KIFC1 Is Essential for Bipolar Spindle Formation and Genomic Stability in the Primary Human Fibroblast IMR-90 Cell

Abstract

Kinesin family member C1 (KIFC1) is the only member of the minus-end-directed kinesin-14 family in human cells. In cancer cells, KIFC1 plays an essential role in bipolar spindle formation by clustering the multiple poles during mitosis. However, it has not been clearly demonstrated whether KIFC1 also functions to mediate bipolar spindle formation and to maintain genomic stability in normal cells. In this study, by using human primary lung fibroblast IMR-90 cells, we showed that KIFC1 knock-down with lentiviral KIFC1 shRNA induced 17% of cells with multiple microtubule organizing centers (MTOCs) and delayed cyclin A degradation for more than 2 hr in early mitosis. However, these cells eventually carried out mitosis, resulting in 24% of cells with lagging chromosomes and 9% of cells with micronuclei after mitosis. Karyotyping of KIFC1-depleted IMR-90 cells demonstrated that cells with various abnormal numbers of chromosomes are produced. When IMR-90 cells treated with KIFC1 or the control shRNA for 60 hr were compared, 20% less cells were observed in KIFC1-depleted cells without an obvious immediate cell death. As reported for Mad2 depletion in IMR-90 cells, KIFC1-depleted IMR-90 cells showed typical features of senescence, like senescence-associated (SA) β-galactosidase expression, when incubated 6 days or more. However, IMR-90 cells knocked down with both KIFC1 and Mad2 underwent apoptosis, suggesting that KIFC1 and Mad2 likely function in different pathways during mitosis. Taken together, we suggest that KIFC1 plays an essential role for bipolar MTOC formation and maintaining chromosomal stability in the mitosis of human primary fibroblast IMR-90.

Introduction

During mitosis, cells form bipolar spindle for equal segregation of duplicated chromosomes to maintain genomic stability. However, bipolar spindle formation is perturbed and multipolar spindles are formed when centrosomes are overproduced or fragmented. Cells with multiple microtubule organizing centers (MTOCs) are observed in various cancer cells with chromosome instability (Meraldi et al., 2002), which cannot be monitored by spindle assembly checkpoint (SAC) (Gisselsson et al., 2010). If cells fail to recover bipolarity, chromosome instability may happen by increasing possibility of merotelic attachment (Ganem et al., 2009). Failure of equal chromosome segregation results in lagging chromosomes during anaphase (Cimini et al., 2002), and small DNA-containing structures called micronuclei besides each nucleus (El-Zein et al., 2006; Rao et al., 2008). Although it is still not well understood whether micronuclei are finally eliminated or re-enter into the main nucleus, micronuclei-containing cells often produce micronucleated cells during subsequent mitosis (Utani et al., 2010).

Motor proteins have been reported to play important roles in bipolar spindle formation. They mediate separation of duplicated centrosomes at the entry of mitosis and bipolar spindle assembly (Walczak et al., 1998; Quintyne et al., 2005). Eg5, a well-known plus-end directed kinesin-5 motor, plays key roles in bipolar spindle formation. Localizing between antiparallel microtubules, Eg5 controls centrosome separation by making pushing forces (Tanenbaum et al., 2008). Minus-end directed motor dynein is also suggested to participate in centrosome separation. Being present between centrosomes, dynein pulls centrosome together (Tanenbaum et al., 2008). In addition, dynein becomes localized at cell cortex and contributes to its lateral interactions with microtubules to control centrosome separation (Zhu et al., 2010). Dynein on cell cortex has also suggested to control centrosome movement by shrinking microtubule tips (Laan et al., 2012).

Kinesin-14 family is distinct from the other kinesins because it has minus-end-directed motility (Fink et al., 2009). Ncd, the kinesin-14 in Drosophila, is well known for its role in centrosome coalescence (Basto et al., 2008). In mammalian cells, only KIFC1 belongs to this family (Dagenbach and Endow, 2004; Lawrence et al., 2004). Localizing on the mitotic spindle (Quintyne et al., 2005), KIFC1 manages spindle length both in mitosis and meiosis using its sliding activity along microtubules (Cai et al., 2009). In addition, KIFC1 mediates proper cytokinesis by organizing and stabilizing spindles (Cai et al., 2010). Depletion of KIFC1 induced multipolar spindle and genomic instability when large-scale analysis of human kinesin esiRNA library was performed in HeLa cells (Zhu et al., 2005). The importance of KIFC1 for bipolar spindle formation was revealed when people found that KIFC1 has an opposite role to Eg5. Eg5 inhibition results in monopolar spindle and a failure of centrosome separation, but the double inhibition of KIFC1 and Eg5 restores Eg5 knock-down phenotypes (Mountain et al., 1999). A recent paper reported that KIFC1 is essential for the viability of certain cancer cells that contain multiple centrosomes (Kwon et al., 2008). KIFC1 is suggested to play key roles in clustering multiple MTOCs of cancer cells, which usually show centrosome amplifications (Pihan et al., 1998). By this function, KIFC1 maintains spindle bipolarity and finishes mitosis in cancer cells with genomic instability (Murphy, 2003). KIFC1 has been suggested to be a good target of cancer therapy, because cancer cells are unviable when they fail to re-cluster multiple MTOCs and undergo multipolar cell division (Ganem et al., 2009).

Regardless of its interest in cancer cells, the function of KIFC1 in bipolar spindle formation and mitotic progression is not well defined in normal human cells. In this study, we demonstrate that KIFC1 has an important role in maintaining genomic stability by regulating bipolar spindle formation during mitosis in human primary IMR-90 cells that do not have chromosome instability.

Results

KIFC1 depletion induces multiple MTOCs during mitosis in human lung fibroblast IMR-90 cells

To assess the role of KIFC1 in the mitosis of normal human cells, we first analyzed its cellular localization throughout cell cycle in primary human lung fibroblast IMR-90 cells. We fixed and co-stained cells for KIFC1, γ-tubulin, and DAPI. KIFC1 was present throughout the cytoplasm during interphase and translocated into the nucleus just before mitosis when the centrosomes were separated (Fig. 1A). KIFC1 localized to the mitotic spindle from prophase to telophase (Fig. 1B).

Fig. 1

Localization of KIFC1 during cell cycle in the primary IMR-90 cells. Localization of KIFC1 during cell cycle was examined by immunofluorescence microscopy. IMR-90 cells were immunostained for KIFC1 (green) and γ-tubulin (red). The nucleus was stained by DAPI (blue). Scale bar, 10 μm. (A) Interphase; (B) Mitosis.

Previous studies showed that HeLa cells of KIFC1 knock-down have multipolar spindle and KIFC1 functions in centrosome clustering to prevent multipolar spindles in human MDA-MB-231 breast cancer cells (Quintyne et al., 2005; Zhu et al., 2005). For the functional study of KIFC1, we constructed lentiviral KIFC1 shRNA as described in Materials and Methods and tested the knock-down efficiency (Fig. 2A, Supplementary Fig. S1A). We reassured multipolar spindle and mitotic delay in KIFC1-depleted MDA-MB-231 breast cancer cells (Supplementary Fig. S1A).

Fig. 2

Knock-down of KIFC1 induces multiple MTOCs and delays early mitosis in the primary IMR-90 cells. (A) The knock-down of KIFC1 by shRNA was confirmed by immunoblot. Control (left) and KIFC1-depleted IMR-90 cells were arrested by single thymidine after KIFC1 shRNA or control lentiviral infection, and KIFC1 expression was examined 9–12 hr after the release. To assess the mitotic progression of these cells, cyclin A expression was monitored by immunoblot. Actin was used as a loading control. (B) The multiple MTOC phenotypes of the KIFC1-depleted IMR-90 cells were examined by immunofluorescence microscopy for α-tubulin (green) and centrin (red). The nucleus was stained by DAPI (blue). Scale bar, 10 μm. (C) The multiple MTOC phenotypes of the KIFC1-depleted IMR-90 cells were examined by immunofluorescence microscopy for γ-tubulin (green) and centrin (red). The nucleus was stained by DAPI (blue). Scale bar, 10 μm. (D) The cells from (C) with multiple MTOCs were counted (both control and KIFC1-depleted cells (right)). Cells with more than three γ-tubulin stained spots were counted for multiple MTOCs. Three independent experiments were performed, and 80 cells were counted in each experiment.

We then examined the expression of endogenous KIFC1 during cell cycle in the primary IMR-90 cells. IMR-90 cells infected with a GFP control virus for 60 hr were used to compare with the cells of KIFC1 knock-down by lentiviral KIFC1 shRNA. These cells were arrested by double thymidine block and were released. Immunoblots demonstrated that KIFC1 gradually accumulated and was present at the highest levels in early mitosis (Fig. 2A, left). During mitosis, KIFC1 began to decrease as cyclin A diminished (Fig. 2A, left).

We observed KIFC1-depleted IMR-90 cells with lentiviral shRNA to evaluate the function of KIFC1 during cell cycle. When cells were released from the thymidine arrest, KIFC1 expression was not observed, and cyclin A degradation was delayed by 2 hr at least (Fig. 2A, right): cyclin A was decreased at 11 hr after the release in control but was still presented at 12 hr after the release in KIFC1-depleted cells. When we extended the time after the release to 15 hr, we observed the decrease of cyclin A in 14 hr in KIFC1-depleted cells, demonstrating that it was delayed approximately 2 hr (Supplementary Fig. S1A). This result suggests that KIFC1 depletion likely delay cell cycle during early mitosis in the primary IMR-90 cells as in cancer cells. We also examined the phenotype of KIFC1-depleted IMR-90 cells. As reported in previous studies of KIFC1-depleted MDA-MB-231 breast cancer and HeLa cells (Supplementary Fig. S1B, data not shown), multipolar spindle was observed in KIFC1-depleted IMR-90 cells by γ-tubulin and centrin immunofluorescence microscopy (Fig. 2B). Multiple MTOCs are not always derived from multiple centrosomes (Holmfeldt et al., 2005) and non-centrosome-associated microtubule clusters including NuMA and γ-tubulin have been reported to contribute to spindle formation (Tulu et al., 2003). In order to confirm that each pole of the multiple MTOCs observed in KIFC1-depleted IMR-90 cells was derived from centrosomes, we examined the co-localization of γ-tubulin with centrin, a marker for centrioles (Paoletti et al., 1996). Both γ-tubulin and centrin were detected on each spindle pole of multiple MTOCs induced by KIFC1-depletion (Fig. 2C). These data demonstrated that each pole of multiple MTOCs holds centrosome-associated microtubules in KIFC1-depleted IMR-90 cells.

When we counted the number of cells with multiple MTOCs in IMR-90 cells, 17% of the KIFC1-depleted cells displayed more than two spindle poles and γ-tubulin signals (Fig. 2D). Since IMR-90 cells are primary, the control virus-treated IMR-90 cells did not contain any multiple MTOCs (Fig. 2D). These observations demonstrated that KIFC1 depletion induces multiple MTOCs, not only in cancer cells but also in the primary IMR-90 cells. Taken together, these data suggest that KIFC1 is important for early mitotic progression and plays a role in bipolar spindle formation in human primary fibroblasts.

KIFC1 depletion induces lagging chromosomes and micronuclei

Lagging chromosomes are often formed when there are defects in kinetochore-microtubule binding or merotelic attachment (Cimini et al., 2002). It has been reported that micronuclei are often derived from lagging chromosomes (Rao et al., 2008). A previous study suggested that KIFC1-depleted cells with misaligned chromosomes could enter the anaphase and finish mitosis in HeLa cells (Zhu et al., 2005). Lagging chromosomes and micronuclei were observed by this cell division. Our time-lapse images using MDA-MB-231 breast cancer cells suggest that KIFC1-depleted cells with multiple MTOCs show three different fates (Supplementary Movie S1–4). In some, multiple MTOCs were clustered and reformed bipolar or pseudo-bipolar spindles, in which the multiple poles were arranged bi-directionally to segregate chromosomes into two opposite directions in two daughter cells with mitotic delays (Supplementary Fig. S2B, Supplementary Movie S2). In the second group, cells had multiple MTOCs and failed to form bipolar spindles but still divided into two daughter cells (Supplementary Fig. S2C, Supplementary Movie S3). In this case, the cells failed their attempt to fix multiple MTOCs by forming bipolar spindles or by arranging them bi-directionally, but eventually underwent cell division. As a result, these cells appeared to be torn out physically. Another very rare phenotype occurred when cells divided into more than three daughter cells (Supplementary Fig. S2D, Supplementary Movie S4). These cells failed to bundle spindles and formed a pseudo-bipole. They eventually formed multiple daughter cells that matched the number of poles.

We then asked whether KIFC1 depletion might lead to lagging chromosomes and micronuclei in IMR-90 cells due to improper mitotic division. IMR-90 is a primary cell and generally does not produce lagging chromosomes or mitotic defects. Control virus or KIFC1 shRNA virus-infected IMR-90 cells were arrested with thymidine treatment, released for 11–12 h, and observed by immunofluorescence microscopy. We carefully examined lagging chromosomes in KIFC1-depeleted IMR-90 cells with multiple MTOCs after immunostaining for α-tubulin and DNA. 24% of KIFC1-depleted IMR-90 cells showed lagging chromosomes in anaphase (Fig. 3A). We also found micronuclei in KIFC1-depleted IMR-90 cells, which are the hallmark of genomic instability (Fig. 3B). About 9% of the KIFC1-depleted IMR-90 cells contained micronuclei. Importantly, we did not observe any lagging chromosomes or micronuclei in the control virus-infected IMR-90 cells. Taken together, these data suggest that KIFC1 depletion induces lagging chromosomes and micronuclei formation, possibly by executing mitosis with an incomplete bipolar spindle. Thus, genomic instability can be increased by KIFC1 depletion through lagging chromosomes and micronuclei in the primary IMR-90 cells.

Fig. 3

KIFC1-depletion induces chromosome instability in IMR-90 cells. IMR-90 cells were infected with lentiviral KIFC1 shRNA. Cells were stained for α-tubulin (green) and DAPI (blue) for immunofluorescence microscopy. The phenotype (left) and quantitation of lagging and micronuclei formation (right) are presented together. (A) Lagging chromosomes in IMR-90 cells are shown. (B) Micronuclei were observed in IMR-90 cells. For each experiment, n=100. Three independent experiments were performed, and error bars represent standard deviations. *, P<0.05; **, P<0.05. (C, D, E) IMR-90 cells were infected with lentiviral KIFC1 shRNA, and metaphase cells were prepared for karyotype analysis as described in the Materials and Methods. (C) Light microscopic images are shown for characteristic metaphase spread in a control (left) and a KIFC1-depleted cell (right). Scale bar, 10 μm. (D, E) Chromosome number distribution of control (D) and KIFC1-depleted cells (E) are shown as percentages. For this, 100 metaphase cells were analyzed in the control and KIFC1-depleted cell groups.

KIFC1 is required for proper chromosomal segregation and genomic stability

Lagging chromosomes randomly move into daughter cells resulting in abnormal chromosome numbers (Tutt et al., 1999). Because a portion of KIFC1-depleted cells exhibited lagging chromosomes and micronuclei, we suggested that KIFC1 depletion induces genomic instability. To examine the relationship between KIFC1 depletion and increased genomic instability in IMR-90 cells, we counted the number of chromosomes on the metaphase spread using a trypsin-Giemsa banding technique (Pan et al., 2009). We infected IMR-90 cells with lentiviral KIFC1 shRNA and synchronized with a single thymidine treatment. The cells were harvested at 9.5 hr after the release to maximize the number of mitotic cells. Chromosomes were stained by Giemsa and were observed under a light microscope. Fig. 3C presents the representative KIFC1-depleted or the control virus-treated IMR-90 cells in metaphase, and the distribution of chromosome numbers in these cells is shown in Fig. 3D and 3E. Consistent with increased lagging chromosomes and micronuclei, the depletion of KIFC1 resulted in an increase in cells with abnormal chromosome numbers (Fig. 3E). Almost all control cells (>95%) had normal 46 chromosomes (Fig. 3D). However, only 46% of KIFC1-depleted IMR-90 cells contained 46 chromosomes (Fig. 3E). In addition, KIFC1-depleted cells more often had reduced numbers of chromosomes rather than increased. Thus, KIFC1 depletion induced the gain or loss of chromosomes in the primary IMR-90 cells.

Depletion of KIFC1 in IMR-90 cells decreases proliferation and leads to senescence

A previous study demonstrated that KIFC1-depleted cancer cells undergo apoptosis, thereby suggesting KIFC1 a good anti-cancer drug target (Kwon et al., 2008). Since genomic instability is induced in KIFC1-depleted IMR-90 cells, we examined the growth and death of these cells. We counted cell number every 12 hr for 60 hr after lentiviral KIFC1 shRNA or control virus infection. Compared to the control virus-treated cells, the number of cells in KIFC1-depleted IMR-90 was decreased by 20% (Fig. 4A). However, when we examined the possible apoptosis of KIFC1-depleted IMR-90 cells by FACS, we could not detect any obvious evidence for cell death (Fig. 4B). Meanwhile, the control IMR-90 cells for apoptosis (cells irradiated with UV) showed a fraction of cells with fragmented DNA content less than 2N (Fig. 4B).

Fig. 4

KIFC1-depleted IMR-90 cells decrease proliferation and lead to senescence. IMR-90 cells were infected with lentiviral shRNA of KIFC1 or Mad2 or both KIFC1 and Mad2. (A) Cells were collected and counted every 12 hr for 60 hr after lentiviral infection. (B) Cells were collected 48 hr and 60 hr after lentivirus treatment. The DNA content of control virus-treated and KIFC1-depleted IMR-90 cells was measured by flow cytometric analysis. For each, 10,000 cells were counted. For the positive control of apoptosis, IMR-90 cells were harvested 24 hr after UV exposure (40 J/m²). The fraction of cells with 2N and 4N DNA content is indicated. (C) Control virus or KIFC1 shRNA-treated cells were collected for senescence-associated (SA) β-gal assay. Images (200× magnification) were taken at the 6th day after virus infection (left). Bar graph (n=200) shows the percentage of SA β-gal positive cells collected at the 1st, 4th, and 6th day after virus treatment (right). IMR-90 cells treated with 100 μM H₂O₂ was used for the control of positive senescence. (D) Cells were collected 48 hr after lentiviral treatment. Apoptotic cells were counted by flow cytometric analysis after Annexin V/propidium iodide staining. For the positive control of apoptosis, IMR-90 cells were harvested 24 hr after UV exposure (40 J/m²). For each FACS, 30,000 cells were counted. Lower left quadrant represents live cells, lower right the early apoptotic cells, and upper right the late apoptotic cells. The percentage of cells in each category is denoted.

Since KIFC1-depleted IMR-90 cells showed decreased proliferation without any sign of apoptosis, we examined whether these cells undergo senescence. When KIFC1 shRNA lentivirus-treated cells were cultured more than six days after virus infection, we observed senescence-associated (SA) β-galactosidase expression in KIFC1-depleted cells. Both the cell density and the percentage of cells with SA β-galactosidase expression were continuously increased in KIFC1-depleted cells with extended culture days (Fig. 4C). On the other hand, we could not detect any SA β-galactosidase expression in control virus-treated cells with extended culture (Fig. 4C). This observation suggests that genomic instability by KIFC1 depletion decrease cell growth rate by leading cells to senescence, but do not immediately induce apoptosis in the primary IMR-90 cells.

IMR-90 cells normally activate spindle assembly checkpoint (SAC) with functional Mad2 expression (Amato et al., 2009), and cell senescence was observed in Mad2-depleted IMR-90 cells (Lentini et al., 2012). However, previous study showed that spindle assembly checkpoint (SAC) does not monitor multipolar spindle (Gisselsson et al., 2010). KIFC1-depleted IMR-90 cells showed approximately 2 hr delay of cyclin A degradation (Fig. 2A), but undertook mitosis with multiple MTOCs to produce cells with lagging chromosomes and micronuclei (Fig. 3). To understand a possible relationship between the multipolar spindle induced by KIFC1 depletion and SAC, we knocked-down both KIFC1 and Mad2, a major component of SAC, in IMR-90 cells by shRNA (Fig. 4A). Mad2-depleted IMR-90 cells showed decreased cell growth rate, as reported previously (Lentini et al., 2012). Unlike KIFC1-depleted or Mad2-depleted cells, cells blocked with both KIFC1 and Mad2 showed sudden drop of proliferation after 36 hr (Fig. 4A). Our FACS analysis after Annexin V/propidium iodide staining supported that co-depletion of KIFC1 and Mad2 induces apoptosis in IMR-90 cells (Fig. 4D). These observations suggest that genomic instability by KIFC1 depletion might not be controlled by SAC because KIFC1 and Mad2 co-depleted cells showed apoptotic event, which was not detected in KIFC1 or Mad2 single knock-downed cells. If KIFC1 and Mad2 functioned in the same pathway, the phenotype of double knock-down of KIFC1 and Mad2 would be similar to that of each single knock-down.

Discussion

Our results first verify that the depletion of KIFC1 produces multiple MTOCs and delays early mitosis in the human primary fibroblasts. We also show that KIFC1 plays an essential role in maintaining genomic stability during mitosis in the human primary cell. When KIFC1 was completely depleted, 17% of IMR-90 cells had multiple MTOCs (Fig. 2D), strongly supporting that KIFC1 functions in bipole formation during early mitosis of the primary cell. This function of KIFC1 is consistent with a 2 hr delay prior to cyclin A degradation observed in KIFC1-depleted IMR-90 cells. However, KIFC1-depleted IMR-90 cells undergo mitosis with multiple MTOCs and thereby induce genomic instability likely due to chromosome segregation with pseudo-bipole or multiple spindle poles as monitored by live images in cancer cells. Lagging chromosomes and micronuclei observed in KIFC1-depleted IMR-90 cells are markers for chromosome instability and are correlated with multi-polarity during mitosis. Interestingly, when cell proliferation was monitored, we detected 20% decrease in cell number but no obvious cell death for 60 hr after KIFC1 depletion in IMR-90 cells (Fig. 4). With extended incubation for six or more days, we observed typical features of senescence in these cells (Fig. 4C). Thus, the less number of cells by KIFC1 depletion may be due to the slow-down of 2 hr delay in early mitosis and/or the diminished cell proliferation by senescence. Considering that DNA damage is a major source of cell senescence (Bartkova et al., 2006), it is plausible that accumulation of genomic instability by KIFC1 depletion rather activates a senescence pathway than induces apoptosis.

We observed multiple MTOCs in KIFC1-depleted IMR-90 cells. Considering that IMR-90 cells are human primary fibroblasts, we assume that there is no centrosome over-duplication. In fact, we could not observe any IMR-90 cells with amplified centrosomes (data not shown). In addition, most functions of KIFC1 are known in mitosis and not in centrosome duplication. Thus, the multiple MTOCs observed in KIFC1-depleted IMR-90 cells are likely due to centrosome fragmentation.

The spindle assembly checkpoint (SAC) is a surveillance mechanism that monitors the kinetochore-microtubule attachment in eukaryotes (Mansfeld et al., 2011). IMR-90 cells normally activate SAC in response to microtubule defects with functional Mad2 expression (Amato et al., 2009). Nevertheless, Cimini et al. suggested that merotelic attachment, which is a major cause of multipolar spindle, cannot be detected by the SAC. Cells with merotelic attachment proceed mitosis without delay (Cimini et al., 2002). In addition, the activated SAC by kinetochore-spindle mis-attachment is known to be maintained for 4 to 10 hr before cells undergo apoptosis (Chin and Herbst, 2006). However, our live cell images of KIFC1-depleted cells showed that the cells with multiple MTOCs delayed early mitosis for approximately 2 hr and underwent cell division, suggesting that SAC is unlikely to detect multipolar spindle induced by KIFC1 depletion. To further understand whether defects by KIFC1 depletion and SAC may function in the same pathway, we compared the effect of KIFC1 and Mad2 double knock-down with the knock-down of each. Unlike KIFC1-depleted or Mad2-depleted cells, apoptosis is induced in IMR-90 cells blocked with both KIFC1 and Mad2 (Fig. 4D). This result also suggests that genomic instability by KIFC1 depletion may not be controlled by SAC. If KIFC1 and Mad2 functioned in the same pathway, the phenotype of double knock-down of KIFC1 and Mad2 would likely be similar to that of each single knock-down. Considering that depletion of both KIFC1 and Mad2 leads to apoptosis while knock-down of each produces senescent cells, apoptosis might be induced by inactivating the SAC in primary fibroblasts with genetic instability as in KIFC1-depleted cells.

Only 17% of KIFC1-depleted IMR-90 cells showed multiple MTOCs, suggesting that other motor proteins are likely function in bipole formation in the absence of KIFC1 in the primary human cell. Elucidating the mechanism of how KIFC1 mediates the control of bipolar spindle formation would require further investigation, as the mechanism of bipolar spindle formation has not been clearly defined yet. Motor proteins have been suggested to play an important role by controlling the separation of duplicated centrosomes and the spindle assembly (Walczak et al., 1998; Quintyne et al., 2005). In particular, dynein functions in centrosome separation through its minus-end-directed motor activity by pulling centrosomes together (Tanenbaum et al., 2008). In addition, lateral interactions between cell cortex-localized dynein and microtubules have been reported to mediate centrosome separation (Zhu et al., 2010). A recent study also suggested that cortex-localized dynein interacts with shrinking microtubule tips and generates forces to mediate centrosome movement (Laan et al., 2012). Considering that KIFC1 is a minus-directed motor protein, it is tempting to speculate that KIFC1 may function similarly as dynein in centrosome separation by pulling centrosomes together in the primary cell, although the localization of KIFC1 in the cortex has not been detected yet.

Materials and Methods

Cell culture and synchronization

IMR-90 cells were maintained at 37°C with 5% CO₂ in MEM containing 10% fetal bovine serum, 100 units/mL of penicillin, 100 μg/mL of streptomycin, and 0.25 μg/mL of amphotericin B (Gibco). MDA-MB-231 cells were maintained at 37°C with 5% CO₂ in DMEM containing 10% fetal bovine serum, 100 units/mL of penicillin, 100 μg/mL of streptomycin, and 0.25 μg/mL of amphotericin B (Gibco). Cells were synchronized by thymidine (2 mM) arrest (Roberts et al., 2002).

Lentiviral shRNA construction and infection

The KIFC1 target sequence (5′ AACGTTGGACCAAGAGAA-CCA 3′) was subcloned into the pLKO.1 puro lentiviral vector (Addgene, Cambridge, MA, USA) using following primers: 5′ CCGGAACGTTGGACCAAGAGAACCACTCGAGTGGTTCT-CTTGGTCCAACGTTTTTTTG 3′ (forward), 5′AATTCAAAAA-AACGTTGGACCAAGAGAACCACTCGAGTGGTTCTCTTG-GTCCAACGTT 3′ (reverse). The lentiviral vector for Mad2 was purchased from Sigma-Aldrich (Mission shRNA SHCLNG-NM002358). The KIFC1 shRNA virus was produced in 293T cells by co-transfection of KIFC1/pLKO.1 with pMD2.G and psPAX2 (Sigma-Aldrich, St. Louis, MO, USA). Mad2 shRNA virus was produced by co-transfecting the Mad2 vector with pMD2.G and psPAX2 (Sigma-Aldrich) into 293T cells. The transfection was carried out using linear polyethylenimine (L-PEI; Polysciences Asia-Pacific, Inc., Taipei, Taiwan). The Mission Turbo GFP vector (Sigma-Aldrich) was used to produce the GFP control virus, and the pLKO.1 puro vector was used to produce the empty vector control virus. The virus was harvested at 48 and 72 hr after transfection. IMR-90 cells were infected with the shRNA embedded lentivirus for 12 hr for protein depletion. The infection efficiency was assessed by the GFP control lentivirus. Viral titers were checked by a crystal violet stain (Campeau et al., 2009). The GFP virus was used as a control for immunoblots, and the empty virus vector was used as a control for immunofluorescence experiments.

Immunoblot

Cells scraped from the culture dish were washed with cold PBS and lysed for 1 hr with 1X RIPA buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM Na₂EDTA, 1 mM EGTA, 1% NP-40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM beta-glycerophosphate, 1 mM Na₃VO₄, 1 μg/ml leupeptin, 1 mM PMSF). The cellular extracts were then centrifuged for 20 min at 10,000 g to remove cellular debris. For immunoblots, the supernatants were mixed with 4X SDS sample buffer (250 mM Tris pH 6.8, 8% SDS, 40% glycerol, 20% β-mercaptoethanol 0.1% bromophenol blue), run on 10% SDS-PAGE gel, and blotted onto a nitrocellulose membrane. The proteins were detected with anti-KIFC1 (Abnova, Taipei, Taiwan), anti-cyclin A (Cell Signaling/Millipore, Billerica, MA, USA), and anti-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) antibodies. Horseradish peroxidase-conjugated anti-mouse and anti-rabbit (Jackson ImmunoResearch, West Grove, PA, USA) secondary antibodies were used.

Immunofluorescence microscopy

The cells were washed in PBS, fixed in ice-cold methanol for 5min, and rehydrated in PBS for 1min. The fixed cells were per-meabilized with PBS with 0.5% Triton X-100 for 5min, blocked in PBS-BSA (PBS, 3% BSA) for 30min, and incubated with primary antibody in PBS-BSA for 1 h. Primary antibodies used were human anti-α-tubulin (1:3000, Sigma-Aldrich), anti-γ-tubulin (1:3000, Sigma-Aldrich), and anti-centrin (1:6000, Dr. Ryoma. Ohi, Vanderbilt Univ.). All primary antibodies were detected using species-specific fluorescent secondary antibodies (Invitrogen, Carlsbad, CA, USA), and DNA was detected with 0.2 μg/ml of 14,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich). Cover-slips were mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA). Immunofluorescence images were collected at 488 and 594nm on Axio Imager A2 (Carl Zeiss, Zena, Germany) using a 40× 0.75NA EC Plan-Neofluar objective lens with an AxioCam Hrc CCD camera (Carl Zeiss). The acquisition parameters and focus were controlled by Axio-Vision software. Cells were also observed on a Zeiss LSM 510 META laser confocal microscope (Carl Zeiss), and 0.2 μm optical sections were acquired with 40× 1.2w Korr UV-VIS-IR lens. The acquisition parameters and focus were controlled by LSM 510 software. The 405 nM wavelength generated by Diode, 488 wavelength generated by Ar laser, and 543 nm wavelength generated by a He-Ne laser.

Flow cytometry

After trypsin-EDTA treatment, cells were harvested and fixed in 70% cold ethanol. Cells were washed in PBS, treated with 100 μg/ml RNase for 30 min in 37°C, and were stained with 0.5 μg propidium iodide (Sigma-Aldrich). To detect apoptotic cells with Annexin V, cells were washed in PBS and resuspended in 1x binding buffer (10 mM Hepes (pH 7.4), 140 mM NaCl, 2.5 mM CaCl₂). 5 μl FITC Annexin V (BD biosciences, California, USA) and 0.5 μg propidium iodide (Sigma-Aldrich) were used to detect cells in apoptotic status. Flow cytometry was carried out with FACScalibur (BD biosciences) and CellQuest software.

Senescence assay

Following the manufacturer’s protocol, senescence-associated β-galactosidase cell staining kit (Cell Signaling) was used to observe senescent cells. For positive control, cells were treated with 100 μM H₂O₂ for 2 hr (Duan et al., 2005).

Live cell imaging

MDA-MB-231 cells were cultured on a 35-mm glass-bottom culture dish (SPL Life Science Inc., Korea). α-tubulin-EGFP was transfected with 4.5 μg of plasmid in 1.5 μl of 1 mg/ml Lipo-fectamine 2000 (Invitrogen) according to the manufacturer’s instructions. The cells were infected with the appropriate shRNA lentivirus for each target gene. GFP images were obtained using PerkinElmer UltraView spinning disk confocal microscopy on a Nikon TE2000 inverted microscope. A series of 0.2 μm optical sections were acquired every 1.5 min for 2 hr with a 40×/1.2 NA Planapochromatic objective lens. The 488 nm wavelength generated by a krypton-argon laser was selected with a Chroma 488/10 nm Bandpass excitation filter. Z-series optical sections were obtained with a Hamamatsu ORCA ER-cooled charge-coupled device camera containing 1344×1024 pixels. The cells for microscopy were enclosed within a temperature and CO₂-controlled environment that maintained an atmosphere of 37°C and 5% humidified CO₂.

Acknowledgments

The authors appreciate Dr. Kyung S. Lee (NIH) and Dr. Yeon-Sun Sung (Dankook Univ., Korea) for help to construct KIFC1 shRNA lentivirus. We thank Dr. Ryoma.Ohi at Vanderbilt Univ. for kindly providing anti-centrin antibody. This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2009-0074480). N. Kim was partly supported by the fellowship of Brain Korea 21 (BK21) Program.

References

• Amato, A., Lentini, L., Schillaci, T., Iovino, F., and Di Leonardo, A. 2009. RNAi mediated acute depletion of retinoblastoma protein (pRb) promotes aneuploidy in human primary cells via micronuclei formation. BMC Cell Biol., 10: 79.
• Bartkova, J., Rezaei, N., Liontos, M., Karakaidos, P., Kletsas, D., Issaeva, N., Vassiliou, L.V., Kolettas, E., Niforou, K., Zoumpourlis, V.C., Takaoka, M., Nakagawa, H., Tort, F., Fugger, K., Johansson, F., Sehested, M., Andersen, C.L., Dyrskjot, L., Orntoft, T., Lukas, J., Kittas, C., Helleday, T., Halazonetis, T.D., Bartek, J., and Gorgoulis, V.G. 2006. Oncogene-induced senescence is part of the tumorigenesis barrier imposed by DNA damage checkpoints. Nature, 444: 633–637.
• Basto, R., Brunk, K., Vinadogrova, T., Peel, N., Franz, A., Khodjakov, A., and Raff, J.W. 2008. Centrosome amplification can initiate tumorigenesis in flies. Cell, 133: 1032–1042.
• Cai, S., Weaver, L.N., Ems-McClung, S.C., and Walczak, C.E. 2009. Kinesin-14 family proteins HSET/XCTK2 control spindle length by cross-linking and sliding microtubules. Mol. Biol. Cell, 20: 1348–1359.
• Cai, S., Weaver, L.N., Ems-McClung, S.C., and Walczak, C.E. 2010. Proper organization of microtubule minus ends is needed for midzone stability and cytokinesis. Curr. Biol., 20: 880–885.
• Campeau, E., Ruhl, V.E., Rodier, F., Smith, C.L., Rahmberg, B.L., Fuss, J.O., Campisi, J., Yaswen, P., Cooper, P.K., and Kaufman, P.D. 2009. A versatile viral system for expression and depletion of proteins in mammalian cells. PLoS One, 4: e6529.
• Chin, G.M. and Herbst, R. 2006. Induction of apoptosis by monastrol, an inhibitor of the mitotic kinesin Eg5, is independent of the spindle checkpoint. Mol. Cancer Ther., 5: 2580–2591.
• Cimini, D., Fioravanti, D., Salmon, E.D., and Degrassi, F. 2002. Merotelic kinetochore orientation versus chromosome mono-orientation in the origin of lagging chromosomes in human primary cells. J. Cell Sci., 115: 507–515.
• Dagenbach, E.M. and Endow, S.A. 2004. A new kinesin tree. J. Cell Sci., 117: 3–7.
• Duan, J., Duan, J., Zhang, Z., and Tong, T. 2005. Irreversible cellular senescence induced by prolonged exposure to H2O2 involves DNA-damage-and-repair genes and telomere shortening. Int. J. Biochem. Cell Biol., 37: 1407–1420.
• El-Zein, R.A., Schabath, M.B., Etzel, C.J., Lopez, M.S., Franklin, J.D., and Spitz, M.R. 2006. Cytokinesis-blocked micronucleus assay as a novel biomarker for lung cancer risk. Cancer Res., 66: 6449–6456.
• Fink, G., Hajdo, L., Skowronek, K.J., Reuther, C., Kasprzak, A.A., and Diez, S. 2009. The mitotic kinesin-14 Ncd drives directional microtubule-microtubule sliding. Nat. Cell Biol., 11: 717–723.
• Ganem, N.J., Godinho, S.A., and Pellman, D. 2009. A mechanism linking extra centrosomes to chromosomal instability. Nature, 460: 278–282.
• Gisselsson, D., Jin, Y., Lindgren, D., Persson, J., Gisselsson, L., Hanks, S., Sehic, D., Mengelbier, L.H., Ora, I., Rahman, N., Mertens, F., Mitelman, F., and Mandahl, N. 2010. Generation of trisomies in cancer cells by multipolar mitosis and incomplete cytokinesis. Proc. Natl. Acad. Sci. USA, 107: 20489–20493.
• Holmfeldt, P., Zhang, X., Stenmark, S., Walczak, C.E., and Gullberg, M. 2005. CaMKIIgamma-mediated inactivation of the Kin I kinesin MCAK is essential for bipolar spindle formation. Embo J., 24: 1256–1266.
• Kwon, M., Godinho, S.A., Chandhok, N.S., Ganem, N.J., Azioune, A., Thery, M., and Pellman, D. 2008. Mechanisms to suppress multipolar divisions in cancer cells with extra centrosomes. Genes Dev., 22: 2189–2203.
• Laan, L., Pavin, N., Husson, J., Romet-Lemonne, G., van Duijn, M., Lopez, M.P., Vale, R.D., Julicher, F., Reck-Peterson, S.L., and Dogterom, M. 2012. Cortical Dynein Controls Microtubule Dynamics to Generate Pulling Forces that Position Microtubule Asters. Cell, 148: 502–514.
• Lawrence, C.J., Dawe, R.K., Christie, K.R., Cleveland, D.W., Dawson, S.C., Endow, S.A., Goldstein, L.S., Goodson, H.V., Hirokawa, N., Howard, J., Malmberg, R.L., McIntosh, J.R., Miki, H., Mitchison, T.J., Okada, Y., Reddy, A.S., Saxton, W.M., Schliwa, M., Scholey, J.M., Vale, R.D., Walczak, C.E., and Wordeman, L. 2004. A standardized kinesin nomenclature. J. Cell Biol., 167: 19–22.
• Lentini, L., Barra, V., Schillaci, T., and Di Leonardo, A. 2012. MAD2 depletion triggers premature cellular senescence in human primary fibroblasts by activating a p53 pathway preventing aneuploid cells propagation. J. Cell. Physiol., 227: 3324–3332.
• Mansfeld, J., Collin, P., Collins, M.O., Choudhary, J.S., and Pines, J. 2011. APC15 drives the turnover of MCC-CDC20 to make the spindle assembly checkpoint responsive to kinetochore attachment. Nat. Cell Biol., 13: 1234–1243.
• Meraldi, P., Honda, R., and Nigg, E.A. 2002. Aurora-A overexpression reveals tetraploidization as a major route to centrosome amplification in p53–/– cells. Embo J., 21: 483–492.
• Mountain, V., Simerly, C., Howard, L., Ando, A., Schatten, G., and Compton, D.A. 1999. The kinesin-related protein, HSET, opposes the activity of Eg5 and cross-links microtubules in the mammalian mitotic spindle. J. Cell Biol., 147: 351–366.
• Murphy, T.D. 2003. Drosophila skpA, a component of SCF ubiquitin ligases, regulates centrosome duplication independently of cyclin E accumulation. J. Cell Sci., 116: 2321–2332.
• Pan, S.H., Tai, C.C., Lin, C.S., Hsu, W.B., Chou, S.F., Lai, C.C., Chen, J.Y., Tien, H.F., Lee, F.Y., and Wang, W.B. 2009. Epstein-Barr virus nuclear antigen 2 disrupts mitotic checkpoint and causes chromosomal instability. Carcinogenesis, 30: 366–375.
• Paoletti, A., Moudjou, M., Paintrand, M., Salisbury, J.L., and Bornens, M. 1996. Most of centrin in animal cells is not centrosome-associated and centrosomal centrin is confined to the distal lumen of centrioles. J. Cell Sci., 109 (Pt 13): 3089–3102.
• Pihan, G.A., Purohit, A., Wallace, J., Knecht, H., Woda, B., Quesenberry, P., and Doxsey, S.J. 1998. Centrosome defects and genetic instability in malignant tumors. Cancer Res., 58: 3974–3985.
• Quintyne, N.J., Reing, J.E., Hoffelder, D.R., Gollin, S.M., and Saunders, W.S. 2005. Spindle multipolarity is prevented by centrosomal clustering. Science, 307: 127–129.
• Rao, X., Zhang, Y., Yi, Q., Hou, H., Xu, B., Chu, L., Huang, Y., Zhang, W., Fenech, M., and Shi, Q. 2008. Multiple origins of spontaneously arising micronuclei in HeLa cells: direct evidence from long-term live cell imaging. Mutat. Res., 646: 41–49.
• Roberts, E.C., Shapiro, P.S., Nahreini, T.S., Pages, G., Pouyssegur, J., and Ahn, N.G. 2002. Distinct cell cycle timing requirements for extracellular signal-regulated kinase and phosphoinositide 3-kinase signaling pathways in somatic cell mitosis. Mol. Cell. Biol., 22: 7226–7241.
• Tanenbaum, M.E., Macurek, L., Galjart, N., and Medema, R.H. 2008. Dynein, Lis1 and CLIP-170 counteract Eg5-dependent centrosome separation during bipolar spindle assembly. Embo J., 27: 3235–3245.
• Tulu, U.S., Rusan, N.M., and Wadsworth, P. 2003. Peripheral, non-centrosome-associated microtubules contribute to spindle formation in centrosome-containing cells. Curr. Biol., 13: 1894–1899.
• Tutt, A., Gabriel, A., Bertwistle, D., Connor, F., Paterson, H., Peacock, J., Ross, G., and Ashworth, A. 1999. Absence of Brca2 causes genome instability by chromosome breakage and loss associated with centrosome amplification. Curr. Biol., 9: 1107–1110.
• Utani, K., Kohno, Y., Okamoto, A., and Shimizu, N. 2010. Emergence of micronuclei and their effects on the fate of cells under replication stress. PLoS One, 5: e10089.
• Walczak, C.E., Vernos, I., Mitchison, T.J., Karsenti, E., and Heald, R. 1998. A model for the proposed roles of different microtubule-based motor proteins in establishing spindle bipolarity. Curr. Biol., 8: 903–913.
• Zhu, C., Zhao, J., Bibikova, M., Leverson, J.D., Bossy-Wetzel, E., Fan, J.B., Abraham, R.T., and Jiang, W. 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.
• Zhu, J., Burakov, A., Rodionov, V., and Mogilner, A. 2010. Finding the cell center by a balance of dynein and myosin pulling and microtubule pushing: a computational study. Mol. Biol. Cell, 21: 4418–4427.