RESEARCH ARTICLE

mTOR hyperactivity mediates lysosomal dysfunction in Gaucher’sdisease iPSC-neuronal cellsRobert A. Brown1, Antanina Voit1, Manasa P. Srikanth1, Julia A. Thayer2, Tami J. Kingsbury3,4,Marlene A. Jacobson5, Marta M. Lipinski2,6, Ricardo A. Feldman1 and Ola Awad1,*

ABSTRACTBi-allelic GBA1 mutations cause Gaucher’s disease (GD), the mostcommon lysosomal storage disorder. Neuronopathic manifestationsin GD include neurodegeneration, which can be severe and rapidlyprogressive. GBA1 mutations are also the most frequent genetic riskfactors for Parkinson’s disease. Dysfunction of the autophagy-lysosomal pathway represents a key pathogenic event in GBA1-associated neurodegeneration. Using an induced pluripotent stemcell (iPSC) model of GD, we previously demonstrated that lysosomalalterations in GD neurons are linked to dysfunction of the transcriptionfactor EB (TFEB). TFEB controls the coordinated expression ofautophagy and lysosomal genes and is negatively regulated by themammalian target of rapamycin complex 1 (mTORC1). To furtherinvestigate the mechanism of autophagy-lysosomal pathwaydysfunction in neuronopathic GD, we examined mTORC1 kinaseactivity in GD iPSC neuronal progenitors and differentiated neurons.We found that mTORC1 is hyperactive in GD cells as evidenced byincreased phosphorylation of its downstream protein substrates. Wealso found that pharmacological inhibition of glucosylceramidesynthase enzyme reversed mTORC1 hyperactivation, suggestingthat increased mTORC1 activity is mediated by the abnormalaccumulation of glycosphingolipids in the mutant cells. Treatmentwith the mTOR inhibitor Torin1 upregulated lysosomal biogenesisand enhanced autophagic clearance in GD neurons, confirming thatlysosomal dysfunction is mediated by mTOR hyperactivation. Furtheranalysis demonstrated that increased TFEB phosphorylation bymTORC1 results in decreased TFEB stability in GD cells. Our studyuncovers a new mechanism contributing to autophagy-lysosomalpathway dysfunction in GD, and identifies the mTOR complex as apotential therapeutic target for treatment of GBA1-associatedneurodegeneration.

KEY WORDS: Lysosomal storage disorder, GBA1 mutations, TFEB,iPSC, mTORC1

INTRODUCTIONGaucher’s disease (GD) is the most common lysosomal storagedisorder (LSD). It is caused by bi-allelic mutations in the GBA1(also known as GBA) gene, which encodes the lysosomal enzymeβ-glucocerebrosidase (GCase) (Jmoudiak and Futerman, 2005;Cox, 2010; Bendikov-Bar and Horowitz, 2012). GCase isresponsible for the breakdown of the cellular glycolipidsglucosylceramide (GluCer) and glucosylsphingosine (GluSph).Loss of GCase activity results in glycosphingolipid accumulationand cellular dysfunction (Brady et al., 1966; Xu et al., 2010). GD ischaracterized by various degrees of nervous system involvementdepending on the severity of the particularGBA1mutation and otherunknown factors (Mignot et al., 2013; Roshan Lal and Sidransky,2017). Type 1 GD is characterized by absent or minimalneurological manifestations, whereas both types 2 and 3 GDdevelop neuropathological changes that can be fatal (Kaminskyet al., 1996; Bellettato and Scarpa, 2010). Type 2 (acuteneuronopathic) GD patients develop severe rapidly progressiveneurodegeneration leading to death during early childhood. Type 3(chronic neuronopathic) GD patients also exhibit neurodegenerativechanges, albeit with later onset and slower progression (Wong et al.,2004; Weiss et al., 2015; Orvisky et al., 2000). GBA1mutations arealso the most frequent genetic risk factors for Parkinson’s disease(Brockmann and Berg, 2014; Klein and Westenberger, 2012), anddecreased GCase enzyme activity is linked to α-synucleinaccumulation and Parkinson’s disease pathogenesis (Almeida,2012; Mazzulli et al., 2016; Schapira, 2015).

Dysfunction of the autophagy-lysosomal pathway represents acentral pathogenic event in GBA1-associated neurodegeneration(Gan-Or et al., 2015; Zhang et al., 2015; Kinghorn et al., 2017). Theautophagy-lysosomal pathway maintains cellular homeostasis byclearing protein aggregates and damaged organelles, which iscritical for neuronal survival (Puyal et al., 2012; Kesidou et al.,2013;Menzies et al., 2017). Studies using neuronopathic GDmousemodels and induced pluripotent stem cells (iPSCs) derived fromGDpatients indicated defective autophagic clearance and accumulationof undegraded proteins in neuronopathic GD neurons (Sun andGrabowski, 2010; Farfel-Becker et al., 2014; Awad et al., 2015;Liou et al., 2016; Schöndorf et al., 2014; Osellame et al., 2013). Wehave previously demonstrated that autophagy-lysosomal pathwayalterations in neuronopathic GD iPSC neurons are linked todysfunction of the transcription factor EB (TFEB), the masterregulator of lysosomal biogenesis and autophagy. Both TFEB levelsand stability were reduced in neuronopathic GD neurons, whichresulted in lysosomal depletion and autophagy block (Awad et al.,2015). However, the mechanisms of GBA1-mediated TFEBdysfunction remain unknown.

TFEB controls the coordinated expression of lysosomal genesthrough binding to the coordinated lysosomal expression andregulation (CLEAR) element in their promoters (Settembre et al.,Received 18 December 2018; Accepted 6 September 2019

1Department of Microbiology and Immunology, University of Maryland School ofMedicine, Baltimore, MD 21201, USA. 2Department of Anesthesiology, University ofMaryland School of Medicine, Baltimore, MD 21201, USA. 3Department ofPhysiology, University of Maryland School of Medicine, Baltimore, MD 21201, USA.4University of Maryland Center for Stem Cell Biology and Regenerative Medicine,University of Maryland School of Medicine, Baltimore, MD 21201, USA. 5MoulderCenter for Drug Discovery Research, Temple University School of Pharmacy,Philadelphia, PA 19140, USA. 6Department of Anatomy and Neurobiology,University of Maryland School of Medicine, Baltimore, MD 21201, USA.

*Author for correspondence (oawad@som.umaryland.edu)

O.A., 0000-0002-9290-8942

This is an Open Access article distributed under the terms of the Creative Commons AttributionLicense (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use,distribution and reproduction in any medium provided that the original work is properly attributed.

1

© 2019. Published by The Company of Biologists Ltd | Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

2012, 2011). Enhancing TFEB activity upregulates lysosomalfunctions and promotes protein clearance, which makes it apromising therapeutic target for neurodegenerative disorders(Kilpatrick et al., 2015; Decressac et al., 2013; Sardiello andBallabio, 2009). TFEB is phosphorylated by multiple cellularkinases; the main one is the mammalian target of rapamycin(mTOR), which regulates TFEB subcellular localization andactivity (Napolitano and Ballabio, 2016; Puertollano et al., 2018).mTOR kinase is a key regulator of cellular growth and

metabolism (Saxton and Sabatini, 2017a) downstream of thephophatidylinositol-3-kinase (PI3K)/AKT signaling network(Dibble and Cantley, 2015). mTOR exists in two functionally andstructurally distinct complexes: mTORC1 and mTORC2 (Liptonand Sahin, 2014; Laplante and Sabatini, 2013). mTORC1 acts as acellular nutrient sensor and is sensitive to growth factor stimulation,nutrients and cellular energy status (Sengupta et al., 2010). Itconveys the nutrient-sensing signals to the lysosomes, thus couplinglysosomal functions to cellular energy demands (Dunlop and Tee,2014; Rabanal-Ruiz and Korolchuk, 2018; Settembre and Ballabio,2014a). To achieve this, mTORC1 phosphorylates TFEB on at leastthree serine residues, Ser122, Ser142 and Ser211, which in turnmodulates TFEB subcellular localization and transcriptionalactivity (Puertollano et al., 2018; Napolitano and Ballabio, 2016).In a high nutrient/pro-growth state, mTORC1 is active andphosphorylates TFEB at Ser142 and/or Ser211, which preventsTFEB translocation from the cytoplasm to the nucleus throughbinding to the pan-14-3-3 scaffold protein (Martina et al., 2012).Upon mTORC1 inhibition due to cellular starvation or lysosomalstress, dephosphorylated TFEB is translocated to the nucleus whereit upregulates its target genes (Settembre et al., 2012; Sardiello et al.,2009). It has been shown that mTORC1 also regulates TFEBstability and cellular abundance (Puertollano et al., 2018). TFEBphosphorylation at Ser142 and Ser211 targets inactive TFEB forproteasomal degradation through binding to the E3 ubiquitin ligase,STUB1 (Rao et al., 2017; Sha et al., 2017). Therefore, deregulationof mTORC1 kinase activity may alter TFEB phosphorylation status,which in turn affects TFEB levels and stability.We have previously demonstrated altered TFEB-mediated

lysosomal biogenesis and a block in macroautophagy (referred tohereafter as autophagy) in iPSC neurons derived fromneuronopathic GD patients (Awad et al., 2015). In the presentstudy, we investigated whether autophagy-lysosomal pathwaydysfunction in neuronopathic GD iPSC neurons is mediated byderegulation of mTOR signaling. We found that mTOR ishyperactive in neuronopathic GD neuronal cells, as evidenced byincreased levels of phospho-mTOR and excess phosphorylation ofthe mTORC1 downstream targets ribosomal protein S6 (RPS6) andthe eukaryotic translation initiation factor 4E binding protein 1(4EBP1; also known as EIF4EBP1). Treatment with GZ667161(GZ-161), a glucosylceramide synthase inhibitor that preventsthe synthesis of GluCer (Sardi et al., 2017), reduced p-RPS6levels in neuronopathic GD cells, suggesting that mTORC1hyperactivity was mediated by the abnormal accumulation ofglycosphingolipids. We have previously demonstrated thattreatment with the pharmacological mTORC1 inhibitor rapamycinresulted in autophagosome accumulation and decreased survival ofneuronopathic GD neurons (Awad et al., 2015). In the present study,we found that treatment with Torin1, but not with rapamycin,upregulated lysosomal biogenesis in neuronopathic GD neurons.Torin1 also increased autophagosome-lysosome association andimproved autophagic clearance in the mutant neurons. Moreover,we found that decreased TFEB stability in GD cells is mediated by

mTORC1 hyperactivity. Our results suggest that autophagy-lysosomalpathway dysfunction in neuronopathic GD is linked to altered lipid-sensing by mTORC1, resulting in TFEB deregulation. This study alsoidentifies the mTOR complex as a potential therapeutic target fortreatment of GBA1-associated neurodegeneration.

RESULTSIncreased mTORC1 activity in neuronopathic GD iPSCneuronal progenitor cells (NPCs)We have previously demonstrated lysosomal alterations in bothNPCs and differentiated neurons derived from neuronopathic GDiPSC lines (Awad et al., 2015, 2017). To investigate whether thesealterations are linked to mTOR deregulation, we examined mTORlevels and kinase activity in neuronopathic GD NPCs. NPCs weregenerated from control and neuronopathic GD iPSC lines aspreviously described (Awad et al., 2015). The neuronopathic GDiPSC lines were derived from two neuronopathic type 2 GD patientsharboring the bi-allelic mutations L444P/RecNciI and W184R/D409H, and from one neuronopathic type 3 GD patient with L444P/L444P mutations. Immunofluorescence analysis showed nodifference in total mTOR levels between control and mutantNPCs (Fig. 1A). However, levels of mTOR phosphorylated atSer2448 (active p-mTOR) were markedly increased in mutant NPCsas indicated by increased fluorescence signal intensity (Fig. 1B).Treatment with the catalytic mTOR inhibitor Torin1 reducedp-mTOR fluorescence signal intensity in both control and mutantNPCs (Fig. 1B). Western blot analysis showed no difference in totalmTOR levels between control and GD NPCs (Fig. S1A,B) but asignificant increase in p-mTOR levels in neuronopathic GD NPCs,which were efficiently reduced by Torin1 treatment (Fig. 1C). AsmTORC1 is involved in regulating lysosomal functions (Settembreet al., 2012; Rabanal-Ruiz et al., 2017), we compared mTORC1activity in control and neuronopathic GD NPCs. We examined thelevels of phosphorylated 4EBP1 (p-4EBP1) and phosphorylatedRPS6 (p-RPS6). Both proteins are known downstream targets ofmTORC1, and their phosphorylation status reflects mTORC1kinase activity (Ikenoue et al., 2009). We found amarked increase inthe fluorescence signal intensity of both p-RPS6 and p-4EBP1 inneuronopathic GD NPCs as compared with control cells (Fig. 1D,E). The levels of total RPS6 were not increased in neuronopathicGD NPCs as compared with control cells (Figs S1A,C and S2A).Treatment with the allosteric mTORC1 inhibitor rapamycin reducedthe fluorescence signal intensity of both p-RPS6 and p-4EBP1 inmutant NPCs (Fig. 1D,E). Conversely, mTORC1 activation byinsulin increased the levels of p-RPS6 (Fig. 1D,F). Western blotanalysis showed significant increase in p-RPS6 and p-4EBP1 levelsin neuronopathic GD NPCs (Fig. 1F,G). Torin1 treatmenteffectively reduced p-RPS6 and p-4EBP1 levels in both controland mutant NPCs (Fig. 1E,F; Fig. S2B); similarly, rapamycinsignificantly reducedmTORC1 activity in GDNPCs as shown by thedecrease in p-4EBP1 levels following treatment (Fig. 1E,G). Toconfirm mTORC1 hyperactivity in neuronopathic GD NPCs, weused two additional control lines, the iPSC line (7TA) and the humanembryonic stem cell (hESC) line (H9). Immunofluorescence analysisof NPCs generated from all three control lines showed similarbaseline levels of p-mTOR, p-RPS6 and p-4EBP1 (Fig. S3A-D).Western blot analysis confirmed the comparable low levels ofmTORC1 substrates in control NPCs compared with theirincreased levels in GD NPCs (Fig. S3E). Together, our dataindicate that mTORC1 is hyperactive in neuronopathic GD NPCs,but is still responsive to pharmacological inhibition by eitherTorin1 or rapamycin.

2

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

mTORC1 hyperactivity in neuronopathic GD NPCs ismediated by lipid substrate accumulationIt is known that nutrient sufficiency promotes mTORC1 activationand phosphorylation of its downstream targets (Rabanal-Ruiz andKorolchuk, 2018). In GD, decreased GCase enzyme activity resultsin accumulation of its lipid substrate, GluCer, in mutant neurons(Schueler et al., 2003; Farfel-Becker et al., 2014). We havepreviously shown GluCer accumulation in GD iPSC neurons(Panicker et al., 2012). To test whether lipid substrate accumulation

is responsible for the increased mTORC1 activity, we treatedneuronopathic GD NPCs with the glucosylceramide synthaseinhibitor GZ667161 (quinuclidin-3-yl-{2-[4′-fluor-(1,1′-biphenyl)-3-yl]propan-2-yl} carbamate) (GZ-161) for 72 h. This compoundinhibits the glucosylceramide synthase enzyme, which catalyzes thesynthesis of GluCer from ceramide, thus preventing its accumulation(Aerts et al., 2006; Lachmann and Platt, 2001). We then comparedmTORC1 activity in treated and untreated NPCs. Our resultsindicated that GZ-161 treatment significantly reduced p-RPS6

Fig. 1. Increased mTOR activity in GD iPSC NPCs. (A) Representative immunofluorescence images of control and GD2a NPCs labeled for mTOR (green)or DAPI (blue). Bar graph showsmeanmTOR fluorescence signal intensity in control and GDNPCs (GD2a, GD2b and GD3 combined). Data were collected from>100 cells per group, assayed in two different fields per experiment, in two independent experiments. P>0.05 between control and GD NPCs (Student’s t-test).(B) Representative immunofluorescence images of control and GD2a NPCs labeled for phospho-mTOR-Ser2448 (p-mTOR) (green) and DAPI (blue). Cells wereeither untreated (NT) or treated with 100 nM Torin1 for 6 h. Bar graph shows mean p-mTOR fluorescence signal intensity in control and GD NPCs (GD2a andGD2b combined). Data were collected from >30 cells per group, assayed in three different fields in a representative experiment. (C) Representative western blotshowing p-mTOR protein levels in control and GD NPCs derived from two GD iPSC lines (GD3 and GD2a). Cells were either untreated or treated with 100 nMTorin1 for 6 h. Also shown is β-actin loading control. Bar graph shows fold p-mTOR in treated relative to untreated control (GD2a, GD2b and GD3 NPCscombined), n=3-5 per group. (D) Representative immunofluorescence images of control and GD2a NPCs labeled for phospho-RPS6 Ser235/236 (p-RPS6).Cells were either untreated, treated with 200 nM rapamycin for 6 h or with 10 nM insulin for 30 min as indicated. (E) Representative immunofluorescence imagesof control and GD2b NPCs labeled for phospho-4EBP1 Thr37/46 (p-4EBP1). Cells were either untreated, treated with 200 nM rapamycin or 100 nM Torin for 6 h.Bar graph showsmean p-RPS6 and p-4EBP1 fluorescence signal intensity in control and GDNPCs (GD2a, GD2b and GD3 combined). Datawere collected from>100 cells per group, assayed in two or three different fields in a representative experiment. (F) Representative western blot showing p-RPS6 proteinlevels in control and GD2a NPCs with or without Torin1 or insulin treatment. Also shown is β-actin loading control. Bar graph shows fold p-RPS6 (data fromGD2a,GD2b and GD3 combined) relative to untreated control, n=3 per group. (G) Representative western blot showing p-4EBP1 levels in control and GD2a NPCswith or without rapamycin treatment. Also shown is β-actin loading control. Bar graph shows fold p-4EBP1 in GD NPCs (data from GD2a and GD2b combined)relative to untreated control, n=3 per group. Data are mean±s.e.m. *P<0.05, **P<0.005, ***P<0.0005, ****P<0.00005 (one-way ANOVA between indicatedgroups). Scale bars: 100 µm in A,D,E (magnification 20×); 50 µm in B (magnification 60×).

3

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

levels in neuronopathic GD NPCs to almost control levels as shownby quantitation of p-RPS6 fluorescence signal intensity (Fig. 2A,B).Western blot analysis also showed that GZ-161 treatment ofneuronopathic GD NPCs caused a significant reduction in bothp-mTOR and p-RPS6 levels (Fig. 2C). Thus, our results suggestthat increased mTORC1 activity in neuronopathic GD NPCs ismediated by the abnormal accumulation of glycosphingolipids in themutant cells.

Increased mTORC1 activity in neuronopathic GD iPSCneuronsWe then examined mTOR activity and p-mTOR levels in iPSCneurons differentiated from control and neuronopathic GD NPCs aspreviously described (Awad et al., 2015). We found that, similar toour findings in NPCs, levels of p-mTOR were significantlyincreased in neuronopathic GD neurons compared with controlneurons (Fig. 3A,B). We also found that Torin1 treatmentsignificantly reduced p-mTOR levels in both control and mutantneurons (Fig. 3B). We then examined mTORC1 activity bycomparing the levels of p-RPS6 and p-4EBP1 in control andneuronopathic GD neurons. Both control and GD NPCs weredifferentiated to neurons as shown by the expression of the neuron-specific marker class III β-tubulin (Tuj1) (Fig. 3C).

Immunofluorescence analysis showed increased fluorescencesignal intensity of p-RPS6 in neuronopathic GD neuronscompared with controls (Fig. 3C). Western blot analysisconfirmed the significant increase in p-RPS6 but not total RPS6levels in neuronopathic GD neurons (Fig. 3D). The level ofp-4EBP1 was also significantly increased in neuronopathic GDneurons (Fig. 3E), and both p-RPS6 and p-4EBP1 levels werereduced to control levels in response to Torin1 treatment (Fig. 3D,E).

Treatment of neuronopathic GD iPSC neurons with Torin1upregulates lysosomal marker expressionAs dysfunction of the autophagy-lysosomal pathway is directlylinked to neuronal loss in neuronopathic GD (Sun and Grabowski,2010), it is important to elucidate its underlying mechanism. Wehave previously demonstrated altered TFEB-mediated lysosomalbiogenesis and decreased lysosomal marker expression inneuronopathic GD iPSC neurons (Awad et al., 2015). Toinvestigate whether those alterations are linked to mTORhyperactivity, we examined the effects of pharmacological mTORinhibition (using rapamycin or Torin1) on lysosomal geneexpression in neuronopathic GD neurons by qRT-PCR analysis.The lysosomal genes examined were: LAMP1, GBA1, cathepsin D,cathepsin B, hexosaminidase A (HEXA) and glucosamine

Fig. 2. Lipid substrate accumulation mediates mTOR hyperactivity in GD iPSC NPCs. (A) Representative immunofluorescence images of control andGD2a NPCs co-labeled for p-RPS6 (green) and DAPI (blue). GD NPCs were either untreated or treated with 5 µM of the substrate reduction compound (SRC)GZ-161 for 72 h. (B) Quantitation of p-RPS6 fluorescence signal intensity in control and GD2a NPCs with and without GZ-161 treatment. Data from >100 cellsper group, assayed in at least four different fields per experiment, in three experiments. Bar graph represents fold change in the mean p-RPS6 fluorescenceintensity relative to control. (C) Representative western blot showing p-mTOR (left) and p-RPS6 (right) levels in control, GD2a and GD2a NPCs treated with theSRC. Also shown is β-actin loading control. Bar graphs represent fold p-mTOR and p-RPS6 in GDNPCs (data fromGD2a andGD2b combined) relative to control,n=3-5 per group. Data are mean±s.e.m.*P<0.05 and **P<0.005 (one-way ANOVA between indicated groups). Scale bar: 50 µm (magnification 20×).

4

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

(N-acetyl)-6-sulfatase (GNS). These genes are known TFEBtargets, which we have previously shown to be downregulated inneuronopathic GD iPSC neurons (Awad et al., 2015). We found thatTorin1 significantly increased the expression levels of most ofthe lysosomal genes examined in treated versus untreatedneuronopathic GD neurons (Fig. 4A), whereas rapamycin did notinduce significant change in the expression levels of these genes(Fig. 4A), suggesting a differential effect of Torin1 and rapamycinon lysosomal biogenesis in neuronopathic GD neurons. To furtherexplore this possibility, we compared the levels of the lysosomal-associated membrane protein 1 (LAMP1) in neurons treated witheither Torin1 or rapamycin. As shown in Fig. 4B, western blotanalysis indicated that LAMP1 levels were significantly reduced inmutant versus control neurons, and that Torin1, but not rapamycin,

increased LAMP1 levels in neuronopathic GD neurons. Thus, ourdata suggest that the altered lysosomal biogenesis in neuronopathicGD neurons is linked to mTOR hyperactivity and that Torin1 is ableto rescue this phenotype.

Treatment with Torin1 induces TFEB-GFP nucleartranslocation and upregulates lysosomal biogenesis inneuronopathic GD iPSC neuronsTo test whether the effect of mTOR inhibition on lysosomalbiogenesis in neuronopathic GD neurons is mediated by TFEB, weoverexpressed a TFEB-green fluorescent protein (GFP) fusionprotein in control and neuronopathic GD neurons as previouslydescribed (Awad et al., 2015). We then treated neuronal cultureswith either Torin1 or rapamycin and examined the changes in

Fig. 3. Increased mTORC1 activity in GD iPSC neurons. (A) Representative immunofluorescence images of control and GD3 neurons labeled for p-mTOR(red) and DAPI (blue). Bar graph (right) shows p-mTOR mean fluorescence signal intensity in control and GD neurons (GD2a and GD3 combined). Data werecollected from >25 cells per group, assayed in two different fields, in a representative experiment. P<0.05 (Student’s t-test). (B) Representative western blotshowing p-mTOR (top) andmTOR (bottom) protein levels in control and GD2b neurons untreated (NT) or treated with 100 nM Torin1 for 6 h. Also shown is β-actinloading control. Bar graph represents fold p-mTOR in GD2 neurons (data from GD2a and GD2b combined) relative to untreated control, n=3-5 per group.(C) Representative immunofluorescence images of control and GD2b neurons labeled for p-RPS6 and Tuj1. The overlay of both markers is shown in the lastpanel. Bar graph (below) represents p-RPS6 mean fluorescence signal intensity in control and GD2 neurons (data from GD2a and GD2b combined). Data werecollected from >100 cells per group, assayed in two to four different fields, in two experiments.P<0.005 (Student’s t-test). (D) Representativewestern blot showingp-RPS6 (top) and RPS6 (bottom) levels in control and GD2a neurons with or without 100 nM Torin1 treatment for 6 h. Also shown is β-actin loading control. Bargraph represents fold p-RPS6 in GD2 neurons (data fromGD2a andGD2b combined) relative to untreated control, n=3 per group. (E) Representativewestern blotshowing p-4EBP1 (top) and 4EBP1 (bottom) protein levels in control and GD2a neurons with or without 100 nM Torin1 treatment for 6 h. Also shown is β-actinloading control. Bar graph represents fold p-4EBP1in GD2 neurons (data fromGD2a andGD2b combined) relative to untreated control, n=3-4 per group. Data aremean±s.e.m. *P<0.05, **P<0.005, ****P<0.00005 (one-way ANOVA between indicated groups). Scale bars: 50 µm in A (magnification 60×); 100 µm in C(magnification 20×).

5

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

TFEB-GFP nuclear translocation, which is required for TFEBtranscriptional activity. We found that Torin1 induced strongnuclear translocation of the TFEB-GFP fluorescence signal in bothcontrol and neuronopathic GD neurons compared with bothuntreated and rapamycin-treated cells (Fig. 5A). Almost allneurons treated with Torin1 had TFEB-GFP fluorescence signalprimarily in the nucleus as shown by TFEB-GFP colocalized withnuclear DAPI staining (Fig. 5A,B). Quantitation of TFEB-GFPfluorescence intensity demonstrated an increase in the ratio ofnuclear/cytoplasmic TFEB-GFP in control and neuronopathic GDneurons treated with Torin1, compared with both untreated andrapamycin-treated cells (Fig. 5C). Thus, consistent with previousreports, mTOR inhibition by Torin1 is more effective thanrapamycin in inducing TFEB nuclear translocation (Settembreet al., 2012). We have previously shown that TFEB overexpressionis unable to upregulate lysosomal biogenesis in neuronopathic GD

neurons, except in the presence of recombinant GCase enzyme(Awad et al., 2015). Here, we wanted to test whetherpharmacological inhibition of mTOR would overcome therequirement for active GCase and rescue TFEB-mediatedlysosomal biogenesis in mutant neurons. To this end, we treatedTFEB-GFP-expressing neurons with either Torin1 or rapamycin,and examined LAMP1 expression, which corresponds withlysosomal abundance. We noticed that Torin1, and to a lesserextent rapamycin, increased LAMP1 fluorescence intensity inneuronopathic GD neurons compared with untreated neurons(Fig. 5D). Quantitation of LAMP1 signal demonstrated thatTorin1 treatment significantly increased LAMP1 area in bothcontrol and neuronopathic GD neurons (Fig. 5E). Thus, our dataindicate that mTOR inhibition with Torin1 is able to enhanceTFEB-mediated lysosomal biogenesis in neuronopathic GDneurons.

Fig. 4. Torin1 upregulates lysosomal marker expression in GD iPSC neurons. (A) qRT-PCR analysis showing fold expression of lysosomal genes incontrol and GD2 neurons (data from GD2a and GD2b combined). GD neurons were untreated (NT) or treated with either 100 nMTorin1 or 200 nM rapamycin for18 h. The lysosomal genes examined were cathepsin B (CATB), cathepsin D (CATD), GNS, LAMP1, HEXA and GBA, n=3-6 per group. (B) Representativewestern blot showing LAMP1 protein levels in control and GD2a neurons that were either untreated (NT) or treated with Torin1 or rapamycin. Also shown is β-actinloading control. Bar graph represents fold LAMP1 in control and GD2 neurons (data from GD2a and GD2b combined) relative to untreated control,n=3-4 per group. Data are mean±s.e.m. *P<0.05, **P<0.005, ***P<0.0005 (one-way ANOVA between indicated groups).

6

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

Treatment with Torin1 improves autophagic clearancein GD iPSC neuronsNext, we tested whether the observed upregulation of TFEB-mediated lysosomal biogenesis by Torin1 would also improveautophagic clearance in neuronopathic GD neurons. We expressed aGFP-LC3 fusion protein in control and neuronopathic GD neuronsas previously described (Awad et al., 2015) and treated the cells withTorin1. We then examined the number and fluorescence intensity ofGFP-LC3 puncta, which corresponds with autophagosomeabundance (Klionsky et al., 2012). There was a significantincrease in both the number and fluorescence intensity of GFP-LC3 puncta in neuronopathic GD compared with control neurons(Fig. 6A,B), indicating autophagosome accumulation in the basalstate. We have previously demonstrated that this autophagosome

accumulation in neuronopathic GD neurons is due to defects inlysosomal degradation rather than an increase in autophagy flux(Awad et al., 2015). Treatment with Torin1 did not cause astatistically significant increase in either the number or thefluorescence intensity of GFP-LC3 puncta in control or mutantneurons compared with the corresponding untreated cells (Fig. 6B).This suggests that autophagy induction by Torin1 did not result inexcessive autophagosome accumulation in the mutant neurons,which is opposite to the effect of rapamycin that we have previouslyreported (Awad et al., 2015). Similar results were obtained when wecompared endogenous levels of LC3II (which corresponds toautophagosome abundance) in NPCs by western blot analysis(Fig. S4A). We then examined the effect of Torin1 on autophagosomeassociation with the lysosomes. We labeled GFP-LC3-expressing

Fig. 5. Torin1 upregulates TFEB-mediated lysosomal biogenesis in GD iPSC neurons. (A) Representative fluorescence images for control (left) and GD2a(right) neurons expressing TFEB-GFP fusion protein (green). Neurons were either untreated (NT) or treated with 100 nM Torin1 or 200 nM rapamycin for18 h. Arrows point to the overlap of TFEB-GFP signal with nuclear DAPI (blue). (B) Quantitation of TFEB-GFP nuclear translocation in control and GD2 neurons(data from GD2a and GD2b combined) that were either untreated or treated with Torin1 or rapamycin. Bar graph represents the percentage of neurons withnuclear TFEB-GFP normalized to the number of GFP-expressing neurons in the same field, in three independent experiments. (C) Fluorescence quantitation ofthe ratio of nuclear/ cytoplasmic TFEB-GFP signal in control and GD2 neurons (data from GD2a and GD2b combined) that were either untreated or treated withTorin1 or rapamycin. Data from >50 cells, assayed in at least four different fields per group in two independent experiments. (D) Representative z-stackfluorescence images of control andGD2a neurons expressing TFEB-GFP fusion protein labeled for LAMP1. Neurons were either untreated or treatedwith 100 nMTorin1 or 200 nM rapamycin for 18 h. (E) Fluorescence quantitation of the LAMP1 area in control and GD2 neurons (data from GD2a and GD2b combined)that were either untreated or treated with Torin1 or rapamycin. Data from >50 cells, assayed in at least four different fields in three independent experiments.Data are mean±s.e.m. *P<0.05, **P<0.005, ***P<0.0005, ****P<0.00005 (one-way ANOVA between indicated groups). Scale bars: 50 μm in A (magnification20×); 50 μm in D (magnification 60×).

7

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

neurons with anti-LAMP1 antibody and determined the percentageof GFP-LC3 puncta colocalization with LAMP1 using fluorescencesignal quantitation (Fig. 6C; Fig. S5). As shown in Fig. 6D, wefound a significant increase in the fraction of GFP-LC3 punctacolocalized with LAMP1 in neuronopathic GD neurons treated withTorin1 compared with untreated cells, indicating increasedautophagosome-lysosome association in response to treatment. Totest whether the enhanced autophagosome-lysosomal associationinduced by Torin1 would improve autophagic clearance, weexamined levels of the autophagic substrate p62/SQSTM1 (p62)protein in neuronopathic GD neurons expressing the GFP-LC3fusion protein. Immunofluorescence analysis showed an increase inp62 fluorescence signal intensity in neuronopathic GD neurons

compared with controls, which was significantly reduced bytreatment with Torin1 (Fig. S4B,C). To further explore the effectof Torin1 on autophagic clearance in neuronopathic GD cells; wemeasured endogenous protein levels of p62 and NBR1 (anotherautophagic substrate that also targets ubiquitinated proteins forautophagic degradation). Western blot analysis showed a significantincrease in the basal levels of p62 and NBR1 proteins inneuronopathic GD neurons, indicative of an autophagic defect,and that treatment with Torin1 reduced the levels of both proteins(Fig. 6E). Both p62 and NBR1 proteins are known to promoteautophagic degradation of the ubiquitinated targets (Kirkin et al.,2009), thus a decrease in their levels reflects efficient degradationvia the lysosomes. Taken together, our results indicate that treatment

Fig. 6. Torin1 improves autophagic clearance in GD iPSC neurons. (A) Representative fluorescence images for control and GD2b neurons, expressing GFP-LC3 fusion protein. Neurons were either untreated (NT) or treated with 100 nM Torin1 for 18 h. Insets are enlargement of a small area in each panel. Arrows pointto GFP-LC3-labeled puncta. (B) Fluorescence quantitation of GFP-LC3 puncta in GD2 neurons (data from GD2a and GD2b combined) with and without Torin1treatment. The bar graphs show average GFP-LC3 puncta number (left) and average GFP-LC3 puncta fluorescence intensity (right). (C) Representativefluorescence images for control and GD2a neurons, expressing GFP-LC3 fusion protein (green) and LAMP1 (red). Neurons were either untreated or treated with100 nM Torin1 for 18 h. The merged image of GFP-LC3 puncti and LAMP1-labeled lysosomes is also shown. Insets to the right are enlargement of the boxedareas to show the fluorescence signal colocalization. (D) Quantitation of the percentage colocalization of GFP-LC3 puncti and LAMP1 fluorescence signal incontrol andGD2 neurons (data fromGD2a andGD2b combined) with andwithout Torin1 treatment for 18 h. Data from >100 cells per group assayed in at least fourdifferent fields in two independent experiments. (E) Representative western blot showing p62 and NBR1 protein levels in control and GD2b neurons that wereeither untreated or treated with 100 µmTorin1 for 18 h. Also shown is β-actin loading control. Bar graphs show fold p62 (left) and NBR1 (right) in GD neurons (DatafromGD2a, GD2b andGD3 combined) relative to the untreated control, n=3-4 per group. Data aremean±s.e.m. *P<0.05, **P<0.005, ***P<0.0005, ****P<0.00005(one-way ANOVA between indicated groups). Scale bars: 50 µm in A (magnification 40×); 25 µm in A (inset); 25 µm in C (magnification 20×); 10 µm in C (inset).

8

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

with Torin1 improves both lysosomal biogenesis and autophagicclearance in neuronopathic GD neurons.

DecreasedTFEBstability in neuronopathicGDNPCs is linkedto mTOR hyperactivityWe have previously shown decreased TFEB protein levels inneuronopathic GD iPSC-derived neurons and NPCs (Awad et al.,2015, 2017). To further confirm the effect of decreased GCaseactivity on TFEB levels, we treated a human neuroglioma cell line(H4), which stably expresses a TFEB-GFP fusion protein with thepharmacological irreversible GCase inhibitor conduritol B-epoxide(CBE). This model enables us to directly monitor the effect ofdecreasing GCase activity by treatment with CBE on TFEB levels.We then examined GFP fluorescence signal intensity, whichcorresponds with TFEB levels. We found that as early as 48 hafter CBE treatment, TFEB-GFP fluorescence signal intensitysignificantly declined in treated versus untreated H4 cells (Fig.S6A). Treatment with the proteasome inhibitor (PSI) Clasto-lactacystin β-lactone restored the GFP signal intensity in CBE

treated H4 cells, suggesting that the decline in TFEB level was dueto excess TFEB degradation by the proteasome (Fig. S6B). We alsofound a significant increase in p-mTOR levels in GFP-TFEB-expressing H4 cells in response to CBE treatment (Fig. S6C),suggesting that decreased GCase activity affects mTORphosphorylation status.

It is known that TFEB phosphorylation by mTORC1 at Ser142targets inactive TFEB for proteasomal degradation (Sha et al.,2017). Wewanted to examine whether the decreased TFEB stabilityin neuronopathic GD cells is linked to mTOR hyperactivity. First,we confirmed the decrease in TFEB levels in GD NPCs by westernblot analysis (Fig. S6D). Then we treated NPCs with a PSI toprevent proteasomal degradation of phosphorylated TFEB andcompared the levels of p-TFEB(Ser142) in control andneuronopathic GD NPCs using a phospho-specific antibody.Western blot analysis of protein lysates from neuronopathic GDNPCs treated with a PSI showed multiple p-TFEB(Ser142) bands ofhigh molecular weight (Fig. 7A). Western blot quantificationshowed a significant increase in p-TFEB (Ser142) levels in

Fig. 7. Increased p-TEFB(Ser142) level in GD iPSC NPCs. (A) Representative western blot showing p-TEFB(Ser142) levels in control and GD2a NPCs thatwere treated with the proteasome inhibitor Clasto-lactacystin β-lactone for 18 h. Also shown is β-actin loading control. (B) Bar graph shows western blotquantitation of fold p-TEFB(Ser142) level in GD2 NPCS (data from GD2a and GD2b combined) relative to control, n=3 per group. (C) Representativeimmunoprecipitation for control, GD2a, and GD2b NPCs using p-TFEB(Ser142) antibody and probed with anti-ubiquitin antibody. Cells were treated with theproteasome inhibitor for 18 h and either untreated or co-treated with Torin1 for 8 h. β-Actin loading levels in the input lysates are also shown. (D) Representativewestern blot (left) showing total TEFB level in control and GD2a NPCs. Cells were treated with the proteasome inhibitor for 18 h and were either untreated orco-treated with Torin1 for 8 h. Bar graph (right) shows western blot quantitation of total TEFB in GD NPCs (GD2a and GD3) relative to untreated control, n=3 pergroup. (E) Schematic for a proposedmechanism of TFEB dysfunction in neuronopathic GD, in which glycosphingolipid accumulation leads to increasedmTORC1activity. mTORC1 hyperactivation results in increased TFEB phosphorylation, which targets TFEB for proteasomal degradation. mTOR inhibition by Torin1stabilizes TFEB and allows its nuclear translocation, thus upregulating lysosomal functions. Data are mean±s.e.m. *P<0.05, ***P<0.0005 (Student’s t-test).

9

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

neuronopathic GDNPCS compared with control cells (Fig. 7B). Wethen examined whether polyubiquitination of p-TFEB isresponsible for this ladder pattern detected in neuronopathic GDNPCs lysates. To this end, we treated control and mutant NPCs witha PSI and performed immunoprecipitation of p-TFEB(Ser142)followed by western blot analysis using anti-Ubiquitin antibody.Our results showed increased ubiquitination of p-TFEB inneuronopathic GD NPCs compared with control cells (Fig. 7C).The intensity of the polyubiquitinated bands in neuronopathic GDcells was efficiently reduced by Torin1 treatment, suggesting thatmTOR inhibition in neuronopathic GD NPCs decreased p-TFEBubiquitination (Fig. 7C). Moreover, western blot analysis of totalTFEB levels in NPCs that were treated with PSI and were eitheruntreated or co-treated with Torin1 showed a significant increase intotal TFEB levels in GD NPCs treated with Torin1 compared withuntreated cells (Fig. 7D). This further supports that Torin1 increasesTFEB stability in GD cells. Taken together, our results support amodel in which neuronopathic GBA1 mutations result in lipidsubstrate accumulation and increased mTORC1 activity. mTORC1hyperactivity causes excess TFEB phosphorylation, which in turntargets TFEB for ubiquitination and proteasomal degradation. Thismodel is shown in Fig. 7E.

DISCUSSIONIn this study we provide evidence that autophagy-lysosomal pathwaydysfunction in neuronopathic GD iPSC-derived neurons is, at least inpart, mediated by increased mTOR activity. mTOR is a highlyconserved serine-threonine kinase that exists in two multiproteincomplexes, mTORC1 and mTORC2 (Laplante and Sabatini, 2009).mTORC1 integrates multiple signals such as growth factors, energystatus and nutrient abundance, to control cellular growth andmetabolism (Saxton and Sabatini, 2017b; Sengupta et al., 2010). It isnow known that mTORC1 also regulates the anabolic-catabolicequilibrium of many classes of lipids (Ricoult and Manning, 2013;Menon et al., 2017). This is achieved, in part, by controllingautophagic degradation of cellular lipids, or lipophagy (Antikainenet al., 2017; Settembre and Ballabio, 2014b). This process involvescoordination between lysosomal nutrient-sensing machinery bymTORC1 and TFEB activity (Settembre and Ballabio, 2014b;Jaishy and Abel, 2016). Thus, it is likely that the abnormal lipidstorage in LSDs disrupts the signaling regulation of this machinery.In support of this idea, studies have demonstrated mTOR signaldysregulation in various LSD models such as Fabry disease, Pompedisease and mucopolysaccharidoses (Ward et al., 2016; Bartolomeoet al., 2017; Lim et al., 2017). In this study, we found that treatmentof neuronopathic GD NPCs with the substrate reduction compoundGZ-161 normalized mTORC1 activity, suggesting that the increasein mTORC1 activity was indeed mediated by the abnormalglycosphingolipid accumulation. Similarly, it has been shown thatGlcCer synthase inhibitors act on Akt-mTOR signaling and reducephosphorylation of RPS6 (Natoli et al., 2010). Substrate reductiontreatment also increased lysosomal abundance and improvedautophagic flux in mouse primary neurons in an mTOR-dependentmanner (Shen et al., 2014). In addition, GZ-161 has been shown tocross the blood brain barrier in neuronopathic GD mouse models(Cabrera-Salazar et al., 2012), and preventing GlcCer accumulationprovides neuroprotective effects in some models of LSD (Stein et al.,2012; Ashe et al., 2011). Substrate reduction compounds are shownto provide clinical improvement in type 1 GD patients (Mistry et al.,2017; Zimran et al., 2018). Our study suggests that it may also be aneffective approach to restore autophagy-lysosomal pathwayfunctions in neuronopathic GD.

It is known that the recruitment and retention of mTORC1 to thelysosome is required for its activation (Yao et al., 2017). Thus, thedefective lysosomes in LSDs may diminish rather than increasemTORC1 activity. However, it has been shown that in some LSDmodels, lysosomal storage enhances rather than inhibits mTORC1signaling (Bartolomeo et al., 2017; Vidal-Donet et al., 2013). Theco-existence of mTORC1 hyperactivation and lysosomaldysfunction has also been found in some genetic models ofParkinson’s disease, which was linked to impaired TFEB activity(Bento et al., 2016). Our study suggests that in neuronopathic GD,mTORC1 hyperactivation may also occur without fully functionallysosomes. The potential involvement of mTOR signalingderegulation in neuronopathic GD was also reported in mousemodels. Dasgupta et al. have previously demonstrated increasedexpression of mTOR pathway mRNAs in the brains ofneuronopathic GD mice (Dasgupta et al., 2015). However, ourfindings contradict previous reports using other genetic models ofGBA1 deficiency, which show mTORC1 signaling attenuation, notactivation. Magalhaes et al. reported that in primary neurons fromGBA1 transgenic mice and in fibroblasts from Parkinson’s diseasepatients with GBA1 mutations, there was altered lysosomalrecycling and a decrease in phosphorylated S6K levels(Magalhaes et al., 2016). Similarly, in a Drosophila model ofneuronopathic GD, mTOR signaling was found to bedownregulated in the brains of mutant flies. Yet, rapamycintreatment partially rescued some of the observed phenotypes,raising the possibility that inhibiting the activity of mTORmay be oftherapeutic value (Kinghorn et al., 2016). As these reports on therelation between mTOR activity and GCase enzyme were carriedout in different models of GBA1 deficiency, the reasons behind thisapparent discrepancy are unclear. One possibility is differentialspecies-dependent (mouse and fly versus human) and cell type-dependent (fibroblasts versus NPCs and neurons) regulation of themTOR and TFEB activity in response to GlcCer accumulation.Further studies are needed to clarify the role of mTORC1 in GBA1-associated neurodegeneration, and to explore the differenttherapeutic approaches for treatment of those conditions.

We have previously demonstrated altered TFEB-mediatedlysosomal biogenesis in neuronopathic GD neurons (Awad et al.,2015). mTORC1 is a key upstream negative regulator of TFEB(Settembre et al., 2012), thus our data indicate that TFEB alterationsin iPSC neuronopathic GD neurons are likely mediated bymTORC1 hyperactivity. A similar correlation between mTORhyperactivity and TFEB downregulation was also reported inAMPK-deficient embryoid bodies, which exhibit elevated mTORsignaling (Young et al., 2016). Young et al. found that AMPK−/−

embryoid bodies have reduced levels of TFEB and alterations inTFEB phosphorylation status. They also found that treatmentwith a pharmacological mTOR inhibitor reversed TFEBhyperphosphorylation status in those knockout embryoid bodies(Young et al., 2016). It is known that when mTORC1 is active, itphosphorylates TFEB on Ser142 and/or Ser211, which preventsTFEB nuclear translocation and sequesters it in the cytosol(Puertollano et al., 2018; Napolitano and Ballabio, 2016).However, although mTORC1 was hyperactive in our GD iPSCneurons, we did not detect excess cytoplasmic sequestration ofTFEB. Instead, total TFEB levels were reduced and, when detected,TFEBwas mainly in the nucleus (Awad et al., 2015). These findingsmight be explained by studies showing that mTORC1 also regulatesTFEB stability (Puertollano et al., 2018). mTORC1phosphorylation of Ser142 and Ser211 residues targets inactivephosphorylated TFEB for ubiquitination and proteasomal

10

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

degradation (Sha et al., 2017). In agreement with this report, wefound increased levels of phosphorylated TFEB Ser142 inneuronopathic GD NPCs, which can explain the decreased TFEBlevels and stability that we previously reported (Awad et al., 2015).Our results showed that treatment of neuronopathic GD neurons

with the pharmacological mTOR inhibitor Torin1 rescued TFEB-mediated lysosomal biogenesis and improved autophagic clearance.This is in contrast to our previous results demonstrating thatrapamycin is unable to rescue TFEB function in neuronopathic GDneurons, instead resulting in increased cell death. It is known thatrapamycin is an allosteric mTOR inhibitor and only suppresses partof mTORC1 function, whereas Torin1 is a catalytic inhibitor thatcompletely suppresses mTOR activity (Thoreen and Sabatini, 2009;Thoreen et al., 2009). As Torin1 inhibits both mTORC1 andmTORC2 activity (Zheng and Jiang, 2015), we cannot rule out thepossibility of mTORC2 involvement in our results. However, it isknown that many mTORC1 functions that involve autophagyregulation are resistant to inhibition by rapamycin (Thoreen andSabatini, 2009). Also, the differential effects of Torin1 andrapamycin on TFEB regulation and lysosomal functions areconsistent with previous studies (Thoreen and Sabatini, 2009;Zhou et al., 2013). For instance, it has been shown that Torin1 is apotent activator of TFEB nuclear translocation, whereas rapamycinhas little or no effect on TFEB subcellular localization (Zhou et al.,2013; Settembre et al., 2012; Li et al., 2018). It has also been shownthat the TFEB Ser142 residue is a rapamycin-resistant mTORC1phosphorylation site, and that Torin1, but not rapamycin, is able toblock nutrient-induced phosphorylation of TFEB on Ser142(Settembre et al., 2012). Moreover, it was found that rapamycin isunable to increase TFEB transcriptional activity, and thatsuppression of mTOR activity by Torin1, but not rapamycin,leads to activation of lysosomal function (Zhou et al., 2013). Thus,our results further support that TFEB is a rapamycin-resistantsubstrate of mTORC1. Our previous finding that rapamycintreatment resulted in decreased survival of neuronopathic GDneurons can be explained by the inability of rapamycin to improvelysosomal functions in those cells (Awad et al., 2015). AlthoughmTOR inhibition by rapamycin induced autophagosome formationin GD iPSC neurons, it was not associated with increasedautophagosome-lysosome fusion or improvement in autophagicclearance (Awad et al., 2015). Here, we found that Torin1 didincrease autophagosome-lysosome association and improvedautophagic clearance in neuronopathic GD neurons and that theseeffects were likely mediated through enhancement of TFEB activity.It would be interesting to investigate the effect of Torin1 on thesurvival of GD neurons in future studies.Although our study focuses on the effect of GBA1 mutations on

TFEB regulation by mTOR, it does not explore other upstreamregulatory signals, which may also be affected by GBA1. Onepotential pathway is the unfolded protein response and endoplasmicreticulum stress, which is shown to activate TFEB through PERK(PKR-like endoplasmic reticulum kinase) and calcineurin, but isindependent of mTOR (Martina et al., 2016).In conclusion, our study provides a novel mechanism

contributing to autophagy-lysosomal pathway dysfunction inneuronopathic GD and identifies the mTOR complex as apotential therapeutic target in GBA1-associated neurodegeneration.

MATERIALS AND METHODSGeneration of neuronopathic GD iPSC NPCs and neuronsAll the control and GD iPSC lines used in this study have been previouslydescribed (Sgambato et al., 2015; Panicker et al., 2012; Awad et al., 2017).

The DF4-7T.A (7TA) iPSC line and the hESC line H9/WA09 werepurchased from the WiCell Repository. The neuronopathic GD iPSC lineswere derived from two neuronopathic type 2 GD patients harboring thebi-allelic mutationsW184R/D409H (GD2a) and L444P/RecNciI (GD2b) andfrom one neuronopathic type 3 GD patient with L444P/L444P mutations(GD3). Experiments were performed using the three neuronopathic GDiPSC lines and data from GD2 and GD3 were combined where indicated inthe figure legends. For each neuronopathic GD iPSC line, two clones weregenerated and assayed. NPCs were generated from embryoid bodies afterneuronal induction and rosette formation as previously described (Awadet al., 2015). NPCs were maintained in culture media consisting ofNeurobasal medium (Life Technologies), containing 1× (vol/vol) MEMnon-essential amino acids (Life Technologies), 1× (vol/vol) GlutaMAX-ICTS (Life Technologies), 1× (vol/vol) B27 supplement (LifeTechnologies), 1× (vol/vol) penicillin/streptomycin and 20 ng/ml basicfibroblast growth factor (Stemgent), with media changed every other day.NPCs were differentiated to neurons according to our previously describedprotocol and maintained in neuronal differentiation medium for 3-4 weeksbefore being assayed (Awad et al., 2015). All control and GD iPSC NPCsused in the study had been recently tested for contamination.

Immunofluorescence analysisFor immunofluorescence analysis, NPCs were plated on chamber slides (Lab-Tek) or differentiated into neurons on poly-L-ornithine/laminin-coated glass-bottom culture dishes (MatTek). Cells were fixed in 4% (vol/vol)paraformaldehyde for 15 min followed by blocking in phosphate-bufferedsaline (PBS) containing 8% fetal bovine serum (vol/vol) for 1 h. The primaryantibodies were prepared in PBS solutionwith 2 mg/ml saponin and incubatedfor 2 h at room temperature or overnight at 4°C. The cells were then washed inPBS and incubated with the corresponding fluorochrome-conjugatedsecondary antibodies for 1 h. Vectashield mounting medium plus DAPI(Vector Laboratories) was applied to label the nuclei. The following primaryantibodies were used: anti-mTOR (Cell Signaling Technology, 2972) 1:100;anti-phospho-mTOR-Ser2448 (Cell Signaling Technology, 5536) 1:100;anti-S6 Ribosomal Protein (Cell Signaling Technology, 2217) 1: 200; anti-phosphoS6-Ser235/236 (Cell Signaling Technology, 2211) 1:100;anti-phospho-4EBP1-Thr37/46 (Cell Signaling Technology, 2855) 1:100;anti-LAMP1 (Developmental Studies Hybridoma Bank, H4A3) 1:200; anti-p62 (BD Biosciences, 610832) 1:100; anti-Tuj1 (Neuromics, MO15013)1:200; anti-TFEB (MyBioSource, MBS855552) 1:50. The secondaryantibodies used were Alexa fluor 488- or 594-conjugated mouse or rabbit(Life Technologies), both at a 1:200 or 1:400 dilution.

Chemical reagents and cell treatmentFor mTOR pathway analysis, NPCs and neurons were treated with 200 nMrapamycin (Sigma-Aldrich) or 100 nM Torin1 (Sigma-Aldrich) for 2-6 h.NPCs were also treated with 10 nM insulin (Sigma-Aldrich) for 20-40 minwhere indicated. For lysosomal biogenesis and autophagy studies, culturedneurons were treated with the same dose of rapamycin or Torin1 as above for18-24 h before analysis. For substrate reduction treatment, we treated NPCswith GZ-161. This compound was synthesized according to WO 2012/129084 at The Moulder Center for Drug Discovery Research, TempleUniversity School of Pharmacy, Philadelphia, PA, USA. NPCs wereincubated with 5 µM GZ-161 for 72 h before being assayed. We also usedthe proteasome inhibitor, Clasto-lactacystin β-lactone (Cayman Chemical),at a concentration of 0.5 μg/ml for 18 h where indicated.

Lentiviral infectionThe TFEB-GFP and GFP-LC3 lentiviral constructs used in this study havebeen previously described (Awad et al., 2015). For lentiviral infection,control and GD neurons were incubated with the lentiviral particles in mediacontaining 6 μg/ml Polybrene. After 48 h the culture media was replacedwith fresh media and GFP expression was analyzed. Three days afterinfection, neurons expressing TFEB-GFP or GFP-LC3 fusion proteins weretreated with 200 nM rapamycin or 100 nM Torin1 for 18-24 h as indicated.Neuronal cultures were then fixed in 4% paraformaldehyde and immune-labeled with anti-LAMP1 or anti-p62 antibodies, followed by fluorescenceimage acquisition and analysis.

11

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

Fluorescence images acquisition and quantitationFluorescence images were captured using an inverted Nikon EclipseTE-2000 microscope, a Nikon Ti-E inverted microscope with NikonImaging Systems (NIS)-Elements, or a Revolve microscope with Olympusoptics and 8MP color camera (Echo Laboratories). Images were analyzedusing ImageJ software (National Institutes of Health), after backgroundsubtraction and threshold detection. For LAMP1 fluorescence signalquantitation, z-stack images were acquired using a Nikon Ti-E invertedmicroscope at 60× (CFI Plan APO VC 60× NA 1.4 Oil) and focused usingthe extended depth of focus (EDF) module of the Nikon Elements software.All images within each experiment were acquired at the same microscopesettings. A minimum of 50 neurons were analyzed per cell line in at leastfour high power fields (HPF) per experiment. The percentage of neuronswith TFEB-GFP-positive nuclei was calculated as the number of cells withTFEB-GFP fluorescence signal that colocalized with nuclear DAPI, dividedby the number of GFP-positive cells in the same vision field. The ratio ofnuclear/cytoplasmic TFEB-GFP signal was quantified using Nikon Elementssoftware. The nuclei were defined based on DAPI staining using spotdetection; TFEB-GFP intensity was quantified in areas overlapping versusnon-overlapping with DAPI and expressed as a ratio. For GFP-LC3 punctaanalysis, images were acquired using a Nikon Ti-E inverted microscope (CFIPlanAPOVC20×NA0.75WD1 mm) or a Revolvemicroscope (UCPLFLN40× or 20×), and analyzed using ImageJ software after backgroundsubtraction and threshold detection. The number and fluorescence intensityof GFP-LC3 puncta were analyzed in at least 100 neurons per group in 3-4fields in each experiment and normalized to the number of GFP-expressingcells in the same field of vision. GFP-LC3 puncta-LAMP1 fluorescencesignal colocalization was determined by measuring the percentage of overlapin the area of integrated intensity for the two markers, which corresponds tothe colocalization coefficient (Manders et al., 1993).

Western blot analysisiPSC-derived NPCs and neurons were lysed in RIPA buffer supplementedwith protease inhibitor (Roche) and phosphatase inhibitors (Pierce), ordirectly lysed in Laemmli buffer (Bio-Rad), followed by sonication. Proteinextracts were denatured in loading buffer at 95°C for 5 min before loadingonto 4-20% polyacrylamide gels (Bio-Rad) and analyzed byelectrophoresis. The proteins were then transferred to PVDF ornitrocellulose membranes and incubated overnight with the primaryantibodies at 4°C, followed by incubation with the corresponding HRP-conjugated secondary antibody for 1 h. Membranes were developed usingSuperSignal West Pico PLUS Chemiluminescent Substrate (Thermo FisherScientific) and imaged using a Chemidoc imager and Imagelab software(Bio-Rad). Primary antibodies used were: anti-mTOR, anti-phospho-mTOR-Ser2448, anti-S6 Ribosomal Protein, anti-phospho-S6-Ser235/236, anti-phospho-4EBP1-Thr37/46, anti-LAMP1, anti-p62 (all at1:1000), anti-NBR1 (Cell Signaling Technology, 9891, 1:1000), anti-TFEB (MyBioSource, MBS855552, 1:1000), anti-phosphoTFEB-Ser142(Millipore, ABE1971, 1:500), anti-ubiquitin (Cell Signaling Technology,3933, 1:1000), and anti-β-actin (Sigma-Aldrich, 1:4000).

Immunoprecipitation (IP)NPCs were lysed in IP lysis buffer, followed by pre-clearing the lysate for1 h with Protein A/G/Agarose Beads (Thermo Fisher Scientific, 20421).The cell lysates were then incubated overnight with the primary antibody at4°C. The following day, Protein A/G Agarose Beads were added andincubated with rotation for 2 h. Beads were then washed and protein waseluted for western blotting. The primary antibody used was anti-phospho-TFEB-Ser142 at 1:200 dilution.

Real-time PCRFor gene expression analysis, neurons were cultured in 12-well plates andtreated with rapamycin or Torin1 as described above. mRNAwas extractedusing the RNAeasy kit (Qiagen). cDNA was synthesized using iScriptcDNA synthesis kit (Bio-Rad). Gene expression was determined byquantitative PCR (7900 HT, Applied Biosystems) in duplicate wells usingSYBR Green PCR Master Mix (Thermo Fisher Scientific). Relative gene

expression was normalized to the corresponding value of GAPDHexpression, and the fold changes relative to the control values within thesame experiment were determined. The sequences of the primers are listedin Table S1.

TFEB-GFP H4 neuroglioma cellsThe neuroglioma cell line H4 (ATCC), was infected with TFEB-GFPlentiviral vector (previously described) (Awad et al., 2015), and selected forstable TFEB-GFP expression using 1 µg/ml Puromycin. Cells weremaintained in DMEM supplemented with 10% fetal bovine serum and1% penicillin/streptomycin. Cells were treated with 1 mM conduritolB-epoxide (CBE) (Sigma-Aldrich) for 48 h alone or with the proteasomeinhibitor Clasto-lactacystin β-lactone as indicated. Cells were either lysedfor western blot analysis or imaged for GFP fluorescence expression.

Statistical analysisData are mean±s.e.m and analyzed using one-way ANOVA followed byTukey’s or Sidak’s post-test to determine statistical differences betweenmultiple groups. Two-tailed unpaired Student’s t-tests were used forcomparison between two groups when appropriate. P-values <0.05 wereconsidered statistically significant. The confidence level for significancewas 95%. Data were analyzed using Prism software version 7.0a (GraphPadSoftware).

Competing interestsThe authors declare no competing or financial interests.

Author contributionsConceptualization: R.A.B., M.M.L., R.A.F., O.A.; Methodology: R.A.B., A.V., M.M.L.,R.A.F., O.A.; Validation: R.A.B., A.V., M.P.S., J.A.T., O.A.; Formal analysis: M.M.L.,O.A.; Investigation: R.A.B., A.V., M.P.S., J.A.T., O.A.; Resources: T.J.K., M.A.J.,M.M.L., R.A.F., O.A.; Data curation: R.A.B., A.V., O.A.; Writing - original draft: R.A.B.,O.A.; Writing - review & editing: R.A.B., J.A.T., T.J.K., M.A.J., M.M.L., R.A.F., O.A.;Visualization: R.A.B., A.V., M.P.S., O.A.; Supervision: M.M.L., R.A.F., O.A.; Projectadministration: O.A.; Funding acquisition: R.A.F., O.A.

FundingThis work was supported by grants from the Michael J. Fox Foundation forParkinson’s Research (Priority Target Award 11577 to O.A.), theMaryland StemCellResearch Fund (2018-MSCRFD-4246 to R.A.F.) and the Children’s GaucherResearch Fund (to R.A.F.).

Supplementary informationSupplementary information available online athttp://dmm.biologists.org/lookup/doi/10.1242/dmm.038596.supplemental

ReferencesAerts, J. M. F. G., Hollak, C. E. M., Boot, R. G., Groener, J. E. M. and Maas, M.

(2006). Substrate reduction therapy of glycosphingolipid storage disorders.J. Inherit. Metab. Dis. 29, 449-456. doi:10.1007/s10545-006-0272-5

Almeida, M. R. (2012). Glucocerebrosidase involvement in Parkinson disease andother synucleinopathies. Front. Neurol. 3, 65. doi:10.3389/fneur.2012.00065

Antikainen, H., Driscoll, M., Haspel, G. and Dobrowolski, R. (2017). TOR-mediated regulation of metabolism in aging. Aging Cell 16, 1219-1233. doi:10.1111/acel.12689

Ashe, K.M., Bangari, D., Li, L., Cabrera-Salazar, M. A., Bercury, S. D., Nietupski,J. B., Cooper, C. G. F., Aerts, J. M. F. G., Lee, E. R., Copeland, D. P. et al.(2011). Iminosugar-based inhibitors of glucosylceramide synthase increase brainglycosphingolipids and survival in a mouse model of Sandhoff disease. PLoSONE 6, e21758. doi:10.1371/journal.pone.0021758

Awad, O., Sarkar, C., Panicker, L. M., Miller, D., Zeng, X., Sgambato, J. A.,Lipinski, M. M. and Feldman, R. A. (2015). Altered TFEB-mediated lysosomalbiogenesis in Gaucher disease iPSC-derived neuronal cells. Hum. Mol. Genet.24, 5775-5788. doi:10.1093/hmg/ddv297

Awad, O., Panicker, L. M., Deranieh, R. M., Srikanth, M. P., Brown, R. A., Voit, A.,Peesay, T., Park, T. S., Zambidis, E. T. and Feldman, R. A. (2017). Altereddifferentiation potential of Gaucher’s disease iPSC neuronal progenitors due toWnt/beta-catenin downregulation. Stem Cell Rep. 9, 1853-1867. doi:10.1016/j.stemcr.2017.10.029

Bartolomeo, R., Cinque, L., DE Leonibus, C., Forrester, A., Salzano, A. C.,Monfregola, J., DEGennaro, E., Nusco, E., Azario, I., Lanzara, C. et al. (2017).mTORC1 hyperactivation arrests bone growth in lysosomal storage disorders bysuppressing autophagy. J. Clin. Invest. 127, 3717-3729. doi:10.1172/JCI94130

12

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

Bellettato, C. M. and Scarpa, M. (2010). Pathophysiology of neuropathic lysosomalstorage disorders. J. Inherit. Metab. Dis. 33, 347-362. doi:10.1007/s10545-010-9075-9

Bendikov-Bar, I. and Horowitz, M. (2012). Gaucher disease paradigm: from ERADto comorbidity. Hum. Mutat. 33, 1398-1407. doi:10.1002/humu.22124

Bento, C. F., Ashkenazi, A., Jimenez-Sanchez, M. andRubinsztein, D. C. (2016).The Parkinson’s disease-associated genes ATP13A2 and SYT11 regulateautophagy via a common pathway. Nat. Commun. 7, 11803. doi:10.1038/ncomms11803

Brady, R. O., Kanfer, J. N., Bradley, R. M. and Shapiro, D. (1966). Demonstrationof a deficiency of glucocerebroside-cleaving enzyme in Gaucher’s disease.J. Clin. Invest. 45, 1112-1115. doi:10.1172/JCI105417

Brockmann, K. and Berg, D. (2014). The significance of GBA for Parkinson’sdisease. J. Inherit. Metab. Dis. 37, 643-648. doi:10.1007/s10545-014-9714-7

Cabrera-Salazar, M. A., Deriso, M., Bercury, S. D., Li, L., Lydon, J. T., Weber,W.,Pande, N., Cromwell, M. A., Copeland, D., Leonard, J. et al. (2012). Systemicdelivery of a glucosylceramide synthase inhibitor reduces CNS substrates andincreases lifespan in a mouse model of type 2 Gaucher disease. PLoS ONE 7,e43310. doi:10.1371/journal.pone.0043310

Cox, T. M. (2010). Gaucher disease: clinical profile and therapeutic developments.Biologics 4, 299-313. doi:10.2147/BTT.S7582

Dasgupta, N., Xu, Y.-H., Li, R., Peng, Y., Pandey, M. K., Tinch, S. L., Liou, B.,Inskeep, V., Zhang, W., Setchell, K. D. R. et al. (2015). Neuronopathic Gaucherdisease: dysregulated mRNAs and miRNAs in brain pathogenesis and effects ofpharmacologic chaperone treatment in a mouse model. Hum. Mol. Genet. 24,7031-7048. doi:10.1093/hmg/ddv404

Decressac, M., Mattsson, B., Weikop, P., Lundblad, M., Jakobsson, J. andBjorklund, A. (2013). TFEB-mediated autophagy rescues midbrain dopamineneurons from alpha-synuclein toxicity. Proc. Natl. Acad. Sci. USA 110,E1817-E1826. doi:10.1073/pnas.1305623110

Dibble, C. C. and Cantley, L. C. (2015). Regulation of mTORC1 by PI3K signaling.Trends Cell Biol. 25, 545-555. doi:10.1016/j.tcb.2015.06.002

Dunlop, E. A. and Tee, A. R. (2014). mTOR and autophagy: a dynamic relationshipgoverned by nutrients and energy. Semin. Cell Dev. Biol. 36, 121-129. doi:10.1016/j.semcdb.2014.08.006

Farfel-Becker, T., Vitner, E. B., Kelly, S. L., Bame, J. R., Duan, J., Shinder, V.,Merrill, A. H., Jr, Dobrenis, K. and Futerman, A. H. (2014). Neuronalaccumulation of glucosylceramide in a mouse model of neuronopathic Gaucherdisease leads to neurodegeneration. Hum. Mol. Genet. 23, 843-854. doi:10.1093/hmg/ddt468

Gan-Or, Z., Dion, P. A. and Rouleau, G. A. (2015). Genetic perspective on the roleof the autophagy-lysosome pathway in Parkinson disease. Autophagy 11,1443-1457. doi:10.1080/15548627.2015.1067364

Ikenoue, T., Hong, S. and Inoki, K. (2009). Monitoring mammalian target ofrapamycin (mTOR) activity. Methods Enzymol. 452, 165-180. doi:10.1016/S0076-6879(08)03611-2

Jaishy, B. and Abel, E. D. (2016). Lipids, lysosomes, and autophagy. J. Lipid Res.57, 1619-1635. doi:10.1194/jlr.R067520

Jmoudiak, M. and Futerman, A. H. (2005). Gaucher disease: pathologicalmechanisms and modern management. Br. J. Haematol. 129, 178-188. doi:10.1111/j.1365-2141.2004.05351.x

Kaminsky, P., Belmatoug, N. and Billette de Villemeur, T. (1996). [Gaucherdisease]. Presse Med. 25, 108-112.

Kesidou, E., Lagoudaki, R., Touloumi, O., Poulatsidou, K. N. and Simeonidou,C. (2013). Autophagy and neurodegenerative disorders. Neural Regen. Res. 8,2275-2283.

Kilpatrick, K., Zeng, Y., Hancock, T. and Segatori, L. (2015). Genetic andchemical activation of TFEB mediates clearance of aggregated alpha-synuclein.PLoS ONE 10, e0120819. doi:10.1371/journal.pone.0120819

Kinghorn, K. J., Gronke, S., Castillo-Quan, J. I., Woodling, N. S., Li, L., Sirka, E.,Gegg, M., Mills, K., Hardy, J., Bjedov, I. et al. (2016). A Drosophila model ofneuronopathicGaucher disease demonstrates lysosomal-autophagic defects andaltered mTOR signalling and is functionally rescued by rapamycin. J. Neurosci.36, 11654-11670. doi:10.1523/JNEUROSCI.4527-15.2016

Kinghorn, K. J., Asghari, A. M. and Castillo-Quan, J. I. (2017). The emerging roleof autophagic-lysosomal dysfunction in Gaucher disease and Parkinson’sdisease. Neural Regen. Res. 12, 380-384. doi:10.4103/1673-5374.202934

Kirkin, V., Lamark, T., Johansen, T. and Dikic, I. (2009). NBR1 cooperates withp62 in selective autophagy of ubiquitinated targets.Autophagy 5, 732-733. doi:10.4161/auto.5.5.8566

Klein, C. and Westenberger, A. (2012). Genetics of Parkinson’s disease. ColdSpring Harb. Perspect. Med. 2, a008888. doi:10.1101/cshperspect.a008888

Klionsky, D. J., Abdalla, F. C., Abeliovich, H., Abraham, R. T., Acevedo-Arozena, A., Adeli, K., Agholme, L., Agnello, M., Agostinis, P., Aguirre-Ghiso,J. A., et al. (2012). Guidelines for the use and interpretation of assays formonitoring autophagy. Autophagy 8, 445-544. doi:10.4161/auto.19496

Lachmann, R. H. and Platt, F. M. (2001). Substrate reduction therapy forglycosphingolipid storage disorders. Expert Opin Investig. Drugs 10, 455-466.doi:10.1517/13543784.10.3.455

Laplante, M. and Sabatini, D. M. (2009). mTOR signaling at a glance. J. Cell Sci.122, 3589-3594. doi:10.1242/jcs.051011

Laplante, M. and Sabatini, D. M. (2013). Regulation of mTORC1 and its impact ongene expression at a glance. J. Cell Sci. 126, 1713-1719. doi:10.1242/jcs.125773

Li, L., Friedrichsen, H. J., Andrews, S., Picaud, S., Volpon, L., Ngeow, K.,Berridge, G., Fischer, R., Borden, K. L. B., Filippakopoulos, P. et al. (2018). ATFEB nuclear export signal integrates amino acid supply and glucose availability.Nat. Commun. 9, 2685. doi:10.1038/s41467-018-04849-7

Lim, J. A., Li, L., Shirihai, O. S., Trudeau, K. M., Puertollano, R. and Raben, N.(2017). Modulation of mTOR signaling as a strategy for the treatment of Pompedisease. EMBO Mol. Med. 9, 353-370. doi:10.15252/emmm.201606547

Liou, B., Peng, Y., Li, R., Inskeep, V., Zhang, W., Quinn, B., Dasgupta, N.,Blackwood, R., Setchell, K. D. R., Fleming, S. et al. (2016). Modulatingryanodine receptors with dantrolene attenuates neuronopathic phenotype inGaucher disease mice. Hum. Mol. Genet. 25, 5126-5141. doi:10.1093/hmg/ddw322

Lipton, J. O. and Sahin, M. (2014). The neurology of mTOR. Neuron 84, 275-291.doi:10.1016/j.neuron.2014.09.034

Magalhaes, J., Gegg, M. E., Migdalska-Richards, A., Doherty, M. K., Whitfield,P. D. and Schapira, A. H. V. (2016). Autophagic lysosome reformationdysfunction in glucocerebrosidase deficient cells: relevance to Parkinsondisease. Hum. Mol. Genet. 25, 3432-3445. doi:10.1093/hmg/ddw185

Manders, E. M., Verbeek, F. J. and Aten, J. A. (1993). Measurement of co-localization of objects in dual-colour confocal images. J. Microsc. 169, 375-382.doi:10.1111/j.1365-2818.1993.tb03313.x

Martina, J. A., Chen, Y., Gucek, M. and Puertollano, R. (2012). MTORC1functions as a transcriptional regulator of autophagy by preventing nucleartransport of TFEB. Autophagy 8, 903-914. doi:10.4161/auto.19653

Martina, J. A., Diab, H. I., Brady, O. A. and Puertollano, R. (2016). TFEB andTFE3 are novel components of the integrated stress response. EMBO J. 35,479-495. doi:10.15252/embj.201593428

Mazzulli, J. R., Zunke, F., Tsunemi, T., Toker, N. J., Jeon, S., Burbulla, L. F.,Patnaik, S., Sidransky, E., Marugan, J. J., Sue, C. M. et al. (2016). Activation ofbeta-glucocerebrosidase reduces pathological alpha-synuclein and restoreslysosomal function in Parkinson’s patient midbrain neurons. J. Neurosci. 36,7693-7706. doi:10.1523/JNEUROSCI.0628-16.2016

Menon, D., Salloum, D., Bernfeld, E., Gorodetsky, E., Akselrod, A., Frias, M. A.,Sudderth, J., Chen, P.-H., Deberardinis, R. and Foster, D. A. (2017). Lipidsensing by mTOR complexes via de novo synthesis of phosphatidic acid. J. Biol.Chem. 292, 6303-6311. doi:10.1074/jbc.M116.772988

Menzies, F. M., Fleming, A., Caricasole, A., Bento, C. F., Andrews, S. P.,Ashkenazi, A., Fullgrabe, J., Jackson, A., Jimenez Sanchez, M., Karabiyik, C.et al. (2017). Autophagy and neurodegeneration: pathogenic mechanisms andtherapeutic opportunities. Neuron 93, 1015-1034. doi:10.1016/j.neuron.2017.01.022

Mignot, C., Gelot, A. and DE Villemeur, T. B. (2013). Gaucher disease. Handb.Clin. Neurol. 113, 1709-1715. doi:10.1016/B978-0-444-59565-2.00040-X

Mistry, P. K., Lukina, E., BEN Turkia, H., Shankar, S. P., Baris, H., Ghosn, M.,Mehta, A., Packman, S., Pastores, G., Petakov, M. et al. (2017). Outcomes after18 months of eliglustat therapy in treatment-naïve adults with Gaucher diseasetype 1: The phase 3 ENGAGE trial. Am. J. Hematol. 92, 1170-1176. doi:10.1002/ajh.24877

Napolitano, G. and Ballabio, A. (2016). TFEB at a glance. J. Cell Sci. 129,2475-2481. doi:10.1242/jcs.146365

Natoli, T. A., Smith, L. A., Rogers, K. A., Wang, B., Komarnitsky, S., Budman, Y.,Belenky, A., Bukanov, N. O., Dackowski, W. R., Husson, H. et al. (2010).Inhibition of glucosylceramide accumulation results in effective blockade ofpolycystic kidney disease in mouse models. Nat. Med. 16, 788-792. doi:10.1038/nm.2171

Orvisky, E., Sidransky, E., Mckinney, C. E., Lamarca, M. E., Samimi, R.,Krasnewich, D., Martin, B. M. and Ginns, E. I. (2000). Glucosylsphingosineaccumulation in mice and patients with type 2 Gaucher disease begins early ingestation. Pediatr. Res. 48, 233-237. doi:10.1203/00006450-200008000-00018

Osellame, L. D., Rahim, A. A., Hargreaves, I. P., Gegg, M. E., Richard-Londt, A.,Brandner, S., Waddington, S. N., Schapira, A. H. V. and Duchen, M. R. (2013).Mitochondria and quality control defects in a mouse model of Gaucher disease–links to Parkinson’s disease. Cell Metab. 17, 941-953. doi:10.1016/j.cmet.2013.04.014

Panicker, L. M., Miller, D., Park, T. S., Patel, B., Azevedo, J. L., Awad, O.,Masood, M. A., Veenstra, T. D., Goldin, E., Stubblefield, B. K. et al. (2012).Induced pluripotent stem cell model recapitulates pathologic hallmarks ofGaucher disease. Proc. Natl. Acad. Sci. USA 109, 18054-18059. doi:10.1073/pnas.1207889109

Puertollano, R., Ferguson, S. M., Brugarolas, J. and Ballabio, A. (2018). Thecomplex relationship between TFEB transcription factor phosphorylation andsubcellular localization. EMBO J. 37, e98804. doi:10.15252/embj.201798804

Puyal, J., Ginet, V., Grishchuk, Y., Truttmann, A. C. and Clarke, P. G. H. (2012).Neuronal autophagy as a mediator of life and death: contrasting roles in chronicneurodegenerative and acute neural disorders. Neuroscientist 18, 224-236.doi:10.1177/1073858411404948

13

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms

Rabanal-Ruiz, Y. and Korolchuk, V. I. (2018). mTORC1 and nutrient homeostasis:the central role of the lysosome. Int. J. Mol. Sci. 19, 818. doi:10.3390/ijms19030818

Rabanal-Ruiz, Y., Otten, E. G. and Korolchuk, V. I. (2017). mTORC1 as the maingateway to autophagy. Essays Biochem. 61, 565-584. doi:10.1042/EBC20170027

Rao, L., Sha, Y. and Eissa, N. T. (2017). The E3 ubiquitin ligase STUB1 regulatesautophagy and mitochondrial biogenesis by modulating TFEB activity. Mol. CellOncol. 4, e1372867. doi:10.1080/23723556.2017.1372867

Ricoult, S. J. H. and Manning, B. D. (2013). The multifaceted role of mTORC1 inthe control of lipid metabolism. EMBO Rep. 14, 242-251. doi:10.1038/embor.2013.5

Roshan Lal, T. and Sidransky, E. (2017). The spectrum of neurologicalmanifestations associated with Gaucher disease. Diseases 5, 10. doi:10.3390/diseases5010010

Sardi, S. P., Viel, C., Clarke, J., Treleaven, C. M., Richards, A. M., Park, H.,Olszewski, M. A., Dodge, J. C., Marshall, J., Makino, E. et al. (2017).Glucosylceramide synthase inhibition alleviates aberrations in synucleinopathymodels. Proc. Natl. Acad. Sci. USA 114, 2699-2704. doi:10.1073/pnas.1616152114

Sardiello, M. and Ballabio, A. (2009). Lysosomal enhancement: a CLEAR answerto cellular degradative needs. Cell Cycle 8, 4021-4022. doi:10.4161/cc.8.24.10263

Sardiello, M., Palmieri, M., DI Ronza, A., Medina, D. L., Valenza, M., Gennarino,V. A., DI Malta, C., Donaudy, F., Embrione, V., Polishchuk, R. S. et al. (2009).A gene network regulating lysosomal biogenesis and function. Science 325,473-477. doi:10.1126/science.1174447

Saxton, R. A. and Sabatini, D. M. (2017a). mTOR signaling in growth, metabolism,and disease. Cell 169, 361-371. doi:10.1016/j.cell.2017.03.035

Saxton, R. A. and Sabatini, D. M. (2017b). mTOR signaling in growth, metabolism,and disease. Cell 168, 960-976. doi:10.1016/j.cell.2017.02.004

Schapira, A. H. V. (2015). Glucocerebrosidase and Parkinson disease: Recentadvances. Mol. Cell. Neurosci. 66, 37-42. doi:10.1016/j.mcn.2015.03.013

Schondorf, D. C., Aureli, M., Mcallister, F. E., Hindley, C. J., Mayer, F., Schmid,B., Sardi, S. P., Valsecchi, M., Hoffmann, S., Schwarz, L. K. et al. (2014). iPSC-derived neurons from GBA1-associated Parkinson’s disease patients showautophagic defects and impaired calcium homeostasis. Nat. Commun. 5, 4028.doi:10.1038/ncomms5028

Schueler, U. H., Kolter, T., Kaneski, C. R., Blusztajn, J. K., Herkenham, M.,Sandhoff, K. and Brady, R. O. (2003). Toxicity of glucosylsphingosine(glucopsychosine) to cultured neuronal cells: a model system for assessingneuronal damage in Gaucher disease type 2 and 3. Neurobiol. Dis. 14, 595-601.doi:10.1016/j.nbd.2003.08.016

Sengupta, S., Peterson, T. R. and Sabatini, D. M. (2010). Regulation of the mTORcomplex 1 pathway by nutrients, growth factors, and stress.Mol. Cell 40, 310-322.doi:10.1016/j.molcel.2010.09.026

Settembre, C. and Ballabio, A. (2014a). Lysosomal adaptation: how the lysosomeresponds to external cues. Cold Spring Harb. Perspect. Biol. 6, a016907. doi:10.1101/cshperspect.a016907

Settembre, C. and Ballabio, A. (2014b). Lysosome: regulator of lipid degradationpathways. Trends Cell Biol. 24, 743-750. doi:10.1016/j.tcb.2014.06.006

Settembre, C., Di Malta, C., Polito, V. A., Garcia Arencibia, M., Vetrini, F., Erdin,S., Erdin, S. U., Huynh, T., Medina, D., Colella, P. et al. (2011). TFEB linksautophagy to lysosomal biogenesis. Science 332, 1429-1433. doi:10.1126/science.1204592

Settembre, C., Zoncu, R., Medina, D. L., Vetrini, F., Erdin, S., Erdin, S. U.,Huynh, T., Ferron, M., Karsenty, G., Vellard, M. C. et al. (2012). A lysosome-to-nucleus signalling mechanism senses and regulates the lysosome via mTOR andTFEB. EMBO J. 31, 1095-1108. doi:10.1038/emboj.2012.32

Sgambato, J. A., Park, T. S., Miller, D., Panicker, L. M., Sidransky, E., Lun, Y.,Awad, O., Bentzen, S. M., Zambidis, E. T. and Feldman, R. A. (2015). Gaucher

disease-induced pluripotent stem cells display decreased erythroid potential andaberrant myelopoiesis. Stem Cells Transl. Med. 4, 878-886. doi:10.5966/sctm.2014-0213

Sha, Y., Rao, L., Settembre, C., Ballabio, A. and Eissa, N. T. (2017). STUB1regulates TFEB-induced autophagy-lysosome pathway. EMBO J. 36, 2544-2552.doi:10.15252/embj.201796699

Shen, W., Henry, A. G., Paumier, K. L., Li, L., Mou, K., Dunlop, J., Berger, Z. andHirst,W. D. (2014). Inhibition of glucosylceramide synthase stimulates autophagyflux in neurons. J. Neurochem. 129, 884-894. doi:10.1111/jnc.12672

Stein, V. M., Crooks, A., Ding, W., Prociuk, M., O’donnell, P., Bryan, C., Sikora,T., Dingemanse, J., Vanier, M. T.,Walkley, S. U. et al. (2012). Miglustat improvespurkinje cell survival and alters microglial phenotype in feline Niemann-Pickdisease type C. J. Neuropathol. Exp. Neurol. 71, 434-448. doi:10.1097/NEN.0b013e31825414a6

Sun, Y. andGrabowski, G. A. (2010). Impaired autophagosomes and lysosomes inneuronopathic Gaucher disease. Autophagy 6, 648-649. doi:10.4161/auto.6.5.12047

Thoreen, C. C. and Sabatini, D. M. (2009). Rapamycin inhibits mTORC1, but notcompletely. Autophagy 5, 725-726. doi:10.4161/auto.5.5.8504

Thoreen, C. C., Kang, S. A., Chang, J. W., Liu, Q., Zhang, J., Gao, Y., Reichling,L. J., Sim, T., Sabatini, D. M. and Gray, N. S. (2009). An ATP-competitivemammalian target of rapamycin inhibitor reveals rapamycin-resistant functions ofmTORC1. J. Biol. Chem. 284, 8023-8032. doi:10.1074/jbc.M900301200

Vidal-Donet, J. M., Carcel-Trullols, J., Casanova, B., Aguado, C. and Knecht, E.(2013). Alterations in ROS activity and lysosomal pH account for distinct patternsof macroautophagy in LINCL and JNCL fibroblasts. PLoS ONE 8, e55526. doi:10.1371/journal.pone.0055526

Ward, C., Martinez-Lopez, N., Otten, E. G., Carroll, B., Maetzel, D., Singh, R.,Sarkar, S. and Korolchuk, V. I. (2016). Autophagy, lipophagy and lysosomal lipidstorage disorders. Biochim. Biophys. Acta 1861, 269-284. doi:10.1016/j.bbalip.2016.01.006

Weiss, K., Gonzalez, A. N., Lopez, G., Pedoeim, L., Groden, C. and Sidransky,E. (2015). The clinical management of Type 2 Gaucher disease. Mol. Genet.Metab. 114, 110-122. doi:10.1016/j.ymgme.2014.11.008

Wong, K., Sidransky, E., Verma, A., Mixon, T., Sandberg, G. D., Wakefield, L. K.,Morrison, A., Lwin, A., Colegial, C., Allman, J. M. et al. (2004). Neuropathologyprovides clues to the pathophysiology of Gaucher disease.Mol. Genet. Metab. 82,192-207. doi:10.1016/j.ymgme.2004.04.011

Xu, Y.-H., Barnes, S., Sun, Y. andGrabowski, G. A. (2010). Multi-system disordersof glycosphingolipid and ganglioside metabolism. J. Lipid Res. 51, 1643-1675.doi:10.1194/jlr.R003996

Yao, Y., Jones, E. and Inoki, K. (2017). Lysosomal Regulation of mTORC1 byAmino Acids in Mammalian Cells. Biomolecules 7, 51. doi:10.3390/biom7030051

Young, N. P., Kamireddy, A., VAN Nostrand, J. L., Eichner, L. J., Shokhirev,M. N., Dayn, Y. and Shaw, R. J. (2016). AMPK governs lineage specificationthrough Tfeb-dependent regulation of lysosomes. Genes Dev. 30, 535-552.doi:10.1101/gad.274142.115

Zhang, H., Duan, C. and Yang, H. (2015). Defective autophagy in Parkinson’sdisease: lessons from genetics.Mol. Neurobiol. 51, 89-104. doi:10.1007/s12035-014-8787-5

Zheng, Y. and Jiang, Y. (2015). mTOR inhibitors at a glance.Mol. Cell. Pharmacol.7, 15-20.

Zhou, J., Tan, S.-H., Nicolas, V., Bauvy, C., Yang, N.-D., Zhang, J., Xue, Y.,Codogno, P. and Shen, H.-M. (2013). Activation of lysosomal function in thecourse of autophagy via mTORC1 suppression and autophagosome-lysosomefusion. Cell Res. 23, 508-523. doi:10.1038/cr.2013.11

Zimran, A., Goldblatt, J. and Szer, J. (2018). Should eliglustat be first line therapyfor patients with type 1 Gaucher disease? Definitions of safety and efficacy. BloodCells Mol. Dis. 68, 14-16. doi:10.1016/j.bcmd.2017.09.003

14

RESEARCH ARTICLE Disease Models & Mechanisms (2019) 12, dmm038596. doi:10.1242/dmm.038596

Disea

seModels&Mechan

isms