Jour

nal o

f Cel

l Sci

ence

RESEARCH ARTICLE

HPS6 interacts with dynactin p150Glued to mediate retrogradetrafficking and maturation of lysosomes

Ke Li1,2, Lin Yang1, Cheng Zhang1, Yang Niu1,2, Wei Li1 and Jia-Jia Liu1,*

ABSTRACT

Hermansky-Pudlak syndrome 6 protein (HPS6) has originally

been identified as a subunit of the BLOC-2 protein complex that

is involved in the biogenesis of lysosome-related organelles.

Here, we demonstrate that HPS6 directly interacts with the

dynactin p150Glued subunit of the dynein–dynactin motor complex

and acts as cargo adaptor for the retrograde motor to mediate the

transport of lysosomes from the cell periphery to the perinuclear

region. Small interfering RNA (siRNA)-mediated knockdown of

HPS6 in HeLa cells not only partially blocks centripetal movement of

lysosomes but also causes delay in lysosome-mediated protein

degradation. Moreover, lysosomal acidification and degradative

capacity, as well as fusion between late endosomes and/or

multivesicular bodies and lysosomes are also impaired when

HPS6 is depleted, suggesting that perinuclear positioning

mediated by the dynein–dynactin motor complex is required for

lysosome maturation and activity. Our results have uncovered a so-

far-unknown specific role for HPS6 in the spatial distribution of the

lysosomal compartment.

KEY WORDS: HPS6, p150Glued, DCTN1, lysosome, dynein–dynactin

motor complex

INTRODUCTIONThe intracellular transport driven by motor proteins plays crucial

roles in membrane trafficking, which is essential for organellebiogenesis, homeostasis and function. Cytoplasmic dynein, amicrotubule-based motor, drives the movement of a wide range ofcargoes towards the minus ends of microtubules and is primarily

responsible for retrograde transport of materials from the cellperiphery to the center (Allan, 2011). Functions of dynein requirethe activity of dynactin, a multisubunit protein complex that not

only enhances its processivity along microtubule tracks (Culver-Hanlon et al., 2006; King and Schroer, 2000; Moore et al., 2009)but also links it to cargoes. Although several proteins that tether

the dynein–dynactin complex to its membranous cargoes havebeen identified and their functions in intracellular trafficking havebeen characterized (Engelender et al., 1997; Hong et al., 2009;

Hoogenraad et al., 2001; Johansson et al., 2007; Tai et al., 1999;Traer et al., 2007; Wassmer et al., 2009; Watson et al.,2005; Yano and Chao, 2005), adaptor proteins for many other

organelles and vesicular cargoes that are retrogradely transported

from the cell periphery to the center remain to be discovered.

Lysosomes are catabolic organelles of eukaryotic cellsthat contain hydrolytic enzymes required for degradation of

macromolecules, including proteins, lipids, nucleic acids andpolysaccharides. They play a central role in the degradation ofmacromolecules that are received from endocytic, phagocytic and

autophagic pathways, and are essential for many cellular andphysiological processes, such as receptor-mediated signaling,antigen presentation, pathogen clearance and cholesterolhomeostasis (Saftig and Klumperman, 2009; Settembre et al.,

2013). The intracellular trafficking and steady-state distributionof lysosomes rely upon molecular motors, and both the actin andmicrotubule cytoskeletons (Burkhardt et al., 1997; Cantalupo

et al., 2001; Cordonnier et al., 2001; Harada et al., 1998; Jordenset al., 2001; Matsushita et al., 2004; Rosa-Ferreira and Munro,2011; Santama et al., 1998). Although it has been reported that

lysosomes change their intracellular positioning in orderto coordinate cellular responses to nutrients through mTORsignaling and autophagic flux (Korolchuk et al., 2011), thephysiological relevance of perinuclear steady-state clustering of

lysosomes in cells remains largely unexplored.

Lysosome-related organelles are a class of tissue-specificcompartments that share many features with lysosomes, but store

and secret proteins for cell-type-specific functions (Marks et al.,2013; Raposo et al., 2007). They include melanosomes, platelet-dense granules, lytic granules, lamellar bodies and many others

(Raposo et al., 2007). Defects in the biogenesis of theseorganelles cause Hermansky–Pudlak syndrome (HPS), a humanautosomal recessive disorder that is characterized by partial loss

of pigmentation and prolonged bleeding, and that occurs in nineforms (Wei and Li, 2013). Except for the HPS2 gene (also knownas AP3B1) – which encodes the b-subunit of the adaptor protein(AP) complex 3 (AP-3), genes mutated in the other eight forms of

HPS encode subunits of three protein complexes that are knownas biogenesis of lysosome-related organelles complex (BLOC)-1,BLOC-2 and BLOC-3. Of these, HPS7, HPS8 and HPS9

constitute BLOC-1; HPS3, HPS5 and HPS6 constitute BLOC-2;and HPS1 and HPS4 constitute BLOC-3 (Di Pietro et al., 2004;Gautam et al., 2004; Li et al., 2003; Wei and Li, 2013; Zhang

et al., 2002; Zhang et al., 2003). Although it has been establishedthat the BLOC complexes are important for the biogenesis oflysosome-related organelles (Chiang et al., 2003; Di Pietro et al.,2006; Martina et al., 2003; Setty et al., 2007) – biological

functions of BLOC-1 and BLOC-3 have been identified mostrecently (Gerondopoulos et al., 2012; John Peter et al., 2013;Kloer et al., 2010) – the exact molecular functions of BLOC-2

subunits remain elusive.

In this study, we found that HPS6 is a cargo adaptor for thedynein–dynactin motor complex. We present evidence that HPS6

binds to dynactin p150Glued (also known as DCTN1), and that this

1State Key Laboratory of Molecular Developmental Biology, Institute of Geneticsand Developmental Biology, Chinese Academy of Sciences, Beijing 100101,China. 2University of Chinese Academy of Sciences, Beijing 100039, China.

*Author for correspondence (jjliu@genetics.ac.cn)

Received 30 August 2013; Accepted 22 August 2014

� 2014. Published by The Company of Biologists Ltd | Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4574

Jour

nal o

f Cel

l Sci

ence

interaction is required for the perinuclear positioning oflysosomes. Depletion of HPS6 in HeLa cells not only caused

mislocalization of LAMP1- and LAMP2-positive lysosomes tothe cell periphery but also impaired lysosomal degradation ofendocytic cargoes. This – however – can be rescued byoverexpressing full-length HPS6 but not by HPS6 truncation

mutants that are unable to bind p150Glued. Furthermore, lysosomalacidification as well as fusion between late endosomes and/ormultivesicular bodies (hereafter referred to as LEs/MVBs) and

lysosomes were also impaired in HPS6-depleted cells. We,therefore, propose that HSP6 facilitates retrograde lysosomaltrafficking by linking the retrograde motor to lysosomes, and that

perinuclear positioning of lysosomes is crucial for the delivery ofendocytic cargoes to lysosomes, for lysosome maturation andfunctioning.

RESULTSHPS6 directly interacts with the dynactin p150Glued subunitof the dynein–dynactin complexIn search of cargo adaptors for the dynein–dynactin motor, weperformed a yeast two-hybrid screen of a human fetal braincDNA library by using the C-terminus of human dynactin

p150Glued, amino acid residues (aa) 718–1278, as bait (Honget al., 2009). Of 20 positive clones, one was found to encode anN-terminal fragment (aa 66–197) of HPS6. To verify the

interaction between HPS6 and p150Glued in mammalian cells,we transiently co-transfected HEK293 cells with plasmidsencoding epitope-tagged full-length p150Glued and HPS6,

respectively, and performed co-immunoprecipitation (co-IP)with immobilized antibodies against the Flag tag. HPS6 andp150Glued reciprocally co-immunoprecipitated with each other(Fig. 1A), indicating that the interaction between HPS6 and

p150Glued also occurs in mammalian cells. To further verify theinteraction, we performed co-IP with HeLa cell lysates anddetected not only HPS3 and HPS5 – the other two subunits of the

BLOC-2 complex – but also p150Glued and the dyneinintermediate chain (hereafter referred to as DIC) in theimmunoprecipitates of HPS6 (Fig. 1B).

To map the HPS6 interaction site on p150Glued, we againperformed yeast two-hybrid assays (Fig. 1D) and co-IPexperiments in cultured mammalian cells with a series ofp150Glued fragments. In mammalian cells, although the extreme

C-terminus of p150Glued was not expressed, co-IP of cell extractsfrom cells that co-expressed HPS6 and p150Glued fragmentsshowed that the C-terminal p150Glued fragment of aa 911–1281

interacts strongly with HPS6 (Fig. 1E,F). Together, these dataindicate that the HPS6-binding site on p150Glued resides in itsextreme C-terminus. We also attempted to map the p150Glued

interaction site(s) on HPS6 by yeast two-hybrid assays and co-IP ofcell extracts from cells that co-expressed p150Glued and a variety ofHPS6 fragments. Interaction between the N-terminal region of

HPS6 (aa 66–200) and the extreme C-terminus of p150Glued (aa1138–1281) was verified in the yeast two-hybrid assay (Fig. 1G).Although the extreme N-terminal fragments of HPS6 were notexpressed in mammalian cells, surprisingly, all of the N-terminal

fragments (aa 1–400, 1–500 and 1–700) that were expressed incultured cells failed to co-immunoprecipitate with p150Glued

(Fig. 1H), suggesting that the region adjacent to the extreme N-

terminus inhibits its interaction with p150Glued through aberrantprotein folding. Intriguingly, among the truncation mutantsexpressed, an N-terminally truncated fragment (aa 201–805) was

able to bind p150Glued, and another N-terminal truncation fragment

(aa 301–805) showed weak p150Glued binding (Fig. 1H). Takentogether, the finding that HPS6 interacts with p150Glued prompted

us to hypothesize that HPS6 serves as cargo adaptor to mediate theretrograde transport driven by the dynein–dynactin motor complex.

The perinuclear distribution of HPS6 requires activity of thedynein–dynactin complexImmunofluorescence staining of HeLa cells with antibodiesagainst HPS6 revealed that it primarily localized to perinuclear

vesicular structures (supplementary material Fig. S1). Totest whether the perinuclear distribution of HPS6-associatedmembranous structures requires retrograde transport mediated

by the dynein–dynactin motor complex, we treated cells withnocodazole to disrupt the microtubule cytoskeleton, whichserves as track for the retrograde motor. Indeed, nocodazole

treatment resulted in disassembly of the microtubule networkand dispersal of HPS6 signals to the cell periphery (Fig. 2A).Moreover, overexpression of a Flag-tagged p50 dynamitin(DCTN2) subunit of dynactin, which disassembles the dynein–

dynactin protein complex (Echeverri et al., 1996; Melkonianet al., 2007), also caused dispersal of HPS6 signals to the cellperiphery (Fig. 2B), indicating that the perinuclear distribution

of HPS6-associated vesicles requires both an intact microtubulecytoskeleton and activity of the dynein–dynactin motorcomplex.

Because p150Glued binds to HPS6 through its C-terminus, andbecause its N-terminal and middle regions mediate itsinteraction with DIC and the Arp1 subunit of dynactin,

respectively (Fig. 1C) (Schroer, 2004), we reasoned that inthe absence of the cargo-recognition site, the N-terminalfragment of p150Glued alone could serve as a dominant-negative mutant to compete with the full-length protein in the

motor complex. In our previous study, we found that SNX6-mediated, dynein–dynactin-complex-driven retrograde transportof CI-MPR from endosomes to the TGN is disrupted by

overexpression of a p150Glued truncation mutant (aa 1–910,Flag-p150Glued-N) that lacks the SNX6-binding C-terminus(Hong et al., 2009). In agreement with this, we found in our

current study that overexpression of the same protein fragmentperturbed HPS6 distribution, thereby causing cytoplasmicdispersal of HPS6. However, overexpression of its HPS6-binding C-terminus (aa 911–1281, Flag-p150Glued-C) had no

obvious effect (Fig. 2B), probably because it cannot assembleinto the dynactin complex by itself and, thus, cannot competewith the endogenous, wild-type protein for vesicular cargoes

(Johansson et al., 2007). Therefore, our data establish that theperinuclear distribution of HPS6 requires retrograde transportdriven by the dynein–dynactin motor complex.

HPS6 associates with the lysosomal compartmentTo investigate the cellular function(s) of HPS6, we first analyzed

its subcellular compartmental distribution in HeLa cells byimmunofluorescence staining and confocal microscopy. HPS6partially colocalized with the LE/MVB marker CD63 (alsoknown as LAMP-3), (Fig. 3A,B). Intriguingly, HPS6 colocalized

to a higher extent with lysosome-associated membrane proteins 1and 2 (LAMP1 and LAMP2) (Fig. 3A,B), two of the mostabundant lysosomal proteins (Eskelinen, 2006). In contrast, little

colocalization between HPS6 and the early endosome markerEEA1 or the trans-Golgi network (TGN)-resident protein p230(GOLGA4) was observed (Fig. 3A,B). However, not only

p150Glued, but also LAMP1 and LAMP2 were detected on

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4575

Jour

nal o

f Cel

l Sci

ence

HPS6-positive vesicles that had been immunoisolated frommembrane fractions of HeLa cells (Fig. 3C). Moreover,immunoisolation of LAMP1-positive vesicles from membrane

fractions showed that not only LAMP2 but also HPS6 andp150Glued associate with LAMP1-positive lysosomes (Fig. 3D).To further determine whether HPS6 is, indeed, associated with

Fig. 1. HPS6 directly interacts with p150Glued. (A) HEK293 cells overexpressing Flag-tagged p150Glued and Myc-tagged HPS6, SNX6 or SNX1 were lysed forco-IP with immobilized Flag antibody (left). Conversely, HEK293 cells overexpressing Flag-tagged HPS6 and Myc-tagged p150Glued, HPS5 or SNX6 were lysedfor co-IP with immobilized Flag antibody (right). Input and bound proteins were analyzed by SDS-PAGE and immunoblotting with antibodies against Myc and Flag.(B) Co-IP of endogenous p150Glued and DIC from HeLa cell lysates with antibodies against HPS6. Normal IgG and antibodies to SNX1 serve as controls.Input and bound proteins were analyzed by SDS-PAGE and immunoblotting with antibodies against p150Glued, DIC, HPS6, HPS3, HPS5, SNX1 and SNX6.(C) Diagrams depicting the primary structures of p150Glued and HPS6. Domains and binding sites in p150Glued are indicated. For HPS6 no known domains andbinding sites have been identified. (D)Yeast two-hybrid assay showing the interaction of HPS6 with p150Glued C-terminus. Interaction was detected by growth ofco-transformed yeast cells on high-stringency medium and the a-Galactosidase colorimetric assay as previously described (Hong et al., 2009). Yeast cells co-transformed with pGBKT7 and pTD1-1 or pGBKT7-53 and pTD1-1 serve as negative or positive controls, respectively. (E) Co-IP from HEK293 cells overexpressingMyc-tagged HPS6 and various Flag-tagged p150Glued deletion mutants. (F) Quantification of protein-protein interaction between HPS6 and p150Glued fragments inE was determined with NIH ImageJ. Data were from three independent co-IP experiments. (G) Yeast two-hybrid assay showing the interaction between HPS6deletion mutants and p150Glued C-terminus. (H) Co-IP from HEK293 cells overexpressing Myc-tagged p150Glued and various Flag-tagged HPS6 deletion mutants.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4576

Jour

nal o

f Cel

l Sci

ence

lysosomes, we performed ultrastructural analysis of immunoisolated

HPS6-associated vesicles on magnetic beads by transmissionelectron microscopy. Electron-dense lamellar lysosomal structureswere found to be associated with HPS6-conjugated beads, whereas

no vesicular structures were detected in association with IgG-conjugated beads (Fig. 3E). In contrast, when immunoisolation wasperformed with antibodies against the cis-Golgi marker GM130,

Golgi-cisternae-like structures were found associated with GM130-conjugated beads (Fig. 3E). Together these data indicate that HPS6associates with lysosomes.

HPS6 mediates minus-end-directed microtubule motility andperinuclear positioning of lysosomesHaving established that HPS6 requires the dynein–dynactin motor

complex for perinuclear distribution and that HPS6 associateswith lysosomes, we reasoned that HPS6 plays a role in theretrograde transport and subcellular distribution of lysosomes. To

investigate the role(s) of HPS6 in lysosomal transport andpositioning, we depleted HPS6 by small interfering RNA(siRNA)-mediated RNA interference (RNAi) (supplementary

material Fig. S1E,F). Immunofluorescence microscopy showedthat lysosomes labeled with LAMP1 or LAMP2 were evenlydistributed in cells that had been depleted of HPS6 (Fig. 4A,B).

The HPS6-deficiency-induced subcellular distribution phenotypeof LAMP1 was rescued by Flag-tagged, siRNA-resistant

HPS6 (Fig. 4D,E), indicating that HPS6 is specificallyrequired for theperinuclear positioning of lysosomes.Furthermore, overexpression of the HPS6 N-terminal fragment(aa 1–700), which cannot bind to p150Glued but retains the ability to

bind to the other BLOC-2 subunit HPS5 (Fig. 4C), did not rescuethe lysosome dispersal phenotype (Fig. 4D,E), indicating that thep150Glued-binding ability of HPS6 is required for the perinuclear

positioning of lysosomes. Overexpression of Flag-HPS6, however,did not affect the subcellular distribution of LAMP1 lysosomes inHeLa cells (Fig. 4F,H).

Having established that HPS6 is required for lysosomeperinuclear positioning and that HPS6 interacts with thedynein-dynactin motor complex, we next asked whether HPS6

mediates retrograde transport of lysosomes by linking the motorcomplex to lysosomal cargoes. Since we have demonstrated thatoverexpression of p150Glued-N caused cytoplasmic dispersal ofHPS6, possibly by competing with endogenous p150Glued for

binding of the dynein–dynactin motor complex, we reasoned thatthis p150Glued fragment can also block retrograde transportmediated by HPS6. Indeed, immunofluorescence microscopy

showed that overexpression of Flag-p150Glued-N but not Flag-p150Glued-C caused dispersal of LAMP1-labeled lysosomes to thecell periphery (Fig. 4G,H), indicating that the interaction of HPS6

with p150Glued is required for lysosomal positioning in theperinuclear region. Similarly, consistent with previous studies(Burkhardt et al., 1997), overexpression of Flag-p50, which

disrupts the dynein–dynactin protein complex, caused peripheraldistribution of lysosomes (Fig. 4G,H).

To test whether HPS6 is required for the association of thedynein–dynactin motor complex with lysosomes, we performed

immunoisolation of lysosomes from HeLa cell lysates withantibodies to LAMP1. Immunoblot analysis indicated that there isa significant decrease in the levels of p150Glued as well as DIC

co-immunoisolated with LAMP1 in cells depleted of HPS6(Fig. 4I). It has been reported that, in the fibroblasts from Hps1

mutant mice (pale ear, BLOC-3 deficient), LAMP1-positive

compartments are less concentrated than wild-type, whereas theirperinuclear clustering and surface levels were normal infibroblasts from Hps3 mutant mice (cocoa, BLOC-2 deficient)(Falcon-Perez et al., 2005; Salazar et al., 2006). To determine

whether loss of HPS6 function in mouse causes a phenotype ofLAMP1 mislocalization similar to that caused by HPS6 geneknockdown, we performed immunofluorescence staining of

mouse embryonic fibroblasts (MEFs) from ruby-eye (ru) mice,which harbor an internal deletion (aa 187–189) mutant of HPS6that cannot interact with p150Glued (Fig. 5A). Confocal microscopy

analysis showed that, compared to wild-type (WT) MEFs, bothLAMP1 and LAMP2 were more peripherally distributed in ru

MEFs (Fig. 5B), indicating that loss of HPS6 function, indeed,

leads to the cytoplasmic dispersal of lysosomes. Consistently,immunoisolation from brain lysates of ru mice by using antibodiesagainst LAMP1 detected less p150Glued and DIC on LAMP1vesicles compared with those from of wild-type mice (Fig. 5C).

Moreover, the cytoplasmic dispersal phenotype of LAMP1 in ru

MEFs was also rescued by the full-length HPS6 fused to enhancedgreen fluorescent proten (EGFP) (Fig. 5D).

To further verify that HPS6, indeed, mediates retrogradetransport of lysosomes driven by the dynein–dynactin complex,we performed live imaging experiments. Image analyses of

trajectories of mCherry-LAMP1-labeled vesicles indicate that,

Fig. 2. The perinuclear distribution of HPS6 requires an intactmicrotubule network and dynein-dynactin activity. (A) HeLa cells weretreated with DMSO or 20 mM nocodazole for 1 h, fixed and immunostainedwith antibodies against HPS6 and a-tubulin. (B) HeLa cells were transientlytransfected with Flag-tagged p50, p150Glued-N (aa1–910) or p150Glued -C(aa911–1281) for 24 h and immunostained with antibodies against HPS6and Flag. Scale bar: 10 mm.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4577

Jour

nal o

f Cel

l Sci

ence

compared with control cells, there was, indeed, a significant decrease

in movement of LAMP1 vesicles towards the cell center in HPS6-depleted cells (supplementary material Fig. S2, Movies 1 and 2).Taken together, these results indicate that the perinuclear positioning

of lysosomes requires retrograde transport that is driven by thedynein–dynactin complex and that HPS6-p150Glued interactionmediates the retrograde transport of lysosomes to the cell center.

That HPS6 is required for perinuclear distribution of lysosomes

prompted us to ask whether it is also required for the subcellulardistribution of late endosomes and multivesicular bodies.Immunofluorescence microscopy of HeLa cells by using

antibodies against Rab7 and CD63 – markers for lateendosomes and multivesicular bodies, respectively – did notdetect any changes in the distribution pattern of Rab7 and CD63-

positive structures in HPS6-depleted cells (supplementarymaterial Fig. S3A-D). Moreover, no change was detected in thesubcellular distribution of CD-MPR and CI-MPR (supplementarymaterial Fig. S3E-H), carrier proteins for lysosomal hydrolases

that shuttle between the TGN and endosomes (Dıaz and Pfeffer,

1998; Saftig and Klumperman, 2009; Seaman, 2004). Thus, HPS6is not required for the positioning of LEs/MVBs, and traffickingbetween endosomes and the TGN.

Previously it has been reported that RILP binds to p150Glued andmediates retrograde transport of lysosomes towards the cell center(Johansson et al., 2007). To test whether there is functionalredundancy between HPS6 and RILP in lysosomal movement and

positioning, we performed double knockdown of HPS6 and RILPand examined lysosome distribution by immunostaining followed byquantitative analysis. Radial profile plots showed that lysosomes

were more peripherally distributed in RILP- or HPS6-depleted cells,and double knockdown caused a further shift of LAMP1fluorescence towards the cell periphery (supplementary material

Fig. S4A-C), suggesting that partially redundant or compensatorymechanisms exist to mediate centripetal movement and perinuclearpositioning of lysosomes. Moreover, we examined colocalization ofHPS6 with RILP-labeled lysosomes and found that, consistent with

Fig. 3. HPS6 associates with the lysosomal compartment. (A) HeLa cells were fixed and immunostained for HPS6 and LAMP1, LAMP2, CD63, EEA1 orp230. Scale bar: 10 mm. (B) Quantification of HPS6 colocalization with various markers in A. Asterisks indicate statistical significance of differences(compared with LAMP1) obtained from one-way analysis of variance with a Tukey post test. Data represent mean 6 s.e.m. (n535–45 cells, *P,0.05,***P,0.001). n.s., not significant. (C,D) HPS6- or LAMP1-associated organelles were immunoisolated from membrane fractions of HeLa cells with magneticDynabeads protein G conjugated with either anti-HPS6 (C) or anti-LAMP1 (D) antibody. Normal IgG and anti-GM130 serve as controls. Input and bound proteinswere resolved by SDS-PAGE and detected with antibodies against EEA1, p150Glued, GM130, LAMP1, LAMP2 and HPS6. (E) Representative electronmicrographs of ultrathin sections of Dynabeads from immunoisolation in C. Right panels are higher magnification images of boxed regions in left panels. Arrowsand arrowheads indicate membranous structures associated with magnetic beads. Scale bars: 500 nm (left panels), 200 nm (right panels).

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4578

Jour

nal o

f Cel

l Sci

ence

Fig. 4. See next page for legend.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4579

Jour

nal o

f Cel

l Sci

ence

previous reports (Cantalupo et al., 2001; Jordens et al., 2001), RILP

colocalized with LAMP1 and Rab7 (supplementary materialFig. S4D). In contrast, RILP did not colocalize with HPS6

(supplementary material Fig. S4D), suggesting that they mediatedifferent retrograde lysosomal transport pathways.

HPS6 and its interaction with p150Glued are required forlysosomal degradation of endocytosed cargoes andlysosomal enzyme activityPrevious studies have found that lysosomal positioningcoordinates nutrient responses through the regulation ofmTORC1 signaling and autophagic flux (Korolchuk et al.,

2011). To assess the physiological relevance of HPS6-mediatedperinuclear positioning of lysosomes, first we examinedlysosomal activity in HPS6-depleted cells. To this end, we

incubated HeLa cells with the fluorogenic substrate DQ-BSA andmeasured its fluorescence intensity, which is an indication oflysosomal degradation capacity as endocytosed DQ-BSA became

fluorescent upon proteolytic cleavage in lysosomes (Vazquez andColombo, 2009). Compared with control cells, weaker DQ-BSAfluorescence was detected in HPS6-depleted cells up until 3 hoursof incubation (Fig. 6A,B), indicating that HPS6 depletion delays

the degradation of endocytosed cargo in lysosomes. Similarly,when we incubated cells with Magic Red cathepsin B (MR catB),a membrane permeable probe that fluoresces upon cleavage by

the lysosomal protease cathepsin B (Creasy et al., 2007a), adecrease in the mean intensity of fluorescent puncta was detectedin HPS6-depleted cells (Fig. 6C,D), indicating that lysosomal

enzyme activity was impaired. Overexpression of siRNA-resistant full-length HPS6 restored lysosomal degradation ofDQ-BSA and MR catB, whereas the truncation mutant that failed

to interact with p150Glued but retained HPS5-binding abilityhad no effect (Fig. 6E-H). Moreover, overexpression of p50 or

Fig. 4. The perinuclear distribution of lysosomes requires HPS6-mediated dynein-dynactin activity. (A) HeLa cells were transfected with acontrol non-targeting siRNA, or siRNAs targeting HPS6 and stained withphalloidin and antibodies against LAMP1 or LAMP2, respectively. KD,knockdown. (B) Radial profile plots of signal distribution of LAMP1 andLAMP2 relative to the nucleus in A. Shown is the distribution of fluorescentsignals starting from the centroid and extending toward the cell periphery.Values are averages corresponding to the fraction of total fluorescencepresent in each concentric circle drawn from the centroid (Mean 6 s.e.m.,n53 experiments, 20–22 cells/experiment). (C) HEK293 cellsoverexpressing Flag-tagged HPS5, Myc-tagged HPS6, HPS6-N (aa1–700),or the internal deletion mutant of HPS6 (D187–189) were lysed for co-IP withimmobilized Flag antibody. Input and bound proteins were analyzed by SDS-PAGE and immunoblotting with antibodies against Myc and Flag. (D) HPS6-depleted cells were transfected with Flag vector, Flag-tagged siRNA-resistant murine HPS6 or HPS6-N and stained with antibodies againstLAMP1. (E) Radial profile plots of signal distribution of LAMP1 relative to thenucleus in D (mean 6 s.e.m., n53 experiments, 20–22 cells/experiment).(F) HeLa cells were transiently transfected with Flag-tagged HPS6 for 24 hand immunostained with antibodies against Flag and LAMP1. (G) HeLa cellswere transiently transfected with Flag vector, Flag-tagged p50, p150Glued-Nor p150Glued-C for 24 h and immunostained with antibodies against Flag andLAMP1. (H) Radial profile plots of signal distribution of LAMP1 relative to thenucleus in F and G (mean 6 s.e.m., n53 experiments, 20–22 cells/experiment). (I) Immunoisolation of LAMP1-associated membranousorganelles from HeLa cells. Input and bound proteins were analyzed bySDS-PAGE and immunoblotting with antibodies to p150Glued, EEA1, LAMP1,LAMP2, HPS6, DIC and b-actin. Relative amount of p150Glued and DIC co-immunoisolated with LAMP1 was determined with NIH ImageJ. Datarepresent mean 6 s.e.m (n53 experiments, **P,0.01). Scale bars: 10 mm.

Fig. 5. Overexpression of HPS6 rescues lysosome dispersal phenotype in MEF cells from ru mouse. (A) HEK293 cells overexpressing Flag-taggedp150Glued or Flag-tagged HPS5 and Myc-tagged HPS6 or Myc-tagged HPS6 deletion mutant (D187–189) were lysed for co-IP with immobilized Flag antibody.Input and bound proteins were analyzed by SDS-PAGE and immunoblotting with antibodies to Flag and Myc. (B) Primary cultured MEF from WT or ru mousewere immunostained with antibodies against LAMP1 and LAMP2. (C) Immunoisolation of LAMP1-associated membranous organelles from WT or ru mousebrain. Input and bound proteins were analyzed by SDS-PAGE and immunoblotting with antibodies to p150Glued, DIC, LAMP1 and HPS6. Relative amount ofp150Glued and DIC co-immunoisolated with LAMP1 was determined with NIH ImageJ. Data represent mean 6 s.e.m. (n53 experiments, *P,0.05, **P,0.01).(D) MEF cells from ru mouse were transfected with constructs expressing EGFP or EGFP-HPS6 for 24 h and immunostained with antibodies against LAMP1.Scale bar: 10 mm.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4580

Jour

nal o

f Cel

l Sci

ence

p150Glued-N fused to EGFP caused a decrease in the fluorescenceintensity of MR catB substrate puncta, wherease the EGFP-

p150Glued-C fusion proten did not (Fig. 6I,J). Together these data

indicate that HPS6 and its interaction with p150Glued are requiredfor both degradation of endocytic cargo and enzyme activity of

lysosomes.

Fig. 6. See next page for legend.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4581

Jour

nal o

f Cel

l Sci

ence

HPS6 depletion and disruption of dynein–dynactin complexactivity impair fusion between LEs/MVBs and lysosomesDirect fusion between LEs/MVBs and lysosomes was observed, and

is required for delivery of endocytic cargoes to lysosomes fordegradation (Bright et al., 2005; Bright et al., 1997; Futter et al.,1996; Mullock et al., 1998). To determine whether the decreased

degradation of endocytic cargo in HPS6-depleted cells is caused by afailure in the fusion between LEs/MVBs and lysosomes, we labeledlysosomes by incubating cells in medium containing dextranconjugated to Alexa-Fluor-647 (Dextran–647) for 12 h followed

by an 8 h chase in conjugate-free medium. Cells were thenincubated in medium containing dextran conjugated to Alexa-Fluor-488 (Dextran–488) for 3 h followed by confocal imaging

(Bright et al., 2005). Confocal microscopy analysis showed nodifference in the mean intensity and number of fluorescent puncta(Fig. 7A,B), indicating that HPS6 knockdown did not affect the

uptake of dextran. However, there was a decrease in thecolocalization of Dextran–488 and Dextran–647 in HPS6-depletedcells (Fig. 7A,B), suggesting that fusion between LEs/MVBs andlysosomes requires perinuclear distribution of lysosomes. Similarly,

less Dextran–488 colocalized with mCherry–LAMP1 in HPS6-depleted cells (Fig. 7C,D), indicating that there was less fusionbetween LEs/MVBs and lysosomes. Furthermore, ultrastructural

analysis of HPS6-depleted cells revealed that, although nosignificant difference in the size of lysosomes was detected(supplementary material Fig. S3I), compared with control cells,

there was an increase in the number of MVBs and a decrease in thenumber of lamellar-structured and electron-dense lysosomes(Fig. 7E,F). Consistently, overexpression of mCherry-p50 or

mCherry-p150Glued-N caused a decrease in the colocalizationbetween Dextran–488 and Dextran–647 (Fig. 7G,H), whereas withmCherry fused to p150Glued-C, no change in colocalization was seen.Together, these data indicate that the fusion between endocytic

cargoes and lysosomes requires HPS6 and an active dynein–dynactincomplex.

Lysosomes fuse with LEs/MVBs and mature in theperinuclear regionThe fact that knockdown of HPS6 caused both cytoplasmicdispersal of lysosomes and impaired delivery of endocyticcargoes to lysosomes prompted us to speculate that fusionbetween LEs/MVBs and lysosomes takes place in the perinuclear

region. When we then plotted the extent of colocalization ofDextran–488 and Dextran–647 as a function of radial distancestarting from the center of the nucleus determined by its measured

centroid (Niu et al., 2013), the radial profile plot revealed that theextent of colocalization of Dextran–488 and Dextran–647 peakednear the nucleus in both control and HPS6-depleted cells

(Fig. 8A), as well as in cells that overexpressed p50 orp150Glued-N (Fig. 8B).

Similarly, we investigated the relationship between lysosomal

enzyme activity and subcellular distribution with radial profileplots of the fluorescent signals of MR catB substrate. The meanintensity of MR catB fluorescence peaked near the nucleus not onlyin control knockdown and HPS6-depleted cells (Fig. 8C), but also

in cells that overexpressed p50 or p150Glued-N (Fig. 8D), whichindicates that lysosomal enzyme activity is highest in theperinuclear region. Because enzymatic activity of lysosomal

proteases is dependent on the pH of lysosomes, and becausethere is evidence that lysosomal acidification is dynamic andtightly regulated (Majumdar et al., 2011; Majumdar et al., 2007;

Mindell, 2012), we speculated that lysosomal acidificationalso takes place in the perinuclear region. To this end, we firstincubated cells overexpressing mCherry–LAMP1 with LysoSensor

(LysoSensorTM green DND-189), a membrane-permeable pHprobe (pKa55.2) that accumulates and fluoresces green inacidified compartments within a pH range of 4–5 (Lin et al.,2001; Siemasko et al., 1998). Intriguingly, compared with control

cells, there was a decrease in both the intensity of LysoSensorfluorescent puncta and in the colocalization of LysoSensor withmCherry–LAMP1 in HPS6-depleted cells (Fig. 8E,F). This

indicates that HPS6 knockdown also impaired the acidificationof lysosomes, thereby contributing to the lower lysosomal enzymeactivity that was detected by using the MR catB substrate

(Fig. 6C,D). Furthermore, the radial profile plots of LysoSensorshowed that its mean fluorescence intensity peaked near thenucleus in both control and HPS6-depleted cells (Fig. 8G),indicating also a positive correlation between lysosomal

acidification and perinuclear positioning. Together, these datasuggest that heterotypic fusion between LEs/MVBs and lysosomesas well as lysosome acidification and maturation most frequently

take place in the perinuclear region.The fact that lysosomes fuse with endocytic cargoes and

mature in the perinuclear region prompted us to speculate that

fusion of lysosomes with LEs/MVBs plays a role in lysosomalmaturation and/or acidification. It has been reported previouslythat Rab7 is required for the transfer of endocytic cargo from the

LEs/MVBs to the lysosomes (Vanlandingham and Ceresa, 2009),probably by regulating the fusion competence of the LEs /MVBs.To determine whether fusion between LE/MVB and lysosome isrequired for lysosomal maturation and acquisition of enzyme

activity, we depleted Rab7 in HeLa cells by knockdown withsiRNA (Fig. 8H). We detected lysosomal degradative capacitywith MR catB fluorescence. Compared to control cells, there was

a significant decrease in MR catB fluorescence in Rab7-depletedcells (Fig. 8I,J), indicating that Rab7-mediated fusion betweenLEs/MVBs and lysosomes, indeed, plays an important role in

lysosomal maturation.

Fig. 6. Lysosomal function is impaired in HPS6-depleted cells.(A) Snapshots of control or HPS6 knockdown cells incubated with DQ-BSAfor 3 h. The appearance of fluorescent spots indicates proteolytic cleavage ofDQ-BSA in lysosomes. Treatment of mock-transfected cells with thevacuolar-type H+-ATPase inhibitor Bafilomycin A1 (Baf) (Yoshimori et al.,1991) abolishes DQ-BSA fluorescence and serves as negative control.(B) Quantification of mean fluorescence intensity of DQ-BSA in HPS6 KDcells at different time points relative to that in control knockdown cells at 1 hin A. Data represent mean 6 s.e.m. (ctrl KD, 32 cells; HPS6 KD, 48 cells.n53 experiments, *P,0.05, **P,0.01). (C) Snapshots of control or HPS6konckdown cells incubated with MR catB substrate for 2 h. Bafilomycintreatment prevents activation of cathepsin B and abolishes MR catBsubstrate fluorescence. (D) Quantification of MR catB substrate meanintensity relative to that of control knockdown in C. Data represent mean 6

s.e.m. (ctrl KD, 49 cells; HPS6 KD, 46 cells. n53 experiments, ***P,0.001).(E) Snapshots of control or HPS6 knockdown cells expressing mCherry,mCherry fused to siRNA-resistant HPS6, or HPS6-N (aa 1–700) andincubated with DQ-BSA for 1 h. (F) Quantification of DQ-BSA mean intensityrelative to control knockdown in E. Data represent mean 6 s.e.m. (n53experiments, 11–14 cells/experiment, **P,0.01, ***P,0.001).(G) Snapshots of control or HPS6 knockdown cells expressing EGFP, EGFPfused to siRNA-resistant HPS6, or HPS6-N (aa 1–700) and incubated withMR catB substrate for 2 h. (H) Quantification of MR catB substrate meanintensity relative to control knockdown in G. Data represent mean 6 s.e.m.(n53 experiments, 11–13 cells/experiment, ***P,0.001). (I) Snapshots ofcells expressing EGFP, EGFP fused to p150Glued-N or p150Glued -C andincubated with MR catB substrate for 2 h. (J) Quantification of MR catBsubstrate mean intensity relative to EGFP in I. Data represent mean 6 s.e.m.(n53 experiments, 17–24 cells/experiment, ***P,0.001). Scale bar: 10 mm.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4582

Jour

nal o

f Cel

l Sci

ence

Fig. 7. Knockdown of HPS6 results in a decrease in fusion between LE/MVB and lysosome. (A) Snapshots of control or HPS6 knockdown cells preloadedwith Dextran-647 to label lysosomes and then incubated for 3 h with Dextran-488 to allow labeling of endocytic endosomes. (B) Quantification of the punctanumber and mean intensity of endo-lysosomal compartments labeled with Dextran-488 or 2647 and their degree of overlap in A. Data represent mean 6 s.e.m.(ctrl KD, n538 cells; HPS6 KD, n539 cells. ***P,0.001). (C) Snapshots of control or HPS6 knockdown cells transfected with mCherry-LAMP1 for 24 h and thenincubated with Dextran-488 for 3 h. (D) Quantification of colocalization between Dextran-488 and mCherry-LAMP1 in C. Data represent mean 6 s.e.m.(n535 cells, *P ,0.05). (E) Representative electron micrographs of control (left) or HPS6 knockdown (middle) cells. Arrows indicate MVBs, solid arrowheadsindicate lysosomes. Right panels are representative higher magnification images of structures with a typical morphology of MVB or lysosomes. (F) Quantificationof indicated organelles in E. Data represent mean 6 s.e.m. (ctrl KD, n5101 cells; HPS6 KD, n5103 cells. *P50.0221, ***P,0.001). (G) Cells transfected withmCherry, mCherry-p50, mCherry-p150Glued-N or mCherry-p150Glued-C for 16 h were loaded with Dextran-647 and Dextran-488 sequentially as in A, stained livewith Hoechst 33342 for 30 min and imaged by confocal microscope. (H) Quantification of colocalization of Dextran-488 and Dextran-647-labeled compartmentsin G. Data represent mean 6 s.e.m. (n540–60 cells, ***P ,0.001). Scale bars: 10 mm (A,C,G).

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4583

Jour

nal o

f Cel

l Sci

ence

DISCUSSIONIn this study, we set out to find the interaction partner(s) ofp150Glued, a subunit of the dynein–dynactin complex. Weidentified HPS6, a subunit of the BLOC-2 that is involved in

LRO biogenesis and Hermansky–Pudlak syndrome. In HeLacells, HPS6 is mostly absent from early endosomes and the TGN;knockdown of HPS6 has no effect on the distribution of these

organelles, neither does it affect the distribution of late

endosomes and multivesicular bodies. Moreover, depletion ofHPS6 has no effect on endocytic trafficking or traffickingbetween endosomes and the TGN. In contrast, althoughdisruption of the activity of the dynein–dynactin complex

results in a block of retrograde transport and peripheraldistribution of lysosomes, HPS6 depletion causes cytoplasmicdispersal of lysosomes. Disruption of the interaction between

HPS6-p150Glued and the dominant negative p150Glued N-terminus

Fig. 8. See next page for legend.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4584

Jour

nal o

f Cel

l Sci

ence

has a similar effect, which indicates that HPS6 is specifically

involved in retrograde transport of lysosomes towards the cellcenter that is driven by the dynein–dynactin complex (Fig. 8K).

Lysosomes are dispersed in Lamp12/2/Lamp22/2 cells (Huynh

et al., 2007) and the dynein–dynactin complex is required for thecentripetal movement of lysosomes (Burkhardt et al., 1997;Cantalupo et al., 2001; Harada et al., 1998; Jordens et al., 2001).However, mechanisms by which the retrograde motor is recruited

to lysosomes are only partially understood. Previous studies haveidentified multiple dynein-recruitment and -regulatory factorsthat include RILP (Cantalupo et al., 2001; Johansson et al., 2007;

Jordens et al., 2001) and LIC1/2 (Tan et al., 2011) – dynein cargoadaptors for late endosomes and lysosomes in HeLa cells.Moreover, in HPS1-deficient (pale ear) murine fibroblasts, not

only the perinuclear clustering of LAMP1-positive LEs andlysosomes but also the frequency of the movement of LAMP1-GFP-labeled organelles along microtubules was reduced,

suggesting that BLOC-3 regulates the attachment of lateendosomes and lysosomes to microtubule motors (Falcon-Perezet al., 2005; Nazarian et al., 2003). In this study, we identifiedHPS6 as yet another dynein cargo adaptor for lysosomes that is

required for lysosome perinuclear positioning. Conceivably, theseresults indicate that multiple dynein–dynactin recruitmentmechanisms exist for lysosome trafficking, which explains the

mild phenotypes caused by HPS6 depletion. Alternatively,different dynein adaptors for late endosomes and/or lysosomesor lysosome-related organelles might provide cargo specificity/

selectivity that is tissue or cell type-specific and/or organellespecific.

Most recently, BLOC-3 was identified as a guanine nucleotideexchange factor for Rab32 and Rab38, two small GTPases that

are required for melanosome biogenesis (Gerondopoulos et al.,

2012). BLOC-1 was shown to promote endosomal maturation byrecruiting the Rab5 GTPase-activating protein Msb3 to the

endosomal membrane in yeast (John Peter et al., 2013). However,molecular function(s) of BLOC-2 subunits remain elusive. Ourstudy on HPS6 reveals that it has a role in mediating lysosomaltrafficking and functioning in the HeLa cells, which is illustrated

in the model in Fig. 8K. It remains to be elucidated whetherHPS6, as the subunit of the BLOC-2 complex, also serves as adynein adaptor in the biogenesis and trafficking of lysosomes,

and of lysosome-related organelles in specialized cell types, suchas fibroblasts and melanocytes.

Fusion between LEs/MVBs and lysosomes is a crucial step

in lysosomal biogenesis and function. Previous studies havedemonstrated that Rab7 is required for efficient fusion betweenLEs/MVBs and lysosomes in HeLa cells, probably through its

interaction with the HOPS (homotypic fusion and vacuole proteinsorting) complex (Luzio et al., 2007; Vanlandingham and Ceresa,2009). In our study, fusion between LEs/MVBs and lysosomes isimpaired in cells depleted of HPS6, accompanying the more

peripheral distribution of LAMP1- and/or LAMP2-postiviecompartments, suggesting that the perinuclear clustering oflysosomes is required for the correct fusion between LEs/MVBs

and lysosomes (Fig. 8). Further, blocking LE/MVB-lysosomefusion by knocking down Rab7 also causes a reduction inlysosomal degradative capacity, supporting the notion that fusion

between LEs/MVBs and lysosomes to form endo-lysosomes isrequired for full maturation and activation of the lysosomalcompartment. Previous studies on lysosomal degradation of

amyloid Ab in microglial cells have provided strong evidencethat lysosomal acidification is tightly regulated (Majumdar et al.,2011; Majumdar et al., 2008; Majumdar et al., 2007). It will, thus,be important to investigate mechanism(s) by which heterotypic

fusion between LEs/MVBs and lysosomes triggers furtheracidification and enzyme activation in the lysosomal compartment.

MATERIALS AND METHODSAntibodies and reagentsRabbit polyclonal antibodies against HPS6 were raised as described

previously (Gautam et al., 2004). Rabbit polyclonal antibodies against

human HPS5 were raised using recombinant His-HPS5 fragment (aa 1–

333) purified from E.coli. Antibodies were affinity purified with the

antigen immobilized on AminoLinkagarose gel beads (Affi-Gel; Bio-Rad

Laboratories, Hercules, CA). Other antibodies used in this study include

mouse anti-p150Glued, mouse anti-EEA1, mouse anti-p230, mouse anti-

CD63, mouse anti-GM130 (BD Biosciences, Mississauga, ON, Canada);

mouse anti-CI-MPR (AbDSeroTec, Oxford, UK); mouse anti-CD-MPR

(22d4), mouse anti-LAMP1 (H4A3), mouse anti-LAMP2 (H4B4) (the

Developmental Studies Hybridoma Bank, Iowa City, IA); mouse anti-b-

actin, rabbit anti-Flag and rabbit anti-LAMP1 (Sigma-Aldrich, St Louis,

MO); rabbit anti-SNX1, goat anti-HPS3 (Abcam, CA); goat anti-

cathepsin D (R&D,MN); mouse anti-DIC, rabbit anti-Rab7,goat anti-

SNX6, goat anti-cathepsin B and mouse anti-c-Myc (Santa Cruz Biotech,

Santa Cruz, CA). Fluorescent conjugates of secondary antibodies and

Alexa-Fluor-488–phalloidin were purchased from Invitrogen (Carlsbad,

CA), horse radish peroxidase (HRP)-conjugated secondary antibodies

were from Cell Signaling Technology (Mississauga, ON, Canada). Sulfo-

NHS-LC-Biotin and streptavidin-agarose resins were from Thermo

Scientific (Waltham, MA).

ConstructsHPS6 full-length cDNA and fragments were PCR-amplified from the

pGBKT7-mHPS6 construct, and cloned into pCMV-Tag2b and pCMV-

Tag3b (Stratagene, La Jolla) for co-immunoprecipitation assay, or cloned

into pmCherry-C1 and pEGFP-C1 (Clontech, CA) for other experimental

Fig. 8. Lysosomes fuse with LEs/MVBs and mature in the perinuclearregion. (A) Radial profile plots of colocalized signals of Dextran-488 andDextran-647 in Fig. 7A. Shown is the distribution of colocalized fluorescentsignals starting from the centroid and extending toward the cell periphery.Values are averages corresponding to the fraction of total fluorescencepresent in each concentric circle drawn from the centroid (mean 6 s.e.m.,n53 experiments, .11 cells/experiment). (B) Radial profile plots ofcolocalized signals of Dextran-488 and Dextran-647 in Fig. 7G (mean 6

s.e.m., n53 experiments, 15 cells/experiment). (C) Radial profile plots ofsignal distribution of MR catB substrate relative to the nucleus in Fig. 6C(mean 6 s.e.m., n53 experiments, .11 cells/experiment). (D) Radial profileplots of signal distribution of MR catB substrate relative to the nucleus inFig. 6I (mean 6 s.e.m., n53 experiments, 15 cells/experiment).(E) Snapshots of control or HPS6 knockdown cells expressing mCherry-LAMP1 and incubated with LysoSensor DND-189 for 30 min.(F) Quantification of LysoSensor DND-189 mean intensity (left) and co-localization (Mander\s overlap) between LysoSensor DND-189 andmCherry-LAMP1 (right) in E. Data represent mean 6 s.e.m. (ctrl KD, n531cells; HPS6 KD, n540 cells. ***P,0.001). (G) Radial profile plots ofLysoSensor signal distribution relative to the nucleus in E (mean 6 s.e.m.,n53 experiments, .11 cells/experiment). (H) Immunostaining analysis ofHeLa cell showing Rab7 silencing with a mixture of Rab7-targeting siRNAduplexes. (I) Snapshots of control or Rab7 knockdown cells incubated withMR catB substrate for 2 h. (J) Quantification of mean intensity of MR catBsubstrate per cell. Data represent mean 6 s.e.m. (n53 experiments,***P,0.001). (K) Model for lysosomal degradation defects in the absence ofHPS6. Lysosomes move along microtubules in a bidirectional manner. HPS6serves as dynein–dynactin cargo adaptor by interacting with p150Glued andtethering the dynein–dyanctin complex to lysosomes. Depletion of HPS6impairs retrograde transport of lysosomes, thus resulting in cytoplasmicdispersal of lysosomes. Moreover, it delays lysosomal degradation ofendocytosed proteins, possibly resulted from reduced fusion of LEs/MVBswith perinuclear lysosomes. Scale bars: 10 mm.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4585

Jour

nal o

f Cel

l Sci

ence

purposes. The p150Glued fragments were PCR-amplified from the Flag-

mp150Glued plasmid and cloned into pCMV-Tag2b (Stratagene, La Jolla)

and pEGFP-C2 (Clontech, CA). The p50 full-length cDNA was PCR-

amplified from the Flag-p50 construct and cloned into pmCherry-C1 and

pEGFP-C2. The mCherry-LAMP1 and Myc–RILP constructs were

generous gifts from Li Yu (Tsinghua University, China) and Tuanlao

Wang (Xiamen University, China), respectively.

Mouse colonies and cellsThe ruby-eye (ru) mutant mouse strain and the wild-type C57BL/6J cells

were originally from the Jackson Laboratory and maintained in Richard

Swank’s laboratory at Roswell Park Cancer Institute (Buffalo, New

York) and at the animal facility of the Institute of Genetics and

Developmental Biology (IGDB), Chinese Academy of Sciences. All

procedures were approved by the Institutional Animal Care and Use

Committee of IGDB. Mouse embryonic fibroblasts (MEFs) were isolated

from these mice by following a standard protocol.

Cell culture, siRNA and transfectionHeLa, HEK293 or MEF cells were grown in DMEM (HyClone, Logan,

UT) supplemented with 10% fetal bovine serum (HyClone, Logan, UT),

2 mM L-glutamine at 37 C with 5% CO2. For DNA transfection, cells

were grown to 50% confluency and transfected with plasmid DNA using

LipofectamineTM 2000 (Invitrogen, Carlsbad, CA) following the

manufacturer’s instructions. For transfection of small interfering RNA

(siRNA), HeLa cells seeded on day 0 were transfected with 50 nM

siRNA duplex by using LipofectamineTM 2000 following the

manufacturer’s instructions on day 1, and seeded and transfected again

on day 2; experiments were performed 72 h post-transfection. For Rab7

knockdown, a mixture of oligonucleotide duplexes (Vanlandingham and

Ceresa, 2009) were transfected as previously described. For rescue

experiments, cells were transfected with plasmid DNA by using

LipofectamineTM 2000 on day 3 after transfection with siRNA, and

analyzed for the experiment on day 4. siRNA sequences used in this

study were: 59-GAGGCUUAGAACAGCUGAATT-39 (HPS6) and 59-

UUCUCCGAACGUGUCACGUTT-39 (non-targeting control). siRNA

sequences targeting Rab7 were 59-CUAGAUAGCUGGAGAGAUGUU-

39 and 59-GUACAAAGCCACAAUAGGAUU-39 ; and 59-AAACGGA-

GGUGGAGCUGUAUU-39 and 59-CGAAUUUCCUGAACCUAUCUU-39.

ImmunoprecipitationImmunoprecipitation experiments were performed as previously

described (Hong et al., 2009). For immunoprecipitation assay, HeLa or

HEK293 cells were washed once with PBS and lysed either with lysis

buffer A (0.05% [vol/vol] NP-40, 20 mMTris-HCl pH 7.4, 50 mMNaCl,

protease inhibitors) for endogenous IP or with lysis buffer B (1% [vol/

vol] Triton X-100, 50 mMTris-HCl pH 7.4, 150 mMNaCl, 1 mM EDTA

and protease inhibitors) for all other IPs.

Immunofluorescence staining and confocal microscopyCells were grown on coverslips, fixed in 4% paraformaldehyde/PBS

(pH 7.4), and permeabilized with either PBS containing 0.1% saponin

(Sigma-Aldrich, St Louis, MO), 1% goat serum, 1% BSA for staining of

LAMP1, LAMP2 and CD63 or PBS containing 0.3% Triton X-100, 1%

BSA for staining of other proteins. Cells were incubated with primary

antibodies for 1 h at room temperature; secondary antibodies conjugated

to Alexa-Fluor-488, Alexa-Fluor-555 or Alexa-Fluor-647 were used for

detection. Images were taken with the Confocal Microscope DIGITAL

ECLIPSE C1Si (Nikon, Japan).

Yeast two-hybrid assayMatchmaker Gal4 two-hybrid system 3 (Clontech, Mountain View, CA)

was used as previously described (Hong et al., 2009). Yeast strain AH109

was transformed with pGBKT7 and pGADT7 vectors, and plated on

tryptophan- and leucine-deficient SD medium (SD-Leu-Trp). The

transformed colonies were further plated on tryptophan-, leucine-,

histidine- and adenine-deficient plates (SD-Leu-Trp-Ade-His) for

interaction assay. The interaction was assessed by comparing the

growth and blue color of colonies on SD-Leu-Trp-Ade-His containing

2 mg/ml X-a-Gal.

ImmunoisolationHeLa cells were homogenized in buffer (10 mM HEPES pH 7.4, 1 mM

EDTA, 0.25 M sucrose, and protease inhibitors) and centrifuged at 800 g

for 10 min to collect the supernatant. The supernatant was pooled with

the supernatant from previous centrifugation and subjected to high-speed

centrifugation at 100,000 g at 4 C for 1 h (TLS-55 rotor, OptimaTM MAX

Ultracentrifuge; Beckman Coulter, Germany). The pellet was

resuspended in homogenization buffer (membrane fraction) and

subjected to immunoisolation with Dynabeads Protein G (Invitrogen,

Carlsbad, CA) coupled with specific antibodies or control IgG as

previously described (Niu et al., 2013). Beads were washed four times

with 5% BSA in PBS, and once with 0.1% BSA in PBS before elution.

For detection of organelle markers by immunoblotting, the beads were

eluted by boiling in 26SDS gel loading buffer and samples were analyzed

by SDS-PAGE and immunoblotting. For electron microscopy, beads

were processed as follows.

Electron microscopyImmunoisolated bead-bound vesicles were fixed with 2.5%

glutaraldehyde (Electronic Microscope Sciences, Hatfield, PA), washed

three times with 0.1M PB buffer (pH 7.4), post-fixed with 1% OsO4 (SPI

Supplies, West Chester, OH), dehydrated in ethanol of ascending purity

(30%, 50%, 70%, 80% and 90%) at 3 min per dilution, and in 100%

ethanol three times for 3 min, followed by dehydration in pure ethanol:

pure acetone (1:1) for 3 min, and dehydration in pure acetone for 3 min.

The samples were then embedded in Embed 812 (Electronic Microscope

Sciences, Hatfield, PA).

The ultrastructural analysis of cells was performed as described

(Vanlandingham and Ceresa, 2009). HeLa cells were transfected with

control or HPS6 siRNA and 72 h post-transfection, cells were washed

once with ice-cold PBS and fixed in 2.5% glutaraldehyde (Electronic

Microscope Sciences, Hatfield, PA), 0.2% tannic acid (Electronic

Microscope Sciences, Hatfield, PA), and 2.5% sucrose in 0.1 M PB

buffer (pH 7.4) at 4 C for 2 h, washed three times with 0.1M PB buffer

(pH 7.4), scraped, and pelleted. Cells were then post-fixed with 1% OsO4

in 0.1M PB for 1.5 h at 4 C. Samples were then dehydrated in ascending

acetone series (30%, 50%, 70%, 90%, 5 min per dilution, and 100% three

times every 5 min) and embedded in Embed 812.

Ultrathin sections (65 nm) were cut with a DiATOME diamond knife

(Biel, Switzerland) on a Leica ultramicrotome (Wien, Austria) and

stained with 2% uranyl acetate and lead citrate. Images were acquired on

JEOL JEM-1400 (Tokyo, Japan) transmission electron microscope

equipped with 4K62.7K pixels Gatan-832 CCD camera (Warrendale,

PA).

Live cell imagingDextran internalization assay was performed as previously described

(Bright et al., 2005). Cells were loaded with 0.2 mg/ml Alexa-Fluor-647-

conjugated dextran (10,000 MW, Invitrogen, Carlsbad, CA) at 37 C for

8 h and incubated in conjugate-free medium for 12 h. The cells were then

incubated with 0.5 mg/ml Alexa-Fluor-488-conjugated dextran for 3 h,

rinsed with dextran-free medium and processed for imaging using

confocal microscope DIGITAL ECLIPSE C1Si (Nikon, Japan).

To test lysosomal activity, cells transfected with control or siRNA

targeting HPS6 for 48 h were seeded in 4-well coverglass chambers

(Thermo Fisher Scientific, Waltham, MA). After 16 h, cells were

incubated for 1–6 h with 0.05 mg/ml DQ-BSA Green (Invitrogen)

(Vazquez and Colombo, 2009), or 2 h with Magic Red cathepsin B

substrate (MR catB, ImmunoChemistry Technology, Bloomington, MN)

(Creasy et al., 2007b) following manufacturer’s instructions. In brief,

26X MR catB solution was added directly to culture medium at a ratio of

1:26, and then the overlay cell medium was gently mixed to ensure even

exposure to MR catB. After incubating for 2 h at 37 C, culture medium

containing MR catB was removed, and cells were rinsed twice with PBS.

To inhibit the lysosomal V-ATPase activity, 1 mM bafilomycin A1

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4586

Jour

nal o

f Cel

l Sci

ence

(Sigma-Aldrich, St Louis, MO) was added to the cells for 2 h. Live cells

were stained with Hoechst 33342 for 30 min and imaged as described

above. To test cathepsin B activity in cells overexpressing p50 or

p150Glued fragments, cells transfected with constructs expressing EGFP,

EGFP-p50, EGFP-p150Glued-N, or EGFP-p150Glued-C were incubated

with MR catB for 2 h as described above, and imaged by confocal

microscope DIGITAL ECLIPSE C1Si.

To examine lysosome movement in HeLa cells, control or HPS6-

knockdown cells seeded in 4-well coverglass chambers were transfected

with mCherry-LAMP1 for 24 h and stained with Hoechst 33342 for

30 min. Time-lapse fluorescence images were acquired using

DeltaVision OMX V4.0TMSysterms (Applied Precision) with the

following settings: 11 sections; spacing: 0.25 mm; time interval: 4 s;

total duration: ,5 min.

Image analysisSingle-slice confocal images were analyzed using NIS-Elements AR 3.1

(Nikon, Japan). For quantification of protein signals on the same

immunoblot, images were analyzed using NIH ImageJ. For quantification

of signals in confocal images, images were subtracted the random

fluorescence intensity, which were averaged over multiple cells.

Blinded quantitative analysis of fluorescent signal distribution was

performed on 7026702 confocal images using the NIH ImageJ plugin

Radial profile as described (Sengupta et al., 2009). The centroid of the

nuclear fluorescent signal was defined as the center of the concentric

circles and was determined by using the ‘Measure’ function of NIH

ImageJ. For each cell, the algorithm measured the mean signal intensity

in a series of concentric circles drawn from the calculated centroid.

Average values from cells over three experiments were then used to

generate radial profile plots in which the fraction of total mean

fluorescence is expressed as a function of radial distance. For

quantification of the distribution of signal overlap between Dextran-

488 and Dextran-647, the overlapped fluorescent signals were generated

using the ‘Build Coloc Channel’ function of Imaris version 7.6.5

(Bitplane, Zurich, Switzerland). To measure the radius of the nucleus, the

freehand selection was first used to trace the outline of the Hoechst 33342

signal, then the radial profile plug-in of NIH ImageJ was used to generate

a circle exactly surrounding the Hoechst 33342 signal; the value of

nuclear radius was then calculated automatically. The mean value of

nuclear radius was generated from all cells.

The trajectories of mCherry-LAMP1 vesicles were analyzed basically

as previously described (Schuh, 2011). Nucleus and vesicles were

detected with the ‘cell function’ of Imaris and spots were tracked in three

dimensions over time. To ensure that vesicles can be tracked long enough

to determine the overall directionality of their movement, vesicles with

displacement length more than 0.35 mm were evaluated. ‘Vesicle

distance to the closest nucleus center’ at each time point was

automatically calculated by the ‘cell function’ and exported into Excel.

The direction of vesicle movement was determined by comparing the

beginning and final distance relative to the center of the nucleus. Since

fluorescent signals in live cells photobleached fast during imaging, only

the first 30 images captured in each live imaging experiment were

analyzed.

Quantification of EM images measuring the number of lysosomes,

multivesicular bodies and endo-lysosomes per cell profile were done in

.100 randomly selected cell profiles obtained from multiple different

grids (.3) containing ultrathin sections. Structures of lysosomes,

multivesicular bodies, and endo-lysosomes in EM images were

identified according to previous reports (Huotari and Helenius, 2011;

Luzio et al., 2007; Saftig and Klumperman, 2009). In brief, lysosomes

have an electron-dense lumen with a diameter ranging from 100–

1200 nm or lamellar structures with some electron-dense areas;

multivesicular bodies contain numerous intraluminal vesicles (ILVs)

with a diameter of ,50–100 nm.

Statistical analysisStatistical analyses were performed using GraphPad Prism 5 (GraphPad

Software, La Jolla, CA). Pairwise significance was calculated using two-tailed

unpaired Student’s t-test or nonparametric Student’s t-test (Mann–Whitney

test). For evaluation of statistical significance of three or more groups of

samples, one-way analysis of variance with a Tukeypost test was used.

Results are described as mean 6 s.e.m.. Data were considered statistically

significant at P,0.05. *P,0.05; **P,0.01; ***P,0.001.

AcknowledgementsWe are indebted to our colleagues Li Yu (Tsinghua University) and Tuanlao Wang(Xiamen University) for providing reagents, and Shilai Bao (IGDB, CAS) for criticalcomments on the manuscript. We thank Ling Zhu for technical assistance, andZhen-Hua Hao for the preparation of MEF cells and tissues from wild-type andruby-eye C57Bl/6J mice.

Competing interestsThe authors declare no competing interests.

Author contributionsExperiments were designed by K.L., W.L. and J.-J.L. Experiments wereconducted by K.L., L.Y. and C.Z. Data were analyzed by K.L. and Y.N. Themanuscript was written by J.-J.L. with help from K.L., L.Y., C.Z., Y.N. and W.L.

FundingThis work was supported by grants from the Ministry of Science and Technology[grant numbers 2011CB965002 and 2014CB942802] and the National NaturalScience Foundation of China [grant numbers 31325017 and 31071175 to J-J.L.and 31230046 to W.L.].

Supplementary materialSupplementary material available online athttp://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.141978/-/DC1

ReferencesAllan, V. J. (2011). Cytoplasmic dynein. Biochem. Soc. Trans. 39, 1169-1178.Bright, N. A., Reaves, B. J., Mullock, B. M. and Luzio, J. P. (1997). Dense corelysosomes can fuse with late endosomes and are re-formed from the resultanthybrid organelles. J. Cell Sci. 110, 2027-2040.

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.

Burkhardt, J. K., Echeverri, C. J., Nilsson, T. and Vallee, R. B. (1997).Overexpression of the dynamitin (p50) subunit of the dynactin complex disruptsdynein-dependent maintenance of membrane organelle distribution. J. Cell Biol.139, 469-484.

Cantalupo, G., Alifano, P., Roberti, V., Bruni, C. B. and Bucci, C. (2001). Rab-interacting lysosomal protein (RILP): the Rab7 effector required for transport tolysosomes. EMBO J. 20, 683-693.

Chiang, P. W., Oiso, N., Gautam, R., Suzuki, T., Swank, R. T. and Spritz, R. A.(2003). The Hermansky-Pudlak syndrome 1 (HPS1) and HPS4 proteins arecomponents of two complexes, BLOC-3 and BLOC-4, involved in the biogenesisof lysosome-related organelles. J. Biol. Chem. 278, 20332-20337.

Cordonnier, M.-N., Dauzonne, D., Louvard, D. and Coudrier, E. (2001). Actinfilaments and myosin I alpha cooperate with microtubules for the movement oflysosomes. Mol. Biol. Cell 12, 4013-4029.

Creasy, B. M., Hartmann, C. B., HigginsWhite, F. K. andMcCoy, K. L. (2007). Newassay using fluorogenic substrates and immunofluorescence staining to measurecysteine cathepsin activity in live cell subpopulations. Cytometry A 71A, 114-123.

Culver-Hanlon, T. L., Lex, S. A., Stephens, A. D., Quintyne, N. J. and King,S. J. (2006). A microtubule-binding domain in dynactin increases dyneinprocessivity by skating along microtubules. Nat. Cell Biol. 8, 264-270.

Di Pietro, S. M., Falcon-Perez, J. M. and Dell’Angelica, E. C. (2004).Characterization of BLOC-2, a complex containing the Hermansky-Pudlaksyndrome proteins HPS3, HPS5 and HPS6. Traffic 5, 276-283.

Di Pietro, S. M., Falcon-Perez, J. M., Tenza, D., Setty, S. R., Marks, M. S., Raposo,G. and Dell’Angelica, E. C. (2006). BLOC-1 interacts with BLOC-2 and the AP-3complex to facilitate protein trafficking on endosomes.Mol. Biol. Cell 17, 4027-4038.

Dıaz, E. and Pfeffer, S. R. (1998). TIP47: a cargo selection device for mannose 6-phosphate receptor trafficking. Cell 93, 433-443.

Echeverri, C. J., Paschal, B. M., Vaughan, K. T. and Vallee, R. B. (1996).Molecular characterization of the 50-kD subunit of dynactin reveals function forthe complex in chromosome alignment and spindle organization during mitosis.J. Cell Biol. 132, 617-633.

Engelender, S., Sharp, A. H., Colomer, V., Tokito, M. K., Lanahan, A., Worley, P.,Holzbaur, E. L. and Ross, C. A. (1997). Huntingtin-associated protein 1 (HAP1)interacts with the p150Glued subunit of dynactin.Hum. Mol. Genet. 6, 2205-2212.

Eskelinen, E.-L. (2006). Roles of LAMP-1 and LAMP-2 in lysosome biogenesisand autophagy. Mol. Aspects Med. 27, 495-502.

Falcon-Perez, J. M., Nazarian, R., Sabatti, C. and Dell’Angelica, E. C. (2005).Distribution and dynamics of Lamp1-containing endocytic organelles infibroblasts deficient in BLOC-3. J. Cell Sci. 118, 5243-5255.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4587

Jour

nal o

f Cel

l Sci

ence

Futter, C. E., Pearse, A., Hewlett, L. J. and Hopkins, C. R. (1996). Multivesicularendosomes containing internalized EGF-EGF receptor complexes mature andthen fuse directly with lysosomes. J. Cell Biol. 132, 1011-1023.

Gautam, R., Chintala, S., Li, W., Zhang, Q., Tan, J., Novak, E. K., Di Pietro,S. M., Dell’Angelica, E. C. and Swank, R. T. (2004). The Hermansky-Pudlaksyndrome 3 (cocoa) protein is a component of the biogenesis of lysosome-related organelles complex-2 (BLOC-2). J. Biol. Chem. 279, 12935-12942.

Gerondopoulos, A., Langemeyer, L., Liang, J. R., Linford, A. and Barr, F. A.(2012). BLOC-3 mutated in Hermansky-Pudlak syndrome is a Rab32/38guanine nucleotide exchange factor. Curr. Biol. 22, 2135-2139.

Harada, A., Takei, Y., Kanai, Y., Tanaka, Y., Nonaka, S. and Hirokawa,N. (1998). Golgi vesiculation and lysosome dispersion in cells lackingcytoplasmic dynein. J. Cell Biol. 141, 51-59.

Hong, Z., Yang, Y., Zhang, C., Niu, Y., Li, K., Zhao, X. and Liu, J. J. (2009). Theretromer component SNX6 interacts with dynactin p150(Glued) and mediatesendosome-to-TGN transport. Cell Res. 19, 1334-1349.

Hoogenraad, C. C., Akhmanova, A., Howell, S. A., Dortland, B. R., De Zeeuw,C. I., Willemsen, R., Visser, P., Grosveld, F. and Galjart, N. (2001).Mammalian Golgi-associated Bicaudal-D2 functions in the dynein-dynactinpathway by interacting with these complexes. EMBO J. 20, 4041-4054.

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

Huynh, K. K., Eskelinen, E. L., Scott, C. C., Malevanets, A., Saftig, P. andGrinstein, S. (2007). LAMP proteins are required for fusion of lysosomes withphagosomes. EMBO J. 26, 313-324.

Johansson, M., Rocha, N., Zwart, W., Jordens, I., Janssen, L., Kuijl, C.,Olkkonen, V. M. and Neefjes, J. (2007). Activation of endosomal dynein motorsby stepwise assembly of Rab7-RILP-p150Glued, ORP1L, and the receptorbetalll spectrin. J. Cell Biol. 176, 459-471.

John Peter, A. T., Lachmann, J., Rana, M., Bunge, M., Cabrera, M. andUngermann, C. (2013). The BLOC-1 complex promotes endosomal maturationby recruiting the Rab5 GTPase-activating protein Msb3. J. Cell Biol. 201, 97-111.

Jordens, I., Fernandez-Borja, M., Marsman, M., Dusseljee, S., Janssen, L.,Calafat, J., Janssen, H., Wubbolts, R. and Neefjes, J. (2001). The Rab7effector protein RILP controls lysosomal transport by inducing the recruitment ofdynein-dynactin motors. Curr. Biol. 11, 1680-1685.

King, S. J. and Schroer, T. A. (2000). Dynactin increases the processivity of thecytoplasmic dynein motor. Nat. Cell Biol. 2, 20-24.

Kloer, D. P., Rojas, R., Ivan, V., Moriyama, K., van Vlijmen, T., Murthy, N.,Ghirlando, R., van der Sluijs, P., Hurley, J. H. and Bonifacino, J. S. (2010).Assembly of the biogenesis of lysosome-related organelles complex-3 (BLOC-3) and its interaction with Rab9. J. Biol. Chem. 285, 7794-7804.

Korolchuk, V. I., Saiki, S., Lichtenberg, M., Siddiqi, F. H., Roberts, E. A.,Imarisio, S., Jahreiss, L., Sarkar, S., Futter, M., Menzies, F. M. et al. (2011).Lysosomal positioning coordinates cellular nutrient responses. Nat. Cell Biol.13, 453-460.

Li, W., Zhang, Q., Oiso, N., Novak, E. K., Gautam, R., O’Brien, E. P., Tinsley, C. L.,Blake, D. J., Spritz, R. A., Copeland, N. G. et al. (2003). Hermansky-Pudlaksyndrome type 7 (HPS-7) results frommutant dysbindin, a member of the biogenesisof lysosome-related organelles complex 1 (BLOC-1). Nat. Genet. 35, 84-89.

Lin, H. J., Herman, P., Kang, J. S. and Lakowicz, J. R. (2001). Fluorescencelifetime characterization of novel low-pH probes. Anal. Biochem. 294, 118-125.

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

Majumdar, A., Cruz, D., Asamoah, N., Buxbaum, A., Sohar, I., Lobel, P. andMaxfield, F. R. (2007). Activation of microglia acidifies lysosomes and leads todegradation of Alzheimer amyloid fibrils. Mol. Biol. Cell 18, 1490-1496.

Majumdar, A., Chung, H., Dolios, G., Wang, R., Asamoah, N., Lobel, P. andMaxfield, F. R. (2008). Degradation of fibrillar forms of Alzheimer’s amyloidbeta-peptide by macrophages. Neurobiol. Aging 29, 707-715.

Majumdar, A., Capetillo-Zarate, E., Cruz, D., Gouras, G. K. and Maxfield, F. R.(2011). Degradation of Alzheimer’s amyloid fibrils by microglia requires deliveryof ClC-7 to lysosomes. Mol. Biol. Cell 22, 1664-1676.

Marks, M. S., Heijnen, H. F. and Raposo, G. (2013). Lysosome-relatedorganelles: unusual compartments become mainstream. Curr. Opin. Cell Biol.25, 495-505.

Martina, J. A., Moriyama, K. and Bonifacino, J. S. (2003). BLOC-3, a proteincomplex containing the Hermansky-Pudlak syndrome gene products HPS1 andHPS4. J. Biol. Chem. 278, 29376-29384.

Matsushita, M., Tanaka, S., Nakamura, N., Inoue, H. and Kanazawa, H. (2004).A novel kinesin-like protein, KIF1Bbeta3 is involved in the movement oflysosomes to the cell periphery in non-neuronal cells. Traffic 5, 140-151.

Melkonian, K. A., Maier, K. C., Godfrey, J. E., Rodgers, M. and Schroer, T. A.(2007). Mechanism of dynamitin-mediated disruption of dynactin. J. Biol. Chem.282, 19355-19364.

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

Moore, J. K., Sept, D. and Cooper, J. A. (2009). Neurodegeneration mutations indynactin impair dynein-dependent nuclear migration. Proc. Natl. Acad. Sci. USA106, 5147-5152.

Mullock, B. M., Bright, N. A., Fearon, C. W., Gray, S. R. and Luzio, J. P. (1998).Fusion of lysosomes with late endosomes produces a hybrid organelle ofintermediate density and is NSF dependent. J. Cell Biol. 140, 591-601.

Nazarian, R., Falcon-Perez, J. M. and Dell’Angelica, E. C. (2003). Biogenesis oflysosome-related organelles complex 3 (BLOC-3): a complex containing theHermansky-Pudlak syndrome (HPS) proteins HPS1 and HPS4. Proc. Natl.Acad. Sci. USA 100, 8770-8775.

Niu, Y., Zhang, C., Sun, Z., Hong, Z., Li, K., Sun, D., Yang, Y., Tian, C., Gong, W.and Liu, J. J. (2013). PtdIns(4)P regulates retromer-motor interaction tofacilitate dynein-cargo dissociation at the trans-Golgi network. Nat. Cell Biol. 15,417-429.

Raposo, G., Marks, M. S. and Cutler, D. F. (2007). Lysosome-related organelles:driving post-Golgi compartments into specialisation. Curr. Opin. Cell Biol. 19,394-401.

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

Saftig, P. and Klumperman, J. (2009). Lysosome biogenesis and lysosomalmembrane proteins: trafficking meets function. Nat. Rev. Mol. Cell Biol. 10, 623-635.

Salazar, G., Craige, B., Styers, M. L., Newell-Litwa, K. A., Doucette, M. M.,Wainer, B. H., Falcon-Perez, J. M., Dell’Angelica, E. C., Peden, A. A.,Werner, E. et al. (2006). BLOC-1 complex deficiency alters the targeting ofadaptor protein complex-3 cargoes. Mol. Biol. Cell 17, 4014-4026.

Santama, N., Krijnse-Locker, J., Griffiths, G., Noda, Y., Hirokawa, N. and Dotti,C. G. (1998). KIF2beta, a new kinesin superfamily protein in non-neuronal cells,is associated with lysosomes and may be implicated in their centrifugaltranslocation. EMBO J. 17, 5855-5867.

Schroer, T. A. (2004). Dynactin. Annu. Rev. Cell Dev. Biol. 20, 759-779.Schuh, M. (2011). An actin-dependent mechanism for long-range vesicletransport. Nat. Cell Biol. 13, 1431-1436.

Seaman, M. N. (2004). Cargo-selective endosomal sorting for retrieval to the Golgirequires retromer. J. Cell Biol. 165, 111-122.

Sengupta, D., Truschel, S., Bachert, C. and Linstedt, A. D. (2009). Organelletethering by a homotypic PDZ interaction underlies formation of the Golgimembrane network. J. Cell Biol. 186, 41-55.

Settembre, C., Fraldi, A., Medina, D. L. and Ballabio, A. (2013). Signals from thelysosome: a control centre for cellular clearance and energy metabolism. Nat.Rev. Mol. Cell Biol. 14, 283-296.

Setty, S. R., Tenza, D., Truschel, S. T., Chou, E., Sviderskaya, E. V., Theos,A. C., Lamoreux, M. L., Di Pietro, S. M., Starcevic, M., Bennett, D. C. et al.(2007). BLOC-1 is required for cargo-specific sorting from vacuolar earlyendosomes toward lysosome-related organelles. Mol. Biol. Cell 18, 768-780.

Siemasko, K., Eisfelder, B. J., Williamson, E., Kabak, S. and Clark, M. R.(1998). Cutting edge: signals from the B lymphocyte antigen receptor regulateMHC class II containing late endosomes. J. Immunol. 160, 5203-5208.

Tai, A. W., Chuang, J. Z., Bode, C., Wolfrum, U. and Sung, C. H. (1999).Rhodopsin’s carboxy-terminal cytoplasmic tail acts as a membrane receptor forcytoplasmic dynein by binding to the dynein light chain Tctex-1. Cell 97, 877-887.

Tan, S. C., Scherer, J. and Vallee, R. B. (2011). Recruitment of dynein to lateendosomes and lysosomes through light intermediate chains. Mol. Biol. Cell 22,467-477.

Traer, C. J., Rutherford, A. C., Palmer, K. J., Wassmer, T., Oakley, J., Attar, N.,Carlton, J. G., Kremerskothen, J., Stephens, D. J. and Cullen, P. J. (2007).SNX4 coordinates endosomal sorting of TfnR with dynein-mediated transportinto the endocytic recycling compartment. Nat. Cell Biol. 9, 1370-1380.

Vanlandingham, P. A. and Ceresa, B. P. (2009). Rab7 regulates late endocytictrafficking downstream of multivesicular body biogenesis and cargosequestration. J. Biol. Chem. 284, 12110-12124.

Vazquez, C. L. and Colombo, M. I. (2009). Assays to assess autophagy inductionand fusion of autophagic vacuoles with a degradative compartment, usingmonodansylcadaverine (MDC) and DQ-BSA. Methods Enzymol. 452, 85-95.

Wassmer, T., Attar, N., Harterink, M., van Weering, J. R., Traer, C. J., Oakley, J.,Goud, B., Stephens, D. J., Verkade, P., Korswagen, H. C. et al. (2009). Theretromer coat complex coordinates endosomal sorting and dynein-mediatedtransport, with carrier recognition by the trans-Golgi network.Dev. Cell 17, 110-122.

Watson, P., Forster, R., Palmer, K. J., Pepperkok, R. and Stephens, D. J.(2005). Coupling of ER exit to microtubules through direct interaction of COPIIwith dynactin. Nat. Cell Biol. 7, 48-55.

Wei, A. H. and Li, W. (2013). Hermansky-Pudlak syndrome: pigmentary and non-pigmentary defects and their pathogenesis. Pigment Cell Melanoma Res. 26,176-192.

Yano, H. and Chao, M. V. (2005). Biochemical characterization of intracellularmembranes bearing Trk neurotrophin receptors. Neurochem. Res. 30, 767-777.

Yoshimori, T., Yamamoto, A., Moriyama, Y., Futai, M. and Tashiro, Y. (1991).Bafilomycin A1, a specific inhibitor of vacuolar-type H(+)-ATPase, inhibitsacidification and protein degradation in lysosomes of cultured cells. J. Biol.Chem. 266, 17707-17712.

Zhang, Q., Li, W., Novak, E. K., Karim, A., Mishra, V. S., Kingsmore, S. F., Roe,B. A., Suzuki, T. and Swank, R. T. (2002). The gene for the muted (mu) mouse,a model for Hermansky-Pudlak syndrome, defines a novel protein whichregulates vesicle trafficking. Hum. Mol. Genet. 11, 697-706.

Zhang, Q., Zhao, B., Li, W., Oiso, N., Novak, E. K., Rusiniak, M. E., Gautam, R.,Chintala, S., O’Brien, E. P., Zhang, Y. et al. (2003). Ru2 and Ru encode mouseorthologs of the genes mutated in human Hermansky-Pudlak syndrome types 5and 6. Nat. Genet. 33, 145-153.

RESEARCH ARTICLE Journal of Cell Science (2014) 127, 4574–4588 doi:10.1242/jcs.141978

4588