MCM family in HCC: MCM6 indicates adverse tumor features and poor outcomes and promotes S/G2 cell cycle progression

The samples used in this study were categorized into three independent Cohorts. Cohort 1 was used to profile the mRNA expression of MCMs and included 105 samples (15 normal livers, 40 HBV cirrhotic livers and 50 HBV-related HCC). Cohort 2 consisted of 102 HBV-related HCC and matched pritumoral livers, and it was used to investigate the clinical implications of MCMs. The samples of Cohort 1 and 2 were immediately snap-frozen in liquid nitrogen after surgical excision and stored at − 80 °C. The tissue microarrays HLiv-HCC1805ur-02 and HLiv-HCC1805ur-03 (OUTDO BIOTECH CO., LTD, China) and part of formalin-fixed paraffin-embedded samples in our hospital were used for immunohistochemistry as Cohort 3 (n = 345). Cohort 3 contained various lesion types (60 normal livers, 110 HBV cirrhotic livers, and 175 HCC). And the major etiology of HCC in cohort 3 were HBV but the accurate proportion was unclear. The normal hepatic samples were from the patients who underwent operation for hemangioma. A diagnosis of cirrhosis was defined histologically as having fibrosis stage 5/6 [30]. The cirrhotic livers were from HCC-absence cirrhotic patients who underwent liver transplantation. The diagnosis of HCC was made by pathological examination of the resected tissues. Approval for these studies was obtained from the Ethics Committee of the First Hospital of Zhejiang University, and all subjects in this study provided written informed consent. All aspects of the study related to human participants were in accordance with the ethical standards of the national research committee as well as with the Helsinki declaration.

Total RNA was isolated from tissue samples preserved at − 80 °C using Trizol (Invitrogen, USA). Good quality RNA (as confirmed by the integrity of 28S and 18S rRNA on agarose gel and A260/A280 ratio) was reverse transcribed by the cDNA kit (vazyme, China) according to manufacturer’s protocol. Quantitative polymerase chain reaction (qPCR) assays were performed using the ABI 7500 fast system (Applied Biosystems, USA). The gene specific primers are shown in Additional file 1. Gene expression was measured in triplicate in the optimized PCR condition as described previously [31]: one cycle of denaturing at 95 °C for 3 min, followed by 40 cycles of amplification at 95 °C for 15 s and 60 °C for 30 s, and last cycle along the melting curve at 95 °C for 15 s, 60 °C for 15 s and 95 °C for 15 s. Relative expression of genes was normalized to GAPDH and reported as 2-△CT, and △CT = Ct(target gene)-Ct(GAPDH).

Four μm thick sections of samples were cut and mounted on poly L-lysine coated slides. Expression of the MCM2, MCM6 and MCM7 proteins were detected in paraffin-embedded samples in Cohort 3. As described previously [32], the sections were de-waxed and antigen retrieval was performed. After blanching of endogenous peroxidase, the sections were blocked and then incubated with the primary antibody at 4 °C overnight, and subsequently washed by PBS buffer at room temperature. On the next day, the slides were incubated with the secondary antibody (Biotech Inc., China) for 60 min at room temperature and the DAB detection was followed by Mayer’s haematoxylin nuclei counterstaining. Immunoreactivity score was assessed semi-quantitatively by determining the number of positive cells over the total number of liver cells: 0%, 5%, 10%, up to 100%, as reported [32, 33]. The assessment was performed by two independent pathologists in a double-blind manner. The antibodies and the dilution were detailed in the Additional file 2.

Human HCC cell line Huh 7 (TCHu182) was purchased from the Institute of Biochemistry and Cell Biology, Chinese Academy of S ciences (Shanghai, China) and maintained in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal bovine serum (Gibco, CA, USA) in a 37 °C incubator with 5% CO₂. The cells in logarithmic growth phase were harvested, seeded into 6-well plates (2 × 10⁵/well) and transfected with Si-MCM6 or SiRNA control (Additional file 3). After 48 h, the cells were collected for flow cytometry. This experiment was repeated three times. For the detailed methods, please refer to the previous publication [34].

The mRNA profiles of MCM2–7, MCM8, MCM9, MCM10 and RECQL4 were investigated in the Cohort 1 (normal livers, n = 15; cirrhotic livers, n = 40; and HCC, n = 50) using qPCR. As shown in Fig. 1, mRNA levels of MCM2–7, MCM8 were significantly up-regulated in HCC compared to normal or cirrhotic livers, and there was no significant difference in expression in normal versus cirrhotic livers. MCM10 was only significantly up-regulated in HCC relative to normal livers. However, MCM9 and RECQL4 remained unchanged throughout the process of hepatocarcinogenesis. The relative changes, medians and interquartile ranges (25th percentile to 75th percentile) of all MCMs in each sample type were reported in Additional file 4. The top three up-regulated MCMs were MCM2, MCM6 and MCM8, which were up-regulated 4.57-, 3.11- and 2.79-fold in HCC relative to noncancerous liver, respectively. These data indicate that MCM2–7, MCM8 and MCM10 mRNA levels increase in HCC, and are candidate drivers of hepatocarcinogenesis.

The aberrant expression of MCM2–7, MCM8 and MCM10 in mRNA levels and their clinical relevance in HCC patients were further studied in HCC and matched peritumoral livers in the Cohort 2 (n = 102). We observed the similar results as in Cohort 1, mRNA levels of MCM2–7, MCM8 and MCM10 all significantly up-regulated in HCC than peritumoral livers (Fig. 2a). First, we investigated the correlations between the expression levels of MCMs. As shown in Additional file 4, the mRNA expression levels of these MCMs were significantly positively correlated with each other (Kendall correlation test, P < 0.05). These results indicate that the MCMs may be transcriptionally regulated together. Indeed, MCM members MCM2–7 are known to work as a complex to regulate DNA replication. Next, their clinical implications were also analyzed. As shown in Additional file 5, TNM stage correlated with MCM2–4, MCM6, MCM7 and MCM10, AFP was associated with MCM2, MCM4, MCM6 and MCM7. We proceeded to investigate whether mRNA levels of MCMs could predict the prognosis of HCC patients. Kaplan-Meier plots showed that patients with high MCM2, MCM6 and MCM7 expression had poorer outcomes (P = 0.018, 0.002, and 0.005, respectively; Fig. 2b-d). There was no correlation between other MCM mRNA levels and patient outcome (data not shown). These results suggest that MCM2, MCM6 and MCM7 mRNA levels could be potential prognostic markers for human HCC.

To examine whether MCM2, MCM6 and MCM7 proteins are also exclusively overexpressed in HCC, we analyzed their expression patterns in Cohort 3 using immunohistochemistry. We detected them with antibodies in 60 normal livers, 110 cirrhotic livers and 175 HCC. The immunoreactivities of MCM2, MCM6 and MCM7 proteins were observed primarily in the hepatocellular cell nucleus and partly in the cytoplasm (Fig. 3a). Immunohistochemistry results were concordant with qPCR expression profiles. MCM2, MCM6 and MCM7 proteins were expressed at significantly higher levels in HCC compared to non-tumor specimens (P < 0.01; Fig. 3b-d), and the degree of their immunoreactivity gradually increased from normal and cirrhotic livers to HCC (Fig. 3b-d). These results suggest that increased MCM2, MCM6 and MCM7 proteins are associated with human HCC development. ROC curves were constructed to evaluate the area under the curve (AUC) for these potential diagnostic markers. The AUCs for MCM2, MCM6 and MCM7 proteins were 0.675, 0.896, and 0.771, respectively, and all the AUCs were significant compared with a Reference Line. MCM6 demonstrated optimal diagnostic performance, with an AUC significantly higher than that of MCM2 and MCM7 (Fig. 3e). These data indicate that MCM2, MCM6 and MCM7 proteins are potential diagnostic tissue markers for HCC, with MCM6 protein emerging as the primary candidate.

Next, we determined the association between MCM2, MCM6, and MCM7 protein levels and specific pathologic features and outcomes in 175 HCC patients. Correlation analysis showed that MCM6 protein levels were significantly associated with Ki67 expression and differentiation, MCM7 with tumor size and Ki67 (P < 0.05, Table 1). However, no significant association between MCM2 protein levels and tumor characteristics such as tumor stage, tumor size, etc. was observed. Kaplan-Meier analysis showed that patients with high MCM2, MCM6 and MCM7 protein levels had significantly poorer prognosis than those with low expression (P = 0.020, 0.001, and 0.001, respectively; Fig. 4a-c). In addition, overexpression of multiple MCMs was found to be a very strong prognostic indictor (P = 0.001; Fig. 4d), as demonstrated when MCM2, MCM6 and MCM7 were combined (three-marker panel). These results strongly suggest that the combined use of MCM2, MCM6 and MCM7 is a reliable prognostic indicator for HCC patients. Finally, further univariate and multivariate Cox regression analyses revealed that MCM6 protein is a significant and independent predictor for poor outcome in HCC patients (Table 2).

Because MCMs play a central role in the replication of DNA, we further studied whether the potential effects of MCM6 on the cell cycle in Huh 7 cells. Compared with control group (54.6 ± 5.1), the proportion of cells in S phase increased markedly (63.6 ± 6.0) while in G2 phase reduced dramatically in Si-MCM6 group (Fig. 5a and b). This suggests that cells in the Si-MCM6 group were arrested in the S phase and failed to enter the G2 phase. Mechanistically, CDK2, CDK4, CyclinA, CyclinB1, CyclinD1, and CyclinE were lower in Si-MCM6 treated cells (Fig. 5c), suggesting that inhibition of MCM6 can delay the cell cycle S/G2 progression through down-regulating the cell cycle checkpoint.