autophagy governs protumorigenic effects of mitotic ...max (invitrogen) according to...

Cell Cycle and Senescence

Autophagy Governs Protumorigenic Effects ofMitotic Slippage–induced SenescenceRekha Jakhar1, Monique N.H. Luijten1, Alex X.F.Wong1, Bing Cheng1, Ke Guo1,Suat P. Neo2, Bijin Au3, Madhura Kulkarni1, Kah J. Lim4, Jiamila Maimaiti4,Han C. Chong1, Elaine H. Lim5, Tee B.K. Tan6, Kong W. Ong6, Yirong Sim6, Jill S.L.Wong7,James B.K. Khoo7, Juliana T.S. Ho7, Boon T. Chua8, Indrajit Sinha4, Xiaomeng Wang1,9,John E. Connolly3,10,11, Jayantha Gunaratne2,12, and Karen C. Crasta1,13,14,15

Abstract

The most commonly utilized class of chemotherapeuticagents administered as a first-line therapy are antimitoticdrugs; however, their clinical success is often impededby chemoresistance and disease relapse. Hence, a better under-standing of the cellular pathways underlying escape from celldeath is critical. Mitotic slippage describes the cellular processwhere cells exit antimitotic drug-enforced mitotic arrest and"slip" into interphase without proper chromosome segrega-tion and cytokinesis. The current report explores the cell fateconsequence following mitotic slippage and assesses a majoroutcome following treatment with many chemotherapies,therapy-induced senescence. It was found that cells postslip-page entered senescence and could impart the senescence-associated secretory phenotype (SASP). SASP factor produc-tion elicited paracrine protumorigenic effects, such as migra-tion, invasion, and vascularization. Both senescence and SASPfactor development were found to be dependent on autop-

hagy. Autophagy induction during mitotic slippage involvedthe autophagy activator AMPK and endoplasmic reticulumstress response protein PERK. Pharmacologic inhibition ofautophagy or silencing of autophagy-related ATG5 led to abypass of G1 arrest senescence, reduced SASP-associated para-crine tumorigenic effects, and increased DNA damage after S-phase entry with a concomitant increase in apoptosis. Con-sistent with this, the autophagy inhibitor chloroquine andmicrotubule-stabilizing drug paclitaxel synergistically inhib-ited tumor growth in mice. Sensitivity to this combinatorialtreatmentwas dependent on p53 status, an important factor toconsider before treatment.

Implications: Clinical regimens targeting senescence andSASP could provide a potential effective combinatorial strat-egy with antimitotic drugs. Mol Cancer Res; 16(11); 1625–40.�2018 AACR.

IntroductionAntimitotic drugs are used extensively in the treatment of a

variety of malignancies (1). The most commonly utilized class ofantimitotic drugs are the microtubule poisons. These drugs inter-fere with cellular proliferation by disrupting microtubules (MT),thereby inducing a mitotic arrest that culminates in mitotic celldeath (MCD; ref. 2). Unfortunately, the clinical success ofMT-targeting drugs is often impeded by chemoresistance anddisease relapse. Thus far, the majority of studies addressingmechanisms underlying therapy resistance have focused on

tubulin mutations and drug efflux pumps, and have yet to yieldimproved outcomes for patients. Hence, a better understanding ofpossible cellular pathways and alternativemolecularmechanismsunderlying escape from cell death are critical to address tumorprogression following MT-targeting drug treatment.

An alternative outcome to MCD following prolonged mitoticarrest is mitotic slippage. Here, cells prematurely exit frommitosis and enter a pseudo-G1 phase without proper chromo-some segregation and cytokinesis. Postslippage cells are oftencharacterized as tetraploid multinucleated cells due to nuclear

1Lee Kong Chian School of Medicine, Nanyang Technological University, Singa-pore. 2Quantitative Proteomics Group, Institute of Molecular and Cell Biology,Agency for Science, Technology, and Research, Singapore. 3TranslationalImmunology Group, Institute of Molecular and Cell Biology, Agency for Science,Technology, and Research, Singapore. 4Acenzia Inc, Windsor, Ontario, Canada.5Division of Medical Oncology, National Cancer Centre Singapore, Singapore.6Division of Surgical Oncology, National Cancer Centre Singapore, Singapore.7Division of Oncologic Imaging, National Cancer Centre Singapore, Singapore.8Singapore Oncogenome Program, Institute of Molecular and Cell Biology,Agency for Science, Technology, and Research, Singapore. 9Vascular BiologyGroup, Institute of Molecular and Cell Biology, Agency for Science, Technology,and Research, Singapore. 10Department of Microbiology and Immunology, YongLoo Lin School of Medicine, National University of Singapore, Singapore.11Institute of Biomedical Studies, Baylor University, Waco, Texas. 12Departmentof Anatomy, Yong Loo Lin School of Medicine, National University of Singapore,Singapore. 13School of Biological Sciences, Nanyang Technological University,Singapore. 14Genomic Instability and Cancer Group, Institute of Molecular

and Cell Biology, Agency for Science, Technology, and Research, Singapore.15Department of Medicine, Imperial College London, London, United Kingdom.

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

R. Jakhar, M.N.H. Luijten, and A.X.F. Wong contributed equally to this article.

Current address for M. Kulkarni: Transnational Cancer Research Centre,Prashanti Cancer Care Mission, Pune and Indian Institute of Science Educationand Research, Pune, India.

Corresponding Author: Karen C. Crasta, Nanyang Technological University, LeeKong Chian School of Medicine, 59 Nanyang Drive, Experimental MedicineBuilding, #04-12-01, Singapore 636921. Phone: 656-592-7870; Fax: 656-339-2889; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-18-0024

�2018 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org 1625

on November 15, 2020. © 2018 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst July 23, 2018; DOI: 10.1158/1541-7786.MCR-18-0024

http://crossmark.crossref.org/dialog/?doi=10.1158/1541-7786.MCR-18-0024&domain=pdf&date_stamp=2018-10-16

http://mcr.aacrjournals.org/

envelope assembly around condensed scattered chromosomes(3). Cells tend to undergo mitotic slippage rather than MCDwhen cyclin B1 degradation precedes proapoptotic signal accu-mulation during prolonged mitotic arrest, as posited by theprevailing "competing networks-threshold" model (4). Therecan be several cell fates following mitotic slippage. One pos-sible outcome is cell death postslippage, that fulfils the cyto-toxic goal of therapy in addition to MCD (5). Cells can alsocontinue to proliferate as genomically unstable cells (5), there-by constituting a potential source of chemoresistance. In addi-tion, cells have been shown to arrest in the next interphasepostslippage and eventually enter cellular senescence (5).Indeed, cellular senescence has emerged as a major outcomeof a variety of chemotherapies in a process known as therapy-induced senescence (6).

Classically, senescence is considered to be a barrier againsttumorigenesis, as it restricts cell proliferation (7). This ismediatedby the senescent cell secretome, known as the senescence-associ-ated secretory phenotype (SASP), which consists of a variety ofcytokines, chemokines, growth factors, and matrix metallopro-teases (8). In addition to reinforcing stable growth arrest via bothautocrine and paracrine signaling, SASP factors also promoteimmunosurveillance of senescent cells, leading to tumor remis-sion (9). In this way, senescence serves as an important tumor-suppressive mechanism. However, senescent cells also possessoncogenic potential via paracrine effects of the SASP. SASP factorshave been shown to engender protumorigenic effects such ascellular motility (invasion, migration, and metastasis), epitheli-al–mesenchymal transition (EMT), proliferation, angiogenesis, aswell as inflammation in neighboring cells (8).

Molecular mechanisms underlying senescence followingmitotic slippage and their consequential cell fate significancefor MT-targeting therapies have not been well-explored. Here,we examine the senescent cell fate following nocodazole(MT-destabilizing agent) or paclitaxel (MT-stabilizing agent)-induced mitotic slippage. Our results demonstrate that mitoticslippage–induced senescence could confer paracrine tumori-genic effects via the SASP. We sought to improve therapeuticoutcomes and identified ER stress–triggered autophagy as aneffector of mitotic slippage–induced senescence. We foundthat inhibition of autophagy via pharmacologic means orsilencing of autophagy-associated genes could override senes-cence, leading to cell death upon antimitotic drug treatment.Importantly, combination treatment of tumors in mice withoften-used paclitaxel and autophagy inhibitor chloroquineshowed significant tumor growth arrest, underscoring autop-hagic inhibition as a potential strategy of tackling resistance.We further found that xenograft tumors with wild-type p53exhibited a superior response toward combinatorial therapycompared with tumors with p53 knockout. This suggests thatsensitivity toward this treatment is dictated by p53 status,which could serve as a potential biomarker in predictingclinical response.

Materials and MethodsCell culture and reagents

U2OS, HCT116, MCF7, MDA-MB-231, PanC1, MIA PaCa2,HEK293T, and hTERT-RPE-1 cells were purchased from ATCCat the start of the project in 2014. All cell lines were cultured inDMEM supplemented with 10% FBS (Hyclone GE), with the

exception of hTERT-RPE-1 (DMEM-F12). Cells were freshlythawed monthly and Mycoplasma testing was performed usingEZ-PCR Mycoplasma Test Kit (Biological Industries). Reagentsused in this study are included as Supplementary Tables S2and S3.

siRNA transfectionCells were transfected with siRNA using Lipofectamine RNAi-

MAX (Invitrogen) according to manufacturer's instructions. Cellswere treated with drugs for 16 hours after transfection. siRNAtargeting scrambled sequence (#D-001810-10-05) and ATG5(#M-004374-04) were purchased from Dharmacon and usedaccording to the manufacturer's instructions.

Senescence-associated b-galactosidase stainingCells were stained using the senescence b-galactosidase

(SA-b-gal) Staining Kit (#9860; Cell Signaling Technology)according to manufacturer's instructions.

BrdU incorporation assayCells were incubated with BrdU (BrdU Labeling and Detection

Kit, Roche) for 16 hours and visualized according to the manu-facturer's instructions.

ImmunoblottingCells were lysed in RIPA buffer (Pierce) containing Protease/

Phosphatase Inhibitor Cocktail (Thermo Scientific) and equalamounts of total protein were subjected to SDS-PAGE.

Flow cytometryAfter fixation in ice-cold 70% ethanol overnight, cells were

stained with propidium iodide, and incubated with phospho-Histone H3 antibody conjugated to Alexa Fluor 488 (CellSignaling Technology). Samples were analyzed using Accuri C6flow cytometer (BD Biosciences).

ImmunofluorescenceCells werefixed in 4%paraformaldehyde in PBS for 10minutes

at room temperature. Antibody incubation and staining wereperformed as described previously (10).

Live-cell imagingCells were imaged on a Nikon inverted fluorescence micro-

scope at multiple sites every 30 minutes for 48–72 hours using aPlan Apo 40� objective.

Preparation of conditioned mediaCellswere treatedwithdrugs for twodays followedby culture in

drug-free media containing 0.5% FBS for additional two days.Culturemediawere then centrifuged at 5,000� g, filtered through0.22-mm pore filter (Pall Corporation) and mixed with mediacontaining 40% FBS in a proportion of 3:1 to generate condi-tioned media (CM) containing 10% FBS.

Clonogenic assayCells were treated with drugs for 72 hours and reseeded at

5,000 cells per 6-well, followed by culture for 10 days with freshmedia supplemented every other day. Colonies were stainedwith 0.05% crystal violet (Sigma). Number of colonies wasanalyzed using GelCount (Oxford Optronix) according tomanufacturer's instructions.

Jakhar et al.

Mol Cancer Res; 16(11) November 2018 Molecular Cancer Research1626




Cell viability assayCells were seeded at 4,000 cells per 96-well and treated with

drugs for 72 hours. Cell viability was assessed using CellTiter96 One Solution Proliferation Assays Kit (G3580; Promega)according to manufacturer's instructions.

RNA extraction and qRT-PCRTotal RNAwas extracted using RNeasy plusMini Kit (QIAGEN)

according to manufacturer's instructions. cDNA was reverse tran-scribed using iScript RT Supermix (Bio-Rad) and subjected toSYBR Kit (#4472942; Life Technologies). Relative expressionvalues of each gene were normalized to GAPDH expression.Primers used are in Supplementary Table S4.

Multiplex cytokine analysisForty-one analytes from Human Cytokine Panel 1 (Merck

Millipore) were measured as per manufacturer's instructions.Plates were washed using Tecan Hydrospeed Washer (Tecan) andread with Flexmap 3D system (Luminex Corp). Data were ana-lyzed using Bio-Plex manager 6.0 (Bio-Rad) with a 5-parametercurve-fitting algorithm applied for standard curve calculations.

Tumorigenic phenotypic assaysFor scratch wound migration assay, cells were incubated with

CM for two days. At 90% confluency, the cell monolayer wasscratched and the wound closure rate tracked. For cell invasionassay, cells expressing H2B-GFP were incubated with CM for twodays, plated on top surface of transwell filter chambers precoatedor uncoatedwithMatrigel (BDBiosciences), and the percentage ofinvasive cells determined after 24 hours. For choroid angiogenesisassay, segments of the peripheral choroid layer from eyes of P3mice were incubated with CM 1:3 diluted with EGM2 media(Lonza) over four days and imaged under phase contrast.

Quantification of cellular migration and metastasis inzebrafish (ZgraftTM)

U2OS H2B-GFP cells were incubated with CM for two daysand stained with DiI (Vybrant, Life Technologies) before injec-tion into zebrafish embryos. Embryos were imaged 48 hoursafter injection to determine the metastatic tumor foci positionrelative to injection site.

Human tumor xenograftsAll animal studies were approved by the Institutional Animal

Care and Use Committee (IACUC no. 151103) of the BiologicalResource Centre and were carried out under the policies from theAnimal Facility Centre of the Agency for Science, Technology andResearch (A�STAR, Singapore). HCT116 colon cancer cells wereinjected subcutaneously into both dorsal flanks of female BALB/cathymic nude mice (nu/nu) (InVivos; n ¼ 5 mice in each treat-ment group). When tumor volume reached around 300 mm3,mice were injected intraperitoneally with respective treatmentevery three days for two weeks. Tumor volume was measuredevery three days from start of treatment.Micewere sacrificed at theend of the treatment schedule and all tumors were harvested forweight measurement, IHC, and immunofluorescence.

Patient breast cancer tissuesThis studywas conducted in accordancewith recognized ethical

guidelines (Singapore Guideline for Good Clinical Practice,Declaration of Helsinki and Belmont Report) and approved by

Nanyang Technological University Institutional Review Board(IRB-2018-04-009). Written informed consents were obtainedfrom participating patients. Biopsies were collected using coreneedle biopsy procedure and were snap-frozen in liquid nitrogenimmediately after aspiration. Frozen biopsies were thawed slow-ly, fixed, and processed accordingly for IHC.

IHCAfter fixation, sections were incubated with primary antibodies

overnight at 4�C, followed by incubation with biotinlyated sec-ondary antibodies (Vector Laboratories). Sections were thenincubated in Avidin:Biotinylated Enzyme Complex (ABC; VectorLaboratories) for 30 minutes, developed with 3,30-diaminoben-zidine (DAB) substrate (Vector Laboratories), and nuclei counter-stained with hematoxylin. Slides weremounted with Fluka Eukittquick-hardening mounting medium (Sigma-Aldrich).

SILAC labelingRPE-1 cells were cultured in SILAC DMEM/F12 media, con-

taining light or heavy arginine and lysine ["R0K0 DMEM/F12"with 12C6-L-arginine (Sigma-Aldrich) and 12C6-L-lysine (Sigma-Aldrich) or "R10K8 DMEM/F12" with 13C6-L-arginine(Cambridge Isotope) and 13C6-L-lysine (Cambridge Isotope)]and supplemented with dialyzed FBS. SILAC labeling wasperformed as described previously (11). After lysis in RIPAbuffer (Thermo Scientific) with Protease/Phosphatase InhibitorCocktail (Thermo Scientific), equal amounts of protein fromR0K0 and R10K8 were mixed and boiled with loading buffer at95�C for 5 minutes. Protein complexes were separated by one-dimensional 4%–12% NuPage Novex Bis–Tris Gel (Invitrogen),stained with Colloidal Blue Staining Kit (Invitrogen), and sub-jected to in-gel digestion as described previously (11).

Additional Materials and Methods can be found under Sup-plementary Methods.

ResultsPostslippage tetraploid cells can enter senescence

To study the fate of cells postslippage, we chose mitotic slip-page–prone cells, namely osteosarcoma U2OS cells, colorectalcarcinoma HCT116 cells, and nontransformed telomerase-immortalized hTERT-RPE-1 cells (hereafter referred to as RPE-1).Mitotic slippage, instead of MCD, was previously found to be thepreferred pathway in these cells upon treatment with MT poisons(4). To achieve maximum homogeneity in the cell population,U2OS and RPE-1 cells were first synchronized before treatmentwith the microtubule-depolymerizing drug, nocodazole for aprolonged period of 72 hours to allow cells to undergo mitoticslippage (experimental scheme in Supplementary Fig. S1A). Cellswere monitored by time-lapse microscopy from the start ofnocodazole treatment (t ¼ 0 hours). U2OS and RPE-1 cellspredominantly displayed a "rounded up" morphology, typicalof mitotic arrest, at 16 hours post-nocodazole treatment (Sup-plementary Fig. S1A). By 72 hours post-nocodazole treatment,70.9% of U2OS cells and 87.8% of RPE-1 cells were observed tohave undergone blebbing and reflattening without cytokinesis(characteristics ofmitotic slippage). Thiswas further confirmedbytime-course experiments in synchronized nocodazole-treatedU2OS cells that showed decreased levels of mitotic cyclin B1,mitotic marker phosphorylated Histone H3, and spindle assem-bly checkpoint markers BubR1 and Bub1 by immunoblotting

Mitotic Slippage–induced Autophagy Facilitates Tumorigenesis

www.aacrjournals.org Mol Cancer Res; 16(11) November 2018 1627




Figure 1.

Postslippage cells elicit SASP that confers tumorigenic effects. A, Experimental scheme and representative images of U2OS and RPE-1 cells treated withnocodazole (Noc) for 0, 3, 6, and 9 days and stained for SA-b-gal. Plot shows percentage of SA-b-gal–positive cells. (Continued on the following page.)

Jakhar et al.





(Supplementary Fig. S1B). In addition, a mitigated decline inlevels of antiapoptotic protein Mcl-1 and corresponding increasein cell deathmarker cleaved PARPwas detected. An accumulationof multinucleated tetraploid G1 U2OS cells was also observed, asshown by 4C-G1 population by flow cytometry and multiplenuclei by phase-contrast microscopy (Supplementary Fig. S1Cand S1D). A similar increase in multinucleation was seen innocodazole-treated HCT116 cells postslippage (SupplementaryFig. S1E).

Because cells following prolonged nocodazole treatment dis-played an enlarged, flattened morphology reminiscent of senes-cence, we stained cells for the SA-b-gal. As shown in Fig. 1A, at day3 post-nocodazole treatment, about 45% of U2OS cells and 70%of RPE-1 cells stained positive for SA-b-gal, indicating entry intosenescence. Negligible BrdU staining in multinucleated cellscompared with control indicating a lack of cell proliferationfurther supporting this finding (Supplementary Fig. S1F). Toconfirm this as a bona fide senescence phenotype, nocodazolewas removed and cells were cultured in fresh media for anadditional 6 days (experimental scheme in Fig. 1A). A progressiveincrease in the number of cells that entered senescence wasobserved (Fig. 1A), confirming that cells underwent a stablecell-cycle arrest. Loss of lamin B1 has been associated withsenescence (12). Our data show a decrease in lamin B1 levelsafter day 3 (Fig. 1B). Increased cyclin-dependent kinase inhibitorp21, tumor suppressors p53 and total retinoblastoma protein(pRb), and hypophosphorylation of pRb at Ser780 (13) com-paredwith control further confirmed the senescence phenotype atday 3 (Fig. 1B). Interestingly, we observed a decline in p53 levelsfrom day 6 onwards even though the senescence phenotypepersisted. This is consistentwith a report showingdownregulationof p53 to be crucial for induction of SASP (14). By inference,because p21 is part of the p53–p21 senescence pathway wherep53 activates p21, this could also explain the correspondingdecrease observed for p21 (15). Taken together, our resultsconfirmed that cells postslippage entered senescence followingG1 arrest.

To determine whether antimitotic drugs broadly induce senes-cence, cells were treated with the MT-stabilizing drug paclitaxel(PTX), Aurora kinase B inhibitor ZM447439 (ZM), or kinesin-related Eg5 inhibitor Monastrol (Mon). SA-b-gal staining showeddiscernible differences in the extent of senescence, as paclitaxeland ZM significantly increased the percentage of SA-b-gal–positive cells while Mon-treated cells hardly displayed signs ofsenescence (Supplementary Fig. S2A). Increased SA-b-gal stainingwas also observed for nocodazole- and paclitaxel-treatedHCT116

cells (Supplementary Fig. S2B). Interestingly, treatments thatresulted in senescence, namely paclitaxel and ZM, showed highdegrees ofmultinucleation, whereasMon-treated cells weremost-ly mononucleated (Supplementary Fig. S2C). This could implythat postslippage multinucleated cells may have enhanced pro-clivity to enter senescence. Indeed, a previous report demonstrat-ed that tetraploid cells with irregular-shaped nuclei progressivelydeveloped senescence following Aurora kinase B inhibition (16).

To determine whether mitotic slippage–induced senescencecould be observed in vivo, we treated HCT116 xenograft micewith either nocodazole or paclitaxel. Hematoxylin and eosinstaining of tissue sections revealed a discernible increase inmultinucleated cells in the nocodazole- or paclitaxel-treatedxenografts compared with vehicle-treated control (Supplementa-ry Fig. S3A). In addition, themajority of cells that stained positivefor SA-b-gal were observed to be multinucleated (SupplementaryFig. S3A), suggesting possible correlation between multinuclea-tion and senescence. Nocodazole- or paclitaxel-treated xenograftsalso showed a significant increase in both p21mRNA and proteinlevels compared with vehicle-treated control (SupplementaryFig. S3B and S3C). In addition, tissue biopsies from invasiveductal breast carcinoma from patients treated with paclitaxelshowed increased populations of cells in the humoral region thatstained positive for p21 expression as compared with biopsiesobtain from patients before treatment (Supplementary Fig. S3D).

Postslippage secretory factors facilitate paracrine tumorigenicphenotypes

SASP factors secreted by cells which have undergone senescencedue to replicative exhaustion, oncogene activation, or irradiationcan modulate the tissue microenvironment to stimulate tumorprogression (8). As a first step to ascertain potential tumorigenicrole for SASP following mitotic slippage, we performed geneexpression microarray analysis on U2OS cells treated with eitherDMSO (control) or nocodazole for 48 hours. Microarray dataconfirmed SASP factor expression, including cytokines and che-mokines IL1a, IL1b, CXCL8, and CCL3 among the upregulatedgenes (Supplementary Fig. S4A). This was further confirmed at theprotein level with stable isotope labeling with amino acids in cellculture (SILAC) followed by high-resolution mass spectrometry(MS; Supplementary Fig. S4B; ref. 11). Notably, mass spectro-metric analysis revealed upregulation of SASP factors includingIL1b and IL8, and senescence-associated histone H3.3 variant(ref. 17; Supplementary Fig. S4C; Supplementary Table S1).In addition, both quantitative real-time PCR (qRT-PCR) andLuminex assays confirmed upregulated expression and secretion

(Continued.) One-hundred cells per condition and data are mean � SD of two independent experiments. Scale bar, 10 mm. B, Cell lysates collected as per Awere subjected to immunoblotting (IB) and probed for senescence-associated markers. The signal intensity ratios are shown at the bottom of the relevant lanes.C, Left, qRT-PCR analysis shows relative mRNA expression of SASP components in DMSO-treated or nocodazole-treated U2OS cells for 72 hours. mRNAlevels were normalized to GAPDH and DMSO control. Right, Plot shows relative fold change of cytokine secretion from U2OS cells treated with DMSO or nocodazolefor 72 hours using Luminex analysis. Data are mean � SD of three independent experiments. D, Scheme depicts the preparation of conditioned media (CM)for determining paracrine tumorigenic effects in Fig. 1E–H. E, U2OS and RPE-1 cells carrying chromosomal marker H2B-GFP were incubated with their respectiveCM for 2 days and subjected to the scratch wound healing assay. Left, Representative images of CM-treated U2OS H2B-GFP cells taken at 0 hours (immediatelyafter scratching) and at the indicated time intervals. Right, Plot shows rate of wound closure in U2OS and RPE-1 cells carrying H2B-GFP. Data are mean � SDof three independent experiments. Scale bar, 200 mm. F, U2OS H2B-GFP cells were treated as per E. Invasive ability was assessed by transwell Matrigel assay.Scale bar, 200 mm. Plot shows percentage of invasive cells. n¼ 5 random fieldswere taken per condition and data aremean� SD of three independent experiments.G, Cells collected from F were subjected to qRT-PCR. Plot shows the relative mRNA expression of migration-associated markers following incubation withindicated CM. mRNA levels were normalized to GAPDH and DMSO control. Data are mean � SD of three independent experiments. H, Representative imagesof choroidal sprouts formed after culturing with indicated CM for 2 days. Plot shows choroid sprouting ratio (area of vascular sprouting/area of the explant).Minimum of 6 explants analyzed per treatment. Data are mean � SD of three independent experiments. � , P < 0.05; �� , P < 0.005; �� , P < 0.001; �� , P < 0.0001;ns, not significant by Student t test.






of various SASP factors in nocodazole-treated U2OS cells at72 hours postslippage compared with DMSO-treated controlcells (Fig. 1C).

The observed increase in SASP factor secretion prompted us totest whether SASP factors could mediate tumorigenic phenotypessuch as migration, invasion, and angiogenesis. To this end, weprepared conditioned media (CM) from RPE-1 and U2OS cellstreated with nocodazole for 48 hours for use in various pheno-typic assays (experimental scheme in Fig. 1D). To assessmigratorycapability, RPE-1 and U2OS cells expressing chromatin markerH2B-GFP exposed to CM from their respective postslippageparental cells were subjected to the scratch wound healing assay.As shown in Fig. 1E, postslippage CM (nocodazole CM) promot-ed wound closure at a rate faster than control CM (DMSO CM),indicating increased induction of migration. Notably, cell prolif-eration assays by BrdU labeling and cell count experimentsdemonstrated that factors secreted did not alter rate of cellproliferation in CM-exposed cells (Supplementary Fig. S5A andS5B), indicating increased wound closure was not influenced byproliferation. To evaluate invasive capability, U2OS H2B-GFPcells incubated with either control or postslippage CM were usedin transwell invasion assays. An increased number of cells exposedto postslippage CM invaded the bottom of the filter chambercompared with control CM (Fig. 1F). An upregulation of invasionand migration-related markers fibronectin and MMP9 in U2OScells exposed to postslippage CM compared with control CMwasalso observed (Fig. 1G), confirming migratory and invasive capa-bilities. To test for angiogenic capability, we used the choroidangiogenesis assay, an ex vivo model of angiogenesis. Increasedinduction of vascular sprouting from choroid explants incubatedwith postslippage CM compared with control CM was observedfour days after incubation (Fig. 1H), demonstrating angiogeniccapability conferred by factors from postslippage cells. Takentogether, we conclude that postslippage SASP proteins conferparacrine protumorigenic potential.

Autophagy activity increases following mitotic slippageWe next wished to investigate the molecular determinants that

induce senescence in postslippage cells. Mass spectrophotometricanalysis of U2OS cells treated with nocodazole for 48 hours,revealed four autophagy-related proteins, namely MAP1A,MAP1B, MAP1LC3B (also known as LC3B), and GABARAP to bedownregulated postslippage comparedwith control (Supplemen-tary Fig. S4C; Supplementary Table S1). Because all four of theseproteins serve as autophagic substrates, their downregulationsuggests increased autophagic activity. Autophagy is a catabolicprocess in which cytoplasmic constituents are targeted for remov-al or recycling in autophagosomes that fuse with the lysosome(18). MAP1A and MAP1B are microtubule-associated proteins,which are precursor polypeptides that undergo proteolytic pro-cessing to generate heavy and light chain subunits such asMAP1LC3A and MAP1LC3B (19). MAP1LC3B and GABARAPbelong to the ATG8 orthologs of mammalian cells (19). Consis-tentwith upregulated autophagy postslippage, time-lapsemicros-copy revealed increased autophagic marker GFP-LC3 punctatefoci in U2OS cells postslippage (t ¼ 48 hours), indicating autop-hagosome accumulation (Fig. 2A).

To further evaluate autophagic flux, we assessed conversionof the nonlipidated formLC3B-I to the lipidated autophagosome-associated formLC3B-II (hereafter LC3-I andLC3-II, respectively),as well as degradation of the ubiquitin-binding autophagic

adaptor protein p62/SQSTM1. To ensure that autophagy flux wasassessed only in adherent postslippage cells destined to entersenescence, and that quantitation was not affected by cell deathpostslippage, cells were washed 36 hours post nocodazoletreatment to remove apoptotic cells. As shown by phosphorylatedHistone H3 levels in Fig. 2B and C, HCT116 and U2OS cells wereinmitosis (M) at t¼ 16 hours and postslippage (PS) from t¼ 24–36 hours. Increased autophagy flux postslippage was observed asshown by an increase in LC3-II isoforms at t ¼ 24–36 hourscompared with t ¼ 16 hours (Fig. 2B and C). To confirm this,U2OS cells were treated with chemical inhibitor of autophagyBafilomycin A1 (Baf A1), which blocked lysosome acidificationand prevented autophagosome clearance. Increased LC3-II wasobserved at t ¼ 36 and 48 hours (Fig. 2C) compared withnocodazole control, suggesting LC3-II accumulation postslippageto be due to autophagosome accumulation and not impairmentof downstream autophagic processes such as autophagosome–lysosome fusion or lysosomal degradation. Concomitantly, adecrease in p62, which can be blocked by Baf A1, further con-firmed active autophagy (Fig. 2B and C). LC3-II accumulation inpostslippage cells was also blocked by knockdown of autophagyby stable expression of short hairpin RNA targeting ATG5(shATG5) required for autophagy elongation (ref. 20; Fig. 2D).

Autophagic induction occurs via a number of pathways thatfinally converge on regulation of the AMPK/ULK/mTOR axis,which integrates growth factor and nutrient signals to regulatecellular metabolism and maintain energy homeostasis (21).Phosphorylation of AMP-activated protein kinase (AMPK) onThr172 activates autophagy by directly activating Unc-51 LikeAutophagy Activating Kinase 1 (ULK1) through phosphorylationof Ser317 and Ser777 (21, 22). In contrast, high activity of themTOR negatively regulates autophagy by preventing ULK1 acti-vation via ULK1 Ser757 phosphorylation and disruption of ULK1and AMPK interaction (21). To test whether cells postslippageinvoked the AMPK/ULK/mTOR axis for autophagy induction,nocodazole-treated U2OS cells were subjected to immunoblot-ting. We observed that in mitotic cells 16 hours post-nocodazoletreatment, both mTOR and AMPK were activated as shown byincreased phosphorylation of the mTOR substrate p70 S6K,conversion of total ULK1 to phosphorylated Ser757 ULK1 as wellas increased phosphorylated Thr172 AMPK (Fig. 2E). This corre-lated with enhanced autophagy, and was consistent with anobserved increase in LC3-II (Fig. 2E). As it was previously reportedthat autophagy can be activated in a mTOR-independent manner(23, 24), our findings suggest presence of autophagic activitydespite mTOR activation during mitosis. On the other hand,diminished levels of phosphorylated-p70 S6k indicatingdecreased mTOR activity with concomitant decrease in ULK1phosphorylated on Ser757 residue was observed in postslippagecells (Fig. 2E). These findings together with the observed increasefor phosphorylated Thr172-AMPK and LC3-II levels, stronglysuggested that the AMPK/ULK/mTOR axis promotes autophagyinduction postslippage.

AMPK activation is induced upon ER proteotoxic stresspostslippage

We next sought to understand the molecular mechanism bywhich mitotic slippage could lead to AMPK activation andautophagy induction. A recent report showed that aneuploidycaused substantial endoplasmic reticulum (ER) proteotoxic stressand unfolded protein response (UPR) activation in cells that

Jakhar et al.





underwent cell death postslippage (25). We therefore examinedthe expression of UPR-related protein kinase RNA-like ER kinase(PERK) and inositol-requiring protein 1 (IRE1) in U2OS cellsafter nocodazole and paclitaxel treatment. Upon accumulationof misfolded or unfolded proteins in the ER, PERK is activatedthrough dimerisation and trans-autophosphorylation on multi-ple residues including Thr980 (26). Activated PERK phosphor-

ylates eukaryotic translation initiator factor 2a (eIF2a) on Ser51,which then attenuates global protein synthesis. PhosphorylatedeIF2a leads to an increase in transcription factor CCAAT/enhancerbinding protein (C/EBP) homologous protein (CHOP) at bothtranscriptional and translational level. Our data revealed thatnocodazole and paclitaxel treatment increased the phosphoryla-tion of PERK at Thr980 and eIF2a at Ser51 postslippage compared

Figure 2.

Autophagy is induced via AMPK/ULK1/mTOR axis during mitotic slippage. A, U2OS cells stably expressing GFP-LC3 were assessed for GFP puncta formationfollowing 48-hour nocodazole (Noc) treatment. Plot shows percentage of GFP-LC3 puncta-positive cells (>5 puncta per cell) following indicated treatments.Fifty cells per condition. Scale bar, 20mm.B,HCT116 cellswere treatedwith nocodazole for indicatedduration. Lysates frommitotic cells (M) or postslippage cells (PS)were subjected to IB and probed for autophagy- and mitotic slippage–related markers. The respective signal intensity ratios are shown at the bottom of therelevant lanes. C, U2OS cells were treated with nocodazole for the indicated duration with or without autophagy inhibitor Bafilomycin A1 (Baf A1). Lysates frommitotic cells (M) or postslippage cells (PS) were subjected to IB analysis and probed for autophagy-related markers LC3 and p62. D, U2OS cells stably expressingshCtrl or shATG5 were treated with nocodazole for the indicated time and subjected to IB analysis. The LC3-II/GAPDH and p62/GAPDH signal intensityratios are shown at the bottom of the relevant lanes. E, U2OS cells were treated with nocodazole for the indicated duration, subjected to IB and probedwith AMPK/mTOR/ULK1 axis–related antibodies. The respective signal intensity ratios are shown at the bottom of the relevant lanes. All data are mean � SDof three independent experiments (�� , P < 0.001 by Student t test).






Figure 3.

ER stress mediates postslippage AMPK activation. A, U2OS cells were treated with nocodazole (Noc) or paclitaxel (PTX) for the indicated duration.Lysates from mitotic cells (M) or postslippage cells (PS) were subjected to IB and probed for ER stress–related markers. The respective signal intensityratios are shown at the bottom of the relevant lanes. B and C, U2OS cells were treated with nocodazole or paclitaxel for the indicated time points. Cells weretreated with 1 mg/mL tunicamycin (TM) for 24 hours as a positive control. Relative CHOP expression and splicing of XBP1 mRNA was detected by RT-PCRand quantified by qRT-PCR. D, U2OS cells were treated with nocodazole in absence or presence of 10 mmol/L PERK inhibitor (PERKi; GSK2656157), subjectedto IB and probed for autophagy-related markers. E, U2OS cells were treated with nocodazole in absence or presence of 10 mmol/L IRE1 inhibitor (IRE1i; 4m8C),were collected at the indicated time points and analyzed by IB with the indicated antibodies. All data are mean� SD of three independent experiments (� , P < 0.05;�� , P < 0.005; �� , P < 0.001 by Student t test).

Jakhar et al.





with control (t¼ 0 hours, start of treatment; Fig. 3A). An increasein CHOP at both protein and mRNA levels was also observedpostslippage compared with control (Fig. 3A and B). An alterna-tive signaling branch of the UPR is the IRE1–XBP1 pathway (27).Phosphorylation of IRE1 leads to cleavage of XBP1 mRNA andgenerates a translational frameshift which acts as a potent tran-scriptional activator (26). Increased IRE1 phosphorylation atSer724 and XBP1 mRNA splicing was observed postslippagecomparedwithDMSO-treated control (Fig. 3C). Thus, in additionto cell death postslippage, our results indicate induction of ERstress and UPR induction with activation of PERK and IRE1 UPRpathways in postslippage senescent cells as well.

To investigate if either of these pathways regulate autophagy,PERK or IRE1 activity was inhibited with PERK inhibitorGSK2656157 or IRE1 inhibitor 4m8C. GSK2656157 reduces bothtotal and phosphorylated PERK levels, while 4m8C blocks sub-strate access to the active site of IRE1 and selectively inactivatesboth XBP1 splicing and IRE1-mediatedmRNAdegradation. Inter-estingly, while inhibition of IRE1 did not prevent the accumula-tion of LC3-II, PERK inhibition decreased LC3-II levels inpostslippage cells to that of cycling cells (Fig. 3D and E), revealingthat autophagy induction postslippage occurred through PERKactivation. As described previously (Fig. 2), autophagy was affect-ed via AMPK activation, which is downstream of PERK (28). Wefind that PERK inhibition led to decreased AMPK phosphoryla-tion (Fig. 3D), further confirming that the increase in LC3-II waspredominantly induced by the PERK–AMPK axis of the ER stressresponse in postslippage cells.

Autophagy inhibition in postslippage cells leads to senescencebypass and cell death

Because autophagy is active postslippage, we sought to inves-tigate whether the senescence cell fate could be influencedby autophagic inhibition. Inhibition of autophagy by Baf A1 ortransfection with shATG5 resulted in senescence bypass asobserved by decreased SA-b-gal staining (Fig. 4A; SupplementaryFig. S6A).Concomitantly, a substantial increase in cells entering S-phase was suggested by increased BrdU labeling upon inhibitionof autophagy (Fig. 4B). Transfection of siATG5 clearly suppressedthe increase in LC3-II indicating autophagy was indeed blocked(Fig. 4C). This unequivocally confirmed that postslippage senes-cence was dependent on autophagy. Time-lapse microscopyrevealed that autophagy inhibition in U2OS cells treated withnocodazole or paclitaxel increased cell death postslippage(Fig. 4D) and reduced cell viability (Fig. 4E). Consistent withthis, clonogenic assays showed reduction of long-term cell sur-vival following autophagic inhibition (Fig. 4F). Autophagy hasbeen described to mediate degradation of lamin B1 (29). Inter-estingly, nocodazole-treated RPE-1 cells pretreated with Baf A1prevented lamin B1 degradation at day 3 and day 6 (Supplemen-tary Fig. S6B and S6C). Notably, these differences were not due tochanges at the transcriptional levels as our RT-PCR results exclud-ed this possibility (Supplementary Fig. S6D).

To determine whether these findings could be extrapolated toother antimitotic drugs, we assayed cells treated with nocodazole,paclitaxel, Mon and ZM for the apoptotic marker cleaved PARPfollowing ATG5 knockdown. Interestingly, the increase in celldeath was restricted to MT-targeting drugs nocodazole and pac-litaxel (Fig. 4G). In addition, autophagy inhibition increasedDNA damage (as determined by gH2AX foci and 53BP1 proteinlevels) compared with control (Fig. 4G–I). Enhanced replication

stress shown by increased RPA32 phosphorylation at 72 hourspost-nocodazole treatment compared with control was alsoobserved (Fig. 4I). Taken together, our findings indicate thatautophagy modulates cell fate upon antimitotic drug treatmentand that inhibition of autophagy engenders cells to bypasssenescence postslippage and enter S phase. These cells thenundergo increased replication stress and DNA damage, culminat-ing in cell death.

A recent study reported that autophagy plays a role in promot-ing MCD during prolonged mitotic arrest (30), suggesting thatentry intomitotic slippage per se could potentially be inhibited byautophagy. We therefore tracked individual cell fate (MCD vsmitotic slippage) in U2OS cells stably transfected with shATG5treated with either nocodazole or paclitaxel for 48 hours usingtime-lapse microscopy. Intriguingly, there was no significantdifference in the proportion of cells undergoing MCD or slippageupon autophagic inhibition compared with control (Supplemen-tary Fig. S7A). Consistently, the duration from mitosis to mitoticslippage (Supplementary Fig. S7B), cyclin B1 degradation, anddephosphorylation of BubR1 and Histone H3 were not affectedby autophagy inhibition postslippage (24–36 hours post-nocodazole treatment; Supplementary Fig. S7C and S7D). Ourresults thus indicate that autophagy modulates cell fate specifi-cally after escape from mitotic arrest through mitotic slippage.

Autophagy inhibition attenuates protumorigenic effectsof SASP

We observed a discernible decrease in expression of SASPcytokines CXCL3, IL1b, IL6, IL8, CCL7, and PDGF-AB/BB andincreased expression of BMP2 postslippage following autophagyinhibition compared with control (Fig. 5A; Supplementary Fig.S8A and S8B). This suggests that autophagy enables senescenceand consequently modulates SASP following mitotic slippage.Notably, IL1b regulatory control via the autophagy axis wasobserved to be more prominent posttranscriptionally as shownin Supplementary Fig. S8B (compare with mRNA level in Sup-plementary Fig. S8A). In addition, SASP-induced tumorigenicfunction was attenuated following autophagic inhibition (Fig. 5).Postslippage CM promoted increased vascular sprouting com-pared with control CM from cycling cells in choroid explants,whereas autophagic inhibition significantly inhibited sproutingactivity (Fig. 5B). In addition, transwell assays using cells incu-bated with CM from postslippage U2OS cells expressing shATG5showed a reduction of these cells capable of invasion comparedwith control CM (Fig. 5C). To assess whether cytokines wererequired for SASP-mediated paracrine signaling functions in vitro,we used CM from IL1b and IL8-deficient postslippage cells (Sup-plementary Fig. S8C) for gene expression analysis of invasionand migration-related markers. Depletion of either IL1b or IL8or both cytokines from the CM resulted in a significant decreasein invasiveness compared with control CM (SupplementaryFig. S8D). In addition, decreased fibronectin expression wasobserved in all conditions (Supplementary Fig. S8E), whereas adecrease in MMP-9 was only observed upon double knockdownof IL1b and IL8 (Supplementary Fig. S8E). Supplementation ofIL1b or IL8 to postslippage CM collected under autophagy inhi-bition conditions proved sufficient to rescue the reduction ininvasiveness (Supplementary Fig. S8F). Taken together, thissuggests that autophagy-enabled production of IL1b and IL8 iscrucial for postslippage senescent cells to promote paracrinetumorigenic progression.






Jakhar et al.





To examine whether the in vitro migratory and invasivereduction following autophagic inhibition could be extrapo-lated to metastatic inhibition in vivo, we used the zebrafishmodel of malignancy to monitor migration. The transparencyof the zebrafish embryo provides the unique ability to visualizein vivo migratory changes of tumor cells in real time. U2OSH2B-GFP–expressing cells were incubated with CM from eithercycling or postslippage cells expressing shATG5 or control fortwo days before injection into zebrafish embryos. H2B-GFP–expressing cells incubated with postslippage CM were observedto migrate out from the injection site and metastasize furtherthan those with control CM (Fig. 5D and E). This metastaticpotential was significantly reduced following autophagic inhi-bition via shATG5 (Fig. 5D and E). These observations furthersupport the role of autophagy-dependent SASP secretion inparacrine migration and invasion in vivo.

Combination treatment of antimitotic drugs and autophagyinhibitor arrests tumor growth and is dependent on p53 status

As p53 is awell-known regulator of cellular senescence (31), wehypothesized that p53 status could be an important predictor ofresponse to combinatorial treatment of nocodazole or paclitaxelwith autophagy inhibition. Increased DNA damage (as measuredby gH2AX protein levels) and a corresponding increase in celldeath (by cell viability assays and cleaved PARP) were detected inU2OS cells postslippage with depleted p53 (shp53) comparedwith control (Supplementary Fig. S9A and S9B). In addition, weobserved that HCT116 cells with wild-type (WT) p53 showedincreased cell death upon combination treatment compared withp53-deficient cells (Fig. 6A). Cancer cells lines with dominant-negative p53mutations, namelyMDA-MB-231 breast cancer cellswith p53 R280K, PanC1 pancreatic cancer cells with p53 R273Hand PaCa2 pancreatic cancer cells with p53 R248H also showedless sensitivity to combination treatment compared with MCF7breast carcinoma cells with WT p53 (Supplementary Fig. S10A).This suggests that cells with intact p53 might respond morefavorably to combinatorial treatment.

To testwhether thesefindings could be extrapolated to an in vivomodel,we treatedmice xenograftedwithHCT116 cells (p53þ/þorp53�/�) with combination treatment of either nocodazole orpaclitaxel and autophagic inhibitor chloroquine (CQ). Combi-nation treatment (nocodazole þ CQ or nocodazole þ CQ)compared with control nocodazole or paclitaxel alone signifi-cantly decreased tumor growth and final tumor weight (Fig. 6B–D), particularly in mice xenografts with WT p53, suggesting thattumors with intact p53 were most sensitive to combination

treatment. Molecular assessment of tumor sections derived frommice xenografts showed more robust LC3 and p62 puncta accu-mulation in the combination drug-treated mice compared withnocodazole or paclitaxel treatment alone (SupplementaryFig. S10B and S10C), indicating efficient autophagic inhibition.

In conclusion, we propose the model outlined in Fig. 6E. Inresponse to the prolonged mitotic arrest induced by antimitoticdrugs, cells undergo either mitotic cell death or mitotic slippage.In cells postslippage, autophagy is induced through ER stress- andUPR-mediated regulation of the AMPK/mTOR/ULK1 axis. Thiscontributes to senescence and SASP production, which confersprotumorigenic potential in a paracrine manner. Conversely,following autophagy inhibition, senescence is bypassed andpostslippage cells undergo death in a p53-dependent manner.

DiscussionThe success of antimitotic therapies, often used as first-line

treatment of several malignancies (1), is limited due to acquiredresistance. The fate of cells following mitotic slippage, albeitrepresenting a route of escape from mitotic arrest and mitoticcell death, has not been extensively studied. Here, we show thatmultinucleated tetraploid cells that accumulate postslippage canundergo senescence and drive paracrine tumorigenic effects bothin vitro and in vivo in an autophagy-dependent manner.

Although prior seminal work on the autophagy-senescenceconnection by Young and colleagues (32) showed that autopha-gic inhibition delayed the senescence phenotype following onco-gene activation, our work demonstrates that inhibition of autop-hagypostslippage results in senescence bypass and accelerated celldeath. The most compelling evidence for senescence bypass afterautophagy inhibitionwas increased entry into S-phase, the induc-tion of DNA damage and replication stress, and reduction in cellviability in postslippage cells collaterally.

Autophagy has also been implicated as a potential inhibitoryregulator of senescence (33). How does one reconcile the fact thatautophagy can both activate and inhibit cellular senescence? Kangand colleagues (34) provide an explanation for these conflictingresults by describing the differential regulation of senescence viaselective versus general autophagy. This is achieved by transcrip-tion factor GATA4, which the authors established to be a senes-cence regulator (34). GATA4 bound to autophagy adaptor p62 isdegraded by selective autophagy under nonsenescent conditions.This selective autophagy is suppressed upon senescence inductionwhere GATA4 is stabilized due to decreased GATA4–p62 inter-action. GATA4 stability triggers a series of events which ultimately

Figure 4.Autophagy mediates postslippage cell fate. A, Representative images of U2OS cells stably expressing shCtrl or shATG5 stained with SA-b-gal after 72-hournocodazole (Noc) treatment. Plot shows percentage of SA-b-gal–positive cells. One-hundred cells per condition and data are mean � SD of three independentexperiments. Scale bar, 10 mm.B,U2OS cells were transfectedwith siCtrl or siATG5 followed by nocodazole treatment for 72 hours. Proliferationwas assessed by theBrdU incorporation assay. Plot shows percentage of BrdU-positive multinucleated cells. One-hundred cells per condition and data are mean � SD of threeindependent experiments. Scale bar, 20 mm. C, IB shows efficiency of ATG5 knockdown in U2OS cells transfected with siCtrl or siATG5 followed by nocodazoletreatment for 72 hours. D, U2OS shCtrl or shATG5 cells carrying H2B-GFP were treated with nocodazole for 48 hours. Postslippage multinucleated cells weretracked individually by time-lapse microscopy for another 2 days. Red, cells that die postslippage. Black, cells that arrest postslippage. Fifty cells per condition anddata aremean� SD of two independent experiments. E,U2OS or RPE-1 cells were treatedwith nocodazole or paclitaxel in absence or presence of Baf A1 for 72 hours.Cell viability was assessed by CellTiter 96 One Solution Proliferation assay. F, Experimental scheme of clonogenic cell survival assay. Left, representativeimages of U2OS cells treated with the indicated drugs followed by crystal violet staining. Right, plot shows number of colonies counted using GelCount software.Eight wells per condition of 2 independent experiments. G, U2OS shCtrl or shATG5 cells were treated with indicated drugs for 72 hours and subjected to IB analysis.H, U2OS shCtrl or shATG5 cells were treated with nocodazole for 72 hours and subjected to IF. Plot shows percentage of gH2AX-positive multinucleated cells.One-hundred cells per condition and data are mean � SD of two independent experiments. Scale bar, 20 mm. I, U2OS shCtrl or shATG5 cells were treated withnocodazole for 72 hours and subjected to IB analysis (� , P < 0.05; �� , P < 0.005; �� , P < 0.001; �� , P < 0.0001; ns, not significant by Student t test).






Figure 5.

Autophagy confers paracrine tumorigenic potential by regulating SASP. A, Luminex cytokine assay of supernatant obtained from U2OS cells treated withnocodazole in absence or presence of Baf A1 for 72 hours. B, Representative images of choroidal sprouts formed using indicated CM generated from U2OSshCtrl or shATG5 cells. Plot shows sprouting ratio (area of vascular sprouting/area of the explant). Minimum 6 explants analyzed per treatment. C, U2OS H2B-GFPcells were cultured with indicated CM from U2OS shCtrl or shATG5 for 48 hours. Invasive ability was assessed by transwell assay. Scale bar, 200 mm.Plot shows percentage of invasive cells. Five random fields were taken per condition and data are mean� SD of three independent experiments. D, U2OS H2B-GFPcells were cultured with indicated CM generated from U2OS shCtrl or shATG5 cells, followed by injection into zebrafish embryo. All tumor foci observed in alltested larvae, belonging to each treatment group (presented as n on the graph), is presented as a scatter plot. Foci in one larva is represented by one color. "n"denotes number of injected embryos from three biological replicates. E, Quantitative comparison of metastatic ability upon indicated treatments wasmeasured by invasive index (top) and migration index (bottom). � , P < 0.05; ��, P < 0.005; ��, P < 0.001; �� , P < 0.0001 and ns, not significant by Student t test.


Jakhar et al.




Figure 6.

Synergistic inhibition of tumor growth by antimitotic drugs and autophagy inhibitors depends on p53 status. A, HCT116 wt or p53�/� cells were treatedwith the indicated drugs for 72 hours. Cell viability was assessed by CellTiter 96 One Solution Proliferation assay. Data are mean � SD of three independentexperiments. B, Representative images of xenograft tumors harvested from nude mice after two weeks of indicated treatment. Drug dosage used: autophagyinhibitor, Chloroquine (CQ): 30 mg/kg, nocodazole: 10 mg/kg, and paclitaxel (PTX): 10 mg/kg. C, Effect of combinatorial treatment on the growth ofxenograft tumors generated by subcutaneous injection of HCT116 wt or p53�/� cells in nude mice. Data are mean of � SD (n ¼ 10 tumors per group). D, Theweight of tumor after two weeks of indicated treatment. Dots represent individual tumor weight, n¼ 10 tumors per group (� , P < 0.05; �� , P < 0.005; �� , P < 0.001,ns, not significant by Student t test). E, Proposed model of autophagy-mediated tumorigenesis upon antimitotic treatment. Upon treatment with antimitotic drugs,autophagy is induced via AMPK/mTOR/ULK1 axis activated by ER stress. Autophagy contributes to senescence and SASP production. This confersprotumorigenic potential in a paracrine manner. Conversely, following autophagy inhibition senescence is bypassed and postslippage cells undergo death.






leads to activation of NFkB for SASP production, thus facilitatingsenescence (34). In addition, autophagy has been described tomediate degradation of lamin B1 (29). The loss of lamin B1, asdescribed before (12) and also shown in our study, is a senes-cence-associated marker. Inhibition of autophagy or LC3-laminB1 interaction prevented oncogene-induced lamin B1 loss andattenuated senescence (29), and we observed similar block inlaminB1degradationupon autophagy inhibition. Itwill beworthexploring whether LC3–lamin B1 interaction and GATA4 con-tribute to autophagy-induced senescence in our context.

Our findings demonstrate that autophagy is induced duringmitotic slippage through the PERK–AMPK axis of the ER stressresponse. Inhibition of ER stress decreased the rate of LC3-IIconversion by preventing AMPK phosphorylation. Althoughsome studies indicate that ER stress prevents cellular senescenceby upregulating autophagy (35), there is compelling evidence tosupport the opposite. Using a mouse model of B-cell lymphomatreated with the DNA-damaging agent cyclophosphamide, theSchmitt and colleagues' group (36) recently reported that senes-cent cells rely on autophagy to cope with SASP-coupled proteo-toxic stress. This proteotoxic stress can lead to transcriptionalactivation of UPR genes and autophagy induction (25), renderingthe senescent cells sensitive to autophagic inhibition. In addition,autophagy might further be maintained as part of the innateimmune response against cytoplasmic chromatin fragments asso-ciatedwith senescence through the cGAS–STINGpathway (37), ascGASdirectly interactswithBeclin-1 (38) to clear cytosolicDNA inan autophagy-dependent manner. In support of this, the cGASand/or STING pathway has been reported by multiple groups tobe involved in induction of senescence and/or SASP (37, 39).

Postslippage SASP factors conferred paracrine protumorigenicphenotypic effects such as migration, invasion, and vasculariza-tion on neighboring cells. Proliferation of cells incubated withCM from postslippage cells was unaffected. As it has been shownthat SASP composition and quantity is dependent on cell typeand mode of senescence induction (8), this suggests that post-slippage SASP may serve to play a "secondary" protumorigenicrole in cells where migration, invasion, and angiogenesis areuncoupled from cell proliferation thereby leading to the devel-opment of a more malignant phenotype in cancer cells (40). It ispossible that once neighboring cells become transformed andstimulated to proliferate, postslippage SASP then plays a roleto further enhance the tumorigenic capabilities of these cells,demonstrating that mitotic slippage–induced senescence couldserve as a conduit for malignant transformation and antimitotictherapy resistance.

In contrast, multiple studies have reported the role of autop-hagy in promoting the elimination of tumor cells through mod-ulation of the inflammatory response (41). The autophagy-dependent inflammatory response is bidirectional and context-dependent. To enhance efficacy of antimitotic therapies, onemight not only aim to eliminate protumorigenic SASP, but mayalso provoke desirable immunogenic cell death and clearanceby the induction of antitumorigenic SASP. Therefore, it will beimportant that future studies test the impact of mitotic slippage–induced SASP expression on the tumor microenvironment andhost immune system in a spontaneous tumor model.

In addition, targeting of senescent cells seems not only relevantfor tumor cells and their microenvironment, but also for non-cancerous cells. The Campisi and colleagues' group (9) recentlyshowed that cytotoxic drugs induce senescence in normal cells

thereby contributing to the systemic side effects of chemotherapy.Hence, several approaches targeting or bypassing senescenceduring chemotherapy to reduce the likelihood of acquired resis-tance and chemotoxicity by redirecting cells toward apoptosishave been attempted (42). In addition, the conviction that senes-cence is a truly irreversible process has been challenged as stem-like aggressive tumor-initiating cells have emerged from senescentcells (43, 44). Interestingly, senescence revertants (without stem-like properties) have recently been shown to contain a subset ofsenescence-activated genes that can contribute to aggressivenessof the revertants (45).

Our results establish a key mechanism underlying cell fateafter mitotic slippage and suggest a novel strategy combiningcytotoxic drugs with autophagy inhibition to tackle undesiredeffects of mitotic slippage–induced senescence. It is conceivablethat senescent cells rely on autophagy as a nutrient sourceto support cellular metabolism and survival, as well as forautophagy-dependent clearance of targeted substrates. Thisdependence could render them selectively vulnerable to autop-hagic inhibition. Our in vivo zebrafish and mice xenografttumor models showed tumor growth arrest upon synergisticautophagy inhibition and antimitotic drug treatment. Thesuccess of drug combination treatment was dependent on thestatus of p53, where tumors with wild-type p53 show greatersensitivity to combinatorial treatment (46). This lends supportto the idea that p53 status is an important factor influencingtherapeutic success and could be used as a potential biomarkerfor patient stratification, with exclusion of patients carryingp53-mutant tumours. Currently, the autophagy inhibitorsCQ and hydroxychloroquine (HCQ), clinically approvedfor treatment of malaria, and additional newly discovered,more potent autophagy-modulating compounds are showingpromising results in clinical trials for the safe use to overcomechemotherapeutic resistance (47, 48).

Our studies also point to regimens targeting senescence andSASP as potential combinatorial strategies with antimitotic drugs.Although our current work focuses on MT-targeting drugs, weanticipate other classes of chemotherapeutic drugs to have similarroles in engendering non-cell–autonomous tumor progression.Hence, it will be of interest to identify and delineate signalingmolecules and pathways leading to SASP and protumorigenicfunctionpostslippage. Future studieswill focus onnodal points inthe "slippage-autophagy-senescence-SASP-tumor progression"axis as these could potentially serve as beneficial therapeutictargets in combination with antimitotic drug therapy. Interest-ingly, induction of cellular senescence is often correlated withpolyploidy (49). It has been posited that polyploid tumors withelevated genomic content tend to become resistant to chemo-therapy due to rapid tumor evolution (49). In addition, near-tetraploid cancer cells have recently been shown to exhibitenhanced invasiveness (50). It will be of interest to determinewhether high ploidy status per se is able to contribute to SASP-induced tumor progression.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: B. Cheng, K. Guo, X. Wang, J.E. Connolly, K.C. CrastaDevelopment of methodology: A.X.F. Wong, B. Cheng, K. Guo, K.J. Lim,J.B.K. Khoo, B.T. Chua, I. Sinha, J.E. Connolly, K.C. Crasta

Jakhar et al.





Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): R. Jakhar, M.N.H. Luijten, A.X.F. Wong,B. Cheng, K. Guo, S.P. Neo, B. Au, M. Kulkarni, K.J. Lim, J. Maimaiti,H.C. Chong, E.H. Lim, T.B.K. Tan, K.W. Ong, Y. Sim, J.B.K. Khoo, I. Sinha,J.E. Connolly, J. Gunaratne, K.C. CrastaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): R. Jakhar, M.N.H. Luijten, A.X.F. Wong, B. Cheng,K. Guo, M. Kulkarni, K.J. Lim, H.C. Chong, B.T. Chua, I. Sinha, X. Wang,J.E. Connolly, J. Gunaratne, K.C. CrastaWriting, review, and/or revision of the manuscript: R. Jakhar, M.N.H. Luijten,A.X.F.Wong, B. Cheng, K.J. Lim, I. Sinha, J.E. Connolly, J. Gunaratne, K.C.CrastaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): A.X.F. Wong, B. Cheng, K. Guo, S.P. Neo,J. Maimaiti, J.T.S. Ho, J.E. Connolly, J. Gunaratne, K.C. CrastaStudy supervision: J.E. Connolly, K.C. CrastaOther (contributed to mass spectrometry analysis): S.P. Neo, J. GunaratneOther (performed biopsies to obtain material): J.S.L. Wong

AcknowledgmentsThis research is supported by the National Research Foundation, Prime

Minister's Office, Singapore, under its NRF Fellowship Programme (NRF Awardno. NRF-NRFF2013-10), Nanyang Assistant Professorship Grant, NanyangTechnological University and the Singapore Ministry of Education AcademicResearch Fund Tier 1 (Grant no.: 2015-T1-002-046-01; provided to K. Crasta).We are grateful to Sixun Chen and other members of the Crasta Lab forvaluable comments and help with manuscript preparation. We thank AMPL forhistology services, andG.L. Siok andX.L.R.Hai for technical support inMSanalysis.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received January 8, 2018; revised June 3, 2018; accepted July 10, 2018;published first July 23, 2018.

References1. Dumontet C, Jordan MA. Microtubule-binding agents: a dynamic field of

cancer therapeutics. Nat Rev Drug Discov 2010;9:790–803.2. Rieder CL, Maiato H. Stuck in division or passing through: what happens

when cells cannot satisfy the spindle assembly checkpoint. Dev Cell2004;7:637–51.

3. Brito DA, Rieder CL. Mitotic checkpoint slippage in humans occurs viacyclin B destruction in the presence of an active checkpoint. Curr Biol2006;16:1194–200.

4. Gascoigne KE, Taylor SS. Cancer cells display profound intra- and interlinevariation following prolonged exposure to antimitotic drugs. Cancer Cell2008;14:111–22.

5. Cheng B, Crasta K. Consequences of mitotic slippage for antimicrotubuledrug therapy. Endocr Relat Cancer 2017;24:T97–T106.

6. Ewald JA, Desotelle JA, Wilding G, Jarrard DF. Therapy-induced senescencein cancer. J Natl Cancer Inst 2010;102:1536–46.

7. Prieur A, Peeper DS. Cellular senescence in vivo: a barrier to tumorigenesis.Curr Opin Cell Biol 2008;20:150–5.

8. Coppe JP, Desprez PY, Krtolica A, Campisi J. The senescence-associatedsecretory phenotype: the dark side of tumor suppression. Annu Rev Pathol2010;5:99–118.

9. Demaria M, O'Leary MN, Chang J, Shao L, Liu S, Alimirah F, et al. Cellularsenescence promotes adverse effects of chemotherapy and cancer relapse.Cancer Discov 2017;7:165–76.

10. Crasta K, Ganem NJ, Dagher R, Lantermann AB, Ivanova EV, Pan Y, et al.DNA breaks and chromosome pulverization from errors inmitosis. Nature2012;482:53–8.

11. Swa HL, Shaik AA, Lim LH, Gunaratne J. Mass spectrometry based quan-titative proteomics and integrative network analysis accentuates modulat-ing roles of annexin-1 in mammary tumorigenesis. Proteomics 2015;15:408–18.

12. Freund A, Laberge RM, Demaria M, Campisi J. Lamin B1 loss is a senes-cence-associated biomarker. Mol Biol Cell 2012;23:2066–75.

13. Lin SP, Chiu FY, Wang Y, Yen ML, Kao SY, Hung SC. RB maintainsquiescence and prevents premature senescence through upregulation ofDNMT1 in mesenchymal stromal cells. Stem Cell Reports 2014;3:975–86.

14. Johmura Y, Nakanishi M. Multiple facets of p53 in senescence inductionand maintenance. Cancer Sci 2016;107:1550–55.

15. Georgakilas AG, Martin OA, Bonner WM. p21: a two-faced genomeguardian. Trends Mol Med 2017;23:310–19.

16. SadaieM,Dillon C,NaritaM, Young AR, Cairney CJ, Godwin LS, et al. Cell-based screen for altered nuclear phenotypes reveals senescence progressionin polyploid cells after Aurora kinase B inhibition. Mol Biol Cell 2015;26:2971–85.

17. Duarte LF, Young AR, Wang Z, WuHA, Panda T, Kou Y, et al. Histone H3.3and its proteolytically processed form drive a cellular senescenceprogramme. Nat Commun 2014;5:5210.

18. MizushimaN, Levine B, Cuervo AM, Klionsky DJ. Autophagy fights diseasethrough cellular self-digestion. Nature 2008;451:1069–75.

19. Halpain S, Dehmelt L. The MAP1 family of microtubule-associatedproteins. Genome Biol 2006;7:224.

20. Mizushima N, Sugita H, Yoshimori T, Ohsumi Y. A new protein conjuga-tion system in human. The counterpart of the yeast Apg12p conjugationsystem essential for autophagy. J Biol Chem 1998;273:33889–92.

21. Kim J, Kundu M, Viollet B, Guan KL. AMPK and mTOR regulateautophagy through direct phosphorylation of Ulk1. Nat Cell Biol 2011;13:132–41.

22. Meley D, Bauvy C, Houben-Weerts JH, Dubbelhuis PF, Helmond MT,Codogno P, et al. AMP-activated protein kinase and the regulation ofautophagic proteolysis. J Biol Chem 2006;281:34870–9.

23. Chen X, Li M, Li L, Xu S, Huang D, Ju M, et al. Trehalose, sucrose andraffinose are novel activators of autophagy inhuman keratinocytes throughan mTOR-independent pathway. Sci Rep 2016;6:28423.

24. Williams A, Sarkar S, Cuddon P, Ttofi EK, Saiki S, Siddiqi FH, et al.Novel targets for Huntington's disease in an mTOR-independent autop-hagy pathway. Nat Chem Biol 2008;4:295–305.

25. Ohashi A, Ohori M, Iwai K, Nakayama Y, Nambu T, Morishita D, et al.Aneuploidy generates proteotoxic stress and DNA damage concurrentlywith p53-mediated post-mitotic apoptosis in SAC-impaired cells. NatCommun 2015;6:7668.

26. Oslowski CM, Urano F. Measuring ER stress and the unfolded proteinresponse using mammalian tissue culture system. Methods Enzymol2011;490:71–92.

27. Li Y, Guo Y, Tang J, Jiang J, Chen Z. New insights into the roles of CHOP-induced apoptosis in ER stress. Acta Biochim Biophys Sin (Shanghai)2014;46:629–40.

28. Avivar-Valderas A, Bobrovnikova-Marjon E, Alan Diehl J, Bardeesy N,Debnath J, Aguirre-Ghiso JA. Regulation of autophagy during ECMdetach-ment is linked to a selective inhibition of mTORC1 by PERK. Oncogene2013;32:4932–40.

29. Dou Z, XuC,DonahueG, Shimi T, Pan JA, Zhu J, et al. Autophagymediatesdegradation of nuclear lamina. Nature 2015;527:105–9.

30. Domenech E, Maestre C, Esteban-Martinez L, Partida D, Pascual R,Fernandez-Miranda G, et al. AMPK and PFKFB3 mediate glycolysis andsurvival in response to mitophagy during mitotic arrest. Nat Cell Biol2015;17:1304–16.

31. Rufini A, Tucci P, Celardo I, Melino G. Senescence and aging: the criticalroles of p53. Oncogene 2013;32:5129–43.

32. Young AR, Narita M, Ferreira M, Kirschner K, Sadaie M, Darot JF, et al.Autophagy mediates the mitotic senescence transition. Genes Dev 2009;23:798–803.

33. Gewirtz DA. Autophagy and senescence: a partnership in search of defi-nition. Autophagy 2013;9:808–12.

34. Kang C, Xu Q, Martin TD, Li MZ, Demaria M, Aron L, et al. The DNAdamage response induces inflammation and senescence by inhibitingautophagy of GATA4. Science 2015;349:aaa5612.

35. Qin L, Wang Z, Tao L, Wang Y. ER stress negatively regulates AKT/TSC/mTOR pathway to enhance autophagy. Autophagy 2010;6:239–47.

36. Dorr JR, Yu Y,MilanovicM, Beuster G, Zasada C,Dabritz JH, et al. Syntheticlethal metabolic targeting of cellular senescence in cancer therapy. Nature2013;501:421–5.






37. Dou Z, Ghosh K, Vizioli MG, Zhu J, Sen P, Wangensteen KJ, et al.Cytoplasmic chromatin triggers inflammation in senescence and cancer.Nature 2017;550:402–06.

38. LiangQ,SeoGJ, Choi YJ, KwakMJ,Ge J, RodgersMA, et al. Crosstalk betweenthe cGAS DNA sensor and Beclin-1 autophagy protein shapes innate anti-microbial immune responses. Cell Host Microbe 2014;15:228–38.

39. Yang H, Wang H, Ren J, Chen Q, Chen ZJ. cGAS is essential for cellularsenescence. Proc Natl Acad Sci U S A 2017;114:E4612–E20.

40. Evdokimova V, Tognon C, Ng T, Sorensen PH. Reduced proliferation andenhancedmigration: two sides of the same coin?Molecularmechanisms ofmetastatic progression by YB-1. Cell Cycle 2009;8:2901–6.

41. Galluzzi L, Bravo-San Pedro JM, Demaria S, Formenti SC, Kroemer G.Activating autophagy to potentiate immunogenic chemotherapy and radi-ation therapy. Nat Rev Clin Oncol 2017;14:247–58.

42. BalachandranRS,HeighingtonCS, StarostinaNG,Anderson JW,OwenDL,Vasudevan S, et al. The ubiquitin ligase CRL2ZYG11 targets cyclin B1for degradation in a conserved pathway that facilitates mitotic slippage.J Cell Biol 2016;215:151–66.

43. Leikam C, Hufnagel AL, Otto C, Murphy DJ, Muhling B, KneitzS, et al. In vitro evidence for senescent multinucleated melano-

cytes as a source for tumor-initiating cells. Cell Death Dis 2015;6:e1711.

44. Achuthan S, Santhoshkumar TR, Prabhakar J, Nair SA, Pillai MR. Drug-induced senescence generates chemoresistant stemlike cells with lowreactive oxygen species. J Biol Chem 2011;286:37813–29.

45. Yang L, Fang J, Chen J. Tumor cell senescence response produces aggressivevariants. Cell Death Discov 2017;3:17049.

46. Li DD, Sun T, Wu XQ, Chen SP, Deng R, Jiang S, et al. The inhibition ofautophagy sensitises colon cancer cells with wild-type p53 but not mutantp53 to topotecan treatment. PLoS One 2012;7:e45058.

47. ChudeCI, Amaravadi RK. Targeting autophagy in cancer: update on clinicaltrials and novel inhibitors. Int J Mol Sci 2017;18:1279.

48. Levy JMM, Towers CG, Thorburn A. Targeting autophagy in cancer. Nat RevCancer 2017;17:528–42.

49. MosieniakG, Sikora E. Polyploidy: the link between senescence and cancer.Curr Pharm Des 2010;16:734–40.

50. WangsaD,Quintanilla I, Torabi K, Vila-CasadesusM, Ercilla A, KlusG, et al.Near-tetraploid cancer cells show chromosome instability triggered byreplication stress and exhibit enhanced invasiveness. FASEB J 2018;32:3502–17.


Jakhar et al.




2018;16:1625-1640. Published OnlineFirst July 23, 2018.Mol Cancer Res Rekha Jakhar, Monique N.H. Luijten, Alex X.F. Wong, et al. induced Senescence

−Autophagy Governs Protumorigenic Effects of Mitotic Slippage

Updated version

10.1158/1541-7786.MCR-18-0024doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2018/07/21/1541-7786.MCR-18-0024.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/16/11/1625.full#ref-list-1

This article cites 50 articles, 11 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mcr.aacrjournals.org/content/16/11/1625To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-18-0024

http://mcr.aacrjournals.org/content/suppl/2018/07/21/1541-7786.MCR-18-0024.DC1

http://mcr.aacrjournals.org/content/16/11/1625.full#ref-list-1

http://mcr.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mcr.aacrjournals.org/content/16/11/1625


autophagy governs protumorigenic effects of mitotic ...max (invitrogen) according to...

Documents