A Mechanism for Asymmetric Cell Division Resulting in … · 2017-10-12 · A Mechanism for Asymmetric Cell Division Resulting in Proliferative Asynchronicity Ipsita Dey-Guha1,2,

A Mechanism for Asymmetric Cell DivisionResulting in Proliferative AsynchronicityIpsita Dey-Guha1,2, Cleidson P. Alves1,2, Albert C. Yeh2, Salony1,2, Xavier Sole1,2,Revati Darp1, and Sridhar Ramaswamy1,2,3,4,5

Abstract

All cancers contain an admixture of rapidly and slowly prolif-erating cancer cells. This proliferative heterogeneity complicatesthe diagnosis and treatment of patients with cancer because slowproliferators are hard to eradicate, can be difficult to detect,and may cause disease relapse sometimes years after apparentlycurative treatment. While clonal selection theory explains thepresence and evolution of rapid proliferators within cancer cellpopulations, the circumstances and molecular details of howslow proliferators are produced is not well understood. Here, ab1-integrin/FAK/mTORC2/AKT1–associated signaling pathwayis discovered that can be triggered for rapidly proliferating cancercells to undergo asymmetric cell division and produce slowlyproliferating AKT1low daughter cells. In addition, evidence indi-cates that the proliferative output of this signaling cascadeinvolves a proteasome-dependent degradation process mediated

by the E3 ubiquitin ligase TTC3. These findings reveal thatproliferative heterogeneity within cancer cell populations, in part,is produced through a targetable signaling mechanism, withpotential implications for understanding cancer progression,dormancy, and therapeutic resistance.

Implications: These findings provide a deeper understanding ofthe proliferative heterogeneity that exists in the tumor environ-ment and highlight the importance of designing future therapiesagainst multiple proliferative contexts.

Visual Overview. A proposed mechanism for producing slowlyproliferating cancer cells. http://mcr.aacrjournals.org/content/early/2015/01/09/1541-7786.MCR-14-0474/F1.large.jpg.Mol Cancer Res; 13(2); 223–30. �2015 AACR.

IntroductionIn cell culture, dividing cancer cells usually produce two

daughter cells that divide again in relative synchrony within afew hours of each other. Occasionally, however, a cancer celldivides to produce progeny that are asynchronous with respect tothe next cell cycle, with one daughter cell having a markedlyslower cell division time than the other, on the order of days. Werecently found that this proliferative heterogeneity correlates withcancer cells asymmetrically suppressing AKT protein kinase levelsby about ninety percent duringmitosis just before cytokinesis (1).These rare asymmetries produce one AKTnormal daughter cell thatrapidly enters the next cell cycle and another AKTlow cell thatremains dormant for a more prolonged time before dividingagain. Slowly cycling AKTlow cells reduce their production ofreactive oxygen species (i.e., ROSlow), downregulate proliferationproteins (e.g., MKI67low, MCM2low), suppress multiple nuclear

histone marks similar to quiescent cell populations (e.g.,H3K9me2low), and transcriptionally upregulate the HES1 tran-scription factor thatmaymark exit from the cell cycle intoG0 (i.e.,HES1high; ref. 1). As AKTlow cells do eventually divide, convertingto an AKTnormal proliferative phenotype over time, we tentativelyhave used the term "G0-like" to describe this temporary andreversible cell state. Significantly, we have also found AKTlow

cancer cells within actual human breast tumors where they appearhighly resistant to prolonged treatment with combination che-motherapy using adriamycin, cyclophosphamide, and paclitaxel,suggesting these slow proliferators may constitute an importantbut unappreciated reservoir of treatment resistance in patientswith breast cancer. We therefore reasoned that understandingmore precisely how AKTlow cancer cells arise at a molecular levelmight provide fundamental insight into cancer biology withpotential clinical relevance.

Materials and MethodsCell culture

HCT116 colon and MCF7 breast were purchased from theATCC where they were authenticated. HCT116-AKT1/2�/� cellswere purchased fromHorizonDiscovery where they were authen-ticated. MCF7 cells were maintained in DMEM, 10% FCS,40 mmol/L glutamine, 100 U/mL penicillin, and 100 mg/mLstreptomycin. HCT116 and HCT116-AKT1/2�/� cells were main-tained in McCoy 5a medium supplemented with 10% FCS,100 U/mL penicillin, and 100 mg/mL streptomycin. Cells weregrown in a humidified atmosphere at 37�C and 5% CO2.

1Massachusetts General Hospital Cancer Center, Boston, Massachu-setts. 2Harvard Medical School, Boston, Massachusetts. 3Broad Insti-tute of Harvard and MIT, Cambridge, Massachusetts. 4Harvard StemCell Institute,Cambridge,Massachusetts. 5Harvard-Ludwig Center forCancer Research, Boston, Massachusetts.

I. Dey-Guha and C.P. Alves contributed equally to this article.

Corresponding Author: Sridhar Ramaswamy, Massachusetts General HospitalCancer Center, 185 Cambridge Street, Boston, MA 02114. E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-14-0474

�2015 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org 223

on July 24, 2020. © 2015 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst January 12, 2015; DOI: 10.1158/1541-7786.MCR-14-0474

http://mcr.aacrjournals.org/

Generation of AKT1-mutant cell linespDD AKT1(WT) and pMSCV-puro-Ctag-mCherry were gifts

from Joan Brugge (Harvard Medical School, Boston, MA).AKT1(WT) cDNA was purified using PCR after cutting PDDAKT1(WT) with restriction enzymes BamHI and XhoI. Afterpurification, the product was ligated into pMSCVpuro-C-tag-mCherry cut with BglII and SalI. All the AKT1 mutants weregenerated using the QuikChange Site Directed Mutagenesis Kit(Agilent Technologies) and the product was ligated intopMSCVpuro-C-tag-mCherry. The resulting vector pMSCV-puro-AKT1-mCherry was subcloned into DH5a-competent cells (Invi-trogen). Sequencing verification of the fusion product was per-formed by the MGH DNA Core Facility with primers pMSCV 50-CCCTTGAACCTCCTCGTTCGACC-30 and pMSCV 30-GAGACG-TGCTACTTCCATTTGTC-50. Virus carrying the desired fusion genewas produced by transfecting HEK 293T cells with target vectorpMSCV-puro-AKT1-mCherry and packaging vector pCL-Amphousing the Mirus TransIT-293 transfection reagent and establishedprotocols. Virus was collected 24 hours after transfection. Beforeinfection, cells were plated in a 6-well plate in DMEM, 10% FCS.Infection was performed 24 hours later by adding 0.5mLDMEM,10% FCS, 0.5 mL pooled virus, and 1 mL 1,000� polybrene perwell. A media change was performed the following day and cellswere allowed to grow to confluency before splitting into a 10 cmdish and selection with 2 mmol/L puromycin. After selection, cellswere allowed to grow to confluency before clones were selectedusing single-cell sorting (Becton Dickinson FACSAria II). Singlecells were filtered by gating on the brightest 5% of cells in the PETexas red channel and sorted into individual wells of a 96-wellplate. Clones were harvested between 14 and 21 days.

Drug treatment in vitroCells were seeded onto collagen IV–coated coverslips, allowed

to attach overnight, and treated with vehicle (DMSO) or AKT1/2inhibitor (HCT116:20mmol/L;MCF7:2mmol/L; Sigma),MK2206(HCT116: 10 mmol/L; MCF7: 3 mmol/L; Selleck Chemicals),TORIN1 (HCT116: 0.5 mmol/L; MCF7: 0.25 mmol/L; Tocris Bio-science), AZD8055 (HCT116: 0.7 mmol/L; MCF7: 0.1 mmol/L;Selleck Chemicals), INK128 (HCT116: 0.05 mmol/L; MCF7:0.01 mmol/L; Active Biochem), Palomid 529 (HCT116:10 mmol/L; MCF7: 20 mmol/L; Selleck Chemicals), Rapamycin(HCT116: 20 mmol/L; MCF7: 20 mmol/L; Sigma), RAD-001(HCT116: 10 mmol/L; MCF7: 5 mmol/L; Selleck Chemicals),BKM-120 (HCT116: 1.5 mmol/L; MCF7: 0.5 mmol/L; Active Bio-chem), FAK inhibitors [PF-562271: 1 mmol/L (Pfizer) and NVP-TAE226 : 1 mmol/L (Novartis), for both cell lines] for 72 hours or144 hours and bortezemib (HCT116: 1 mmol/L; MCF7: 4 mmol/L;Selleck Chemicals)MG-132 (vehicle: ethanol; HCT116: 1 mmol/L;MCF7: 10 mmol/L), for 24 hours.

shRNA constructsHuman TRIPZ lentiviral inducible shRNAmirs for RICTOR

(clone ID: V2THS_120392, V2THS_120389, V2THS_38014,V2THS_225915), FAK (clone ID: V2THS_57326, V2THS_325805),b1-integrin (clone ID:V2THS_133469,V2THS_390997), nonsilen-cing, and empty vector were purchased fromOpen Biosystems andvirus was generated using a standard protocol. Infection wasperformed 24 hours later in MCF7 and HCT116 cells with thelentiviral particles followed by selectionwith 2 mmol/L puromycin.After selection, cells were allowed to grow to confluency. TheshRNAs were induced using 2 mg/mL doxycycline for 72 hours.

The TTC3 virus was purchased from Sigma-Aldrich and infected inHCT116 and MCF7 cells and the standard protocol for selectionwas followed.

Antibody activation and inhibitionCells were incubated in media containing 10% FBS and the

respective b1-integrin antibody: inhibiting [AIIB2:20 mg/mL(Developmental Studies Hybridoma Bank), P4C10: 10 mg/mL(Millipore)], and activating (TS2/16 and 12G10: 10 mg/mL; SantaCruz Biotechnology), for 1 hour at 4�C. Cells were then plated oncollagen IV–coated coverslips (Sigma) and incubated in theantibody at 37�C for 24 hours.

Immunofluorescence stainingCells were grown directly on collagen IV–coated coverslips

(Sigma). Cells were fixed in 3.7% formalin, permeabilized using0.1%TritonX-100, and treatedwith 0.1%SDS. Theywere blockedin 1% BSA and then incubated with primary antibody(a-H3K9me2, a-Hes1 a-TTC3, and a-AKT(phospho-T308;Abcam); a-MCM2, a-Tubulin, a-pan-AKT, and a-AKT(phos-pho-S473; Cell Signaling Technology), diluted in blocking solu-tion, washed, and incubated with the respective secondary anti-body. Cells were mounted using hardset mounting media con-tainingDAPI (Vector Laboratories). All secondary antibodieswereAlexa Fluor conjugates (488, 555, 568, 633, and 647; Invitrogen).

Collagen matrix studiesSlides coated with type-I collagen (control) and AlignCol

woven with large collagen fibers (100–200 nm; Advanced Bio-matrix) were incubated for 1 hour in 70% ethanol and thenwashed with PBS. Cells were then plated on the slides, incubatedfor 24 hours, and processed for immunofluorescence.

Confocal imagingImmunofluorescence imaging was performed on a Nikon

Eclipse Ti A1R-A1 confocal microscope. G0-like slow prolifera-tors were specifically identified as cells in the bottom 10% ofcoincident staining for MCM2, H3K9me2, and HES1. Cellswere scored by counting G0-like versus other proliferatingcancer cells among 10,000 cells from multiple fields of viewat 20� magnification.

Western blotting and immunoprecipitationWe used standard protocols for SDS-PAGE electrophoresis and

the following primary antibodies:a-RICTOR,a-mTOR, Phospho-AKT-Ser473 (D9E; Cell Signaling Technology) and a-RAPTOR,a-FAK, a-b1-integrin, a-TTC3, Pan-AKT, and GAPDH (Abcam).For immunoprecipitation studies, cells were synchronized with200 ng/mL of nocodazole for 12 hours and then released for2 hours. Cells were rinsed with PBS, fixed with 0.37% formalde-hyde, and quenched with 0.25 mol/L glycine. Cell lysates wereprepared in lysis buffer [1% Triton X-100, 150 mmol/L NaCl,3 mmol/L MgCl, 40 mmol/L HEPES (pH 7.5), 50 mmol/L NaF,EDTA-free protease inhibitor and phosphatase inhibitor (Roche)]and centrifuged at 14,000 � g for 10 minutes. Supernatant (250mg) was incubated with the indicated antibodies [a-FAK(AbCam), a- RICTOR (Santa Cruz Biotechnology)], for 4 hoursat 4�Cwith rotation and thenwith 50mLof a 50%slurry of proteinG-sepharose (Roche) for 1 hour. Immunoprecipitates werewashed and resolved by SDS-PAGE electrophoresis.

Dey-Guha et al.

Mol Cancer Res; 13(2) February 2015 Molecular Cancer Research224




ResultsAs AKTlow cancer cells only partially suppress total AKT protein

levels, we first asked whether asymmetric division occurs in thecomplete absence of all three AKT isoforms (i.e., AKT1, AKT2, andAKT3). To do so, we obtainedHCT116 colorectal cancer cells withadeno-associated virus (AAV)-mediated disruption of the AKT1and AKT2 gene loci (i.e., AKT1/2�/� cells; ref. 2). Importantly,AKT1/2�/� cells do not express either AKT1 or AKT2, nor do theyexpress AKT3, and thus survive and proliferate in the completeabsence of AKT signaling, presumably through compensatorychanges that arose during their initial selection. We used confocalmicroscopy to score AKT1/2�/� cell populations for rare, asym-metrically dividing, and G0-like cancer cells that express thepreviously validated H3K9me2low/MCM2 low/HES1high molecu-lar profile, which specifically marks AKTlow slow proliferatorsas we had previously shown (Fig. 1A; ref. 1). Interestingly, thisAKT1/2�/� line had virtually no asymmetrically dividing or G0-like cells compared with wild-type HCT116 (the parental linefrom which AKT1/2�/� cells are derived; Fig. 1B). However,

lentiviral-mediated overexpression of an AKT1 cDNA in AKT1/2�/� cells completely restored formation of both asymmetricallydividing and G0-like cells, while overexpression of AKT2 did not,suggesting that AKT1 is both necessary and sufficient for theproduction of G0-like cells (Fig. 1B).

On the basis of this result, we used site-directed mutagenesis toidentify AKT1 domains that might be required for its partialsuppression during asymmetric division. We created a seriesof AKT1 cDNA constructs with mutations in critical amino acidsknown to be important for various aspects of AKT1 signaling(Fig. 1C). We then overexpressed each mutant AKT1 construct inAKT1/2�/� cells and scored these engineered cells for both asym-metrically dividing and G0-like cancer cells. We first asked wheth-er AKT1 kinase activity was necessary for production of these slowproliferators. We found that AKT1-K179M (a commonly studiedmutation in the kinase pocket that renders AKT1 catalyticallydead) failed to restore production of asymmetrically dividing andG0-like cells in the AKT1/2�/� line (Fig. 1B). In addition, AKT1-D292A (a mutant hypomorph with diminished kinase catalyticactivity) did so only weakly compared with wild-type AKT1

Asymmetricdividing

cell

Slowproliferator

3

2

1

0

3

2

1

0% A

sym

met

ric

mit

ose

s

% G

0-lik

e ce

lls

WT

WT

AKT1

AKT1

AKT2

AKT2

AKT1/2

-/-

AKT1/2

-/-

HCT116 MCF7

AKT1-K179M AKT1-K179MAKT1-D292A

AKT1/2MK2206

AKT1-D292A

AKT1/2MK2206

shTTC3 hp3 shTTC3 hp3shTTC3 hp4shTTC3 hp5

shTTC3 hp4shTTC3 hp5AKT1-K8R AKT1-K8R

AKT1-K14RAKT1-K8R/K14R

AKT1-K14RAKT1-K8R/K14RBortezomibMG-132

BortezomibMG-132

% of Control % of Control

N/AN/A

N/AN/AN/A

TTC3+

slow

proliferator

−100% −50% 500% 1,000%0 −100% −50% 500% 1,000%0

mTORC2 complex

K8 K14 K179 D292 T308 T450 S473

TTC3

GAPDH

NS hp1

hp2

hp3

hp4

hp5

A

B C

D

E F

Figure 1.A mechanism for AKT1low slowproliferators: AKT1, TTC3, andproteasome. A, HCT116 cells stainedfor DAPI, H3K9me2, HES1, and TUBB.Merged image represents respectivestains with underlying DAPI stain.Arrows indicate a G0-like cell. Bar,10 mm. B, bar graph of percentages ofH3K9me2low/MCM2low/HES1high

asymmetric mitoses and G0-like cellsin AKT1/2�/�HCT116 cells with cDNAsfor AKT1 or AKT2 or AKT1-K179M orAKT1-D292A. C, schematic model ofAKT1 protein with C, catalytic; P,phosphorylation; Ub, ubiquitination;PH, pleckstrin homology; HD,hydrophobic domain. D, graphicalrepresentation of percentage changein H3K9me2low/MCM2low/HES1high

asymmetrically dividing and G0-likecells relative to control in HCT116 andMCF7 cell lines. Solid bars representasymmetrically dividing and clearbars represent G0-like cancer cells.Error bars indicate mean � SEM for 3replicates. E, MCF7 cells stained forDAPI, H3K9me2, and TTC3. Mergedimage represents respective stainswith underlying DAPI stain. Arrowindicates a G0-like TTC3þ cell. Bar,10 mm. F, Western blot analysis ofshort hairpin TTC3 knockdown.

A Mechanism that Produces Slowly Proliferating Cancer Cells

www.aacrjournals.org Mol Cancer Res; 13(2) February 2015 225




(Fig. 1B; refs. 3, 4). These results were consistent with AKT1 kinaseactivity being necessary for asymmetric division.

We previously noted that treating wild-type cancer cells withallosteric AKT inhibitors at low doses dramatically increasesthe frequency of both asymmetrically dividing and G0-likecells in HCT116 and MCF7 breast cancer cells (i.e., AKT1/2,MK2206; Fig. 1D; ref. 1). These allosteric inhibitors are knownto bind to the AKT1 pleckstrin homology domain, inducingconformational change and protein displacement from the cellmembrane, thus promoting its ubiquitination and protea-some-mediated degradation (5). We therefore hypothesizedthat asymmetric division might depend on targeted degrada-tion of AKT1 protein. TTC3 is a RING-type E3 protein-ligaseknown to ubiquitinate AKT1 at the lysine-8 and lysine-14residues leading to its destruction by the proteasome (6). We

found that G0-like cells from wild-type MCF7 express highlevels of TTC3 protein compared with proliferating cells,consistent with a potential role for this E3 ligase in producingAKT1low cells (Fig. 1E). In addition, inducible shRNA knock-down of TTC3 suppressed the frequency of G0-like cells inboth wild-type HCT116 and MCF7 (Fig. 1D). Furthermore,AKT1-K8R, AKT1-K14R, and AKT1-K8R/K14R double mutantproteins (which cannot be ubiquitinated by TTC3) failed torescue the formation of G0-like cells in the AKT1/2�/� line(Fig. 1D-left). Moreover, two different small molecules thatinhibit proteasome function reduced the frequency of G0-likecells in both wild-type HCT116 and MCF7 when used at dosesthat do not affect overall cell proliferation (i.e., MG-132,bortezomib; Fig. 1D). Overall, these results were consistentwith enzymatically active AKT1 being ubiquitinated by TTC3

HCT 116 MCF7

AKT1-T308AAKT1-S473AAKT1-T450AAKT1-T308A/S473A

Torin 1AZD8055INK-128Palomid-529RapamycinRAD-001BKM-120

shRICTOR hp1shRICTOR hp4

Torin 1AZD8055INK-128Palomid-529RapamycinRAD-001BKM-120

shRICTOR hp1shRICTOR hp4

N/A

N/AN/AN/A

−100% −50% 500% 1,000%0 −100% −50% 500% 1,000%0

A

B

C

% of Control % of Control

RICTOR

GAPDH

NS hp1

hp2

hp3

hp4

HCT116 MCF7

FAK

FAK

mTOR

Rictor

RictorRICTOR

Raptor

FAK

FAK

mTOR

Rictor

Rictor

Raptor

IP

IPFAK

RICTORIP

IPFAK

10%

inpu

t

No Ab

IgG

IP 10%

Inpu

t

No Ab

IgG

IP

Figure 2.A mechanism for AKT1low slowproliferators: FAK, mTORC2, andAKT1. A, graphical representation ofpercentage change in H3K9me2low/MCM2low/HES1high asymmetricallydividing and G0-like cells relative tocontrol in HCT116 and MCF7 cell lines.Solid bars represent asymmetricallydividing and clear bars representG0-like cancer cells. Error barsindicate mean � SEM for 3 replicates.B, Western blot analysis of shorthairpin RICTOR knockdown. C, HCT116and MCF7 cells in M-phase of the cellcycle, FAK IP with anti-FAK andimmunoblotted with anti-FAK, anti-mTOR, anti-RICTOR ,and anti-RAPTOR antibody. Reciprocally,RICTOR IP with anti-RICTOR andimmunoblotted with anti-RICTOR oranti-FAK antibody.

Dey-Guha et al.





and degraded by the proteasome during cell division to pro-duce slow proliferators.

AKT1 is usually activated by two different upstream kinases:PDPK1 phosphorylates AKT1 at the T308 residue, whereas themTORC2 kinase complex phosphorylates the AKT1-S473and AKT1-T450 sites (7, 8). Similar to AKT1 cDNA, overexpres-sion of the AKT1-T308A cDNA mutant (which cannot be phos-phorylated by PDPK1) completely restored the productionof asymmetrically dividing and G0-like cells in AKT1/2�/� cells(Fig. 2A, left). In contrast, AKT1-S473A, AKT1-T450A, and anAKT1-T308A/AKT1-S473Adoublemutant (all ofwhich cannot bephosphorylated bymTORC2) did not produce phenotypic rescuein these cells (Fig. 2A, left). We also found that four structurallydifferent small molecules that inhibit both mTORC2 andmTORC1 signaling reduced the frequency of asymmetricallydividing and G0-like cells in both wild-type HCT116 and MCF7cancer cells at low doses that did not appreciably inhibitcell proliferation (i.e., TORIN1, AZD8055, INK-128, Palomid-529; Fig. 2A). In contrast, the production of G0-like cells was notsuppressed either by inhibitors that preferentially target theTORC1 signaling complex alone (i.e., rapamycin, RAD-001) orby a pan-class I PI3K inhibitor (i.e., BKM-120), when used attarget-suppressing doses in these wild-type cells (Fig. 2A). Inaddition, inducible shRNA knockdown of RICTOR (an obligatemember of the mTORC2 signaling complex) suppressed theproduction of both asymmetrically dividing and slowly prolifer-ating G0-like cells in both wild-type HCT116 and MCF7 (Fig. 2Aand B). We also found in asymmetrically dividing cells, that theslow proliferator daughter cells (i.e., H3K9me2low) were phos-pho-AKT1-S473high but phospho-AKT1-T308normal (Fig. 3A). Incontrast, after cytokinesis these slow proliferators (i.e.,

H3K9me2low) were AKTlow and commensurately phospho-AKT1-S473low and phospho-AKT1-T308low (Fig. 3B; ref. 1). Inaggregate, these results supported a dynamic model wherebydifferential phosphorylation of AKT1 by mTORC2 may precedethe production of slow proliferators with low levels of AKT1protein.

To identify an upstream regulator that might activatemTORC2 signaling during asymmetric division, we next usedan immunoprecipitation approach to identify proteins thatphysically interact with the mTORC2 complex during mitosis.We first treated HCT116 and MCF7 cells with nocodazole, tosynchronize cells in metaphase, and then prepared whole-cellprotein lysates 2 hours after release of this synchronization withthe cells still in mitosis. We found that immunoprecipitationwith a RICTOR antibody (under conditions that maintainintegrity of the mTORC2 complex in whole-cell lysates) pulleddown focal adhesion kinase (FAK) protein in both HCT116 andMCF7. Reciprocally, immunoprecipitation with a FAK antibodypulled down both mTOR kinase and RICTOR, but not RAPTOR(an obligate member of the related mTORC1 complex), con-firming the specific interaction of FAK with mTORC2 complexin these cells (Fig. 2C). This observation suggested that FAKactivity might regulate mTORC2-associated AKT1 degradationand asymmetric cancer cell division. Furthermore, inducibleshRNA knockdown of FAK increased both asymmetricallydividing and G0-like cells in HCT116 and MCF7 (Fig. 4Aand D). Similarly, inhibiting FAK enzymatic activity with twodifferent small molecules increased the frequency of bothasymmetrically dividing and G0-like cells (i.e., PF-562271,NVP-TAE226; Fig. 4A). However, FAK inhibitors failed toincrease asymmetries or slow proliferators after RICTOR

Asymmetricdividing

cell

Slowproliferator

A

B

Figure 3.A mechanism for AKT1low slowproliferators: phospho-AKT1. A and B,MCF7 cells stained for DAPI,H3K9me2, TUBB, and phospho-AKT1-S473, phospho-AKT1-T308, or pan-AKT. Merged image representrespective stainswith underlyingDAPIstain. Arrows indicate a G0-like cell.Bar, 10 mm.






knockdown (Fig. 4A). These findings were consistent with amodel whereby a loss of FAK activity might induce mTORC2-mediated asymmetric cancer cell division.

Integrins are a family of heterodimeric transmembranereceptors that transduce signals from the extracellular matrix,by activating signaling intermediaries, including FAK, toincrease the cell cycle, survival, and motility of cancer andnormal cells (9). We therefore reasoned that decreased integrinsignaling might be the proximate cause for a loss in FAKactivity resulting in asymmetric mitosis. In fact, shRNA knock-down of b1-integrin (i.e., ITGB1) increased the fraction ofasymmetrically dividing and G0-like cells in both HCT116and MCF7 (Fig. 4B and E). In addition, blocking b1-integrinfunction with two different monoclonal antibodies alsoincreased both asymmetrically dividing and G0-like cells (i.e., A2B2, P4C10; Fig. 4B; ref. 10). However, activating b1-integrin signaling with two other monoclonal antibodies,which force b1-integrin into a constitutive "on" state by

imposing a conformational change, eliminated both asymme-tries and slow proliferators in these cell lines (i.e., TS2/16,12G10; Fig. 4B; ref. 10).

These observations suggested that the asymmetric cancercell divisions might result from random variation in b1-integ-rin signaling related to extracellular irregularities withincell culture. We therefore grew cancer cells on engineeredmatrices displaying type-I collagen (a major extracellularmatrix protein that activates b1-integrin) closely aligned in aregular fibrillar pattern, to assure uniform b1-integrin activa-tion in any cancer cell undergoing mitosis (11). Notably,cancer cells proliferating in this structured collagen matrix didnot produce asymmetries or G0-like cells, in contrast withtypical cell culture (Fig. 4C). In the aggregate, our results wereconsistent with loss in b1-integrin signaling during mitosis(likely resulting from random irregularity in extracellular type-I collagen) triggering a conserved pathway to produce slowproliferators in vitro.

HCT116 MCF7

HCT116 MCF7

shFAK hp1shFAK hp2PF-562271NVP-TAE226

shFAK hp1shFAK hp2PF-562271NVP-TAE226

shRICTOR hp1 and PF-562271shRICTOR hp4 and PF-562271shRICTOR hp1 and NVP-TAE226shRICTOR hp4 and NVP-TAE226

shRICTOR hp1 and PF-562271shRICTOR hp4 and PF-562271shRICTOR hp1 and NVP-TAE226shRICTOR hp4 and NVP-TAE226

−100% −50% 500% 1,000%0 −100% −50% 500% 1,000%0

% of Control

−100% −50% 500% 1,000%0

% of Control

−100% −50% 500% 1,000%0

% of Control

% of Control

A2B2P4C10TS2/1612G10

A2B2P4C10TS2/1612G10

shβ1-Integrin hp1shβ1-Integrin hp2

shβ1-Integrin hp1shβ1-Integrin hp2

8

6

4

2

0

3

2

1

0

Contro

lAlig

ned

Contro

lAlig

ned

% A

sym

met

ric

mit

ose

s

% G

0-lik

e ce

lls

FAK

GAPDH GAPDH

β1-Integrin

NS hp1

hp2

NS hp1

hp2

A

B

C D E

Figure 4.A mechanism for AKT1low slow proliferators: b1-integrin and FAK. A and B, graphical representation of percentage change in H3K9me2low/MCM2low/HES1high

asymmetrically dividing and G0-like cells relative to control in HCT116 and MCF7 cell lines. Solid bars represent asymmetrically dividing and clear bars representG0-like cancer cells. Error bars indicate mean � SEM for 3 replicates. C, bar graph of percentages of H3K9me2low/MCM2low/HES1high asymmetric mitoses andG0-like cells in MCF7 cells plated on control (random) or aligned type-I collagen fibrils (aligned). D and E, Western blots of short hairpin FAK and b1-integrinknockdown in HCT116 cells with nonsilencing shRNA (NS) as control.


Dey-Guha et al.




DiscussionThe proliferative heterogeneity among cancer cells within

tumors generally correlates with differences in growth, responseto treatment, and disease relapse in patients with cancer (12).Despite recent progress, however, we do not understand clearlyhow this heterogeneity is generated in the first place. We previ-ously discovered that cancer cells occasionally divide asymmet-rically to spawn AKTlow, MCM2low, H3K9me2low, HES1high prog-eny that proliferate slowly and are resistant to cytotoxic chemo-therapy in cell culture (�1% of cell divisions; ref. 1). We alsodemonstrated the existence of these AKTlow cancer cells withinactual human breast tumors where they appear to survive inten-sive, combination chemotherapy, suggesting that these cells maymediate clinically important chemoresistance (1). In our currentwork, we reveal a signaling pathway that is triggered in dividingcancer cells to spawn these slowproliferators in vitro. This pathwayinvolves a decrease in b1-integrin/FAK activity, activation of themTORC2 complex, and suppression of AKT1 protein levelsthrough TTC3/proteasome-mediated degradation. Interestingly,any dividing cancer cell appears capable of triggering the b1-integrin pathway that we describe to produce AKT1low slowproliferators. This facultative behavior presumably occurs if divid-ing cancer cells encounter irregularities in extracellular type Icollagen, although additional cooperative factors yet to be dis-coveredmay also be required.Moreover, wefind that activation ofb1-integrin signaling with monoclonal antibodies or inhibitionof mTORC2 signaling with small molecules reduces asymmetriccancer cell division and the production of these slowproliferators.Our findings might therefore suggest potentially new avenues forexperimentally or therapeutically manipulating and studying theproduction of AKT1low slow proliferators both in vitro and in vivo.

These results also offer potentially useful molecular insightinto different signaling molecules that are under intensiveinvestigation as therapeutic targets for various cancer types,which may carry implications for the development and use ofclinical inhibitors that target these important molecules. Forexample, the MCF7 and HCT116 cancer cells we study haveactivating mutations in PIK3CA, and are thus dependent onconstitutive PI3K/AKT signaling for their proliferation andsurvival. Despite this dependency, however, we find that theseERþ breast and colorectal cancers retain the b1-integrin path-way that we describe to produce AKT1low slow proliferators.This suggests that cancer cells may actually derive some indis-pensible advantage from suppressing AKT1 to produce slowproliferators in this way, although further work will determinewhether our findings extend to additional molecular subtypesof cancer. In addition, we find that a quantitative reduction inb1-integrin, FAK, or AKT1 (rather than AKT2/3) signaling incancer cells produces this reversible cell-cycle arrest through aconserved pathway, compared with complete suppression ofthese targets that generally results in cell death or senescence.Our results also suggest that FAK may physically interact with

and functionally suppress the mTORC2 signaling complexduring cell division, which we believe is not generally appre-ciated. Moreover, while mTORC2 activity is normally requiredfor AKT1 activation, we find that this multifunctional signalingcomplex is also necessary for triggering AKT1 degradationduring asymmetric cancer cell division, although additionalexperiments will be necessary to understand exactly how thishappens (8). Finally, we find that TTC3-mediated proteasomedegradation of AKT1 is necessary for the production ofAKT1low slow proliferators, although we do not yet understandprecisely how the expression and activity of this E3-ubiquitingligase and the proteasome is regulated during asymmetricmitosis. Deeper insight into these molecular interactions, theprecise cellular contexts in which they occur, and why cancercells retain this functional pathway for producing slow pro-liferators may thus provide further useful insight into cancerbiology.

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

Authors' ContributionsConception and design: I. Dey-Guha, C.P. Alves, S. RamaswamyDevelopment of methodology: I. Dey-Guha, C.P. Alves, S. RamaswamyAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): I. Dey-Guha, C.P. Alves, A.C. Yeh, SalonyAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): I. Dey-Guha, C.P. Alves, A.C. Yeh, Salony, X. Sole,S. RamaswamyWriting, review, and/or revision of the manuscript: I. Dey-Guha, C.P. Alves,X. Sole, S. RamaswamyAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): R. Darp, S. RamaswamyStudy supervision: S. Ramaswamy

AcknowledgmentsThe authors thank Jim DeCaprio, Richard Hynes, Bill Sellers, and Bob

Weinberg for helpful discussions.

Grant SupportS. Ramaswamy is supported by a Stand Up to Cancer Innovative Research

Grant (Grant number SU2C-AACR-IRG0911). StandUp ToCancer is a programof the Entertainment Industry Foundation administered by the AmericanAssociation for Cancer Research. This work is also supported by awards fromthe National Cancer Institute, Howard Hughes Medical Institute (Physician-Scientist Early Career Award), SusanG. Komen for theCure, andProstate CancerFoundation (to S. Ramaswamy). C.P. Alves is supported by an award fromCNPq "Ciencia sem Fronteiras"—Brazil (202620/2012-3). A.C. Yeh was sup-ported by anHHMIMedical Student Research Fellowship (2010–2012). X. Soleis supportedby a "Bolsa deAmpliaci�ondeEstudios, Instituto de SaludCarlos III,Ministerio de Economía y Competitividad (BA12/00021)"—Spanish postdoc-toral fellowship award.

ReceivedAugust 25, 2014; revisedNovember 5, 2014; acceptedNovember 17,2014; published OnlineFirst January 12, 2015.

References1. Dey-Guha I,Wolfer A, Yeh AC, Albeck JG, Darp R, Leon E, et al. Asymmetric

cancer cell division regulated by AKT. Proc Natl Acad Sci U S A2011;108:12845–50.

2. Ericson K,GanC,Cheong I, RagoC, Samuels Y, Velculescu VE, et al. Geneticinactivation of AKT1, AKT2, and PDPK1 in human colorectal cancer cells

clarifies their roles in tumor growth regulation. Proc Natl Acad Sci U S A2010;107:2598–603.

3. Okuzumi T, Fiedler D, Zhang C, Gray DC, Aizenstein B, Hoffman R,et al. Inhibitor hijacking of Akt activation. Nat Chem Biol 2009;5:484–93.






4. Aoki M, Batista O, Bellacosa A, Tsichlis P, Vogt PK. The akt kinase:molecular determinants of oncogenicity. Proc Natl Acad Sci U S A 1998;95:14950–5.

5. Jo H, Mondal S, Tan D, Nagata E, Takizawa S, Sharma AK, et al. Smallmolecule-induced cytosolic activation of protein kinase Akt rescuesischemia-elicited neuronal death. Proc Natl Acad Sci U S A 2012;109:10581–6.

6. Suizu F, Hiramuki Y, Okumura F, MatsudaM, Okumura AJ, Hirata N, et al.The E3 ligase TTC3 facilitates ubiquitination and degradation of phos-phorylated Akt. Dev Cell 2009;17:800–10.

7. Sarbassov DD, Guertin DA, Ali SM, Sabatini DM. Phosphorylation andregulation of Akt/PKB by the rictor-mTOR complex. Science 2005;307:1098–101.

8. Manning BD,Cantley LC. AKT/PKB signaling: navigating downstream. Cell2007;129:1261–74.

9. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell2002;110:673–87.

10. Byron A, Humphries JD, Askari JA, Craig SE, Mould AP, Humphries MJ.Anti-integrin monoclonal antibodies. J Cell Sci 2009;122(Pt 22):4009–11.

11. Bessea L, CoulombB, Lebreton-Decoster C,Giraud-GuilleMM.Productionof ordered collagen matrices for three-dimensional cell culture. Biomater-ials 2002;23:27–36.

12. Hong WK, Bast RC, Hait WN, Kufe DW, Pollock RE, Weichselbaum RR,et al. Cancer Medicine, 8th Edition. Shelton, CT: People's MedicalPublishing House, 2010. xxv, 2021 pp.


Dey-Guha et al.




2015;13:223-230. Published OnlineFirst January 12, 2015.Mol Cancer Res Ipsita Dey-Guha, Cleidson P. Alves, Albert C. Yeh, et al. Proliferative AsynchronicityA Mechanism for Asymmetric Cell Division Resulting in

Updated version

10.1158/1541-7786.MCR-14-0474doi:

Access the most recent version of this article at:

Overview

Visual

http://mcr.aacrjournals.org/content/13/2/223/F1.large.jpgA diagrammatic summary of the major findings and biological implications:

Cited articles

http://mcr.aacrjournals.org/content/13/2/223.full#ref-list-1

This article cites 11 articles, 6 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/13/2/223.full#related-urls

This article has been cited by 5 HighWire-hosted articles. Access the articles 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/13/2/223To 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-14-0474

http://mcr.aacrjournals.org/content/13/2/223/F1.large.jpg

http://mcr.aacrjournals.org/content/13/2/223.full#ref-list-1

http://mcr.aacrjournals.org/content/13/2/223.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/13/2/223


A Mechanism for Asymmetric Cell Division Resulting in … · 2017-10-12 · A Mechanism for Asymmetric Cell Division Resulting in Proliferative Asynchronicity Ipsita Dey-Guha1,2,

Documents