Lysosomal Transmembrane Protein LAPTM4B Promotes Autophagy and Tolerance to Metabolic Stress in Cancer Cells

Published OnlineFirst October 28, 2011; DOI: 10.1158/0008-5472.CAN-11-0940

Cancer Molecular and Cellular Pathobiology Research

Yang Li1, Qing Zhang1, Ruiyang Tian1, Qi Wang1, Jean J. Zhao1, J. Dirk Iglehart1,2, Zhigang Charles Wang1, and Andrea L. Richardson1,2

Abstract Amplification of chromosome 8q22, which includes the gene for lysosomal associated transmembrane protein LAPTM4B, has been linked to de novo anthracycline resistance in primary breast cancers with poor prognosis. LAPTM4B overexpression can induce cytosolic retention of anthracyclines and decrease drug-induced DNA damage. In this study, we tested the hypothesis that LAPTM4B may contribute to tumor cell growth or survival in the absence of a chemotherapeutic exposure. In mammary cells, LAPTM4B protein was localized in lysosomes where its depletion increased membrane permeability, pH, cathepsin release, and cellular apoptosis. Loss of LAPTM4B also inhibited later stages of autophagy by blocking maturation of the autophagosome, thereby rendering cells more sensitive to nutrient deprivation or hypoxia. Conversely, enforced overexpression of LAPTM4B promoted autophagic flux and cell survival during in vitro starvation and stimulated more rapid tumor growth in vivo. Together, our results indicate that LAPTM4B is required for lysosome homeostasis, acidification, and function, and that LAPTM4B renders tumor cells resistant to lysosome-mediated cell death triggered by environmental and genotoxic stresses. Cancer Res; 71(24); 7481–9. 2011 AACR.

Introduction oxygen species and lipid mediators or triggering mitochondrial outer membrane permeabilization. Cathepsin B is a major Chemoresistance is often an acquired characteristic of mediator of apoptotic pathways activating the proapoptotic recurrent tumors either induced or selected by exposure to protein Bid by truncation. Truncated Bid (tBid) translocates to a drug. Alternatively, de novo drug resistance can occur due to mitochondria to active the caspase cascade and promotes intrinsic features of the primary tumor which are selected by apoptosis. Cathepsin D is an aspartate protease that directly contributing a proliferative or survival advantage to tumor activates caspase (3, 4). The current knowledge of the proteins cells. The possible mechanisms of chemoresistance may on lysosomal membranes critical for affecting or regulating include altered drug uptake, intracellular distribution, efflux, these various lysosomal functions is limited. and turnover. Lysosomal retention of drugs is associated with Autophagy is a conserved lysosome-mediated intracellular drug resistance and lysosomal concentration of anthracyclines trafficking pathway that degrades and recycles intracellular is thought to increase drug efflux and decrease drug nuclear components (5). Autophagy is also a homeostatic mechanism localization, thereby preventing effective chemotherapy- that regulates metabolism and energy production and may be induced DNA damage (1). upregulated in response to a variety of cell stresses (5–7). As Lysosomes are organelles that contain hydrolytic enzymes cancers develop and disseminate, the process of autophagy such as proteases, nucleases, and lipases. Lysosomal mem- (autophagy flux) may be upregulated to support tumor cell brane permeabilization (LMP) can pose a threat to cellular survival and allow tumors to adapt to these stresses (8–10). In homeostasis through release of lysosomal contents and is a addition, autophagy has been shown to promote cell survival in recognized trigger of cell death (2). For example, a selective and response to chemotherapeutic agents (11–13). Conversely, too limited release of lysosomal cathepsins B or D may activate the much autophagy may catabolize essential components result- mitochondrial cell death pathway by either generating reactive ing in autophagic cell death (14, 15). Autophagy is regulated by a signaling cascade involving mTOR pathway inhibition, the autophagy proteins (Atgs), and two ubiquitin-like conjugation Authors' Affiliations: 1Dana-Farber Cancer Institute, and 2Brigham and – Women's Hospital, Harvard Medical School, Boston, Massachusetts systems (16 20). During autophagy initiation, a portion of the cytosol is surrounded by a cisternal membrane, the phago- Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). phore (21), that closes to form a double-membraned vesicle, the autophagosome (22). During autophagosome formation, Corresponding Author: Andrea L. Richardson, Dana Farber Cancer Insti- tute, 450 Brookline Ave, Smith 936A, Boston, MA 02215. Phone: 617-582- cytosolic LC3 is cleaved by a protease and then conjugated to 7352; Fax: 617-632-3709; E-mail: [email protected] and Zhigang phosphatidylethanolamine to form autophagosomal mem- Charles Wang. E-mail: [email protected] brane–associated LC3II (23); the level of LC3II correlates with doi: 10.1158/0008-5472.CAN-11-0940 the number of autophagosomes (24, 25). After their formation, 2011 American Association for Cancer Research. autophagosomes fuse with endosomes to form amphisomes

www.aacrjournals.org 7481

Li et al.

(26, 27) and then with lysosomes to form autolysosomes where copy number and expression of LAPTM4B (43) and were sequestered material is degraded (5, 28–30). Blockade of chosen for LAPTM4B transfection for overexpression and autophagosome maturation and fusion with the lysosomal tumor xenograft experiments. Cell culture condition, siRNA compartment results in accumulation of autophagosomes and transfection, expression construct carrying LAPTM4B cDNA, interruption in the flux of material through the autophagic and stable transfer of this construct into MDA468 were con- pathway (31). Autophagic flux can be monitored by measuring ducted as described (36). the level of substrates normally degraded by autophagy such as p62/SQSTM1 (25). The p62 protein is a ubiquitin-binding Analysis of LMP scaffold protein that binds directly to LC3 and is itself degraded Cells were transfected with control and LAPTM4B siRNA. in autolysosomes (32). Inhibition of autophagic degradation After 48 hours, control and LAPTM4B expression–modified results in an increase in p62 levels and increased autophagy cells were incubated with fluorescein isothiocyanate (FITC)- flux is indicated by decreased p62 levels (33). Many molecules labeled dextrans of different molecular weights from 4, 40, and have been implicated in these later stages of autophagosome 70 kDa (Sigma) for 2 or 12 hours, followed by wash out of excess maturation (30). dextran with culture medium, examination on a Zeiss HAL100 Lysosomal associated protein transmembrane-4 b fluorescence microscope and photographed with a Zeiss (LAPTM4B) is a putative novel oncogene (34, 35) frequently camera. amplified and overexpressed in primary treatment-naive breast cancers (36). Like other LAPTM family members, Tumorigenicity assay LAPTM4B protein has a lysosome localization motif (34) and Two groups of mice were injected with LAPTM4B expres- colocalizes with markers of late endosomes and lysosomes sion–manipulated MDA468 cells or the control counterpart (37). Increased expression of LAPTM4B has been shown in cells into the inguinal mammary fat pads. Mice were routinely hepatocellular carcinoma and lung, ovarian, uterine, and gas- monitored for health and tumor size; mice were sacrificed tric cancers (35, 38–40). Overexpression of the LAPTM4B-35 when tumors reached 2 cm. Tumor growth rates were mea- isoform in hepatocellular carcinoma cell lines promotes apo- sured and compared between LAPTM4B-manipulated and ptosis resistance in vitro and growth and metastasis of hepa- control tumors. LAPTM4B expression level and the autopha- tocellular carcinoma xenografts in mice (41). We reported gosome marker LC3 were measured by immunoblot in tumor overexpression of LAPTM4B in primary breast tumors is lysates prepared from xenograft tumors. associated with resistance to chemotherapy, specifically anthracyclines, and may serve as a predictive biomarker for Statistical analysis distant recurrence in patients treated with adjuvant chemo- Difference in cell viability, tumor size, and ratios of LC3II/ therapy (36). By sequestering drug in the cytoplasmic com- LC3I, or of p62/actin between control and testing groups was partment and enhancing efflux of drugs from cancer cells, evaluated by the Student t test and P values derived from this overexpression of LAPTM4B decreases nuclear localization of test were used to determine the significance of the difference. drug and drug-induced DNA damage and thereby reduces drug Additional methods are described in Supplementary effectiveness (36, 42). However, the potential mechanisms by Material. which LAPTM4B promotes tumor growth and survival in treatment-naive cancers is not well studied. Results This study tests the effect of modulating LAPTM4B expression on cell survival under various stress conditions including LAPTM4B localizes to lysosomes and downregulation of nutrient deprivation, pH alteration, hypoxia, chemotherapy LAPTM4B triggers LMP exposure, and in vivo tumor growth. Our results show that high LAPTM4B is a member of a family of proteins which contain expression of LAPTM4B inhibits lysosome-mediated death a lysosome targeting motif at the C terminus (34). To deter- pathways, promotes autophagy, and leads to stress tolerance, mine lysosome localization in mammary epithelial cells, we thereby enhancing tumor cell growth and survival. LAPTM4B examined HMECs stably expressing exogenous 6-His epitope– may be an important new therapeutic target for inhibiting tagged LAPTM4B by immunofluorescence microscopy and cancer growth or sensitizing tumors to chemotherapy. showed expression of His-LAPTM4B in the lysosomal compartment defined by the lysosome-marker Lysotracker DND- Materials and Methods 99 (Supplementary Fig. S1A). To understand how LAPTM4B might modulate lysosome Cell lines, siRNA transfection, and gene transfer function, we investigated the relationship between LAPTM4B Breast cell lines BT549 and MDA-MB-468 were obtained expression and LMP. LMP was evaluated by measuring endo- from American Type Culture Collection and were authenti- cytic uptake and release of a nondigestible fluid-phase sub- cated by Promega PowerPlex 1.2 short tandem repeat profiling. strate, FITC-dextran. In BT549 breast cancer cells transfected Tert-immortalized human mammary epithelial cells (HMEC) with a scramble control siRNA, after 2 hours of exposure to were from Clontech and provided by the Zhao laboratory. FITC-dextran, the fluorescent dextran particles were detected BT549 cancer cells have amplification and overexpression of in a punctate cytoplasmic pattern consistent with lysosomes LAPTM4B (43) and were used for siRNA knockdown experi- and were retained in the lysosomal compartment 12 hours ments. MDA-MB-468 (MDA468) cancer cells have normal DNA later (Fig. 1A; Supplementary Fig. S2A and S2B). In BT549 cells

7482 Cancer Res; 71(24) December 15, 2011 Cancer Research

LAPTM4B Promotes Autophagy and Cell Tolerance to Stress

Control siRNA-LAPTM4B Figure 1. Suppression of LAPTM4B A B expression triggers LMP. A and B, 2 h 12 h 2 h 12 h merged fluorescence analysis for dextran (FITC, green) and nuclear 4kDa staining (DAPI, blue) to show intracellular distribution of 4 kDa (top row), 40 kDa (middle row), or 70 kDa (bottom row) dextran particles. A, 40kDa BT549 cells transfected with a scrambled control siRNA. B, BT549 cells transfected with LAPTM4B- specific siRNA. The times after dextran exposure are indicated on 70kDa top. C and D, merged fluorescence analysis for doxorubicin (drug autofluorescence, red), lysosomes C 2 h D 24 h (Lysotracker DND-26, green), and nuclear staining (DAPI, blue) in BT549 cells transfected with control Control siRNA (Ctrl, top row) and with LAPTM4B-specific siRNA (siRNA, bottom row) after 2 or 24 siRNA- hours of exposure to doxorubicin. LAPTM4B

transfected with LAPTM4B-specific siRNA, the fluorescence cells treated with LAPTM4B-siRNA, after 12 to 24 hours, pattern was similar to control cells at 2 hours; however, after 12 doxorubicin has been redistributed and predominantly colo- hours, the 4 and 40 kDa FITC-dextran molecules were in a calizes with nuclear 40,6-diamidino-2-phenylindole (DAPI) diffuse distribution throughout the cell consistent with release staining (Fig. 1D, bottom). This result suggests that knockdown from the lysosomal compartment (Fig. 1B; Supplementary Fig. of LAPTM4B does not completely abolish the uptake of doxo- S2C and S2D). The distribution of the larger molecular weight rubicin by cytoplasmic organelles but significantly weakens the 70 kDa FITC-dextrans were mostly unchanged after 12 hours capability of lysosomes to retain the drug, resulting in its more (Fig. 1B, bottom). These results suggest that LAPTM4B exerts rapid release and redistribution to the nucleus. lysosome-stabilizing properties to retain low and intermediate Doxorubicin is also a weak base that can be trapped in molecular weight macromolecules. Doxorubicin is small mol- acidified lysosomes. To examine whether LAPTM4B expres- ecule with a molecular weight of approximately 580 Da. As sion has an effect on lysosomal pH, we used a pH sensing dye, shown in Fig. 1C, after 2 hours of drug exposure, doxorubicin LysoSensor Green DND-189, which is taken up into lysosomes autofluorescence is colocalized with Lysotracker in both the and fluoresces with greater intensity in lower pH. As shown control-treated and LAPTM4B-siRNA–treated BT549 cells in Fig. 2A, BT549 cells treated with LAPTM4B-siRNA have a consistent with initial uptake of drug into lysosomes. Doxo- higher lysosomal pH (lower fluorescent intensity) than control rubicin is retained in the lysosomal compartment 24 hours cells. The degree of increase in lysosomal pH is similar to BT549 later in the control siRNA–treated or untreated BT549 cells cells treated with 50 mmol/L chloroquine, a lysosomotropic (Fig. 1D, top; Supplementary Fig. S1B). However, in the BT549 agent that increases the lysosomal pH (Fig. 2B). However,

Figure 2. LAPTM4B depletion changes acidification of lysosomes. ACB Cell counts (y-axis) by fluorescence intensity (x-axis, log scale) of cells stained with LysoSensor Green DND-189. Cell treatments are indicated as follows: purple filled area, cells stained with PBS (1% bovine serum albumin); green line, cells transfected with siRNA-control or empty vector; pink line, BT549 cells transfected with siRNA- LAPTM4B (A) or treated with 50 Control Control Control mmol/L chloroquine for 24 hours (B), siRNA- Chloroquine LAPTM4B or SKBR3 cells with forced stable LAPTM4B expression of LAPTM4B (C).

www.aacrjournals.org Cancer Res; 71(24) December 15, 2011 7483

Li et al.

A B Ctrl siRNA L4B siRNA Ctrl siRNA L4B siRNA --- Doxo --- Doxo

Cathepsin B Cath inh + Cath inh +

Doxo --- Doxo --- Cath InhCasp inh Casp inh + DoxoCath inhCasp inh Casp inh + Doxo

Cathepsin D Bid tBid

Actin Active caspase-3

cl PARP

Actin

Figure 3. Suppression of LAPTM4B induces cathepsin-dependent cleavage of Bid, caspase-3, and PARP in breast cancer cells that overexpress LAPTM4B. A, BT549 cells transfected with a scrambled control siRNA (Ctrl) or LAPTM4B-speciﬁc siRNA (L4B siRNA) incubated in medium alone (—) or with 1 mmol/L doxorubicin (Doxo). Proteins (100 mg per lane) from cytosol preparations were fractionated on SDS-PAGE; blots were probed with anti-cathepsins B, D, and anti-actin. B, BT549 cells pretreated with cathepsin inhibitors, pepstatin A and EST (Cath inh), or pan-caspase inhibitor z-VAD-FMK (Casp inh) then transfected with control or LAPTM4B-speciﬁc siRNA and treated with medium alone (—)or1mmol/L doxorubicin (Doxo). Proteins (100 mg per lane) from cell lysates were fractionated on SDS-PAGE; blots were probed with anti-Bid, anti-caspase-3, anti-PARP, and anti-actin antibodies.

SKBR3 breast cancer cells with forced stable overexpression of gate caspase-3 activation but did not prevent Bid truncation. LAPTM4B had similar lysosomal pH compared with control Treatment with 50 mmol/L chloroquine alone was not suffi- cells (Fig. 2C). To determine whether change in lysosomal pH cient to induce cathepsin release and resulted in no added alone could account for increased permeability of the lyso- effect to LAPTM4B knockdown on cathepsin release or apo- some, we treated BT549 cells with 50 mmol/L chloroquine to ptosis induction (Supplementary Fig. S3, lanes 4 and 8). How- increase lysosomal pH and found no alteration in the uptake of ever, the release of cathepsin B and D and the activation of dextran at 2 hours and dextran was not released into the caspase-3 were more pronounced in cells also treated with the cytosol at 12 hours (Supplementary Fig. S2E and S2F). These DNA-damaging drug, doxorubicin (Fig. 3A and B; Supplemen- results suggest that LAPTM4B may not directly regulate lyso- tary Fig. S3, lanes 2 and 6). There was no potentiation of somal pH, but the increase of lysosomal pH in LAPTM4B cathepsin release or caspase-3 activation in cells also treated knockdown cells may be a secondary effect of increased LMP with the microtubule-stabilizing agent taxol (Supplementary resulting in deacidification of the lysosome. Fig. S3, lanes 3 and 7), consistent with previous studies showing that taxol-induced apoptosis may be independent of caspase-3 Knockdown of LAPTM4B by siRNA provokes lysosomal and -9 activation (44). These results suggest that a certain level mediated programmed cell death through induction of of LAPTM4B is required to prevent lysosome-mediated initi- cathepsin release ation of apoptosis. This requirement is more notable in the The maintenance of lysosomal membrane integrity is impor- setting of additional cellular insult, such as exposure to certain tant for cell survival. LMP leads to cathepsin release followed chemotherapy agents. by caspase activation. This process initiates apoptosis which triggers further lysosomal destabilization to induce lysosomal LAPTM4B overexpression promotes autophagy mediated cell death in a positive feedback loop (2). To deter- Autophagy plays a critical survival role by supporting mine whether LAPTM4B knockdown led to cathepsin release energy requirements and sustaining viability under adverse from lysosomes, we conducted immunoblot analysis for cathe- conditions and may also promote resistance to chemother- psins on cell cytosol fractions of BT549 tumor cells after apy-induced genotoxic stress. To determine whether mod- transfection with control or LAPTM4B-specific siRNA. Deple- ulation of LAPTM4B expression has an effect on autophagy, tion of LAPTM4B resulted in appearance of cathepsin B and we examined markers of autophagosome maturation and cathepsin D in the cytosol of cells (Fig. 3A) and resulted in Bid flux during starvation-induced autophagy in BT549 breast truncation, caspase-3 activation, and PARP cleavage (Fig. 3B) cancer cells with or without depletion of LAPTM4B. In consistent with initiation of apoptosis. Pan inhibitors of cathe- parental BT549 cells (with inherent overexpression of psins (EST and pepstatin A) protected BT549 cells from LAPTM4B), autophagy was induced by starvation as indi- LAPTM4B knockdown–induced Bid truncation and apoptosis. cated by the appearance of punctate LC3 staining in the The pan-caspase inhibitor z-VAD-FMK was sufficient to abro- cytoplasm. Many of the LC3-puncta colocalized with the

7484 Cancer Res; 71(24) December 15, 2011 Cancer Research

LAPTM4B Promotes Autophagy and Cell Tolerance to Stress

A Control LC3 LAMP2 DAPI Merge

Normal

Figure 4. Effects of modulating LAPT4B expression on autophagy maturation and flux. A and B, merged immunofluorescence analysis for Starvation autophagosomes (anti-LC3, red), lysosomes (anti-LAMP2, green), and nuclear staining (DAPI, blue) in BT549 cells transfected with (A) siRNA-LAPTM4B control scramble siRNA and (B) B LAPTM4B-specific siRNA. Cells LC3 LAMP2 DAPI Merge were cultured in nutrient-rich medium (Normal, top of A and B) or in Normal low serum medium (Starvation, bottom of A and B). C, immunoblot of control siRNA–treated BT549 cells in medium alone (—) or with chloroquine (CQ) or LAPTM4B- siRNA treated BT549 in medium Starvation alone (si-L4B). D, immunoblot of MDA468 cells transfected with control vector cultured with medium alone (—) or chloroquine or MDA468 cells transfected LAPTM4B C Normal Starvation D Normal Starvation expression vector cultured in L4B+ L4B+ --- L4B CQ --- L4B CQ medium alone (L4B) or with --- CQ si-L4B --- CQ si-L4B CQ CQ chloroquine (L4B þ CQ). C and D, LC3I LC3I cells were cultured in nutrient-rich LC3II LC3II medium (Normal, left side) or low serum medium (Starvation, right side). Proteins (100 mg per lane) from p62 p62 cell lysates were fractionated on SDS-PAGE; blots were probed with anti-LC3, anti-p62, anti-active Active Actin caspase-3, anti-PARP, and anti- Caspase-3 actin as indicated.

cl PARP

Actin

lysosome marker LAMP2 indicating normal autophagosome some number which can increase due to increased autophago- maturation and fusion to lysosomes (Fig. 4A). In contrast, in some formation or from a block in autophagosome maturation cells in which LAPTM4B has been depleted by siRNA, or both (31). To determine whether autophagy flux was starvation resulted in marked cytoplasmic accumulation of increased or blocked, we measured levels of the autophagy enlarged LC3-positive autophagosomes, some of which were substrate, p62. In control cells, serum starvation increased not colocalized with LAMP2-positive lysosomes (Fig. 4B). autophagy flux as indicated by lower p62 levels in the serum- This suggests that depletion of LAPTM4B resulted in a block starved cells than in cells grown in nutrient-rich medium. in autophagosome–lysosome fusion and blocked autolyso- However, in cells depleted of LAPTM4B, there was no change some formation. in the level of p62 after serum starvation, indicating a block in We next analyzed autophagy in these cells by immunoblot- starvation-induced autophagy flux (Fig. 4C; Supplementary ting for LC3II and p62. The siRNA depletion of LAPTM4B had Fig. S4B). The effect of knockdown of LAPTM4B on LC3II and no effect on the level of LC3II in BT549 cells grown in nutrient- p62 levels was similar to what was observed in serum-starved rich medium but resulted in significant increase in the level of cells treated with chloroquine. In conjunction with the immu- LC3II when cells were stressed by serum starvation (Fig. 4C; nofluorescence findings, these results suggest that LAPTM4B Supplementary Fig. S4A). LC3II is an indicator of autophago- is required for the later stages of autophagy maturation in

www.aacrjournals.org Cancer Res; 71(24) December 15, 2011 7485

Li et al.

which autophagosomes are fused with lysosomes to form autolysosomes. A Abortive accumulation of autophagosomes may act as a cell Nutrient rich Low serum Low glucose ** *** ** death messenger to trigger caspase-dependent or -indepen- 6 6 6 Ctrl dent cell death. We found an increase in cell death as indicated siRNA-L4B by caspase-3 activation and cleaved PARP in LAPTM4B-deplet- 4 4 4 BT549 ed cells with starvation-induced aggregation of autophago- 2 2 2 somes (Fig. 4C). Although downregulation of LAPTM4B alone Ratio of viable cells Ratio of viable Ratio of viable cells Ratio of viable cells Ratio of viable may trigger LMP and to a lesser extent caspase activation, the 0 0 0 01234 01234 01234 combination of LMP and blocked autophagy results in more Days Days Days pronounced caspase activation and cell death. B In a complimentary experiment, stable overexpression of Nutrient rich Low serum Low glucose 8 8 8 * LAPTM4B in MDA468 cells with inherently low levels of ** Ctrl LAPTM4B, resulted in greater starvation-induced autophagy 6 6 6 LAPTM4B as indicated by higher levels of LC3II and lower levels of p62 4 4 4

(Fig. 4D; Supplementary Fig. S4C and S4D). This effect was MDA468 blocked by the autolysosome inhibitor chloroquine. These 2 2 2 results are consistent with the notion that overexpression of cells Ratio of viable 0 cells Ratio of viable 0 cells Ratio of viable 0 ﬂ 0 123 4 0123 4 0 123 4 LAPTM4B promotes increased autophagy ux in cancer cells Days Days Days exposed to metabolic stress. C MDA468 BT549 Overexpression of LAPTM4B promotes tumor cell survival and resistance to apoptosis induced by low serum concentration or glucose deprivation

The previous experiments suggest that overexpression of Nutrient richLow serumLow glucoseNutrient richLow serumLow glucose LAPTM4B may result in decreased sensitivity of tumor cells to Ctrl L4B Ctrl L4B Ctrl L4B Ctrl si-L4B Ctrl si-L4B Ctrl si-L4B insults that trigger LMP, lysosome-mediated programmed cell Active Caspase-3 death, or induce autophagy. We examined the effect of mod-

ulating LAPTM4B expression levels on cell survival in cells cl PARP exposed to various environmental stressors. In BT549 cells with amplification and overexpression of LAPTM4B, siRNA Actin depletion of LAPTM4B resulted in approximately 50% decrease (P ¼ 0.0267) in viable cells when cultured in nutrient-rich medium. Depletion of LAPTM4B resulted in a similar or more Figure 5. Effects of modulating LAPTM4B expression on tolerance to pronounced decrease in viable cells when grown in low serum nutrient stress. Nutrient- and glucose-dependent cell survival curves in or low glucose medium (P ¼ 0.001 and 0.0260, respectively; Fig. (A) BT549 cells transfected with scramble control (––) and LATPM4B- specific siRNA (–*–) and (B) MDA468 cells transfected with control GFP 5A) MDA468 breast cancer cells have inherently low normal vector (––) and LAPTM4B vector (–*–). The ratio of viable cells at each levels of LAPTM4B expression. MDA468 cells with forced time point versus time zero are indicated on y-axis (mean of triplicates overexpression of LAPTM4B had similar cell viability to paren- SD). P values from t test for difference between control (Ctrl) and tal cells when cultured in nutrient-rich media (P ¼ 0.105). LAPTM4B-manipulated cells are indicated as follows: , P < 0.1; , P < P ¼ These results suggest that a certain level of LAPTM4B may be 0.03; , 0.001. C, immunoblot of MDA468 cells transfected with control GFP vector (Ctrl) and LAPTM4B vector (L4B) and BT549 cells required for tumor cell survival, but its overexpression does not transfected with scramble control and LATPM4B-specific siRNA (si- directly promote in vitro cell growth under normal culture L4B). Cell lines are indicated along the top. Proteins (100 mg per lane) from conditions. However, the cells with overexpression of cell lysates were fractionated on SDS-PAGE; blots were probed with anti- LAPTM4B showed higher cell viability than control cells when caspase-3, anti-PARP, and anti-actin antibodies. Cells were cultured in P ¼ nutrient-rich, low serum, and low glucose medium as indicated above the cultured in low serum or low glucose medium ( 0.0256 and panels or lanes. 0.0960, respectively; Fig. 5B). As shown in Fig. 5C, the decreased cell viability of BT549 cells after knockdown of LAPTM4B was associated with increased activation of caspase-3 and PARP had no effect on cell viability when cells were cultured in cleavage. Similarly, the enhanced survival of MDA468 cells acidified medium. Exposure to hypoxic stress showed variable overexpressing LAPTM4B in nutrient-deprived conditions was results. Forced overexpression of LAPTM4B in MDA468 cells associated with much lower levels of active caspase-3 and had no significant effect on cell viability of cells grown in low PARP cleavage. These results support our hypothesis that high oxygen (Supplementary Fig. S5C). However, BT549 cells with expression of LAPTM4B potentiates the growth and survival of LAPTM4B depletion were more sensitive to hypoxia and had breast cancer cells under metabolic stress. reduced viability compared with parental BT549 cells that Next, we tested whether modulation of LAPTM4B affects overexpress LAPTM4B (Supplementary Fig. S5D). survival of cells exposed to other types of stress. As shown in We further examined the effects of LAPTM4B modulation on Supplementary Fig. S5A and S5B, expression levels of LAPTM4B tumor growth and survival in an in vivo xenograft model.

7486 Cancer Res; 71(24) December 15, 2011 Cancer Research

LAPTM4B Promotes Autophagy and Cell Tolerance to Stress

Figure 6. Overexpression of A MDA468 xenograft B LAPTM4B promotes increased autophagy and faster tumor growth 4,000 in vivo. A, tumor growth of MDA468- LAPTM4B ) expressing control vector (––)or 3 3,000 LAPTM4B vector (–*–)in Control immunodeﬁcient mice. B, LAPTM4B immunoblotting for His epitope– 2,000 tagged LAPTM4B in tumor explant Actin lysates derived from MDA468 cells expressing a control vector (left lane) Tumor size (mm Tumor 1,000 or LAPTM4B vector (right lanes). Full- length LAPTM4B (black arrow) and smaller proteolytic fragments (white 0 arrows) indicate delivery of 0102030405060 LAPTM4B to lysosomes. C, Days immunoblotting for LC3 and p62 in tumor explant lysates derived from C D EF cells expressing a control vector (left 80 200 60 lanes, n ¼ 4) or LAPTM4B vector Control LAPTM4B /p62 (right lanes, n ¼ 5). D–F, LC3I 60 150 I quantiﬁcation of LC3II to LC3I ratio, LC3II 40 p62 to actin ratio, and LC3II/LC3I to 40 100 p62 ratio, respectively. Data p62 represent mean þ SD of MDA468 20 20 50 xenograft tumors expressing a control vector (white bars, n ¼ 4) or Actin Ratio of p62/actin Ratio of LC3II/LC3I Ratio of LC3II/LC3 LAPTM4B vector (black bars, n ¼ 5). 0 0 0 Ctrl L4B Ctrl L4B Ctrl L4B

MDA468 cells with stable expression of either a control vector that LAPTM4B may provide a growth or survival advantage to or a vector driving LAPTM4B overexpression were inoculated tumor cells in the absence of a therapy challenge, resulting in into the cleared mammary fat pad of nude mice. After 2 its selection and upregulation in primary (untreated) cancers. months, we found that the tumor xenografts that overex- In this study, we provide evidence that LAPTM4B is localized in pressed LAPTM4B grew much faster than control xenografts lysosomes of mammary cells and promotes lysosome mem- (t test, P ¼ 0.033; Fig. 6A). Examination of the xenograft tumor brane stability. Decreased expression of LAPTM4B leads to explants showed that each continued to overexpress increase in LMP, lysosomal pH, lysosomal release of cathe- LAPTM4B compared with control xenograft tumors (Fig. psins, and lysosome-mediated cell death. This effect is more 6B); the presence of the smaller proteolytic fragments indicates pronounced when cells are exposed to anthracycline chemo- proper delivery of LAPTM4B to the lysosomes (37). We exam- therapy but not when exposed to a taxane, consistent with the ined tumor lysates for markers of autophagy by immunoblot increased sensitivity to anthracyclines in tumors with low (Fig. 6C). LAPTM4B-overexpressing xenografts showed higher LAPTM4B expression (36). We find that LAPTM4B plays a LC3II/LC3I ratio (P ¼ 0.0040; Fig. 6D), decreased p62 levels (P ¼ critical role in autophagy and insufficient levels of LAPTM4B 0.0363; Fig. 6E), and higher LC3II/LC3I/p62 ratio (P ¼ results in a failure of autophagosome–lysosome fusion causing 0.0074; Fig. 6F) consistent with increased autophagic flux. This a block in autolysosome formation and decreased autophagy result suggests that LAPTM4B overexpression promotes tumor flux. In contrast, overexpression of LAPTM4B in breast cancer growth in vivo, perhaps by promoting greater tolerance to cells that have normal DNA copy number of 8q22 and express stress through increased induction of autophagy. normal levels of LAPTM4B results in increased autophagy flux. These observations suggest that the proper function of Discussion LAPTM4B is required for lysosome-mediated autophagy maturation. A recent study showed that transient overexpression The cancer gene LAPTM4B is one of two genes amplified and of LAPTM4B without coordinate increase of the expression of overexpressed from chromosome 8q22 which predicts for de its partner MCOLN1 in retinal epithelial cells led to increased novo anthracycline resistance and metastatic recurrence in LC3II levels because of an accumulation of autophagosomes patients with breast cancer (36). Our previous work suggested (37). This finding could result from increased autophagy one mechanism by which LAPTM4B may promote chemother- induction or a block in autophagy flux but is consistent with apy resistance is by retention of anthracyclines in a cyto- the notion that appropriate levels of LAPTM4B are crucial to plasmic compartment thereby preventing nuclear drug proper lysosome-mediated autophagy. localization and drug-induced DNA damage. The frequent Fusion of lysosome with autophagosome is a critical step for occurrence of 8q amplification and overexpression of autophagy maturation. Rab7, SKD1, Lamp2, UVRAG, and LAPTM4B in treatment-naive cancers raised the possibility Rebicon have been shown to be essential for this process

www.aacrjournals.org Cancer Res; 71(24) December 15, 2011 7487

Li et al.

(5, 29, 30, 37, 45, 46). Frequent frameshift mutations have been As normal tissues are seldom under nutrient deficiency, there identified in the UVRAG gene in colon and gastric cancers (47); may be a significant therapeutic window for inhibiting however, no significant aberrations in autophagy were found in LAPTM4B in tumors that overexpress this gene and are cancer cells carrying this mutation (48). To the best of our dependent on it for survival. Over-reliance of cancer cells on knowledge, LAPTM4B represents the first gene crucial for LAPTM4B for tolerance to stress may represent the "Achilles autophagy maturation and flux that is amplified in cancer heel" for malignancies that have amplified this gene and (36, 49) and is associated with treatment resistance and poor provide a new therapeutic strategy for targeting these cancers. clinical outcome in various cancers (34, 36, 39–41, 49). Metad- herin was recently shown to induce autophagy in different cell Disclosure of Potential Conflicts of Interest lineages by increasing expression of atg5 and activating AMP kinase (50). Interestingly, the gene for metadherin, MTDH,isa A.L. Richardson and Z.C. Wang have filed a patent on the detection of chr fi 8q22 genes including LAPTM4B to predict anthracycline resistance in cancer neighbor to LAPTM4B and is coampli ed with LAPTM4B in (Dana-Farber Cancer Institute and Brigham and Women's Hospital). The other cancers. We propose that the coordinate amplification of the 2 authors disclosed no potential conflicts of interest. genes may activate simultaneously the 2 major stages of the autophagy pathway and thereby enhance tumor survival. Acknowledgments Tumors frequently experience elevated metabolic stress fi The authors thank Lisa Cameron (DF/hepatocellular carcinoma confocal from nutrient and oxygen deprivation due to insuf cient core facility) and members of the Richardson-Wang laboratory for their advice angiogenesis (8). The high metabolic demands of cell prolif- and assistance and Dr. William Kaelin for allowing us to use the hypoxia chamber eration and altered metabolism provide additional tumor cell in his laboratory. intrinsic stress. Thus, tumors may be more dependent on survival mechanisms such as autophagy to maintain growth Grant Support or to disseminate (10, 51). We show that the roles of LAPTM4B This work was supported by Susan G. Komen For the Cure (KG080939; to A.L. in limiting lysosome-mediated cell death and promoting Richardson, Z.C. Wang, Y. Li, R. Tian), Breast Cancer Research Foundation (J.D. autophagy have significant survival effects in cancer cells, Iglehart, A.L. Richardson, Z.C. Wang), Terri Brodeur Breast Cancer Foundation (Y. Li), and friends of Dana-Farber Cancer Institute (Y. Li). The work was also including greater resistance to nutrient deprivation, hypoxia, supported by the NCI Harvard SPORE in Breast Cancer (2 P50 CA89393-06) and a and chemotherapy-induced genotoxic stress. We further show DOD concept award (BC053041). that LAPTM4B prosurvival effect is associated with greater in The costs of publication of this article were defrayed in part by the payment of vivo page charges. This article must therefore be hereby marked advertisement in tumor growth. The new knowledge from this study con- accordance with 18 U.S.C. Section 1734 solely to indicate this fact. tributes to a better understanding of the molecular mechanisms of lysosome-mediated drug resistance and the particular Received March 16, 2011; revised September 7, 2011; accepted October 7, 2011; dependency of tumors on lysosomal function for cell survival. published OnlineFirst October 28, 2011.

References 1. Larsen AK, Escargueil AE, Skladanowski A. Resistance mechanisms 10. Kenific CM, Thorburn A, Debnath J. Autophagy and metastasis: associated with altered intracellular distribution of anticancer agents. another double-edged sword. Curr Opin Cell Biol 2010;22: Pharmacol Ther 2000;85:217–29. 241–5. 2. Johansson AC, Appelqvist H, Nilsson C, Kagedal K, Roberg K, Ollinger 11. Fan QW, Cheng C, Hackett C, Feldman M, Houseman BT, Nicolaides K. Regulation of apoptosis-associated lysosomal membrane permea- T, et al. Akt and autophagy cooperate to promote survival of drug- bilization. Apoptosis 2010;15:527–40. resistant glioma. Sci Signal 2010;3:ra81. 3. Baumgartner HK, Gerasimenko JV, Thorne C, Ashurst LH, Barrow SL, 12. Rubin BP, Debnath J. Therapeutic implications of autophagy-mediat- Chvanov MA, et al. Caspase-8-mediated apoptosis induced by oxi- ed cell survival in gastrointestinal stromal tumor after treatment with dative stress is independent of the intrinsic pathway and dependent on imatinib mesylate. Autophagy 2010;6:1190–1. cathepsins. Am J Physiol Gastrointest Liver Physiol 2007;293: 13. Levine B. Cell biology: autophagy and cancer. Nature 2007;446: G296–307. 745–7. 4. Conus S, Perozzo R, Reinheckel T, Peters C, Scapozza L, Yousefi S, 14. Scarlatti F, Granata R, Meijer AJ, Codogno P. Does autophagy have a et al. Caspase-8 is activated by cathepsin D initiating neutrophil license to kill mammalian cells? Cell Death Differ 2009;16:12–20. apoptosis during the resolution of inflammation. J Exp Med 2008; 15. Eisenberg-Lerner A, Bialik S, Simon HU, Kimchi A. Life and death 205:685–98. partners: apoptosis, autophagy and the cross-talk between them. Cell 5. Burman C, Ktistakis NT. Autophagosome formation in mammalian Death Differ 2009;16:966–75. cells. Semin Immunopathol 2010;32:397–413. 16. Klionsky DJ, Cregg JM, Dunn WA Jr, Emr SD, Sakai Y, Sandoval IV, 6. He C, Klionsky DJ. Regulation mechanisms and signaling pathways of et al. A unified nomenclature for yeast autophagy-related genes. Dev autophagy. Annu Rev Genet 2009;43:67–93. Cell 2003;5:539–45. 7. Rabinowitz JD, White E. Autophagy and metabolism. Science 2010; 17. Yorimitsu T, Klionsky DJ. Autophagy: molecular machinery for self- 330:1344–8. eating. Cell Death Differ 2005;12 Suppl 2:1542–52. 8. Jin S, White E. Role of autophagy in cancer: management of metabolic 18. Wullschleger S, Loewith R, Hall MN. TOR signaling in growth and stress. Autophagy 2007;3:28–31. metabolism. Cell 2006;124:471–84. 9. Degenhardt K, Mathew R, Beaudoin B, Bray K, Anderson D, Chen G, 19. Hanada T, Noda NN, Satomi Y, Ichimura Y, Fujioka Y, Takao T, et al. et al. Autophagy promotes tumor cell survival and restricts necrosis, The Atg12-Atg5 conjugate has a novel E3-like activity for protein inflammation, and tumorigenesis. Cancer Cell 2006;10:51–64. lipidation in autophagy. J Biol Chem 2007;282:37298–302.

7488 Cancer Res; 71(24) December 15, 2011 Cancer Research

LAPTM4B Promotes Autophagy and Cell Tolerance to Stress

20. Sou YS, Waguri S, Iwata J, Ueno T, Fujimura T, Hara T, et al. The 37. Vergarajauregui S, Martina JA, Puertollano R. LAPTMs regulate Atg8 conjugation system is indispensable for proper development lysosomal function and interact with mucolipin 1: new clues for of autophagic isolation membranes in mice. Mol Biol Cell 2008;19: understanding mucolipidosis type IV. J Cell Sci 2010;124: 4762–75. 459–68. 21. Seglen PO, Gordon PB, Holen I. Non-selective autophagy. Semin Cell 38. Liu Y, Zhang QY, Qian N, Zhou RL. Relationship between LAPTM4B Biol 1990;1:441–8. gene polymorphism and susceptibility of gastric cancer. Ann Oncol 22. Juhasz G, Neufeld TP. Autophagy: a forty-year search for a missing 2007;18:311–6. membrane source. PLoS Biol 2006;4:e36. 39. Yang H, Xiong FX, Lin M, Yang Y, Nie X, Zhou RL. LAPTM4B-35 23. Kabeya Y, Mizushima N, Ueno T, Yamamoto A, Kirisako T, Noda T, overexpression is a risk factor for tumor recurrence and poor prognosis et al. LC3, a mammalian homologue of yeast Apg8p, is localized in in hepatocellular carcinoma. J Cancer Res Clin Oncol 2010;136:275–81. autophagosome membranes after processing. EMBO J 2000;19: 40. Meng FL, Yin MZ, Song HT, Yang H, Lou G, Zhou RL. LAPTM4B-35 5720–8. overexpression is an independent prognostic marker in endometrial 24. Klionsky DJ, Cuervo AM, Seglen PO. Methods for monitoring autop- carcinoma. Int J Gynecol Cancer 2010;20:745–50. hagy from yeast to human. Autophagy 2007;3:181–206. 41. Yang H, Xiong F, Wei X, Yang Y, McNutt MA, Zhou R. Overexpression 25. Klionsky DJ, Abeliovich H, Agostinis P, Agrawal DK, Aliev G, Askew of LAPTM4B-35 promotes growth and metastasis of hepatocellular DS, et al. Guidelines for the use and interpretation of assays for carcinoma in vitro and in vivo. Cancer Lett 2010;294:236–44. monitoring autophagy in higher eukaryotes. Autophagy 2008;4: 42. Li L, Wei XH, Pan YP, Li HC, Yang H, He QH, et al. LAPTM4B: a novel 151–75. cancer-associated gene motivates multidrug resistance through efflux 26. Razi M, Chan EY, Tooze SA. Early endosomes and endosomal coat- and activating PI3K/AKT signaling. Oncogene 2010;29:5785–95. omer are required for autophagy. J Cell Biol 2009;185:305–21. 43. Neve RM, Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, et al. A 27. Stromhaug PE, Seglen PO. Evidence for acidity of prelysosomal collection of breast cancer cell lines for the study of functionally distinct autophagic/endocytic vacuoles (amphisomes). Biochem J 1993; cancer subtypes. Cancer Cell 2006;10:515–27. 291:115–21. 44. Ofir R, Seidman R, Rabinski T, Krup M, Yavelsky V, Weinstein Y, et al. 28. Dunn WA Jr. Studies on the mechanisms of autophagy: maturation of Taxol-induced apoptosis in human SKOV3 ovarian and MCF7 breast the autophagic vacuole. J Cell Biol 1990;110:1935–45. carcinoma cells is caspase-3 and caspase-9 independent. Cell Death 29. Eskelinen EL. Maturation of autophagic vacuoles in Mammalian cells. Differ 2002;9:636–42. Autophagy 2005;1:1–10. 45. Liang C, Lee JS, Inn KS, Gack MU, Li Q, Roberts EA, et al. Beclin1- 30. Mehrpour M, Esclatine A, Beau I, Codogno P. Overview of macro- binding UVRAG targets the class C Vps complex to coordinate autop- autophagy regulation in mammalian cells. Cell Res 2010;20:748–62. hagosome maturation and endocytic trafficking. Nat Cell Biol 2008; 31. Rubinsztein DC, Cuervo AM, Ravikumar B, Sarkar S, Korolchuk V, 10:776–87. Kaushik S, et al. In search of an "autophagomometer". Autophagy 46. Matsunaga K, Saitoh T, Tabata K, Omori H, Satoh T, Kurotori N, et al. 2009;5:585–9. Two Beclin 1-binding proteins, Atg14L and Rubicon, reciprocally 32. Pankiv S, Clausen TH, Lamark T, Brech A, Bruun JA, Outzen H, et al. regulate autophagy at different stages. Nat Cell Biol 2009;11:385–96. p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degradation of 47. Kim MS, Jeong EG, Ahn CH, Kim SS, Lee SH, Yoo NJ. Frameshift ubiquitinated protein aggregates by autophagy. J Biol Chem 2007; mutation of UVRAG, an autophagy-related gene, in gastric carcinomas 282:24131–45. with microsatellite instability. Hum Pathol 2008;39:1059–63. 33. Bjorkoy G, Lamark T, Pankiv S, Overvatn A, Brech A, Johansen T. 48. Knaevelsrud H, Ahlquist T, Merok MA, Nesbakken A, Stenmark H, Monitoring autophagic degradation of p62/SQSTM1. Methods Enzy- Lothe RA, et al. UVRAG mutations associated with microsatellite mol 2009;452:181–97. unstable colon cancer do not affect autophagy. Autophagy 2010;6: 34. Shao GZ, Zhou RL, Zhang QY, Zhang Y, Liu JJ, Rui JA, et al. Molecular 863–70. cloning and characterization of LAPTM4B, a novel gene upregulated in 49. Hu G, Chong RA, Yang Q, Wei Y, Blanco MA, Li F, et al. MTDH hepatocellular carcinoma. Oncogene 2003;22:5060–9. activation by 8q22 genomic gain promotes chemoresistance and 35. Kasper G, Vogel A, Klaman I, Grone J, Petersen I, Weber B, et al. The metastasis of poor-prognosis breast cancer. Cancer Cell 2009;15: human LAPTM4b transcript is upregulated in various types of solid 9–20. tumours and seems to play a dual functional role during tumour 50. Bhutia SK, Kegelman TP, Das SK, Azab B, Su ZZ, Lee SG, et al. progression. Cancer Lett 2005;224:93–103. Astrocyte elevated gene-1 induces protective autophagy. Proc Natl 36. Li Y, Zou L, Li Q, Haibe-Kains B, Tian R, Li Y, et al. Amplification of Acad Sci U S A 2010;107:22243–8. LAPTM4B and YWHAZ contributes to chemotherapy resistance and 51. White E, DiPaola RS. The double-edged sword of autophagy modu- recurrence of breast cancer. Nat Med 2010;16:214–8. lation in cancer. Clin Cancer Res 2009;15:5308–16.

www.aacrjournals.org Cancer Res; 71(24) December 15, 2011 7489

Lysosomal Transmembrane Protein LAPTM4B Promotes Autophagy and Tolerance to Metabolic Stress in Cancer Cells

Yang Li, Qing Zhang, Ruiyang Tian, et al.

Cancer Res 2011;71:7481-7489. Published OnlineFirst October 28, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-11-0940

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2011/10/28/0008-5472.CAN-11-0940.DC1

Cited articles This article cites 51 articles, 12 of which you can access for free at: http://cancerres.aacrjournals.org/content/71/24/7481.full#ref-list-1

Citing articles This article has been cited by 8 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/71/24/7481.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/71/24/7481. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.