITF are dramatically decreased in UOK257 cells (Figure 7A and B). UOK257–2 cells used as a control were generated by retrovirally restoring FLCN gene in UOK257 cells. Finally, in UOK124 cells, the levels of the PRCC-TFE3 fusion are dramatically decreased along with the levels of TFEB and MiTF in WDR24 KDs (Figure 7C and D). Taken together these data confirm that GATOR2 maintains MiT/TFEs protein levels independent of the non-canonical activity of TORC1 and the RAG GTPases. Moreover, they indicate that the GATOR2 complex protects MiT/TFE proteins from proteolytic degradation in renal cell carcinomas.

Figure 7. The GATOR2 complex maintains protein levels of MiT/TFEs in renal cell carcinoma cells.

(A) Immunoblot showing that protein levels of MiT/TFEs and GPNMB are decreased in UOK257 and UOK257–2 cells after siRNA knockdowns of WDR24.

(B) Quantifications of the relative protein levels of TFEB, TFE3, MITF and GPNMB from (A). Error bars represent standard deviation.

(C) Immunoblot showing that protein levels of the PRCC-TFE3 (fusion TFE3), TFEB, MITF and GPNMB are decreased in UOK124 (Xp11.2 translocation renal cell carcinoma) after siRNA knockdowns of WDR24.

(D) Quantifications of the relative protein levels of TFEB, TFE3, MITF and GPNMB from (C). Error bars represent standard deviation.

(E) The GATOR2 complex maintains lysosomal-based catabolism by preventing MiT/TFEs from being ubiquitinated and targeted to the proteosome for degradation. In addition, the GATOR2 complex is required in anabolism by activating TORC1 through opposing GATOR1.

The protein GPNMB (Glycoprotein nonmetastatic B) is a direct transcriptional target of TFE3 and is a pathological marker for renal cell carcinomas associated with increased transcription of TFE3, TFEB or MITF^{61 62}. Therefore, we use an antibody against GPNMB to examine its protein level after knockdowns of WDR24. As expected, the loss of WDR24 results in a marked decrease in the protein level of GPNMB in both FLCN-deficient UOK257 and TFE3-fusion UOK124 cells (Figure 7A–D), further suggesting that in cells lacking a function GATOR2, the expression of MiT/TFEs transcriptional targets is impaired.

DISCUSSION

The regulation of the lysosomal autophagic compartment is critical to cellular health and the maintenance of metabolic homeostasis. MiT/TFE transcription factors promote the expression of lysosomal and autophagic genes and are frequently deregulated in cancer^11,12 Here we show that the TORC1 activator GATOR2 contains a conserved TORC1-independent function required for the maintenance of lysosomal homeostasis. In cells with deletions of the GATOR2 subunits WDR24, MIOS or Seh1, members of the MiT/TFEs family of transcription factors, TFEB, TFE3 and MITF, are ubiquitylated and degraded by the proteasome. We find that low levels of MiT/TFEs in GATOR2-KO cells result in a decrease in the transcription of a wide range of lysosomal-autophagic genes, including V-ATPase subunits, lysosomal enzymes (Cathepsins), structure proteins (LAMPs and LAMTORs) and several ATG genes (ATG9), causing a significantly increased lysosomal pH and a systematic failure in digestion of lysosomal cargoes (Figure 1). Additionally, we demonstrate that GATOR2 promotes MiT/TFE protein stability by preventing a trio of E3 ligases, HERC2, UBE3A and STUB1 from targeting the proteins for ubiquitin-mediated degradation. MiT/TFE are frequently deregulated in cancer. Thus, we expanded our study to several tumor cell types defined by increased MiT/TFE activity. We find that GATOR2 is required to maintain MiT/TFE protein levels in both autophagy-addicted PDAC and FLCN-deficient as well as oncogenic fusion TFE3 translocations. In summary, we demonstrate that the GATOR2 complex has independent roles in both TORC1 regulation and the maintenance of MiT/TFE protein stability and thus is central to coordinating cellular metabolism with the control of the lysosomal-autophagic system. Our data support the model that the GATOR2 complex regulates the balance of cell metabolism by fulfilling its dual roles: activation of TORC1-dependent anabolism and protection of MiT/TFEs-dependent catabolism (Figure 7E).

The activity of MiT/TFE transcription factors is tightly coupled to cellular metabolism. In the non-canonical TORC1 signaling pathway, low TORC1 activity promotes both catabolic metabolism and the activation of MiT/TFE transcription factors, thus coupling the activation of catabolic pathways with the expression of genes that promote the expansion of the lysosomal-autophagic compartment. Here we describe a regulatory pathway that controls MiT/TFE activity independent of non-canonical TORC1 signaling. We can think of at least two models to explain the physiologic significance of the GATOR2-MiT/TFE protein stability pathway. First, the GATOR2-MiT/TFEs pathway may prevent the constitutive activation of MiT/TFE transcription factors in cellular contexts that have limited TORC1 activity and thus cannot use non-canonical TORC1 signaling to restrict MiT/TFE activity. Alternatively, the absence of the GATOR2 complex may trigger the targeting of MiT/TFE transcription factors for proteasomal degradation by a quality control pathway. In this second model, GATOR2 is required for the production of functional MiT/TFE proteins. Notably, the three E3 ligases HERC2, UBE3A and STUB1 have been implicated in protein quality control^63–67. Delineating the physiological significance of GATOR2-MiT/TFEs protein stability is an important goal of future investigations.

We find that the GATOR2-MiT/TFEs stability pathway does not require the GATOR2 subunit Wdr59. These data are consistent with work from Drosophila, where wdr59 null mutants do not exhibit the lysosomal defects observed in wdr24, mio and seh1 mutant tissues^18,23. Indeed, recent evidence indicates that in Drosophila and S. pombe, Wdr59 acts as an inhibitor of GATOR2 and is not required for GATOR2 to oppose the activity of GATOR1^41,68. In mammalian cells, WDR59 is required for the stability of other GATOR2 components, including SEH1, MIO and WDR24 making it difficult to assay the role of WDR59 in the regulation of GATOR2 using loss of function strategies^21,41. Intriguingly, however, the addition of proteasomal inhibitors rescues the levels of other GATOR2 components and TORC1 activity in WDR59-KO cells suggesting that WDR59 may not be an absolute requirement for GATOR2 activity in mammals. Considering these observations, we believe there are two possible explanations for why WDR59 is required to inhibit the GATOR1 complex but is not required for GATOR2-dependent MiT/TFEs stability in mammalian cells. The first model posits that there are two “pools” of GATOR2, one pool of GATOR2 inhibits GATOR1 while a second pool of GATOR2 acts to protect MiT/TFE proteins from proteolytic destruction. These two pools might have alternative subcellular localizations or compositions. Alternatively, the instability of other GATOR2 components in WDR59-KO cells, may reduce the overall levels of functional GATOR2 complex below the threshold required to effectively inhibit GATOR1 but not below the levels of GATOR2 required to protect MiT/TFE proteins from proteolytic destruction.

In a number of tumor types the activity of MiT/TFE transcription factors is decoupled from non-canonical TORC1 signaling allowing cancer cells to use catabolic pathways to promote growth¹¹. KRas-driven PDAC is a devastating and aggressive disease with a poor prognosis⁶⁹. Like many cancers, PDAC cells display a high basal level of autophagy, which facilitates their aggressive growth and invasive behavior. Notably, in PDAC cells, MiT/TFEs are not controlled by TORC1 phosphorylation but instead are retained in the nucleus and active independent of TORC1 status^11,13. Here we demonstrate that knockdowns of GATOR2 components in PDAC cells result in the degradation of MiT/TFE proteins and reduction in numerous malignant features including increased cell proliferation, invasion and EMT hallmarks (Figure 6). Additionally, we demonstrate that GATOR2 is required for maintaining protein levels of MiT/TFEs in kidney cancer cells derived from patients with Birt-Hogg-Dubé as well as Xp11.2 translocation renal cell carcinoma, including the PRCC-TFE3 fusion protein, which is stable and resistant to TORC1-dependent degradation (Figure 7). These findings suggest that targeting the dual functions of GATOR2, TORC1 activation and MiT/TFEs protection may provide potential therapeutic targets for the treatment of pancreatic ductal adenocarcinoma and other cancers.

Shortly prior to the original submission of this manuscript, Nardone, Elledge and colleagues described a pathway that regulates the stability of the MiT/TFE family members TFE3 and MITF in response to nutrients²⁵. Unlike the regulation of MiT/TFEs by GATOR2 reported here, the pathway described by Nardone et al., is TORC1 sensitive and requires the activity of the Rag GTPase. Specifically, they identified and characterize degrons in TFE3 and MITF, which target the proteins for proteolytic degradation upon recruitment by the Rags to the lysosomes where they are phosphorylated by TORC1. Notably, the oncogenic PRCC-TFE3 fusion protein loses the degron region as a result of a genomic rearrangement. Thus, the fusion TFE3 protein is resistant to TORC1 and RAG GTPase-dependent degradation. We find that the loss of GATOR2 subunit WDR24 causes a dramatic decrease in the levels of the PRCC-TFE3 fusion protein in UOK124 tRCC cells. We also find that in kidney cancer cells from a BHD syndrome patient, the loss of WDR24 also markedly decreased protein levels of MiT/TFEs. Notably, BHD syndrome is caused by a pathogenic variant in the FLCN gene, which disables the recruitment of MiT/TFEs to the lysosomes by RAGC and thus impedes TORC1’s non-canonical inhibitory activity toward MiT/TFEs²⁸. These data confirm that GATOR2 regulates MiT/TFEs protein levels independent of both TORC1 and the RAG GTPase. Moreover, it indicates that in addition to the TORC1 sensitive degron, there are additional sites present in TFE3 that mediate protein stability that are regulated by GATOR2.

Taken together our data indicates that the GATOR2-MiT/TFE stability pathway is independent of the regulation of MiT/TFE via non-canonical TORC1 signaling. Thus, the GATOR2 complex is a potential target for pharmaceutical intervention to diminish the impact of MiT/TFE proteins in cancer.

Limitations of Study

Here we define a TORC1-independent role for the GATOR2 complex in the regulation of MIT/TFE protein stability. We have yet to answer several important questions. First, while our data indicate that GATOR2 inhibits the ubiquitin ligases HERC2, UBE3 and STUB1, we have not defined a full molecular mechanism for this process. Similarly, although recent evidence suggests that the GATOR2 complex regulates Nprl2 protein in part through ubiquitylation, currently there is not a full understanding of how the GATOR2 complex inhibits the intensely studied GATOR1 complex⁷⁰. Additionally, little is known about how the ubiquitylation-deubiquitylation cycle is regulated for MiT/TFEs proteins. Second, TORC1’s non-canonical activity promotes the ubiquitylation of degrons in TFE3 and MITF which targets the proteins for degradation²⁵. We showed that TFE3 (also the fusion PRCC-TFE3 in tRCC cells) and MITF are degraded in GATOR2 mutant cells independent of non-canonical TORC1 activity. Consistent with these findings, fusion PRCC-TFE3 protein in tRCC cells that lacks the degron is degraded in GATOR2 mutant cells. In the current study, we did not identify the sites in TFE3（and the fusion TFE3) or MITF that are ubiquitylated in GATOR2 mutant cells. Finally, although we showed that loss of GATOR2 function impeded the growth of PDAC cells and decreased the level of pathological markers in kidney cancer cells, we did not formally investigate how GATOR2 affects tumorigenesis of pancreatic cancer and kidney cancer. Also, tissue-specific knockouts of GATOR2 subunits in mouse models will need to be established to examine whether the GATOR2-MiT/TFE stability pathway can be targeted to prevent RCC kidney tumors driven by TFE3 translocations and FLCN mutations in Birt-Hogg-Dubé syndrome.

STAR METHODS

RESOURCE AVAILABILITY

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Mary Lilly (lillym@nih.gov).

Materials Availability

Drosophila stocks, HeLa cell knockout lines, and other reagents generated in this study will be available upon request from the Lead Contact Mary Lilly (lillym@nih.gov).

Data and Code Availability

• Original images from immunoblots, immunofluorescence staining and raw data for statistical analysis are deposited in Mendeley Data: DOI: 10.17632/tpgdh5zdts.1.

• This paper does not report original code.

• Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS

Mammalian cell culture and treatment

Cell lines (WT HeLa, Panc-1 and Panc 03.27) were obtained from American Type Culture Collection. UOK257, UOK257–2 and UOK124 cells were kindly provided by W. Marston Linehan from the Urological Oncology branch of the National Cancer Institute (Bethesda, MD). CRISPR knockout WDR24-KO, MIOS-KO and WDR59-KO HeLa cell lines were made previously^23,75. HeLa cells and UOK257, UOK257–2, UOK124 cells were cultured in DMEM medium plus 10% FBS. Panc-1 and Panc 03.27 were cultured in RPMI medium plus 10% FBS. For amino acid starvation, cells were cultured in the completed DMEM medium until 80% confluency, then were washed twice with 1x phosphate buffered saline (PBS), medium was changed to DMEM without amino acids (glucose was adjusted to 4.5g/l), with 10% dialyzed FBS for 2 hr. Negative mycoplasma contamination status of all cell lines was confirmed using LookOut Mycoplasma PCR Kit (MP0035, Sigma-Aldrich). For MG132 treatment: Briefly, MG132 (Sigma 474790) was dissolved in DMSO (Sigma D2650) and was pre-diluted with media to 10 μg/ml. Media containing MG132 was then used to replace the original media and cells were cultured in the presence of MG132 for 3 hours.

Generation of the DEPDC5, RAGA, RAGC CRISPR-Cas9 knockout cell lines

TrueGuide Synthetic gRNA (DEPDC5, A35533; RAGA, A35533; RAGC, A35533) from Thermo Fisher were used in CRISPR-Cas9 gene editing. On day one, 1 million HeLa cells were seeded into 10-cm dish in the completed DMEM medium without antibiotics. The next day, gRNA was transfected into the cells along with the TrueCut HiFi Cas9 Protein (A50576, Thermo Fisher), by using the Lipofectamine^™ CRISPRMAX^™ Cas9 Transfection Reagent (CMAX00003, Thermo Fisher). Cells were incubated for 2 days. The GeneArt^™ Genomic Cleavage Detection Kit (A24372, Thermo Fisher) was used to estimate gene editing efficiency. Then, limiting dilution cloning was used to isolate the single cell colony. Cells were disassociated by trypsin and serial diluted to the concentration of 0.5 cells per 100 ul medium. The diluted cells were passed into 96-well plates in the amount of 100 ul per well. 1 week after plating, the plates were examined and wells with a single cell colony were marked. After expansion for another 2 weeks, the genome of each cell colony was collected, and their gRNA targeted genome loci were analyzed by PCR. Cells with PCR products of the expected size after editing by Cas9 were collected and their genomes were sent for further sequencing. Finally, the protein level of DEPDC5, RAGA or RAGC was verified by immunoblot using the antibodies against these three proteins respectively.

Drosophila Strains and Genetics

All Drosophila stocks were maintained at 25°C and 60% humidity with a 12 h on/off light cycle on JAZZ-mix Drosophila food (Fisher Scientific). The experimental animals, female adult Drosophila, were collected for 3–5 days and well fed with dry yeast. The following Drosophila strains were used: w; ; ; Mitf-7xV5–3xGFP₁₁, w; ; FRT82, wdr24¹; Mitf-7xV5–3xGFP₁₁. RagC mutant clone: Nos-Flp; FRT42, ragC-13/FRT42, ubi-GFPnls; ; Mitf-7xV5/+. BHD mutant clone: Nos-Flp; ; FRT2A, bhd-8/FRT2A, His2av-mRFP; Mitf-7xV5/+

Generation of the BHD CRISPR Knockout fly:

To generate the BHD CRISPR knock-out fly, the BHD TKO line (BDSC 77094) was crossed to the nos-Cas9 fly to create the knock-out event and further confirmed by PCR. The gRNA sequence is TGGGTGCAGAAGATGGCGCAGGG.

Generation of the RagC CRISPR Knockout fly:

The CRISPR target sites were designed via flyCRISPR Target Finder Tool on https://flycrispr.org/target-finder/. Two Tandem gRNA fragments (see below) were amplified by using pCFD4–3xP3DsRed (Addgene 86864) as a template and the PCR products were subsequently cloned into the BbsI site cut pCFD4–3xP3DsRed via In-Fusion cloning (Takara). The construct was targeted into attp40 to make a transgenic line. To generate the CRISPR knockout fly, we cross gRNA transgenic line to nos-Cas9 fly to create the knock-out event and further confirmed by PCR.

RagCD-KO-Fw primer containing gRNA sequence:

GATATCCGGGTGAACTTCGTTCCAGATCTGGGACTTCCCGTTTTAGAGCTAGAAATAGCAAG

RagCD-KO-Re primer containing gRNA sequence:

GCTATTTCTAGCTCTAAAACTAACGAGGCTGGCCTCGATCGACGTTAAATTGAAAATAGGTC

Generation of the Mitf-V5 CRISPR knock-in fly:

gRNA construct:

CRISPR gRNA was made by self-annealing of the two oligos and sub-cloned into BbsI site of pU6-BbsI-chiRNA⁷⁶ (Addgene 45946). GMR-3xP3-DtagRed cassette excision: Eye specific marker was removed by crossing the tub-PiggyBac, 3xp3ECFP flies. The marker excised flies were diagnostic by PCR and Sanger sequencing.

Mitf-7xV5–3xGFP₁₁ gRNA construct:

CRISPR targeting site: TACGGTGCCAAGTGACCTATGGG.

Sense oligo: CTTCGTACGGTGCCAAGTGACCTAT.

Antisense oligo: AAACATAGGTCACTTGGCACCGTAC.

Mitf 5’ Homologous (HR) fragment cloning:

Two DNA fragments were PCR amplified from y[1], sc[1], v[1]; ;nos-cas9, attp2 fly genomic DNA and one from synthesized fragment containing 7XV5–3xGFP11 tag. Both fragments were cloned into the EcoRV/BglII site of pUC57–3XGFP11-piggy-5XGMR-3xp3-DtagRed via In-Fusion cloning (Takara) to obtain 5’ HR-7xV5 3XGFP11-piggy-GMR-3xp3-TagRed construct.

Primer set for genomic PCR:

Mitf-5HF-1F: CGAATGCATCTAGATATCGTAGTTCGAGACATTCGTCC

Mitf-5HF-1R: GTGACCTATGcGAAGATGAGAG

Mitf-5HF2F: TCTTCGCATAGGTCACTTGG

Mitf-5HF2R: TTTTCCCGAGCCGATATCTAACTCAATATCCAATGTGTCAGATG

Synthesized fragment: (7V5 + 3xGFP11)

TATTGAGTTAGATATCGGCTCGGGAAAACCAATACCGAATCCGCTCTTGGGCTTGGATTCGACAGGTAGTGGCAAGCCGATACCTAACCCGTTGCTCGGCTTGGATAGTACGGGTAGCGGTAAACCAATACCCAACCCACTCCTGGGATTGGACTCGACAGGCAGTGGAAAGCCCATCCCTAACCCCTTGCTGGGACTCGACAGCACCGGTTCCGGCAAACCGATACCCAATCCCCTGCTCGGCCTGGATAGCACCGGCAGCGGCAAGCCAATCCCCAACCCTCTGCTGGGCCTCGACTCGACAGGATCGGGCAAGCCCATACCAAATCCCTTGCTGGGCCTCGACAGCACTGGGGGCAGTGGCGGCAGAGATCACATGGTGCTCCATGAGTACGTTAATGCGGCCGGCATTACTGGAGGCTCTGGCGGTCGAGATCACATGGTCCTTCACGAGTACGTTAACGCTGCCGGAATCACGGGCGGTAGCGGCGGCCGCGACCACATGGTCCTGCACGAATACGTGAATGCAGCGGGTATCACATGAAGATCTccttaaccct

Mitf 3’ Homologous (HR) fragment cloning:

3’ Homologous fragment was PCR amplified from y[1], sc[1], v[1]; ;nos-cas9, attp2 fly genomic DNA and subsequently subcloned into SnaBI cut 5’ HR-7xV5–3XGFP11-piggy-GMR-3xp3-TagRed construct to obtain 5’ HR-7xV5–3XGFP11-piggy-GMR-3xp3-TagRed-3HR construct.

Mio3HF-F: TTCTAGGGTTAATACGTAATGATAAAGTTGTTAAAATTCGATATAA

mio3HF-R: CCGGGATCCGATTACGTATACATACCCATTACTCGCTG

METHOD DETAILS

Immunofluorescence Microscopy for Mammalian Cells

HeLa cells were seeded onto the 8-well Nunc Lab-Tek II Chamber Slide System (Thermo Fisher, 154534) coated with fibronectin in the density of 10⁴ per well and cultured in the completed DMEM medium overnight. Cells were washed 3 times with 1 x PBS and fixed with 4% paraformaldehyde (PFA) in 1 x PBS at room temperature for 15 mins. Cells were washed 3 times with PBS and permeabilized with 0.2% Trition X-100 in PBS for 10 mins at room temperature. Then cells were blocked in blocking buffer (0.1% BSA, 0.2% Trition X-100 in PBS) for 1 hr at room temperature. The chamber slides were incubated overnight at 4°C in blocking buffer containing a diluted primary antibody. Then they were washed 3 times in 1 x PBS and incubated in blocking buffer containing a secondary antibody conjugated with fluorophores for 1 hr in dark at room temperature, followed by a 10 mins incubation with 300 nM 4’,6-diamidino-2-phenylindole (DAPI). The chamber slides were washed 3 times in PBS and then mounted on slides using the ProLong Diamond Antifade solution (Thermo Fisher, P36970) after removing the chamber well. All confocal images were captured by a Zeiss 880 laser scanning microscopy with a Plan-Apochromat 63X/1.4 oil immersion objective. Each set of immunofluorescence experiments was repeated at least three times.

Mammalian Protein Extraction and Immunoblot

Cells were seeded into 6-well plates. 1.2 million cells from each well were washed twice by PBS. Cells were then covered by 600 μl of the M-PER mammalian protein extraction buffer plus proteinases inhibitor cocktail with or without phosphatases inhibitor cocktail, followed by gently shaking at room temperature for 5 mins. The solutions were collected, and the soluble parts were separated by centrifugation. Target proteins in the soluble part were detected by immunoblot using specific antibodies. To detect phosphorylated protein, Pierce Protein-Free T20 (TBS) Blocking Buffer was used to block the membrane and dilute the antibodies. HRP signals were visualized by using a Clarity Western ECL substrate kit (Biorad) and detected with a Biorad ChemiDoc MP imaging system. The grey scale of each band was quantified by Photoshop CC. Each set of immunoblot experiments was repeated at least three times. Representative examples are shown in each figure.

Mammalian cell siRNA Transfections and Knockdowns

ON-TARGETplus siRNAs SMARTpools against human genes were obtained from Horizon Discovery as followings: WDR24 (L-014822-00-0005), NPRL3 (L-020050-02-0005), HERC2 (L-007180-00-0005), STUB1 (L-007201-00-0005), UBE3A (L-005137-00-0005), non-targeting siRNA control (D-001810-01-05). All siRNAs were transfected by the Xfect RNA Transfection reagent, according to manufacturer’s protocol. 0.3 million cells were seeded into 6-well plates and cultured in completed medium overnight. For the next two days, each day a combination of 5 μl siRNA stock (20 mM) and 9 μl of the transfection reagent was added into the culture medium per well. Cells were assayed on the third day for immunoblotting or imaging.

Mammalian Protein immunoprecipitation

For pull-down epitope tagged proteins, GFP-Trap Magnetic Agarose kit (gtma-100) from Chromotek and HA-Tag Magnetic IP/Co-IP Kit was used to pulldown GFP or HA tagged proteins respectively based on manufacturer’s instruction. Briefly, cell lysate containing GFP tagged proteins and other epitope tagged proteins were incubated with the magnetic beads conjugated with an anti-GFP nanobody or anti-HA magnetic beads for 1 hr at room temperature. Unbound proteins were washed away. Then proteins were eluted by heating the beads in protein sample buffer [60 mM Tris (pH 6.8), 10% glycerol, 100 mM DTT, 0.2% bromophenol blue, 2% SDS] at 90°C for 10 mins, followed by immunoblot.

Endogenous Co-immunoprecipitation

HeLa cells were cultured in a 15 cm cell culture dish. Proteins from HeLa cells were extracted in the M-PER^™ Mammalian Protein Extraction Reagent (78505, Thermo Fisher) with proteinase and phosphatase inhibitor cocktail as mentioned above. Antibodies against endogenous WDR24, MIOS, TFEB, TFE3 or MITF was crosslinked to Dynabeads Protein A by disuccinimidyl suberate (DSS) according to manufacturer’s protocol. Protein extracts were incubated overnight at 4°C with the antibody linked protein A beads. After incubation beads were washed three times with protein extraction buffer, then proteins were eluted by heating the beads in protein sample buffer [60 mM Tris (pH 6.8), 10% glycerol, 100 mM DTT, 0.2% bromophenol blue, 2% SDS] at 70°C for 10 mins, followed by immunoblot.

DNA construct and transfection

pRK5-HA-WDR24 (46335), pRK5-FLAG-MIOS (46326), pRK5-HA-WDR59 (46328), pEGFP-N1-MITF (38132), pEGFP-N1-TFE3 (38120), pEGFP-N1-TFEB (38119), HA-Ubiquitin (18712), pRK5-HA-Ubiquitin-K48 (17605), pRK5-HA-Ubiquitin-K48R (17604) from Addgene.

GFP-UBE3A was constructed by cloning UBE3A cDNA (2604 bp) from the pCMV3-UBE3A-Myc plasmid (Sino Biological, HG11130-CM) into pEGFP-C1 vector by using the in-fusion cloning kit (Takara, 638947). pEGFP-C1 plasmid was cut by SacI. UBE3A cDNA was amplified by using the following primers: CGGACTCAGATCTCGATGGAGAAGCTGCACCAGTG and AGAATTCGAAGCTTGTTACAGCATGCCAAATCCTTTG.

GFP-STUB1 was constructed by cloning STUB1 cDNA (912 bp) from the pCMV3-FLAG-STUB1 plasmid (SinoBiological, HG12496-NF) into pEGFP-C1 vector by using the in-fusion cloning kit (Takara, 638947). pEGFP-C1 plasmid was cut by XhoI. STUB1 cDNA was amplified by using the following primers: GGACTCAGATCTCGAATGCAGCAGCACGAGCAGG and GAAGCTTGAGCTCGATCAGTAGTCCTCCACCCAGCC.

GFP-HERC2 was constructed by cloning HERC2 cDNA (14640 bp) from pcDNA5 FRT/TO SF-HERC2 plasmid (Addgene, 55613) into pEGFP-C1 vector by using the in-fusion cloning kit (Takara, 638947). pEGFP-C1 plasmid was cut by XhoI. HERC2 cDNA was amplified by using the following primers: GGACTCAGATCTCGAATGCCCTCTGAATCTTTCTGTT and GAAGCTTGAGCTCGATTAGTGTCCTGTTAAATAATCTTGT. Herculase II Fusion DNA Polymerases (Agilent, 600677) was used for HERC2 cDNA amplification.

All the plasmid constructs were verified by DNA sequencing. Plasmid transfections into cells were performed by using the Xfect Transfection reagent, according to manufacturer’s protocol. A total of 5 mg plasmid was used in transfection for 0.8 million cells.

In vitro ubiquitinylation assay

In vitro ubiquitinylation assay was performed by using the Ubiquitinylation kit (BML-UW9920–0001) from Enzo based on manufacturer’s instructions. Briefly, purified TFEB, TFE3 and MITF was individually mixed with purified E3 ligases HERC2 (GFP-HERC2 was expressed in HeLa cells and purified by using the GFP-Trap kit), UBE3A and STUB1, together with E1, E2 (UbcH5b for STUB1, UbcH7 for UBE3A, UbcH13 for HERC2), ubiquitin, ATP and reaction buffer provided in the kit. The reaction was processed by a 30 mins incubation at 37°C. An anti-ubiquitin antibody was used to detect ubiquitinylation patterns on substrate proteins by immunoblot. An anti-TFEB, anti-TFE3 and anti-MITF antibody was used to detect substrate proteins by immunoblot.

Quantitative RT–PCR

Total cellular RNA was extracted using TRIzol^™ Reagent (Thermo Fisher) and reverse transcription was performed from 1 mg of total RNA using the SuperScript^™ III Reverse Transcriptase (Thermo Fisher). Quantitative RT–PCR was performed with the PowerUp^™ SYBR^™ Green Master Mix (Thermo Fisher) in a StepOnePlus^™ Real-Time PCR System (Thermo Fisher). PCR reactions were performed in triplicate and the relative amount of cDNA was calculated by the comparative CT method using the GAPDH mRNA as the housekeeping gene to calculate the relative cycle number. Primer sequences are available in the Supplemental information.

Live cell imaging and DQ-BSA green assay

For live cell imaging, cells were seeded onto the 8-well Nunc Lab-Tek II Chamber Slide System (Thermo Fisher, 154534) coated with fibronectin in the density of 10⁴ per well and cultured in the completed DMEM medium overnight. DQ-BSA assay was performed as described³⁵. Briefly, cells were incubated in a medium with DQ-BSA (10 μg/ml), then washed twice with PBS. Cells were cultured in amino acids starvation medium for 1 hr to induce autophagy. The DQ-BSA fluorescence was detected by a Zeiss 880 laser scanning microscopy with a Plan-Apochromat 63X/1.4 oil immersion objective. Each set of immunofluorescence experiments was repeated at least three times.

Cell lysosomal pH measurement

Lysosomal pH measurements were performed as describe²³. In brief, cells were stained with 1 μM LysoSensor Green DND-189 in medium for 20 min at 37°C, 5% CO₂. Subsequently, cells were washed with PBS twice and analyzed by a microplate reader (485/530 nm) in triplicate. Lysosomal pH quantification was performed using LysoSensor Yellow/Blue DND-160. Cells were labeled with 1 μM LysoSensor Yellow/Blue DND-160 for 30 min at 37°C, 5% CO₂ in medium and then washed twice with PBS. To generate a calibration curve, cells were treated for 10 min with 10 μM monensin and 10 μM nigericin in 25 mM MES calibration buffer, pH 3.5–6.0, containing 5 mM NaCl, 115 mM KCl, and 1.2 mM MgSO₄. The fluorescence was measured with a microplate reader (340/440 nm and 380/530 nm) at 37°C. These experiments were performed in triplicate.

Cell proliferation measurement

Cell proliferation and viability were measured using CellTiter-Glo 2.0 Assay (Promega, G9241) based on manufacturer’s instruction. This assay determines the number of viable cells in culture after experimental treatments (siRNA, protein overexpression) by quantifying ATP, which indicates the presence of metabolically active cells. ATP is measured by the fluorescence intensity of oxyluciferin, which is generated by oxidizing luciferin with the consumption of ATP. Briefly, cells were seeded into a 96-well plate at a density of 10,000 cells per well. As a control, wells containing culture medium without cells were prepared to determine background luminescence. After 1 day, siRNAs and plasmids were transfected into experimental wells. After a two-days incubation, CellTiter-Glo 2.0 Reagent equal to the volume of cell culture medium was added into each well. Contents were mixed for 2 minutes on an orbital shaker to induce cell lysis and incubated at room temperature for 10 mins. Luminescence was measured by using a Tecan infinite M1000 Pro plate reader. To generate an ATP standard curve, a serial of 10-fold dilution of ATP solution was prepared from a culture medium containing 1 μM ATP (ATP disodium salt, Sigma A7699). CellTiter-Glo 2.0 Reagent was added, and luminescence was recorded after a 10 mins incubation.

Cell invasion measurement

PDAC cell invasion was measured by using the CytoSelect 96-well cell invasion assay based on manufacturer’s instruction. This assay utilizes a special 96-well plate consisting of two chambers (membrane chamber and feeder tray). The membrane chamber surface was pre-coated with the Matrigel Matrix solution. It serves as a barrier to separate invasive cells from non-invasive cells. PDAC cells with the ability to degrade the matrix protein are capable of passing through the pores of the polycarbonate membrane. Finally, the invaded cells are dissociated from the membrane and their number is determined with CyQuant GR Dye. siRNA and/or plasmid in transfection solution were added to cells (10⁶ cells/ml) in a serum-free medium. 150 μL completed RPMI medium with 10% FBS was added into the wells of the feeder tray. 100 μL cell suspension was then added to the membrane chamber. The plate was incubated for 48 hrs. Prior to the end of incubation, 150 μL Cell detachment solution was added into wells of the 96-well cell harvesting tray. After the incubation, the membrane chamber was separated from the feeder tray, and the medium and solutions from the top side of the membrane chamber were removed. Then the membrane chamber was inserted into the cell harvesting tray containing the cell detachment solution, followed by a 30 mins incubation at 37°C. After the incubation, CyQuant GR Dye solution was added to each well of the cell harvesting tray, followed by a 20 mins incubation at room temperature. Then fluorescence at 480 nm/520 nm was measured by using the Tecan infinite M1000 Pro plate reader. The number of invaded cells is proportional to the fluorescence intensity.

Drosophila immunoblot analysis

Drosophila ovaries and heads were homogenized in RIPA buffer containing Complete Protease Inhibitors and Phosphatase Inhibitors (Thermo Fisher). Antibodies were used at the following concentrations: mouse anti-Actin JLA20 (1:1000) (DSHB), mouse anti-V5 (1:1000) (Thermo Fisher), rabbit Drosophila anti-Vha68 (1:1000) (Genescript). The band intensity was quantified using Image J analysis tool (NIH).

QUANTIFICATION AND STATISTICAL ANALYSIS

Sample Size and Statistical Analysis

Detailed sample size can be found in figure legends. All graphs report the mean ± S.D. and represent data from three independent experiments. Bar height indicates mean. Unless otherwise indicated. Statistical comparisons were made using the Unpaired Student’s t-test with Welch’s correction by GraphPad Prism 8 software.

Supplementary Material

2

Table S1: Primers of the lysosomal-autophagic genes for qRT-PCR, related to Figure 1.

KEY RESOURCES TABLE

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
LAMP1 Mouse monoclonal, H4A3	Developmental Studies Hybridoma Bank	Cat# H4A3, RRID:AB_2296838
Phospho-p70 S6 Kinase (Thr389) (D5U1O) Rabbit mAb	Cell Signaling Technology	Cat# 97596, RRID:AB_2800283
p70 S6 Kinase (E8K6T) XP® Rabbit mAb	Cell Signaling Technology	Cat# 34475, RRID:AB_2943679
LC3B (D11) XP® Rabbit mAb	Cell Signaling Technology	Cat# 3868, RRID:AB_2137707
SQSTM1/p62 (D1Q5S) Rabbit mAb	Cell Signaling Technology	Cat# 39749, RRID:AB_2799160
GAPDH (D16H11) XP® Rabbit mAb	Cell Signaling Technology	Cat# 5174, RRID:AB_10622025
ATP6V1A (E5N9E) Rabbit mAb	Cell Signaling Technology	Cat# 39517
ATP6V1B2 (D3O7Q) Rabbit mAb	Cell Signaling Technology	Cat# 14488, RRID:AB_2798496
Cathepsin D Antibody	Cell Signaling Technology	Cat# 2284, RRID:AB_10694258
LAMP2 (D5C2P) Rabbit mAb	Cell Signaling Technology	Cat# 49067, RRID:AB_2799349
TFEB (D2O7D) Rabbit mAb	Cell Signaling Technology	Cat# 37785, RRID:AB_2799119
Phospho-TFEB (Ser211) (E9S8N) Rabbit mAb	Cell Signaling Technology	Cat# 37681, RRID:AB_2799117
MITF (D3B4T) Rabbit mAb	Cell Signaling Technology	Cat# 97800, RRID:AB_2800289
TFE3 Antibody	Cell Signaling Technology	Cat# 14779, RRID:AB_2687582
RagA (D8B5) Rabbit mAb	Cell Signaling Technology	Cat# 4357, RRID:AB_10545136
RagC (D8H5) Rabbit mAb	Cell Signaling Technology	Cat# 9480, RRID:AB_10614716
GFP (D5.1) Rabbit mAb	Cell Signaling Technology	Cat# 2956, RRID:AB_1196615
Ubiquitin (P37) Antibody	Cell Signaling Technology	Cat# 58395
DYKDDDDK Tag (D6W5B) Rabbit mAb	Cell Signaling Technology	Cat# 14793, RRID:AB_2572291
HA-Tag (C29F4) Rabbit mAb	Cell Signaling Technology	Cat# 3724, RRID:AB_1549585
E-Cadherin (24E10) Rabbit mAb	Cell Signaling Technology	Cat# 3195, RRID:AB_2291471
Slug (C19G7) Rabbit mAb	Cell Signaling Technology	Cat# 9585, RRID:AB_2239535
ZEB1 (E2G6Y) XP® Rabbit mAb	Cell Signaling Technology	Cat# 70512, RRID:AB_2935802
K48-linkage Specific Polyubiquitin (D9D5) Rabbit mAb	Cell Signaling Technology	Cat# 8081, RRID:AB_10859893
α/β-Tubulin Antibody	Cell Signaling Technology	Cat# 2148, RRID:AB_2288042
Mios (D12C6) Rabbit mAb	Cell Signaling Technology	Cat# 13557, RRID:AB_2798254
WDR59 (D4Z7A) Rabbit mAb	Cell Signaling Technology	Cat# 53385, RRID:AB_2799432
Histone H3 (D1H2) XP® Rabbit mAb	Cell Signaling Technology	Cat# 4499, RRID:AB_10544537
GPNMB (E4D7P) XP® Rabbit mAb	Cell Signaling Technology	Cat# 38313, RRID:AB_2799131
Anti-Rabbit IgG, HRP-linked	Cell Signaling Technology	Cat# 7074, RRID:AB_2099233
Anti-Mouse IgG, HRP-linked	Cell Signaling Technology	Cat# 7076, RRID:AB_330924
NPRL3 Antibody	Novus Biologicals	Cat# NBP1-88447, RRID:AB_11038206
STUB1 Antibody	Novus Biologicals	Cat# NB100-77315, RRID:AB_1084262
UBE3A Antibody	Novus Biologicals	Cat# NB500-240, RRID:AB_10002763
Anti-ATP6V1D antibody	Abcam	Cat# ab157458, RRID:AB_2732041
Recombinant Anti-ATP6V0D1/P39 antibody	Abcam	Cat# ab202897, RRID:AB_2802121
Recombinant Anti-SEH1L antibody	Abcam	Cat# ab218531
Mouse Anti-beta Actin	Abcam	Cat# ab8224, RRID:AB_449644
Recombinant Anti-DEPDC5 antibody	Abcam	Cat# ab213181
WDR24 Antibody	Proteintech	Cat# 20778-1-AP, RRID:AB_10696183
Mouse Anti-HERC2 monoclonal Antibody	BD Biosciences	Cat# 20778-1-AP, RRID:AB_399728
V-ATPase Subunit A Antibody (also recognizes Drosophila Vha68)	Genescript	Cat# A00938
V5 Tag monoclonal Antibody	Thermo Fisher	Cat# R960-25, RRID:AB_2556564
Guinea Pig Drosophila RagC Antibody	This study	N/A
Mouse anti-Actin JLA20	DSHB	RRID:AB_528068
Anti-Rabbit IgG AlexaFluor 488 conjugated	Thermo Fisher	Cat# A-11008, RRID:AB_143165
Anti-Rabbit IgG AlexaFluor 594 conjugated	Thermo Fisher	Cat# A-11037, RRID:AB_2534095
Anti-Mouse IgG AlexaFluor 488 conjugated	Thermo Fisher	Cat# A-11029, RRID:AB_138404
Anti-Mouse IgG AlexaFluor 594 conjugated	Thermo Fisher	Cat# A-11005, RRID:AB_141372
Goat anti-Guinea pig IgG AlexaFluor 568 conjugated	Thermo Fisher	Cat# A-11075, RRID:AB_2534119
Chemicals, Peptides, and Recombinant Proteins
Fibronectin human plasma	Sigma-Aldrich	Cat# F0895
Paraformaldehyde	Electron Microscopy Sciences	Cat# 15714
Triton X-100	Sigma-Aldrich	Cat# T8787
Disuccinimidyl suberate	Thermo Fisher	Cat# 21555
Proteinases inhibitor cocktail	Thermo Fisher	Cat# 78430
Phosphatases inhibitor cocktail	Thermo Fisher	Cat# 78420
4’,6-diamidino-2-phenylindole (DAPI)	Thermo Fisher	Cat# D1306
Hoechst 33342	Thermo Fisher	Cat# H3570
LysoSensor Green DND-189	Thermo Fisher	Cat# L7535
LysoSensor Yellow/Blue DND-160	Thermo Fisher	Cat# L7545
DQ-BSA green	Thermo Fisher	Cat# D12050
TRIzol™ Reagent	Thermo Fisher	Cat# 15596026
MG132	Sigma-Aldrich	Cat# 474790
DMSO	Sigma-Aldrich	Cat# D2650
Matrigel Matrix	Corning	Cat# 354234
Critical Commercial Assays
HA-Tag Magnetic IP/Co-IP Kit	Thermo Fisher	Cat# 88838
GeneArt™ Genomic Cleavage Detection Kit	Thermo Fisher	Cat# A24372
GFP-Trap Magnetic Agarose kit	Chromotek	Cat# gtma-100
CellTiter-Glo 2.0 cell proliferation assay	Promega	Cat# G9241
CytoSelect 96-well cell invasion assay	Cell BioLabs	Cat# CBA-112
In vitro ubiquitination kit	Enzo	Cat# BML-UW9920-0001
LookOut Mycoplasma PCR Kit	Sigma-Aldrich	Cat# MP0035
Experimental Models: Cell Lines
HeLa (Wild type)	ATCC	Cat# CCL-2
HeLa, WDR24-KO	Cai et al.23	N/A
HeLa, WDR59-KO	Yang et al.71	N/A
HeLa, MIOS-KO	Yang et al.71	N/A
HeLa, DEPDC5-KO	This study	N/A
HeLa, WDR24-KO/DEPDC5-KO	This study	N/A
HeLa, RAGA-KO	This study	N/A
HeLa, WDR24-KO/RAGA-KO	This study	N/A
HeLa, RAGC-KO	This study	N/A
HeLa, WDR24-KO/RAGC-KO	This study	N/A
Panc-1	ATCC	Cat# CRL-1469
Panc 03.27	ATCC	Cat# CRL-2549
UOK 257	Yang et al.58	N/A
UOK 257-2	Yang et al.58	N/A
UOK 124	Shipley et al.72	N/A
Experimental Models: Drosophila Stocks
Mitf-7xV5-3xGFP11 knock-in (w; ; ; Mitf-7xV5-3xGFP11)	This study	N/A
FRT82, wdr241	Cai et al.23	N/A
Nos-Flp	Wei et al.32	N/A
w; ; FRT82, wdr241; Mitf-7xV5-3xGFP11	This study	N/A
RagC mutant clone: Nos-Flp/+; FRT42D, ragC13/FRT42D, ubi-GFPnls; ; Mitf-7xV5-3xGFP11/+	This study	N/A
BHD mutant clone: Nos-Flp/+; ; FRT2A, bhd8/FRT2A, His2av-mRFP; Mitf-7xV5-3xGFP11/+	This study	N/A
Oligonucleotides
ON-TARGETplus Non-targeting siRNA control	Horizon Discovery	D-001810-01-05
SMARTpool: ON-TARGETplus NPRL3 siRNA	Horizon Discovery	L-020050-02-0005
SMARTpool: ON-TARGETplus WDR24 siRNA	Horizon Discovery	L-014822-00-0005
SMARTpool: ON-TARGETplus HERC2 siRNA	Horizon Discovery	L-007180-00-0005
SMARTpool: ON-TARGETplus STUB1 siRNA	Horizon Discovery	L-007201-00-0005
SMARTpool: ON-TARGETplus UBE3A siRNA	Horizon Discovery	L-005137-00-0005
Recombinant DNA
pRK5-HA-WDR24	Bar-Peled et al.15	Addgene, Cat# 46335
pRK5-FLAG-MIOS	Bar-Peled et al.15	Addgene, Cat# 46326
pRK5-HA-WDR59	Bar-Peled et al.15	Addgene, Cat# 46328
pEGFP-N1-MITF	Roczniak-Ferguson et al.6	Addgene, Cat# 38132
pEGFP-N1-TFE3	Roczniak-Ferguson et al.6	Addgene, Cat# 38120
pEGFP-N1-TFEB	Roczniak-Ferguson et al.6	Addgene, Cat# 38119
pRK5-HA-Ubiquitin-WT	Lim et al.73	Addgene, Cat# 17608
pRK5-HA-Ubiquitin-K48	Lim et al.73	Addgene, Cat# 17605
pRK5-HA-Ubiquitin-K48R	Lim et al.73	Addgene, Cat# 17604
pcDNA5 FRT/TO SF-HERC2	Chan et al.74	Addgene, Cat# 55613
pCMV3-UBE3A-Myc	Sino Biological	Cat# HG11130-CM
pCMV3-FLAG-STUB1	Sino Biological	Cat# HG12496-NF
pEGFP-N1-TFEB4KR	This study	N/A
Deposited data
Original images from immunoblots, immunofluorescence staining and raw data for statistical analysis	This study	Mendeley Data DOI: 10.17632/tpgdh5zdts.1
Software and Algorithms
Microsoft Excel	Microsoft	https://products.office.com/en-us/excel
GraphPad Prism 8	GraphPad Software Inc.	https://www.graphpad.com/scientific-software/prism/
Imaris 9.3.0	Bitplane Inc.	https://imaris.oxinst.com/products/imaris-for-cell-biologists
ImageJ	National Institutes of Health	https://imagej.nih.gov/ij/
Photoshop 2024	Adobe	https://www.adobe.com/products/photoshop.html
Other
Dulbecco’s modified Eagle’s medium GlutaMAX™-I with pyruvate	Thermo Fisher	Cat# 10569010
Fetal bovine serum	Thermo Fisher	Cat# A3160602
Penicillin-Streptomycin	Thermo Fisher	Cat# 15140122
DMEM without amino acids	MyBioSource	Cat# MBS653087
TrypLE™ Express	Thermo Fisher	Cat# 12605010
SuperScript™ III Reverse Transcriptase	Thermo Fisher	Cat# 18080044
PowerUp™ SYBR™ Green Master Mix	Thermo Fisher	Cat# A25778
TrueCut HiFi Cas9 Protein	Thermo Fisher	Cat# A50576
Lipofectamine™ CRISPRMAX™ Cas9 Transfection Reagent	Thermo Fisher	Cat# CMAX00003
Xfect™ Transfection reagent	Takara	Cat# 631318
Xfect™ RNA Transfection reagent	Takara	Cat# 631450
In-Fusion® Snap Assembly Master Mix	Takara	Cat# 638948
Herculase II Fusion DNA Polymerases	Agilent	Cat# 600677
PfuUltra II Fusion High-fidelity DNA Polymerase	Agilent	Cat# 600387
Dynabeads™ Protein A	Thermo Fisher	Cat# 1001D
M-PER mammalian protein extraction buffer	Thermo Fisher	Cat# 78503
Pierce™ Protein-Free T20 (TBS) Blocking Buffer	Thermo Fisher	Cat# 37571
Clarity™ Western ECL substrate	Biorad	Cat# 170-5060

Highlights.

• GATOR2 maintains lysosomal function by regulating proteasomal degradation of MiT/TFEs

• GATOR2 regulates MiT/TFEs protein levels independent of GATOR1 and the Rag GTPase

• A trio of E3 ligases targets MiT/TFE proteins for degradation in GATOR2 mutants

• Loss of GATOR2 reduces fusion TFE3 levels in Xp11 translocation renal cell carcinoma