Aurora A Kinase amplifies a midzone phosphorylation gradient to promote high-fidelity cytokinesis.
Mitosis is the process by which a cell divides its duplicated genetic material into two daughter cells. Equal segregation of the DNA is required for cell viability, and, thus, it is critical that this process is orchestrated flawlessly every time. Cytokinesis is achieved by an actin-myosin contractile ring that physically divides the cell into two daughter cells following separation of the sister chromatids during anaphase. Proper positioning of the contractile ring and, hence. the cleavage furrow is critically important for cytokinesis, but current understanding of the cues that spatially determine where the furrow forms is incomplete.
The aurora family of proteins is a group of mitotic serine/threonine kinases that regulate many aspects of cell division (Carmena et al., 2009: Hochegger et al., 2013). Aurora A kinase (AAK) and aurora B kinase (ABK), the two members found in Drosophila melanogaster, have highly conserved C-terminal kinase domains that phosphorylate many of the same substrates. However, their cellular functions are distinct and dictated by their different cellular localizations, which are determined by the divergent N-terminal regulatory domain (Li et al., 2015). Aurora B kinase is part of a multi-subunit protein complex called the chromosomal passenger complex (CPC). Prior to anaphase, the CPC is highly enriched at the inner centromere, and ABK activity contributes to spindle and kinetochore assembly, spindle assembly checkpoint signaling, and error correction (Tanaka et al., 2002; Emanuele et al., 2008; Lampson and Cheeseman, 2011; Moutinho-Pereira et al., 2013). Aurora B kinase activity is essential for proper cytokinesis (Adams et al., 2001; Echard et al., 2004; Eggert et al., 2004; Smurnyy et al., 2010) and, at onset of anaphase, the CPC complex relocalizes to the spindle midzone (Adams et al., 2000), an area comprised of stable, overlapping microtubules (MTs). The midzone-associated CPC generates an activity gradient (Fuller et al., 2008; Tan and Kapoor, 2011) that helps ensure complete sister chromatid separation (Afonso et al., 2014), and that is proposed to contribute to positioning of the cleavage furrow (Adams et al., 2000; Terada, 2001; Tan and Kapoor, 2011). Aurora A kinase is required for centrosome maturation, separation, and function (Glover et al., 1995; Hannak et al., 2001; Giet et al., 2002), proper kinetochore-MT attachment (Bakhoum et al., 2014; Barisic et al., 2014; Chmatal et al., 2015; Ye et al., 2015), chromosome segregation (Hegarat et al., 2011), MT nucleation (Scrofani et al., 2015), and robust assembly of midzone MTs (Lioutas and Vernos, 2013; Reboutier et al., 2013).
While it is clear that MTs are required for positioning of the cleavage furrow (Rappaport, 1971), MT-dependent and MT-independent mechanisms have been proposed to contribute to faithful cytokinesis (Alsop and Zhang, 2003; Canman et al., 2003; Foe and von Dassow, 2008; Murthy and Wadsworth, 2008; Nguyen et al., 2014; Field et al., 2015; Rodrigues et al., 2015). Broadly speaking, two major models are evoked to explain how cells define where the cleavage furrow is positioned during cytokinesis. Equatorial stimulation holds that a positive signal promotes cortical contractility through activation of myosin in between the spindle poles, while polar and/or astral relaxation posits that a signal coming from the vicinity of the spindle poles inhibits contractility, causing the polar cortex to relax. While there is evidence that astral MTs in the polar region contribute to polar relaxation (Murthy and Wadsworth, 2008), recent work suggests that the signal does not require astral MTs, but rather is mediated by a kinetochore-derived phosphatase gradient (Rodrigues et al., 2015). Equatorial stimulation depends on signals from both the midzone, which is enriched for essential cytokinesis regulators, including the CPC (Glotzer, 2005; Nguyen et al., 2014), and an "astral" stimulation signal that requires a subpopulation of stable MTs in the vicinity of the furrow site (Canman et al., 2003; Foe and von Dassow, 2008; Field et al., 2015). Thus, three potentially redundant pathways regulate the formation and positioning of the cleavage furrow: 1) polar relaxation, 2) midzone stimulation, and 3) astral stimulation. Here, we apply a variety of live-cell imaging techniques in Drosophila S2 cells to explore the contribution of each of these pathways to successful cytokinesis.
Materials and Methods
Drosophila S2 cell culture
All cell lines were grown in Schneider's medium (Life Technologies, Carlsbad, CA), supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 0.5X antibiotic/antimycotic cocktail (Sigma-Aldrich, St. Louis, MO), and maintained at 25 [degrees]C. All cell lines were generated by transfecting the plasmid with Effectene Transfection Reagent system (Qiagen, Hilden, Germany), following the manufacturer's protocol. Expression of the proteins was checked by fluorescence microscopy. To select the cell expressing the constructs, cells were split in the presence of Blasticidin S HC1 (Thermo Fisher Scientific, Waltham, MA) and/or Hygromycin (Sigma-Aldrich). (Spaghetti squash (Dm MRLQ-GFP, a mCherry-[alpha]-tubulin cell line, was a generous gift from Eric Griffis, University of Dundee.)
A soluble, Forster resonance energy transfer (FRET)-based aurora phosphorylation sensor was previously generated (Ye et al., 2015). To target this sensor to MTs, Tau (CG45110) was amplified from complementary DNA (cDNA) with flanking SpeI sites and inserted into the soluble reporter construct via Gibson assembly (Gibson et al., 2009). The Tau sequence was inserted downstream of the centromere protein C (CENP-C) promoter to drive expression of the reporter.
RNA interference (RNAi) experiments
DNA templates for subito (CGI2298), aurora A kinase (CG3068), Sds22 (CG5851), and centrosomin (CNN) (CG4832) were produced to contain ~500 base pairs (bps) of complementary sequence flanked by T7 promoter sequence. Double-stranded RNAs (dsRNAs) were synthesized from the DNA templates overnight at 37 [degrees]C, using the T7 RiboMax Express Large Scale RNA Production System (Promega Corp., Madison, WI), following the manufacturer's protocol. For RNAi experiments, media was aspirated off semi-adhered cell at 25% confluency and replaced with 1 ml of serum-free Schneider's medium containing 20 [micro]g of dsRNA. After 1 h, 1 ml of fresh Schneider's media plus FBS was added to the wells and incubated for 2 days at 24 [degrees]C. See Table 1 for primer sequences.
Drosophila S2 cells were allowed to adhere to acid-washed Concanavalin A (Sigma-Aldrich)-coated coverslips, then treated with 5 [micro]mol [1.sup.-1] MG132 and 125 nmol [1.sup.-1] MLN8237 (MLN) (Selleck Chemicals, Houston, TX) or dimethyl sulfoxide (DMSO) as control for 1 h before being quickly rinsed with BRB80 buffer, and then fixed with 10% paraformaldehyde for 10 minutes. Cells were then permeabilized with phosphate-buffered saline (PBS) containing 1% Triton X-100 for 8 min, rinsed 3 times with PBS plus 0.1% Triton X-100, and blocked with boiled donkey serum for 60 min. All primary antibodies were diluted in boiled donkey serum. Anti-Phospho-aurora A/B/C and Phosphoaurora A (Cell Signaling Technology, Danvers, MA) was used at a concentration of 1:1000, and anti-tubulin antibody (DMla; Sigma-Aldrich) at 1:1000. All secondary antibodies (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) were diluted in boiled donkey serum at 1:200. After secondary treatment, coverslips were washed 2 times with PBS plus 0.1% Triton X-100, followed by incubation with 4',6'-diamidino-2-phenylindole (DAPI) at a concentration of 1:1000 for 5 min, and 2 additional washes. Coverslips were sealed in mounting media containing 20 mmol [1.sup.-1] Tris, pH 8.0, 0.5% N-propyl gallate, and 90% glycerol.
Three- or four-color Z-series consisting of ~30 planes at 0.2-[micro]m intervals were acquired for green fluorescent protein (GFP), Rhodamine, Cy5 (where appropriate), and DAPI channels. Fluorescence intensities were obtained by drawing larger and small regions manually around the maximum-intensity projection of the Z-series images. To obtain the ratio intensities, regions were drawn manually on the tubulin channel and transferred to a phosphorylated ABK (pABK) channel. The following equations were used: background signal = (integrated fluorescence intensity of big area--integrated fluorescence intensity of small area)/(big area--small area). Total intensity = integrated fluorescence intensity of small area--(background signal X small area).
Forster resonance energy transfer (FRET)
Cells were treated with 125 nmol [1.sup.-1] MLN8237 (MLN) or DMSO as control, diluted 1:1000, in a tissue culture dish for 1 h. Cells were allowed to adhere to acid-washed, Concanavalin A (Sigma-Aldrich)-coated coverslip (Corning, Inc., Corning. NY) for exactly 1 h, then assembled in a rose chamber containing Schneider's medium with 125 nmol [1.sup.-1] MLN or DMSO, and subjected to imaging at 25 [degrees]C. Cells were imaged for a maximum of 1 h on an eclipse Ti-E inverted microscope (Nikon, Tokyo, Japan) equipped with an iXON EMCCD camera (Andor Technology, Belfast, U.K.), using a 100 x 1.4 NA Plan Apo violet-corrected series differential interference contrast objective (Nikon). Metamorph software (Molecular Devices, Sunnyvale, CA) was used to control the imaging system.
Mitotic cells were found in the red fluorescent protein (RFP) channel, and images of the best focal plane were acquired in RFP, CFP, and YFP, and FRET (CFP excitation and YFP emission) channels with equal exposure times. The ratios of the fluorescence intensities of FRET to the mTurqoise2 fluorescence intensity (CFP spectral channel) were obtained by drawing larger and smaller regions in Metamorph around the central spindle in the FRET images, then transferred to the CFP image. The background signal and the total intensity equations (see previous section) were used.
A total of 10 [micro]g of protein was loaded into a 10% SDS-PAGE gel, run out, and transferred to a nitrocellulose membrane on the Trans-Blot Turbo transfer system (Bio-Rad Laboratories. Inc., Hercules, CA), using the manufacturer's preprogrammed 7-min "MIXED MW" protocol. All antibodies were diluted in Tris-buffered saline (TBS) with 0.1% Tween and 5% milk. The membrane was first incubated with anti-aurora A serum (gift of Marcin Przewloka and David Glover, University of Cambridge) at a 1:5000 dilution, followed by DM1 [alpha] at 1:5000 dilution. Rabbit and mouse horseradish peroxidase (HRP) secondary antibodies (Jackson ImmunoResearch Laboratories, Inc.), diluted at 1:5000, were used in conjunction with their respective primaries and imaged with a GBox system controlled by GeneSnap software (Syngene, Cambridge, U.K.).
Total internal reflection fluorescence (TIRF) microscopic and spinning disk confocal imaging
Cells were allowed to adhere to Concanavalin A-treated coverslips, and mounted into a rose chamber. Cells were imaged on a TIRF-Spinning Disk system assembled on an Eclipse Ti-E inverted microscope (Nikon), equipped with a Borealis (Andor Technology, Ltd., Belfast, U.K.) retrofitted CSU-10 (Yokogawa Electric Corp., Tokyo, Japan) spinning disk head and two ORCA-Flash4.0 LT Digital CMOS (Hamamatsu Corp., Bridgewater, NJ), using a 100X 1.49 NA Apo differential interference contrast objective (Nikon). Metamorph software was used to control the imaging system. Dual TIRF (myosin regulatory light chain-green fluorescent protein; MRLC-GFP) and widefield (mCherry-[alpha]-tubulin) images were acquired at 1-min intervals. To quantify the myosin dynamic, three 10-[pixel.sup.2] boxes were placed in the pole region or the equator and averaged for each cell. To compare myosin accumulation at the equator and depletion at the pole, half-maximum change at the equator was normalized to be time 0 for each cell, and 10 before-and-after time points were reported. For correlative TIRF-spinning disk confocal imaging, cells were followed by time-lapse TIRF imaging; once the cell had assembled a contractile ring (based on MRLC imaging) the system was switched to spinning disk mode and 0.2-[micro].m confocal z-sections were taken. All quantifications and 3D-reconstructions were done using Metamorph software. Microtubule dynamics in interphase cells were manually tracked using the MTrackJ plug-in (Meijering et al., 2012) in ImageJ (Schneider et al., 2012). The catastrophe and rescue frequencies are determined by the number of catastrophe events divided by the total amount of time spent polymerizing and the number of rescue events divided by the total amount of time spent depolymerizing, respectively.
To determine if aurora A kinase (AAK) plays a post-metaphase role in Drosophila S2 cell division, AAK was knocked down by RNA interference (RNAi), and microtubule (MT) intensity in the spindle midzone during late anaphase was quantified (Fig. 1A-C). Consistent with previous reports in other cell types (Lioutas and Vernos, 2013; Reboutier et al., 2013), a 34% decrease in MT midzone population was observed in AAK RNAi-depleted cells compared to control cells. Similar to the AAK-depleted conditions, cells treated with a 125 nmol [1.sup.-1] concentration of the inhibitor, MLN8237--which specifically inhibits AAK in Drosophila S2 cells (Ye et al., 2015), resulting in a loss of phosphorylated AAK (pAAK) from centrosomes (Fig. 1G)--exhibited a 40% decrease in midzone MT intensity compared to DMSO-treated cells (Fig. 1F-H). The observed decrease in midzone MT density in late anaphase following AAK depletion or chemical inhibition may be a result of inhibiting aurora B kinase (ABK) as it relocalizes to the overlapping central spindle MTs after anaphase. To examine if the observed results were due to mislocalizing or inhibiting ABK, levels of active phosphorylated ABK (pABK) at midzones were quantified in control and AAK-inhibited conditions. When normalized to the control levels, pABK did not exhibit a significant change at the midzone in either AAK knockdown (Fig. 1D) or MLN8237-treated cells (Fig. 1I) compared to the control conditions. Thus, while the MT intensities were significantly decreased relative to controls in the absence of AAK activity, the total level of midzone pABK was unaffected by either chemical inhibition of AAK or its depletion. It is noteworthy that while the total amount of pABK did not change, when normalized to the midzone tubulin signal., the amount of midzone pABK per MT increased significantly compared to the controls in both AAK-depleted and MLN8237-treated cells (Fig. 1E, J).
The midzone aurora kinase activity gradient was next investigated. A FRET-based aurora phosphorylation sensor (Fuller et al., 2008; Ye et al., 2015) was generated and targeted to the MTs by fusing it to the MT-associated protein Tau (full-length protein). Expression of the Tau-FRET reporter did not obviously bundle MTs or alter interphase MT dynamics relative to cells in the same imaging chamber that were not expressing the reporter (Table 2). The sensor is designed such that the FRET ratio decreases when it is phosphorylated by aurora kinases (Fig. 2A). Cells treated with 125 nmol [1.sup.-1] MLN8237, which specifically inhibits AAK activity and has the same effects as AAK knockdown (Fig. 1), exhibited a 5% higher FRET emission ratio (Fig. 2B, C). This indicates that substrates are less phosphorylated by the midzone aurora kinase activity gradient when AAK activity is inhibited in S2 cells--despite having normal levels (or even higher levels per MT) of midzone pABK. This finding is consistent with previous work that identified transforming acidic coiled-coil containing protein 3 (TACC3) and p150Glued as midzone AAK substrates in human cells that were less phosphorylated and mislocalized, respectively, following AAK inhibition (Lioutas and Vernos. 2013; Reboutier et al., 2013). Importantly, the behavior of a non-phosphorylatable FRET sensor was unaffected by MLN8237 (Fig. 2D). We posit that the measured decrease in kinase activity must be due to inhibition of AAK, because pABK levels are actually higher per MT in this condition (Fig. 1E, J). Thus, we conclude that AAK amplifies a midzone aurora kinase activity gradient.
Where is AAK functionally relevant post-anaphase, during midzone assembly and cytokinesis? While we and others (Berdnik and Knoblich, 2002; Giet et al., 2002; Ye et al., 2015) have described the localization pattern of AAK through Drosophila cell division, we more closely examined AAK relative to MTs by imaging cells co-expressing mCherry-tagged AAK and GFP-[alpha]-tubulin. Aurora A kinase was highly enriched at centrosomes throughout mitosis and localized to spindle MTs to varying degrees, depending on the level of overexpression, with a tendency to be enriched near spindle poles in low to moderately expressing cells. In cells with the highest levels of AAK overexpression, a slight enrichment of AAK was sometimes observed in the vicinity of kinetochores/centromeres, although not to the extent previously seen in mouse oocytes overexpressing AAK (Chmatal et al., 2015). Aurora A kinase remained enriched at centrosomes throughout anaphase and during cytokinesis, although--even in high-expressing cells--AAK only faintly associated with MTs but was not enriched on midzone MTs (Fig. 2E and Supplementary video 1. http://www.biolbull.org/content/supplemental). Aurora B kinase, on the other hand, is highly enriched on midzone MTs, and. when visualized by total internal reflection fluorescence (TIRF) microscopy (Vale et al., 2009), ABK also exhibits MT tip-tracking behavior at onset of anaphase. We next employed TIRF imaging of cells co-expressing mCherry-tagged AAK and GFP-[alpha]-tubulin, to determine if AAK exhibited behavior similar to that of ABK (Fig. 2F and Supplemental video 2, http://www.biolbull.org/content/supplemental). While AAK was evident at a centrosome that entered the TIRF field during anaphase, it neither tip-tracked nor became enriched on midzone or astral MTs in the cortical region. Thus, we posit that the midzone-derived aurora kinase activity gradient is amplified largely by a soluble rather than MT-associated pool of AAK.
Our data indicate that AAK amplifies a midzone phosphorylation gradient to promote robust midzone assembly in Drosophila S2 cells. To examine whether the observed defects in midzone assembly impacted furrow formation or assembly of the actin-myosin contractile ring during cytokinesis. myosin dynamics were visualized in living cells expressing GFP-tagged myosin regulatory light chain (MRLC: spaghetti squash in Drosophila) and mCherry-[alpha]-tubulin by TIRF microscopy. In this image-based assay, cells are allowed to adhere to Concanavalin A, which prevents successful completion of cytokinesis but allows for impressive visualization of myosin at the cortex after anaphase onset (Vale et al., 2009). In accord with previous TIRF imaging of this cell line (Vale et al., 2009), MRLC was lost from the equator and became enriched at the site of the cleavage furrow following anaphase onset. There was no significant difference between DMSO- and MLN8237-treated cells in myosin dynamics at the polar relative to the equatorial regions (Fig. 2G, H). We next employed correlative TIRF-spinning disk confocal imaging on MLN-treated cells to better observe MRLC organization. A field of cells that contained a metaphase cell was imaged by TIRF microscopy for 25 min to capture the metaphase cell progress through anaphase, enrich MRLC at the equator, and assemble a contractile ring (Supplementary video 3, http://www.biolbull.org/content/supplemental). At that point, the field was imaged by confocal spinning disk microscopy (Fig. 21), 0.2-[micro]m confocal z-sections were acquired (Supplementary video 4, http://www.biolbull.org/content/supplemental), and 3D reconstructions were generated for each of the three cells in the field of view (Fig. 2 J-L). The cell that progressed from metaphase to anaphase had formed a cleavage furrow and assembled a clear ring of MRLC that deformed the upper part of the midzone (Fig. 2J and Supplementary video 5, http://www.biolbull.org/content/supplemental). A neighboring cell possessed a midbody. and a small ring of MRLC could be seen around the midbody (Fig. 2K and Supplementary video 6. http://www.biolbull.org/content/supplemental). Since the cells had been treated with MLN for several hours at this point, this cell very likely had progressed through cytokinesis in the absence of AAK activity. An interphase cell with MRLC highly enriched at the cortex, contacting the coverslip, was also present in the field of view (Fig. 2L and Supplementary video 7. http://www.biolbull.org/content/supplemental). Taken together, the data indicate that loss of AAK activity does not dramatically alter MRLC dynamics or organization during cytokinesis.
It is widely accepted that aurora B kinase activity is required for cytokinesis. In support of this theory, numerous key cytokinesis regulators have been shown to be ABK targets (Adams et al., 2001; Echard et al., 2004; Eggert et al., 2004; Guse et al., 2005; Smurnyy et al., 2010; Nunes Bastos et al., 2013). Similar to the phenotype associated with compromising critical cytokinesis regulators in Drosophila, such as Pavarotti (Dm MKLP1) (Adams et al., 1998), large cells are prevalent following treatment with the Drosophila ABK-specific inhibitor Binucleine-2 (Bin2) (Eggert et al., 2004). However, as previously described (Moutinho-Pereira et al., 2013), we have observed that Bin2-treated cells fail to fully condense their chromosomes or assemble bipolar spindles. In addition, the spindle assembly checkpoint is severely compromised, leading to rapid mitotic exit in the absence of any semblance of normal mitotic timing or organization. Furthermore, while failure in cytokinesis following depletion of bona fide regulators such as Pavarotti results in cells with two comparably sized nuclei, Bin2-treated cells rarely exhibit the typical binutieate phenotype. Rather, these cells have a single, large nucleus or large nuclei with numerous smaller nuclei. Similar to previous observations (Afonso et al., 2014), the addition of Bin2 in early anaphase led to rapid mitotic exit and the formation of a single, large nucleus (Fig 3A, Supplementary videos 8 and 9, http://www.biolbull.org/content/supplemental). The fact that this Bin2-treated anaphase cell did not complete cytokinesis was likely due to the consequences of rapid mitotic exit, as well as the positioning of the nucleus in the midzone position. Thus, the pleiotropic effects of globally inhibiting ABK render defects of cytokinesis difficult to interpret. To better isolate the contribution of midzone ABK (mABK), the Drosophila mitotic kinesin-like protein 2 (MKLP2) homologue, Subito, which localizes chromosomal passenger complex (CPC) to the midzone in numerous model systems, including Drosophila (Gruneberg et al., 2004; Cesario et al., 2006; Nguyen et al., 2014), was depleted by RNAi. Subito depletion resulted in a substantial reduction in midzone levels of pABK (Fig. 3B, C). Mislocalizing mABK by Subito depletion did not result in a significant increase in the number of binucleated cells compared to control conditions (Fig. 3D). While AAK contributes to the midzone phosphorylation gradient, the frequency of binucleated cells was not increased in MLN-treated cells relative to DMSO controls (Fig. 3D). We reasoned that if AAK amplifies the midzone activity gradient to promote cytokinesis, then inhibition of AAK in the absence of mABK would result in a higher frequency of cytokinesis failure. Indeed, MLN-treated cells that were depleted of mABK resulted in a ~4-fold increase in binucleate cells compared to controls (Fig. 3D). The additive effect of losing both AAK and mABK activities suggests that the two kinases work together to increase the fidelity of cytokinesis. In the double inhibited cells, the midzone aurora kinase activity gradient is compromised and, while a 4-fold increase in cytokinesis failure is not desirable, most cells manage to complete cytokinesis. Thus, redundant pathways must compensate for the loss of the midzone aurora phosphorylation gradient.
Previous work has shown that a kinetochore-based PP1-Sds22 phosphatase gradient mediates the polar relaxation signal (Kunda et al., 2012; Rodrigues et al., 2015). To inhibit the relaxation gradient and examine its functional redundancy with the equatorial stimulation signal from aurora kinases, Sds22 was depleted from cells alone and in combination with treatments that disrupt the midzone aurora kinase activity gradient. As previously reported (Rodrigues et al., 2015), depletion of Sds22 did not lead to a measurable increase in cytokinesis failure (Fig. 3E). Interestingly, cells lacking Sds22, and in which AAK was also inhibited, were no more prone to fail cytokinesis than control cells, while Sds22 depletion combined with loss of the midzone aurora kinase activity gradient were indistinguishable from Subito-depleted cells treated with 125 nmol [1.sup.-1] MLN8237 (Fig. 3D, E). Thus, Sds22-dependent polar relaxation is dispensable for proper cytokinesis, even in the absence of the auroramediated equatorial stimulation signal. Astral microtubules (MTs), presumably derived from centrosomes, have been thought to contribute to both polar relaxation (Murthy and Wadsworth, 2008) and equatorial stimulation (Rappaport, 1961). Centrosomin (CNN) was depleted in order to examine the contribution of centrosome-derived astral MTs to cytokinesis and their redundancy with the midzone aurora phosphorylation gradient. In CNN RNAi cells treated with either dimethyl sulfoxide (DMSO) or MLN8237, the percentage of binucleated cells was not different from control cells (Fig. 3F). As was the case for Sds22 depletion. CNN depletion in combination with Subito RNAi and 125 nmol [1.sup.-1] MLN8237 exhibited the same phenotype as cells lacking mABK and AAK activity (Fig. 3D-F). This suggested that centrosomal MTs do not make a major contribution to cytokinesis in Drosophila S2 cells, even when the midzone aurora kinase activity gradient is compromised.
The aurora family of kinases regulates many mitotic pathways, and their critical functions in early mitosis make it difficult to study their involvement in late anaphase and cytokinesis. Lioutas and Vernos (2013) and Reboutier et al. (2013) have reported the importance of aurora A kinase (AAK) in the formation of a central spindle by regulating microtubule (MT) growth through phosphorylation of TACC3 and p150Glued, respectively. Here, we report that AAK activity also contributes to robust midzone MT assembly and to phosphorylation of midzone substrates in Drosophila S2 cells. Interestingly, total midzone levels of phosphorylated aurora B kinase (pABK) were normal in AAK-inhibited cells and actually higher when normalized to MTs despite having less robust midzones. Thus, there may be a regulatory mechanism that ensures adequate levels of chromosomal passenger complex (CPC) enrichment at overlapping midzone MTs that buffer against variations in MT density.
The experiments conducted in this study have allowed us to dissect the contributions of equatorial stimulation and polar relaxation pathways (Fig. 3G). While depletion of MKLP2 and Subito leads to loss of mABK in human and Drosophila cells, respectively, their depletion leads to a significantly higher incidence of cytokinesis failure in human cells (Zhu et al., 2005; Kitagawa et al., 2013, 2014). Since MKLP2-depleted cells fail during the late stages of cytokinesis (Zhu et al., 2005), this discrepancy may be due to differences in abscission mechanisms between human and Drosophila. Interestingly, the fact that the cleavage furrow ingresses between segregating chromosomes in MKLP2-depleted human cells suggests that, as in Drosophila S2 cells, midzone ABK (mABK) is likely also dispensable for spatially positioning the cleavage furrow in human cells. In Drosophila, both midzone-derived and astral stimulation signals likely contribute to equatorial stimulation, while recent work (Rodrigues et al., 2015) suggests that polar relaxation is mediated by a phosphatase gradient emanating from kinetochores. We have shown that a midzone-based aurora kinase activity gradient that requires both mABK and a predominantly soluble pool of AAK contributes to the fidelity of cytokinesis. However, redundant and/or dominant pathways must exist, since a majority of cells lacking the midzone aurora activity gradient successfully complete cytokinesis. Polar relaxation signals from the phosphatase gradient or by centrosomal MTs are neither dominant--since single depletions of Sds22 or CNN do not lead to cytokinesis defects--nor redundant to the midzone aurora kinase activity gradient; the effects of depleting CNN or Sds22 combined with loss of the aurora phosphorylation gradient was identical to removing the aurora activity gradient alone. We infer from these data that the dominant pathway for spatially defining furrow formation in Drosophila S2 cells is the astral stimulation pathway.
Interestingly, astral stimulation, in this case, does not require centrosome-derived MTs, because CNN depletion has no effect on cytokinesis. We acknowledge that the concept of "astral" MTs without centrosomes seems counter-intuitive; however, we propose that, in Drosophila S2 cells, what is historically referred to as the astral stimulation pathway is actually mediated by a population of critically important "furrow MTs" that do not require centrosomes to assemble. Furthermore, since inhibiting Pavarotti (Dm MKLP1) in Drosophila cells leads to complete failure in cytokinesis (Adams et al., 1998; Goshima and Vale, 2003; Echard et al., 2004; Eggert et al., 2004) it is likely that Pavarotti, as part of the Centralspindlin complex with RacGAP50C (Dm MgcRacGAP) (Glotzer, 2005) and other regulatory components at the ends of furrow MTs, are the central determinants of where the cleavage furrow forms. Pavarotti has been observed to tip-track on MTs at onset of anaphase. It eventually concentrates on the ends of MTs in the vicinity of the furrow, where it bundles and stabilizes intersecting MTs in the equatorial region (Vale et al., 2009). Both Pavarotti and ABK also accumulate as a band at the equatorial cortex, distinct from their localization to cortical MTs (Minestrini et al., 2003; Hu et al., 2008, 2011; Vale et al., 2009). Like Pavarotti, ABK, which has been reported to phosphorylate and regulate MKLP1 (Guse et al., 2005), accumulates on MT plus-ends at anaphase onset, and inhibition of ABK blocked the equatorial accumulation--but not the tip-localizing behavior--of Pavarotti (Vale et al., 2009). We envision that a midzone-derived aurora kinase activity gradient contributes to cytokinesis via crosstalk with a more dominant astral stimulation and/or furrow MT pathway through multiple, non-mutually exclusive mechanisms: 1) phosphorylating tip-tracking regulatory complexes, 2) rendering cortical components competent for equatorial accumulation of cytokinesis regulators such as RhoA. RhoGEF, ABK, and Pavarotti, and 3) stabilization of furrow MTs, for example, through inhibition of catastrophe factors (Sampath et al., 2004; Gadea and Ruderman, 2006; Kelly et al., 2007).
In conclusion, we favor a model in which furrow positioning in Drosophila S2 cells is driven largely by equatorial stimulation that is comprised of 1) a midzone-derived stimulation signal to which an aurora kinase activity gradient contributes, and 2) what has traditionally been referred to as an "astral" stimulation signal., which is mediated by a non-centrosomal population of stable furrow MTs in the vicinity of the equatorial cortex (Canman et al., 2003; Foe and von Dassow, 2008; Field et al., 2015).
We would like to thank Eric Griffis for generously providing Drosophila S2 cell lines, and Marcin Przewloka and David Glover for the generous gift of the Drosophila antiaurora A. We are also grateful to Pat Wadsworth for sharing many insightful discussions on cytokinesis. This work was supported by an NIH grant (No. 5 R01 GM107026) to TJM and by Research Grant No. 5-FY13-205 from the March of Dimes Foundation to TJM, as well as support from the Charles H. Hood Foundation, Inc., Boston, MA, to TJM.
Adams, R. R., A. A. M. Tavares, A. Salzberg, H. J. Bellen, and D. M. Glover. 1998, pavarotti encodes a kinesin-like protein required to organize the central spindle and contractile ring for cytokinesis. Genes Dev. 12: 1483-1494.
Adams, R. R., S. P. Wheatley, A. M. Gouldsworthy, S. E. Kandels-Lewis, M. Carmena, C. Smythe, D. L. Gerloff, and W. C. Earn-shaw. 2000. 1NCENP binds the Aurora-related kinase A1RK2 and is required to target it to chromosomes, the central spindle and cleavage furrow. Curr. Biol. 10: 1075-1078.
Adams, R. R., H. Maiato. W. C. Earnshaw, and M. Carmena. 2001. Essential roles of Drosophila inner centromere protein (Incenp) and Aurora B in histone H3 phosphorylation, metaphase chromosome alignment, kinetochore disjunction, and chromosome segregation. J. Cell Biol. 153: 865-880.
Afonso. O., I. Matos, A. J. Pereira. P. Aguiar, M. A. Lampson, and H. Maiato. 2014. Feedback control of chromosome separation by a midzone Aurora B gradient. Science 345: 332-336.
Alsop, G. B., and D. Zhang. 2003. Microtubules are the only structural constituent of the spindle apparatus required for induction of cell cleavage. J. Cell Biol. 162: 383-390.
Bakhoum, S. F., L. Kabeche. J. P. Murnane, B. I. Zaki, and D. A. Compton. 2014. DNA-damage response during mitosis induces whole-chromosome missegregation. Cancer Discov. 4: 1281-1289.
Barisic, M., P. Aguiar, S. Geley, and H. Maiato. 2014. Kinetochore motors drive congression of peripheral polar chromosomes by overcoming random arm-ejection forces. Nat. Cell Biol. 16: 1249-1256.
Berdnik, D., and J. A. Knoblich. 2002. Drosophila Aurora-A is required for centrosome maturation and actin-dependent asymmetric protein localization during mitosis. Curr. Biol. 12: 640-647.
Canman, J. C, L. A. Cameron, P. S. Maddox, A. Straight, J. S. Tirnaucr, T. J. Mitchison, G. Fang, T. M. Kapoor, and E. D. Salmon. 2003. Determining the position of the cell division plane. Nature 424: 1074-1078.
Carmena, ML., S. Ruchaud, and W. C. Earnshaw. 2009. Making the Auroras glow: regulation of Aurora A and B kinase function by interacting proteins. Curr. Opin. Cell Biol. 21: 796-805.
Cesario, J. ML., J. K. Jang, B. Redding. N. Shah, T. Rahman, and K. S. McKim. 2006. Kinesin 6 family member Subito participates in mitotic spindle assembly and interacts with mitotic regulators. J. Cell Sci. 119: 4770-4780.
Chmatal, L., K. Yang, R. M. Schultz, and ML A. Lampson. 2015. Spatial regulation of kinetochore microtubule attachments by destabilization at spindle poles in meiosis 1. Curr. Biol. 25: 1835-1841.
Echard, A., G. R. X. Hickson, E. Foley, and P. H. O'Farrell. 2004. Terminal cytokinesis events uncovered after an RNAi screen. Curr. Biol. 14: 1685-1693.
Eggert, V. S., A. A. Kiger, C. Richter, Z. E. Perlman. N. Perrimon, T. J. Mitchison, and C. M. Field. 2004. Parallel chemical genetic and genome-wide RNAi screens identify cytokinesis inhibitors and targets. PLoS Biol. 2: e379.
Emanuele, M. J., W. Lan, M. Jwa, S. A. Miller, C. S. M. Chan, and P. T. Stukenberg. 2008. Aurora B kinase and protein phosphatase 1 have opposing roles in modulating kinetochore assembly. J. Cell Biol. 181: 241-254.
Field, C. M., A. C. Groen, P. A. Nguyen, and T. J. Mitchison. 2015. Spindie-to-cortex communication in cleaving, polyspermic Xenopus eggs. Mol. Biol. Cell 26: 3628-3640.
Foe, V. E., and G. von Dassow. 2008. Stable and dynamic microtubules coordinately shape the myosin activation zone during cytokinetic furrow formation. J. Cell Biol. 183: 457-470.
Fuller. B. G., M. A. Lampson. E. A. Foley, S. Rosasco-Nitcher, K. V. Le, P. Tobelmann, D. L. Brautigan, P. T. Stukenberg, and T. M. Kapoor. 2008. Midzone activation of Aurora B in anaphase produces an intracellular phosphorylation gradient. Nature 453: 1132-1136.
Gadea, B. B., and J. V. Ruderman. 2006. Aurora B is required for mitotic chromatin-induced phosphorylation of Op 18/Stathmin. Proc. Natl. Acad. Sci. USA 103: 4493-4498.
Gibson, D. G., L. Young, R.-Y. Chuang, J. C. Venter, C. A. Hutchison, 3rd, and H. O. Smith. 2009. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nat. Methods 6: 343-345.
Giet, R., D. McLean, S. Descamps, M. J. Lee, J. W. Raff, C. Prigent, and D. M. Glover. 2002. Drosophila Aurora A kinase is required to localize D-TACC to centrosomes and to regulate astral microtubules. J. Cell Biol. 156: 437-451.
Glotzer, M. 2005. The molecular requirements for cytokinesis. Science 307: 1735-1739.
Glover. D. M., M. H. Leibowitz, D. A. McLean, and H. Parry. 1995. Mutations in aurora prevent centrosome separation leading to the formation of monopolar spindles. Cell 81: 95-105.
Goshima, G., and R. D. Vale. 2003. The roles of microtubule-based motor proteins in mitosis: comprehensive RNAi analysis in the Drosophila S2 cell line. J. Cell Biol. 162: 1003-1016.
Gruneberg, U., R. Neef, R. Honda, E. A. Nigg, and F. A. Barr. 2004. Relocation of Aurora B from centromeres to the central spindle at the metaphase to anaphase transition requires MKlp2. J. Cell Biol. 166: 167-172.
Guse, A., M. Mishima, and M. Glotzer. 2005. Phosphorylation of ZEN-4/MKLP1 by Aurora B regulates completion of cytokinesis. Curr. Biol. 15: 778-786.
Hannak, E., M. Kirkham, A. A. Hyman, and K. Oegema. 2001. Aurora-A kinase is required for centrosome maturation in Caenorhabditis elegans. J. Cell Biol. 155: 1109-1116.
Hegarat, N., E. Smith, G. Nayak, S. Takeda, P. A. Eyers, and H. Hochegger. 2011. Aurora A and Aurora B jointly coordinate chromosome segregation and anaphase microtubule dynamics. J. Cell Biol. 195: 1103-1113.
Hochegger. H.. N. Hegarat, and J. B. Pereira-Leal. 2013. Aurora at the pole and equator: overlapping functions of Aurora kinases in the mitotic spindle. Open Biol. 3: 120185.
Hu, C.-K.. M. Coughlin, C. M. Field, and T. J. Mitchison. 2008. Cell polarization during monopolar cytokinesis. J. Cell Biol. 181: 195-202.
Hu. C.-K., M. Coughlin, C. ML Field, and T. J. Mitchison. 2011. KIF4 regulates midzone length during cytokinesis. Curr. Biol. 21: 815-824.
Kelly. A. E., S. C. Sampath, T. A. Maniar, E. M. Woo, B. T. Chait, and H. Funabiki. 2007. Chromosomal enrichment and activation of the Aurora B pathway are coupled to spatially regulate spindle assembly. Dev. Cell 12: 31-43.
Kitagawa, M., S. Y. S. Fung, N. Onishi, H. Saya, and S. H. Lee. 2013. Targeting Aurora B to the equatorial cortex by MKlp2 is required for cytokinesis. PLoS One 8: e64826.
Kitagawa, M., S. Y. S. Fung, U. F. S. Hameed, H. Goto, M. Inagaki, and S. H. Lee. 2014. Cdkl coordinates timely activation of MK1p2 kinesin with relocation of the chromosome passenger complex for cytokinesis. Cell Rep. 7: 166-179.
Kunda, P., N. T. L. Rodrigues, E. Moeendarbary, T. Liu, A. Ivetic, G. Charras, and B. Baum. 2012. PP1-mediated moesin dephosphorylation couples polar relaxation to mitotic exit. Curr. Biol. 22: 231-236.
Lampson, M. A., and I. M. Cheeseman. 2011. Sensing centromere tension: Aurora B and the regulation of kinetochore function. Trends Cell Biol. 21: 133-140.
Li, S., Z. Deng, J. Fu, C. Xu, G. Xin, Z. Wu, J. Luo, G. Wang, S. Zhang. B. Zhang et al. 2015. Spatial compartmentalization specializes the function of Aurora A and Aurora B. J. Biol Chem. 290: 17546-17558.
Lioutas, A., and I. Vernos. 2013. Aurora A kinase and its substrate TACC3 are required for central spindle assembly. EMBO Rep. 14: 829-836.
Meijering. E., O. Dzyubachyk, and I. Smal. 2012. Methods for cell and particle tracking. Methods Enzymol. 504: 183-200.
Minestrini, G., A. S. Harley, and D. M. Glover. 2003. Localization of Pavarotti-KLP in living Drosophila embryos suggests roles in reorganizing the cortical cytoskeleton during the mitotic cycle. Mol. Biol. Cell 14: 4028-4038.
Moutinho-Pereira, S., N. Stuurman, O. Afonso, M. Hornsveld, P. Aguiar. G. Goshima. R. D. Vale, and H. Maiato. 2013. Genes involved in centrosome-independent mitotic spindle assembly in Drosophila S2 cells. Proc. Natl. Acad. Sci. USA 110: 19808-19813.
Murthy, K., and P. Wadsworth. 2008. Dual role lor microtubules in regulating cortical contractility during cytokinesis. J. Cell Sci 121: 2350-2359.
Nguyen, P. A., A. C. Groen, M. Loose, K. Ishihara, M. Wiihr, C. M. Field, and T. J. Mitchison. 2014. Spatial organization of cytokinesis signaling reconstituted in a cell-free system. Science 346: 244-247.
Nunes Bastos, R., S. R. Gandhi, R. D. Baron, U. Gruneberg, E. A. Nigg., and F. A. Barr. 2013. Aurora B suppresses microtubule dynamics and limits central spindle size by locally activating KIF4A.J. Cell Biol. 202: 605-621.
Rappaport, R. 1961. Experiments concerning the cleavage stimulus in sand dollar eggs. J. Exp. Tool 148: 81-89.
Rappaport. R. 1971. Cytokinesis in animal cells. Int. Rev. Cytol. 31: 169-213.
Reboutier, D., M.-B. Troadec, J.-Y. Cremet, L. Chauvin, V. Guen, P. Salaun, and C Prigent. 2013. Aurora A is involved in central spindle assembly through phosphorylation of Ser 19 in P150Glued. .J. Cell Biol. 201: 65-79.
Rodrigues, N. T. L., S. Lekomtsev, S. Jananji., J. Kriston-Vizi, G. R. X. Hickson, and B. Baum. 2015. Kinetochore-localized PPI-Sds22 couples chromosome segregation to polar relaxation. Nature 524: 489-492.
Sampath, S. C., R. Ohi, O. Leismann, A. Salic, A. Pozniakovski, and H. Funabiki. 2004. The chromosomal passenger complex is required for chromatin-induced microtubule stabilization and spindle assembly. Cell 118: 187-202.
Schneider, C. A., W. S. Rasband, and K. W. Eliceiri. 2012. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9: 671-675.
Scrofani, J., T. Sardon, S. Meunier, and I. Vernos. 2015. Microtubule nucleation in mitosis by a RanGTP-dependent protein complex. Curr. Biol. 25: 131-140.
Smurnyy, Y., A. V. Toms, G. R. Hickson, M. J. Eck, and U. S. Eggert. 2010. Binucleine 2, an isoform-specilic inhibitor of Drosophila Aurora B kinase, provides insights into the mechanism of cytokinesis. ACS Client. Biol. 5: 1015-1020.
Tan, L., and T. M. Kapoor. 2011. Examining the dynamics of chromosomal passenger complex (CPC)-dependent phosphorylation during cell division. Proc. Natl. Acad. Sci. USA 108: 16675-16680.
Tanaka. T. U., N. Rachidi, C. Janke, G. Pereira, M. Galova, E. Schiebel, M. J. R. Stark, and K. Nasmyth. 2002. Evidence that the Ip11-Sli15 (Aurora kinase-INCENP) complex promotes chromosome bi-orientation by altering kinetochore-spindle pole connections. Cell 108: 317-329.
Terada, Y. 2001. Role of chromosomal passenger complex in chromosome segregation and cytokinesis. Cell Struct. Fund. 26: 653-657.
Vale, R. D., J. A. Spudich, and E. R. Griffis. 2009. Dynamics of myosin, microtubules, and Kinesin-6 at the cortex during cytokinesis in Drosophila S2 cells. J. Cell Biol. 186: 727-738.
Ye. A. A., J. Deretic, C. M. Hoel, A. W. Hinman. D. Cimini, J. P. Welburn, and T. J. Maresca. 2015. Aurora A Kinase contributes to a pole-based error correction Pathway. Curr. Biol. 25: 1842-1851.
Zhu, C., J. Zhao. M. Bibikova. J. D. Leverson, E. Bossy-Wetzel. J.-B. Fan, R. T. Abraham, and W. Jiang. 2005. Functional analysis of human microtubule-based motor proteins, the kinesins and dyneins, in mitosis/cytokinesis using RNA interference. Mol. Biol. Cell 16: 3187-3199.
ANNA A. YE (1,2,[dagger]). JULIA TORABI (1,[dagger]), and THOMAS J. MARESCA (1,2,*)
(1) Biology Department, and (2) Molecular and Cellular Biology Graduate Group, University of Massachusetts, Amherst, Massachusetts, 01003
Received 16 March 2016; accepted 12 July. 2016.
(*) To whom correspondence should be addressed. E-mail: tmaresca@bio. umass.edu
([dagger]) These authors contributed equally to the manuscript.
Abbreviations: AAK. aurora A kinase; ABK. aurora B kinase; cDNA. complementary DNA; CNN, centrosomin: CPC, chromosomal passenger complex; DAPI, 4',6'-diamidino-2-phenylindole: DMSO, dimethyl sulfoxide: dsRNA, double-stranded RNA; FBS, fetal bovine serum: FRET. Forster (or fluorescence) resonance energy transfer; GFP. green fluorescent protein: INCENP. Drosophila inner centromere protein: mABK, midzone ABK: MKLPI. MKLP2. mitotic kinesin-like protein 1, 2; MLN, MLN8237, AAK inhibitor; MRLC, myosin regulatory light chain; MT, microtubule; pAAK, pABK, phosphorylated AAK. ABK: p150Glued. Dynactin subunit; PBS, phosphate-buffered saline: RFP, red fluorescent protein: RNAi. RNA interference: SDS-PAGE. sodium dodecyl sulfate polyacrylamide gel electrophoresis; SdsS22. regulatory subunit of protein phosphatase 1; TACC3. transforming acidic coiled-coil containing protein 3: TBS. Tris-buffered saline: TIRF. total internal reflection fluorescence (microscopy).
Table 1 The primers used in this study Primer Sequence (5'[right arrow]3') Aurora A RNAi F TAATACGACTCACTATAGGGATGTCCCATCCGTCTGACCA Aurora A RNAi R TAATACGACTCACTATAGGGTTGCCAAACTTGCCCCGTCC Subito RNAi F TAATACGACTCACTATAGGGCTGAAGCTAATCAATGGCAGC Subito RNAi R TAATACGACTCACTATAGGGTTTTCTGAACTGTACTGGCCG CNN RNAi F TAATACGACTCACTATAGGGCTATGGGAGAAAAATCGTGAGG CNN RNAi R TAATACGACTCACTATAGGGTCTTTCCTCTGTCCTTCTGAGC Sds22 RNAi F TAATACGACTCACTATAGGGAGAACTTGTCCAGCCTGAAA Sds22 RNAi R TAATACGACTCACTATAGGGTTCTTGGCCAAGTCGAGGGT F, forward: R, reverse. Table 2 Interphase microtubule dynamics are unaffected by expression of the Tau-FRET reporter No FRET FL Tau-FRET expression expression Growth velocity ([micro]m/min) 4.57 4.21 Shrink velocity ([micro]m/min) 5.56 5.34 Rescue frequency ([s.sup.-1]) 0.014 0.014 Catastrophe frequency ([s.sup.-1]) 0.057 0.06 FRET, Forster resonance energy transfer; FL, full length.
Please note: Some tables or figures were omitted from this article.
|Printer friendly Cite/link Email Feedback|
|Author:||Ye, Anna A.; Torabi, Julia; Maresca, Thomas J.|
|Publication:||The Biological Bulletin|
|Date:||Aug 1, 2016|
|Previous Article:||Using stage- and slit-scanning to improve contrast and optical sectioning in dual-view inverted light sheet microscopy (diSPIM).|
|Next Article:||Imaging adenosine triphosphate (ATP).|