Kohzaburo Fujikawa-Yamamoto

Induction of centrosome amplification and chromosome instability in human bladder cancer cells by p53 mutation and cyclin E overexpression

Centrosome amplification frequently occurs in human cancers and is a major cause of chromosome instability (CIN). In mouse cells, centrosome amplification can be readily induced by loss or mutational inactivation of p53. In human cells, however, silencing of endogenous p53 alone does not induce centrosome amplification or CIN, although high degrees of correlation between p53 mutation and CIN/centrosome amplification in human cancer can be detected, suggesting the presence of additional regulatory mechanism(s) in human cells that ensures the numeral integrity of centrosomes and genomic integrity. Cyclin E, a regulatory subunit for CDK2 that plays a key role in centrosome duplication, frequently is overexpressed in human cancers. We found that cyclin E overexpression, together with loss of p53, efficiently induces centrosome amplification and CIN in human bladder cancer cells but not by either cyclin E overexpression or loss of p53 alone. We extended these findings to bladder cancer specimens and found that centrosome amplification is strongly correlated with concomitant occurrence of cyclin E overexpression and p53 inactivation but not with either cyclin E overexpression or p53 inactivation alone. Because cyclin E expression is strictly controlled in human cells compared with mouse cells, our findings suggest that this stringent regulation of cyclin E expression plays an additional role underlying numeral homeostasis of centrosomes in human cells and that deregulation of cyclin E expression, together with inactivation of p53, results in centrosome amplification.

Centrosome Hyperamplification and Chromosomal Instability in Bladder Cancer

OBJECTIVE Chromosomal instability (CIN) is a common feature of malignant tumors. Centrosome hyperamplification (CH) occurs frequently in human cancers, and may be a contributing factor in CIN. In this study, we investigated the relationship between CH and CIN in bladder cancer. METHODS Clinical samples obtained by transurethral resection from 22 patients with bladder cancer were examined (histological grade G1, 5 cases; G2, 6 cases; G3, 11 cases). CH was evaluated by immunohistochemistry using anti-pericentrin antibody. CIN was evaluated by fluorescence in situ hybridization (FISH). FISH probes for pericentromeric regions of chromosomes 3, 7, and 17 were hybridized to touch preparations of nuclei from frozen tissues. We also analyzed the centrosome replication cycle of bladder cancer by laser scanning cytometry (LSC). RESULTS Of the 22 cases examined, 18 (81.8%) had centrosome hyperamplification: CH 0, 4 cases (18.1%); CH I, 5 cases (22.7%); CH II, 5 cases (22.7%); CH III, 8 cases (36.4%). The grade of CH was directly proportional to the histological grade (p=0.03, chi(2) test). LSC analysis showed that the centrosome replication cycle was well regulated in pathologically low-grade bladder cancer, which did not have chromosomal instability. In contrast, we found marked variability of centrosomes in pathologically high-grade bladder cancer, which had chromosomal instability. CH and CIN were both detected in pathologically high-grade tumors. The grade of CH was directly proportional to the CIN grade (p=0.0079, chi(2) test). CONCLUSION The results of the present study suggest that CH may be involved in CIN in bladder cancer.

Centrosome Hyperamplification and Chromosomal Damage after Exposure to Radiation

Objective: In order to elucidate the effects of radiation on centrosome hyperamplification (CH), we examined the centrosome duplication cycle in KK47 bladder cancer cells following irradiation. Methods: KK47 cells were irradiated with various doses of radiation and were examined for CH immunostaining for γ-tubulin. Results: Nearly all control cells contained one or two centrosomes, and mitotic cells displayed typical bipolar spindles. The centrosome replication cycle is well regulated in KK47. Twenty-four hours after 5-Gy irradiation, ∼80% of irradiated cells were arrested in G2 phase, and at 48 h after irradiation, 56.9% of cells contained more than two centrosomes. Laser scanning cytometry performed 48 h after irradiation showed the following two pathways: (1) unequal distribution of chromosomes to daughter cells, or (2) failure to undergo cytokinesis, resulting in polyploidy. With mitotic collection, M-phase cells with CH could be divided into G1 cells with micronuclei and polyploidal cells. Fluorescence in situ hybridization analysis showed clear signs of chromosomal instability (CIN) at 48 h after irradiation. The present study had two major findings: (1) continual duplication of centrosomes occurred in the cell cycle-arrested cells upon irradiation, leading to centrosome amplification; (2) cytokinesis failure was due to aberrant mitotic spindle formation caused by the presence of amplified centrosomes. Abnormal mitosis with amplified centrosomes was detected in the accumulating G2/M population after irradiation, showing that this amplification of centrosomes was not caused by failure to undergo cytokinesis, but rather that abnormal mitosis resulting from amplification of centrosomes leads to cytokinesis block. Conclusion: These results suggest that CH is a critical event leading to CIN following exposure to radiation.

Establishment of a tetraploid cell line from mouse H‐1 (ES) cells highly polyploidized with demecolcine

Abstract.  Objective: Establishment of tetraploid ES cells. Materials and methods: Mouse H‐1 (ES) cells were polyploidized by demecolcine and released from the drug. Results: A tetraploid cell line (4nH1 cells) was established from mouse H‐1 (ES) cells (2nH1 cells) highly polyploidized by treatment with demecolcine. Cell cycle parameters of 4nH1 cells were almost the same as those of 2nH1 cells, suggesting that the rate of DNA synthesis was about twice that of the diploid cells. Mode of chromosome number of 4nH1 cells was 76, about twice that of 2nH1 cells. Cell volume of 4nH1 cells was about twice of that of diploid cells, indicating that 4nH1 cells contained about twice as much total intracellular material as 2nH1 cells. Morphology of the 4nH1 cells was flagstone‐like, thus differing from that of the spindle‐shaped 2nH1 cells, suggesting that the transformation had occurred during the diploid–tetraploid transition. 4nH1 cells exhibited alkaline phosphatase activity and formed teratocarcinomas, implying that they would be pluripotent. Conclusion: A pluripotent tetraploid cell line (4nH1 cells) was established.

Establishment of a tetraploid Meth-A cell line through polyploidization by demecolcine but not by staurosporine, K-252a and paclitaxel

Polyploid cells are made by DNA reduplication without cell division, however, it is not easy to establish polyploid mammalian cell lines. It is worth studying the difference in cell character between hyperploid and parent cell lines. Meth‐A cells were polyploidized by demecolcine, K‐252a, staurosporine and paclitaxel. The cell‐cycle responses of highly polyploid Meth‐A cells after the removal of the drugs were examined by flow cytometry (FCM). Meth‐A cells were highly polyploidized by these drugs. The polyploid Meth‐A cells gradually decreased in ploidy after the drug release. A tetraploid Meth‐A cell line was established only from the demecolcine‐induced polyploid Meth‐A cells. The duration of G1, S and G2/M phases of the tetraploid cell line were mostly the same as those of the parent diploid cells, except that the G2/M phase was 1.5 h longer. The chromosome number of tetraploid Meth‐A cell line was about twice of the diploidy. A tetraploid Meth‐A cell line was established.

Establishment of a triploid V79 cell line from tetraploid cells obtained through polyploidization using K‐252a

Abstract. Triploid V79 cells were established from tetraploid cells. Diploid V79 cells were polyploidized by K‐252a, an inhibitor of protein kinases, and then released from the drug for 10 days. At that time, the cell population was a mixture of diploid and tetraploid cells. Triploid cells were obtained through the cloning of tetraploid cells. They had 33 chromosomes (1.5 times the diploid number) and showed a karyotype of three homologueous chromosomes. The duration of the G1, S and G2/M phases was almost the same as for diploid cells. The cell volume of triploid V79 cells was about two times that of the diploid cells. An explanation for the diploid‐tetraploid‐triploid transition is proposed.

Involvement of protein kinase C in taxol-induced polyploidization in a cultured sarcoma cell line

Taxol was found to inhibit the proliferation and to induce the polyploidization of cultured methylcholanthrene-induced sarcoma cells (Meth-A cells). To investigate whether protein kinase C is involved in taxol-induced polyploidization, phorbol 12-myristate 13-acetate (PMA), which regulates the activity of protein kinase C, was used along with taxol to treat the cells. We found that PMA did not interfere with the proliferation and did not induce polyploidization by itself. However, at low concentration, taxol, which by itself did not induce polyploidization, clearly induced polyploidization in the presence of PMA. To explore the mechanism by which PMA potentiates polyploidization, the levels of the G1 checkpoint-related proteins cyclin E and cdk2, and those of the G2 checkpoint-related proteins cyclin B and cdc2 were determined by flow cytometry. We found that both G1 and G2 checkpoint-related proteins increased during the induction of polyploidization. To verify the relationship between protein kinase C and tubulin polymerization, flow cytometry was used to determine the total content of tubulin protein, and morphological observation was used to examine spindle organization. PMA did not affect the taxol-induced increase in tubulin protein, but markedly potentiated taxol-induced spindle disorganization. These findings suggest that protein kinase C plays an important role in regulating the induction of polyploidization in Meth-A cells.

Alteration and preservation of cellular characteristics in long‐term culture of tetraploid H‐1 (ES) cells

To examine the alteration in cellular characteristics of polyploid ES cells during long-term culturing, tetraploid H-1 (ES) cells were continuously cultured for 180 days. Cellular DNA content of the tetraploid cells decreased and reached a plateau of 3.3 C, where C represents the complement of haploid chromosomes. The chromosome number also decreased, indicating that the DNA loss was induced by chromosome loss. Cell volume was maintained, suggesting that the DNA loss did not involve cytoplasmic loss. The cell cycle parameters were almost the same during the DNA decay process, indicating that cell cycle progression was independent of the quantity of homologous chromosomes. Hypotetraploid cells showed alkaline phosphatase activity and formed teratocarcinomas in mouse abdomens, suggesting that the pluripotent potential was maintained. Cellular morphology was also retained, suggesting that the gene expression specifying morphological characteristics was conserved. We conclude that these initial cellular characteristics of tetraploid H1 (ES) cells were preserved in long-term culture, irrespective of chromosome loss.

Octaploid Meth-A cells are established from a highly polyploidized cell population

Abstract.  Tetraploid Meth‐A cells were polyploidized by demecolcin, an inhibitor of spindle fibre formation in M phase, and then released from the drug 1, 2, 3 and 4 days after the addition. Octaploid cells were successfully established from cell populations including hexadecaploid cells produced by 2, 3 and 4 days of exposure to demecolcin. One‐day‐treated cells were polyploidized octaploid cells, but they returned to tetraploid cells. All of the octaploid Meth‐A cells showed essentially the same features. The octaploid Meth‐A cells had eight homologous chromosomes and double the DNA content of the parent tetraploid cells. The doubling time of octaploid Meth‐A cells was 30.2 h, somewhat longer than the 28.3 and 24.0 h of tetraploid and diploid cells, respectively. The fractions of cells in the G1, S and G2/M phases were essentially the same in diploid, tetraploid and octaploid Meth‐A cells. The cell volume of octaploid Meth‐A cells was about two times that of the tetraploid cells. It was concluded that octaploid Meth‐A cells were established from transient hexadecaploid cells produced by the polyploidization of tetraploid cells that had been established from diploid cells.

DNA stable pentaploid H1 (ES) cells obtained from an octaploid cell induced from tetraploid cells polyploidized using demecolcine

Pentaploid H1 (ES) cells (5H1 cells) were accidentally obtained through one‐cell cloning of octaploid H1 (ES) cells (8H1 cells) that were established from tetraploid H1 (ES) cells (4H1 cells) polyploidized using demecolcine. The number of chromosomes of 5H1 cells was 100, unlike the 40 of diploid H1 (ES) cells (2H1 cells), 80 of 4H1, and 160 of 8H1 cells. The durations of G1, S, and G2/M phases of 5H1 cells were 3, 7, and 6 h, respectively, almost the same as those of 2H1, 4H1, and 8H1 cells. The cell volume of 5H1 cells was half of that of 8H1 cells, suggesting that 5H1 cells were created through abnormal cell divisions of 8H1 cells. The morphology of growing 5H1 cells was a spherical cluster similar to that of 2H1 cells and differing from the flagstone‐like shape of 4H1 and 8H1 cells. Pentaploid solid tumors were formed from 5H1 cells after interperitoneal injection into the mouse abdomen, and they contained endodermal, mesodermal, and ectodermal cells as well as undifferentiated cells, suggesting both that the DNA content of 5H1 cells was retained during tumor formation and that the 5H1 cells were pluripotent. The DNA content of 5H1 cells was stable in long‐term culturing as 2H1 cells, meaning that 5H1 and 2H1 cells shared similarities in DNA structure. The excellent stability of the DNA content of 5H1 cells was explained using a hypothesis for the DNA structure of polyploid cells because the pairing of homologous chromosomes in 5H1 cells is spatially forbidden. J. Cell. Physiol. 223: 369–375, 2010.