g protein-coupled receptors participate in cytokinesis

G Protein-Coupled Receptors Participate in Cytokinesis

Xin Zhang,1,2,3 Anne V. Bedigian,1,2 Wenchao Wang,1,3 and Ulrike S. Eggert1,2,4*1Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts2Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts3High Magnetic Field Laboratory, Chinese Academy of Sciences, Hefei, People’s Republic of China4Department of Chemistry and Randall Division of Cell and Molecular Biophysics, King’s College London, SE1 1UL, United Kingdom

Received 21 May 2012; Accepted 16 July 2012Monitoring Editor: Douglas Robinson

Cytokinesis, the last step during cell division, is a highlycoordinated process that involves the relay of signalsfrom both the outside and inside of the cell. We have abasic understanding of how cells regulate internal events,but how cells respond to extracellular cues is less explored.In a systematic RNAi screen of G protein-coupled receptors(GPCRs) and their effectors, we found that someGPCRs areinvolved in cytokinesis. RNAi knockdown of these GPCRscaused increased binucleated cell formation, and live cellimaging showed that most formed midbodies but failed atthe abscission stage. OR2A4 (olfactory receptor, family 2,subfamily A, member 4) localized to cytokinetic structuresin cells and its knockdown caused cytokinesis failure at anearlier stage, likely due to effects on the actin cytoskeleton.Identifying the downstream components that transmitGPCR signals during cytokinesis will be the next step andwe show that GIPC1 (GIPC PDZ domain containing fam-ily, member 1), an adaptor protein for GPCRs, may play apart. RNAi knockdown of GIPC1 significantly increasedbinucleated cell formation. Understanding the moleculardetails of GPCRs and their interaction proteins in cytokine-sis regulation will give us important clues about GPCRs sig-naling as well as how cells communicate with theirenvironment during division. VC 2012 Wiley Periodicals, Inc

KeyWords: cytokinesis, GPCRs

Introduction

Cytokinesis is the last step of cell division, in whichcells physically separate their content into two daugh-

ter cells. It involves many cellular structures and compart-

ments, including microtubules and its associated proteins,a contractile ring that is composed of actin, myosin II andmany other proteins, intracellular vesicles as well as cellmembrane [Rappaport, 1986, 1996; Eggert et al., 2006;Atilla-Gokcumen et al., 2010, 2011; Normand and King,2010]. Successful cytokinesis requires temporal and spatialcontrol of multiple cellular events. The cell needs toaccurately coordinate these different components to ensurethe proper positioning of the contractile ring, ingression ofthe furrow, equal partitioning of cellular contents, and themembrane sealing between two daughter cells. For mostcell types, cytokinesis is a symmetric process in whichgenetic materials and cellular contents are divided evenly.In some specialized cells, for example stem cells, cytokinesisneeds to be asymmetric so that daughter cells can bedifferent sizes and adopt different fates, which is importantfor organism development and tissue homeostasis inmulticellular organisms [Oliferenko et al., 2009]. For singlecell organisms, one obvious example is budding yeast,which uses asymmetric cytokinesis to sequester damagedproteins in aging mother cells [Aguilaniu et al., 2003].

In multicellular organisms, cells mostly undergo cytoki-nesis in a three-dimensional tissue environment. Althoughcytokinesis has been studied for decades, not much isknown about how signals from outside of the cell com-municate with intracellular events. The contractile ringlies right beneath the plasma membrane, a key module incleavage furrow positioning and ingression. Although it isknown that extracellular matrix proteins are required forcytokinesis [Xu and Vogel, 2011], how or if cells respondto extracellular signals is not well understood. How or ifthe cell membrane passes along signals from the outsideto the inside of the cell is also unclear. One class ofobvious candidates for such signal transduction events arethe G protein-coupled receptors (GPCRs), the most abun-dant integral membrane protein superfamily in mamma-lian cells. We show that several GPCRs appear to play arole during cytokinesis, suggesting that external cues doplay a role in this important process.

Additional Supporting Information may be found in the onlineversion of this article.

*Address correspondence to: Ulrike S. Eggert, Department ofChemistry and Randall Division of Cell and MolecularBiophysics, King’s College London, SE1 1UL, United Kingdom.E-mail: [email protected]

Published online in Wiley Online Library(wileyonlinelibrary.com).

SHORT REPORTCytoskeleton, Month 2012 000:000–000 (doi: 10.1002/cm.21055)VC 2012 Wiley Periodicals, Inc.

1 n

GPCRs are also known as 7-transmembrane receptorsbecause they share similar cross membrane structures. It isestimated that the human genome has around 1000GPCRs and they are one of the most studied target fami-lies in the pharmaceutical industry [Filmore, 2004; Over-ington et al., 2006; Gilchrist, 2010]. GPCRs can befound in eukaryotes from amoeba and fungi to plants,invertebrates, and vertebrates. On ligand binding at thecell surface, GPCRs undergo conformational changes andsend signals across the cell membrane by interacting withheterotrimeric G proteins. Different subclasses of Ga pro-teins, such as Gas, Gai, Gaq, and Ga12, signal throughdistinct pathways [Neubig and Siderovski, 2002]. G pro-teins transmit signaling cascades in cells through a largenumber of effectors, including adenylyl cyclases, ion chan-nels, calcium, protein kinase C, and Rho GTPases.

In addition to the traditional G protein-dependentsecond messenger signaling cascades triggered by GPCRactivation, GPCRs can also stimulate G protein-independ-ent signaling events such as arrestin recruitment [Defea,2008] and activate a broad set of intracellular signalingmolecules, such as JNK (c-Jun N-terminal kinase), Akt,PI3 kinase, and RhoA [DeWire et al., 2007]. Upon ligandbinding, GPCR kinases phosphorylate GPCRs and recruitb-arrestins, which results in termination or attenuation ofsignaling by blocking G proteins from further interactionwith the receptors [Hupfeld and Olefsky, 2007]. Thus,the b-arrestins are central players for desensitization,sequestration, and intracellular trafficking of GPCRs,which prevents cells from undergoing excessive receptorstimulation. Recent findings also show that b-arrestinshave additional functions, such as interacting with andcontrolling cytoskeletal actin and the F-actin severingprotein cofilin [Min and Defea, 2011; Pontrello et al.,2011], which may mediate some GPCRs’ function inactin regulation.

GPCRs are the major way for cells and organisms tosense the environment and their ligands include neuro-transmitters, nonsteroid hormones, biogenic amines, odor-ants, and light. Some of the best characterized examplesof GPCR-mediated signaling cascades are in neuronalfunction (e.g., dopamine, serotonin, and adrenergic recep-tors) and in sensory responses, for example, to taste, smell,and light. GPCRs are usually categorized by the ligandsthey bind to or by their functions, but both ligands andfunctions remain unknown for many receptors. Many ofthese GPCRs of unknown function are named by the tis-sues in which they were originally identified, by analogyto known GPCRs, or because they are genetically linkedto loci that respond to and influence a certain type of per-ception. For example, olfactory receptors’ classical func-tion appears to be interaction with odorant molecules andinitiation of neuronal responses to smells. Taste receptorsrespond to different tastes and were originally found intaste receptor cells of the tongue and palate epithelia. We

conducted a screen to investigate if GPCRs play a role incytokinesis and report that knockdown of several GPCRsthat were traditionally thought to be involved in sensorysignal transduction also cause cell division defects, suggest-ing that some of these GPCRs may play broader roles.

Results and Discussion

To systematically investigate whether GPCRs play a rolein cytokinesis, we screened a GPCR siRNA library pro-vided by Dharmacon (Lafayette, CO) for the ICCB-Long-wood screening center at Harvard Medical School (Fig.1A). This library includes siRNA SMARTpools (a mixtureof four individual siRNAs) for 516 GPCRs and relatedproteins (423 GPCRs, 19 G proteins, 42 GPCR regula-tors and effectors, 11 other receptors, and 21 ligands). Weused an increased number of binucleated and multi-nucleated cells as a readout for cytokinesis failure [Eggertet al., 2004; Castoreno et al., 2010]. After siRNA trans-fection, HeLa cells were incubated for 3 days before fixa-tion. Whole cells were then visualized with amine reactivetetramethyl rhodamine-N-hydroxysuccinimide (TAMRA-NHS) ester and DAPI (40,6-diamidino-2-phenylindole)was used to visualize DNA. We then imaged the screeningplates by automated high content confocal fluorescencemicroscopy and identified wells with elevated levels ofbinucleated cells by visual inspection. To control forpotential off-target effects due to unintended knockdownof mRNAs corresponding to other genes, we confirmedpositive hits from the screen by testing the individualsiRNAs from the SMARTpools in subsequent cherrypicks. We only followed up potential hits where severalindividual siRNAs targeting different regions on the targetgene resulted in a failed cytokinesis phenotype.

Fig. 1. Systematic screen for cytokinesis inhibition with siRNAstargeting GPCRs and effector proteins. Schematic flow chart illus-trates our systematic GPCR RNAi screen. HeLa cells were platedin 384 well plates before they were transfected with GPCR siRNAsat ICCB-Longwood. Cells were incubated for 72 h before fixingand staining. Images were acquired using automated microscopyand wells containing binucleated cells were identified by visualinspection.

n 2 Zhang et al. CYTOSKELETON

After examining all 516 gene knockdown wells induplicate experiments, we found that RNAi knockdownof 27 genes resulted in an increased number ofbinucleated and multinucleated cells. These include bothGPCRs and interacting proteins, and gene symbols, num-ber of confirmed individual siRNAs, and the siRNAsequences used for follow-up are listed in Supporting In-formation, Table S1. For follow-up experiments, wefocused mostly on GPCRs rather than effector proteinsand selected those genes that resulted in the highest per-centage of binucleated and multinucleated cells per well.Using these criteria, we selected 13 genes (12 GPCRs anda GPCR-interacting protein GIPC, see below), where atleast two out of four individual siRNAs showed obviouslyincreased binucleated and multinucleated cells. Quantifica-tion revealed that knockdown of these GPCRs increasedbinucleated and multinucleated cells by 4–10-fold relativeto the control (Fig. 2B). We subsequently confirmed thatall GPCRs are expressed in HeLa cells and used live imag-ing to evaluate their effects on cytokinesis.

The 12 GPCRs we selected for follow-up range fromtaste receptors, opsins, olfactory receptors, to dopaminereceptors. They include: BLR1/CXCR5 (Burkitt lym-phoma receptor 1/C-X-C chemokine receptor type 5),DRD2 (dopamine receptor D2), DRD3 (dopamine recep-tor D3), MRGPRX2 (MAS-related GPR, member X2),OPN1MW (opsin 1, medium-wave-sensitive), OPN1LW(opsin 1, long-wave-sensitive), OR1A2 (olfactory receptor,family 1, subfamily A, member 2), OR2A4 (olfactory re-ceptor, family 2, subfamily A, member 4), RXFP3(Relaxin/insulin-like family peptide receptor 3), SSTR5(somatostatin receptor 5), TAS1R2 (taste receptor, type 1,member 2), and TAS2R13 (taste receptor, type 2, member13) (Fig. 2A; Supporting Information, Table S1). Otherthan our recent report showing that DRD3 is involved inendocytic sorting and cytokinesis [Zhang et al., 2012], toour knowledge, none of these receptors have previouslybeen connected to cytokinesis, although the Mitocheckproject reported a lethal phenotype for MRGPRX2knockdown (www.mitocheck.org) [Neumann et al., 2010].

Fig. 2. Knockdown of several GPCRs causes binucleated and multinucleated cell formation. (A) Control or GPCR RNAi cellswere stained with TAMRA-NHS (shown in red) to visualize whole HeLa cells, and DAPI (shown in green) for DNA. Images were takenusing an OPERA high content confocal microscope. Representative images for RNAi of SSTR5, OPN1MW, and OR2A4 are shown. (B)Quantification of binucleated and multinucleated cells in RNAi knockdown cells of 12 selected GPCRs. Quantification was done bycounting 100 HeLa cells each in two duplicates from two independent experiments. Error bars show standard deviations.

CYTOSKELETON GPCRs in Cytokinesis 3 n

Surprisingly, many of these GPCRs belong to subfamiliesthat function in sensation. Some receptors have well-estab-lished and characterized roles in transmitting sensory sig-nals, for example, taste receptor TAS1R2 is highlyexpressed in specialized cells on the tongue and is respon-sible for sensing sweet tastes [Hoon et al., 1999; Matsu-nami et al., 2000; Zhao et al., 2003]. The opsinsOPN1LW and OPN1MW are important for color visionand are highly expressed in optical cones and mutationsin these receptors are associated with eye diseases [Gardneret al., 2012]. Interestingly, the third opsin, OPN1SW, re-sponsible for processing short-wave light, did not score inour screen. Our data suggest that these GPCRs have addi-tional functions independent of their sensory roles. OtherGPCRs on our list, for example, the putative odorant re-ceptor OR2A4, are uncharacterized and appear to havebeen named mainly based on similarity. It is very possiblethat they have unknown cellular functions other than theirpredicted role in sensation.

Most of what is known in the literature about theGPCRs we identified here is focused on their roles in spe-cialized tissues such as sensory organs or the brain. How-ever, we discovered that they are involved in cytokinesis inHeLa cells, a human cervical cancer cell line. To test thegenerality of our results, we investigated if knocking downthese 12 GPCRs in other cancer cell lines also causedcytokinesis failure (Supporting Information, Table S3). Asexpected, knowing that different tissues express differentGPCRs, we found that there was partial overlap betweencell lines. Knockdown of 8/12 GPCRs in a colon cancercell line (HCT116) and 5/12 in a prostate cancer cell line(DU145) resulted in cytokinesis failure, whereas knock-down 3/12 GPCRs caused cytokinesis failure in all threecell lines. The plasticity of GPCR expression and functionamong different cell lines might be an interesting consid-eration in the development of specific cancer therapeutics.

We next tested the GPCRs’ expression in HeLa cells.Because it can be difficult to obtain sensitive ligands and/orspecific antibodies to determine protein levels, we used real-time quantitative PCR (RT-qPCR, or qPCR) to examinethe expression as well as the knockdown efficiency of theGPCR hits from the screen (Supporting Information, Fig.S1A). Primer/probe sets were validated as described in theMaterials and Methods and their sequences are listed inSupporting Information, Table S2. qPCR confirmed expres-sion and knockdown of all 12 GPCRs in HeLa cells. Torule out the possibility that the effects we observed were dueto a general perturbation of GPCR signaling, we also usedqPCR to confirm the expression in HeLa and knockdownsof several GPCRs that did not cause cytokinesis failure afterRNAi (Supporting Information, Fig. S1B).

As expected for proteins of low abundance, the GPCRs’threshold cycle (Ct) values, which are the numbers ofPCR cycles required for the sample fluorescence to reachthe threshold level in qPCR, were relatively high (� 25–

37), but consistent with � 40 cycles of qPCR usually rec-ommended for GPCRs [Regard et al. 2008; Brattelid andLevy, 2011]. This indicates that their expression levels are100–1000-fold lower than abundant proteins such astubulin (Ct value � 18). The different Ct values amongthese GPCRs also indicate their differential expression.For example, the olfactory receptor OR2A4 (Ct value �26) seems to express at higher level than dopamine recep-tor DRD2 (Ct value � 37). We also confirmed knock-down efficiency of the different GPCRs by qPCR. All theGPCRs showed knockdown, but were only partiallydepleted. This could be due to high protein stability, andwas amplified by using an iterative method such as qPCRto measure changes in a population of mRNA correspond-ing to proteins that are of low abundance even in controlcells. For example, OR2A4, for which we could obtain afunctional antibody, appears to be quite efficientlyknocked down at the protein level (Fig. 4A).

While the formation of binucleated cells is a generalreadout for cytokinesis failure, live imaging is needed todetermine which step of cytokinesis is affected, for exam-ple, cleavage furrow positioning, furrow ingression, or ab-scission. We analyzed cell division in control or the 12GPCR knockdown cells by live imaging for 48 h, whichallowed each cell to complete at least one full cell cycle.We found that knockdown of the GPCRs (except forOR2A4, see below) caused cytokinesis failure at a very latestage (Fig. 3 and Supporting Information, Movies 1–4).Control cells usually divide quickly after furrow ingressionand midbody formation (Fig. 3 and Supporting Informa-tion, Movie 1). GPCR RNAi knockdown cells (expect forOR2A4) were able to go through normal mitosis, furrowingression, and midbody formation until the final cuttingstep. The two daughter cells then remained connected bymidbody bridges for extended periods of time, rangingfrom 1–20 h, before their cleavage furrows regressed toform binucleated cells (Fig. 3, Supporting Information,Movies 2 and 4). The times cells spent at the abscissionstage before binucleated cell formation were not correlatedwith the GPCR that was knocked down and are likely dueto cell by cell variations in RNAi efficiency and the stageof the cell cycle when knockdown was first achieved. Tobetter understand how these GPCRs might participate incytokinesis, we used immunofluorescence to assess thelocalization of several key players of cytokinesis in controlor GPCR RNAi-treated cells. Except for OR2A4 (seebelow), the actin cytoskeleton was not affected. Microtu-bule structures, including the mitotic spindle and the mid-zone, were also not affected, and neither were microtubule-associated proteins with functions in cytokinesis, such asthe mitotic kinesin CENPE (centromere-protein E) andMKLP1 (mitotic kinesin-like protein 1) [Raich et al.,1998; Zhu et al., 2005; Liu et al., 2006]. Ray Rappaport[1996] showed that microtubule structures are essentialduring early cytokinesis, but they are less important during


the final steps of cytokinesis, so a lack of perturbation isconsistent with our observation that most GPCR RNAicells fail cytokinesis at the abscission stage. GPCRs aremembrane proteins and enter cells through the endocyticpathway. Endocytic vesicles and proteins are involved inthe abscission process [Strickland and Burgess, 2004; Carl-ton and Martin-Serrano, 2007; Morita et al., 2007; Leeet al., 2008; Horgan and McCaffrey, 2012]. It is possiblethat the GPCRs we identified here are transported to themidbody via vesicle trafficking, where they may play a rolein regulating the final steps of cytokinesis.

Knockdown of OR2A4 caused a minority of cells(�20–30%) to fail to form midbody bridges, but the ma-jority of remaining cells failed during early cytokinesis

(Fig. 3 and Supporting Information, Movie 3). OR2A4has been predicted to belong to the olfactory receptorfamily, but it is completely uncharacterized (0 PubMedentries in July 2012). Western blot and immunofluores-cence with an OR2A4 antibody confirmed the expressionand knockdown of OR2A4 (Fig. 4A and 4B). Cell cycleanalysis of OR2A4 RNAi cells using flow cytometry showsan increase in G2/M (4n) cells as well as a peak of poly-ploid/multinucleated cells (8n or more), an indication ofcytokinesis failure (Fig. S2). The increased G2/M peak inflow cytometry is likely due to an increase in binucleatedcells, which have the same cellular DNA content as andare therefore indistinguishable from G2 phase or normalmitotic cells (Supporting Information, Fig. S2).

Fig. 3. GPCR knockdowns cause cytokinesis failure at early (OR2A4) and late stages (other GPCRs). (A) Schematic illustrationof cytokinesis failure after RNAi depletion of the 12 hit GPCRs. 11/12 GPCR knockdowns show cytokinesis failure at the abscissionstage, where midbody bridges were formed between two daughter cells before they bounced back and became binucleated. ForOR2A4 knockdown cells, a majority of cells fail cytokinesis at an earlier stage, where cleavage furrows formed but did not progressto form midbodies. (B) Images from live imaging of OR2A4, SSTR5, OPN1MW RNAi. Dotted lines outline cell boundaries, twodifferent colors are used (yellow and red) to illustrate individual cells within the same field. Live imaging was done every 20 minfor 48 h, starting 24 h after time of transfection. The complete movies are available in Supporting Information. Scale bar: 10 lm.


Immunofluorescence of OR2A4 shows spindle pole, mid-zone, and midbody ring localizations (Fig. 4B), furthersupporting a role for this GPCR in cytokinesis. Althoughit may seem counterintuitive that an integral membraneprotein would localize to cytokinetic structures, this is notuncommon because membrane trafficking plays a key roleduring cell division [Montagnac et al., 2008]. We investi-gated if OR2A4 associated with specific vesicle types andfound that it colocalizes at the spindle poles and the mid-body with the vesicle-associated membrane protein 3(VAMP3) and transferrin receptor, markers of recycling

endosomes and endosome-derived vesicles (Supporting In-formation, Fig. S3), suggesting that these vesicles may beinvolved in signal transduction in these compartments.

We examined F-actin by phalloidin staining and foundthat the actin cytoskeleton is perturbed in OR2A4 knock-down cells. Compared to control cells and other GPCRknockdown cells at the same stage, these cells havereduced polar blebs, which are dynamic actin-rich cellprotrusions on membranes, in early cytokinesis (Fig. 4Cand 4D). Although actin-enriched membrane blebs weretraditionally considered as a characteristic of apoptosis,

Fig. 4. OR2A4 localizes to cytokinetic structures and affects the actin cytoskeleton. (A) Western blots with OR2A4 antibodyshow the knockdown of OR2A4. Tubulin and actin were used as loading controls. (B) OR2A4 localizes to the spindle poles (redarrows) and chromosomes (green arrow) during mitosis and to the cleavage furrow (yellow arrow) and midbody ring/Flemming body(blue arrow in cell adjacent to metaphase cell) in cytokinesis. OR2A4 immunofluorescence in control and OR2A4 knockdown HeLaat different cell cycle stages are shown. (C) OR2A4 RNAi inhibits polar blebbing in cytokinesis cells. TRITC-phalloidin staining tovisualize F-actin in representative control, OR2A4 or OPN1MW RNAi telophase/early cytokinesis cells is shown. Inserts magnifymarked regions. (D) Quantification of percentage of cells in cytokinesis with polar blebbing in control or OR2A4 RNAi cells. Fiftycells in each condition were counted in two independent experiments. (E) OR2A4 RNAi affects actin structures in interphase cells.TRITC-phalloidin staining to visualize actin in control, OR2A4, or OPN1MW RNAi interphase cells is shown, representative imageswere taken at the same exposure times. Note the increase in the brightness of actin structures in OR2A4 RNAi cells.


they have been more recently observed under other cir-cumstances, for example, during cell division and migra-tion [Charras and Paluch, 2008]. Blebs are produced byactomyosin contractions at the cell membrane cortex andare thought to be involved in a spindle-independentmechanism that cells use to regulate cleavage furrow posi-tioning. Disturbing the actin polymer by actin stabilizingreagents prevents cells from forming polar blebs duringearly cytokinesis [Sedzinski et al., 2011]. Increased bright-ness of F-actin staining and the appearance of more highlybundled actin filaments in OR2A4 knockdown interphasecells indicate that actin polymers may have been stabilized(Fig. 4D and 4E), while actin protein expression was notaffected (Fig. 4A). As other GPCRs have been shown toplay a role in actin organization [Davies et al. 2006; Cot-ton and Claing, 2009], we propose that OR2A4 isinvolved in cytokinesis by exerting a regulatory role onthe actin cytoskeleton. Given that OR2A4 is a GPCR, itis likely to be upstream in any signaling cascades it mightcontrol. Therefore, actin regulation is probably just one of

the pathways that OR2A4 affects during cytokinesis.Based on its localization at the spindle pole, midzone andmidbody, it is likely that it participates in additional path-ways during cytokinesis.

We show here that some GPCRs play a role in cytoki-nesis, but how they send signals into cells is an open ques-tion. It is highly likely that there are other proteins, orindeed entire signaling cascades, that regulate crosstalkbetween these GPCRs and their final effectors. We showedrecently that small molecule agonists/antagonists of dopa-mine receptor D3 do not induce cytokinesis failure whileRNAi depletion of this GPCR does, suggesting that someof the signaling may be nontraditional [Zhang et al.,2012]. Some GPCR effector proteins were included inour RNAi library, including one strong hit, GIPC1(GIPC PDZ domain containing family, member 1).GIPC1, also called synectin, is an adaptor and scaffoldingprotein that interacts with many GPCRs [Katoh, 2002;Hu et al., 2003; Jeanneteau et al., 2004b]. Western blotand qPCR show that it is expressed in HeLa cells and

Fig. 5. GIPC1 RNAi knockdown causes cytokinesis inhibition. (A) Western blots with GIPC1 antibody show knockdownefficiency of GIPC1 RNAi. Tubulin was used as loading control. (B) RT-qPCR shows knockdown efficiency of GIPC1 RNAi. 28SRNA was used as internal control. (C) GIPC1 RNAi induces binucleated cell formation. HeLa cells were stained with TAMRA-NHS(shown in red) to visualize whole cells, and DAPI (shown in green) for DNA. Images from the screen were taken on an OPERAhigh content confocal microscope. Representative images for RNAi of GIPC1 are shown. (A–C) are in HeLa cells. (D) Quantificationof binucleated and multinucleated cells in control versus GIPC1 RNAi-treated HeLa, DU145, and HCT116 cells. Quantificationwas done by counting 100 cells each from two duplicates from two independent experiments. Error bars show standard deviations.


that knockdown is efficient (Fig. 5A–5B). The low Ctvalue of GIPC (Ct value � 21) in our qPCR analysisindicates that it expresses at higher levels than the testedGPCRs. GIPC1 knockdown significantly increasesbinucleated and multinucleated cell formation from 4 to64% (Fig. 5D). GIPC1 knockdown also caused penetrantcytokinesis failure in the other cell lines we tested (colonand prostate cancer, Fig. 5D). Live cell imaging in HeLashows that GIPC RNAi caused late cytokinesis failure, asdid all the GPCRs we identified except for OR2A4.GIPC may act as a mediator between GPCRs and theirdownstream players in cytokinesis regulation. This is con-sistent with the observation that GIPC interacts withDRD2 and DRD3, but not DRD4 [Jeanneteau et al.,2004a]. We show that knockdown of DRD2 and DRD3,but not DRD4, results in cytokinesis defects. GIPC1 notonly interacts with different GPCRs, but also with numer-ous other proteins, such as MyoGEF [Wu et al., 2010],suggesting that it may serves as a bridge between GPCRsand their effectors during cytokinesis.

This study opens a new door for cytokinesis regulation.Not only do we add to the list of proteins that play a roleduring cytokinesis, but we also show that GPCR signalingis involved, suggesting a role for communication betweenthe exterior and interior of cells. Examining GPCRs andtheir interactions with extracellular stimuli and intracellu-lar proteins will help us understand how cells divide in athree-dimensional context, where external cues are likelyto be important. Our results raise many questions abouthow GPCRs might participate in cytokinesis that will beimportant to address in the future. For example, what arethe downstream components that mediate the cytokinesiseffects of these GPCRs? Do they involve different G pro-teins, b-arrestin, or Rho family GTPases? Gai proteinslocalize to centrosomes, the spindle midzone, and mid-bodies and are involved in cell division [Cho and Kehrl,2007]. A role for some heterotrimeric G proteins in celldivision has also been proposed in some model organisms[Schaefer et al. 2001; Bringmann, 2008]. It is possiblethat the GPCRs we found exert their effects through oneof these G proteins. Do GPCRs sit on the membrane andsend signals into the cells to regulate cytokinesis or dothey need to be endocytosed inside the cells to participate?GPCRs are internalized through endosomes, which canlocalize to the mitotic apparatus including spindle polesand midbodies [Emery et al., 2005; Fielding et al., 2005;Montagnac et al., 2008; Morita et al., 2010; Neto et al.,2011]. OR2A4 localizes to spindle poles and midbodies,indicating that it may function at these locations, insteadof at the distant cytoplasmic membrane.

Although it is estimated that up to 40% of the drugsused in the clinic act via GPCRs [Filmore, 2004] many ofthe older drugs were not originally designed as GPCRligands, but were developed based on functional activityobserved. GPCR signaling is far more diverse than origi-

nally thought. One GPCR can couple to multiple G pro-teins, as well as signal through other adaptor proteinsindependently of G protein coupling [Seasholtz et al.,1999; Marinissen and Gutkind, 2001; Defea, 2008]. Abetter understanding of the physiological roles of GPCRswill be essential in the design of next generation therapeu-tics, not only in aiming to selectively modulate GPCRsfor drug development, but also in understanding the con-sequences of GPCR modulation on different biologicalsystems.

Acknowledgment

We thank the Nikon Imaging Center at Harvard MedicalSchool for assistance with microscopy and the Flow Cytom-etry facility at Dana-Farber Cancer Institute for assistancewith flow cytometry. We also thank ICCB-Longwood, thescreening center at Harvard Medical School for the RNAiscreen and the Drosophila RNAi Screening Center at Har-vard Medical School for the use of the OPERA microscope.This project was funded by NIH grant R01 GM082834 (toUE) and the Dana-Farber Cancer Institute.

References

Aguilaniu H, Gustafsson L, Rigoulet M, Nystrom T. 2003. Asym-metric inheritance of oxidatively damaged proteins during cytokine-sis. Science 299:1751–1753.

Atilla-Gokcumen GE, Castoreno AB, Sasse S, Eggert US. 2010.Making the cut: the chemical biology of cytokinesis. ACS ChemBiol 5:79–90.

Atilla-Gokcumen GE, Bedigian AV, Sasse S, Eggert US. 2011. Inhi-bition of glycosphingolipid biosynthesis induces cytokinesis failure.J Am Chem Soc 133:10010–10013.

Brattelid T, Levy FO. 2011. Quantification of GPCR mRNA usingreal-time RT-PCR. Methods Mol Biol 746:165–193.

Bringmann H. 2008. Mechanical and genetic separation of aster-and midzone-positioned cytokinesis. Biochem Soc Trans 36:381–383.

Carlton JG, Martin-Serrano J. 2007. Parallels between cytokinesisand retroviral budding: a role for the ESCRT machinery. Science316:1908–1912.

Castoreno AB, Smurnyy Y, Torres AD, Vokes MS, Jones TR, Car-penter AE, Eggert US. 2010. Small molecules discovered in a path-way screen target the Rho pathway in cytokinesis. Nat Chem Biol6:457–463.

Charras G, Paluch E. 2008. Blebs lead the way: how to migratewithout lamellipodia. Nat Rev Mol Cell Biol 9:730–736.

Cho H, Kehrl JH. 2007. Localization of Gi alpha proteins in thecentrosomes and at the midbody: implication for their role in celldivision. J Cell Biol 178:245–255.

Cotton M, Claing A. 2009. G protein-coupled receptors stimula-tion and the control of cell migration. Cell Signal 21:1045–1053.

Davies SL, Gibbons CE, Vizard T, Ward DT. 2006. Ca2þ-sensingreceptor induces Rho kinase-mediated actin stress fiber assemblyand altered cell morphology, but not in response to aromatic aminoacids. Am J Physiol Cell Physiol 290:C1543–C1551.

Defea K. 2008. Beta-arrestins and heterotrimeric G-proteins: collab-orators and competitors in signal transduction. Br J Pharmacol153(Suppl 1):S298–S309.


DeWire SM, Ahn S, Lefkowitz RJ, Shenoy SK. 2007. Beta-arrestinsand cell signaling. Annu Rev Physiol 69:483–510.

Eggert US, Kiger AA, Richter C, Perlman ZE, Perrimon N, MitchisonTJ, Field CM. 2004. Parallel chemical genetic and genome-wide RNAiscreens identify cytokinesis inhibitors and targets. PLoS Biol 2:e379.

Eggert US, Mitchison TJ, Field CM. 2006. Animal cytokinesis:from parts list to mechanisms. Annu Rev Biochem 75:543–566.

Emery G, Hutterer A, Berdnik D, Mayer B, Wirtz-Peitz F, GaitanMG, Knoblich JA. 2005. Asymmetric Rab 11 endosomes regulatedelta recycling and specify cell fate in the Drosophila nervous sys-tem. Cell 122:763–773.

Fielding AB, Schonteich E, Matheson J, Wilson G, Yu X, HicksonGR, Srivastava S, Baldwin SA, Prekeris R, Gould GW. 2005.Rab11-FIP3 and FIP4 interact with Arf6 and the exocyst to controlmembrane traffic in cytokinesis. Embo J 24:3389–3399.

Filmore D. 2004. It’s a GPCR world. Modern Drug Discovery(American Chemical Society):7:24–28.

Gardner JC, Webb TR, Kanuga N, Robson AG, Holder GE, Stockman A,Ripamonti C, Ebenezer ND, Ogun O, Devery S, et al. 2012. A novel mis-sense mutation in both OPN1LWand OPN1MW cone opsin genes causesX-linked cone dystrophy (XLCOD5). Adv Exp Med Biol 723:595–601.

Gilchrist A. 2010. GPCR Molecular Pharmacology and Drug Tar-geting: Shifting Paradigms and New Directions. Wiley. ISBN 978-0-470-30778-6.

Hoon MA, Adler E, Lindemeier J, Battey JF, Ryba NJ, Zuker CS.1999. Putative mammalian taste receptors: a class of taste-specificGPCRs with distinct topographic selectivity. Cell 96:541–551.

Horgan CP, McCaffrey MW. 2012. Endosomal trafficking in ani-mal cytokinesis. Front Biosci (Schol Ed) 4:547–555.

Hu LA, Chen W, Martin NP, Whalen EJ, Premont RT, LefkowitzRJ. 2003. GIPC interacts with the beta1-adrenergic receptor andregulates beta1-adrenergic receptor-mediated ERK activation. J BiolChem 278:26295–26301.

Hupfeld CJ, Olefsky JM. 2007. Regulation of receptor tyrosine kinasesignaling by GRKs and beta-arrestins. Annu Rev Physiol 69:561–577.

Jeanneteau F, Diaz J, Sokoloff P, Griffon N. 2004a. Interactions of GIPCwith dopamine D2, D3 but not D4 receptors define a novel mode of reg-ulation of G protein-coupled receptors. Mol Biol Cell 15:696–705.

Jeanneteau F, Guillin O, Diaz J, Griffon N, Sokoloff P. 2004b.GIPC recruits GAIP (RGS19) to attenuate dopamine D2 receptorsignaling. Mol Biol Cell 15:4926–4937.

Katoh M. 2002. GIPC gene family (Review). Int J Mol Med 9:585–589.

Lee HH, Elia N, Ghirlando R, Lippincott-Schwartz J, Hurley JH.2008. Midbody targeting of the ESCRT machinery by a noncanoni-cal coiled coil in CEP55. Science 322:576–580.

Liu D, Zhang N, Du J, Cai X, Zhu M, Jin C, Dou Z, Feng C, Yang Y,Liu L, et al. 2006. Interaction of Skp1 with CENP-E at the midbody isessential for cytokinesis. Biochem Biophys Res Commun 345:394–402.

Marinissen MJ, Gutkind JS. 2001. G-protein-coupled receptors andsignaling networks: emerging paradigms. Trends Pharmacol Sci 22:368–376.

Matsunami H, Montmayeur JP, Buck LB. 2000. A family of candi-date taste receptors in human and mouse. Nature 404:601–604.

Min J, Defea K. 2011. Beta-arrestin-dependent actin reorganization:bringing the right players together at the leading edge. Mol Phar-macol 80:760–768.

Montagnac G, Echard A, Chavrier P. 2008. Endocytic traffic in ani-mal cell cytokinesis. Curr Opin Cell Biol 20:454–461.

Morita E, Sandrin V, Chung HY, Morham SG, Gygi SP, RodeschCK, Sundquist WI. 2007. Human ESCRT and ALIX proteins

interact with proteins of the midbody and function in cytokinesis.EMBO J 26:4215–4227.

Morita E, Colf LA, Karren MA, Sandrin V, Rodesch CK, SundquistWI. 2010. Human ESCRT-III and VPS4 proteins are required forcentrosome and spindle maintenance. Proc Natl Acad Sci USA 107:12889–12894.

Neto H, Collins LL, Gould GW. 2011. Vesicle trafficking andmembrane remodelling in cytokinesis. Biochem J 437:13–24.

Neubig RR, Siderovski DP. 2002. Regulators of G-protein signal-ling as new central nervous system drug targets. Nat Rev Drug Dis-cov 1:187–197.

Neumann B, Walter T, Heriche JK, Bulkescher J, Erfle H, ConradC, Rogers P, Poser I, Held M, Liebel U, et al. 2010. Phenotypicprofiling of the human genome by time-lapse microscopy revealscell division genes. Nature 464:721–727.

Normand G, King RW. 2010. Understanding cytokinesis failure.Adv Exp Med Biol 676:27–55.

Oliferenko S, Chew TG, Balasubramanian MK. 2009. Positioningcytokinesis. Genes Dev 23:660–674.

Overington JP, Al-Lazikani B, Hopkins AL. 2006. How many drugtargets are there? Nat Rev Drug Discov 5:993–996.

Pontrello CG, Sun MY, Lin A, Fiacco TA, DeFea KA, Ethell IM.2011. Cofilin under control of beta-arrestin-2 in NMDA-dependentdendritic spine plasticity, long-term depression (LTD), and learning.Proc Natl Acad Sci USA 109:E442–E451.

Raich WB, Moran AN, Rothman JH, Hardin J. 1998. Cytokinesisand midzone microtubule organization in Caenorhabditis elegansrequire the kinesin-like protein ZEN-4. Mol Biol Cell 9:2037–2049.

Rappaport R. 1986. Establishment of the mechanism of cytokinesisin animal cells. Int Rev Cytol 105:245–281.

Rappaport R. 1996. Cytokinesis in Animal Cells. In: Developmentaland Cell Biology Series. Cambridge: Cambridge University Press.

Regard JB, Sato IT, Coughlin SR. 2008. Anatomical profiling of Gprotein-coupled receptor expression. Cell 135:561–571.

Schaefer M, Petronczki M, Dorner D, Forte M, Knoblich JA.2001. Heterotrimeric G proteins direct two modes of asymmetriccell division in the Drosophila nervous system. Cell 107:183–94.

Seasholtz TM, Majumdar M, Brown JH. 1999. Rho as a mediator ofG protein-coupled receptor signaling. Mol Pharmacol 55:949–956.

Sedzinski J, Biro M, Oswald A, Tinevez JY, Salbreux G, Paluch E.2011. Polar actomyosin contractility destabilizes the position of thecytokinetic furrow. Nature 476:462–466.

Strickland LI, Burgess DR. 2004. Pathways for membrane traffick-ing during cytokinesis. Trends Cell Biol 14:115–118.

Wu D, Haruta A, Wei Q. 2010. GIPC1 interacts with MyoGEFand promotes MDA-MB-231 breast cancer cell invasion. J BiolChem 285:28643–28650.

Xu X, Vogel BE. 2011. A secreted protein promotes cleavage furrowmaturation during cytokinesis. Curr Biol 21:114–119.

Zhang X, Wang. W, Bedigian AV, Coughlin ML, Mitchison TJ,Eggert US. 2012. Dopamine receptor D3 regulates endocytic sort-ing by a Prazosin-sensitive interaction with COPI. Proc Natl AcadSci USA 109:12485–12490.

Zhao GQ, Zhang Y, Hoon MA, Chandrashekar J, Erlenbach I,Ryba NJ, Zuker CS. 2003. The receptors for mammalian sweet andumami taste. Cell 115:255–266.

Zhu C, Zhao J, Bibikova M, Leverson JD, Bossy-Wetzel E, Fan JB,Abraham RT, Jiang W. 2005. Functional analysis of human micro-tubule-based motor proteins, the kinesins and dyneins, in mitosis/cytokinesis using RNA interference. Mol Biol Cell 16:3187–3199.


g protein-coupled receptors participate in cytokinesis

Documents