• 검색 결과가 없습니다.

Chromosomal instability induced by cell-to-cell fusion using HeLa cells

N/A
N/A
Protected

Academic year: 2021

Share "Chromosomal instability induced by cell-to-cell fusion using HeLa cells"

Copied!
63
0
0

로드 중.... (전체 텍스트 보기)

전체 글

(1)

저작자표시-비영리-변경금지 2.0 대한민국 이용자는 아래의 조건을 따르는 경우에 한하여 자유롭게 l 이 저작물을 복제, 배포, 전송, 전시, 공연 및 방송할 수 있습니다. 다음과 같은 조건을 따라야 합니다: l 귀하는, 이 저작물의 재이용이나 배포의 경우, 이 저작물에 적용된 이용허락조건 을 명확하게 나타내어야 합니다. l 저작권자로부터 별도의 허가를 받으면 이러한 조건들은 적용되지 않습니다. 저작권법에 따른 이용자의 권리는 위의 내용에 의하여 영향을 받지 않습니다. 이것은 이용허락규약(Legal Code)을 이해하기 쉽게 요약한 것입니다. Disclaimer 저작자표시. 귀하는 원저작자를 표시하여야 합니다. 비영리. 귀하는 이 저작물을 영리 목적으로 이용할 수 없습니다. 변경금지. 귀하는 이 저작물을 개작, 변형 또는 가공할 수 없습니다.

(2)

Chromosomal instability

induced by cell-to-cell fusion

using HeLa cells

by

Mihyang Do

Major in Cancer Biology

Department of Biomedical Sciences

The Graduate School, Ajou University

(3)

Chromosomal instability

induced by cell-to-cell fusion using HeLa cells

by

Mihyang Do

A Dissertation Submitted to The Graduate School of

Ajou University in Partial Fulfillment of the Requirements

for the Degree of Ph.D. of Biomedical Sciences

Supervised by

Jae-Ho Lee, M.D., Ph.D.

Major in Cancer Biology

Department of Biomedical Sciences

The Graduate School, Ajou University

(4)

This certifies that the dissertation

of Mihyang Do is approved.

SUPERVISORY COMMITTEE

Hyeseong Cho

Gyesoon Yoon

Myung-Hee Kwon

Soo-Youl Kim

Jae-Ho Lee

The Graduate School, Ajou University

June, 22th, 2017

(5)

i

-ABSTRACT-

Chromosomal instability induced by cell-to-cell fusion

using HeLa cells

Tetraploidy, a potential precursor of cancer-associated aneuploidy, is produced either by cell fusion or cytokinesis failure. These cells undergo cell-cycle arrest or apoptosis in a p53-dependent manner. Here, I used low p53-expressing HeLa cells as a model system to address the fate of cancer cells after fusion in the context of decreased influence of p53. I found that massive apoptotic cell death or growth arrest occurred a few days after fusion and was accompanied by an increase in p53. In addition, cells with larger nuclei preferentially died after fusion, suggesting that a larger deviation from normal DNA content is a strong inducer of apoptosis. Closer observation of the cells revealed that the division of fused cells immediately after cell fusion showed the formation of multipolar spindles due to increased number of centrosomes, resulting in various mitotic defects (DNA bridges, lagging chromosome, micronucleus, etc.) and asymmetric division with subsequent cytokinesis failure. After a series of unstable cell division processes, the surviving cells showed depolyploidization to have a little bit more chromosomes comparing to parental cells. The CIN (chromosomal instability) assay measuring the centromere of chromosomes 8 and 18 confirmed that chromosomal instability was increased in fused cells, and even in stable fused cell lines. Notably, a fraction of cells escaped from cell death and proliferated. These surviving fused cells were characterized by upregulation of survivin, reflecting increased survivin protein stability. Moreover, in fused cells, survivin became preferentially localized to the cytosol, where it is known to exert its anti-apoptotic function. Knockdown of survivin decreased survival to a greater extent in fused cells than in unfused cells, suggesting that fused cells became more dependent on survivin. Therefore, above findings indicate that, after cancer cell fusion, a subpopulation of fused cells with a higher level of cytosolic survivin are able to avoid apoptotic crisis and survive to proliferate continuously, a process that might contribute to human cancer progression. Regarding cancer progression, fused cells were superior to

(6)

ii

unfused cells in their overall cell migration ability, and their abilities varied according to each cell line. In addition, when the cells were grown under low concentration of cisplatin, fused cells got resistance more readily than unfused cells. Collectively, the fusion of cancer cells leads to chromosomal instability that induce p53-dependent cell death; however, a fraction of cells such as the cells having more cytosolic survivin survive and get genetic diversity, which probably confers the cancer cells more progressed characteristics such as acquisition of chemoresistance as well as migration ability.

(7)

iii

TABLE OF CONTENTS

ABSTRACT ... i

TABLE OF CONTENTS ... iii

LIST OF FIGURES ... v

LIST OF TABLES ... vi

I. INTRODUCTION ... 1

II. MATERIALS AND METHODS ... 4

1. Cell cultures and plasmids ... 4

2. RNA extraction, reverse transcription-polymerase chain reaction ... 4

3. Antibodies ... 4

4. Cell fusion and FACS ... 4

5. Measurement of cellular DNA content ... 5

6. Clonogenic cell proliferation assays ... 5

7. Establishment of Cisplatin-resistance cells ... 5

8. Immunocytochemistry ... 6

9. Cytogenetics: karyotypes and FISH ... 6

10. Live cell imaging ... 6

III. RESULTS ... 8

1. Generation and isolation of fused cells. ... 8

2. Fused cells experience massive cell death. ...10

3. Cells with enlarged nuclei are prone to apoptosis. ...12

(8)

iv

5. Fused cells showed asymmetric division showing chromosome missegregations

with spindle multipolarity. ...19

6. Cell fusion induces ROS production and DNA damages. ...32

7. Survivin is necessary for the survival of fused cells that escape apoptotic crisis. ...35

8. Survivin protein in fused cells is localized in the cytosol and shows increased stability. ...39

9. Fused cells increased cell migration and showed higher resistance than unfused cells in cisplatin treatment. ...42

IV. DISCUSSION ...45

V. CONCLUSION ...48

REFFERENCE ...49

(9)

v

LIST OF FIGURES

Fig. 1. Generation and isolation of fused cells. ……….………9

Fig. 2. Analysis of cell fate after cancer cell fusion. ………11

Fig. 3. Preferential elimination of the meganucleated cells with larger nuclei through apoptosis. ………14

Fig. 4. Accumulation of p53 on nucleus of fused cells and depletion of p53 attenuated cell death in fused cells. ………16

Fig. 5. Depolyploidization of fused cells. ………18

Fig. 6. Fused cells showed supernumerary centrosomes. ……….……….23

Fig. 7. Mutipolar division resulted in depolyploidization of fused cells. ………24

Fig. 8. Accumulated mitotic defects in fused cells. ………26

Fig. 9. Chromosomal instability in the early stage of the fused cell. ………28

Fig. 10. Chromosomal instability in fused lines. ………29

Fig. 11. Fused cells increased the ROS level and induced DNA damages. .………33

Fig. 12. Survivin, anti-apoptotic protein increased in surviving fused cells. ……….….……36

Fig. 13. Depletion of survivin decreased cell growth in fused cells. ……….…….…….37

Fig. 14. Increased protein stability of survivin upon cell fusion. ………40

Fig. 15. Increased cytosolic localization of survivin upon cell fusion. ………41

Fig. 16. Fused cells increased cell migration. ………...43

Fig. 17. Fused cells showed more chemoresistant than unfused cells treating cisplatin by selective pressure.………44

(10)

vi

LIST OF TABLES

(11)

1

I.

INTRODUCTION

Physiological cell fusion results in terminally differentiated cells, such as syncytiotrophoblasts, myocytes and osteoclasts, whereas unphysiological cell fusion induced by a variety of agents, including viruses and chemicals, produce fused cells with proliferative capacity (Duelli et al., 2005). As a result of subsequent cell divisions, these fused tetraploid cells give rise to daughter cells that exhibit genomic instability, a process similar to the genomic instability that follows cytokinesis failure, which causes daughter cells to become aneuploid and carcinogenic (Ganem et al., 2007). Physiological cell fusion results in terminally differentiated cells, such as syncytiotrophoblasts, myocytes and osteoclasts, whereas unphysiological cell fusion induced by a variety of agents, including viruses and chemicals, produce fused cells with proliferative capacity (Duelli et al., 2005). As a result of subsequent cell divisions, these fused tetraploid cells give rise to daughter cells that exhibit genomic instability, a process similar to the genomic instability that follows cytokinesis failure, which causes daughter cells to become aneuploid and carcinogenic (Ganem et al., 2007). Unphysiological cell fusion has also been considered a mechanism by which cancer cells acquire more aggressive phenotypes (Lu and Kang 2009). For example, fusion of cancer cells with macrophages has been reported to confer on cancer cells the capacity to invade and metastasize (Kerbel et al., 1983). It has also been suggested that fusion of cancer cells with endothelial cells may enable cancer cells to more easily penetrate the endothelial cell layer (Mortensen et al., 2004). Importantly, fusion between cancer cells can induce genomic instability, which can be a driving force for these cells to obtain diverse tumor-progression phenotypes (Lu and Kang 2009). On the other hand, promoting cancer cell fusion has been considered a potential therapeutic strategy, because the profound genomic instability induced by cell fusion can induce cell death (Horn et al., 2005).

However, the mechanism by which tetraploidy makes tumorigenesis has not yet been fully elucidated. Based on previous studies, tetraploid cells are known to have an increased number of centrosomes, which leads to multipolar division during mitosis and eventually divides CIN into higher aneuploid cells (Ganem and Pellman 2007; Levine et al., 1991). This aneuploidy also affects the cancer progression and may lead to phenotypic diversification of cancer, such as

(12)

2

metastasis and anti-cancer drug resistance (Vitale et al., 2010).

The tetraploid cells produced by either cell fusion or cytokinesis failure become cell cycle-arrested or apoptotic through a process that has been considered to be p53 dependent (Andreassen et al., 2001; Aylon and Oren 2011; Vogel et al., 2004). Activation of p53 induces p21-dependent cell-cycle arrest or increases proapoptotic Bcl-2 family proteins, such as Bax and Puma/BBC3, and thus induces apoptosis in a cell context-dependent manner (Vousden and Lu 2002;Wei et al., 2001;Yu et al., 2003). Therefore, after cell fusion or cytokinesis failure, those cells with increased p53 activity are removed in a p53-dependent manner (Aylon and Oren 2011), whereas cells in which p53 is less activated survive and even proliferate, demonstrating an ability to form colonies in soft agar (Ho et al., 2010).

Considering the tendency of cancer cells to inactivate p53, fusion between cancer cells may result in a high probability of escaping cell death after fusion, while simultaneously allowing acquisition of proliferative potential and genomic instability. Therefore, understanding the fate of cells arising from the fusion of cancer cells with lower p53 activity is important to understanding the role of cancer cell fusion in cancer progression. In addition, the factors that determine the fate of fused cells are also important, but have not yet been identified. Here, we used HeLa cells, which harbor low levels of p53 owing to enhanced p53 degradation in the presence of the E6 viral oncoprotein, as a model system to address the fate of cancer cells after fusion in the context of decreased influence of p53 (Scheffner et al., 1990). Interestingly, massive cell death occurred a few days after fusion, followed by the emergence of proliferating cells. These proliferating cells were mainly tetraploid and appeared to have escaped apoptotic cell death, which had eliminated cells with a higher DNA content. Furthermore, upregulation and cytosolic localization of survivin was at least partly responsible for the escape of these proliferating cells from apoptotic crisis. Observing with a time-lapse microscopy for a long time following cell fusion, we will examine in detail how the fused cells proliferate and pass the cell division process.

Interestingly, a large change of DNA contents occurred a few days after fusion through the cell division provoking chromosome missegregation with spindle multipolarity. Therefore, fused cells resulted in continuous karyotypic rearrangements, favoring the generation of progeny increased CIN and hence, the spontaneous evolution of tumors to malignancy such as acquire the

(13)

3

resistance to chemotherapy or cell migration ability.

(14)

4

II.

MATERIALS AND METHODS

1. Cell cultures and plasmids

HeLa cells were cultured in Dulbecco’s modified Eagle’s medium nutrient mixture (Sigma Chemical Co.) supplemented with 10% (vol/vol) fetal bovine serum (Hyclone). G418 resistant - or Hygromycin resistant - HeLa cells were cultured in the presence of 0.3 mg/ml G418 or 0.2 mg/ml Hygromycin B respectively. GFP- H2B or DsRed- H1 HeLa cells were cultured in the presence of 0.3 mg/ml G418. DiI (D282) and DiO (D275) were purchased from Gibco-Invitrogen (Carlsbad, USA). Cycloheximide (CHX) and z-VAD fmk were purchased from Sigma Chemical Co.. NAC (N-acetyl-L-cysteine) and cisplatin were purchased from Sigma-Aldrich Co. (Korea).

.

2. RNA extraction, reverse transcription-polymerase chain reaction

Total RNA was extracted from cells using Trizol reagent (Ambion) according to the manufacturer’s instructions. First strand cDNA was synthesized from 1 µg of total RNA by using Rever Tra Ace qPCR (TOYOBO). The mRNA expression level of survivin was normalized using β-actin as an internal control. The 2-△△CT relative quantification method was used to calculate the mean fold expression difference between the groups. The following primers were used:

survivin, 5’-CTGCCTGGCAGCCCTTTCTCAA-3’ (forward) and

5’-AATAAACCCTGGAAGTGGTGCA-3’ (reverse); β-actin,

5′-GTGGCATCCATGAAACTACAT-3′ (forward) 5′-AACGCAGCTCAGTAACAGTC-3′

(reverse).

3. Antibodies

We purchased following antibodies from Cell Signaling Technology (PARP-1, caspase 3, survivin, Bcl2, BAG1, p53), Santa Cruz (lamin-B, actin, α-tubulin, GAPDH) or Invitrogen (HRP-conjugated secondary antibodies). Alexa Fluor 488 (Molecular Probes)-(HRP-conjugated secondary antibody was used for immunocytochemistry.

(15)

5

HeLa cells were stained with DiI (10 μg/ml) or DiO (10 μg/ml) for 15 min at 37°C. Stained HeLa cells were washed with Ca2+/Mg2+-free phosphate buffered saline (PBS) and resuspended

with fresh nonelectrolyte solution (1mM MgSO4, 1mM CaCl2, 0.3M Mannitol) at a density of

2x106 cells per ml. Same numbers of each stained cells were mixed and then used immediately

for electrofusion by using ECM 2001 Electro Cell Manipulator (BTX, Harvard Apparatus). Optimized condition of electrofusion was as follows: AC 20 V for 30 sec, DC 700 V for 70 sec, post-fusion AC 20 V for 9 sec. After then, cells were plated into 10 cm culture dish in normal growth media. After 2 hours, cell sorting was performed by using a FACS Aria III Flow cytometer (BD Biosciences).

5. Measurement of cellular DNA content

Cells were trypsinized, centrifuged and fixed with 70% cold ethanol for 1 hr. After samples were washed with cold-phosphate-buffered saline (PBS) and then resuspended in a solution containing RNase A (100 μg/ml) for 5 min, propidium iodide (50 μg/ml) was added. After 30 min of incubation, samples were subjected to FACS analysis using a FACS Vantage flow cytometer (Becton Dickinson, Franklin Lakes, NJ).

6. Clonogenic cell proliferation assays

Parental HeLa cells were used as control cells, and cultured in normal growth media without selective agents. Colonies were fixed/stained with an aqueous solution containing 0.25% (w/v) crystal violet, 20% (v/v) methanol and counted (Vitale et al., 2010). Only colonies consist of >30 cells were counted. The effect of survivin knockdown was expressed as the survival fraction (%): survival fraction (%) = 100 x # of colonies in survivin knockdown plate/ # of colonies in control knockdown plate.

7. Establishment of Cisplatin-resistance cells

Cisplatin-resistant variants of unfused and fused HeLa cells were obtained by exposure to cisplatin as follows. Cells were cultured at an initial density of 1x105 cells per 6 well plate

containing 3 ml medium for 1 day, was treated with 0.5, 1.0 or 2.0 μg/ml cisplatin for 1 day. Cells were then cultured in cisplatin-free medium for 1 weeks to recover cell density. The above treatment was repeated 8 times for 8 weeks.

(16)

6

8. Immunocytochemistry

Cells were fixed with 3.7% formaldehyde in PBS and permeabilized with 0.1% Triton X-100 in PBS. Fixed cells were incubated with primary antibody for overnight at 4°C. Cells were then washed with PBS and incubated with secondary antibody for 1hr at room temperature. For DNA counterstaining, DAPI (Molecular probes) was used.

9. Cytogenetics: karyotypes and FISH

For karyotypes, cells were exposed to 100 ng/ml nocodazole for 12 h, and mitotic cells were then collected, incubated at 37°C for 15 min in 0.075 M hypotonic KCl, pelleted and fixed with methanol:acetic acid (3:1) solution. After washing three times with PBS, samples were spreaded on slide glass, dried at room temperature (RT) for 24 h and then stained with 3% (wt/vol) Giemsa staining solution. Photographs were taken using confocal microscope (LSM710; Carl Zeiss, Germany).

For FISH analysis, centromeric DNA probes for chromosomes 8 (CEP 8 Spectrum Orange, Direct Labeled Fluorescent DNA Probe Kit, Vysis, Abbott Molecular, USA) was used to determine the number of chromosome 8 constitution of interphase cells, according to manufacturer’s instruction. Cells were plated into chamber glass slides (MA Nunc Lab-Tek II, Thermo Scientific, Waltham, MA) at a density of 2.0 x 104 cells/ml. Cell were fixed coated with

3:1 methanol:acetic acid, and allowed to dry. Samples were analysed and scored under a Zeiss Axiovert 200M microscope (Carl Zeiss). For analysis of interphase nuclei, signals were counted for as many nuclei available, up to 200 for each sample. Signals were scored as separate if they were further than a signal’s diameter apart in distance, according to established clinical practices of FISH signal interpretation and as described (Bayani and Squire 2005).

10. Live cell imaging

Sorted cells were seeded into 6 cm culture dishes at a concentration of 1.0×104 cells per dish.

Phase contrast live cell imaging was performed as previously described with a Zeiss Axiovert 200M microscope (Carl Zeiss) (Nam et al., 2014). DIC, GFP and DsRed fluorescence images were acquired with Motorized Inverted Fluorescence Microscope Nikon Ti-E (Nikon

(17)

7

Corporation). Microscope was enclosed within temperature and CO2-controlled environments

that maintained an atmosphere of 37℃ and 3–5% humidified CO2. GFP, DsRed and bright field

images were captured at multiple points every 3 min for 4-5 days with objectives. Captured images from each experiment were analysed using NIS-Elements software.

(18)

8

III.

RESULTS

1. Generation and isolation of fused cells.

Separate populations of Geneticin-resistant and hygromycin-resistant HeLa cells were stained with the vital fluorescence dye DiO and DiI, respectively, and then subjected to electrofusion. Fused cells and unfused cells were separated and isolated by fluorescence-activated cell sorting (FACS). DiO/DiI double-positive [Dio(+)/DiI(+)] cells were identified as fused cells, whereas DiO (-) DiI (+) cells corresponded to unfused cells, which were used as control cells that had undergone the electrofusion procedure but without the resultant cell fusion (Fig. 1A). Fused cells and unfused cells were easily differentiated under a fluorescence microscope (Fig. 1B), and FACS analysis revealed that ~99% of the FACS-sorted fused cells were DiO (+) DiI (+) (Fig. 1C), indicating the reliability of the FACS procedure. A further analysis of fused cells 1 day after fusion revealed that about 67% had a single, enlarged nucleus or two nuclei, as expected for the fusion of two cells, whereas the remaining ~33% had more than three nuclei, suggesting fusion of more than three cells (Fig. 1D).

(19)

9

Fig. 1. Generation and isolation of fused cells. (A) Schematic diagram of the cell fusion and selection procedure is shown. Detailed procedure is described in Materials and Methods. (B) Nucleus number of fused and unfused cells was counted with images taken by fluorescent time-lapse microscope (Ti-E Nikon). After cell fusion between H2B-GFP-HeLa and H1-DsRed-HeLa, isolated cells were seeded in a 4-well Lab-Tek chamber slide. Results are given as the mean ± SD from three independent experiments. (unfused; n=63, fused; n=23). (C) At 2 hours after fusion, cells were sorted by FACS and the purity of each isolated population was measured by analytical FACS.

(20)

10

2. Fused cells experience massive cell death.

To address the fate of fused cells, we monitored cell proliferation and cell death after fusion. As shown in figure 2A, the growth rate of fused cells was significantly lower than that of unfused cells 4 days after cell fusion. Thereafter, the proliferation of fused cells modestly increased, whereas that of unfused cells increased steeply. Trypan blue staining revealed that fused cells underwent massive cell death, peaking at 33.7% ± 4.0% cell death on day 4 after fusion (Fig. 2B-C). The subsequent decrease in the cell death rate at day 6 coincided with a gradual increase in the proliferation of fused cells. These observations suggest that the observed differences in cell proliferation might be at least partly attributable to differences in cell death.

We next followed the fate of individual cells by counting cell numbers over time in each well of a 96-well plate after limiting dilution. Whereas most unfused cells showed a steep increase in cell number per well during this period (Fig. 2D, left panel), a majority of fused cells showed a steep decline in cell number per well after quite variable times of division (Fig. 2D, right panel). An analysis of clones 9 days after fusion classified 81.4% of unfused cells as proliferative, 6.8% as growth arrested and placed 11.9% in the cell death category. The classification of clones of fused cells revealed a quite different picture: 26.8% were proliferative, 32.1% were in growth arrest and 41.1% belonged to the cell death category (Fig. 2E). These data clearly demonstrate that a major population of fused cells underwent cell death or growth arrest, whereas a fraction escaped the apoptotic crisis and continued to proliferate.

(21)

11

Fig. 2. Analysis of cell fate after cancer cell fusion. (A) Representative images of fused and unfused cells obtained from fusion of DiI+- and DiO+-HeLa cells as in Fig.1A by fluorescence

microscopy. HeLa cells were fused as described in Materials and Methods. Scale bar: 50 μm. (B) and (C) After cell fusion, cell death and proliferation rate were measured after trypan blue staining. Cells were seeded at a density of 1x104cells per well in 12-well plates and counted at the indicated

times. Mean ± SD from three independent experiments; ***, p<0.001; *, p<0.05 by Student's t-test. (D) The fate of individual cell was monitored under microscope. Unfused or fused cells were seeded at a density of 0.5 cell per well in 96 well plate and cultured for 9 days in growth media without selection. The cell number in each well was counted at the indicated times. (B-D) These were performed by IH Kwak. (E) At 9 day after cell seeding, cell clones were classified as either growth (wells having more than two cells/well), arrest (1~2 cells/well), or death (no cells in the well, but used to have cells at earlier time points).

(22)

12

3. Cells with enlarged nuclei are prone to apoptosis.

To more precisely describe cell fates after cell fusion, I continuously monitored cells by time-lapse microscopy. This analysis clearly revealed a major population of dead (Fig. 3A, row 3) or growth-arrested cells, and a minor population of continuously proliferating cells (Fig. 3A, row 2). In contrast, most unfused cells were proliferative (Fig. 3A, row 1). I then addressed whether DNA content affected the fate of fused cells. Since it is difficult to directly measure DNA content from time-lapse data using GFP-H2B- and DsRed-H1-HeLa cells, we instead measured nuclear size as a surrogate marker for DNA content. As expected, the extent of the increase in nuclear size 1 day after fusion varied considerably. Interestingly, the heterogeneity of nuclear size observed on the day after fusion decreased gradually, together with a decrease in the average nuclear size (Fig. 3B), suggesting the disappearance of cells with larger nuclei. Indeed, tracking the fate of daughter cells having same parent cells by time-lapse analysis enabled us to clearly determine that dead cells had larger nuclei compared with live cells (Fig. 3D-E).

It has previously been reported that tetraploid cells induced by cell fusion or cytokinesis failure tend to die via apoptosis (Castedo et al., 2006; Fukuta et al., 2008). To address this aspect, I assessed apoptosis by measuring the levels of the active (cleaved) form of the apoptosis-inducing factors, poly (ADP-ribose) polymerase 1 (PARP-1) and caspase-3, in fused and unfused cells. As shown in figure 3F. cleavage of PARP-1 and caspase-3 robustly increased in fused cells but not in unfused cells, suggesting that the death of fused cells was at least partly attributable to apoptosis. Moreover, zVAD-fmk, a pan-caspase inhibitor, clearly abolished the cleavage of both PARP-1 and caspase-3 (Fig. 3F), and partly prevented the death of fused cells, but not unfused cells, thereby suggesting a partial involvement of caspase-dependent apoptosis in the death of fused cells. Interestingly, both immunoblot and immunocytochemical assessment of p53 shows an increase in p53 levels in fused cells, as compared to unfused cells (Fig. 4A-B), indicating that even in HeLa cells, p53 levels increase after fusion and probably contributed to the massive cell death after cell fusion, thus suggesting that the decrease in p53 in HeLa cells caused by HPV E6 protein can be overcome by strong apoptotic stimuli, one of which is cell fusion.

To confirm the above hypothesis, we evaluated the effect of p53 depletion on growth and death of fused cells. After p53 depletion (Fig. 4C), cell growth significantly increased in

(23)

13

fused cells after day 5 post-fusion, while there were no changes due to p53 depletion in unfused cells (Fig. 4D). In addition, the death of fused cells after p53 depletion decreased at day 3 post-fusion, and significantly dropped at day 5. On the contrary, unfused cells did not show significant decrease in cell death due to p53 depletion (Fig. 4E). Collectively, our data indicate that a majority of fused cells succumbed to death, probably owing to an increase in p53, whereas only a few cells that overcame this apoptotic crisis ultimately attained the capacity to grow continuously.

(24)

14

Fig. 3. Preferential elimination of the meganucleated cells with larger nuclei through apoptosis. (A) Representative images of each cell fate. Time-lapse images were acquired by using the Zeiss Axiovert 200M microscope. Phase-contrast images are shown (upper panels). Representative time-lapse images captured at the indicated time points show that some daughter cells that originated from fused cells undergo apoptosis (black arrow in row 3) whereas others can proliferate like unfused cells (white arrow in row 3). (B) Change of nuclear size following cell fusion. Images of DAPI-stained nuclei were measured by using Axiovision Rel 4.5 software (n=700). Bar indicates median value. (C) Representative images of meganucleated cell death following cell fusion. Dotted circles denote a single cell membrane.

(25)

15

The size (D) and number (E) of nucleus of cells at interphase was measured by the images of cells under microscopic observation (Nikon Ti-E). For data analysis we used NIS elements software. (F) Immunoblots of fused and unfused cells were probed with antibodies to PARP-1 and caspase 3. The level of cleaved PARP-1, activated caspase 3 (black arrows) and p53 were increased in fused HeLa cells. Cells were treated with or without z-VAD fmk (10 μM) and harvested at 3 day following cell fusion. β-actin: loading control. (G) Cell death was partially abrogated by the z-VAD-fmk in fused cells. Cells were seeded in a 12-well plate at a density of 1x104 cells/well and counted by 0.4% trypan blue staining at 3 day after cell fusion. Scale bar: 20 μm. Mean ± SD from three independent experiments; **, p<0.01; *, p<0.05 by Student's t-test.

(26)

16

Fig. 4. Accumulation of p53 on nucleus of fused cells and depletion of p53 attenuated cell death in fused cells. (A) Immunocytochemical staining of p53 in unfused and fused cells at day 2 following cell fusion. (B) The percentage of p53-positive nucleus in unfused and fused cells. (C-E) Cells were transfected with indicated siRNAs (100 nM) and then performed cell fusion process. (C) Cells were harvested at day 5 following cell fusion. Protein blots of fused and unfused cells were probed with indicated antibodies. GAPDH: loading control. (D) and (E) Cell death and proliferation rate were measured after trypan blue staining. Cells were seeded at a density of 1x104cells per well in 12-well plates and counted at the indicated times. (C-E) These were

performed by JH Ahn. Mean ± SD from three independent experiments; ***, p<0.001; **, P < 0.01; *, p<0.05 by Student's t-test.

(27)

17

4. Depolyploidization after cell fusion

To assume the depolyploidization in fused cells, we isolated unfused and fused cells with FACS sorting after cell fusion between HeLa cells expressing H2B-GFP and H1-DsRed (Fig. 5A). To monitor the alteration of DNA contents following cell fusion, we first performed DNA FACS analysis over time. As expected, the fused cells initially showed > 2-fold DNA content to comparing to unfused cells. We also observed considerable amount of sub-G1 population around day 3 and day 5, as we have mentioned. Interestingly, increased DNA contents altered with reducing at 7 day (Fig. 5B) and reduction of DNA was kept maintaining for 21 days (data not shown). According with decreasing DNA contents, the percentage of bi- or multinucleated cells significantly decreased in fused cells by 5 days (Fig.5D).

In order to chase each cell division, images were taken through a florescence time-lapse microscope every 3 minutes for 80 hours of fused and unfused cells and measured the nucleus area with NIS Elements ver. 4.0 (Nikon). As a result, the nucleus area increased with a large variation at first but showed the decease of median value in fused cells (Fig.5C). It was clearly informed that the depolyploidization induced in fused cells passing through cell divisions.

(28)

18

Fig. 5. Depolyploidization of fused cells. (A) The first fluorescence images of unfused (H1-DsRed-HeLa) and fused (H2B-GFP-/H1-(H1-DsRed-HeLa) cells using fluorescence time-lapse microscopy (Nikon Ti-E). Scale bar, 20 μm. (B) Changes in the DNA histograms of fused and unfused cells for 10 days. Cells were harvested and fixed with ethanol at indicated time point followed by propidium iodide staining for DNA content. Dotted lines indicate G1 peak of unfused cells at 3 day, as a control. (C) Alterations of nucleus area of G2 phase cells from time-lapse analysis. Images of unfused (H1-DsRed-HeLa) and fused (H2B-GFP-/H1-DsRed-HeLa) cells were acquired every 3 minutes for 80 hours using by fluorescence time- lapse microscopy (Nikon Ti-E). Nucleus area of G2 phase cells before chromosome condensation was measured by using NIS viewer version 4.0 (Nikon). F0 means the cells just after cell sorting. F1 is the daughter cells of F0, as a next generation. (D) Quantification of the bi- or multinucleated cells following cell fusion. Cells were harvested and fixed at indicated time point. α-tubulin and DAPI staining was used for ICC. Results are given as the mean ± SD from three independent experiments; ***, p<0.001 by Student's t-test.

(29)

19

5. Fused cells showed asymmetric division showing chromosome missegregations with spindle multipolarity.

Cell fusion actually induces the increase of centrosomes. As expected, fused cells were increased supernumerary centrosomes than unfused cells showing 1 or 2 centrosomes. But numerical aberration of centrosomes gradually disappeared and converged with 1 or 2 centrosomes in fused cells for 7 days (Fig. 6B). Normally in diploid cells, at the beginning of mitosis, single centrosome duplicates and the mother and daughter organelles migrate to opposite cell poles, directing the formation of the spindle, to guarantee a balanced chromosomal segregation (Nigg and Stearns 2011). However, we expected that supernumerary centrosomes induced by cell fusion can cluster together, acting as two single units mimicking a bipolar spindle, or as single entities that generate multipolar spindles in which chromosomes are improperly segregated into two or more daughter cells. Cells had been specially stained centrosome with using gamma-tubulin Ab by time after cell fusion and nucleus had been counter-stained with DAPI (Fig. 6A).

To understand the reduction of DNA contents in fused cells, I chased the fate of cells every three minute for 80hrs after their generation by time-lapse microscopy. First, it was analyzed that how many cells entered into mitosis as setting F0 for the first cells right after fusion. More than 90% of fused cells except 9.5% of cells showing interphasic death proceeded in mitosis similarly with unfused cells (Table 1).

And we found that fused cells underwent multipolar (mostly tri- or tetrapolar) mitosis — which are associated with Y- and X-shaped metaphases—much more frequently than unfused control cells (Fig. 7A and B).

Following the time-lapse data for about 80hours revealed that most of unfused cells (about 90%) showed bipolar spindles, whereas fused cells conducted forming multipolar spindles in approximately more than 85% of cells during the first mitosis. And it was also provided that the fused cells with spindle multipolarity were reduced up to 30% through next mitosis, on the contrary, the cells with bipolar spindles were increased closed to 70% as the time passed (Fig. 7B).

(30)

20

anaphase onset (AO), which is under the control of the spindle assembly checkpoint (SAC), which inhibits the anaphase-promoting complex/cyclosome and delays anaphase onset if kinetochores are not properly attached to spindle microtubules (Foley and Kapoor 2013; Khodjakov and Pines 2010). Consistent with these observations, we found that mitotic duration of fused cells was significantly lengthen and it might delay anaphase onset until all kinetochores are attached to spindle microtubules (Fig. 7C).

And the process of cytokinesis is very important for the maintenance of genome integrity, however, it was exhibited that as for fused cells, cytokinesis failures, which was coordinated with furrow regression and abscission failure, were occurred and the first generated F0 cells displayed higher frequency of 76%, while unfused did slightly about 6% (Table 1). By such cytokinesis failure, it was found that daughter cells were not divided evenly as spindle pole numbers, and eventually ended in asymmetric division in fused cells (Fig.8).

Taken together, apart from bipolar division in unfused cells, fused cells harboring supernumerary centrosomes accumulated spindle multipolarity and chromosome missegregation and led to cytokinesis failures and asymmetric division. These results presumably constituted the major generator of aneuploidy and genomic instability in fused cells.

To further characterize the chromosomal instability in the fused lines, we performed cell fusion between G418R-HeLa and hygromicinR-HeLa and it was cultured for more than one

month in selection media with limiting dilution by plating on 96 well plate at a density of 0.5 cell per well after sorting. The alteration of numerical chromosomes was performed with Giemsa staining of chromosome spreads. The number of chromosomes per cell showed in all 12 cell lines, an abnormal deviation from the average number of hypertriploid HeLa cells. In Fig. 10A and B, examples of Giemsa-stained metaphase spreads of fused cell lines (#301~360) and unfused, as control cell lines (#225~244) are shown. The number of chromosomes per cell revealed a widespread variability in chromosome number (aneuploidy), distributing around doubled within all fused cell lines. In contrast, the control cell line revealed predominantly a hypertriploid karyotype.

As a confirmation experiment for chromosomal instability, CIN assay was performed by FISH analysis on each cell line using a centromere specific probe. Cells were isolated with

(31)

21

fused and unfused cells and cultured with selection media for 1 month to exclude any contamination with unfused cells. And cells were subcultured with or without selection media for 2 weeks.

To assess mitotic fidelity in fused lines, I quantified the modal chromosome number and the percentage of cells deviating from that mode in unfused and fused cells using FISH with the centromere specific probe of chromosome 18 (Fig. 10D). As the result, the percentage of fused cells deviating from modal Cep18 number (n=4) was only about 15%, while fused lines which were cultured without selection media significantly increased the deviation from modal number by 25%. It provided that fused cells displayed a large alteration of chromosome number through cell division and revealed that chromosomal instability was induced in fused cells by cell-to-cell fusion.

Next, I checked that the CIN phenotypes were observed with stable clones in fused cells. I imaged fixed cells and found that near 2-fold increase on centrosome amplification was observed in fused lines. In addition, fused lines displayed not only increase of spindle multipolarity but also segregation aberrations with ~2.0-fold and 1.5-fold, respectively (Fig. 10E-G). These data provided that mitotic errors frequently occur in fused cell lines as well as in primary fused cells immediately after cell fusion, indicating that the proliferation of fused lines must combine with elevated chromosome missegregation rates to generate aneuploid cells with CIN.

Along with this, CIN phenomenon was not only led in fused cell right after fusion but also in stable lines showing significant increment of centrosomes, lagging chromosomes, late segregating defects (Fig. 10G). It suggests that CIN is induced by cancer cell fusion in both fused cells and established lines.

(32)

22

(33)

23

Fig. 6. Fused cells showed supernumerary centrosomes. (A) Confocal microscopy images from unfused and fused cells stained for centrosomal γ-tubulin (green) and nucleus (DAPI, blue). Scale bar, 20 μm. (B) Quantification of centrosomes (γ-tubulin spots) and nucleus (DAPI) number in a cell. Cells were seeded at a density of 1x104 cells per well in 12-well plate and fixed at the

indicated time point. More than 300 cells from (A) were counted for each time, from three times of independent experiments. Data pooled from three independent experiments.

(34)

24

Fig.7. Mutipolar division resulted in depolyploidization of fused cells. (A) Images of unfused (H1-DsRed-HeLa) and fused (H2B-GFP-/H1-DsRed-HeLa) cells were acquired by fluorescence time-lapse microscopy (Nikon Ti-E). Still images of time-lapse microscopy experiments show representative cells at NEBD, metaphase, anaphase and telophase (from left to right). The time of initial round-up was set to 00:00 (hr:min). Scale bar, 10 μm. (B) Representative images of multipolar spindles in fused cells at metaphase. Confocal microscopy images from unfused and fused cells stained for centrosomal γ-tubulin (green) and nucleus

(35)

25

(DAPI, blue). Scale bar, 20 μm. (C) Quantification of the cells showing bi- or multipolar division was performed using the images at metaphase. M and number mean mitosis and its order. (D) Measuring the time of mitotic duration from NEBD to anaphase onset in period of the first mitosis after observation. Mitotic duration was analyzed with images from time-lapse data (n=25). (E) Representatives images show centrosome clustering of fused cells (hh:mm). ( F ) Bar graph was quantified by measuring the mitotic duration time according to the number of spindle poles. Scale bar, 20 μm. Results are given as the mean ± SD from three independent experiments; ***, p<0.001 by Student's t-test.

(36)

26

Fig. 8. Accumulated mitotic defects in fused cells. (A) Representative frames from time- lapse microscopy of cytokinesis progression of cells expressing H2B-GFP and H1-DsRed after cell fusion. Three daughter cells (white dots) became two cells (white and black dots) owing to abscission failure due to DNA bridge in magnified view of the dashed box in (A). (B) Quantification of the cytokinesis failure (CF) with or without DNA bridges during

(37)

27

around 80 hours immediately after cell fusion. Results are given as the mean ± SD from three independent experiments; ***, p<0.001 by Student's t-test. (C) Representative images from time-lapse microscopy showing unaligned chromosome (yellow arrow), anaphase lagging (green) and micronucleus (blue). (D) Representative images of successful cytokinesis in unfused cells (upper panel) and cytokinesis failure resulting in asymmetric division in fused cells (lower panel). The time indicates from metaphase to end of cytokinesis. The character a, b, c, d each means the precursor of daughter cells in telophase. Finally, two daughter cells were produced after the process of abscission. The images were taken by fluorescence time-lapse microscopy (Nikon Ti-E). Scale bar, 20 μm. (E) Horizontal columns represent single cells over the time, as indicated in minutes. The color code depicting bipolar (green) or multipolar (orange) cell divisions, as well as cell death (black) is used. The increase in cell ploidy following late cytokinesis failure is represented by grey darkening.

(38)

28

Fig. 9. Chromosomal instability in the early stage of fused cells. (A) FISH was performed using the Cep8 (orange) probe as a fixed sample on the first day after cell fusion. Results are given as the mean ± SD from three independent experiments; ***, p<0.001; *, p<0.05 by Student's t-test.

(39)
(40)

30

Fig. 10. Chromosomal instability in fused lines. (A) Representative metaphase spreads from unfused and fused lines. The stable lines of cells were established by with selection media (0.8 mg/ml G418 and 0.4 mg/ml hygromycin) for more than 2 weeks. Images were taken by confocal microscopy and scored using ZEN2011 software. Scale bar, 10 μm. (B) Distribution of the chromosome number per cell in fused cell lines (#301-#360) and unfused cell lines (#225-#244). (C) DNA contents of asynchronous cells, assayed by flow cytometry with PI staining. (D) CIN assay in stable lines. Right graph shows quantification of cells with chromosomal instability and left, representative images show that FISH signals for chromosomes 18 (orange) in cells. FI and FII mean fused lines, which were cultured in selection media (with G418 and hygromycin B) or not, respectively. Cells were stained with DAPI (blue) to visualize nuclei. Centrosome (E) and spindle pole (F) numbers in the indicated cell lines, determined by microscopic analysis of pooled

(41)

31

populations stained with γ-tubulin and DAPI. (E) Interphase cells were scored as having a normal number of centrosomes (1 or 2 γ-tubulin spots) or numbers in excess of two. (F) Mitotic cells were counted as having either two or more than two spindle poles. (G) The percentage of cells with chromosomal abnormality at anaphase. Representative images show DNA bridges, late segregating, lagging chromosomes (right). Results are given as the mean ± SD from three independent experiments; ***, p<0.001; **, p<0.01; *, p<0.05 by Student's t-test.

(42)

32

6. Cell fusion induces ROS production and DNA damages.

To gain insight into the impaired mitosis by cell fusion may negatively affect to genome stability, we evaluate the H2AX phosphorylation at ser 139 using γH2AX antibody to measure DNA damages. We performed ICC and immunoblots to determine the γ-H2AX level at indicated time points. In fused cells, both ICC and immunoblotting, nuclear foci formed by γ -H2AX were dramatically increased at early times and diminished in a time-dependent manner, on the contrary not detectable of γ -H2AX level in unfused cells (Fig. 11A-C).

It is well known that DNA is targeted by reactive oxygen species (ROS) which originate multiple oxidative products able to mutate purine, pyrimidine and deoxyribose, such as the oxidative damage DNA marker 8-hydroxy-2′-deoxyguanosine (8-OHdG) (Halliwell 2007). To investigate whether the increment of DNA damages was related to intracellular ROS level following cell fusion, we analyzed intracellular ROS level for three days that was performed by flow cytometry and confocal microscopy, using the oxidation-sensitive fluorescent probe DCFH-DA. The diacetate form of DCF (H2-DCFDA) is taken up by cells and hydrolyzed to

membrane-impermeant dichlorofluorescein (DCFH), which is oxidized in the presence of ROS to form the highly fluorescent dichlorofluorescein (DCF).

Also, to test whether only the procedure of cell fusion has the probability of ROS production, we checked ROS level in unfused and fused immediately after cell fusion. It was proved not to effect on intracellular ROS level, both unfused and fused cells (data not shown).

Interestingly, intracellular ROS level in fused cells dramatically increased after 24h and attenuated with the treatment of N-acetyl-cycteine (NAC), ROS scavenger, decreasing DNA damages (Fig. 11D-G). It suggests that DNA damages in fused cells are related increased ROS level immediately after cell fusion.

Next, we checked the cell viability whether the treatment of NAC could affect to cell fate in fused cells. But we could not have any effect of NAC on cell fate (data not shown). All the results demonstrated that oxidative DNA damage could be lead by cell-to-cell fusion and the prevention of DNA damage is important to maintain genomic integrity.

(43)

33

Fig. 11. Fused cells increased the ROS level and induced DNA damages. Cells were harvested at indicated time points following cell fusion. (A) Representative images of immunostaining with phosphorylated histone H2AX (γ-H2AX, green), and nuclear DAPI staining (blue) of unfused and fused cells. Images were taken by fluorescence microscopy (ZEISS). (B) Quantification of

(44)

34

the γ-H2AX level at interphase cells. γ-H2AX foci positive cells were determined by the number of foci (more than 5). (C) Immunoblots of fused and unfused cells were probed with antibodies to γ-H2AX and α-tubulin, as a loading control. (D) Intracellular ROS level was measured by using H2DCF-DA. (E) Intracellular ROS level was measured by using H2DCF-DA. Cells were treated

with or without NAC (5mM) and harvested at day 3 following cell fusion. (F) Immunoblots of fused and unfused cells were probed with antibodies to γ-H2AX and α-tubulin, as a loading control. Cells were treated with or without NAC (5 mM) and harvested at 3 day following cell fusion. (G) Quantification of the γ-H2AX level at interphase cells. High leveled γ-H2AX was determined by the intensity of foci. Cells were harvested at day 3 with or without NAC treatment following cell fusion. Results are given as the mean ± SD from three independent experiments, *** p< 0.001, ** p< 0.01 by Student's t-test.

(45)

35

7. Survivin is necessary for the survival of fused cells that escape apoptotic crisis. Since a major fraction of fused cells died through caspase-dependent apoptosis, averting apoptosis might be very important for those cells that managed to escape this crisis after cell fusion. We therefore measured the expression of various anti- and pro-apoptotic proteins after cell fusion. As shown in Figure 12A, the expression level of survivin, a well-known anti-apoptotic protein, was clearly increased in fused cells compared to unfused cells as early as 3 days after fusion and remained elevated throughout the experimental period. This expression pattern was quite different from that of Bcl2, another anti-apoptotic protein, and BAG1 (BCL2 associated athanogene 1), a pro-apoptotic protein. Interestingly, Bcl2 expression increased on day 7 after fusion in both unfused and fused cells for reasons that are not yet clear. To address whether the increased expression of survivin is a characteristic of surviving fused cells, we measured survivin expression levels in stable clones of fused cells established by the limiting-dilution procedure. As shown in Figure 12B, all four established cell lines of fused cells showed a variable, but clear, increase in the expression of survivin, but not Bcl2 or BAG1, compared with stable clones of unfused cells, strongly suggesting that overexpression of survivin is a common characteristic of surviving fused cells.

Moreover, in colony-forming assay, two different small inhibitory RNA (siRNA) that effectively decreased survivin expression, reduced the survival fraction of fused cells by 70~80% compared to control siRNA treatment, whereas knockdown of survivin in unfused HeLa cells resulted in approximately a 40~50% decrease in the survival fraction after fusion compared with control siRNA treatment (Fig. 13A). Therefore, although siSurvivin reduced cell survival in unfused cells, it exerted more deleterious effects on the survival of fused cells. Further analysis using time-lapse monitoring confirmed that knockdown of survivin reduced the cell survival significantly more in fused cells, as compared to unfused cells (Fig. 13B). These data suggest that the fused cells that overexpress survivin have the potential to avert the apoptotic crisis of fused cells, and thereby survive to become more stable cells that can proliferate continuously.

(46)

36

Fig. 12. Survivin, anti-apoptotic protein increased in surviving fused cells. (A) Cells were harvested at indicated time points following cell fusion. Western blots of fused and unfused cells were probed with indicated antibodies. GAPDH: loading control. (B) Immunoblotting was performed with cell lysates derived from fused or unfused stable lines, which were established by using selective medium containing G418 (1 mg/ml) and Hygromycin (0.8 mg/ml) for 3 weeks. Numbers indicate different stable clones.

(47)

37

Fig. 13. Depletion of survivin decreased cell growth in fused cells. (A) The absence of survivin decreases the clonogenic survival of freshly generated fused cells. Parent HeLa cells and mixture of fused and unfused cells were transiently transfected with si-Survivin (40 nM) and si-Control (40 nM) at 1 day after cell fusion. After 24 hours, cells were reseeded and cultured in with or without selective medium for 10 days. Representative pictures of colonies formed by survivin-depleted cells and quantification was counted as observed upon crystal violet staining. Cell lysates from the indicated samples at 24 h after transfection were subjected to western blot analysis using anti-survivin and anti-GAPDH antibodies. Results are given as the mean ± SD from three independent experiments ; ***, p<0.001 by Student's t-test. (B) Cell viability was checked by

(48)

38

using time-lapse images. Cells were transfected with indicated siRNAs at day 1 after fusion, placed in a stage-top incubation chamber, and imaged using an automated fluorescence microscope for 60 hours. Black bars mean the time of mitosis.

(49)

39

8. Survivin protein in fused cells is localized in the cytosol and shows increased stability.

We subsequently examined how survivin increased in fused cells. The quantitative RT-PCR analysis clearly showed no significant increase in survivin mRNA in cells following fusion (Fig. 14A), suggesting a post-transcriptional mechanism. Therefore, we compared survivin degradation rates in fused and unfused cells. In unfused cells, the half-life (t1/2) of survivin protein was less

than 30 minutes (26.3 ± 3.9 min), in agreement with previously published reports (Zhao et al., 2000). However, in the case of fused cells, t1/2 was increased significantly (48.4 ± 3.0 min) (Fig.

14B).

Since cytoplasmic and/or mitochondrial survivin is considered to be cytoprotective (Li et al., 2005), we assessed the subcellular localization of survivin by western blotting and immunocytochemical analysis. Interestingly, we observed a robust increase of survivin in the cytoplasmic fraction of fused cells, resulting in approximately a 3-fold increase in the survivin cytosolic-to-nuclear ratio (Fig. 15A-B). These data clearly suggest that both the increase in protein stability of survivin and preferential localization to the cytosol contribute to the survival of some fraction of fused cells.

(50)

40

Fig. 14. Increased protein stability of survivin upon cell fusion. (A) Expression of Survivin mRNA for 7 days after cell fusion. Cells were harvested at indicated time points following cell fusion. mRNA levels were measured by real-time PCR. Values represent the mean ± SD of 3 independent experiments and were expressed as the relative mRNA accumulation corrected using β-actin mRNA as an internal standard. **, p<0.01 by Student's t-test. (B) The stability of survivin protein was analyzed by Western blotting of the whole cell lysates prepared from cells after addition of 50 μM cycloheximide (CHX). α-tubulin was used as the loading control.

(51)

41

Fig. 15. Increased cytosolic localization of survivin upon cell fusion. (A) Immunocytochemistry of unfused and fused cells against survivin, F-actin and nucleus (DAPI). The graph shows the difference in survivin localization between cytoplasm and nucleus analyzed by the fluorescence intensity of images (Cytoplasmic/Nuclear survivin). (B) The cytoplasmic and nuclear extracts were subjected to western blot analysis using anti-survivin, anti-lamin B1 and anti-tubulin antibodies. The bar graph shows the quantification of western blot images (Cytoplasmic/Nuclear extracts). Band intensity was normalized to tubulin (cytosolic fraction) and lamin B1 (nuclear fraction), respectively. Mean ± SD from three independent experiments; **, p<0.01 by Student's t-test.

(52)

42

9. Fused cells increased cell migration and showed higher resistance than unfused cells in cisplatin treatment.

To measure cell migration ability, I used in vitro scratch assay. As shown in the Fig. 16A, it showed migration ability, compared to unfused cells, more than 4 times in the fused cell. The result was shown that the ability of each fused cell line varied (Fig. 16B).

Cell growth was analyzed during treatment with anticancer drugs in order to examine the effect of fusion cells on cancer progression. Unfused and fused cells were treated with several representative anticancer drugs, cisplatin, etoposide and adriamycin to confirm cell viability by XTT assay. It was proved that fused cells showed much higher viability than unfused cells as a result from after 24 hours treatment of cisplatin and it means that fused cells higher intrinsic resistance on cisplatin (Fig. 16A).

And selective pressure, which results from natural selection of tumor cell clones carrying (epi) genetic alterations that confer resistance to cytotoxic treatment was exposed with cisplatin for 2 month. Resistant-HeLa cells, cisR0, cisR1 and cisR2 cells were obtained by treating cisplatin with 0, 1 or 2 µg/ml concentrations for 24 hours, twice every week on unfused and fused cells. The experiment was that the cells with or without selective pressure were seeded on 96well with a different concentration (0, 3 or 20 µg/ml) of cisplatin for 24 hours, and counted living cells after 48 hours. As a result, it was known that resistance was acquired by selective pressure for 8 weeks showing the increment of cell proliferation in cisR1 and cisR2 of unfused and fused cells at a 3 µg/ml cisplatin. In particular, when cisplatin was treated at low concentration (3 µg/ml) or high concentration (20 µg/ml), the cell growth of fused cells under selective pressure with cisplatin was remarkably increased. This result suggests that fused cells are superior not only in acquiring resistance to anticancer drugs but also in acquisition rate (Fig. 17B).

(53)

43

Fig. 16. Fused cells increased cell migration. (A) Confluent monolayers of cells were subjected to scratch assay, as described in Materials and methods Section. Phase contrast microscopy images of fused and unfused cells, as a control in the scratch assay at beginning at the 0 h and after 24 h, are shown. Results are given as the mean ± SD are indicated, *** p<0.001 by Student's t-test. (B) In vitro scratch assay in unfused and fused stable lines.

(54)

44

Fig. 17. Fused cells showed more chemoresistant than unfused cells treating cisplatin by selective pressure. (A) Primary resistance was determined by cell viability with XTT assay. Cells were seeded in 96-well plate at a density of 800 cells per well and treated with various concentration of cisplatin, etoposide and adriamycin for 48 hours. (B) Selective resistance was measured by cell counting. Selective pressure was conducted by using the treatment of cisplatin (0, 1, 2 μg/ml) for 24 hours once or twice in a week for 8 weeks. Cells were seeded in a 96-well plate at a density of 500 cells per well and treated with cisplatin (3, 20 μg/ml) for 48 hours. Results are given as the mean ± SD are indicated; **, p<0.01; *, p<0.05 by Student's t-test.

(55)

45

IV. DISCUSSION

Tetraploidy is accepted as a potential precursor of cancer-associated aneuploidy, and considered to be a possible cause of tumor formation as well as tumor progression. Tetraploid cells can be produced either by cell fusion or cytokinesis failure. As a first step to understand the implication of cancer cell fusion in tumor progression, we tried to describe the fate of fused cancer cells and underlying molecular explanations related with cell fate. Actually, researchers already showed that most tetraploid cells resulting from non-cancerous cell fusion would undergo p53-dependent cell cycle arrest or apoptosis (Andreassen et al., 2001; Aylon and Oren 2011; Vogel et al., 2004). We used HeLa cells as a model system because HeLa cells are well-known to have HPV E6 protein (Scheffner et al., 1990), thus have low amount/activity of p53, which condition we frequently encounter in cancer. Interestingly, we found that massive apoptotic cell death or growth arrest occurred a few days after fusion even in HeLa cells (Fig. 2 and 4), and surprisingly, it was accompanied by an increase of p53 stability (Fig. 3 and 4F). Induction of p53 and p53-induced cell death processes in HeLa cells has been reported previously (Behera et al., 2014; Guo et al., 2014), suggesting that strong apoptotic stimuli could overcome E6-induced downregulation of p53. In addition, we also observed that fused cells with larger nuclei, indicating larger DNA contents, preferentially died after fusion (Fig. 4B-E), suggesting that a larger deviation from normal DNA content is a strong inducer of apoptosis.

Notably, a fraction of cells escaped cell death and proliferated, and these surviving fused cells were characterized by upregulation of survivin (Fig. 5-8). Survivin is the smallest member of the inhibitor of apoptosis (IAP) family proteins, and plays a key role in inhibiting apoptosis by blocking caspase activation (Chen et al., 2016). In addition, survivin has been reported to not only exert anti-apoptotic functions, but also cell proliferative functions, reflecting its involvement in forming the chromosome passenger complex, which is crucial for the normal progression of the cell cycle (Cheung et al., 2013). Therefore, the overexpression of survivin probably affected the survival/proliferation of fused cells shortly after fusion, possibly providing fused cells the power to overcome the apoptotic crisis.

Regarding the mechanism underlying the increase in survivin, both transcriptional and post-translational regulation were majorly considered (Chen et al., 2016). It has been reported that the

(56)

46

transcription factors SP1, E2F and HIF-1α (hypoxia-inducible factor 1) increase survivin expression, whereas p53 and the forkhead box transcription factors, FOXO1 and FOXO3, decrease it (Chen et al., 2016). Notably, hypoxia upregulates both HIF-1α and survivin expression in HeLa cells (Bai et al., 2013). However, our quantitative reverse transcription-polymerase chain reaction (RT-PCR) analysis clearly showed no significant increase in survivin mRNA in cells following fusion, but rather decreased compared to unfused cells (Fig. 7A). In relation to post-translational regulation, it has been previously reported that heatshock protein 90 increases the stability of survivin (Fortugno et al., 2003). Although we observed the delay in survivin degradation (Fig. 7B), we could not observe the changes in the amount of heatshock protein 90 in fused cells (data not shown). Thus, the molecular mechanism responsible for the upregulation of survivin found here remains to be determined.

Anti-apoptotic activities of survivin is also dependent on its intracellular localization. Since cytoplasmic and/or mitochondrial survivin is considered to be cytoprotective (Li et al., 2005), we assessed the subcellular localization of survivin. Interestingly, we observed a robust increase of survivin in the cytoplasmic fraction of fused cells (Fig. 8), suggesting that both the increase in protein stability and preferential localization to the cytosol worked in fused cells. Inhibition of nuclear export of survivin by using leptomycin B, however, did not decrease the survival of fused cells in our hand (data not shown), which might be due to the non-specific effects of leptomycin B to the nuclear export of diverse array of proteins other than survivin.

Here, we confirmed that polyploid cells generated from the cell fusion between HeLa cells might be result in chromosomal instability, which it could provoke the cancer progression.

Above all, we found that the depolymerization process of fused cells resulted in the formation of multipolar spindles with supernumerary centrosomes, which led to DNA bridges among various mitotic defects resulting in asymmetric division due to cytokinesis failure. In addition, in the fate chasing of daughter cells that had undergone asymmetric division, relatively large nuclear cells tended to be removed by cell death, resulting in a decrease in the DNA content after cell fusion (Fig. 5B).

However, it showed genetic instability rather than simple DNA reduction. This suggests that the surviving fusion cells may have a large impact on the cancer progression and may contribute

(57)

47

to more severe oncogenesis.

In addition, it was assured that the ability of cell migration in fused cells is superior to that of unfused cells and that this ability vary for each stable line. A significant finding from these results is to show that fused cells are much more resistant to the cisplatin, a representative anti-cancer drug with or without selective pressure for 8 weeks as well as the ability of cell migration (Fig. 16 and 17).

(58)

48

V. CONCLUSION

Fusion between cervical cancer cells, HeLa cells, led to vast cell death, even though parental cells had low levels of p53. In the process, the fused cells showed mitotic defect such as formation of spindle multipolarity due to increased centrosomes, resulting in asymmetric division. In addition, when the fate of daughter cells was traced, multinucleated cells showed tendency to die. However, surviving fused cells, were characterized by upregulation of survivin, reflecting increased survivin protein stability. Moreover, in fused cells, survivin became preferentially localized to the cytosol, where it is known to exert its anti-apoptotic function. Knockdown of survivin decreased survival to a greater extent in fused cells than in unfused cells, suggesting that fused cells became more dependent on survivin. Therefore, above findings indicate that, after cancer cell fusion, a subpopulation of fused cells with a higher level of cytosolic survivin are able to avoid apoptotic crisis and survive to proliferate continuously, a process that might contribute to human cancer progression. Fused cells occuring depolyploidization had similar amounts of DNA and proliferated but showed genetic instability, which is far superior to cell migration or chemoresistance, which could have detrimental effects on cancer therapy. This suggests that fused cells made from isotypic parental cells may lead to more severe tumor heterogeneity, a difficult problem in cancer treatment, and that cell fusion may result in genome diversity.

(59)

49

REFFERENCE

1. Andreassen PR, Lohez OD, Lacroix FB, Margolis RL (2001) Tetraploid State Induces p53-dependent Arrest of Nontransformed Mammalian Cells in G1. Molecular Biology of the Cell 12:1315-1328

2. Aylon Y, Oren M (2011) p53: guardian of ploidy. Molecular oncology 5:315-323

3. Bai H, Ge S, Lu J, Qian G, Xu R (2013) Hypoxia inducible factor-1alpha-mediated activation of survivin in cervical cancer cells. J Obstet Gynaecol Res 39:555-563 4. Bayani J, Squire JA (2005) Comparative genomic hybridization. Current protocols in cell

biology Chapter 22:Unit 22.26

5. Behera B, Mishra D, Roy B, Devi KS, Narayan R, Das J, Ghosh SK, Maiti TK (2014) Abrus precatorius agglutinin-derived peptides induce ROS-dependent mitochondrial apoptosis through JNK and Akt/P38/P53 pathways in HeLa cells. Chem Biol Interact 222:97-105

6. Castedo M, Coquelle A, Vitale I, Vivet S, Mouhamad S, Viaud S, Zitvogel L, Kroemer G (2006) Selective resistance of tetraploid cancer cells against DNA damage-induced apoptosis. Annals of the New York Academy of Sciences 1090:35-49

7. Chen X, Duan N, Zhang C, Zhang W (2016) Survivin and Tumorigenesis: Molecular Mechanisms and Therapeutic Strategies. Journal of Cancer 7:314-323

8. Cheung CH, Huang CC, Tsai FY, Lee JY, Cheng SM, Chang YC, Huang YC, Chen SH, Chang JY (2013) Survivin - biology and potential as a therapeutic target in oncology. OncoTargets and therapy 6:1453-1462

9. Duelli DM, Hearn S, Myers MP, Lazebnik Y (2005) A primate virus generates transformed human cells by fusion. J Cell Biol 171:493-503

10. Foley EA, Kapoor TM (2013) Microtubule attachment and spindle assembly checkpoint signalling at the kinetochore. Nature reviews Molecular cell biology 14:25-37

11. Fortugno P, Beltrami E, Plescia J, Fontana J, Pradhan D, Marchisio PC, Sessa WC, Altieri DC (2003) Regulation of survivin function by Hsp90. Proceedings of the National

수치

Table 1. Phenotypes of cell cycle in F0 cells…………………………………………22
Fig. 1. Generation and isolation of fused cells. (A) Schematic diagram of the cell fusion and  selection  procedure  is  shown
Fig. 2. Analysis of cell fate after  cancer  cell fusion.  (A) Representative images of fused and  unfused cells obtained from fusion of DiI + - and DiO + -HeLa cells  as in Fig.1A by fluorescence
Fig.  3.  Preferential  elimination  of  the  meganucleated  cells  with  larger  nuclei  through  apoptosis
+7

참조

관련 문서

(1) PCR monitoring of bphC gene of Ralstonia eutropha H850 in soil Amount of bphC gene amplification of H850 cells was not generally proportional to the cell density

Glutamate excitotoxicity induced by excessive activation of NMDA receptor causes various damage to cells, which leads to cell death.. In previous studies, increased ROS

Effect of cell migration of JMJD6 transcript variants over- expressed MCF-7 cells using Transwell.. Effect of cell migration of JMJD6 transcript variants

Effect of LRP6 on survivin expression in hypoxic cardiomyocytes (A) Expression of survivin in Ad-LRP6 and silencing LRP6 infected cardiomyocytes as determined by western

ƒ The amount of a compound per unit cell mass (or cell The amount of a compound per unit cell mass (or cell volume).

Activated Foxo3a can inhibit the activation of p53 by DNA damage [46] and Resulted in a significant down-regulation of both p21 and p53 expression in Foxo3a- siRNA U2OS cells

Recent stem cell studies have reported that cultured hematopoietic stem cells (HSCs) are reactivated through fetal hemoglobin expression by treatment with

– Established the fundamentals of high temperature fuel cells such as the molten carbonate and solid oxide fuel cell.. – In 1958, he demonstrated an alkali cell using a