Implantation is the first maternal-embryo crosstalk that only occurs during a finite period called the, implantation window'. HOXA10, a homeobox transcription factor, plays an important regulatory role during this period. However, the target genes of HOXA10 involved in implantation and decidualization have not been identified. Using a chromatin immunoprecipitation screen, we identified the p300/CBPassociated factor (p/CAF) as a direct HOXA10 target gene in vivo. Adenovirus-mediated overexpression and siRNA-specific knockdown of HOXA10 altered p/CAF promoter activity via interaction with the three consecutive TTAT element units in human endometrial cells. These results indicate that p/CAF is a novel HOXA10 target gene, and HOXA10 promotes human endometrial development, at least in part, through the regulation of p/CAF gene.

Calmodulin (CaM) is a major cytoplasmic calcium receptor that performs multiple functions including cell motility. To investigate the mechanism of the regulation of CaM on cell morphology and motility, first we checked the distribution of CaM in the living cells using GFP-CaM as an indicator. We found that GFP-CaM showed a fiber-like distribution pattern in the cytosol of living Potorous tridactylis kidney (PtK2) cells but not in living HeLa cells. The endogenous CaM in heavily permeabilized HeLa was also found to display a fiber-like distribution pattern. Further examination showed that the distribution pattern of GFP-CaM was same as that of stress fibers, but not microtubules. Co-immunoprecipitation also showed that CaM can interact with actin directly or indirectly. The microinjection of trp peptide, a specific inhibitor of CaM, attenuated the polymerization of stress fibers and induced the alteration of cell morphology. A wound-healing assay and a single cell tracking experiment showed that CaM in PtK2 cells could increase cell motility. The data we have got from living cells suggested that CaM affect cell morphology and motility through binding to stress fibers and regulate f-actin polymerization.

Ultraviolet (UV) irradiation can result in cell cycle arrest. The reactivation of Polo-like kinase 1 (Plk1) is necessary for cell cycle reentry. But the mechanism of how Plk1 regulates p53 in UV-induced mitotic arrest cells remained elusive. Here we find that UV treatment leads HEK293 cells to inverse changes of Plk1 and p53. Over-expression of Plk1 rescue UV-induced mitotic arrest cells by inhibiting p53 activation. Plk1 could also inhibit p53 phosphorylation at Ser15, thus facilitates its nuclear export and degradation. Further examination shows that Plk1, p53 and Cdc25C can form a large complex. Plk1 could bind to the sequence-specific DNA-binding domain of p53 and active Cdc25C by hyperphosphorylation. These results hypothesize that Plk1 and Cdc25C participate in recovery the mitotic arrest through binding to the different domain of p53. Cdc25C may first be actived by Plk1, and then its phosphatase activity makes p53 dephosphorylated at Ser15.

c-Met is highly expressed and constitutively activated in various human tumors. We employed adenovirus-mediated RNA interference technique to knock down c-Met expression in hepatocellular carcinoma cells and observed its effects on hepatocellular carcinoma cell growth in vitro and in vivo. Among the five hepatocellular carcinoma and one normal human liver cell lines we analyzed, c-Met was highly expressed and constitutively tyrosine phosphorylated in only MHCC97-L and HCCLM3 hepatocellular carcinoma cells. Knockdown of c-Met could inhibit MHCC97-L cells proliferation by arresting cells at G0-G1 phase. Soft agar colony formation assay indicated that the colony forming ability of MHCC97-L cells decreased by f70% after adenovirus AdH1-small interfering RNA (siRNA)/met infection. In vivo experiments showed that adenovirus AdH1-siRNA/met inhibited the tumorigenicity of MHCC97-L cells and significantly suppressed tumor growth when injected directly into tumors. These results suggest that knockdown of c-Met by adenovirus-delivered siRNA may be a potential therapeutic strategy for treatment of hepatocellular carcinoma in which c-Met is overexpressed.

A transient postanaphase repositioning of the centriole is found to control the completion of cytokinesis. Using a green fluorescent protein-calmodulin fusion protein as a living cell probe, we have previously found that calmodulin is associated with the initiation and progression of cytokinesis. In this study, we further studied the effect of calmodulin on the repositioning of the centriole and subsequent cell cycle progression. When activity of calmodulin is inhibited, the regression of the centriole from the intercellular bridge to the cell center is blocked, and thus the completion of cell division is repressed and two daughter cells are linked by longer cell bridge in perturbed cells. W7 treatment during cytokinesis also results in unfinished cytokinesis and stopped G1 phase. These results suggest that calmodulin activity is required for centriole repositioning and can affect the completion of cytokinesis and cell cycle progression.