The kinase LATS/WARTS is a tumor suppressor protein conserved in evolution, but its function at the molecular level is not well understood. We report here that human LATS1 interacts with MOB1A, a protein whose homologue in budding yeast associates with kinases involved in mitotic exit. This suggested that LATS1 may be a component of the previously uncharacterized mitotic exit network in higher eukaryotes. Indeed, moderate overexpression of human LATS1 in cells exposed to microtubule poisons facilitated mitotic exit, and this activity required MOB1A. Reciprocally, small interfering RNA–mediated suppression of LATS1 or MOB1A prolonged telophase, but had no effect on the length of the earlier phases of mitosis. A role of LATS1 in mitotic exit may explain its previously described abilities to induce G2 arrest and promote cytokinesis.

LATS (large tumor suppressor, also known as WARTS) was originally identified in a screen for genes whose inactivation leads to overproliferation of cells in Drosophila (1, 2). Two homologues of LATS have been identified in mammals and are called LATS1/WARTS and LATS2/KPM (36). Inactivation of Lats1 in mice leads to development of various tumors, including sarcomas and ovarian cancer (4). Further, promoter methylation, loss of heterozygosity, and missense mutations targeting LATS1 have been reported in human sarcomas and ovarian carcinomas (7, 8), suggesting that LATS functions as a tumor suppressor across evolution from Drosophila to humans.

At the molecular level, the function of LATS1 is not well understood. LATS1 has been reported to inhibit the CDC2 kinase by associating with CDC2 and displacing its cyclin subunit (3). More recently, LATS1 has been reported to inhibit the LIMK1 kinase again by directly interacting with LIMK1 (9). Inhibition of CDC2 may explain why LATS1 overexpression inhibits the transition from G2 to mitosis (3, 10, 11), whereas inhibition of LIMK1 may explain the ability of LATS1 to promote cytokinesis because LIMK1 inhibits actin polymerization (9).

Interestingly, LATS1 itself is a protein kinase, but its kinase activity is not required for inhibition of LIMK1 and possibly of CDC2 (3, 9). In turn, this suggests that LATS1 has functions that require its kinase activity and are independent of CDC2 and LIMK1. Based on its amino acid sequence, LATS1 has been assigned to the nuclear Dbf2-related (NDR) family of protein kinases (12). In yeast, NDR kinases are involved in cell cycle control. In Saccharomyces cerevisiae, there are three members in the NDR family. Dbf2 and its nearly identical and redundant Dbf20 kinase are key components of the mitotic exit network (MEN), a signaling pathway that coordinates cyclin-dependent kinase (CDK) inactivation, sister chromatid decondensation, and cytokinesis at the end of mitosis (1315); on the other hand, Cbk1, the third member, promotes polarized growth and activates daughter-specific gene transcription after cytokinesis (1619). In Schizosaccharomyces pombe, there are two members in the NDR family. Sid2, the orthologue of Dbf2/Dbf20, functions in the septation initiation network (SIN), which is analogous to the S. cerevisiae MEN (1315); on the other hand, Orb6, like its orthologue Cbk1, is required for maintenance of cell polarity after completion of mitosis (20).

The yeast NDR kinases require regulatory subunits for catalytic activity. In S. cerevisiae, there are two such subunits, Mob1 and Mob2. Mob1 interacts with Dbf2 and Dbf20, whereas Mob2 interacts with Cbk1. These interactions are conserved in S. pombe, such that S. pombe Mob1 and Mob2 interact with Sid2 and Orb6, respectively (18, 19, 2125).

The assignment of LATS1 to the NDR protein kinase family raises the possibility that LATS1 is an orthologue of yeast NDR kinases (either the Mob1- or the Mob2-interacting kinases). Indeed, we report here that in human cells, LATS1 interacts with MOB1A and has functional similarities to the yeast NDR kinases that interact with Mob1 (S. cerevisiae Dbf2/Dbf20 and S. pombe Sid2). Thus, LATS1 is a component of the previously uncharacterized mammalian MEN.

Expression plasmids and cell transfections. The coding sequences of human NDR1, MOB1A, and MOB2 were amplified from expressed sequence tags by PCR and subcloned into a pCDNAzeo3.1 vector, which was modified to contain either hemagglutinin (HA) or FLAG tags within its polylinker. A pCDNA3.1 plasmid expressing myc-tagged human LATS1 was provided by Dr. Wufan Tao (The Stem Cell Institute, University of Minnesota Medical School, Minneapolis, MN). These plasmids were transfected in U2OS osteosarcoma cells by calcium phosphate precipitation and their protein products were analyzed 72 hours after transfection. For stable expression, the inserts encoding myc-tagged LATS1 and HA-tagged MOB1A were subcloned in the pIRESN2 vector. U2OS cells transfected with these vectors or with a pIRESN2 vector that had no insert (neo control) were selected with 0.5 mg/mL G418 for 4 weeks. G418-resistant colonies were then pooled by trypsinization and propagated in tissue culture media containing 0.25 mg/mL G418.

Small interfering RNA transfection. U2OS cells (150,000) were seeded in 60 mm tissue culture dishes and the following day were transfected with small interfering RNA (siRNA) specific for human MOB1A (AAUGCAGAAGCAACUCUAGGA) or LATS1 (9) or luciferase (Dharmacon, Lafayette, CO) as a control. For analysis of cell cycle progression, the cells were synchronized 1 day later with a single thymidine block as described below.

Cell cycle synchronization. The cells were synchronized by adding 1.3 mmol/L thymidine in the tissue culture media. After 20 hours, the cells were washed twice with PBS and released into media containing 250 μmol/L thymidine and 250 μmol/L deoxycytidine. Where indicated, 8 hours after release into S phase, the cells were exposed to 1 μmol/L nocodazole.

Live cell videomicroscopy. Cells that had been cultured on 12 mm autoclaved glass slides were synchronized and placed on the microscope stage 8 hours after release from the G1-S block. The cells were observed using a phase contrast lens and maintained at 37°C using a stage heater (Bioptechs, Butler, PA). Images were acquired every 5 minutes using an ORCA ER digital camera (Hamamatsu, Hamamatsu City, Japan).

Antibodies. A monoclonal antibody against human MOB1A was prepared using as antigen full-length MOB1A fused to the COOH terminus of glutathione S-transferase. The other antibodies were obtained commercially: γ-tubulin polyclonal (Sigma, St. Louis, MO), LATS1 polyclonal (Santa Cruz Biotechnologies, Santa Cruz, CA), HA monoclonal (Covance, Princeton, NJ), FLAG monoclonal (Sigma), 9E10-Myc monoclonal (Upstate, Charlottesville, VA), and histone H3 Ser10 phosphospecific polyclonal (Upstate).

Flow cytometry. For flow cytometry analysis, 1 × 106 million cells were trypsinized, washed once in PBS, and fixed in 70% ethanol at 4°C for at least 15 minutes. The cells were then washed in PBS, incubated in PBS supplemented with 0.25% Triton X-100 on ice for 5 minutes and then in 200 μL 1% bovine serum albumin (BSA)/PBS containing 1 μg anti–phospho-Ser10 histone H3 antibody at room temperature for 1 hour. After washing, the cells were incubated in 200 μL 1% BSA/PBS containing 1 μg Alexafluor 488–conjugated goat anti-rabbit antibody (Molecular Probes, Eugene, OR) for 30 minutes at room temperature. The cells were then stained with propidium iodide for 20 minutes at 37°C in the presence of DNase-free RNase and analyzed on a Becton Dickinson fluorescence-activated cell sorting with CellQuest software (Becton Dickinson, Franklin Lakes, NJ).

Immunofluorescence. For immunofluorescence, cells that had been cultured on 12 mm autoclaved glass slides were washed with PBS containing Ca and Mg and fixed in 100% methanol at −20°C for 7 minutes. The cells were subsequently incubated in PBS supplemented with 0.5% Triton X-100 for 15 minutes at 4°C. After two 5-minute washes in PBS, the coverslips were blocked with 2.5% gelatin for 30 minutes at room temperature. Primary antibodies were diluted in 1% BSA/PBS and incubated overnight at 4°C. The next day, the coverslips were washed in 1% BSA/PBS supplemented with 0.1% Tween 20 (PBST) and then in PBS. They were then incubated with secondary antibody conjugated to Alexafluor 488 or 568 at a dilution of 1:500 in 1% BSA/PBS for 30 minutes at room temperature. After washing with PBST and PBS, the cells were stained with 4′,6-diamidino-2-phenylindole (DAPI) and mounted with Fluorimount G (Southern Biotechnology Associates, Birmingham, AL). Images were acquired using a fluorescence microscope (Leica, Wetzlar, Germany) equipped with an ORCA ER digital camera (Hamamatsu) and were processed using IRIX Imagevision Libary Tools (SGI, Mountain View, CA).

Preparation of cell extracts, immunoprecipitation, and immunoblotting. Cells were lysed in buffer consisting of 50 mmol/L Tris (pH 8.0), 120 mmol/L NaCl, 0.5% NP40, 1 mmol/L DTT, 0.4 μg/mL Pefabloc SC, 2 μg/mL pepstatin, 15 mmol/L NaF, 0.1 μmol/L staurosporine, and 1 mmol/L sodium vanadate. The lysates were cleared by centrifugation and used for immunoprecipitation or were resolved by SDS-PAGE and immunoblotted. For immunoprecipitation, 500 μg whole cell extract was incubated with 0.3 μg primary antibody in a total volume of 500 μL immunoprecipitation buffer [50 mmol/L HEPES (pH 7.4), 150 mmol/L NaCl, 0.5 mmol/L EDTA, 0.1 mmol/L MgCl2, 0.5% Tween 20] for 2 hours at 4°C. Then, 40 μL Protein G-Sepharose beads (Pharmacia, Piscataway, NJ) were added and incubated for an additional hour at 4°C. The beads were then washed five times in immunoprecipitation buffer, resuspended in 2× SDS loading buffer, and resolved by SDS-PAGE.

LATS1 interacts with human MOB1A. By analogy to the yeast NDR kinases, we reasoned that LATS1 may interact with the human homologues of the yeast Mob1 or Mob2 proteins. Three such homologues have been identified: MOB1A and MOB1B have 95% amino acid sequence identity to each other and high sequence similarity to yeast Mob1; whereas MOB2 (also known as MMh) has high sequence similarity to yeast Mob2 (26, 27). Interactions between LATS1 and human MOB proteins were assayed by transiently expressing in cells myc-tagged LATS1 together with either HA-tagged MOB1A or HA-tagged MOB2 and performing coimmunoprecipitations. As controls, we also examined the interaction of MOB1A and MOB2 with human FLAG-tagged NDR1 because NDR1 was recently reported to associate with human MOB2 (28). Myc-tagged LATS1 associated preferentially with HA-tagged MOB1A, whereas FLAG-tagged NDR1 associated preferentially with HA-tagged MOB2 (Fig. 1A).

Figure 1.

Specific interaction between human LATS1 and MOB1A proteins. A, interaction between ectopically expressed tagged proteins. U2OS cells were transfected with a vector without insert (−) or vectors expressing human HA-tagged MOB1A (M1ha), HA-tagged MOB2 (M2ha), myc-tagged LATS1 (L1my), or FLAG-tagged NDR1 (N1fl) or combinations thereof as indicated. Cell lysates were prepared 48 hours after transfection. Expression of the tagged proteins was monitored by immunoblotting (IB) with antibodies that recognize the HA (ha), FLAG (fl), and myc (my) tags. Interactions between the various tagged proteins were monitored by immunoprecipitation (IP) followed by immunoblotting using antibodies that recognize the tags. B, expression of myc-tagged LATS1 in two stably transfected clones (U2OS L1my, clones 1 and 2) was monitored by immunoblotting with antibodies that recognize the myc tag or LATS1. Cells transfected with vector without insert (U2OS neo) served as controls. C, coimmunoprecipitation between endogenous MOB1A and ectopically expressed myc-tagged LATS1 in stably transfected U2OS cells (clone L1my 1). An antibody that recognizes the HA tag and U2OS cells expressing neo served as controls. D, interaction between endogenous human MOB1A and LATS1 proteins in parental U2OS cells.

Figure 1.

Specific interaction between human LATS1 and MOB1A proteins. A, interaction between ectopically expressed tagged proteins. U2OS cells were transfected with a vector without insert (−) or vectors expressing human HA-tagged MOB1A (M1ha), HA-tagged MOB2 (M2ha), myc-tagged LATS1 (L1my), or FLAG-tagged NDR1 (N1fl) or combinations thereof as indicated. Cell lysates were prepared 48 hours after transfection. Expression of the tagged proteins was monitored by immunoblotting (IB) with antibodies that recognize the HA (ha), FLAG (fl), and myc (my) tags. Interactions between the various tagged proteins were monitored by immunoprecipitation (IP) followed by immunoblotting using antibodies that recognize the tags. B, expression of myc-tagged LATS1 in two stably transfected clones (U2OS L1my, clones 1 and 2) was monitored by immunoblotting with antibodies that recognize the myc tag or LATS1. Cells transfected with vector without insert (U2OS neo) served as controls. C, coimmunoprecipitation between endogenous MOB1A and ectopically expressed myc-tagged LATS1 in stably transfected U2OS cells (clone L1my 1). An antibody that recognizes the HA tag and U2OS cells expressing neo served as controls. D, interaction between endogenous human MOB1A and LATS1 proteins in parental U2OS cells.

Close modal

The interaction between human LATS1 and MOB1A was further examined by studying endogenous MOB1A. We prepared U2OS cells that stably express myc-tagged LATS1 (clones 1 and 2; Fig. 1B) and used these cells (clone 1) to show that a monoclonal antibody that recognizes human MOB1A, but not a control antibody, coprecipitated myc-tagged LATS1 (Fig. 1C). We could not show the reciprocal interaction, because untagged human endogenous MOB1A comigrates with immunoglobulin light chain (see Fig. 1A showing HA-tagged MOB1A migrating just slightly slower than the immunoglobulin light chain). Finally, we could also detect an interaction between endogenous MOB1A and LATS1 proteins in parental U2OS cells, although the signal was not as strong as with the tagged proteins (Fig. 1D).

Based on the three-dimensional structure of human MOB1A, we had previously proposed that the interaction between Mob proteins and their partner kinases would be mediated by electrostatic interactions between a negative patch on the surface of Mob proteins and a basic region present in all Mob-associated kinases (27). This prediction has been supported by analysis of NDR kinases implicated in mitotic exit in S. pombe and Xenopus. Specifically, residues 101 to 207 of S. pombe Sid2, which encompass the basic region (Fig. 2A and B), are sufficient for interaction with Mob1 (29) and a peptide corresponding to the basic region of the Xenopus NDR kinase interacts with Mob1 (30). Therefore, we anticipated that the basic region of human LATS1 would be important for the interaction with MOB1A. Indeed, LATS1 mutants with Arg682 and Lys683 or with Arg694, Lys696, and Arg697 substituted with glutamic acid failed to interact with HA-tagged MOB1A, whereas, as a control, LATS1 with a substitution in the kinase domain (Lys734 to Met) interacted with MOB1A as efficiently as wild-type LATS1 (Fig. 2C).

Figure 2.

The interaction between human LATS1 and MOB1A is mediated by the basic region of LATS1. A, domain organization of human LATS1 (1,130 amino acids), NDR1 (465 amino acids), S. cerevisiae Dbf2 (572 amino acids), and S. pombe Sid2 (607 amino acids). BR, basic region; KD, kinase domain; CE, COOH-terminal extension of the kinase domain; MBD, Mob1-binding domain of Sid2 (29). B, sequence alignment of the basic region and small part of the kinase domain of kinases that interact with Mob proteins. The sequences are compared with the sequence of CDK2, which does not interact with Mob proteins. Blue, basic residues; magenta, conserved nonbasic residues. The basic residues substituted with glutamic acid (E) in the LATS1 E682 and E694 mutants are marked above the sequence alignment. The numbers indicate the residue range of the aligned sequences. hs, Homo sapiens; sc, S. cerevisiae; sp, S. pombe. C, interaction between ectopically expressed human HA-tagged MOB1A and wild-type (wt) or mutant myc-tagged LATS1. The LATS1 E682 and E694 mutants have two and three amino acid substitutions, respectively, as shown in (B). The M734 mutant substitutes Lys734 with Met and is a kinase-dead mutant. Expression of the tagged proteins was monitored by immunoblotting with antibodies that recognize the HA and myc tags. MOB1A-LATS1 interactions were monitored by immunoprecipitation followed by immunoblotting.

Figure 2.

The interaction between human LATS1 and MOB1A is mediated by the basic region of LATS1. A, domain organization of human LATS1 (1,130 amino acids), NDR1 (465 amino acids), S. cerevisiae Dbf2 (572 amino acids), and S. pombe Sid2 (607 amino acids). BR, basic region; KD, kinase domain; CE, COOH-terminal extension of the kinase domain; MBD, Mob1-binding domain of Sid2 (29). B, sequence alignment of the basic region and small part of the kinase domain of kinases that interact with Mob proteins. The sequences are compared with the sequence of CDK2, which does not interact with Mob proteins. Blue, basic residues; magenta, conserved nonbasic residues. The basic residues substituted with glutamic acid (E) in the LATS1 E682 and E694 mutants are marked above the sequence alignment. The numbers indicate the residue range of the aligned sequences. hs, Homo sapiens; sc, S. cerevisiae; sp, S. pombe. C, interaction between ectopically expressed human HA-tagged MOB1A and wild-type (wt) or mutant myc-tagged LATS1. The LATS1 E682 and E694 mutants have two and three amino acid substitutions, respectively, as shown in (B). The M734 mutant substitutes Lys734 with Met and is a kinase-dead mutant. Expression of the tagged proteins was monitored by immunoblotting with antibodies that recognize the HA and myc tags. MOB1A-LATS1 interactions were monitored by immunoprecipitation followed by immunoblotting.

Close modal

Human LATS1 colocalizes with MOB1A at the centrosomes and midbody. Human LATS1 localizes at the centrosomes in interphase, at the mitotic spindle in prometaphase/metaphase, and at the midbody in telophase (31, 32). If LATS1 and MOB1A interact in cells, as suggested by the coimmunoprecipitation experiments, then MOB1A should localize to the same intracellular structures as LATS1. We used the monoclonal antibody raised against human MOB1A to monitor the intracellular localization of endogenous MOB1A in U2OS and HeLa cells. In U2OS cells, MOB1A localized at the centrosomes in interphase, at the poles of the mitotic spindle in prometaphase/metaphase, and at the midbody in telophase (Fig. 3A). An identical staining pattern was observed with an anti-HA antibody in U2OS cells that stably express HA-tagged MOB1A (data not shown). Interestingly, in many telophase cells, there was asymmetrical staining of MOB1A at the midbody (Fig. 3A), reminiscent of the asymmetrical spindle pole body staining of SIN proteins in S. pombe (33, 34). Myc-tagged LATS1 also exhibited asymmetrical staining at the midbody in some U2OS cells (Fig. 3B). The significance of this asymmetry is unclear and may be a feature of only some cell lines, because in HeLa cells both MOB1A (Fig. 3C) and LATS1 (9) exhibit symmetrical staining at the midbody. Irrespective of the staining asymmetry or lack thereof, the immunofluorescence analysis suggests that LATS1 and MOB1A colocalize in cells, further supporting the conclusion that these two proteins interact.

Figure 3.

Intracellular localization of human MOB1A. A, intracellular localization of endogenous MOB1A in U2OS cells monitored by immunofluorescence with a monoclonal antibody raised against MOB1A. The cells were also stained for γ-tubulin to mark the centrosomes (in interphase), poles of the mitotic spindle (in metaphase), and midbody (in telophase). Merged images of MOB1A and γ-tubulin staining corresponding to the centrosomes, mitotic poles, and midbodies are shown in the insets. The nuclei and chromosomes were stained with DAPI. B, asymmetrical midbody localization of myc-tagged LATS1 in stably transfected U2OS cells (clone L1my 1) revealed by immunofluorescence using antibodies that recognize the myc tag and γ-tubulin. The inset shows merged images of myc and γ-tubulin staining corresponding to the midbody. C, intracellular localization of endogenous MOB1A in HeLa cells. The positions of the midbodies were detected by differential interference contrast optics. Inset, a higher-magnification image of MOB1A staining corresponding to the midbody.

Figure 3.

Intracellular localization of human MOB1A. A, intracellular localization of endogenous MOB1A in U2OS cells monitored by immunofluorescence with a monoclonal antibody raised against MOB1A. The cells were also stained for γ-tubulin to mark the centrosomes (in interphase), poles of the mitotic spindle (in metaphase), and midbody (in telophase). Merged images of MOB1A and γ-tubulin staining corresponding to the centrosomes, mitotic poles, and midbodies are shown in the insets. The nuclei and chromosomes were stained with DAPI. B, asymmetrical midbody localization of myc-tagged LATS1 in stably transfected U2OS cells (clone L1my 1) revealed by immunofluorescence using antibodies that recognize the myc tag and γ-tubulin. The inset shows merged images of myc and γ-tubulin staining corresponding to the midbody. C, intracellular localization of endogenous MOB1A in HeLa cells. The positions of the midbodies were detected by differential interference contrast optics. Inset, a higher-magnification image of MOB1A staining corresponding to the midbody.

Close modal

Moderate overexpression of LATS1 does not delay entry into mitosis, but induces mitotic checkpoint defects. The specific interaction between human LATS1 and MOB1A raises the possibility that LATS1 is the human orthologue of the S. cerevisiae Dbf2/Dbf20 and S. pombe Sid2 kinases. However, these yeast kinases function at the end of mitosis in the MEN (S. cerevisiae) or analogous SIN (S. pombe), whereas human LATS1 was originally shown to inhibit mitotic entry (10, 11). The experiments demonstrating that LATS1 inhibits the transition from G2 into mitosis utilized adenoviral vectors to induce LATS1 expression. However, adenoviral vectors typically drive very high levels of protein expression. To overcome this limitation, we studied two stably transfected U2OS clones, in which the ectopic myc-tagged LATS1 is expressed at 3- to 5-fold higher levels than endogenous LATS1 and a neo-control clone (Fig. 1B). All clones were synchronized at the G1-S boundary and released into the cell cycle (0 hour). Twelve hours later, some cells were in G2-M (4N DNA content) and some cells had already completed cytokinesis (2N DNA content), whereas 4 hours later all cells had completed cytokinesis and were in G1; however, there was no difference among the clones (Fig. 4A). Thus, moderate overexpression of LATS1 did not affect the kinetics of progression into and out of mitosis.

Figure 4.

Ectopic expression of human LATS1 overrides the spindle checkpoint in a MOB1A-dependent manner. A, ectopic expression of myc-tagged human LATS1 does not delay or inhibit entry into mitosis. Stably transfected U2OS clones expressing myc-tagged LATS1 (clones L1my 1 and 2) were released into the cell cycle after synchronization in G1-S (0 hour) and cell division was monitored by analysis of genomic DNA content by flow cytometry. B, stably transfected U2OS clones were released into the cell cycle after a G1-S synchronization block (0 hour) and the ability of cells with a 4N DNA content to maintain mitotic arrest when challenged with nocodazole was monitored by flow cytometry. Top, propidium iodide (PI) staining. Bottom rows, histone H3 phosphorylation at Ser10 (H3 S10p) of the cells with 4N DNA content. The peaks of the H3 S10p–positive and H3 S10p–negative cells are colored gray and black, respectively. C, suppression of endogenous MOB1A protein levels by siRNA in stably transfected U2OS cells expressing myc-tagged LATS1 (clone L1my 1). The expression of myc-tagged LATS1 was not suppressed by the siRNA directed against MOB1A. ctl, control; *, nonspecific band. D, MOB1A is required for ectopically expressed LATS1 to override the spindle checkpoint. Stably transfected U2OS cells (clone L1my 1) were treated with control or MOB1A-directed siRNA, synchronized, and assayed for mitotic arrest after challenge with nocodazole by monitoring histone H3 phosphorylation at Ser10. Top, propidium iodide staining. Bottom rows, histone H3 phosphorylation at Ser10 of the cells with 4N DNA content. The peaks of the H3 S10p–positive and H3 S10p–negative cells are colored gray and black, respectively.

Figure 4.

Ectopic expression of human LATS1 overrides the spindle checkpoint in a MOB1A-dependent manner. A, ectopic expression of myc-tagged human LATS1 does not delay or inhibit entry into mitosis. Stably transfected U2OS clones expressing myc-tagged LATS1 (clones L1my 1 and 2) were released into the cell cycle after synchronization in G1-S (0 hour) and cell division was monitored by analysis of genomic DNA content by flow cytometry. B, stably transfected U2OS clones were released into the cell cycle after a G1-S synchronization block (0 hour) and the ability of cells with a 4N DNA content to maintain mitotic arrest when challenged with nocodazole was monitored by flow cytometry. Top, propidium iodide (PI) staining. Bottom rows, histone H3 phosphorylation at Ser10 (H3 S10p) of the cells with 4N DNA content. The peaks of the H3 S10p–positive and H3 S10p–negative cells are colored gray and black, respectively. C, suppression of endogenous MOB1A protein levels by siRNA in stably transfected U2OS cells expressing myc-tagged LATS1 (clone L1my 1). The expression of myc-tagged LATS1 was not suppressed by the siRNA directed against MOB1A. ctl, control; *, nonspecific band. D, MOB1A is required for ectopically expressed LATS1 to override the spindle checkpoint. Stably transfected U2OS cells (clone L1my 1) were treated with control or MOB1A-directed siRNA, synchronized, and assayed for mitotic arrest after challenge with nocodazole by monitoring histone H3 phosphorylation at Ser10. Top, propidium iodide staining. Bottom rows, histone H3 phosphorylation at Ser10 of the cells with 4N DNA content. The peaks of the H3 S10p–positive and H3 S10p–negative cells are colored gray and black, respectively.

Close modal

We next examined the effect of LATS1 overexpression in cells treated with microtubule poisons. In S. cerevisiae and S. pombe, mitotic arrest in response to microtubule poisons is compromised when genes that suppress the MEN/SIN are inactivated (35, 36). If LATS1 functions in a pathway equivalent to the yeast MEN/SIN, then LATS1 overexpression should lead to mitotic checkpoint defects in human cells. To examine this hypothesis, the neo and myc-tagged LATS1 U2OS clones were synchronized at the G1-S boundary and then allowed to proceed through the cell cycle in the presence of nocodazole. At 24, 36, and 48 hours after release from the cell cycle arrest, all clones had a 4N DNA content (neo clone; Fig. 4B,, top; myc-tagged LATS1 clones, data not shown). The fraction of 4N DNA cells that was in mitosis at each time point was determined by analysis of histone H3 phosphorylation on Ser10. The neo-expressing control clone maintained a high fraction of cells in prometaphase/metaphase at 24 and 36 hours, but by 48 hours most cells had decondensed their chromosomes and stained negative for histone H3 phosphorylated on Ser10 due to adaptation to the spindle checkpoint. The myc-tagged LATS1-expressing clones had a high fraction of cells staining positive for histone H3 phosphorylation on Ser10 at 24 hours, but by 36 hours the majority of cells stained negative (Fig. 4B). Thus, moderate overexpression of LATS1 compromises mitotic arrest in response to microtubule poisons, similar to the phenotype observed in yeast when genes that suppress the MEN/SIN are inactivated.

The function of the yeast MEN/SIN kinases is dependent on their Mob1 subunits. By analogy, the phenotype of LATS1 overexpression should be dependent on MOB1A. To test this prediction, we suppressed by siRNA the levels of endogenous MOB1A in the stably transfected myc-tagged LATS1-expressing cells (Fig. 4C) and examined its effect on mitotic arrest. Indeed, suppression of MOB1A reverted the ability of LATS1 to cause premature adaptation to the spindle checkpoint (Fig. 4D), suggesting that LATS1 function requires MOB1A.

Suppression of LATS1 prolongs telophase. In yeast, mutations that inactivate the MEN/SIN lead to mitotic arrest. Therefore, we examined whether suppression of LATS1 or MOB1A would have a similar phenotype in human cells. LATS1 or MOB1A protein levels were suppressed by siRNA in U2OS cells and progress through mitosis was monitored by live cell microscopy capturing images every 5 minutes through a phase contrast lens. The morphology of the cells allowed us to divide mitosis into three parts: from the first sign of change in cell shape until the cell became round (probably reflecting the interval from mid-prophase to prometaphase); from the time the cells became round until a cleavage furrow became evident (probably reflecting the interval from prometaphase to anaphase); and from the time the cleavage furrow became evident until the edges of the cell became less refractile (probably reflecting the interval from anaphase until completion of mitosis, i.e., telophase; Fig. 5A). siRNA-mediated suppression of LATS1 or MOB1A did not affect the length of the first two parts of mitosis (data not shown), but extended the length of the third part (Fig. 5B). The magnitude of the effect varied from cell to cell, but this probably reflected the cell-to-cell variability in siRNA-mediated suppression of LATS1 and MOB1A.

Figure 5.

Suppression of LATS1 or MOB1A prolongs telophase. A, time lapse phase contrast images of a U2OS cell treated with MOB1A-directed siRNA undergoing cell division. Mitosis was divided into three parts: part 1, when the cell begins to change morphology (arrowhead; 15-minute time point) until the cell becomes round (40 minutes); part 2, when the cell becomes round until it forms a cleavage furrow (75 minutes); part 3, when the cell forms a cleavage furrow until one or more of its edges become less refractile and cannot be distinguished from neighboring cells (165 minutes). B, length of part 3 of mitosis (defined as in A) in U2OS cells treated with control, LATS1- or MOB1A-directed siRNA. Each dot represents a single cell. Results from one experiment are shown, but the experiment was done in triplicate with identical results. C, comparison of S. cerevisiae MEN and S. pombe SIN to the proposed human MEN. For each node in the pathway, the S. cerevisiae protein is listed first, followed by the S. pombe (italics) and human (underlined) orthologues.

Figure 5.

Suppression of LATS1 or MOB1A prolongs telophase. A, time lapse phase contrast images of a U2OS cell treated with MOB1A-directed siRNA undergoing cell division. Mitosis was divided into three parts: part 1, when the cell begins to change morphology (arrowhead; 15-minute time point) until the cell becomes round (40 minutes); part 2, when the cell becomes round until it forms a cleavage furrow (75 minutes); part 3, when the cell forms a cleavage furrow until one or more of its edges become less refractile and cannot be distinguished from neighboring cells (165 minutes). B, length of part 3 of mitosis (defined as in A) in U2OS cells treated with control, LATS1- or MOB1A-directed siRNA. Each dot represents a single cell. Results from one experiment are shown, but the experiment was done in triplicate with identical results. C, comparison of S. cerevisiae MEN and S. pombe SIN to the proposed human MEN. For each node in the pathway, the S. cerevisiae protein is listed first, followed by the S. pombe (italics) and human (underlined) orthologues.

Close modal

In yeast, the MEN and SIN coordinate CDK inactivation, chromosome decondensation, and cytokinesis at the end of mitosis (1315). In higher eukaryotes, a signaling pathway homologous to the yeast MEN and SIN has not been described. Humans have two homologues of the yeast Mob1 protein (MOB1A and MOB1B), as well as four kinases belonging to the NDR family (LATS1, LATS2, NDR1, and NDR2), yet none of these proteins were previously implicated in mitotic exit. We propose here that LATS1 and MOB1A are components of the previously undefined human MEN. Multiple lines of evidence support this functional assignment. First, among the members of the human NDR kinase family, LATS1 has the highest sequence similarity to S. cerevisiae Dbf2 and S. pombe Sid2 and, among the human Mob family members, MOB1A has the highest sequence similarity to S. cerevisiae and S. pombe Mob1. Second, LATS1 interacts with MOB1A, but not with MOB2, similar to S. cerevisiae Dbf2 and S. pombe Sid2 kinases, which interact specifically with yeast Mob1. Third, human LATS1 and MOB1A localize at the centrosomes and midbody, which are equivalent to the yeast spindle pole bodies and mother-daughter cytokinesis ring, respectively, where the yeast MEN kinases and Mob1 proteins localize. Fourth, moderate overexpression of human LATS1 overcomes the spindle checkpoint, which is consistent with a mitotic exit-inducing function and reminiscent of the phenotype observed in S. cerevisiae and S. pombe when the MEN and SIN, respectively, are not subjected to negative regulation (3538). And fifth, siRNA-mediated suppression of LATS1 or MOB1A extends telophase, but not other phases of mitosis.

The mechanism by which LATS1 mediates mitotic exit remains elusive, but most likely involves the activation of phosphatases homologous to the S. cerevisiae Cdc14 phosphatase. Yeast Cdc14 is sequestered and kept inactive in the nucleolus throughout most of the cell cycle. It is partially released from the nucleolus during early anaphase and completely released at the end of mitosis (3941). Once released, Cdc14 dephosphorylates mitotic CDK substrates (42). The late mitotic release and activation of Cdc14 is dependent on Dbf2-Mob1, although the specific mechanism is not completely resolved. There are two Cdc14 proteins in human cells, hCDC14A and hCDC14B. Although they are not fully characterized, recent data suggests that hCDC14A dephosphorylates and activates APCCDH1, which targets mitotic cyclins for degradation, suggesting that it belongs to a MEN-like signaling network (4345). LATS1 may, therefore, stimulate mitotic exit by activating hCDC14.

A role of LATS1 in hCDC14 activation could explain the previously described ability of LATS1 to arrest cells in G2 (3, 10, 11). These previous experiments relied on adenoviral vectors to drive LATS1 expression. Because these vectors induce very high levels of protein expression, it is possible that hCDC14 was activated prematurely in G2, thereby leading to premature destruction of cyclin B1 and inability of the cells to fully activate CDC2 and enter mitosis. The recently proposed role of LATS1 in cytokinesis (9) is also consistent with LATS1 being a mitotic exit kinase, because the MEN and SIN are required for cytokinesis in S. cerevisiae and S. pombe, respectively (1315). In human cells, the ability of LATS1 to promote cytokinesis could involve LIMK1 (9).

The identification of LATS1 as a mitotic exit kinase will facilitate the elucidation of the entire human MEN. In S. cerevisiae, the MEN consists of a guanine nucleotide exchange factor (Lte1) and a bipartite GTPase-activating protein (Bub2-Bfa1) that regulate the GTPase Tem1; the latter, in its GTP-bound state, activates a protein kinase cascade (Fig. 5C). The kinase activated by Tem1, Cdc15, has sequence similarity to GTPase-activated kinases and contains a region that interacts specifically with the GTP-bound form of Tem1; Cdc15, in turn, activates Dbf2, which associates with Mob1 (1315).

Analysis of human LATS1 and Drosophila LATS suggests the presence of an evolutionarily conserved MEN pathway from yeast to insect and human cells. We showed here that human LATS1 interacts with MOB1A and it was recently shown that Drosophila LATS interacts with a Mob protein called MATS, which is the putative Drosophila orthologue of human MOB1A (46). Further, similar to the yeast MEN, Drosophila LATS and human LATS1 are activated by the Hippo and MST2 kinases, respectively (4751). Hippo and MST2 have high sequence similarity to S. cerevisiae Cdc15 and belong to the Ste20 protein kinase family, many members of which are activated by GTPases (52), similar to S. cerevisiae Cdc15, which is activated by Tem1 (Fig. 5C).

LATS was originally identified as a tumor suppressor in Drosophila (1, 2). Its inactivation in mosaic flies leads to the development of large tumors in many organs. Mammalian LATS1 is also a tumor suppressor. LATS1−/− mice develop soft tissue sarcomas and ovarian cell tumors, especially, when exposed to carcinogens (4). In humans, LATS1 promoter methylation and missense point mutations are frequent in soft tissue sarcomas (7, 8). Drosophila MATS is also a tumor suppressor, as is Salvador, a gene that functions in the same pathway as LATS in Drosophila; further, the human homologue of Salvador is deleted in renal cancer cell lines (46, 53). Taken together, these findings suggest that the MEN functions as a tumor suppressor pathway in Drosophila and mammals. Why inactivation of the MEN promotes tumor formation remains to be elucidated. In yeast, the MEN, in addition to its role in mitotic exit and cytokinesis, prepares cells for entry into G1 by regulating transcription of genes, such as the CDK inhibitor Sic1 (42). In Drosophila, LATS, Hippo, and Salvador allow cells in a developmentally appropriate manner to permanently exit the cell cycle and/or undergo apoptosis after completing mitosis (4749, 53). This is accomplished by suppressing expression of cyclin E and modulating expression of proapoptotic and antiapoptotic genes. Thus, it is possible that the tumor suppressor activity of the MEN may relate to its role in G1. Alternatively, inactivation of the MEN may promote tumor development by leading to genomic instability. In principle, any gene that plays a role in mitosis could lead to errors in sister chromatid segregation and genomic instability. As long as inactivation of the gene is not a lethal event, then the gene in question may be a tumor suppressor. LATS1 fits this profile, because its inactivation does not lead to lethality in mice (4).

Grant support: National Cancer Institute grant CA105160 (T.D. Halazonetis), National Institute General Medical Sciences grant GM60575 (F.C. Luca) and National Cancer Institute training grants CA09677 (J. Bothos) and CA09171 (R.L. Tuttle).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Wufan Tao for the myc-tagged LATS1 expression plasmid, Russel Kaufman for support and helpful discussions, and the Wistar Institute Nucleic Acid and Hybridoma Facilities for DNA sequencing analysis and generating the MOB1A monoclonal antibody, respectively.

1
Justice RW, Zilian O, Woods DF, Noll M, Bryant PJ. The Drosophila tumor suppressor gene warts encodes a homolog of human myotonic dystrophy kinase and is required for the control of cell shape and proliferation.
Genes Dev
1995
;
9
:
534
–46.
2
Xu T, Wang W, Zhang S, Stewart RA, Yu W. Identifying tumor suppressors in genetic mosaics: the Drosophila lats gene encodes a putative protein kinase.
Development
1995
;
121
:
1053
–63.
3
Tao W, Zhang S, Turenchalk GS, et al. Human homologue of the Drosophila melanogaster lats tumour suppressor modulates CDC2 activity.
Nat Genet
1999
;
21
:
177
–81.
4
St. John MA, Tao W, Fei X, et al. Mice deficient of Lats1 develop soft-tissue sarcomas, ovarian tumours and pituitary dysfunction.
Nat Genet
1999
;
21
:
182
–6.
5
Yabuta N, Fujii T, Copeland NG, et al. Structure, expression, and chromosome mapping of LATS2, a mammalian homologue of the Drosophila tumor suppressor gene lats/warts.
Genomics
2000
;
63
:
263
–70.
6
Hori T, Takaori-Kondo A, Kamikubo Y, Uchiyama, T. Molecular cloning of a novel human protein kinase, kpm, that is homologous to warts/lats, a Drosophila tumor suppressor.
Oncogene
2000
;
19
:
3101
–9.
7
Hansen LL, Jensen LL, Dimitrakakis C, et al. Allelic imbalance in selected chromosomal regions in ovarian cancer.
Cancer Genet Cytogenet
2002
;
139
:
1
–8.
8
Hisaoka M, Tanaka A, Hashimoto H. Molecular alterations of h-warts/LATS1 tumor suppressor in human soft tissue sarcoma.
Lab Invest
2002
;
82
:
1427
–35.
9
Yang X, Yu K, Hao Y, et al. LATS1 tumour suppressor affects cytokinesis by inhibiting LIMK1.
Nat Cell Biol
2004
;
6
:
609
–17.
10
Yang X, Li DM, Chen W, Xu T. Human homologue of Drosophila lats, LATS1, negatively regulate growth by inducing G(2)/M arrest or apoptosis.
Oncogene
2001
;
20
:
6516
–23.
11
Xia H, Qi H, Li Y, et al. LATS1 tumor suppressor regulates G2/M transition and apoptosis.
Oncogene
2002
;
21
:
1233
–41.
12
Tamaskovic R, Bichsel SJ, Hemmings BA. NDR family of AGC kinases-essential regulators of the cell cycle and morphogenesis.
FEBS Lett
2003
;
546
:
73
–80.
13
Bardin AJ, Amon A. Men and sin: what's the difference?
Nat Rev Mol Cell Biol
2001
;
2
:
815
–26.
14
McCollum D, Gould KL. Timing is everything: regulation of mitotic exit and cytokinesis by the MEN and SIN.
Trends Cell Biol
2001
;
11
:
89
–95.
15
Simanis V. The mitotic exit and septation initiation networks.
J Cell Sci
2003
;
116
:
4261
–2.
16
Racki WJ, Becam AM, Nasr F, Herbert CJ. Cbk1p, a protein similar to the human myotonic dystrophy kinase, is essential for normal morphogenesis in Saccharomyces cerevisiae.
EMBO J
2000
;
19
:
4524
–32.
17
Bidlingmaier S, Weiss EL, Seidel C, Drubin DG, Snyder M. The Cbk1p pathway is important for polarized cell growth and cell separation in Saccharomyces cerevisiae.
Mol Cell Biol
2001
;
21
:
2449
–62.
18
Colman-Lerner A, Chin TE, Brent R. Yeast Cbk1 and Mob2 activate daughter-specific genetic programs to induce asymmetric cell fates.
Cell
2001
;
107
:
739
–50.
19
Weiss EL, Kurischko C, Zhang C, Shokat K, Drubin DG, Luca FC. The Saccharomyces cerevisiae Mob2p-Cbk1p kinase complex promotes polarized growth and acts with the mitotic exit network to facilitate daughter cell-specific localization of Ace2p transcription factor.
J Cell Biol
2002
;
158
:
885
–900.
20
Verde F, Wiley DJ, Nurse P. Fission yeast orb6, a ser/thr protein kinase related to mammalian rho kinase and myotonic dystrophy kinase, is required for maintenance of cell polarity and coordinates cell morphogenesis with the cell cycle.
Proc Natl Acad Sci U S A
1998
;
95
:
7526
–31.
21
Komarnitsky SI, Chiang YC, Luca FC, et al. DBF2 protein kinase binds to and acts through the cell cycle-regulated MOB1 protein.
Mol Cell Biol
1998
;
18
:
2100
–7.
22
Salimova E, Sohrmann M, Fournier N, Simanis V. The S. pombe orthologue of the S. cerevisiae mob1 gene is essential and functions in signalling the onset of septum formation.
J Cell Sci
2000
;
113
:
1695
–704.
23
Lee SE, Frenz LM, Wells NJ, Johnson AL, Johnston LH. Order of function of the budding-yeast mitotic exit-network proteins Tem1, Cdc15, Mob1, Dbf2, and Cdc5.
Curr Biol
2001
;
11
:
784
–8.
24
Mah AS, Jang J, Deshaies RJ. Protein kinase Cdc15 activates the Dbf2-Mob1 kinase complex.
Proc Natl Acad Sci U S A
2001
;
98
:
7325
–30.
25
Hou MC, Wiley DJ, Verde F, McCollum D. Mob2p interacts with the protein kinase Orb6p to promote coordination of cell polarity with cell cycle progression.
J Cell Sci
2003
;
116
:
125
–35.
26
Hennebold JD, Tanaka M, Saito J, Hanson BR, Adashi EY. Ovary-selective genes I: the generation and characterization of an ovary-selective complementary deoxyribonucleic acid library.
Endocrinology
2000
;
141
:
2725
–34.
27
Stavridi ES, Harris KG, Huyen Y, et al. Crystal structure of a human Mob1 protein: toward understanding Mob-regulated cell cycle pathways.
Structure
2003
;
11
:
1163
–70.
28
Devroe E, Erdjument-Bromage H, Tempst P, Silver PA. Human Mob proteins regulate the NDR1 and NDR2 serine-threonine kinases.
J Biol Chem
2004
;
279
:
24444
–51.
29
Hou MC, Guertin DA, McCollum D. Initiation of cytokinesis is controlled through multiple modes of regulation of the Sid2p-Mob1p kinase complex.
Mol Cell Biol
2004
;
24
:
3262
–76.
30
Ponchon L, Dumas C, Kajava AV, Fesquet D, Padilla A. NMR solution structure of Mob1, a mitotic exit network protein and its interaction with an NDR kinase peptide.
J Mol Biol
2004
;
337
:
167
–82.
31
Nishiyama Y, Hirota T, Morisaki T, et al. A human homolog of Drosophila warts tumor suppressor, h-warts, localized to mitotic apparatus and specifically phosphorylated during mitosis.
FEBS Lett
1999
;
459
:
159
–65.
32
Hirota T, Morisaki T, Nishiyama Y, et al. Zyxin, a regulator of actin filament assembly, targets the mitotic apparatus by interacting with h-warts/LATS1 tumor suppressor.
J Cell Biol
2000
;
149
:
1073
–86.
33
Sohrmann M, Schmidt S, Hagan I, Simanis V. Asymmetric segregation on spindle poles of the Schizosaccharomyces pombe septum-inducing protein kinase Cdc7p.
Genes Dev
1998
;
12
:
84
–94.
34
Cerutti L, Simanis V. Asymmetry of the spindle pole bodies and spg1p GAP segregation during mitosis in fission yeast.
J Cell Sci
1999
;
112
:
2313
–21.
35
Hoyt MA, Totis L, Roberts BT. S. cerevisiae genes required for cell cycle arrest in response to loss of microtubule function.
Cell
1991
;
66
:
507
–17.
36
Murone M, Simanis V. The fission yeast dma1 gene is a component of the spindle assembly checkpoint, required to prevent septum formation and premature exit from mitosis if spindle function is compromised.
EMBO J
1996
;
15
:
6605
–16.
37
Fesquet D, Fitzpatrick PJ, Johnson AL, Kramer KM, Toyn JH, Johnston LH. A Bub2p-dependent spindle checkpoint pathway regulates the Dbf2p kinase in budding yeast.
EMBO J
1999
;
18
:
2424
–34.
38
Guertin DA, Venkatram S, Gould KL, McCollum D. Dma1 prevents mitotic exit and cytokinesis by inhibiting the septation initiation network (SIN).
Dev Cell
2002
;
3
:
779
–90.
39
Shou W, Seol JH, Shevchenko A, et al. Exit from mitosis is triggered by Tem1-dependent release of the protein phosphatase Cdc14 from nucleolar RENT complex.
Cell
1999
;
97
:
233
–44.
40
Visintin R, Hwang ES, Amon A. Cfi1 prevents premature exit from mitosis by anchoring Cdc14 phosphatase in the nucleolus.
Nature
1999
;
398
:
818
–23.
41
Stegmeier F, Visintin R, Amon A. Separase, polo kinase, the kinetochore protein Slk19, and Spo12 function in a network that controls Cdc14 localization during early anaphase.
Cell
2002
;
108
:
207
–20.
42
Visintin R, Craig K, Hwang ES, Prinz S, Tyers M, Amon A. The phosphatase Cdc14 triggers mitotic exit by reversal of Cdk-dependent phosphorylation.
Mol Cell
1998
;
2
:
709
–18.
43
Bembenek J, Yu H. Regulation of the anaphase-promoting complex by the dual specificity phosphatase human Cdc14a.
J Biol Chem
2001
;
276
:
48237
–42.
44
Kaiser BK, Zimmerman ZA, Charbonneau H, Jackson PK. Disruption of centrosome structure, chromosome segregation, and cytokinesis by misexpression of human Cdc14A phosphatase.
Mol Biol Cell
2002
;
13
:
2289
–300.
45
Mailand N, Lukas C, Kaiser BK, Jackson PK, Bartek J, Lukas J. Deregulated human Cdc14A phosphatase disrupts centrosome separation and chromosome segregation.
Nat Cell Biol
2002
;
4
:
317
–22.
46
Lai ZC, Wei X, Shimizu T, et al. Control of cell proliferation and apoptosis by mob as tumor suppressor, mats.
Cell
2005
;
120
:
675
–85.
47
Harvey KF, Pfleger CM, Hariharan IK. The Drosophila Mst ortholog, hippo, restricts growth and cell proliferation and promotes apoptosis.
Cell
2003
;
114
:
457
–67.
48
Udan RS, Kango-Singh M, Nolo R, Tao C, Halder G. Hippo promotes proliferation arrest and apoptosis in the Salvador/Warts pathway.
Nat Cell Biol
2003
;
5
:
914
–20.
49
Wu S, Huang J, Dong J, Pan D. hippo encodes a Ste-20 family protein kinase that restricts cell proliferation and promotes apoptosis in conjunction with salvador and warts.
Cell
2003
;
114
:
445
–56.
50
Creasy CL, Chernoff J. Cloning and characterization of a member of the MST subfamily of Ste20-like kinases.
Gene
1995
;
167
:
303
–6.
51
Chan EH, Nousiainen M, Chalamalasetty RB, Schafer A, Nigg EA, Sillje HH. The Ste20-like kinase Mst2 activates the human large tumor suppressor kinase Lats1.
Oncogene
2005
;
24
:
2076
–86.
52
Dan I, Watanabe NM, Kusumi A. The Ste20 group kinases as regulators of MAP kinase cascades.
Trends Cell Biol
2001
;
11
:
220
–30.
53
Tapon N, Harvey KF, Bell DW, et al. salvador Promotes both cell cycle exit and apoptosis in Drosophila and is mutated in human cancer cell lines.
Cell
2002
;
110
:
467
–78.