RESEARCH ARTICLE

Lysosome enlargement during inhibition of the lipid kinasePIKfyve proceeds through lysosome coalescenceChristopher H. Choy1,2,*, Golam Saffi1,2,*, Matthew A. Gray1, Callen Wallace3, Roya M. Dayam1,2,Zhen-Yi A. Ou1, Guy Lenk4, Rosa Puertollano5, Simon C. Watkins3 and Roberto J. Botelho1,2,‡

ABSTRACTLysosomes receive and degrade cargo from endocytosis,phagocytosis and autophagy. They also play an important role insensing and instructing cells on their metabolic state. The lipidkinase PIKfyve generates phosphatidylinositol-3,5-bisphosphate tomodulate lysosome function. PIKfyve inhibition leads to impaireddegradative capacity, ion dysregulation, abated autophagic flux and amassive enlargement of lysosomes. Collectively, this leads tovarious physiological defects, including embryonic lethality,neurodegeneration and overt inflammation. The reasons for suchdrastic lysosome enlargement remain unclear. Here, we examinedwhether biosynthesis and/or fusion-fission dynamics contribute toswelling. First, we show that PIKfyve inhibition activates TFEB, TFE3and MITF, enhancing lysosome gene expression. However, thisdid not augment lysosomal protein levels during acute PIKfyveinhibition, and deletion of TFEB and/or related proteins did not impairlysosome swelling. Instead, PIKfyve inhibition led to fewer butenlarged lysosomes, suggesting that an imbalance favouringlysosome fusion over fission causes lysosome enlargement.Indeed, conditions that abated fusion curtailed lysosome swelling inPIKfyve-inhibited cells.

KEY WORDS: Transcription factors, Phosphoinositides,Lysosomes, Organelles, Fusion, Fission

INTRODUCTIONLysosomes are organelles responsible for the degradation of a largevariety of molecules that originate from within and from outsidecells. Lysosomes accomplish this by fusing with cargo-containingorganelles such as autophagosomes, Golgi-derived vesicles,endosomes and phagosomes (Luzio et al., 2007, 2009; Schwakeet al., 2013). To enable their degradative capacity, lysosomes areequipped with a plethora of hydrolytic enzymes that digest proteins,lipids, nucleic acids and carbohydrates. In addition, the vacuolarH+-pump ATPase generates a highly acidic lumen optimal forhydrolytic activity (Mindell, 2012; Schwake et al., 2013; Xiong andZhu, 2016; Luzio et al., 2007). Nonetheless, the function of

lysosomes is not limited to molecular degradation. Among otherroles, lysosomes are an important storage organelle, they mediateantigen presentation, they undergo exocytosis to repair membranedamage, and are a genuine signalling platform that senses andinforms the cell of its nutrient and stress conditions (Medina et al.,2015; Settembre and Ballabio, 2014; Sancak et al., 2010; Delamarreet al., 2005; Xiong and Zhu, 2016; Reddy et al., 2001; Settembreet al., 2012; Roczniak-Ferguson et al., 2012).

Strikingly, the lysosome is now considered to be a highlydynamic and heterogeneous organelle. Not only can individual cellshost tens to hundreds of lysosomes, lysosomes themselves arehighly heterogeneous in their pH (Johnson et al., 2016), subcellularposition (Li et al., 2016a; Pu et al., 2015; Bagshaw et al., 2006) andmorphology (Swanson et al., 1987; Saric et al., 2016; Vyas et al.,2007). Lastly, lysosomes can adapt to their environment. Forexample, in macrophages and dendritic cells, lysosomes aretransformed from individual puncta into a highly tubular network(Saric et al., 2016; Vyas et al., 2007; Swanson et al., 1987). Inaddition, nutrient depletion, protein aggregation and phagocytosisactivate a family of transcription factors that can enhance expressionof lysosome genes and adjust lysosome activity (Gray et al., 2016;Pastore et al., 2016; Sardiello et al., 2009; Tsunemi et al., 2012;Medina et al., 2011; Settembre et al., 2012; Roczniak-Fergusonet al., 2012). This is best understood for the related transcriptionfactors, TFEB and TFE3, which are recruited to the surface oflysosomes and subject to phosphorylation by various kinases,including mTORC1 to modulate their nucleocytoplasmic transport(Li et al., 2016b; Martina et al., 2012; Peña-Llopis et al., 2011;Martina et al., 2014, 2016; Roczniak-Ferguson et al., 2012).

Given the above, lysosomes are highly regulated organelles. Amaster modulator of lysosomes is the lipid kinase PIKfyve, whichconverts phosphatidylinositol-3-phosphate [PtdIns(3)P] intophosphatidylinositol-3,5-bisphosphate [PtdIns(3,5)P2] (Sbrissaet al., 1999). Inactivation of PIKfyve or of its regulators, Vac14/ArPIKfyve and Fig4/Sac3, causes a variety of physiologicalproblems, including embryonic lethality, neurodegeneration andimmune malfunction (Cai et al., 2013; Chow et al., 2007; Jin et al.,2008; Ho et al., 2012; Mccartney et al., 2014; Sbrissa et al., 2007;Ferguson et al., 2010; Min et al., 2014; Lenk et al., 2016; Ikonomovet al., 2011). At the cellular level, these defects relate to impairedautophagic flux, altered delivery to lysosomes, dysregulation oflysosomal Ca2+ transport and massive enlargement of lysosomes(Ferguson et al., 2009; Dong et al., 2010; de Lartigue et al.,2009; Ferguson et al., 2010; Kim et al., 2014, 2016; Ho et al., 2012;Mccartney et al., 2014). Remarkably, it remains unclear howlysosomes enlarge in cells with defective PIKfyve activity. It hasbeen suggested that PIKfyve, or loss of the yeast orthologue Fab1impairs recycling from endosomes, lysosomes or from the yeastvacuole, thus causing an influx of membrane into lysosomes andtheir enlargement (Rutherford et al., 2006; Bryant et al., 1998; DoveReceived 28 November 2017; Accepted 10 April 2018

1Department of Chemistry and Biology, Ryerson University, Toronto, ON, Canada,M5B2K3. 2The Graduate Program in Molecular Science, Ryerson University,Toronto, ON, Canada, M5B2K3. 3Department of Cell Biology, University ofPittsburgh, Pittsburgh, PA 15261, USA. 4Department of HumanGenetics, Universityof Michigan, Ann Arbor, MI 48109, USA. 5Cell Biology and Physiology Center,National Heart, Lung, and Blood Institute, National Institutes of Health, 50 SouthDrive, Building 50, Room 3537, Bethesda, MD 20892, USA.*These authors contributed equally to this work

‡Author for correspondence (rbotelho@ryerson.ca)

R.J.B., 0000-0002-7820-0999

1

© 2018. Published by The Company of Biologists Ltd | Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

et al., 2004). This recycling process is assumed to occur throughvesicular-tubular transport intermediates like those induced byclathrin, adaptor proteins and retromer (Currinn et al., 2016; deLartigue et al., 2009; Ho et al., 2012; Phelan et al., 2006), butremains speculative. This type of membrane fission is distinct fromthe ‘kiss-and-run’ process, where lysosomes undertake transienthomotypic and heterotypic fusion events that are followed by fissionto avoid coalescence of lysosomes (Bright et al., 2005). Membranefission can also occur as organelle splitting, rather than budding, asis the case of mitochondrial fission (Chan, 2006).Recently, impairment of PIKfyve was shown to cause nuclear

accumulation of TFEB, suggesting that increased lysosomalbiogenesis may contribute to lysosome swelling (Gayle et al.,2017; Wang et al., 2015). Here, we examined the role of TFEB andthe related proteins TFE3 and MITF in lysosome enlargement andconclude that biosynthesis does not contribute to lysosomeenlargement, at least acutely. Instead, we observed that PIKfyveinhibition causes a significant reduction in lysosome numberconcurrent with lysosome enlargement. We propose that lysosomeswelling proceeds through lysosome coalescence, probably as aresult of reduced ‘run’ or fission events during lysosome kiss-and-run and/or full fusion and fission cycles.

RESULTSPIKfyve inhibition activates the transcription factor TFEBindependently of mTORPIKfyve inhibition causes lysosome enlargement (Ikonomov et al.,2001; Mccartney et al., 2014) but the mechanism by which thishappens remains unclear. There is a growing body of literature thatlinks TFEB and related transcription factors to enhanced lysosomeand protein expression (Settembre and Ballabio, 2014; Raben andPuertollano, 2016). Thus, we set out to test if TFEB activation mightcontribute to lysosome swelling. Indeed, PIKfyve inhibition usingapilimod, an exquisitely specific inhibitor of PIKfyve (Cai et al.,2013; Gayle et al., 2017), in RAW cells triggered a rapid and robustnuclear translocation of GFP-tagged TFEB and endogenous TFEB(Fig. 1A,B). To complement our pharmacological treatment, wealso demonstrated that siRNA-mediated silencing of PIKfyvesignificantly boosted the number of RAW cells with nuclearTFEB-GFP relative to control cells (Fig. 1C). In addition, HeLacells treated with apilimod also exhibited nuclear TFEB relative toresting cells, showing that activation of TFEB in the absence ofPIKfyve occurs across distinct cell types (Fig. 1D). Lastly, weshowed that GFP fusions of TFE3 and MITF, which are related toTFEB (Ploper and De Robertis, 2015; Raben and Puertollano,2016), also translocated to the nucleus upon PIKfyve abrogation inRAW cells (Fig. S1A,B). Overall, our data are consistent with recentreports that PIKfyve activity robustly controls nuclear localizationof TFEB and related factors (Gayle et al., 2017; Wang et al., 2015).TFEB is controlled by several factors including mTOR, which

phosphorylates and maintains TFEB in the cytosol (Martina et al.,2012; Roczniak-Ferguson et al., 2012; Settembre et al., 2012). Inaddition, PIKfyve has been suggested to control mTORC1 inadipocytes (Bridges et al., 2012), although others have noted thatmTOR activity is sustained in PIKfyve-inhibited cells (Wang et al.,2015; Krishna et al., 2016). To better assess the possibility thatPIKfyve controls TFEB by coordinating with mTOR in our cells, weexamined the phosphorylation state of S6K, a major mTORC1 target,in cells treated with apilimod. Strikingly, we could not observe asignificant difference in the phosphorylation levels of S6K betweencontrol and apilimod-treated cells (Fig. 2A). This suggests thatmTORC1 retains its activity in PIKfyve-inhibited RAWmacrophages.

We also wondered whether mTOR could control TFEB by regulatingPIKfyve activity instead. To test this, we exposed cells to torin1, aninhibitor of mTOR. First, we showed that apilimod reducedPtdIns(3,5)P2 by >70% with a concomitant increase in its precursorphosphatidylinositol-3-phosphate (Fig. 2B). In comparison, we didnot observe a decrease in the levels of either phosphatidylinositol-3-

Fig. 1. PIKfyve inhibition causes nuclear translocation of TFEB. (A,B)RAW cells expressing TFEB-GFP (A) or stained for endogenous TFEB (B)treated with vehicle or 20 nM apilimod for 1 h. (C) RAW cells silenced forPIKfyve or mock-silenced and expressing TFEB-GFP. (D) HeLa cells stainedfor endogenous TFEB treated with vehicle or 20 nM apilimod for 1 h. DAPI wasused to stain the nuclei. Nuclear translocation of TFEB was quantified either aspercentage of cells containing nuclear GFP-tagged TFEB (A,C) or as thenuclear-to-cytosol fluorescence intensity ratio for endogenous TFEB (B,D).Shown is the mean and s.e.m. from at least three independent experimentsand based on 50-300 cells per condition per experiment. *P<0.05 comparedwith respective control conditions using one-way ANOVA and Tukey’s post hoctest or with an unpaired Student’s t-test, as appropriate. Scale bars: 5 µm.

2

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

phosphate or PtdIns(3,5)P2 in torin1-treated cells relative to controlcells; in fact, there was a statistically significant increase inPtdIns(3,5)P2 levels in torin1-treated cells (Fig. 2B).Since TFEB is regulated by phosphorylation at various sites

(Dephoure et al., 2008; Mayya et al., 2009), we decided to assesswhether PIKfyve activity might govern TFEB throughphosphorylation. To achieve this, we examined changes in TFEBphosphorylation using Phos-tag gels, which can resolve multiplephospho-isoforms of a target protein. Notably, apilimod treatmentcaused a change in the phosphorylation of TFEBwhen resolved on aPhos-tag gel to a pattern comparable to that in torin1-treated cells

(Fig. 2C). Importantly, lysates from Tfeb−/− RAW cells lacked mostbands resolved by Phos-tag gels and detected with anti-TFEBantibodies, identifying the signal as specific to TFEB (Fig. 2C). Inaddition, pre-incubation of lysates with λ-phosphatase collapsed thesignal into a single fast-running TFEB band (Fig. 2C), suggestingthat we specifically detected a variety of phospho-TFEB isoforms.Overall, these data suggest that PIKfyve regulates kinases orphosphatases that impinge on TFEB, but these are seemingly inparallel and/or independent of mTOR.

PIKfyve inhibition boosts lysosome gene expression but notlysosomal protein levelsOur observations suggest that PIKfyve may stimulate lysosomalgene and protein expression by activating TFEB and related factors.To test this, we measured mRNA levels of genes encoding selectedlysosomal proteins, including LAMP1, MCOLN1, cathepsin D andthe V-ATPase H and D subunits using qRT-PCR. As expected, weobserved augmented mRNA levels for these genes after incubationwith apilimod for 3 h (Fig. 3A). We next examined if the enhancedmRNA levels corresponded to an increase in protein levels.Surprisingly, the levels of the V-ATPase H subunit, LAMP1 andcathepsin D in apilimod-treated RAW macrophages were similar tocontrol cells after 3 or 6 h of PIKfyve inhibition, although the smallincrease in LAMP1 protein levels after treatment for 6 h wasstatistically higher than that in resting cells (Fig. 3B,C). Thedecoupling between enhanced mRNA levels and unchanged proteinlevels may be due to the role of PIKfyve in targeting lysosomalproteins to lysosomes (Ikonomov et al., 2009a; Rutherford et al.,2006; Min et al., 2014). Given this, we suspected that TFEBactivation may not contribute to lysosome swelling in cells acutelysuppressed for PIKfyve.

PIKfyve inhibition induces lysosomeswelling in cells deletedfor TFEB and related proteinsNext, we tested whether expression of TFEB and/or TFE3 isdispensable for lysosome swelling during PIKfyve inhibition. To dothis, RAW macrophages deleted for TFEB, TFE3 or both (Pastoreet al., 2016) were treated with vehicle or apilimod for 1 h andmanually assessed for the average diameter and number of swollenvacuoles using differential interference contrast (DIC) optics.As suspected, RAW cells carrying deletions for TFEB and/orTFE3 were not impaired in apilimod-induced vacuolation, withvacuole number and size comparable to values in wild-type RAWcells (Fig. S2A).

We then followed manual assessment with a more robustvolumetric analysis to quantify the size and number of endo/lysosomes by decorating their luminal volume with fluid-phasepinocytic tracers fluorescent-tagged dextrans and Lucifer Yellow.Notably, our labelling method likely decorates a collection of lateendosomes, lysosomes and endolysosomes, a term that describeshybrid endosome-lysosome organelles (Huotari and Helenius,2011). Thus, we use the term endo/lysosome to reflect theheterogeneous mixture of endosomes and lysosomes likely to belabelled here and to differentiate from endolysosomes. First, weshowed that fluorescent-dextran colocalized to LAMP1-positivestructures similarly well between all RAW strains, confirming thatdeletion of TFEB and/or TFE3 did not disrupt trafficking of thelabel to endo/lysosomes (Fig. S2B). Second, we defined parameterssuch as thresholding and minimum particle size to estimate endo/lysosome number and volume (see Materials and Methods andFig. S3 for optimization testing). Using this approach, we conclude

Fig. 2. mTOR and PIKfyve function independently. (A) Western blotdetection of phosphoThr389-p70S6K and total p70S6K in PIKfyve-inhibitedcells treated as indicated. The ratio of phospho-S6K to total S6Kwas quantifiedfrom three independent blots. Shown is the mean ratio and s.d. normalized toDMSO-treated cells. (B) Quantification of PtdIns(3)P and PtdIns(3,5)P2 levelsusing 3H-myo-inositol incorporation and HPLC-coupled flow scintillation.Results are shown as the mean and s.d. from three independent experiments,normalized to their PtdInsP levels and to the respective DMSO-treatedcontrols. *P<0.05 compared with respective control conditions using multipleStudent’s t-test with the Bonferroni correction. (C) Western blot showing TFEBexpression in whole-cell lysates resolved on a Phos-tag gel. Shown in arepresentative blot from three independent experiments showing differentialmigration of TFEB in lysates from control relative to torin1- and apilimod-treatedcells. Anti-TFEB antibody specificity was confirmed by the loss of most bandsin Tfeb−/− RAW whole-cell lysates. In addition, most bands representphosphorylated isoforms of TFEB since pre-treatment of lysates with λ proteinphosphatase led to a single fast-running band.

3

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

that the average number and volume of individual endo/lysosomeswere similar between apilimod-treated wild-type and Tfeb−/−,Tfe3−/− and Tfeb−/− Tfe3−/− RAW macrophages (Fig. 4A),consistent with the manual analysis (Fig. S2A). Nevertheless,Tfeb−/− Tfe3−/− RAW macrophages retain expression of the relatedMITF protein, which becomes nuclear upon addition of apilimod(Fig. S1); this raised the possibility that MITF may be sufficientlyredundant with TFEB and TFE3 to promote lysosome expansionduring PIKfyve inhibition. To test this, we opted to compare HeLacells to counterpart HeLa cells deleted for all three proteins,previously described by Nezich et al. (2015). Despite deletion ofgenes encoding TFEB, TFE3 and MITF, endo/lysosomeenlargement caused by apilimod was indistinguishable betweenwild-type and mutant HeLa strains (Fig. S4). Finally, we then askedwhether protein biosynthesis was necessary for lysosomes toenlarge during PIKfyve repression. Strikingly, cycloheximide,which arrests protein biosynthesis, did not interfere with

apilimod-induced lysosome swelling (Fig. 4B). As a control forthe effectiveness of cycloheximide inhibition of protein synthesis,we assayed for translation activity with puromycylation and steady-state levels of p53, which is a high turnover protein. As shown inFig. 4C, resting cells exhibited a high rate of puromycylation, whichis a marker for ribosome-mediated protein synthesis, whereascycloheximide ablated this. In addition, cycloheximide rapidlydepleted p53 levels, demonstrating that protein synthesis wasarrested (Fig. 4D). Overall, our data suggest that proteinbiosynthesis and lysosome biogenesis controlled by TFEB andrelated transcription factors do not contribute substantially to endo/lysosome swelling during acute PIKfyve inhibition.

PIKfyve inhibition concurrently increases endo/lysosomevolume while decreasing endo/lysosome numberTo better understand the process of endo/lysosome enlargementduring PIKfyve inhibition, we further employed quantitativevolumetric image analysis. Using this approach, we observed thatthe average volume of individual endo/lysosomes in RAW cellstreated with apilimod increased with incubation period relative tovehicle-only cells (Fig. 5A,B). Remarkably, the average number ofendo/lysosomes per cell in apilimod-treated cells decreasedconsiderably with incubation period and relative to vehicle-exposed cells (Fig. 5A,C). Despite the decrease in lysosomenumber, the total lysosome volume per cell remained unperturbed(Fig. 5A,D). As suggested above, we also observed similar trendswith HeLa cells exposed to apilimod and Fig4−/−mouse embryonicfibroblasts; there was an increase in the average volume ofindividual endo/lysosomes, a concurrent decline in their number,whereas total lysosome volume per cell was not significantly altered(Fig. S4, Fig. 5D-H). Finally, and as implied above, there was nosignificant difference in the average size, number and total volumeof endo/lysosomes between wild-type RAW and HeLa cells andcounterpart strains deleted for TFEB, TFE3 and/or MITF before andafter PIKfyve suppression (Fig. 4; Fig. S4). Interestingly, replacingapilimod-containing medium with fresh medium reversed theeffects on endo/lysosome volume and number per cell (Fig. 6).Overall, our results collectively suggest that acute PIKfyveinhibition not only enlarges endo/lysosomes, but also decreasestheir numbers without disturbing the total endo/lysosome volume.

Imbalanced fusion-fission cycles may underpin endo/lysosome swelling in PIKfyve-inhibited cellsIt is unlikely that endo/lysosome number is reduced during PIKfyveinhibition via lysosome lysis because the total lysosome volume percell was unchanged relative to control cells. Instead, the reduction inendo/lysosome number coupled to their enlargement suggests thatthere is a shift towards endo/lysosome homotypic fusion relative tofission in PIKfyve-abrogated cells. Hence, we predicted thatconditions that abate homotypic lysosome fusion would lessenenlargement and maintain higher lysosome numbers in PIKfyve-inhibited cells.

To test this hypothesis, we disrupted the microtubule system andmotor activity using nocadozole and ciliobrevin, respectively, toimpair lysosome-lysosome fusion (Rosa-Ferreira and Munro, 2011;Deng and Storrie, 1988; Jordens et al., 2001). Indeed, cells treatedwith these compounds resisted apilimod-induced swelling andretained higher endo/lysosome number relative to results in theapilimod-only condition (Fig. 7A-C for nocadozole, Fig. S5A-C forciliobrevin). Conversely, endo/lysosome shrinkage was acceleratedupon washing apilimod in microtubule-disrupted cells (Fig. 7D-F).To complement the pharmacological approach, we transfected cells

Fig. 3. Acute PIKfyve inhibition enhances lysosomal gene transcriptionbut not protein levels. (A) Expression of selected lysosomal genes quantifiedby qRT-PCR and normalized against Abt1. Shown is the mean±s.e.m. fromseven independent experiments. (B) Western blot of selected lysosomalproteins in RAW cells treated as indicated. Numbers on the right representrelative molecular mass (kDa). (C) Quantification of indicated proteinsdemonstrated in B and shown as the mean±s.e.m. intensity from fourindependent blots. *P<0.05 compared with respective control conditions usingmultiple Student’s t-test with the Bonferroni correction.

4

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

with the p50 dynamitin subunit or dominant negative kinesin-1 torespectively impair dynein and kinesin. We observed reducedswelling and increased endo/lysosome numbers during apilimodexposure relative to untransfected cells (Fig. S5D-I). Overall thesedata suggest that endo/lysosomes swell in PIKfyve-inhibited cellsthrough endo/lysosome coalescence.

Lysosomes undergo constant rounds of ‘kiss-and-run’ fusion andfission (Duclos et al., 2003; Bright et al., 2005; Pryor et al., 2000).We propose a model in which PIKfyve, probably through synthesisof PtdIns(3,5)P2, controls endo/lysosome fission. Consequently, inPIKfyve-inhibited cells, lysosomes fail to perform fission and/or‘kiss-and-marry’, resulting in endo/lysosome coalescence,enlargement and a drop in total lysosome number. To support thismodel, we tried to visualize endo/lysosome swelling duringapilimod exposure in RAW cells by using live-cell spinning discconfocal microscopy to obtain z-stacks at high temporal resolution.However, endo/lysosomes under observation failed to swell whenacquiring z-stacks even as infrequently as every 2 min (Fig. S6,Movies 1 and 2). In addition, lysosomes also appeared to becomeless dynamic. Indeed, when we moved to a different field of viewwithin the same coverslip after the period of observation, endo/lysosomes were engorged (Fig. S6). This suggests that photo-toxicity impairs apilimod-dependent endo/lysosome swelling. Wecould observe dynamic endo/lysosome movement and enlargementwhen imaging single planes at 1 frame/2 min (Fig. 8A,B, Movies 3and 4). Unfortunately, this made tracking of individual endo/lysosomes difficult due to their highly dynamic nature.

To bypass the photo-toxicity issue, we switched to swept-fieldconfocal microscopy, which employs dramatically less light energy.Using this technology, we could reconstruct cells at high temporalresolution and using z-stacks. Indeed, we could track the dynamicsof endo/lysosomes in control cells and those undergoingvacuolation in PIKfyve-inhibited cells over 2 h (Fig. 8C,D,Movies 5 and 6). When imaging untreated cells, endo/lysosomesproved to be highly dynamic structures undertaking significantnumber of collisions, splitting and ‘tubulation’ events (Fig. 8C,Movie 5). After adding apilimod, endo/lysosomes underwentenlargement and became fewer in number through coalescence(Fig. 8D, Movie 6). However, it is possible to see that even enlargedendo/lysosome vacuoles in PIKfyve-suppressed cells retain theirdynamic nature. To try to quantify fission and ‘run’ events, we usedparticle tracking and splitting functions in Imaris as a proxy.Although this analysis cannot distinguish between fission andseparation of overlapping but independent particles, we reasonedthat this could estimate differences in fission rates if the defect wassufficiently large. Indeed, while ∼85% of endo/lysosomes ‘split’ incontrol cells over a period of 30 min, this was reduced to ∼50% inapilimod-treated cells (Fig. 8E). Altogether, we propose that

Fig. 4. Volumetric analysis of endo/lysosomes in PIKfyve-curtailed wild-type, Tfeb−/−, Tfe3−/− and Tfeb−/− Tfe3−/− RAW cells. (A) Cells were pre-labelled with Lucifer Yellow and then treated with apilimod to induce lysosomevacuolation. Lysosome number and lysosome volume were quantified from100 cells from three independent experiments. (B) Lysosome morphology inwild-type RAW cells treated as indicated. Lysosome volume and number werequantified. In all cases, data shown are the mean±s.e.m. from threeindependent experiments, with at least 100 cells per condition per experiment.(C) Puromycylation of proteins in the absence and presence of cycloheximide.Puromycin is incorporated into elongating proteins by the ribosome, which canbe detected by anti-puromycin antibodies during western blotting.Cycloheximide treatment eradicated puromycylation of proteins by blockingprotein synthesis. The first two lanes resolved lysates unexposed to puromycinand thus accounts for non-specific bands detected by anti-puromycinantibodies. Representative blot shown from three independent experiments.(D) Cycloheximide reduces p53 abundance relative to β-actin. Representativeblot shown from three independent experiments. Numbers on the rightrepresent relative molecular mass (in kDa). Data are mean±s.e.m. abundanceby measuring p53 signal as a ratio against β-actin. *P<0.05 compared withrespective control conditions using Student’s t-test. Scale bars: 5 µm.

5

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

PIKfyve loss of function causes lysosome enlargement by reducingthe rate of lysosome fission/‘run’ events relative to lysosome fusion/‘kiss’ events, leading to lysosome coalescence.

DISCUSSIONDisruption of PIKfyve activity causes several major physiologicaldefects, including neurological decay, lethal embryonicdevelopment and inflammatory disease, among others (Ikonomovet al., 2011; Min et al., 2014; Chow et al., 2007; McCartney et al.,2014). It is unclear how these defects arise, but they are likely torelate to disruption of endo/lysosome homeostasis, includingimpaired hydrolytic function, reduced rates of autophagic flux andendo/lysosome swelling (Martin et al., 2013; de Lartigue et al.,2009; Min et al., 2014; Kim et al., 2016). Endo/lysosomevacuolation remains the most dramatic and visible defect inPIKfyve-abrogated cells. Yet, the mechanism by which endo/lysosomes become engorged remains poorly defined. Theprevailing model is that recycling from endo/lysosomes isimpaired in PIKfyve/Fab1-deficient cells, thus causing endo/

lysosome and vacuole swelling (McCartney et al., 2014; Ho et al.,2012). Arguably, this is assumed to proceed through tubulo-vesicular transport intermediates formed during membrane fissioncatalysed by coat proteins like the retromer (Bryant et al., 1998; deLartigue et al., 2009; Rutherford et al., 2006; Zhang et al., 2007).Here, we explored two other possibilities that could contribute toendo/lysosome enlargement: enhanced endo/lysosome biosynthesisand/or an imbalance in endo/lysosome fusion-fission dynamics.Collectively, our work intimates that enhanced lysosomebiosynthesis does not drive endo/lysosome swelling. Instead, ourwork suggests that endo/lysosomes coalescence during PIKfyveinhibition, resulting in concurrent enlargement of endo/lysosomesand abatement in their numbers. We provide evidence thatcoalescence is due to a shift away from fission during endo/lysosome fusion-fission cycling (Fig. S7). Our observations areconsistent with recent work by Bissig et al. (2017), who suggest arole for PIKfyve in reforming terminal lysosomes fromendolysosomes, a term they use to define a subset of lysosomesthat are acidic and proteolytic.

Fig. 5. Volumetric analysis of endo/lysosomes during acute and chronic PIKfyve suppression. (A) RAW cells were pre-labelled with Lucifer Yellowand exposed to vehicle only or to 20 nM apilimod for the indicated times. Fluorescence micrographs are z-projections of 45-55 z-planes acquired by spinning discconfocal microscopy. Scale bar: 5 µm. (B-D) Quantification of individual endo/lysosome volume (B), endo/lysosome number per RAW macrophage (C) andtotal endo/lysosome volume (D). (E)Wild-type andFig4−/−mouse embryonic fibroblasts labelled with Lucifer Yellow. Scale bar: 30 µm. (F-H) Analysis of individualendo/lysosome volume (F), endo/lysosome number per cell (G) and total endo/lysosome volume per cell (H) in wild-type and Fig4−/− mouse embryonicfibroblasts. In all cases, data shown are the mean±s.e.m. from three independent experiments, with at least 15-20 cells per condition per experiment.*P<0.05 compared with respective control conditions using one-way ANOVA and Tukey’s post hoc test for B-D, and unpaired Student’s t-test for F-H.

6

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

PIKfyve and TFEB functionInhibition of PIKfyve caused rapid and robust nuclear translocation ofTFEB, TFE3 and MITF. Indeed, others have recently observed thatTFEBshuttles into the nucleus inPIKfyve-inhibited cells (Gayle et al.,2017; Wang et al., 2015). Consistent with nuclear translocation, wealso observed increased mRNA levels of selected lysosomal genesincluding LAMP1, V-ATPase subunits and cathepsins. This increaseis in linewith a recent study that looked at transcriptome-wide changesin mRNA levels in B-cell non-Hodgkins lymphoma cells treated withapilimod (Gayle et al., 2017). This suggests that PIKfyve regulates anegative-feedback loop to control lysosome biogenesis throughsuppression of TFEB and related factors. However, acute inhibitionof PIKfyve led either to no change, or in the case of LAMP1, to asmall increase in protein levels after 6 h of PIKfyve inhibition, despitethe augmented mRNA levels. The decoupling between enhancedtranscription and constant protein levels may result from pleiotropiceffects of PIKfyve inhibition. For example, PIKfyve is known togovern trafficking of newly synthesized lysosomal genes bycontrolling recycling of mannose-6-phosphate receptors (Gayleet al., 2017; Ikonomov et al., 2009a; Rutherford et al., 2006; Zhanget al., 2007). By contrast, cells chronically impaired for PIKfyve haveaugmented LAMP1 levels (Ferguson et al., 2009, 2010). However,this LAMP1 accumulation is thought to occur through impairedautophagosome turnover rather than an increase in biosynthesis ofLAMP1 (Ferguson et al., 2009; Kim et al., 2016). This may beconsistent with the small increase in LAMP1 after prolongedinhibition of PIKfyve (Fig. 3C). Overall, given that endo/lysosomeenlargement was unaffected in cells deleted for TFEB, TFE3 and/or

MITF or treated with cycloheximide, we argue that biosynthesis doesnot significantly contribute to lysosome enlargement in cells acutelyinhibited for PIKfyve, although we do not exclude a role in otherPIKfyve-mediated functions.

PIKfyve and TFEB regulationTFEB and TFE3 are maintained in the cytosol via phosphorylationon specific residues that form docking sites for the 14-3-3scaffolding protein (Roczniak-Ferguson et al., 2012; Martinaet al., 2012). We now know that TFEB and related factors aresubjected to multiple kinases including mTORC1, PKC, PKD,GSK3β and phosphatases such as the Ca2+-dependent calcineurin(Li et al., 2016b; Martina et al., 2014; Roczniak-Ferguson et al.,2012; Settembre et al., 2012; Najibi et al., 2016; Ploper et al.,2015; Marchand et al., 2015; Wang et al., 2015). However,mTORC1 regulation of TFEB remains the best-understoodregulatory circuit. Given that PIKfyve is needed for mTORC1activity in adipocytes (Bridges et al., 2012), we examined whetherthere could be a PIKfyve-mTORC1-TFEB axis. However, S6Kremained phosphorylated in apilimod-treated cells, suggesting thatmTORC1 was active. These observations are consistent with thework of Wang et al. (2015) and Krishna et al. (2016) who alsoshowed that mTORC1 retains its activity in PIKfyve-abrogatedcells. We considered the possibility that mTORC1 and starvationcontrol PIKfyve activity, which could then modulate TFEB,perhaps by PtdIns(3,5)P2-dependent localization of TFEB tolysosomes, as has been shown for Tup1, a transcription factor inyeast that controls galactose metabolism (Han and Emr, 2011).

Fig. 6. Volumetric analysis of endo/lysosome number and volume during PIKfyve reactivation. (A) Cells were pre-labelled with Lucifer Yellowas before andthen exposed to vehicle only or to 20 nM apilimod for 1 h, followed by removal and chase in fresh medium for the indicated times. Fluorescence micrographsare z-projections of 45-55 z-planes acquired by spinning disc confocal microscopy. Scale bar: 5 µm. (B-D) Volumetric quantification of individual endo/lysosomevolume (B), endo/lysosome number per cell (C) and total endo/lysosome volume per cells (D). Data shown are the mean±s.e.m. from three independentexperiments, with at least 15-20 cells per condition per experiment. *P<0.05 between data indicated by lines using one-way ANOVA and Tukey’s post hoc test.

7

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

However, we did not observe a significant decrease in PtdIns(3,5)P2levels in mTOR-inhibited cells, although we cannot excludedisruption of a specific pool of PtdIns(3,5)P2 or of PtdIns(5)P, a

lipid that is directly or indirectly synthesized by PIKfyve (Zolovet al., 2012; Sbrissa et al., 2012). Overall, our data suggest thatmTORC1 and PIKfyve regulate TFEB in parallel. We speculate thatPIKfyve may regulate other modulators of TFEB such as PKD,PKC and GSK3β.

Enlargement of lysosomes in PIKfyve-suppressed cellsLoss of PIKfyve causes massive lysosome swelling. This has beenpartially explained through a defect in membrane recycling fromendosomes and/or lysosomes (Ho et al., 2012; McCartney et al.,2014). This model arose through experiments done in yeastexpressing an engineered ALP protein carrying a Golgi retrievalsequence (RS-ALP). RS-ALP traffics to the vacuole, where it issubjected to proteolytic maturation. Mature RS-ALP appears in theGolgi membrane fraction in wild-type yeast cells, but not in cellswith defective PtdIns(3,5)P2 activity, suggesting a defect in retrievalfrom the vacuole (Bryant et al., 1998; Dove et al., 2004). In addition,PIKfyve appears to regulate retrieval of mannose-6-phosphatereceptors from endo/lysosomes in mammalian cells, which suggestsa link between PIKfyve and coat proteins such as retromer(Ikonomov et al., 2009b; Rutherford et al., 2006; Zhang et al.,2007). Hence, endo/lysosome enlargement may partly occurthrough an imbalance in membrane influx, caused by unabatedanterograde membrane flow and hindered recycling due to defects informing tubulo-vesicular intermediates (Fig. S7). However, ourobservations also suggest that the number of endo/lysosomes inPIKfyve-suppressed cells is significantly reduced. The diminishedendo/lysosome number is unlikely to be due to lysosome lysis sincethe total endo/lysosome volumewas unchanged in PIKfyve-arrestedcells. Instead, we propose that PIKfyve suppression causes animbalance in lysosome fusion-fission cycling, leading to lysosomecoalescence and enlargement (Fig. S7).

Lysosomes constantly interact with each other through full fusionand fission, or through ‘kiss-and-run’ events. During the ‘kiss’event, at least two lysosomes dock and form a transient pore to helpexchange and homogenize their luminal content (Bright et al., 2005;Luzio et al., 2009; Duclos et al., 2000). However, complete fusionand coalescence is pre-empted by the ‘run’ event, when the twolysosomes separate, thus preventing coalescence. Our worksuggests that PIKfyve, probably through PtdIns(3,5)P2, plays arole in mediating lysosome fission to balance fusion. This role mayoccur during complete fusion-fission cycles and/or by catalysing the‘run’ event during ‘kiss-and-run’ (Fig. S7). In either case, failure toperform fission would result in overt coalescence, resulting inlysosome enlargement and reduced numbers. This model isconsistent with recent work by Bissig et al. (2017), who suggest arole for PIKfyve in reforming terminal lysosomes fromendolysosomes. We do not understand how PIKfyve modulateslysosome fission dynamics. However, we speculate thatmechanisms that offset lysosome coalescence are distinct fromthose that drive vesiculation and recycling (Krishna et al., 2016;Bryant et al., 1998; Rutherford et al., 2006). One interestingpossibility may relate to contact sites between the endo/lysosomeand the endoplasmic reticulum since these have been shown todemarcate and assemble molecular machines that catalyseendolysosome fission (Friedman et al., 2013; Rowland et al.,2014). Thus, it is appealing to speculate that PIKfyve andPtdIns(3,5)P2 may mediate this process or be regulated by thesecontact sites to drive endo/lysosome fission.

Overall, we examined two possible mechanisms that might havecontributed to endo/lysosome enlargement in PIKfyve-inhibitedcells. While TFEB and related factors are controlled by PIKfyve,

Fig. 7. Volumetric analysis of endo/lysosome number and volume in cellsdisrupted for microtubule function. (A) Cells were pre-labelled with AlexaFluor 488-conjugated dextran, followed by vehicle alone, 10 µM nocadozolefor 60 min, or co-treated with apilimod and vehicle or nocadozole for 1 h.Fluorescence micrographs are z-projections of 45-55 z-planes acquired byspinning disc confocal microscopy. (B,C) Quantification of individual endo/lysosome volume (B) and endo/lysosome number per RAW macrophage (C).(D) Cells were pre-labelled with Alexa Fluor 488-conjugated dextran as before,followed by 20 nM apilimod for 1 h and then replaced with fresh medium ormedium containing nocadozole for the indicated time period. (E,F) Volumetricquantification of individual endo/lysosome volume (E) and endo/lysosomenumber per cell (D). In all cases, data shown are the mean±s.e.m. from threeindependent experiments, with at least 15-20 cells per condition perexperiment. *P<0.05 statistical difference between conditions indicated bylines using one-way ANOVA and Tukey’s post hoc test. Scale bars: 5 µm.

8

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

which probably establishes a negative-feedback loop for lysosomebiogenesis, we did not obtain evidence to suggest that biosynthesiscontributes to lysosome swelling during acute PIKfyve inhibition.Instead, we observed that loss of PIKfyve activity caused aconcomitant reduction in the number and increase in the volume ofendo/lysosomes, suggesting that endo/lysosomes undergocoalescence through fusion and impaired fission. Future workshould delineate the contribution of lysosome coalescence andvesicular recycling to lysosome enlargement.

MATERIALS AND METHODSCell culture and transfectionRAW 264.7 macrophage strains and HeLa cell strains were cultured inDulbecco’s Modified Eagle Medium (DMEM; Wisent, St Bruno, QC)supplemented with 5% heat-inactivated fetal bovine serum (FBS; Wisent).Mouse embryonic fibroblast (MEF) cells were cultured in Roswell ParkMemorial Institute medium (RPMI-1640; Gibco, Burlington, ON) with

15% FBS. All cells were authenticated, checked for contamination andcultured at 37°C with 5% CO2. Transient transfections were performed withFuGene HD (Promega, Madison, WI) following the manufacturer’sinstructions, using a 3:1 DNA to transfection reagent ratio. Thetransfection mixture was replaced with fresh medium after 4-6 h and cellswere used 24 h post-transfection. Silencing RNA oligonucleotides wereelectroporated into RAW cells with the Neon Electroporation system(Life Technologies, Burlington, ON) following the manufacturer’sinstructions. Briefly, two rounds of knockdown were performed usingON-TARGETplus PIKfyve SMARTpool siRNA (L-040127-00-0005) orON-TARGETplus non-targeting control pool (GE Healthcare, Mississauga,ON) over 48 h.

Pharmacological manipulation of cellsCells were treated with apilimod (Toronto Research Chemicals, Toronto,ON) at indicated concentrations and times. Torin-1 (Tocris Bioscience,Minneapolis, MN) or rapamycin (BioShop, Burlington, ON) were used aspositive controls for TFEB nuclear localization and mTORC1 inhibition.

Fig. 8. Live-cell imaging of apilimod-induced lysosome swelling. (A,B) Cells were pre-labelled with Alexa Fluor 546-conjugated dextran, followed by(A) vehicle alone or (B) 20 nM apilimod at time 0 min. Single-plane images were obtained by spinning disc microscopy at 1 frame/2 min. Time in min refers totime after adding vehicle or apilimod. See Movies 3 and 4 for full capture. Scale bars: 5 µm. (C,D) Swep-tfield imaging of RAW cells exposed to (C) vehicle or(D) 20 nM apilimod. See Movies 5 and 6 for full capture. Shown are z-collapsed frames at 30 s intervals. The inset is a magnified portion of the field-of-viewtracking a few endo/lysosomes (arrows) undergoing contact and split events, which may represent fusion or fission. Scale bars: 3 µm. (E) Rate of ‘splitting’events over 30 min in vehicle- or apilimod-treated cells showing a reduction in splitting in PIKfyve-inhibited cells, which may reflect reduced fission. *P<0.05 byStudent’s t-test.

9

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

Cells were also incubated with 100 µM ciliobrevin D (EMD Millipore,Toronto, ON) or 10 µM nocodazole (Sigma-Aldrich, Oakville, ON).

Western blottingWhole-cell lysates were generated with 2× Laemmli sample buffersupplemented with protease inhibitor cocktail (Sigma-Aldrich) andPhosSTOP phosphatase inhibitor (Roche, Laval, QC) where necessary.Lysates were passed through a 27-gauge needle six times and heated.Puromycylation assays were carried out with the addition of 10 µg/mlpuromycin for 15 min before cell lysis. When protein dephosphorylationwas required, cells were lysed with 1% Triton-X100 in PBS supplementedwith protease inhibitor cocktail, centrifuged at 16,000 g for 10 min, and thesupernatant was subjected to lambda protein phosphatase (NEB, Whitby,ON) as per the manufacturer’s instruction. Briefly, 20 µg protein wasincubated with 50 mMHEPES, 100 mM NaCl, 2 mM DTT, 0.01% Brij 25,1 mMMnCl2 and 200 units of phosphatase at 30°C for 60 min. The reactionwas stopped with 5× Laemmli sample buffer and heated.

Most samples were subjected to SDS-PAGE through a 10% acrylamideresolving gel. To assay for TFEB mobility shift, we used an 8% acrylamideresolving gel supplemented with 12.5 μMPhos-tag (Wako Chemicals USA,Richmond, VA) and 25 μM MnCl2. Phos-tag gels were washed three timeswith 100 mM EDTA for 10 min each, followed by a wash with Transferbuffer (25 mM Tris-HCl, 192 mM glycine, 20% methanol). Proteins weretransferred to PVDF, blocked and incubated with primary and HRP-linkedsecondary antibodies using Tris-buffered saline plus 0.1% Tween-20 witheither 5% skimmed milk or 3% BSA. Proteins were visualized using Clarityenhanced chemiluminescence (Bio-Rad Laboratories, Mississauga, ON)with a ChemiDoc XRS+ or ChemiDoc Touch imaging system (Bio-Rad).Quantification of proteins were determined using Image Lab software (Bio-Rad) by normalizing against a loading control and then normalizing againstthe vehicle-treated control. Rabbit monoclonal antibodies used were againstcathepsin D (1:1000, GTX62603, GeneTex Inc., Irvine, CA), HSP60(1:1000, 12165S, Cell Signaling Technologies, Danvers, MA), and p70 S6kinase (1:1000, 2708S, Cell Signaling). Rabbit polyclonal antibodies wereused against ATP6V1H (1:1000, GTX110778, GeneTex), TFEB (1:2000,A303-673A, Bethyl Laboratories, Montgomery, TX), phosphoThr389-p70S6 kinase (1:1000, 9205S, Cell Signaling) and β-actin (1:1000, 4967S, CellSignaling). Rat monoclonal antibodies were used against LAMP1 (1:200,1D4B, Developmental Studies Hybridoma Bank, Iowa City, IO). Mousemonoclonal antibodies were used against puromycin (1:1000, 540411,Millipore, Burlington, MA) and p53 (1:1000, 2524S, Cell Signaling).Secondary antibodies were raised in donkey (Bethyl) and conjugated toHRP.

Quantitative RT-PCRTotal RNA was extracted from RAW cells using the GeneJET RNApurification kit (Thermo Fisher Scientific, Mississauga, ON). Equalquantities of mRNA were reverse transcribed with SuperScript VILOcDNA synthesis kit (Invitrogen) following the manufacturer’s guidelines.The subsequent cDNA was diluted 1:100 and amplified for quantitativePCR using the TaqMan Fast Advanced Master Mix (Applied Biosystems,Foster City, CA) in the presence of TaqMan assays with a Step One PlusReal-Time PCR thermal cycler (Applied Biosystems) with Step Onesoftware (v.2.2.2; Applied Biosystems). The TaqMan gene expressionassays (Invitrogen) for the reference gene Abt1 (Mm00803824_m1) andtarget genes Ctsd (Mm00515586_m1), Atp6v1h (Mm00505548_m1),ATP6V1D (Mm00445832_m1), LAMP1 (Mm00495262_m1) andMcoln1 (Mm00522550_m1) were done in triplicate. Target geneexpression was determined by relative quantification (ΔΔCt method) toAbt1 and the vehicle-treated control sample.

Endo/lysosomal labellingEndo/lysosomes in RAW cells, HeLa cells and MEFs were labelled byincubating cells with either 200 µg/ml Alexa Fluor 488- or 546-conjugateddextran (ThermoFisher) or with 2.5 mg/ml Lucifer Yellow (ThermoFisher)in cell-specific complete medium for 2 h at 37°C in 5% CO2, followed bywashing in phosphate-buffered saline and replenishing with fresh complete

medium for 1 h before live-cell imaging. This method is likely to label aspectrum of late endosomes, lysosomes and endolysosomes, which areendosome-lysosome hybrids (Huotari and Helenius, 2011). Thus, we referto dextran- or Lucifer Yellow-labelled structures as endo/lysosomes.

Epifluorescence and spinning disc confocal microscopyManual quantification of lysosome size and imaging of GFP-basedtransfections were imaged using a Leica DM5000B system outfitted witha DFC350FX camera and controlled by Leica application suite (LeicaMicrosystems Inc., Concord, ON). TFEB-GFP transfected into PIKfyveknockout cells and endogenous TFEB in RAW cells were imaged using anOlympus IX83 microscope with a Hamamatsu ORCA-Flash4.0 digitalcamera controlled by CellSens Dimensions software (Olympus Canada Inc.,Richmond Hill, ON). Endogenous TFEB was further deconvolved using aconstrained iterative in CellSens Dimensions. All other imaging of fixedcells was done with a Quorum DisKovery Spinning Disc confocalmicroscope system equipped with a Leica DMi8 microscope connected toAndor Zyla 4.2Megapixel sCMOS camera and controlled by QuorumWaveFX Powered byMetaMorph software (Quorum Technologies, Guelph, ON).Live-cell imaging by spinning disc confocal microscopy was accomplishedwith either an Olympus IX81 inverted microscope equipped with aHamamatsu C9100-13 EMCCD camera and a 60×1.35 N.A. objective andcontrolled with Volocity 6.3.0 (PerkinElmer, Bolton, ON). Alternatively,we employed the Quorum DisKovery Spinning Disc confocal microscopesystem above but using the iXON 897 EMCCD camera. Cells were imagedlive using cell-specific complete medium in a 5% CO2 chamber at 37°C. Allmicroscopes were equipped with standard filters appropriate to fluorophoresused in this study.

3H-myo-inositol labelling of phosphoinositides andphosphoinositide measurement by HPLC-coupled flowscintillationRAW cells were incubated for 24 h in inositol-free DMEM (MPBiomedical, CA) with 10 µCi/ml myo-[2-3H(N)] inositol (Perkin Elmer,MA), 10% dialyzed FBS (Gibco), 4 mM L-glutamine (Sigma Aldrich), 1×insulin-transferrin-selenium-ethanolamine (Gibco) and 20 mM HEPES(Gibco). Cells were then treated with torin1 or apilimod for 1 h. Reactionswere arrested in 600 µl of 4.5% perchloric acid (v/v) on ice for 15 min,scraped, and pelleted at 12,000 g for 10 min. Pellets were washed with 1 mlice-cold 0.1 M EDTA and resuspended in 50 µl water. Phospholipiddeacylation was then done in 500 µl of methanol/40% methylamine/1-butanol [45.7% methanol: 10.7% methylamine: 11.4% 1-butanol (v/v)] for50 min at 53°C. Samples were vacuum-dried and washed twice byresuspending in 300 µl water and drying. Dried samples were resuspendedagain in 450 µl water and extracted with 300 µl 1-butanol/ethyl ether/ethylformate (20:4:1), vortexed for 5 min and centrifuged at 12,000 g for 2 min.The bottom aqueous layer was collected and extracted two more times.The aqueous layer was dried in vacuum and resuspended in 50 µl water.Equal counts of 3H were separated by HPLC (Agilent Technologies,Mississauga, ON) through an anion exchange 4.6×250-mm column(Phenomenex, Torrance, CA) with a flow rate of 1 ml/min and subjectedto a gradient of water (buffer A) and 1 M (NH4)2HPO4, pH 3.8 (adjustedwith phosphoric acid) (buffer B) as follows: 0% B for 5 min, 0 to 2% B for15 min, 2% B for 80 min, 2 to 10% B for 20 min, 10% B for 30 min, 10 to80% B for 10 min, 80% B for 5 min, 80 to 0% B for 5 min. Theradiolabelled eluate was detected by β-RAM 4 (LabLogic, Brandon, FL)with a 1:2 ratio of eluate to scintillant (LabLogic) and analyzed by Laura 4software. Phosphoinositides were normalized against the parentphosphatidylinositol peak.

Live-cell imaging with swept-field confocal microscopyCells were seeded into single glass-bottom micro-well dishes (MatTek,Ashland, MA) and grown to 70-80% confluency. Cells were then incubatedfor 2 h in 200 µl DMEM medium supplemented with 10% FBS with a200 µg/ml Alexa Fluor 555-conjugated dextran (ThermoFisher). Sampleswere washed and incubation medium was replaced with fresh completeDMEM followed by a 90 min post-incubation period. Live-cell imaging wasperformed with a Nikon Ti inverted microscope equipped with a Prairie

10

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

Technologies swept-field slit scanning confocal scan head and a 100×1.49N.A. objective (Prairie Technologies, Sioux Falls, SD). Nikon NISElements software equipped with a Photometrics Prime 95B backilluminated sCMOS camera was used for image acquisition while high-speed triggered acquisition was controlled by a National Instruments DAQcard and Mad City Labs Piezo Z stage controller. Sample volumes wereacquired at 100 ms per z-slice at 30 s interval over a 30 min observationperiod before adding 20 nM apilimod and then for 2 h in the presence ofapilimod.

ImmunofluorescenceFollowing treatment, cells were fixed with 4% paraformaldehyde (v/v) for15 min. Cells were blocked and permeabilized with 0.2% (v/v) Triton X-100and 2% (v/v) BSA for 10 min, then subjected to rabbit polyclonal antibodyagainst TFEB (1:250; Bethyl) and Dylight-conjugated donkey polyclonalantibody against rabbit (1:500; Bethyl). Nuclei were counterstained with0.4 μg/ml DAPI. For LAMP1-specific immunostaining, cells were fixed andpermeabilized with 100% ice-cold methanol for 5 min, followed by a 2%BSA block, incubation with Dylight-conjugated rat monoclonal antibodyagainst LAMP1 (1:200, Clone 1D4B; Developmental Studies HybridomaBank) and polyclonal donkey anti-rat IgG antibodies (1:500; Bethyl).

Imaging analysis and volumetricsTo quantify the proportion of cells with nuclear TFEB-GFP (includingTFE3-GFP or MiTF-GFP), RAW cells were visually scored as cytosolic ifthe cytosol had equal or greater staining intensity compared with thenucleus. Alternatively, the nuclear:cytosolic ratio of endogenous TFEB wasmeasured as the ratio of the mean intensity in the nucleus over the meanintensity in the cytosol using ImageJ. To manually quantify vacuolediameter in DIC images, vacuoles were first defined as having a diametergreater than 1.5 µm and then the diameter was measuring using line plot inImageJ.

To quantify average endo/lysosome volume and number per cells, weemployed volumetric and particle detection tools in Volocity (Volocity6.3.0) or Icy BioImage Analysis. Z-stack images were imported into thesoftware and subject to signal thresholding. First, we tried thresholding at1.5×, 2× and 2.5× average cytosolic fluorescence background. This was thensubject to particle detection by defining particles as those with a volume>0.3 µm3, which helped eliminate noise-derived particles. Regions ofinterest were then drawn around each cell for analysis and particles splitusing the watershed function. Importantly, while increasing thresholdinglevels reduced absolute number of particles, particle size and total particlevolume, (i) the reduction occurred at similar levels between control andapilimod-treated cells and (ii) the relative trends between control andapilimod-treated cells remained the same (Fig. S3A,B). Hence, weemployed a threshold of 2× the average cytosol fluorescence intensity forthe remaining analysis. Analysis of endo/lysosome splitting frequency wasperformed using Imaris (BitPlane, Concord, MA) using its‘ImarisTrackLineage’ module, where endo/lysosome splitting was definedas the frequency of events producing two particles from a single apparentparticle. Image adjustments such as enhancing contrast were done withImageJ or Adobe Photoshop (Adobe Systems, San Jose, CA) withoutaltering the relative signals within images. Figures were assembled withAdobe Illustrator (Adobe Systems).

Statistical analysisAll experiments were repeated at least three independent times. Therespective figure legends indicate the total number of experiments andnumber of samples/cells assessed. The mean and measure of variation (s.d.)or error (s.e.m.) are indicated.

All experimental conditions were statistically analysed with either anunpaired Student’s t-test when comparing two parameters only or with aone-way ANOVA test when comparing multiple conditions in non-normalized controls. ANOVA tests were coupled to Tukey’s post hoc testto analyse pairwise conditions. For multiple comparisons with normalizedcontrols, Student’s t-tests were utilized with Bonferroni correction. Here, weaccept P<0.05 as indication of statistical significance.

AcknowledgementsPlasmid constructs expressing the kinesin-1 dominant negative and of the p50dynamitin subunit were obtained from Addgene (Cambridge, MA). Plasmidsencoding GFP fusion proteins of TFEB, TFE3 and MITF were kind gifts from DrShawn Ferguson at Yale University. The Tfeb−/− Tfe3−/− Mitf−/− HeLa cells were akind gift from Dr Richard Youle at the National Institutes of Health.

Competing interestsThe authors declare no competing or financial interests.

Author contributionsConceptualization: R.J.B.; Methodology: C.H.C., G.S., M.A.G., Z.-Y.A.O., G.L., R.P.;Software: G.S.; Formal analysis: C.H.C., G.S., S.C.W., R.J.B.; Investigation: C.H.C.,G.S., M.A.G., C.W., R.M.D., R.J.B.; Resources: G.L., R.P., S.C.W., R.J.B.; Datacuration: C.H.C., G.S., M.A.G., C.W., R.M.D., Z.-Y.A.O.; Writing - original draft:C.H.C., G.S., R.J.B.; Writing - review & editing: R.J.B.; Visualization: C.H.C., G.S.,M.A.G., C.W., R.M.D., Z.-Y.A.O., S.C.W.; Supervision: S.C.W., R.J.B.; Projectadministration: R.J.B.; Funding acquisition: R.J.B.

FundingThe research performed in the lab of R.J.B. was funded by a Discovery Grant fromthe Natural Sciences and Engineering Research Council of Canada, by the EarlyResearcher Award program from the Government of Ontario, by a Tier II CanadaResearch Chairs Award and by contributions from Ryerson University. Research byC.H.C. was made possible through support from an Ontario Graduate Scholarshipand Queen Elizabeth II Graduate Scholarship. G.L. was funded through the NationalInstitutes of Health (GM24872). Deposited in PMC for release after 12 months.

Supplementary informationSupplementary information available online athttp://jcs.biologists.org/lookup/doi/10.1242/jcs.213587.supplemental

ReferencesBagshaw, R. D., Callahan, J.W. andMahuran, D. J. (2006). The Arf-family protein,

Arl8b, is involved in the spatial distribution of lysosomes. Biochem. Biophys. Res.Commun. 344, 1186-1191.

Bissig, C., Hurbain, I., Raposo, G. and van Niel, G. (2017). PIKfyve activityregulates reformation of terminal storage lysosomes from endolysosomes. Traffic.11, 747-757.

Bridges, D., Ma, J.-T., Park, S., Inoki, K., Weisman, L. S. and Saltiel, A. R. (2012).Phosphatidylinositol 3,5-bisphosphate plays a role in the activation andsubcellular localization of mechanistic target of rapamycin 1. Mol. Biol. Cell 23,2955-2962.

Bright, N. A., Gratian, M. J. and Luzio, J. P. (2005). Endocytic delivery tolysosomes mediated by concurrent fusion and kissing events in living cells. Curr.Biol. 15, 360-365.

Bryant, N. J., Piper, R. C., Weisman, L. S. and Stevens, T. H. (1998). Retrogradetraffic out of the yeast vacuole to the TGN occurs via the prevacuolar/endosomalcompartment. J. Cell Biol. 142, 651-663.

Cai, X., Xu, Y., Cheung, A. K., Tomlinson, R. C., Alcazar-Roman, A., Murphy, L.,Billich, A., Zhang, B., Feng, Y., Klumpp, M. et al. (2013). PIKfyve, a class III PIkinase, is the target of the small molecular IL-12/IL-23 inhibitor apilimod and aplayer in Toll-like receptor signaling. Chem. Biol. 20, 912-921.

Chan, Y. H. M., Reyes, L., Sohail, S. M., Tran, N. K. and Marshall, W. F. (2016).Organelle size scaling of the budding yeast vacuole by relative growth andinheritance. Curr. Biol. 26, 1221-1228.

Chow, C. Y., Zhang, Y., Dowling, J. J., Jin, N., Adamska, M., Shiga, K., Szigeti,K., Shy, M. E., Li, J., Zhang, X. et al. (2007). Mutation of FIG4 causesneurodegeneration in the pale tremor mouse and patients with CMT4J. Nature448, 68-72.

Currinn, H., Guscott, B., Balklava, Z., Rothnie, A. and Wassmer, T. (2016). APPcontrols the formation of PI(3,5)P2 vesicles through its binding of the PIKfyvecomplex. Cell. Mol. Life Sci. 73, 393-408.

de Lartigue, J., Polson, H., Feldman, M., Shokat, K., Tooze, S. A., Urbe, S. andClague, M. J. (2009). PIKfyve regulation of endosome-linked pathways.Traffic 10,883-893.

Delamarre, L., Pack, M., Chang, H., Mellman, I. and Trombetta, E. S. (2005).Differential lysosomal proteolysis in antigen-presenting cells determines antigenfate. Science 307, 1630-1634.

Deng, Y. P. and Storrie, B. (1988). Animal cell lysosomes rapidly exchangemembrane proteins. Proc. Natl. Acad. Sci. USA 85, 3860-3864.

Dephoure, N., Zhou, C., Villen, J., Beausoleil, S. A., Bakalarski, C. E., Elledge,S. J. and Gygi, S. P. (2008). A quantitative atlas of mitotic phosphorylation. Proc.Natl. Acad. Sci. USA. 105, 10762-10767.

Dong, X., Shen, D.,Wang, X., Dawson, T., Li, X., Zhang, Q., Cheng, X., Zhang, Y.,Weisman, L. S., Delling, M. et al. (2010). PI(3,5)P(2) controls membrane

11

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

trafficking by direct activation of mucolipin Ca(2+) release channels in theendolysosome. Nat. Commun. 1, 38.

Dove, S. K., Piper, R. C., McEwen, R. K., Yu, J. W., King, M. C., Hughes, D. C.,Thuring, J., Holmes, A. B., Cooke, F. T., Michell, R. H. et al. (2004). Svp1pdefines a family of phosphatidylinositol 3,5-bisphosphate effectors. EMBO J. 23,1922-1933.

Duclos, S., Diez, R., Garin, J., Papadopoulou, B., Descoteaux, A., Stenmark, H.and Desjardins, M. (2000). Rab5 regulates the kiss and run fusion betweenphagosomes and endosomes and the acquisition of phagosome leishmanicidalproperties in RAW 264.7 macrophages. J. Cell Sci. 113, 3531-3541.

Duclos, S., Corsini, R. and Desjardins, M. (2003). Remodeling of endosomesduring lysosome biogenesis involves “kiss and run” fusion events regulated byrab5. J. Cell Sci. 116, 907-918.

Ferguson, C. J., Lenk, G. M. and Meisler, M. H. (2009). Defective autophagy inneurons and astrocytes from mice deficient in PI(3,5)P2. Hum. Mol. Genet. 18,4868-4878.

Ferguson, C. J., Lenk, G. M. and Meisler, M. H. (2010). PtdIns(3,5)P2 andautophagy in mouse models of neurodegeneration. Autophagy 6, 170-171.

Friedman, J. R., DiBenedetto, J. R., West, M., Rowland, A. A. and Voeltz, G. K.(2013). Endoplasmic reticulum-endosome contact increases as endosomestraffic and mature. Mol. Biol. Cell 24, 1030-1040.

Gayle, S., Landrette, S., Beeharry, N., Conrad, C., Hernandez, M., Beckett, P.,Ferguson, S. M., Xu, T., Rothberg, J. and Lichenstein, H. (2017). B-cell non-Hodgkin lymphoma: Selective vulnerability to PIKFYVE inhibition. Autophagy. 13,1768-1778.

Gray, M. A., Choy, C. H., Dayam, R. M., Ospina-Escobar, E., Somerville, A.,Xiao, X., Ferguson, S. M. and Botelho, R. J. (2016). Phagocytosis enhanceslysosomal and bactericidal properties by activating the transcription factor TFEB.Curr. Biol. 26, 1955-1964.

Han, B.-K. and Emr, S. D. (2011). Phosphoinositide [PI(3,5)P2] lipid-dependentregulation of the general transcriptional regulator Tup1. Genes Dev. 25, 984-995.

Ho, C. Y., Alghamdi, T. A. and Botelho, R. J. (2012). Phosphatidylinositol-3,5-bisphosphate: no longer the poor PIP2. Traffic 13, 1-8.

Huotari, J. and Helenius, A. (2011). Endosome maturation. EMBO J. 30,3481-3500.

Ikonomov, O. C., Sbrissa, D. and Shisheva, A. (2001). Mammalian cellmorphology and endocytic membrane homeostasis require enzymatically activephosphoinositide 5-kinase PIKfyve. J. Biol. Chem. 276, 26141-26147.

Ikonomov, O. C., Fligger, J., Sbrissa, D., Dondapati, R., Mlak, K., Deeb, R. andShisheva, A. (2009a). Kinesin adapter JLP links PIKfyve to microtubule-basedendosome-to-trans-Golgi network traffic of furin. J. Biol. Chem. 284, 3750-3761.

Ikonomov, O. C., Sbrissa, D. and Shisheva, A. (2009b). YM201636, an inhibitor ofretroviral budding and PIKfyve-catalyzed PtdIns(3,5)P2 synthesis, halts glucoseentry by insulin in adipocytes. Biochem. Biophys. Res. Commun. 382, 566-570.

Ikonomov,O. C., Sbrissa, D., Delvecchio, K., Xie, Y., Jin, J.-P., Rappolee, D. andShisheva, A. (2011). The phosphoinositide kinase PIKfyve is vital in earlyembryonic development: preimplantation lethality of PIKfyve-/- embryos butnormality of PIKfyve+/- mice. J. Biol. Chem. 286, 13404-13413.

Jin, N., Chow, C. Y., Liu, L., Zolov, S. N., Bronson, R., Davisson, M., Petersen,J. L., Zhang, Y., Park, S., Duex, J. E. et al. (2008). VAC14 nucleates a proteincomplex essential for the acute interconversion of PI3P and PI(3,5)P(2) in yeastand mouse. EMBO J. 27, 3221-3234.

Johnson, D. E., Ostrowski, P., Jaumouille, V. and Grinstein, S. (2016). Theposition of lysosomes within the cell determines their luminal pH. J. Cell Biol. 212,677-692.

Jordens, I., Fernandez-Borja, M., Marsman, M., Dusseljee, S., Janssen, L.,Calafat, J., Janssen, H.,Wubbolts, R. andNeefjes, J. (2001). TheRab7 effectorprotein RILP controls lysosomal transport by inducing the recruitment of dynein-dynactin motors. Curr. Biol. 11, 1680-1685.

Kim, G. H. E., Dayam, R. M., Prashar, A., Terebiznik, M. and Botelho, R. J.(2014). PIKfyve inhibition interferes with phagosome and endosomematuration inmacrophages. Traffic 15, 1143-1163.

Kim, S. M., Roy, S. G., Chen, B., Nguyen, T. M., McMonigle, R. J., McCracken,A. N., Zhang, Y., Kofuji, S., Hou, J., Selwan, E. et al. (2016). Targeting cancermetabolism by simultaneously disrupting parallel nutrient access pathways.J. Clin. Invest. 126, 4088-4102.

Krishna, S., Palm, W., Lee, Y., Yang, W., Bandyopadhyay, U., Xu, H., Florey, O.,Thompson, C. B. and Overholtzer, M. (2016). PIKfyve regulates vacuolematuration and nutrient recovery following engulfment. Dev. Cell 38, 536-547.

Lenk, G. M., Frei, C. M., Miller, A. C., Wallen, R. C., Mironova, Y. A., Giger, R. J.and Meisler, M. H. (2016). Rescue of neurodegeneration in the Fig4 null mouseby a catalytically inactive FIG4 transgene. Hum. Mol. Genet. 25, 340-347.

Li, X., Rydzewski, N., Hider, A., Zhang, X., Yang, J., Wang, W., Gao, Q., Cheng,X. and Xu, H. (2016a). A molecular mechanism to regulate lysosome motility forlysosome positioning and tubulation. Nat. Cell Biol. 18, 404-417.

Li, Y., Xu, M., Ding, X., Yan, C., Song, Z., Chen, L., Huang, X., Wang, X., Jian, Y.,Tang, G. et al. (2016b). Protein kinase C controls lysosome biogenesisindependently of mTORC1. Nat. Cell Biol. 18, 1065-1077.

Luzio, J. P., Pryor, P. R. and Bright, N. A. (2007). Lysosomes: fusion and function.Nat. Rev. Mol. Cell Biol. 8, 622-632.

Luzio, J. P., Parkinson, M. D. J., Gray, S. R. and Bright, N. A. (2009). The deliveryof endocytosed cargo to lysosomes. Biochem. Soc. Trans. 37, 1019-1021.

Marchand, B., Arsenault, D., Raymond-Fleury, A., Boisvert, F.-M. and Boucher,M.-J. (2015). Glycogen Synthase Kinase-3 (GSK3) inhibition induces prosurvivalautophagic signals in human pancreatic cancer cells. J. Biol. Chem. 290,5592-5605.

Martin, S., Harper, C. B., May, L. M., Coulson, E. J., Meunier, F. A. and Osborne,S. L. (2013). Inhibition of PIKfyve by YM-201636 dysregulates autophagy andleads to apoptosis-independent neuronal cell death. PLoS ONE 8, e60152.

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.

Martina, J. A., Diab, H. I., Lishu, L., Jeong-A, L., Patange, S., Raben, N. andPuertollano, R. (2014). The nutrient-responsive transcription factor TFE3promotes autophagy, lysosomal biogenesis, and clearance of cellular debris.Sci. Signal. 7, ra9.

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.

Mayya, V., Lundgren, D. H., Hwang, S.-I., Rezaul, K., Wu, L., Eng, J. K.,Rodionov, V. and Han, D. K. (2009). Quantitative Phosphoproteomic analysis ofT cell receptor signaling reveals system-wide modulation of protein-proteininteractions. Sci. Signal. 2, ra46.

Mccartney, A. J., Zhang, Y. and Weisman, L. S. (2014). Phosphatidylinositol 3,5-bisphosphate: low abundance, high significance. BioEssays 36, 52-64.

Medina, D. L., Fraldi, A., Bouche, V., Annunziata, F., Mansueto, G.,Spampanato, C., Puri, C., Pignata, A., Martina, J. A., Sardiello, M. et al.(2011). Transcriptional activation of lysosomal exocytosis promotes cellularclearance. Dev. Cell 21, 421-430.

Medina, D. L., Di Paola, S., Peluso, I., Armani, A., De Stefani, D., Venditti, R.,Montefusco, S., Scotto-Rosato, A., Prezioso, C., Forrester, A. et al. (2015).Lysosomal calcium signalling regulates autophagy through calcineurin and TFEB.Nat. Cell Biol. 17, 288-299.

Min, S. H., Suzuki, A., Stalker, T. J., Zhao, L., Wang, Y., McKennan, C., Riese,M. J., Guzman, J. F., Zhang, S., Lian, L. et al. (2014). Loss of PIKfyve in plateletscauses a lysosomal disease leading to inflammation and thrombosis in mice. Nat.Commun. 5, 4691.

Mindell, J. A. (2012). Lysosomal acidification mechanisms. Annu. Rev. Physiol. 74,69-86.

Najibi, M., Labed, S. A., Visvikis, O. and Irazoqui, J. E. (2016). An evolutionarilyconserved PLC-PKD-TFEB pathway for host defense. Cell Rep. 15, 1728-1742.

Nezich, C. L., Wang, C., Fogel, A. I. and Youle, R. J. (2015). MiT/TFE transcriptionfactors are activated during mitophagy downstream of Parkin and Atg5. J. Cell.Biol. 210, 435-450.

Pastore, N., Brady, O. A., Diab, H. I., Martina, J. A., Sun, L., Huynh, T., Lim, J.-A.A., Zare, H., Raben, N., Ballabio, A. et al. (2016). TFEB and TFE3 cooperate inthe regulation of the innate immune response in activated macrophages.Autophagy 12, 1240-1258.

Pen a-Llopis, S., Vega-Rubin-de-Celis, S., Schwartz, J. C., Wolff, N. C., Tran,T. A. T., Zou, L., Xie, X.-J., Corey, D. R. and Brugarolas, J. (2011). Regulation ofTFEB and V-ATPases by mTORC1. EMBO J. 30, 3242-3258.

Phelan, J. P., Millson, S. H., Parker, P. J., Piper, P. W. and Cooke, F. T. (2006).Fab1p and AP-1 are required for trafficking of endogenously ubiquitylated cargoesto the vacuole lumen in S. cerevisiae. J. Cell Sci. 119, 4225-4234.

Ploper, D. and De Robertis, E. M. (2015). The MITF family of transcription factors:Role in endolysosomal biogenesis, Wnt signaling, and oncogenesis. Pharmacol.Res. 99, 36-43.

Ploper, D., Taelman, V. F., Robert, L., Perez, B. S., Titz, B., Chen, H.-W., Graeber,T. G., von Euw, E., Ribas, A. and De Robertis, E. M. (2015). MITF drivesendolysosomal biogenesis and potentiates Wnt signaling in melanoma cells.Proc. Natl. Acad. Sci. USA 112, E420-E429.

Pryor, P. R., Mullock, B. M., Bright, N. A., Gray, S. R. and Luzio, J. P. (2000). Therole of intraorganellar Ca(2+) in late endosome-lysosome heterotypic fusion andin the reformation of lysosomes from hybrid organelles. J. Cell Biol. 149,1053-1062.

Pu, J., Schindler, C., Jia, R., Jarnik, M., Backlund, P. and Bonifacino, J. S.(2015). BORC, a multisubunit complex that regulates lysosome positioning. Dev.Cell 33, 176-188.

Raben, N. and Puertollano, R. (2016). TFEB and TFE3: linking lysosomes tocellular adaptation to stress. Annu. Rev. Cell Dev. Biol. 32, 255-278.

Reddy, A., Caler, E. V. and Andrews, N. W. (2001). Plasma membrane repair ismediated by Ca(2+)-regulated exocytosis of lysosomes. Cell 106, 157-169.

Roczniak-Ferguson, A., Petit, C. S., Froehlich, F., Qian, S., Ky, J., Angarola, B.,Walther, T. C. and Ferguson, S. M. (2012). The transcription factor TFEB linksmTORC1 signaling to transcriptional control of lysosome homeostasis. Sci.Signal. 5, ra42.

Rosa-Ferreira, C. and Munro, S. (2011). Arl8 and SKIP act together to linklysosomes to kinesin-1. Dev. Cell 21, 1171-1178.

12

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience

Rowland, A. A., Chitwood, P. J., Phillips, M. J. and Voeltz, G. K. (2014). ERcontact sites define the position and timing of endosome fission. Cell 159,1027-1041.

Rutherford, A. C., Traer, C., Wassmer, T., Pattni, K., Bujny, M. V., Carlton, J. G.,Stenmark, H. and Cullen, P. J. (2006). The mammalian phosphatidylinositol 3-phosphate 5-kinase (PIKfyve) regulates endosome-to-TGN retrograde transport.J. Cell Sci. 119, 3944-3957.

Sancak, Y., Bar-Peled, L., Zoncu, R., Markhard, A. L., Nada, S. and Sabatini,D. M. (2010). Ragulator-Rag complex targets mTORC1 to the lysosomal surfaceand is necessary for its activation by amino acids. Cell 141, 290-303.

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). Agene network regulating lysosomal biogenesis and function. Science 325,473-477.

Saric, A., Hipolito, V. E. B., Kay, J. G., Canton, J., Antonescu, C. N. andBotelho,R. J. (2016). mTOR controls lysosome tubulation and antigen presentation inmacrophages and dendritic cells. Mol. Biol. Cell 27, 321-333.

Sbrissa, D., Ikonomov, O. C. and Shisheva, A. (1999). PIKfyve, a mammalianortholog of yeast Fab1p lipid kinase, synthesizes 5-phosphoinositides. Effect ofinsulin. J. Biol. Chem. 274, 21589-21597.

Sbrissa, D., Ikonomov, O. C., Fu, Z., Ijuin, T., Gruenberg, J., Takenawa, T. andShisheva, A. (2007). Core protein machinery for mammalian phosphatidylinositol3,5-bisphosphate synthesis and turnover that regulates the progression ofendosomal transport: Novel Sac phosphatase joins the ArPIKfyve-PIKfyvecomplex. J. Biol. Chem. 282, 23878-23891.

Sbrissa, D., Ikonomov, O. C., Filios, C., Delvecchio, K. and Shisheva, A. (2012).Functional dissociation between PIKfyve-synthesized PtdIns5P andPtdIns(3,5)P2 by means of the PIKfyve inhibitor YM201636. Am. J. Physiol.Cell Physiol. 303, C436-C446.

Schwake, M., Schroder, B. and Saftig, P. (2013). Lysosomal membrane proteinsand their central role in physiology. Traffic 14, 739-748.

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

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. Eur. Mol. Biol. Organ. J. 31, 1095-1108.

Swanson, J., Burke, E., Silverstein, S. C., Bushnell, A. and Silverstein, S. C.(1987). Tubular lysosome morphology and distribution within macrophagesdepend on the integrity of cytoplasmic microtubules. Proc. Natl. Acad. Sci. USA84, 1921-1925.

Tsunemi, T., Ashe, T. D., Morrison, B. E., Soriano, K. R., Au, J., Roque, R. A. V.,Lazarowski, E. R., Damian, V. A., Masliah, E. and La Spada, A. R. (2012). PGC-1α rescues Huntington’s disease proteotoxicity by preventing oxidative stress andpromoting TFEB function. Sci. Transl. Med. 4, 142ra97.

Vyas, J. M., Kim, Y.-M., Artavanis-Tsakonas, K., Love, J. C., Van der Veen, A. G.and Ploegh, H. L. (2007). Tubulation of class II MHC compartments ismicrotubule dependent and involves multiple endolysosomal membraneproteins in primary dendritic cells. J. Immunol. 178, 7199-7210.

Wang, W., Gao, Q., Yang, M., Zhang, X., Yu, L., Lawas, M., Li, X., Bryant-Genevier, M., Southall, N. T., Marugan, J. et al. (2015). Up-regulation oflysosomal TRPML1 channels is essential for lysosomal adaptation to nutrientstarvation. Proc. Natl. Acad. Sci. USA 112, E1373-E1381.

Xiong, J. and Zhu, M. X. (2016). Regulation of lysosomal ion homeostasis bychannels and transporters. Sci. China Life Sci. 59, 777-791.

Zhang, Y., Zolov, S. N., Chow, C. Y., Slutsky, S. G., Richardson, S. C., Piper,R. C., Yang, B., Nau, J. J., Westrick, R. J., Morrison, S. J., Meisler, M. H. andWeisman, L. S. (2007). Loss of Vac14, a regulator of the signaling lipidphosphatidylinositol 3,5-bisphosphate, results in neurodegeneration in mice.Proc. Natl. Acad. Sci. USA. 104, 17518-17523.

Zolov, S. N., Bridges, D., Zhang, Y., Lee, W.-W., Riehle, E., Verma, R., Lenk,G. M., Converso-Baran, K., Weide, T., Albin, R. L. et al. (2012). In vivo, Pikfyvegenerates PI(3,5)P2, which serves as both a signaling lipid and the majorprecursor for PI5P. Proc. Natl. Acad. Sci. USA 109, 17472-17477.

13

RESEARCH ARTICLE Journal of Cell Science (2018) 131, jcs213587. doi:10.1242/jcs.213587

Journal

ofCe

llScience