Kinase activity-independent role of EphA2 in the regulation of M-phase progression
Yuichiro Kaibori, Kiriko Katayama, Yuka Tanaka, Masayoshi Ikeuchi, Mika Ogawa, Yuki Ikeda, Ryuzaburo Yuki, Youhei Saito, Yuji Nakayama
Highlight:
EphA2 knockdown-induced M-phase delay was rescued by EphA2 D739N mutant EphA2 D739N mutant loses the kinase activity
The membrane localization of EphA2 is independent of its kinase activity
Kinase activity of EphA2 is dispensable for RhoG recruitment to the plasma membrane in M phase
Abbreviations
Cdk1, cyclin-dependent kinase 1; DMEM, Dulbecco’s Modified Eagle Medium; Dox, doxycycline; ERK, extracellular signal-regulated kinase; Eph receptor, erythropoietin-producing human hepatocellular receptor; FBS, fetal bovine serum; GAP, GTPase-activating protein; GEF, guanine nucleotide exchange factor; MAPK, mitogen-activated protein kinase; PMSF, phenylmethylsulfonyl fluoride; PVDF, polyvinylidene difluoride membranes; RhoG, Ras homolog gene family member G; RSK, 90 kDa ribosomal S6 kinase; STLC, S-trityl-L-cysteine; TTBS, tween 20-contained tris-buffered saline; 3′-UTR, 3′-untranslated region
Abstract
Cell division is a tightly regulated, essential process for cell proliferation. Very recently, we reported that EphA2 is phosphorylated at Ser897, via the Cdk1/MEK/ERK/RSK pathway, during M phase and contributes to proper M-phase progression by maintaining cortical rigidity via the EphA2pSer897/ephexin4/RhoG pathway. Here, we show that EphA2 kinase activity is dispensable for M-phase progression. Although EphA2 knockdown delayed this progression, the delay was rescued by an EphA2 mutant expression with an Asp739 to Asn substitution, as well as by wild-type EphA2. Western blotting analysis confirmed that the Asp739Asn mutant lost its EphA2 kinase activity. Like wild-type EphA2, the Asp739Asn mutant was localized to the plasma membrane irrespective of cell cycle. While RhoG localization to the plasma membrane was decreased in EphA2 knockdown cells, it was rescued by re-expression of wild-type EphA2 but not via the mutant containing the Ser897 to Ala substitution. This confirmed our recent report that phosphorylation at Ser897 is responsible for RhoG localization to the plasma membrane. In agreement with the M-phase progression’s rescue effect, the Asp739Asn mutant rescued RhoG localization in EphA2 knockdown cells. These results suggest that EphA2 regulates M-phase progression in a manner independent of its kinase activity.
1. Introduction
Cell division is an essential process involved in cell proliferation. Aberrant cell division can lead to asymmetrical chromosome segregation and give rise to cellular transformation (1); thus, cell division is tightly regulated. Faithful cell division requires the bipolar mitotic spindle to form properly. In the early M phase, following nuclear envelope breakdown, mitotic spindle microtubules emanate from separated centrosomes. Spindle microtubules are classified into three groups: kinetochore, interpolar, and astral microtubules (2). Kinetochore microtubules connect kinetochores on the chromosomes with centrosomes and contribute to chromosome translocation. Interpolar microtubules, emanating from two opposite centrosomes, overlap at the cells’ equatorial region and control spindle length. Astral microtubules radiate from centrosomes and are tethered to the cell cortex, regulating the spindle position via exertion of cortical pulling forces. These spindle microtubules contribute to chromosome alignment at the cellular equator and, thereby, to bipolar spindle formation as well. Once the mitotic spindle properly forms and the spindle assembly checkpoint is satisfied, kinetochore microtubules pull chromosomes toward the opposite poles, an action defined as anaphase onset (3). Importantly, the astral microtubules’ interaction with the cortex can generate pulling forces against chromosomes via centrosomes (4). Therefore, loss of cortical rigidity can cause defects in mitotic spindle morphology and function, which also affects cell rounding at mitotic entry (5–7). Cortical rigidity—which is sustained by cortical actin and actin-membrane interaction(s)—is essential for bipolar spindle formation in addition to spindle microtubules (4).
Eph receptors, the largest family within the receptor-type tyrosine kinases, are classified into two groups based on their affinities for ephrin-A and ephrin-B ligands: EphA and EphB (8). In mammalian cells, there are nine EphA receptors and five EphB receptors, which generally interact with ephrin-A and ephrin-B ligands, respectively. However, EphA4 and EphB2 are exceptions to this trend and bind to ephrin-B and ephrin-A5 ligands, respectively (8). Because ephrin-As are anchored to the plasma membrane by glycosylphosphatidyl inositol (GPI) anchors, and ephrin-Bs are transmembrane proteins, cell-cell interactions are required to introduce ligand-dependent signaling, which is involved in a variety of cellular functions, including: actin cytoskeleton remodeling, cell shape regulation, formation of cell junctions, cell proliferation, and differentiation (8–10). A well-known function elicited by the receptor–ligand interaction is the repulsive response, leading to cell sorting. It prevents cell subpopulations from intermingling for boundary formation during the developmental process (10). EphA2, a member of the Eph family, possesses two distinct signaling pathways: ligand-dependent and -independent (11). Interaction with ephrin-A1 causes phosphorylation at Tyr588/594 within the juxtamembrane domain and elicits the ligand-dependent signaling pathway that regulates the aforementioned events. On the other hand, multiple growth factors and TNF-α cause phosphorylation of EphA2 at Ser897 by activation of Akt, RSK, or PKA without ligand binding to EphA2 (11–14). This signaling promotes cancer cell migration and invasion by forming cortactin protrusions (15).
Very recently, we reported that EphA2 is phosphorylated at Ser897, via the Cdk1/MEK/ERK/RSK pathway during M phase, thereby contributing to proper M-phase progression via maintenance of cortical rigidity (16). NVP-BHG712 — an inhibitor of Eph receptor kinase activity
— causes M-phase delay; we assumed that EphA2’s kinase activity is required for M-phase progression. However, NVP-BHG712 treatment decreases EphA2 phosphorylation at both Ser897 and Tyr588 (16). In addition, the EphA2 mutant harboring a Ser897 to Ala substitution is incapable of rescuing the M-phase delay caused by EphA2 knockdown. We thus concluded that decreased phosphorylation at Ser897, and not kinase activity, is responsible for the delay of M-phase progression within NVP-BHG712-treated cells. Consistently, the kinase-dependent EphA2 phosphorylation at Tyr588 is reduced in the early stage of M phase, even without the inhibitor treatment. However, it remains unclear whether or not EphA2 kinase activity is actually dispensable for M-phase progression. Here we evaluate EphA2 kinase activity’s involvement during M-phase progression by establishing a cell line capable of expressing a kinase-dead EphA2 mutant. There was no difference in M-phase progression between these mutant- and wild-type-expressing cells, even when endogenous EphA2 was knocked down. Based on these results, we concluded that EphA2’s kinase activity is dispensable for proper M-phase progression.
2. Materials and methods
2.1. Cell culture and transfection
HeLa S3 (Japanese Collection of Research Bioresources, Osaka, Japan) and Lenti-X 293T (Clontech Laboratories, Mountain View, CA, USA) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 5% FBS, 20 mM HEPES (pH7.4), and 2 mM L(+)-Gln with humidified air containing 5% CO2 at 37°C. hTERT RPE-1 cells (CRL-4000; American Type Culture Collection, Manassas, VA, USA) were cultured in DMEM/Ham’s F-12 medium. For siRNA transfection, HeLa S3 cells were seeded in 24-well plates and transfected with 20 pmol siRNA per well using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The siRNA targeting the 3′-UTR of EphA2 (5′-CGGACAGACAUAUAGGAUATT-3′) was synthesized by MilliporeSigma (Burlington, MA, USA).
2.2. Plasmid DNA
EphA2 within the pDONR223 vector (pDONR223-EphA2, provided by William Hahn and David Root, plasmid #23926: Addgene, Watertown, MA, USA) (17) was recombined into a pLIX_402 lentiviral vector (provided by David Root, plasmid #41394; Addgene) (18) via Gateway cloning (Invitrogen) as previously described (16). The EphA2 mutants, containing an Asp739 to Asn739 substitution (11) or Lys646 to Met646 substitution (19), were generated using pDONR223-EphA2 as a template with KOD-plus-Neo DNA polymerase (Toyobo, Tokyo, Japan) according to the manufacturer’s instructions. The sense and antisense primers used for introducing the Asp739Asn mutation were 5′- . These mutants were recombined into the pLIX_402 vector. The lentiviral packaging plasmids pCAG-HIVgp and pCMV-VSV-G-RSV-Rev were provided by Dr. Hiroyuki Miyoshi [Rikagaku Kenkyusho Foundation (RIKEN) BioResource Center, Ibaraki, Japan].
2.3. Establishment of clonal cells capable of inducible EphA2 expression
The HeLa S3-derived clones HeLa S3/WT-HA and HeLa S3/S897A-HA, which can induce expression of HA-tagged wild-type EphA2 (WT-HA) and an EphA2 mutant harboring a Ser897 to Ala substitution (S897A-HA), respectively, were established as previously described (16). To establish HeLa S3/D739N-HA cells, which can induce expression of an HA-tagged EphA2 mutant containing an Asp739 to Asn substitution (D739N-HA), Lenti-X 293T cells were transfected with pLIX_402 vector harboring D739N-HA, pCAG-HIVgp and pCMV-VSV-G-RSV-Rev using PEI Max (Polysciences, Warrington, PA, USA). HeLa S3 cells were infected, using a virus-containing medium, and selected with puromycin as described previously (16). To establish hTERT RPE-1/WT-HA and hTERT RPE-1/K646M-HA cells, which can express HA-tagged wild-type EphA2 (WT-HA) and its mutant containing a Lys646 to Met substitution (K646M-HA), respectively, gene transduction was performed as described above using pLIX_402 vector harboring WT-HA or K646M-HA. The infected hTERT RPE-1 cells were selected with 4 µg/ml puromycin.
2.4. Cell cycle synchronization
For synchronization during M phase, HeLa S3 and hTERT RPE-1 cells were treated with 6 and 8 M RO-3306 (Selleck Chemicals, Houston, TX, USA), respectively, for 20 h and washed 4x with pre-warmed PBS containing MgCl2 and CaCl2 [PBS(+)] (20). These cells were cultured in a fresh, pre-warmed medium (37°C) for 1.5 h, and the M-phase cells were collected by mitotic shake-off for use in Western blotting or fixed for immunofluorescence staining. Alternatively, S-trityl-L-cysteine (STLC) was used for cell cycle synchronization in the M phase by incubating for 16 h. To synchronize cells at metaphase, HeLa S3 cells were released for 30 min from 20-h treatment with RO-3306 and then treated with 10 µM MG-132 for 1 h.
2.5. Antibodies
Primary antibodies used for immunoblotting (IB) or immunofluorescence (IF) were as follows: rabbit monoclonal anti-EphA2 (IB, 1:1000; 6997S, Cell Signaling Technology, Danvers, MA, USA), anti-phospho-EphA2 Tyr588 (IB, 1:1000–2000; 12677S, Cell Signaling Technology), anti-phospho-EphA2 Ser897 (IB, 1:1000; 6347S, Cell Signaling Technology), anti-phospho-Aurora A Thr288/Aurora B Thr232/Aurora C Thr198 (IB, 1:1000; 2914S, Cell Signaling Technology), anti-phospho-p90RSK Ser380 (IB, 1:1000; 11989S, Cell Signaling Technology) antibodies, rabbit polyclonal anti-HA (IF, 1:200; sc-805, Santa Cruz Biotechnology, Dallas, TX, USA), anti-ERK2 (IB, 1:500; sc-154, Santa Cruz Biotechnology), and anti-cyclin B1 (IB, 1:2000; sc-752, Santa Cruz Biotechnology) antibodies, mouse monoclonal anti-phospho-p44/42 MAPK Thr202/Tyr204 (pERK1/2; IB, 1:1000; 9106S, Cell Signaling Technology), anti-IAK1 (Aurora A; IB, 1:1000; 610938, BD Biosciences, San Jose, CA, USA), anti-AIM-1 (Aurora B; IB, 1:1000; 611082, BD Biosciences), anti-HA (IB, 1:1000–2000; M180-3, Medical and Biological Laboratories, Nagoya, Japan), anti-RhoG (IF, 1:200; sc-80015, Santa Cruz Biotechnology) antibodies, and rat monoclonal anti--tubulin (IB, 1:2000; MCA78G, Bio-Rad, Hercules, CA, USA) antibody. For immunoblotting analysis, horseradish peroxidase-conjugated goat anti-mouse IgG (light chain specific, 1:8000; 115-035-174), donkey anti-mouse IgG (1:8000; 715-035-151), monoclonal mouse anti-rabbit IgG (light chain specific, 1:8000; 211-032-171), donkey anti-rabbit IgG (1:8000; 711-035-152), and donkey anti-rat IgG (1:8000; 712-035-153) antibodies were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA, USA) to be used as secondary antibodies. For immunofluorescence, Alexa Fluor 488- or 555-conjugated donkey anti-mouse, donkey anti-rabbit, and goat anti-rat (1:1000; Invitrogen) IgG antibodies were used.
2.6. Western blotting
Western blotting analysis was carried out as previously (21, 22). Briefly, cells were washed with PBS(-) and then lysed in SDS-sample buffer containing 2 g/ml aprotinin (Seikagaku Kogyo, Tokyo, Japan), 0.8 g/ml pepstain A (Wako Pure Chemicals, Osaka, Japan), 2 g/ml leupeptin (Nacalai Tesque, Kyoto, Japan), 2 mM phenylmethylsulfonyl fluoride (PMSF, Nacalai Tesque), 20 mM -glycerophosphate (MilliporeSigma), 50 mM NaF (Wako), and 10 mM Na3VO4 (Wako). The lysed proteins were separated by SDS-PAGE and transferred onto a PVDF membrane (Pall Corporation, Port Washington, NY, USA). After blocking with Blocking One (Nacalai Tesque) reagent for 30 min at room temperature, the membrane was incubated with primary, and then secondary, antibodies for 1–3 h at room temperature. Detection was carried out using a ChemiDoc XRSplus image analyzer (Bio-Rad) with Clarity (Bio-Rad) or Chemi-Lumi One L (Nacalai Tesque) as the enhanced chemiluminescence substrate
2.7. in vitro kinase assay
Lenti-X 293T cells were transfected with EphA2-HA (WT-HA) and EphA2-D739N-HA (D739N-HA) using Polyethylenimine “MAX” (PEI-MAX, Polysciences, Warrington, PA, USA) and incubated with 0.25 µg/ml Dox for 24 h. The cells were lysed with 1% Triton lysis buffer containing protease inhibitors (1 mM PMSF, 40 µg/ml aprotinin, 16 µg/ml pepstatin A, 2 µg/ml leupeptin and 10 mM EGTA-KOH) for 15 min on ice. After centrifugation for 10 min, the supernatants were incubated at 37°C for 60 min and then incubated with sepharose beads conjugating anti-HA antibody. Immunoprecipitated HA-tagged wild-type and mutant EphA2 were incubated in the buffer (5 mM MgCl2, 10 mM Na3VO4, 10 mM HEPES, 1 mM PMSF, 40 µg/ml aprotinin, 16 µg/ml pepstatin A, 2 µg/ml leupeptin and 10 mM EGTA-KOH) with or without 1 µM ATP at 30°C for indicated periods and analyzed by western blotting with indicated antibodies.
2.8. Immunofluorescence microscopy
Immunofluorescence staining was carried out as previously described (16, 23). Briefly, cells were seeded on coverslips and fixed with pre-warmed (37°C) 4% formaldehyde in PBS(-) for 20 min at room temperature. Alternatively, cells were fixed with 4% formaldehyde in PBS containing 2 mM PIPES (pH 6.8), 0.2% Triton X-100, 10 mM EGTA, and 1 mM MgCl2 for 20 min. After incubation for 30 min with PBS(−) containing 0.1% saponin and 3% bovine serum albumin for blocking and permeabilization, the cells were incubated with the primary, and then secondary, antibodies for 1 h each. Fluorescence images were captured by a laser scanning microscope (LSM800; Carl Zeiss, Oberkochen, Germany) equipped with a 63x/1.42 NA oil immersion objective lens. Hoechst 33342, Alexa Fluor 488, and Alexa Fluor 555 were excited with the 405, 488, and 561 nm wavelengths, respectively, and their fluorescence were detected with 400–460, 510–550, and 570–620 nm emission filters, respectively. The images were edited by ImageJ (National Institutes of Health, Bethesda, MD, USA), Photoshop Elements 12 (Adobe, San Jose, CA, USA), and Illustrator CC (Adobe) softwares.
2.9. Statistics
The homogeneity of variance was tested by Bartlett’s test. Statistical differences for multiple comparisons with equal variances were analyzed by one-way ANOVA, followed by the Tukey-Kramer test, using Statcel 4 software (OMS Publishing, Tokorozawa, Japan), Microsoft Excel (Redmond, WA, USA), and EZR (Saitama Medical Centre, Jichi Medial University; http://www.jichi.ac.jp/saitama-sct/SaitamaHP.files/statmedEN.html; Kanda, 2012), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria, version 2.13.0).
3. Results
3.1. Kinase activity of EphA2 is dispensable for M-phase progression
As reported previously (16), upon mitotic entry, EphA2 phosphorylation at Tyr588 was decreased, and phosphorylation at Ser897 was subsequently increased (Fig. 1). In agreement, phosphorylation of ERK and RSK kinases, which are the upstream kinases of phosphorylation of EphA2 at Ser897, was also detected in M-phase cells, in which the mitotic kinases Aurora kinases were accumulated and phosphorylated. Because EphA2 phosphorylation at Tyr588 is induced when kinase activity is activated, via binding of the ligand ephrin-A1, we hypothesized that EphA2 kinase activity would be dispensable for proper M-phase progression. To address this issue, HeLa S3-derived cell lines were generated, which express either wild-type EphA2 (WT-HA) or kinase-dead mutant with Asp739 to Asn substitution (D739N-HA) (11) in a Dox-inducible manner. HeLa S3/WT-HA and HeLa S3/D739N-HA cells were knocked down for endogenous EphA2, and WT-HA and D739N-HA expression was induced by treatment with 5 and 0.25 µg/mL Dox, respectively (Fig. 2A). The siRNA treatment reduced endogenous EphA2 levels in both HeLa S3/WT-HA and HeLa S3/D739N-HA cells. Detection with Anti-EphA2 antibody revealed that WT-HA and D739N-HA were expressed with Dox treatment at higher levels than endogenous EphA2, enabling us to detect Tyr588 phosphorylation of exogenously introduced WT-HA. In spite of the slightly higher levels observed in D739N-HA expression, compared to WT-HA, the band corresponding to phosphorylated Tyr588 almost completely disappeared in HeLa S3/D739N-HA cells (Fig. 2A).
Next, these cells were synchronized in the M phase by treatment with the Eg5 inhibitor STLC. Inhibition of Eg5 caused the monopolar spindle to form, resulting in M-phase arrest via activation of the spindle assembly checkpoint. M-phase cells were collected by mitotic shake-off and subjected to western blotting analysis (Fig. 2B). Similar to that occurring during interphase, EphA2 phosphorylation at Tyr588 was not detected in HeLa S3/D739N-HA cells during the M phase (Fig. 2B left), which was confirmed by detecting the mitotic markers (Fig. 2B right). As described, Tyr588 phosphorylation was lower in M phase than that in interphase; however, detection of the signal of overexpressed EphA2 for longer period enabled detection of this phosphorylation. Furthermore, in vitro kinase assay was performed by detecting phosphorylation of EphA2 at Tyr588 (Fig. 2C). Incubation of immunoprecipitated EphA2 without ATP showed no signal of Tyr588 phosphorylation, indicating that phosphorylation in cells were dephosphorylated during purification step. When incubated with ATP, wild-type EphA2 (WT-HA) showed phosphorylation at Tyr588, indicating that this residue was phosphorylated in vitro. On the contrary, EphA2 mutant having D739N substitution (D739N-HA) completely lost this signal, confirming that D739N-HA mutant lacks kinase activity.
To investigate the requirement of the kinase activity for M-phase progression, these cells were arrested in the G2/M border via treatment with the reversible CDK1 inhibitor RO-3306. After washing out RO-3306, the cell cycle resumed, and M-phase progression was examined. As previously (16), endogenous EphA2 knockdown delayed M-phase progression, and WT-HA re-expression rescued this M-phase delay (Fig. 3A, left). Interestingly, such M-phase delay was rescued, not only by WT-HA but also D739N-HA (Fig. 3B, left). Of note, knockdown of endogenous EphA2 and re-expression of either WT-HA or D739N-HA did not affect the mitotic index (Fig. 3A, B, right), suggesting EphA2 is not involved in mitotic entry. No requirement of kinase activity was confirmed by using the hTERT-immortalized retinal pigment epithelial hTERT RPE-1 cells (Fig. 4). At 45 min after release from RO-3306 treatment, only 21% of cells did not yet align chromosomes in prophase/prometaphase, and the rest of cells already aligned chromosomes. Knockdown of EphA2 increased the number of cells in prophase/prometaphase, indicating delay in chromosome alignment; however, expression of the kinase dead K646M mutant (K646M-HA) (19) rescued it as well as wild-type EphA2 (WT-HA). These results suggest that the EphA2 kinase activity is not required for the regulation of M-phase progression.
3.2. Membrane localization of EphA2 is independent of kinase activity
In our previous work, we showed that EphA2 phosphorylation at Ser897 maintains cortical rigidity in M-phase cells, through EphA2pSer897/Ephexin4/RhoG signaling, at the plasma membrane. Thus, we next compared subcellular localization of the D739N-HA mutant with WT-HA. HeLa S3/WT-HA and HeLa S3/D739N-HA cells were treated with Dox and stained for HA-tag. D739N-HA was localized to the plasma membrane, similar to wild-type EphA2, which was primarily localized to the plasma membrane within the interphase cells (Fig. 5A). Both D739N-HA and wild-type EphA2 were localized to the plasma membrane within the M-phase cells as well (Fig. 5B). These results suggest that membrane localization of EphA2 is independent of its kinase activity. Interestingly, WT-HA showed cytoplasmic localization during interphase and M phase; these localizations were reduced with D739N-HA. Since ligand binding with EphA2 activates its endocytosis with activation of EphA2 kinase activity (24), the loss of cytoplasmic localization of D739N-HA may be attributed to inactive kinase activity.
3.3. Membrane localization of RhoG during M phase does not rely on EphA2 kinase activity
We previously reported that EphA2 phosphorylation at Ser897 recruits RhoG to the plasma membrane to regulate cortical rigidity during the M phase, thereby contributing to proper M-phase progression (16). In the present study, we showed that EphA2 kinase activity was not required for proper M-phase progression. Therefore, we hypothesized that its kinase activity may be dispensable for RhoG recruitment to the plasma membrane during the M phase. To investigate this, endogenous EphA2 was knocked down by siRNA targeting 3′UTR of EphA2 in HeLa S3/WT-HA, HeLa S3/D739N-HA and HeLa S3/S897A-HA cells, followed by induction, via Dox treatment, of WT-HA, D739N-HA, or EphA2 mutant containing a Ser897 to Ala substitution (S897A-HA), respectively. In agreement with our previous report (16), RhoG localization to the plasma membrane was decreased in S897A-HA-expressing cells (Fig. 4A, B), confirming that phosphorylation at Ser897 is responsible for RhoG recruitment to the plasma membrane. In contrast, RhoG membrane localization was unaffected in D739N-HA-expressing as well as WT-HA-expressing cells (Fig. 4A, B). These results suggest that EphA2 kinase activity is dispensable for RhoG recruitment to the plasma membrane during the M phase. Taken together, EphA2 regulates M-phase progression in EphA2-expressing cells through recruitment of RhoG to the plasma membrane in a manner independent of its kinase activity.
4. Discussion
Formation of the mitotic spindle is critical for proper chromosome segregation, and defects in its formation results in asymmetrical chromosome segregation. Thus, spindle formation is tightly regulated by multiple molecules (25). We recently reported that EphA2 plays a role in M-phase progression via participation in mitotic spindle formation. The direct target of EphA2 signaling is cortical rigidity. EphA2 phosphorylation at Ser897 maintained cortical rigidity, via the EphA2/Ephexin4/RhoG pathway, leading to proper formation of the mitotic spindle. The CDK1/MEK/ERK/RSK pathway mediates EphA2 phosphorylation at Ser897, and this phosphorylation is required for interaction with Ephexin4 and RhoG activation; phosphorylation at Tyr588—an index of kinase activity of EphA2— is then reduced. Here we demonstrate that EphA2 kinase activity is dispensable for proper M-phase progression.
While wild-type EphA2 was localized at the plasma membrane, it was also detected in the cytoplasm of M-phase cells (Fig. 3). These intracellular localizations were less in the kinase-dead EphA2 mutant than that in the wild-type. Since ligand-dependent activation of EphA2 kinase activity promotes its own endocytosis (24), loss of kinase activity may suppress its own endocytosis and thereby maintain its plasma membrane localization. Although we have, thus far, been unable to determine where Ser897-phosphorylated EphA2 is localized, we predict that phosphorylated EphA2 at Ser897 is localized at the plasma membrane, in which EphA2 regulates cortical rigidity. It was reported that ligand binding to EphA2 reduces phosphorylation at Ser897 (13) by recruitment of RasGAP to the plasma membrane and suppresses of the MAPK pathway during interphase (26). Thus, although further experiments will be required to address this issue, we speculate that decrease in EphA2 kinase activity may facilitate the plasma membrane localization of phosphorylated EphA2 at Ser897 to regulate cortical rigidity.
How does EphA2 lose its kinase activity within the M phase? A plausible possibility is that the loss of cell-cell interaction(s) – i.e. loss of ligand binding – decreases in kinase activity. The remaining kinase-active EphA2 would be endocytosed, since endocytosis is enhanced at mitotic entry (27), and EphA2 kinase activity activates its own endocytosis (24). However, even in the absence of ligand, EphA2 can dimerize and, thereby, increase Tyr phosphorylation (28). Thus, escape from the dimeric state may also contribute to reduce kinase activity at mitotic entry. It is noteworthy that lipid composition regulates dimerization of receptor-type tyrosine kinase (RTK). For example, the glycosphingolipid GM3 ganglioside inhibits RTK dimerization at the plasma membrane (29, 30). Cholesterol depletion causes ligand-independent EGFR activation via dimerization (31). Interestingly, lipid composition fluctuates in a cell cycle-dependent manner, and depletion of some lipid synthetic enzymes causes cell division defects (32). Cell surface cholesterol level is the highest at metaphase throughout the entire cell cycle (33), raising the possibility that elevated cholesterol levels may inhibit EGFR dimerization within the M phase. It is interesting to investigate whether elevated cholesterol levels prevent EphA2 dimerization. Since the dimerization-deficient EphA2 mutant shows decreased tyrosine phosphorylation (Tyr772) (28), elevated cholesterol would decrease EphA2 kinase activity, in addition to loss of ligand binding. In support, the dimerization-deficient EphA2 mutant shows increased Ser897 phosphorylation in interphase (28). We, therefore, speculate that stabilization of the monomeric EphA2 state, possibly by loss of ligand binding and elevation of cholesterol, may suppress its kinase activity, reinforcing EphA2 phosphorylation at Ser897.
Cholesterol depletion inhibits cytokinesis completion, resulting in polyploid cells (34). We predict that cholesterol depletion might maintain a dimerization state of EphA2 at mitotic entry. In addition, cholesterol depletion decreases CDK1 kinase activity (35). Since CDK1 triggers EphA2 phosphorylation through the CDK1/MEK/ERK/RSK pathway, cholesterol depletion leads toinsufficient EphA2 phosphorylation at Ser897 via decreased CDK1 activity and EphA2 dimerization. Insufficient EphA2 phosphorylation at Ser897 might be the reason of abnormal cell division in cholesterol-depleted cells. Cortical rigidity is regulated spatio-temporally and varies during M-phase progression. Many cell types lose their adhesions to the extracellular matrix, upon mitotic entry via Rap1 GTPase inactivation (36), leading to cell shape changes into near-spherical shapes. Cell stiffness is increased by the cortical actomyosin network to counter osmotic swelling (37). At telophase, where segregating chromosomes are positioned close to the poles, the cortex in close proximity to the two poles is relaxed – which is known as polar relaxation – via ERM protein dephosphorylation at the polar cortex and reduced actin fibers (38, 39). On the contrary, the cortex at the cleavage furrow shows tremendous stiffness during cytokinesis. Even if EphA2 phosphorylation at Ser897 would contribute to the cortical rigidity following, as well as prior to, anaphase onset, it would be uneven throughout the plasma membrane. Since CDK1 is inactivated following anaphase onset, localization of active RSK, an upstream kinase of phosphorylation at Ser897, may be responsible for where EphA2 is phosphorylated and regulate cortical rigidity. It was reported that RSK localizes to the spindle midzone (40).
Considering that spindle midzone-localized centralspindlin recruits RhoGEF Ect2, and stimulates RhoA at the equatorial plasma membrane (41), midzone-localized RSK would phosphorylate EphA2 localized at the equatorial plasma membrane; phosphorylated EphA2 at Ser897 may regulate cortical rigidity at the cleavage furrow during cytokinesis. However, exocytosis is activated, and endosomes are fused with the plasma membrane at the cleavage furrow, thereby locally increasing EphA2 density if EphA2 is recycled to the cleavage furrow. Increased receptor density may facilitate transient dimer formation and receptor activation (42). If kinase activity suppresses EphA2 phosphorylation at Ser897 by recruiting RasGAP (26), EphA2 phosphorylation at Ser897 may be suppressed at the cleavage furrow. It would be necessary to investigate localization of phosphorylated EphA2 at Ser897, and that of EphA2 with kinase activity, during M-phase progression, which may provide new insight into M-phase regulation by EphA2.
Funding
This work was supported, in part, by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (Nos. 19K07055, and 16K08253), grants from the Promotion and Mutual Aid Corporation for Private Schools of Japan (Kyoto Pharmaceutical University and Chiba University).
Conflicts of interest
The authors declare that they have no competing interests.
Author Contributions:
YK and YN designed the study and wrote the paper. YK, KK, YT, MI, MO, YI, RY, YS, YN performed experiments. All authors reviewed the results and approved the final version of the manuscript.
References
1. Jallepalli, P. V. and Lengauer, C. (2002) Chromosome segregation and cancer: cutting through the mystery. Nat. Rev. Cancer 1, 109–117
2. Wittmann, T., Hyman, A., and Desai, A. (2001) The spindle: A dynamic assembly of microtubules and motors. Nat. Cell Biol. 3
3. Musacchio, A. and Salmon, E. D. (2007) The spindle-assembly checkpoint in space and time.
Nat Rev Mol Cell Biol 8, 379–393
4. Grill, S. W. and Hyman, A. A. (2005) Spindle positioning by cortical pulling forces. Dev. Cell
8, 461–465
5. Kunda, P., Pelling, A. E., Liu, T., and Baum, B. (2008) Moesin Controls Cortical Rigidity, Cell Rounding, and Spindle Morphogenesis during Mitosis. Curr. Biol. 18, 91–101
6. Carreno, S., Kouranti, I., Glusman, E. S., Fuller, M. T., Echard, A., and Payre, F. (2008) Moesin and its activating kinase Slik are required for cortical stability and microtubule organization in mitotic cells. J. Cell Biol. 180, 739–746
7. Zuo, Y., Oh, W., and Frost, J. A. (2014) Controlling the switches: Rho GTPase regulation during animal cell mitosis. Cell. Signal. 26, 2998–3006
8. Pasquale, E. B. (2005) Eph receptor signalling casts a wide net on cell behaviour. Nat. Rev.
Mol. Cell Biol. 6, 462–475
9. Egea, J. and Klein, R. (2007) Bidirectional Eph-ephrin signaling during axon guidance. Trends Cell Biol. 17, 230–238
10. Pasquale, E. B. (2008) Eph-Ephrin bidirectional signaling in physiology and disease. Cell 133, 38–52
11. Miao, H., Li, D. Q., Mukherjee, A., Guo, H., Petty, A., Cutter, J., Basilion, J. P., Sedor, J., Wu,
J., Danielpour, D., Sloan, A. E., Cohen, M. L., and Wang, B. (2009) EphA2 mediates
ligand-dependent inhibition and ligand-independent promotion of cell migration and invasion via a reciprocal regulatory loop with Akt. Cancer Cell 16, 9–20
12. Zhou, Y., Yamada, N., Tanaka, T., Hori, T., Yokoyama, S., Hayakawa, Y., Yano, S., Fukuoka, J., Koizumi, K., Saiki, I., and Sakurai, H. (2015) Crucial roles of RSK in cell motility by catalysing serine phosphorylation of EphA2. Nat. Commun. 6, 7679
13. Barquilla, A., Lamberto, I., Noberini, R., Heynen-Genel, S., Brill, L. M., and Pasquale, E. B. (2016) Protein kinase A can block EphA2 receptor-mediated cell repulsion by increasing EphA2 S897 phosphorylation. Mol. Biol. Cell 27, mbc.E16-01-0048
14. Hamaoka, Y., Negishi, M., and Katoh, H. (2016) EphA2 is a key effector of the MEK/ERK/RSK pathway regulating glioblastoma cell proliferation. Cell. Signal. 28, 937–945
15. Hiramoto-Yamaki, N., Takeuchi, S., Ueda, S., Harada, K., Fujimoto, S., Negishi, M., and Katoh, H. (2010) Ephexin4 and EphA2 mediate cell migration through a RhoG-dependent mechanism. J. Cell Biol. 190, 461–477
16. Kaibori, Y., Saito, Y., and Nakayama, Y. (2019) EphA2 phosphorylation at Ser897 by the Cdk1/MEK/ERK/RSK pathway regulates M-phase progression via maintenance of cortical rigidity. FASEB J. 33, 5334–5349
17. Johannessen, C. M., Boehm, J. S., Kim, S. Y., Thomas, S. R., Wardwell, L., Johnson, L. A., Emery, C. M., Stransky, N., Cogdill, A. P., Barretina, J., Caponigro, G., Hieronymus, H., Murray, R. R., Salehi-Ashtiani, K., Hill, D. E., Vidal, M., Zhao, J. J., Yang, X., Alkan, O., Kim, S., Harris, J. L., Wilson, C. J., Myer, V. E., Finan, P. M., Root, D. E., Roberts, T. M., Golub, T., Flaherty, K. T., Dummer, R., Weber, B. L., Sellers, W. R., Schlegel, R., Wargo, J. A., Hahn, W. C., and Garraway, L. A. (2010) COT drives resistance to RAF inhibition through MAP kinase
pathway reactivation. Nature 468, 968–972
18. Yang, X., Boehm, J. S., Yang, X., Salehi-Ashtiani, K., Hao, T., Shen, Y., Lubonja, R., Thomas,
S. R., Alkan, O., Bhimdi, T., Green, T. M., Johannessen, C. M., Silver, S. J., Nguyen, C., Murray, R. R., Hieronymus, H., Balcha, D., Fan, C., Lin, C., Ghamsari, L., Vidal, M., Hahn, W. C., Hill, D. E., and Root, D. E. (2011) A public genome-scale lentiviral expression library of human ORFs. Nat. Methods 8, 659–661
19. Wang, Y. jie, Ota, S., Kataoka, H., Kanamori, M., Li, Z. you, Band, H., Tanaka, M., and Sugimura, H. (2002) Negative regulation of EphA2 receptor by Cbl. Biochem. Biophys. Res. Commun. 296, 214–220
20. Okumura, D., Hagino, M., Yamagishi, A., Kaibori, Y., Munira, S., Saito, Y., and Nakayama, Y. (2018) Inhibitors of the VEGF receptor suppress HeLa S3 cell proliferation via misalignment of chromosomes and rotation of the mitotic spindle, causing a delay in M-phase progression. Int. J. Mol. Sci. 19, 4014
21. Horiuchi, M., Kuga, T., Saito, Y., Nagano, M., Adachi, J., Tomonaga, T., Yamaguchi, N., and Nakayama, Y. (2018) The tyrosine kinase v-Src causes mitotic slippage by phosphorylating an inhibitory tyrosine residue of Cdk1. J. Biol. Chem. 293, 15524–15537
22. Kakihana, A., Oto, Y., Saito, Y., and Nakayama, Y. (2019) Heat shock-induced mitotic arrest requires heat shock protein 105 for the activation of spindle assembly checkpoint. FASEB J. 33, 3936–3953
23. Ifuji, A., Kuga, T., Kaibori, Y., Saito, Y., and Nakayama, Y. (2017) A novel immunofluorescence method to visualize microtubules in the antiparallel overlaps of microtubule-plus ends in the anaphase and telophase midzone. Exp. Cell Res. 360, 347–357
24. Boissier, P., Chen, J., and Huynh-Do, U. (2013) Epha2 signaling following endocytosis: Role
of tiam1. Traffic 14, 1255–1271
25. Kline-Smith, S. L. and Walczak, C. E. (2004) Mitotic spindle assembly and chromosome segregation: refocusing on microtubule dynamics. Mol Cell 15, 317–327
26. Miao, H., Wei, B.-R., Peehl, D. M., Li, Q., Alexandrou, T., Schelling, J. R., Rhim, J. S., Sedor,
J. R., Burnett, E., and Wang, B. (2001) Activation of EphA receptor tyrosine kinase inhibits the Ras/MAPK pathway. Nat. Cell Biol. 3
27. Boucrot, E. and Kirchhausen, T. (2007) Endosomal recycling controls plasma membrane area during mitosis. Proc. Natl. Acad. Sci. 104, 7939–7944
28. Singh, D. R., Ahmed, F., King, C., Gupta, N., Salotto, M., Pasquale, E. B., and Hristova, K. (2015) EphA2 Receptor Unliganded Dimers Suppress EphA2 Pro-tumorigenic Signaling. J. Biol. Chem. 290, 27271–27279
29. Zhou, Q., Hakomori, S. I., Kitamura, K., and Igarashi, Y. (1994) G(M3) directly inhibits tyrosine phosphorylation and de-N-acetyl-G(M3) directly enhances serine phosphorylation of epidermal growth factor receptor, independently of receptor-receptor interaction. J. Biol. Chem. 269, 1959–1965
30. Coskun, U., Grzybek, M., Drechsel, D., and Simons, K. (2011) Regulation of human EGF receptor by lipids. Proc. Natl. Acad. Sci. 108, 9044–9048
31. Chen, X. and Resh, M. D. (2002) Cholesterol depletion from the plasma membrane triggers ligand-independent activation of the epidermal growth factor receptor. J. Biol. Chem. 277, 49631–49637
32. Atilla-Gokcumen, G. E., Muro, E., Relat-Goberna, J., Sasse, S., Bedigian, A., Coughlin, M. L., Garcia-Manyes, S., and Eggert, U. S. (2014) Dividing cells regulate their lipid composition and localization. Cell 156, 428–439
33. Santos, A. J. M., Meinecke, M., Fessler, M. B., Holden, D. W., and Boucrot, E. (2013) Preferential invasion of mitotic cells by Salmonella reveals that cell surface cholesterol is maximal during metaphase. J. Cell Sci. 126, 2990–2996
34. Fernández, C., Lobo, M. D. V. T., Gómez-Coronado, D., and Lasunción, M. A. (2004) Cholesterol is essential for mitosis progression and its deficiency induces polyploid cell formation. Exp. Cell Res. 300, 109–120
35. Martínez-botas, J., Suárez, Y., Ferruelo, A. J., Gómez-coronado, D., and Lasunción, M. A. (1999) Cholesterol starvation decreases P34 cdc2 kinase activity and arrests the cell cycle at G2. FASEB J. 13, 1359–1370
36. Dao, V. T., Dupuy, A. G., Gavet, O., Caron, E., and de Gunzburg, J. (2009) Dynamic changes in Rap1 activity are required for cell retraction and spreading during mitosis. J. Cell Sci. 122, 2996–3004
37. Ramkumar, N. and Baum, B. (2016) Coupling changes in cell shape to chromosome segregation. Nat. Rev. Mol. Cell Biol. 17, 511–521
38. Sedzinski, J., Biro, M., Oswald, A., Tinevez, J. Y., Salbreux, G., and Paluch, E. (2011) Polar actomyosin contractility destabilizes the position of the cytokinetic furrow. Nature 476, 462– 468
39. Rodrigues, N. T. L., Lekomtsev, S., Jananji, S., Kriston-Vizi, J., Hickson, G. R. X., and Baum,
B. (2015) Kinetochore-localized PP1-Sds22 couples chromosome segregation to polar relaxation. Nature 524, 489–492
40. Nam, H. J., Lee, I. J., Jang, S. H., Bae, C. D., Kwak, S. J., and Lee, J. H. (2014) P90 ribosomal S6 kinase 1 (RSK1) isoenzyme specifically regulates cytokinesis progression. Cell. Signal. 26, 208–219
41. Green, R. A., Paluch, E., and Oegema, K. (2012) Cytokinesis in animal cells. Annu. Rev. Cell Dev. Biol. 28, 29–58
42. Sawano, A., Takayama, S., Matsuda, M., Ro-3306 and Miyawaki, A. (2002) Lateral Propagation of EGF Signaling after Local Stimulation Is Dependent on Receptor Density. Dev. Cell 3, 245–257