Background
Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and the most common form of liver cancer, being responsible for 80% of primary malignant tumors in adults. HCC causes more than 600,000 deaths annually worldwide [
1] and its endemic prevalence in Asia, including South Korea, makes HCC one of the top causes of death in this region. HCC is a type of tumor that is highly resistant to available chemotherapeutic agents, administered either alone or in combination [
2]. Thus, in many cases, no effective therapy can be offered to patients with HCC. Therefore, it is of vital importance to identify important prognostic factors and novel molecular targets of HCC to develop targeted therapies, ultimately advancing therapeutic strategies of HCC in general.
Current evidence indicates that the precancerous liver and the early stages in HCC development are characterized by certain common traits governed by both genetic and epigenetic mechanisms [
3,
4]. These include the alteration of numerous signaling pathways leading to autonomous and deregulated cell proliferation and resistance to cell death [
4‐
7]. Therefore, it is important to better understand the roles of deregulated genes in hepatocellular carcinogenesis. Derangements in various methylation processes in liver diseases have been identified [
8,
9], including increased nicotinamide methylation in cirrhotic patients [
10]. Nicotinamide N-methyltransferase (NNMT) catalyzes the N-methylation of nicotinamide, pyridines, and other structural analogues [
11]. It is involved in the biotransformation of many drugs and xenobiotic compounds. Although several studies indicated differential expression of NNMT in HCC specimens [
12‐
15], the clincopathologic relevance of NNMT expression has not been fully investigated.
The aim of the present investigation was to examine whether NNMT expression could be used to predict the clinical course of HCC. Using a real-time RT-PCR analysis of NNMT gene expression, we found significant correlation between NNMT mRNA levels and poor prognosis of HCC. Thus, potential biological changes related to NNMT gene expression require further study, as they may have implications in predicting clinical outcome and choosing treatment modalities, due to the central role of NNMT in biotransformation and detoxification.
Discussion
The metabolism of drugs, toxic chemicals, and hormones is important in the fields of pharmacology and endocrinology given its implication in many pathophysiological processes, such as cancer and resistance to chemotherapy [
21]. One of the key enzymes involved in biotransformation and drug metabolism is NNMT, which catalyzes the N-methylation of nicotinamide, pyrimidines, and other structural analogues [
22,
23]. NNMT is predominantly expressed in the liver, where its activity varies with a bimodal frequency distribution, thus raising the possibility that a genetic polymorphism might play a role in regulating the enzyme activity [
23]. Lower expression is observed in other organs such as the kidney, lungs, placenta, heart, and brain. Although several studies indicated differential expression of NNMT in HCC [
12‐
15], the role of NNMT in the molecular pathogenesis of HCC has yet to be elucidated.
This study focused on NNMT as a potential molecular marker responsible for determining clinicopathologic features and the prognosis of HCC. Utilizing a large number of HCC specimens, the quantitative real-time PCR assay showed that the expression of NNMT is markedly reduced in HCCs compared to non-cancerous surrounding tissues, consistent with other studies [
12‐
15]. Stratification of HCC specimens based on NNMT gene expression levels showed that NNMT expression was significantly correlated with tumor stage (P = 0.010). More importantly, the log-rank test showed that patients who expressed higher NNMT mRNA levels tended to have a shorter OS time (P = 0.053) and a significantly shorter DFS time (P = 0.016). Both NNMT expression (P = 0.0096) and high tumor stage (P = 0.0017) were found to be significant prognostic factors for DFS in a multivariate analysis. It is not clear why NNMT expression level was a significant prognostic factor for DFS but not for OS. We believe that the limited follow-up time was not the main cause of lack of correlation between NNMT and OS because the events (death or relapse) were rare after the median follow-up time of 50 months in our cohort. Our analysis of NNMT expression in correlation with the clinicopathologic features and prognosis of HCC yielded the novel finding that NNMT mRNA levels could be used as a prognostic factor for DFS.
The mechanism for reduced expression of NNMT and its relation to HCC progression is not clear. Several metallothionein genes involved in detoxification and drug metabolism are downregulated in HCC especially in tumors with high Edmonson grades, reflecting de-differentiation of cancer cells [
12]. Thus, it is possible that the liver specific function of NNMT is lost during the progression of HCC. On the other hand, a recent in vitro study found that NNMT was necessary for cancer cell migration in bladder cancer cell lines [
24], pointing to a possible involvement in tumor invasion. In 120 HCCs observed in this study, NNMT mRNA was higher in recurrent tumors than in non-recurrent tumors especially in stage III & IV tumors, although the differences were not statistically significant. Thus, there's a possibility that increased NNMT expression is related to cell mobility and tumor invasiveness in high stage HCC. Interestingly, the NNMT expression level was decreased in stage II tumors compared to stage I tumors, while stage III & IV tumors showed a similar NNMT level as stage I tumors. This could be due to tumor de-differentiation preceding tumor invasion. However, we cannot rule out other regulatory mechanisms independent of tumor de-differentiation and invasion.
In tumors, abnormal expression of NNMT has been reported in glioblastoma [
25], stomach cancer [
26,
27], papillary thyroid cancer [
28,
29], colon cancer [
30], and renal carcinoma [
31,
32]. NNMT was identified as a novel serum marker for human colorectal cancers although this protein is not thought to be secreted [
30]. Interestingly, the upregulation of NNMT was found to be inversely correlated with tumor size in renal clear cell carcinoma, suggesting that the enzyme may be significant in an initial phase of malignant conversion [
32]. Increased expression of NNMT in non-tumor cells was reported in a few situations: the cerebellum of patients with Parkinson's disease [
33,
34], human hepatoma cells (Huh7) with expression of the hepatitis C core protein [
35], and the liver of mice transplanted with tumors [
36,
37]. In these situations, the mechanism for deregulated NNMT expression remains unclear.
Recently, NNMT promoter was cloned and studied in papillary thyroid cancer cell lines, where it was shown to be activated by hepatocyte nuclear factor-1β [
29]. Subsequently, it was found that the NNMT promoter region also contains the consensus sequences for signal transducers and activators of transcription (STAT) binding elements and nuclear factor-interleukin (IL) 6 binding elements [
38]. Accordingly, hepatoma cell line (Hep-G2), which expressed low levels of NNMT, increased NNMT expression several fold upon stimulation by IL-6. The stimulation by IL-6 was largely abolished with the expression of dominant-negative STAT3 [
38]. Activation of STAT3 alone caused a four-fold higher induction of NNMT promoter activity in the transformed Hep-G2 cells. Thus, NNMT expression could be regulated by IL-6 and STAT3 in a subclass of HCC. The expression of NNMT analyzed in relation to the expression of related regulatory molecules could improve the predictive power on HCC prognosis.
To our knowledge, this is the first report of NNMT as a prognostic factor of DFS in HCC. The findings herein indicate that NNMT is an attractive target for therapeutic regulation because it is involved in drug metabolism and could alter the efficacy of standard chemotherapeutic drugs. Additional research in larger populations of HCC patients may ultimately determine the ability of NNMT in accurate diagnosis and sub-classification of HCC.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
JK analyzed the RT-PCR data and wrote the manuscript. SH and SK helped write the paper. EL and YY carried out the RT-PCR experiment. JR and ID collected the samples and patients' clinical data. JJ analyzed patients' clinical data and helped write the final version. DK conceived of the study and wrote the manuscript. All authors read and approved the final manuscript.