背景と目的

The processes of DNA replication are tightly regulated so that it occurs only once in each cell cycle. Defective regulation in DNA replication and cell division often leads to genome instability and cancer development. Cdc7 (cell division cycle 7), an evolutionary conserved serine/threonine kinase, is an important DNA replication regulatory protein overexpressed in many human tumors and neoplastic cells. Studies demonstrating cell death in Cdc7-depleted cancer cells but not that in the normal fibroblast highlighted Cdc7 as a useful candidate target for cancer therapy. Nevertheless, a limited knowledge on the function of human Cdc7 which mostly focuses on the S phase of the cell cycle so far may pose a potential limitation to the significance and potential development of Cdc7 inhibitor as an anti-cancer drug. Hence, the present study was undertaken to unravel the biological functions of Cdc7 during mitotic (M) phase of the cell cycle in human cells.

実験結果

Cdc7 is a cell-cycle regulated protein abundantly expressed in mitotic phase.

By using HeLa cells released from double-thymidine block (dTB), I show that human Cdc7 is a cell-cycle regulated protein, accumulating from G1/S phase transition and peaking at M phase (Fig. 1A). Consistent with the endogenous data, time-lapse imaging of HeLa cells expressing monomeric-Azami Green (mAG)-tagged Cdc7 shows similar oscillation during cell cycle with predominant nuclear localization during S phase and cytoplasmic localization during M phase (Fig. 1B). Examination of fluorescent-tagged human Cdc7 regulatory subunit, Dbf4/ASK or Drf1/ASKL1, in HeLa cells also show oscillation and nuclear-cytoplasmic localization similar to that of the mAG-tagged Cdc7-fusion.

Regulation of Cdc7 protein level during cell cycle.

I then examined the mechanism that regulates Cdc7 protein in human cells. Cycloheximide (CHX) at 75 μg/mL, a translation inhibitor, reduces Cdc7 protein level at as soon as 0.5 hr post-treatment, indicating that Cdc7 is an unstable protein (Fig. 1C). In contrast, MG132, a proteasome inhibitor, results in accumulation of Cdc7 after 24 hr-treatment at 5 μM, suggesting regulation of Cdc7 protein by the proteasomal degradation pathway (Fig. 1D). In silico analysis reveals two putative Destruction (D)-box motifs and a D-box Activating Domain (DAD)-box in the Cdc7 sequence. Among these mutants, mAG-tagged Cdc7 R(373)xxT>AxxA mutant shows slight stabilization during the cell cycle, suggesting that R(373)xxT motif may be a potential D-box targeting Cdc7 for APC/C(Cdh1)-dependent proteolysis in early G1 phase. However, this mutant does not cause any significant cell cycle effect possibly due to the absence of sufficient Dbf4/ASK for Cdc7 kinase activation.

Potential roles of Cdc7 during transition from M to G1 phase.

Chromatin binding and band shift of Cdc7 protein in Nocodazole-treated HeLa cells suggest phosphorylation and potential functionality of Cdc7 during M phase (Fig 1E). Here, I show that Cdc7 interacts with both N-term kinase domain (1-330 amino acids) and C-term Polo-box domain (331-603 residues) of human Plk1, an important mitotic kinase and cell cycle regulator (Fig. 2A, B). Cdc7 also interacts with human Plk2 and Plk3 after coexpression albeit with reduced affinity (Fig. 2A). Noteworthy, coexpression of Cdc7 and Dbf4/ASK rescued the G2/M phase arrest caused by Plk1 overexpression, although this rescue is independent of Cdc7 kinase activity since coexpression of Cdc7 kinase-dead form also leads to similar effect (Fig. 2C). Kinase activity of Plk1 is not affected but its interaction with endogenous Cdt1, an important DNA licensing factor, is weakened by Cdc7-ASK overexpression (Fig. 2D, E). Further examination shows that Cdt1 interacts with the N-term 331-603 residues of Plk1, suggesting that Cdc7 and Cdt1 may compete for binding to the N-terminus of Plk1. Based on these observation, I propose that Plk1 overexpression causes cell cycle arrest by inhibiting Cdt1 function through direct interaction, and Cdc7 may counteract the arrest by competing with this interaction.

Functional interactions between Cdc7 and mitotic kinases.

Cdc7 kinase activity appears to be downregulated by Plk1 since the Cdc7 autophosphorylation level was slightly reduced upon coexpression (Fig. 2E). I have identified the T68 residue as a potential Plk1-mediated phosphorylation site on Cdc7. In vitro experiments show that Plk1 reduces the Cdc7-mediated phosphorylation of the MCM complex, and that Cdc7 can stimulate kinase activity of Aurora B. Taken together, these results suggest that Cdc7 may regulate M phase progression through diverse functional interactions with mitotic kinases.

結論

Cdc7 is a cell cycle regulated protein abundantly expressed in M phase and degraded at G1 phase potentially through APC/C(Cdh1) dependent pathway. Interaction between Cdc7 and Plk1 in M phase may be crucial to regulate the timing for DNA origin licensing by releasing Cdt1 from Plk1 (Fig 3). Potential significance of functional interactions between Cdc7 and mitotic kinases for progression of mitosis is also indicated. Thus, findings from this study will provide new insights into the roles of Cdc7 during M phase and facilitate the understanding of the molecular pathways for maintenance of genomic stability.

Figure 1. Dynamics and protein stability of human Cdc7 protein in HeLa cells. (A) Cdc7 protein level during cell cycle. HeLa cells were arrested at G1/S phase by double thymidine block (dTB). Cells were harvested at every 2-hr interval after release and whole cell extract (WCE) was separated by SDS-PAGE and immunoblotted with the antibodies against the proteins indicated. (B) Time-lapse imaging of mAG-tagged Cdc7-fusion (Green) in HeLa cells during cell cycle. (C, D) Levels of Cdc7 after Cycloheximide (CHX) and/or MG132 treatment, as indicated in the figure. Plk1 was run as a control since it is known to be degraded via APC-C dependent pathway. (E) Subcellular localization of Cdc7. HeLa cells were synchronized at G1/S boundary or G2/M phase by using dTB or Nocodazole block, respectively. Triton-soluble and -insoluble extracts were prepared by using CSK buffer containing 0.5% triton-X.

Figure 2. Physical and functional interactions between Cdc7 and Plk1. (A) Interaction between overexpressed Cdc7 and Plk1, Plk2 and Plk3. 293T cells were transfected using polyetherimide (PEI) with the indicated plasmids for 48 hrs. Triton-soluble extracts were prepared and immunoprecipitates by anti-FLAG (Plks) antibody were separated by SDS-PAGE and immunobloted with antibodies indicated. (B) Interaction between endogenous Cdc7 and Plk1. HeLa cells were synchronized as described in (1E). Triton-soluble extracts were prepared and immunoprecipitation was conducted by using anti-Plk1 or anti-Cdc7 antibody. (C) Effects of Plk1, Cdc7 (WT: wildt-ype; KD: kinase-dead) and/or ASK transfection on the cell cycle of 293T cells. Transfection was performed as described in (A) prior to preparation for FACS analysis. (D) Triton-soluble extracts were prepared. The immunoprecipitates by anti-HA (Cdc7) or anti-FLAG (Plk1) antibody were examined by immunoblotting using antibodies against the replication factors indicated. Interaction between Plk1 and Cdt1 as well as that between Plk1 and MCM2 Ser40/41 were affected upon Cdc7-ASK coexpression. (E) Immunoprecipitates by anti-HA (Cdc7) or anti-FLAG (Plk1) antibody were prepared and used for in vitro kinase assays.

Figure 3. A proposed model for regulation of DNA origin licensing by Plk1 and Cdc7. Plk1 and Cdt1 interaction may prevent chromatin loading of Cdt1, thereby reduce recruitment of MCM complex to replication origin to form pre-Replication Complex (pre-RC). Cdc7-ASK may facilitate release of Cdt1 from Plk1 through competitive interaction with Plk1 and thus allow origin licensing.

This thesis comprises of four (4) chapters: (I) Introduction, (II) Results, (III) Discussion and Conclusion, and (IV) Materials and Methods. In the first chapter, background on DNA replication and functions of Cdc7 reported to date are elaborated. In the second chapter, novel functions of human Cdc7 during G2 and M phases of the mammalian cell cycle are discussed. The third chapter includes the general discussion and concluding remarks, while the forth chapter covers the research methodologies related to this work.

Cdc7 (cell division cycle 7), an evolutionary conserved serine/threonine kinase, is an important DNA replication regulatory protein overexpressed in many human cancers and neoplastic cells. Studies demonstrating selective induction of cell death in the human cancer cells but not in the normal fibroblasts after Cdc7 inhibition have highlighted Cdc7 as a useful candidate target for cancer therapy. Nevertheless, limited knowledge on the biological functions of human Cdc7 outside S phase may pose limitations to the potential and specificity of the Cdc7 inhibitors as anti-cancer drugs. In this work, Toh Gaik Theng (トウ ゲック ティング)describes the novel roles of human Cdc7 as an important regulator during G2 and M phase progression through evaluations of the physical and/or functional interactions between Cdc7 and mitotic kinases such as Plk1 and Aurora B.

Section 2.1 and 2.2 of this thesis show the oscillation pattern of human Cdc7 protein during cell cycle through western blot analysis and time-lapse imaging Cdc7-fusion expressing cells. Cdc7 is suggested as an unstable protein which undergoes degradation at mitotic exit. Section 2.3 and 2.4 describe Cdc7 as target of APC/CCdh1-mediated proteolysis which carries a putative degradation box (D-box) in its C-terminal. Section 2.5 and 2.6 demonstrate chromatin localization of Cdc7 in Mitotic (M) phase and the effects of siRNA-mediated Cdc7 depletion on cell cycle progression, particularly on that of G2- and M-phase. Delayed G2 phase progression and inefficient mitotic exit (i.e. M/G1 transition) observed in the Cdc7-depleted cells implicate the involvement of Cdc7 in regulating these cell cycle phases. Consistent with this, section 2.7 shows identification of the Plk1 mitotic kinase as a Cdc7 interacting protein during M phase. Section 2.8 and 2.9 further discuss about the functional significance of the Cdc7 and Plk1 interaction in vivo. Simultaneous expression of Cdc7, Dbf4/ASK (a Cdc7 regulator) and Plk1 alleviates G2/M phase arrest caused by Plk1 overexpression, plausibly by improving M/G1 transition since interaction between Plk1 and Cdt1 DNA licensing factor is weakened in these cells. Cdc7 is suggested to antagonize interaction between Plk1 and Cdt1 via N-terminus of Plk1, thereby releasing Cdt1 necessary for efficient pre-replicative complex formation during DNA origin licensing at M/G1 transition. Stimulation of Aurora B kinase activity by Cdc7 in vitro which implies a potential crosstalk between Cdc7 and multiple mitotic kinases during cell cycle regulation is also discussed. Finally, section 2.10 confirms that depletion of Cdc7 by siRNA reduces Cdt1 chromatin loading and accumulation of G2/M phase population, similar to that observed in the Cdt1-depleted cells. Based on these findings, a mechanism by which Cdc7 regulates DNA origin licensing through interaction with Plk1 is proposed, unraveling novel roles of Cdc7 during cell cycle regulation.

With the abovementioned, I hereby acknowledge that the scope and quality of this thesis is qualified for the award of the degree of Doctor of Philosophy in Life Sciences.