![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 20, Issue 22, 4686-4695, November 15, 2009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Department of Biological Chemistry, School of Medicine, University of California, Irvine, Irvine, CA 92697
Submitted November 17, 2008;
Revised September 3, 2009;
Accepted September 15, 2009
Monitoring Editor: Yixian Zheng
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Coil1, a mutant defective in Hec1 interaction, elicited abnormal spindle morphology often detected in Hec1 and Hice1 deficient cells. Together, the results suggest that Hec1, through cooperation with Hice1, contributes to centrosome-directed microtubule growth to facilitate establishing a proper mitotic spindle. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Karsenti and Vernos, 2001; Kline-Smith and Walczak, 2004).
Hec1, also known as Ndc80, is an evolutionarily conserved coiled-coil protein critical for mitotic progression (Chen et al., 1997a,b; Wigge et al., 1998). It contains three leucine-heptad coiled-coil domains at the C-terminal region, which are thought to mediate multiple protein–protein interactions (Chen et al., 1997b; Kline-Smith et al., 2005; Ciferri et al., 2008). The N-terminal region of Hec1 is a microtubule binding domain structurally similar to the calponin-homology domain of the microtubule binding protein EB1 (Cheeseman et al., 2006; Wei et al., 2007; Ciferri et al., 2008). In cells, Hec1 is a major binding partner for another coiled-coil protein, Nuf2. The Hec1/Nuf2 dimer resides at the outer kinetochore layer and orients in such a way that their N-terminal microtubule binding module projects outward, whereas the C-terminal tail is anchored by the inner layer components Spc24/Spc25 (McCleland et al., 2003; Emanuele et al., 2005; Wilson-Kubalek et al., 2008). These four proteins can form a dumbbell-shaped heterotetramer in an equal molecular ratio in vitro (Ciferri et al., 2005; Wei et al., 2005). At kinetochores, the Hec1 complex associates with the KNL1/Spc105 and the Mtw1/Mis12 complexes, and together they form the KNL-1/Mis12 complex/Ndc80 complex network that can provide two microtubule binding interfaces for kinetochore fiber attachment (Cheeseman et al., 2006). The microtubule binding activity of Hec1 is inhibited upon phosphorylation of its N-terminal tail by kinetochore-associated kinase Aurora B. This serves as a critical mechanism for correcting improper microtubule attachment at the kinetochore (Cheeseman et al., 2006; DeLuca et al., 2006).
Importantly, Hec1 overexpression has been observed in a variety of human cancers and was found to associate with adverse clinical outcomes of primary breast cancers and cases with multiple cancers (Chen et al., 1997a; van't Veer et al., 2002; Glinsky et al., 2005). In an inducible mouse model, overexpression of Hec1 was shown to result in spindle checkpoint hyperactivation correlating with eventual significant tumor formation, mainly lung adenoma and hepatocellular adenoma (Diaz-Rodriguez et al., 2008). This animal phenotype is recapitulative of those observed in Mad2-overexpressing mice (Sotillo et al., 2007). Hec1 has now emerged as a novel therapeutic target for potential cancer intervention by using strategies of RNA interference (RNAi) or small molecular inhibitors in part because of its specific requirement in the mitotic process (Gurzov and Izquierdo, 2006; Li et al., 2007; Wu et al., 2008b). Therefore, the investigation of Hec1 function is becoming an increasingly important area of study.
The precise functions of Hec1 during mitosis are not fully understood at present, although previous studies have revealed important roles of Hec1 at the kinetochore. Interestingly, several reports have documented that a portion of cellular Hec1, as well its binding partners Nuf2 and Spc25, are located at the centrosome during interphase and at the spindle pole region during mitosis (Hori et al., 2003; Sauer et al., 2005; Lin et al., 2006; Goshima et al., 2007; Diaz-Rodriguez et al., 2008). Importantly, significant spindle abnormalities (e.g., multipolarity) were observed in Hec1-depleted cells (Martin-Lluesma et al., 2002; DeLuca et al., 2003; McCleland et al., 2003). However, it remains to be shown whether Hec1 is involved in regulating mitotic spindle assembly by interacting with a spindle-associated factor.
Previous yeast two-hybrid screens have identified multiple Hec1-interacting candidates known to be involved in centrosome or spindle regulation (Chen et al., 2002; Wong et al., 2007; Wu et al., 2008a), one of which is Hice1. Hice1 was shown to be a centrosome- and spindle-associated protein possessing a microtubule binding module at its N-terminal region (Wu et al., 2008a). Remarkably, Hice1 knockdown cells exhibited delayed mitotic progression in conjunction with evident spindle abnormalities, increased chromosome misalignment and segregation errors. This was later confirmed by other reports, which detected Hice1 as a subunit of the eight-member Augmin complex (Lawo et al., 2009; Uehara et al., 2009). Although accumulating evidence suggests the critical importance of Hice1 in maintaining proper spindle morphology, it has not been systematically tested whether Hice1 plays a regulatory role in spindle assembly, for example, microtubule nucleation initiated by the well-documented 
-tubulin ring complex (-TuRC) complex (Wiese and Zheng, 2006).
In this study, we showed that Hec1 colocalizes with Hice1 at the spindle pole region during mitosis and Hec1 interacts with the coiled-coil domain 1 of Hice1. Reproducibly, knockdown of Hice1 by small interfering RNA (siRNA) treatment resulted in reduction of Hec1 in cells, primarily from the centrosome and spindle pole vicinity. Antibody-mediated neutralization of Hec1 or Hice1 impaired microtubule aster formation from isolated mitotic centrosomes. Disruption of the Hec1/Hice1 interaction by overexpressing Hice1Coil1, a mutant defective in Hec1 interaction, triggered spindle multipolarity characteristic of Hec1- and Hice1-deficient cells. These results suggest that the interaction of Hec1 with Hice1 is important for optimal centrosome-directed microtubule formation, so as to facilitate the establishment of a proper mitotic spindle.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Chen et al., 1997a). For glutathione transferase (GST) pull-down assays, Hice1 and Hec1 deletion mutants were tagged with GST and prepared from Escherichia coli by using affinity binding with glutathione-Sepharose. The 35S-labeled Hice1 produced by in vitro translation was used for interacting with various GST-Hec1 fusions in the pull-down assay, and vice versa. Procedures for GST-pull-down assays and coimmunoprecipitation assays have been detailed previously (Xiao et al., 2001). For the Western blot analysis of the resultant coimmunoprecipitates, the Nuf2 blot was developed with a TrueBlot Ultra kit (eBioscience, San Diego, CA) to avoid cross-reactivity with the immunoglobulin (Ig)G heavy chain, because Nuf2 migrates closely to the IgG heavy chain and Hice1. The blot was then stripped with a stripping buffer (Millipore, Billerica, MA) to allow for the subsequent detection of Hice1.
For in vitro interacting assays with further purified proteins, Hec1/GST-Nuf2 dimer encoded by a bicistronic pGEX-6p-1 plasmid or GST-Nuf2 was expressed in BL21 bacteria and purified by affinity binding with glutathione-Sepharose followed by gel filtration chromatography (Superdex 200) under a high salt condition (Ciferri et al., 2005). Six his-tagged Hice1 (wild type) and Hice1 mutants lacking one of the two coiled-coiled regions were expressed in Rosetta bacterium strain and purified using immobilized metal affinity chromatography affinity resin according to manufacturer's instructions (Bio-Rad Laboratories, Hercules, CA). Hice1 proteins were also further purified by gel filtration chromatography (Superdex 200) using the AKATA FPLC system (Pharmacia/GE Healthcare, Chalfont St. Giles, Buckinghamshire, United Kingdom). For the in vitro interaction, equal amounts of each Hice1 versions were mixed with GST-Nuf2 or Hec1/GST-Nuf2 proteins for 30 min at room temperature (RT). Potential interacting proteins were pulled down by glutathione-Sepharose beads blocked with 5% bovine serum albumin in a binding buffer (30 mM Tris-HCl, pH 7.6, 500 mM NaCl, and 0.2% Triton X-100). The complexes were washed adequately and then lysed in the Lamelli buffer for SDS-polyacrylamide gel electrophoresis (PAGE) analysis followed by Western blot with various antibodies.
Cell Lines, Mutant Expression, and RNAi
Human cancer cell lines HeLa and U2OS were cultured in DMEM plus 10% fetal bovine serum (FBS). The KE37 cell line of T lymphoblastic origin (Mayer and Kinkel, 1982) were cultivated in RPMI medium plus 10% FBS. U2OS cells stably expressing Hice1-green fluorescent protein (GFP), Hec1-GFP, or Hice1Coil1-GFP were established by infection of retrovirus produced in GP2 293 packaging cell line. Previously validated Hec1 and Hice1 targeting siRNA sequences were used (Martin-Lluesma et al., 2002; Lin et al., 2006; Wu et al., 2008a). siRNA was transfected into cells with Lipofectamine 2000 (Invitrogen, Carlsbad, CA).
Microscopy
Immunostaining procedure was adapted as described previously (Wu et al., 2000). In brief, cells grown on coverslips were gently washed with the PEMG buffer [80 mM piperazine-N,N'-bis(2-ethanesulfonic acid) (PIPES), pH 6.8, 5 mM EGTA, 1 mM MgCl2, and 4 M glycerol] or phosphate-buffered saline (PBS) before fixation with 100% methanol at –20°C or 4% paraformaldehyde in PEMG or PBS buffer. After permeabilization with 0.4% Triton-X 100, cells were blocked with 5% normal goat serum (NGS) in PBS and then incubated with primary antibodies in PBS with 5% NGS (1–2 h; RT). Secondary antibodies used were conjugated with Alexa 488 or 594 (Invitrogen, Carlsbad, CA). 4',6-Diamidino-2-phenylindole (DAPI) staining was applied after secondary antibody incubation and cells were finally mounted on coverslides with Prolong gold anti-fade reagent (Invitrogen). Images were captured with an Axiovert 200M microscope (Carl Zeiss, Thornwood, NY) equipped with a charge-coupled device camera (Hamamatsu, Bridgewater, NJ) controlled by the Axiovision software. Further image analysis or quantification was performed with Image-Pro Plus (MediaCybernetics, Bethesda, MD) or Adobe Photoshop software (Adobe Systems, Mountain View, CA).
Centrosome Isolation and In Vitro Microtubule Nucleation Assay
Centrosomes were isolated from KE37 cells similarly to the previous report (Gosti-Testu et al., 1986). For mitotic centrosome isolation, HeLa or KE37 cells were first synchronized at the G1/S boundary via a double thymidine arrest procedure, released to progress through S phase, and finally arrested at prometaphase by adding nocodazole 8 h before harvest. The mitotic arrest efficiency was further confirmed by fluorescence-activated cell sorting analysis (>90% of cells with 4N DNA content). Centrosome fractions were collected and analyzed by SDS-PAGE and Western blotting. For centrosome-directed microtubule nucleation assay in vitro, reaction mixtures contained 1x BRB80 (80 mM PIPES, pH 6.9, 2 mM MgCl2, 0.5 mM EGTA, and 1 mM guanosine triphosphate). Before being used for neutralization, antibodies underwent buffer exchange in 1x BRB80 using Amicon centrifugal filter devices (Millipore). Antibodies were incubated with centrosomes for 30 min at 4°C before addition of microtubules containing 20% of rhodamine-labeled tubulins. Nucleation mixture was incubated for 37°C for 30 min and then applied on a coverslide and covered with a coverslip for subsequent microscopic imaging (Bornens et al., 1987).
Microtubule Regrowth Assays
Microtubule regrowth in interphase or mitotic cells was performed similarly to the previous report (Luders et al., 2006). In brief, cells were grown on coverslip and transfected with siRNA for 48 h. For mitotic microtubule growth, cells were treated with 400 ng/ml nocodazole at 37°C for 2 h to depolymerize microtubules, followed by drug washout and an additional incubation on ice for 30 min (cold shock). Note that for interphase cells only cold treatment is required. Cells were recovered in prewarmed growth medium at 37°C to allow microtubule regrowth and subsequently fixed at various time points for further immunostaining to reveal the tubulin structures. Microscopic z-sectioning of mitotic cells was performed to collect the images at varying z-axis focal planes. Images were deconvoluted, the microtubule aster intensity was quantified using ImageJ (National Institutes of Health, Bethesda, MD), and the maximal projection of the z-sections is presented.
Antibodies
Commercial antibodies used for immunostaining or Western blotting were as follows: mouse monoclonal anti-Hec1, rabbit anti-Aurora A, mouse anti-BubR1, mouse anti-Eg5, mouse anti-β-actin, rabbit anti-p150Glued, and rabbit anti-NUMA (GeneTex, San Antonio, TX); mouse anti-
-tubulin and rabbit anti--tubulin (Sigma-Aldrich, St. Louis, MO); sheep anti-β-tubulin (Cytoskeleton, Denver, CO); mouse anti-Nuf2 antibodies (GeneTex and MBL international, Woburn, MA); and mouse anti-GFP monoclonal mixtures (Roche Diagnostics, Mannheim, Germany). Affinity-purified mouse polyclonal antibody for Hice1 was described previously (Wu et al., 2008a).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Hice1 has been demonstrated to be a centrosomal and microtubule binding protein (Wu et al., 2008a), and both proteins are shown to be localized at the centrosome and spindle. This potential interaction of Hec1 with Hice1 provides an opportunity to address the role of Hec1 at the centrosome. To do so, we first consolidated the interaction between Hec1 and Hice1. To refine the precise region of Hec1 responsible for Hice1 interaction, a yeast two-hybrid assay was used using different regions of Hec1 fused to the GAL4 DNA-binding domain, whereas the full-length Hice1 was tagged with the transactivation domain. A region covering part of the Hec1 C-terminal coiled-coil domains was sufficient for Hice1 interaction (Figure 1A). Next, a set of GST-tagged Hec1 or Hice1 deletion mutants were prepared from bacteria for GST-pull-down assays. The resultant GST preparations are mainly products of correct size, although some also contain small amount of degradation species (Figure 1, B and C). With these GST fusions, a preferential interaction of 35S-labeled Hice1 was detected for a Hec1 region spanning from coiled-coil 1–2 (Figure 1B). Inversely, the 35S-labeled Hec1 was found to primarily interact with a fragment encompassing the first coiled-coil domain of Hice1 (Figure 1C). Weak interactions with Hec1 were also observed for the other fragments of Hice1, probably due to nonoptimal binding condition for Hec1. Nonetheless, the data support that the Hec1/Hice1 interaction is primarily mediated through the coiled-coil 1 and 2 regions of Hec1 and the coiled-coil 1 domain of Hice1 in vitro.
|
Coil1 (coiled-coil domain 1 deleted), or GFP only, because the coiled-coil domain 1 of Hice1 is important for Hec1 interaction in vitro (Figure 1). Hice1-FL, but not Hice1Coil1 or GFP alone, was capable of coimmunoprecipitating with endogenous Hec1 (Figure 1E), thereby confirming that the coiled-coil domain 1 of Hice1 is responsible for Hec1 interaction in cells.
Hice1 Interacts with Hec1 of the Hec1/Nuf2 Complex
Nuf2 is thought to be a major binding partner of Hec1 and the Hec1/Nuf2 heterodimer may further complex with Spc24 and Spc25 to form a heterotetramer complex (McCleland et al., 2003; Ciferri et al., 2005; Emanuele et al., 2005; Wei et al., 2005). To test whether Hice1 may interact with the Hec1/Nuf2 dimer (presumably through Hec1), 6his-tagged Hice1-FL, Hice1Coil1, and Hice1Coil2 were individually expressed in bacteria and purified to near homogeneity by performing affinity binding followed by gel filtration chromatography (Figure 2A). GST-Nuf2 alone or the Hec1/GST–Nuf2 complex was also expressed and similarly purified to near homogeneity as described previously (Figure 2, A and B) (Ciferri et al., 2005). Purified Hec1/GST-Nuf2 dimer was able to interact with Hice1-FL and Hice1Coil2, but not Hice1Coil1 (Figure 2C, lanes 1–3). This indicates that Hice1 interacts with Hec1/Nuf2 and that the coiled-coil domain 1 of Hice1 is critically important for this binding event (Figure 1). Furthermore, little or no interaction was detected for GST-Nuf2 toward any version of the purified Hice1 proteins when Hec1 was absent (Figure 2C, lanes 4–6), suggesting that Hec1 but not Nuf2 plays a primary role in mediating the interaction between Hice1 and the Hec1/Nuf2 dimer. Consistently, both anti-Hec1 and anti-Nuf2 antibodies were capable of coimmunoprecipitating with endogenous Hice1 from cell extracts. In contrast, Eg5, an abundant spindle-associated kinesin, failed to associate with Hice1 in a similar assay (Figure 2D). Together, the results clearly demonstrate that Hice1 may use its coiled-coil domain 1 to specifically interact with Hec1 or with the Hec1/Nuf2 dimer via Hec1.
|
). Western blot analysis on the resultant centrifugational fractions indicated that Hec1 and Hice1 cofractionated with the centrosome marker -tubulin (Figure 3C). Last, immunofluorescent staining was performed using isolated centrosomes stuck on coverslips. As revealed by colocalization with -tubulin signals, the centrosomes were positive for Hice1 and Hec1 in 88 and 89% of the population, respectively (Figure 3D). Together, these results suggest that Hec1 and Hice1 may interact with each other at the centrosome.
|
|
-tubulin (Figure 4, C and D). In contrast, the kinetochore-associated Hec1 signal was not affected (Figure 4, C and D), suggesting that Hice1 depletion preferentially affects the centrosomal/spindle pool of Hec1. Fluorescent quantification of a 3-µm circular region around the centrosome revealed a 80–85% decrease of Hec1 signal in Hice1 siRNA-treated cells relative to control treated ones (n = 35 centrosomes for each treatment; p < 0.05). Inversely, knockdown of Hec1 by siRNA did not seem to affect the localization of Hice1 in cells displaying either a bipolar or multipolar spindle (Figure 4E). Importantly, the mitotic distribution of two other well documented spindle molecules, NUMA and p150Glued (a dynactin subunit), was not apparently affected by Hice1 depletion (Supplemental Figure 2), suggesting a specific effect on Hec1 subcellular localization. Together, these findings suggest that Hice1 is critically important for proper Hec1 localization at the centrosome and spindle.
Inactivation of Hec1 and Hice1 Resulted in Reduced Microtubule Regrowth from Centrosomes
The centrosome is a major microtubule nucleation site for initiating spindle assembly. We were intrigued in testing the potential role of Hec1 and Hice1 in centrosome-directed microtubule formation. To test this possibility, a microtubule growth assay was undertaken with enriched centrosomes that were isolated as described in a previous experiment (Figure 3). Both Hec1 and Hice1 have a microtubule binding activity (Cheeseman et al., 2006; DeLuca et al., 2006; Wu et al., 2008a). To neutralize Hec1 and Hice1, specific antibodies were prepared and affinity purified. Isolated centrosomes were first neutralized with Hice1 or Hec1 antibodies and then incubated with rhodamine-labeled tubulin to allow microtubule outgrowth (Figure 5A). The addition of varying doses of normal mouse IgG did not alter the microtubule growth in comparison with the mock-treated reaction (Figure 5, Aa and B). Interestingly, adding Hec1 antibody resulted in shortened microtubule length (maximal reduction 
25%) (Figure 5, Ab and B). In comparison, Hice1 antibody was able to inhibit microtubule growth more dramatically (maximal reduction 60%) (Figure 5, Ac and B). Notably, adding both Hice1 and Hec1 antibodies seemed to have an additive inhibitory effect on microtubule length (Figure 5B), but a low level of microtubule growth (5 µm) still occurred. Because centrosomes isolated from the above procedure represent a mixture of both interphase and mitotic centrosomes, we then proceeded to determine whether the phenotypes can be detected using a more pure mitotic population. To do so, mitotic centrosomes were isolated from KE37 or HeLa cells enriched at prometaphase by a double thymidine block followed by nocodazole arrest (>90% arrested at M phase; see Materials and Methods). Using the mitotic centrosomes, similar phenotypes were recapitulated in the microtubule nucleation assay (Figure 5C). Together, the results suggest that Hice1 and Hec1 are required for optimal microtubule aster formation in vitro, where Hice1 plays a more dominant role than Hec1. To further determine whether the reduction of microtubule length was due to Hec1's and Hice1's role in microtubule nucleation or stabilization of prenucleated microtubules, microtubule asters were first allowed to form from the centrosomes followed by adding polyclonal antibodies against Hice1 or Hec1. No reduction in final microtubule length was detected (data not shown), suggesting that Hice1 and Hec1 are unlikely to play a major role in maintaining the prenucleated microtubules.
|
). Thus, the role of Hec1 and Hice1 in centrosome regulation seems to be subtle during interphase. Alternatively, the remaining residual Hec1 and Hice1 after siRNA treatment are sufficient for the interphase centrosome functionality. Therefore, a specific role of Hec1 and Hice1 in mitotic spindle assembly was then tested by adapting a mitotic microtubule regrowth assay in U2OS cells (Luders et al., 2006). In this assay, the microtubule aster formation was visualized and measured by quantitative confocal fluorescent microscopy. Because Hice1 or Hec1 depletion tends to trigger abnormal spindle pole configurations incompatible for faithful microtubule aster quantification, only those cells with two distinct spindle poles, reflecting bipolarity, were measured. Even under this stringent condition, it was found that, in comparison to the control siRNA-treated cells, the time-dependent microtubule regrowth from mitotic centrosomes was impaired significantly at early time points (
3 min) in cells treated with Hec1 siRNA, Hice1 siRNA, or both (Figure 6, A and B). The relative reduction in microtubule aster intensity is significant (p < 0.0022, two-way analysis of variance [ANOVA] test) (Figure 6C). Consistent with the observation from the microtubule nucleation assay in vitro (Figure 5), Hec1 inactivation in cells seemed to have less effect on centrosomal microtubule growth than Hice1 inactivation. However, treating cells with Hec1 and Hice1 siRNAs together elicited a similar effect to Hice1 siRNA alone (p = 0.77; no significant difference) (Figure 6C). Together, these results suggest that Hec1 functions in microtubule aster formation from mitotic centrosomes, which is probably controlled by Hice1.
|
; Zhu et al., 2008). Importantly, FAM29A, an Augmin subunit, was found to associate with Hice1 but not Hec1 in a coimmunoprecipitation assay (Supplemental Figure 4), suggesting that the Hec1/Hice1 association is distinct to Hice1 itself but not necessarily to the integral Augmin complex. Thus, the role of Hec1/Hice1 in the centrosomal microtubule nucleation is probably specific to a subpopulation of Hice1 different from that in the Augmin complex.
The Interaction of Hec1 with Hice1 Is Required for Proper Mitotic Spindle Formation
To further consolidate the importance of the Hec1/Hice1 interaction for spindle formation, we subsequently examined how the abrogation of Hec1/Hice1 interaction may affect the spindle configuration. To disrupt the Hec1/Hice1 interaction, the Hice1Coil1 mutant defective in Hec1 binding (Figure 1E) was used. It was noted that GFP-tagged Hice1Coil fails to localize to microtubule arrays during interphase (Wu et al., 2008a) but still associates with the mitotic spindle, albeit to a lesser extent than the full-length version (Figure 7A). This suggests a mitotic specific and Hec1-independent mechanism for Hice1 recruitment to spindles. Next, U2OS cells were transiently infected with Hice1-expressing retrovirus (RNAi-resistant Hice1-FL-GFP or Hice1Coil1-GFP), or GFP-expressing control virus, followed by Hice1 siRNA transfection to deplete endogenous Hice1. Expectedly, depletion of Hice1 in GFP-only U2OS cells triggered significant formation of abnormal spindles with multiple poles (Figure 7B), similar to the previous report (Wu et al., 2008a). Expression of Hice1-FL in Hice1-siRNA treated cells restored the bipolar spindle formation to a normal rate (Figure 7, B and C). Remarkably, replacing endogenous Hice1 with the Hice1Coil1 mutant led to a sixfold increase of multipolar spindle formation relative to the Hice1-FL rescued cells (Figure 7, B and C) in part due to defective Hec1 interaction, although non-Hec1 associated negative effects cannot be ruled out at present. The phenotypes are reminiscent of those observed in Hec1 or Hice1 depleted cells (Martin-Lluesma et al., 2002; DeLuca et al., 2003; Hori et al., 2003; McCleland et al., 2003; Lin et al., 2006; Wu et al., 2008a). Together, these results highlight the critical importance of the Hec1/Hice1 interaction for a proper mitotic spindle assembly in vivo, which requires the coiled-coil domain 1 of Hice1.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Microtubule nucleation can be initiated from the centrosome, chromatid, or within the spindle. Hice1 was recently shown to be a subunit of the eight-member Augmin complex. RNAi experiments suggested that some Augmin subunits were involved in microtubule nucleation from within the spindle but not from the centrosome (Goshima et al., 2008; Zhu et al., 2008). Although it remains to be shown whether Augmin exclusively functions as an integral unit, it is likely that a pool of Hice1 may function in the intraspindle microtubule nucleation as an Augmin subunit, similar to FAM29A (Zhu et al., 2008). Consistent with this idea, we noticed that the spindle density in Hice1-depleted cells was often decreased (Wu et al., 2008). In contrast, our results suggest that Hice1 may additionally contribute to the centrosome-directed pathway. This latter activity of Hice1 is in part mediated through Hec1, which does not associate with FAM29A in a coimmunoprecipitation experiment (Supplemental Figure 4), although the potential association with the other six Augmin components remains to be tested. It is possible that the Hec1-associated pool of Hice1 might be distinct from that constituting the Augmin complex. Furthermore, Hice1 could be engaged in chromatid-directed microtubule nucleation pathway, possibly through associating with Aurora A and TPX2 (our unpublished data). Thus, it is very likely that Hice1 forms distinct complexes to regulate microtubule nucleation from various subcellular sites.
The activity of Hec1 seems to be subtle in the microtubule nucleation from mitotic centrosomes, but it can be detected reproducibly and with statistical significance. On the contrary, the activity of Hice1 is much more evident. Our preliminary study reveals that Hec1 and Hice1, when ectopically expressed in cells, can associate with the -TuRC via the GCP2 subunit. Furthermore, both Hice1 and Hec1 have microtubule binding and bundling activities in vitro (Cheeseman et al., 2006; Wu et al., 2008). It is possible that during microtubule nucleation by the -TuRC complex, Hec1 can assist Hice1 to hold or stabilize the microtubules and also help build bundled microtubule fibers important for establishing a bipolar spindle with sufficient microtubule density.
An obvious multipolar phenotype was observed in cells when endogenous Hice1 has been replaced by ectopically expressed Hice1Coil1 (so as to disrupt the Hice1/Hec1 interaction) or in those cells depleted of Hice1 or Hec1. It is apparent that Hice1 and Hec1 play a positive role to maintain the spindle bipolarity. However, it remains to be elucidated with regards to the exact relationship between the microtubule nucleation activity of Hice1/Hec1 and the spindle polarity. It is possible that defects in microtubule nucleation may affect the ratio of free tubulin versus microtubules, which may then adversely affect the spindle microtubule density and consequently the bipolarity as a whole, as proposed in a previous study (Holmfeldt et al., 2003). Moreover, it cannot be ruled out that Hice1Coil1 may also affect some other unknown aspects of Hice1 function, other than recruiting Hec1 to the spindle pole. For example, the coiled-coil domain 1 of Hice1 may be important for stabilizing the K-fibers to maintain the tension on the kinetochores and the spindle, as suggested for some Augmin subunits (Lawo et al., 2009). In this regard, however, the requirement for Hec1 in stabilizing K-fibers is well established (DeLuca, 2006). Whether Hec1 and Hice1 work together to regulate the K-fiber dynamics and stability at the kinetochore/K-fiber interface warrants further investigation.
It is known that overexpression of Hec1 associates with adverse clinical outcomes of human cancers. Consistently, we observed that both Hice1 and Hec1 are overexpressed with similar expression profiles in the NCI-60 panel of cancer lines and an additional collection of breast and ovarian cancer cell lines (total >70 cell lines; our unpublished data), underlying a common regulatory pathway for the expression of these two molecules, the significance of which in tumorigenesis is becoming increasingly clear and seems to implicate their roles in mitosis. The novel activity of Hec1 in mitotic centrosomes may help explain its role in tumorigenesis.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
These authors contributed equally to this work. 
Present address: Department of Microbiology and Molecular Genetics, School of Medicine, University of California, Irvine, Irvine, CA 92697.
Address correspondence to: Wen-Hwa Lee (whlee{at}uci.edu)
Abbreviations used: -TuRC, -tubulin ring complex; DAPI, 4',6-diamidino-2-phenylindole; GFP, green fluorescent protein; GST, glutathione transferase.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cheeseman, I. M., Chappie, J. S., Wilson-Kubalek, E. M., and Desai, A. (2006). The conserved KMN network constitutes the core microtubule-binding site of the kinetochore. Cell 127, 983–997.[CrossRef][Medline]
Chen, Y., Riley, D. J., Chen, P. L., and Lee, W. H. (1997a). HEC, a novel nuclear protein rich in leucine heptad repeats specifically involved in mitosis. Mol. Cell. Biol 17, 6049–6056.
Chen, Y., Riley, D. J., Zheng, L., Chen, P. L., and Lee, W. H. (2002). Phosphorylation of the mitotic regulator protein Hec1 by Nek2 kinase is essential for faithful chromosome segregation. J. Biol. Chem 277, 49408–49416.
Chen, Y., Sharp, Z. D., and Lee, W. H. (1997b). HEC binds to the seventh regulatory subunit of the 26 S proteasome and modulates the proteolysis of mitotic cyclins. J. Biol. Chem 272, 24081–24087.
Ciferri, C., De Luca, J., Monzani, S., Ferrari, K. J., Ristic, D., Wyman, C., Stark, H., Kilmartin, J., Salmon, E. D., and Musacchio, A. (2005). Architecture of the human ndc80-hec1 complex, a critical constituent of the outer kinetochore. J. Biol. Chem 280, 29088–29095.
Ciferri, C. et al. (2008). Implications for kinetochore-microtubule attachment from the structure of an engineered Ndc80 complex. Cell 133, 427–439.[CrossRef][Medline]
Compton, D. A. (2000). Spindle assembly in animal cells. Annu. Rev. Biochem 69, 95–114.[CrossRef][Medline]
DeLuca, J. G., Gall, W. E., Ciferri, C., Cimini, D., Musacchio, A., and Salmon, E. D. (2006). Kinetochore microtubule dynamics and attachment stability are regulated by Hec1. Cell 127, 969–982.[CrossRef][Medline]
DeLuca, J. G., Howell, B. J., Canman, J. C., Hickey, J. M., Fang, G., and Salmon, E. D. (2003). Nuf2 and Hec1 are required for retention of the checkpoint proteins Mad1 and Mad2 to kinetochores. Curr. Biol 13, 2103–2109.[CrossRef][Medline]
Diaz-Rodriguez, E., Sotillo, R., Schvartzman, J. M., and Benezra, R. (2008). Hec1 overexpression hyperactivates the mitotic checkpoint and induces tumor formation in vivo. Proc. Natl. Acad. Sci. USA 105, 16719–16724.
Durfee, T., Becherer, K., Chen, P. L., Yeh, S. H., Yang, Y., Kilburn, A. E., Lee, W. H., and Elledge, S. J. (1993). The retinoblastoma protein associates with the protein phosphatase type 1 catalytic subunit. Genes Dev 7, 555–569.
Emanuele, M. J., McCleland, M. L., Satinover, D. L., and Stukenberg, P. T. (2005). Measuring the stoichiometry and physical interactions between components elucidates the architecture of the vertebrate kinetochore. Mol. Biol. Cell 16, 4882–4892.
Glinsky, G. V., Berezovska, O., and Glinskii, A. B. (2005). Microarray analysis identifies a death-from-cancer signature predicting therapy failure in patients with multiple types of cancer. J. Clin. Invest 115, 1503–1521.[CrossRef][Medline]
Goshima, G., Mayer, M., Zhang, N., Stuurman, N., and Vale, R. D. (2008). Augmin: a protein complex required for centrosome-independent microtubule generation within the spindle. J. Cell Biol 181, 421–429.
Goshima, G., Wollman, R., Goodwin, S. S., Zhang, N., Scholey, J. M., Vale, R. D., and Stuurman, N. (2007). Genes required for mitotic spindle assembly in Drosophila S2 cells. Science 316, 417–421.
Gosti-Testu, F., Marty, M. C., Berges, J., Maunoury, R., and Bornens, M. (1986). Identification of centrosomal proteins in a human lymphoblastic cell line. EMBO J 5, 2545–2550.[Medline]
Gurzov, E. N., and Izquierdo, M. (2006). RNA interference against Hec1 inhibits tumor growth in vivo. Gene Ther 13, 1–7.[CrossRef][Medline]
Holmfeldt, P., Brattsand, G., and Gullberg, M. (2003). Interphase and monoastral-mitotic phenotypes of overexpressed MAP4 are modulated by free tubulin concentrations. J. Cell Sci 116, 3701–3711.
Hori, T., Haraguchi, T., Hiraoka, Y., Kimura, H., and Fukagawa, T. (2003). Dynamic behavior of Nuf2-Hec1 complex that localizes to the centrosome and centromere and is essential for mitotic progression in vertebrate cells. J. Cell Sci 116, 3347–3362.
Karsenti, E., and Vernos, I. (2001). The mitotic spindle: a self-made machine. Science 294, 543–547.
Kline-Smith, S. L., Sandall, S., and Desai, A. (2005). Kinetochore-spindle microtubule interactions during mitosis. Curr. Opin. Cell Biol 17, 35–46.[CrossRef][Medline]
Kline-Smith, S. L., and Walczak, C. E. (2004). Mitotic spindle assembly and chromosome segregation: refocusing on microtubule dynamics. Mol. Cell 15, 317–327.[CrossRef][Medline]
Lawo, S. et al. (2009). HAUS, the 8-subunit human Augmin complex, regulates centrosome and spindle integrity. Curr. Biol 19, 816–826.[CrossRef][Medline]
Li, L., Yang, L., Scudiero, D. A., Miller, S. A., Yu, Z. X., Stukenberg, P. T., Shoemaker, R. H., and Kotin, R. M. (2007). Development of recombinant adeno-associated virus vectors carrying small interfering RNA (shHec1)-mediated depletion of kinetochore Hec1 protein in tumor cells. Gene Ther 14, 814–827.[CrossRef][Medline]
Lin, Y. T., Chen, Y., Wu, G., and Lee, W. H. (2006). Hec1 sequentially recruits Zwint-1 and ZW10 to kinetochores for faithful chromosome segregation and spindle checkpoint control. Oncogene 25, 6901–6914.[CrossRef][Medline]
Luders, J., Patel, U. K., and Stearns, T. (2006). GCP-WD is a gamma-tubulin targeting factor required for centrosomal and chromatin-mediated microtubule nucleation. Nat. Cell Biol 8, 137–147.[CrossRef][Medline]
Martin-Lluesma, S., Stucke, V. M., and Nigg, E. A. (2002). Role of Hec1 in spindle checkpoint signaling and kinetochore recruitment of Mad1/Mad2. Science 297, 2267–2270.
Mayer, L., Fu, S. M., and Kunkel, H. G. (1982). Human T cell hybridomas secreting factors for IgA-specific help, polyclonal B cell activation, and B cell proliferation. J. Exp. Med 1, 1860–1865.
McCleland, M. L., Gardner, R. D., Kallio, M. J., Daum, J. R., Gorbsky, G. J., Burke, D. J., and Stukenberg, P. T. (2003). The highly conserved Ndc80 complex is required for kinetochore assembly, chromosome congression, and spindle checkpoint activity. Genes Dev 17, 101–114.
Sauer, G., Korner, R., Hanisch, A., Ries, A., Nigg, E. A., and Sillje, H. H. (2005). Proteome analysis of the human mitotic spindle. Mol Cell Proteomics 4, 35–43.
Sotillo, R., Hernando, E., Diaz-Rodriguez, E., Teruya-Feldstein, J., Cordon-Cardo, C., Lowe, S. W., and Benezra, R. (2007). Mad2 overexpression promotes aneuploidy and tumorigenesis in mice. Cancer Cell 11, 9–23.[CrossRef][Medline]
Uehara, R., Nozawa, R. S., Tomioka, A., Petry, S., Vale, R. D., Obuse, C., and Goshima, G. (2009). The augmin complex plays a critical role in spindle microtubule generation for mitotic progression and cytokinesis in human cells. Proc. Natl. Acad. Sci. USA 106, 6998–7003.
van 't Veer, L. J. et al. (2002). Gene expression profiling predicts clinical outcome of breast cancer. Nature 415, 530–536.[CrossRef][Medline]
Wei, R. R., Al-Bassam, J., and Harrison, S. C. (2007). The Ndc80/HEC1 complex is a contact point for kinetochore-microtubule attachment. Nat. Struct. Mol. Biol 14, 54–59.[CrossRef][Medline]
Wei, R. R., Sorger, P. K., and Harrison, S. C. (2005). Molecular organization of the Ndc80 complex, an essential kinetochore component. Proc. Natl. Acad. Sci. USA 102, 5363–5367.
Wiese, C., and Zheng, Y. (2006). Microtubule nucleation: gamma-tubulin and beyond. J. Cell Sci 119, 4143–4153.
Wigge, P. A., Jensen, O. N., Holmes, S., Soues, S., Mann, M., and Kilmartin, J. V. (1998). Analysis of the Saccharomyces spindle pole by matrix-assisted laser desorption/ionization (MALDI) mass spectrometry. J. Cell Biol 141, 967–977.
Wilson-Kubalek, E. M., Cheeseman, I. M., Yoshioka, C., Desai, A., and Milligan, R. A. (2008). Orientation and structure of the Ndc80 complex on the microtubule lattice. J. Cell Biol 182, 1055–1061.
Wong, J. et al. (2007). A protein interaction map of the mitotic spindle. Mol. Biol. Cell 18, 3800–3809.
Wu, G., Lee, W. H., and Chen, P. L. (2000). NBS1 and TRF1 colocalize at promyelocytic leukemia bodies during late S/G2 phases in immortalized telomerase-negative cells. J. Biol. Chem 275, 30618–30622.
Wu, G., Lin, Y. T., Wei, R., Chen, Y., Shan, Z., and Lee, W. H. (2008a). Hice1, a novel microtubule-associated protein required for maintenance of spindle integrity and chromosomal stability in human cells. Mol. Cell. Biol 28, 3652–3662.
Wu, G., Qiu, X. L., Zhou, L., Zhu, J., Chamberlin, R., Lau, J., Chen, P. L., and Lee, W. H. (2008b). Small molecule targeting the Hec1/Nek2 mitotic pathway suppresses tumor cell growth in culture and in animal. Cancer Res 68, 8393–8399.
Xiao, J., Liu, C. C., Chen, P. L., and Lee, W. H. (2001). RINT-1, a novel Rad50-interacting protein, participates in radiation-induced G(2)/M checkpoint control. J. Biol. Chem 276, 6105–6111.
Zhu, H., Coppinger, J. A., Jang, C. Y., Yates, J. R., 3rd, and Fang, G. (2008). FAM29A promotes microtubule amplification via recruitment of the NEDD1-gamma-tubulin complex to the mitotic spindle. J. Cell Biol 183, 835–848.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
30 randomly selected mitotic cells with a near bipolar spindle (three independent experiments) and used to fit a growth curve according to a multifactor exponential equation (SigmaPlot, Systat Software, San Jose, CA). The 12-min time point was not measured as the asters had become less distinct. AU, arbitrary units. (C) p values derived from a two-way ANOVA test (factor 1, time; factor 2, siRNAs) to show differences in the mean levels of microtubule aster intensity between various siRNA (factor 2) treatment groups. Differences among different time points (factor 1) are all significant (p < 0.002, data not shown).