GSK923295 as a potential antihepatocellular carcinoma agent causing delay on liver regeneration after partial hepatectomy

Abstract

Background: The clinical trials emerged centromere protein E inhibitor GSK923295 as a promising anticancer drug, but its function in hepatocellular carcinoma (HCC) remain needs to be fully elucidated, especially as chemotherapy after hepatectomy for liver tumors. We aimed to describe anti-HCC activities of GSK923295 and compare its antiproliferative effects on liver regeneration after partial hepatectomy (PH).

Methods: All subjects were randomized to treatment with either vehicle or GSK923295. Antitumor activity of GSK923295 was assessed by xenograft growth assays. The C57BL/6 mice were subjected to 70% PH and the proliferation was calculated by liver coefficient, further confirmed by immunohistochemistry. The proliferation and cell cycle analysis of liver cell AML12 and HCC cells LM3, HUH7, and HepG2 were investigated using the cell counting kit-8 assay and Flow Cytometry. The chromosome misalignment and segregation in AML12 cells were visualized by immunofluorescence.

Results: Treatment with GSK923295 induced antiproliferation in HCC cell lines. It also caused delay on HCC tumor growth instead of regression both in a HCC cell line xenograft model and patient-derived tumor xenograft model. With microarray analysis, CENtromere Protein E was gradually increased in mouse liver after PH. Exposure of liver cells to GSK923295 resulted in delay on a cell cycle in mitosis with a phenotype of misaligned chromosomes and chromosomes clustered. In 70% PH mouse model, GSK923295 treatment also remarkably reduced liver regeneration in later stage, in parallel with the mitotic marker phospho-histone H3 elevation.

Conclusion: The anticancer drug GSK923295 causes a significant delay on HCC tumor growth and liver regeneration after PH in later stage.

Figure 1

GSK923295 causes delay on liver tumor growth. (A) The cell proliferative activity of GSK923295-treated hepatocellular carcinoma (HCC) cells (HepG2; LM3; HUH7) for 24, 48, and 96 h were assessed by the CCK-8 assay. The y-axis represents the proliferation rate that calculated as a ratio to normal control (untreated cells). (B) HCC cells were arrested at the G1/S boundary by a double thymidine block (DTB). After release of DTB, HCC cells were exposure to 5 μmol/L GSK923295 for 8 h and were subjected to flow cytometry. Meanwhile, Western blotting the HCC cells treated as above was stained with cyclin B1 and pHH3. (C) After GSK923295 treatment, tumors from the LM3 were significantly smaller than those treated with vehicle at 19th day, while the total body mass showed no significant difference. (D) After 3 days of (arrowheads) GSK923295 administration, the subcutaneous tumor size was calculated and recorded every 2 days after GSK923295 or vehicle administered IP. (E) Representative photomicrograph of an H&E-stained section of LM3 tumor xenografts removed 24 h after a single injection of vehicle or GSK923295. GSK923295-induced appearance of mitotic figures (right, arrow). Scale bar: 50 μm. (F) GSK923295 treatment caused delay on tumor growth in patient-derived tumor xenograft (PDTX) model from a patient with advanced HCC. (G) After 3 days of (arrowheads) GSK923295 administration to PDTX mice, the folds of change of tumor size were diagrammatically presented.

Figure 2

GSK923295 suppresses proliferation of liver cells results from mitotic arrest. (A) The proliferative activity of GSK923295-treated AML12 for 24, 48, and 96 h were assessed as described in Figure 1A. (B) After release of double thymidine block, cells were subjected to 5 μM GSK923295 for 8, 24, 48, and 96 h and under flow cytometry. (C) Quantitation of cells at different cell cycle phases after GSK923295 exposure (∗P < 0.05, n = 3). (D) Western blotting the AML12 cells treated as in (B) are stained for cyclin B, and pHH3, and for cleavage of full-length poly-(ADP-Ribose) polymerase (PARP; top band) to yield a PARP fragment (bottom band), a marker of apoptosis. (E) When AML12 cells were treated with 5 μmol/L GSK923295 for 8, 24, 48, and 96 h, the relative viable cells are plotted as a histogram based on comparison to cells treated with DMSO for 8 h. (F) Metaphase in AML12 cells 24 h after GSK923295 treatment. Cells were stained with tubulin (red), DAPI (blue), and anti-CENtromere Protein E (CENP-E) antibody (green). Bar graph (right) quantitating the average mitosis duration of AML12 cells with DMSO vs. GSK923295 treatment or control vs. CENP-E knockdown, respectively. Scale bar: 10 μm. (G and H) Western blotting was performed with anti-CENP-E and GAPDH antibodies after CENP-E siRNAs were transfected into AML12 cells. Staining for cyclin B1, and CENP-E was performed as well. (G) AML12 cells were transiently transfected with negative control siRNA or CENP-E siRNA for 24, 48, 72, 96, and 120 h, the relative viable cells were measured (^∗P < 0.01, n = 3).

Figure 3

Expression of CENtromere Protein E (CENP-E) during mouse liver regeneration after PH. (A) Schematic representation of murine liver anatomy (Upper). I-II, Ligatured LLL, ML. III-IV, Resected LLL, ML. Diagram of procedures used to investigate the influence of GSK923295 in PH mouse model (bottom). (B) Each respectively, total RNAs were isolated from regenerative livers of PH mice and livers of sham-operated mice (24, 48, and 96 h), and the amounts of CENP-E mRNAs were determined as described in Materials and Methods by qPCR. (C) Western blotting analysis of CENP-E from lysates of livers at 24, 48, and 96 h after PH. (D) CENP-E expression during liver regeneration in response to PH in GEO database. The ordinates represent relative mRNA amounts and the abscissas represents the periods of time (H) after PH. (E) Expression of CENP-E in liver regeneration after PH using microarray analysis. CL: Caudate lobe; LLL: Left lateral lobe; ML: Middle lobe; qPCR: Real-time quantitative PCR detecting system; RL: Right lobe.

Figure 4

CENtromere Protein E (CENP-E) inhibitor GSK923295 suppresses liver regeneration. (A) Liver recovery after PH was determined by the ratio of liver/body weight ratio at indicated time points. (B and C) Representative image of Ki-67 or proliferating cell nuclear antigen (PCNA) staining by immunohistochemistry at 24, 48, and 96 h. Scale bar: 100 μm. (D) Proliferative index of different groups, which was calculated by the mean of percentage of Ki-67 or PCNA positive particle in four random visual fields (200×) stained with immunohistochemistry (IHC). (E and F) Protein level expression of Ki-67 or PCNA in each group after 48 h. (G) The protein level of ALB was measured by immunofluorescence. Scale bar:100 μm. (H) Immunohistochemistry of pHH3 in the liver sections (upper), which were verified by western blotting the corresponding tissue sample (bottom). Scale bar: 100 μm. (I) In vehicle or GSK923295-treated group, mRNA expression of BubR1 in response to PH was determined by qPCR using the primer pairs listed in Supplementary Table 4, http://links.lww.com/CM9/A7. ^∗P < 0.01. (J) The relationship between CENP-E and BubR1 in vehicle or GSK923295-treated groups was calculated by the ratio of CENP-E/BubR1 mRNA levels. The log₁₀ of the ratio was filled in the form with color key.