VK2 can inhibit the proliferation of cancer cells by inducing the cell-cycle arrest of cancer cells, in which inhibition of nuclear factor-κB (NF-κB) activity has a crucial role. NF-κB is a regulatory factor that can be simulated by cytokines to participate in the immune and inflammatory reaction (29). In addition, as a nuclear transcription factor, NF-κB is associated with cell growth by regulating the cyclin D1 gene (23). Cyclin D1 can contribute to the G₁-S transformation during the cell cycle by binding to CDK4 or CDK6 (30). Reportedly, NF-κB and cyclin D1 are involved in carcinogenesis. In cancer cells, VK2 can downregulate the expression of cyclin D1 by inhibiting the binding of NF-κB to the cyclin D1 promoter, which is followed by cell-cycle arrest in the G₁ phase (23). Regarding the suppression of the aberrant activity of NF-κB by VK2, certain studies reported results using HCC cells as follows. First, VK2 can inhibit the IκB kinase (IKK)/IκB/NF-κB pathway (23). Usually, NF-κB exists in the cytosol and is inactivated by the inhibitor of NF-κB (IκB). IκB can be phosphorylated by IKK in response to certain stimulatory factors prior to degrading IκB, which leads to the nuclear translocation of NF-κB and the activation of associated genes regulated by NF-κB (31). VK2 inhibited the function of IKK, which suppressed the phosphorylation of IkB and activity of NF-κB (23). Second, VK2 can inhibit the protein kinase Cα (PKCα)/NF-κB and PKCε/protein kinase D1 (PKD1)/NF-κB pathways (24). Protein kinase C (PKC) is a phospholipid-dependent kinase family, which is reportedly involved in the activation of NF-kB and cell growth. PKD1 is the effector of PKC and has been confirmed to be phosphorylated entirely by PKCε (32). Xia et al (24) suggested that VK2 inhibited NF-κB activation by inhibiting the catalytic action of activated PKCα, not by phosphorylation of PKCα, and that VK2 could also inhibit NF-κB activation by hindering the phosphorylation of PKCε and thus further suppressing PKD1 phosphorylation. Besides, PKD1 was found to promote IκB phosphorylation, from which the PKC/PKD1 pathway may be one of the upstream pathways through which VK2 restrains IKK activity.

In addition to PKC, protein kinase A (PKA), which can lead to cell-cycle arrest at the G₁ and G₂-M phase, is another type of kinase involved in the mechanism of VK2 against tumor cells. VK2 has been identified to stimulate the phosphorylation of PKA and activate activating protein 2 (AP-2), upstream transcription factor-1 (USF-1), and cAMP-response element binding protein (CREB) transcriptional factors to inhibit the proliferation of HCC cells (33). At present, the activities of these transcriptional factors are possibly the downstream pathways of VK2 activating PKA, because PKA is the only known regulator of AP-2, USF-1, and CREB. However, the details about the mechanism by which VK2 activates PKA continue to require clarification.

The promotion of VK2 on the transcription of p21 and p27 is another way to effectively promote cell-cycle arrest of cancer cells. Cell-cycle regulatory proteins p27 and p21 are the two members of the Cip/Kip family, working as CDK inhibitors, which have a negative role in the cell-cycle progression of the G₁-S transition. In HCC cells, VK2 induces G₁ arrest by activating the promoter of the p21 gene and increasing the expression of p21; however, the activity of p27 is not interfered with by VK2. It has been identified that the primary interaction sites of VK2 and the promoter region of the p21 gene are the sequences between −2,130 and −102 bp of p21 (34). By contrast, VK2 upregulates the expression of p27 to result in the G₀-G₁ arrest in leukemic cells (28). The difference between VK2 regulating p21 and p27 in HCC cells and leukemic cells is likely attributed to cancer cell types, and the role p21 and p27 have in other tumor cell lines inhibited by VK2 warrants further investigation.

Furthermore, suppression of the c-MYC expression is possibly associated with the cell-cycle arrest of cancer cells induced by VK2. c-MYC is highly expressed in a variety of human tumors, and the induction of cell-cycle arrest and differentiation of AML cells by all-trans retinoic acid is primarily attributed to c-MYC downregulation (35). Maniwa et al (35) identified VK2 suppression on c-MYC expression and deduced that 5 µΜ VK2 exposure inhibited c-MYC expression in HL-60 leukemia cells to ~80% that of control cells.

Finally, in HCC cells, it has been identified that VK2 could evidently downregulate the expression of hepatoma-derived growth factor (HDGF) by interfering in the initiation of HDGF gene transcription and subsequently inhibiting the proliferation of cancer cells. The region −1 to −150 bp in the promoter of the HDGF gene was detected to be the binding site of VK2 (36). HDGF is highly expressed in various cancer cells and can transmit a proliferative signal to cells in the early G₁ phase and then stimulate the increase of cyclin D (36). Hence, the inhibitory activity of VK2 on HDGF may lead to G₁ arrest and then suppress the cell growth. There are two problems that remain to be solved: i) Which pathways exert functions in the regulation of VK2 to HDGF; and ii) whether VK2 could suppress cancer cell growth by downregulating other growth factors (Fig. 1).

It has been certified in various cell lines that cell-cycle arrest is closely associated with cell differentiation. VK2 can exhibit differentiation-inducing results by the induction of the G₀-G₁ arrest in cancer cells. For instance, Miyazawa et al (28) reported that VK2 treatment induced monocytic differentiation in HL-60-Bcl-2 leukemia cells, with an evident increase in the proportion of cells in the G₀-G₁ phase. Maniwa et al (35) determined that VK2 reduced c-MYC expression, which subsequently contributed to cell growth arrest and cell differentiation. Hence, the molecular activities regulated by VK2, including the induction of cell-cycle arrest in cancer cells, are likely to be crucial to the mechanism of the VK2-dependent stimulation of cancer cell differentiation. Notably, cell-cycle arrest is only one of the processes that can lead to the differentiation of cancer cells.

VK2 suppresses connexin 43 (Cx43) expression and enhances Cx32 activity, which is another mechanism of cancer cell differentiation. Cx proteins are comprised of gap junction channel structures that mediate intercellular communication to maintain tissue homeostasis. Studies indicate that Cx genes have tumor-suppressing effects and the distribution of Cx is tissue specific (37). Cx32 is mainly expressed in hepatocytes. The risk of tumor development can be raised when knocking out the Cx32 gene in mice. However, Cx43 is not found in regular hepatocytes, and its expression is highly upregulated in HCC cells (38). Regarding cell differentiation induced by VK2, Kaneda et al (37) reported that VK2 exposure could drive Huh7 HCC cells to adopt a normal liver cell phenotype by upregulating Cx32 indirectly via downregulation of Cx43 at the transcriptional level, and the increase of gap junction intercellular communication enhanced by VK2 is potentially due to Cx32 upregulation.

VK2-induced differentiation is partly associated with steroid and xenobiotic receptor (SXR). VK2 can work as the ligand of SXR, a nuclear receptor, and their binding largely exerts osteoprotective function. Sada et al (13) investigated the effects of VK2 on normal hematopoietic progenitor cells and revealed that VK2 promotes the differentiation of myeloid progenitors partly owing to its binding to SXR and the upregulation of the transcriptional factors CCAAT/enhancer-binding protein α (C/EBPα) and PU.1, which are crucial for myeloid development. In addition, VK2 binding to SXR may subsequently improve the expression of C/EBPα and PU.1, which may aid elucidation of the therapeutic effect of VK2 on patients with MDS (Fig. 1) (13).

The mitochondrial pathway is a crucial process through which VK2 exerts its pro-apoptotic function. Following treatment with VK2, the mitochondrial membrane potential is depolarized and cytochrome c is released from the mitochondria into the cytosol to form the apoptosome, driving activation of caspase-9, which ultimately leads to the activation of caspase-3 and initiation of cell apoptosis (39–41). It is anticipated that the selective binding of VK2 and VK2-2,3 epoxide (VK2-O) to the mitochondrial protein Bcl-2 antagonist killer 1 (Bak) is an essential molecular mechanism of VK2 dissipating the mitochondrial membrane potential of leukemia cells. In addition, the direct binding of VK2 to Bak specifically resulted in a post-translational modification of Bak (39). It has been revealed that VK2 and VK2-O metabolized from VK2 could non-covalently bind to Bak through the Arg-169 and Trp-170 residues, and that VK2-O further covalently attached to the Cys-166 residue of Bak in HL-60 leukemia cells (39). In addition, VK2 treatment increased the intracellular level of the reactive oxygen species (ROS) (39,40). Evidence indicates that VK2-induced ROS generation occurred prior to the induction of apoptosis. Assumedly, ROS contributes by converting VK2 to VK2-O (39). Besides Bak, other Bal-2 family members also have crucial roles in the mitochondrial apoptosis pathway induced by VK2. VK2 decreases Bcl-2 expression and increases the expression of Bcl-2-associated X protein (Bax) in a newly established MDS cell line, which was associated with apoptosis (42). Furthermore, Tsujioka et al (41) determined that the ratio of B-cell lymphoma (Bcl)-extra large and Bcl-extra small expression decreased when exposing myeloma cells to 10 µΜ VK2; however, the expression of Bcl-2 and Bax remained unaffected. VK2 can thus alter the expression of the Bcl-2 family, tending to a pro-apoptotic balance and activating the mitochondrial apoptosis pathway of cancer cells. It is hypothesized that the release of cytochrome c from the mitochondria partly results from the acidic phospholipid cardiolipin (CL), abundant in the outer membrane of the mitochondria, being peroxidated by ROS (40), which remains to be confirmed.

Furthermore, the mitogen-activated protein kinase (MAPK) pathway is essential for the VK2-mediated mitochondrial apoptosis pathway (25,26,41,43). In the MAPK superfamily, p38 MAPK and c-Jun N-terminal kinase (JNK) pathways respond to stress and contribute to inflammation or even apoptosis, whereas the extracellular signal-related kinase (ERK) pathway reacts to growth factors or other external mitogenic signals by stimulating cell proliferation and resisting apoptotic signals. VK2 activates p38 MAPK to its phosphorylated form and subsequently results in apoptosis of the neoplastic cells (41,44). Tsujioka et al (41) identified this in myeloma cells, following which they detected that the mitochondrial membrane potential was depolarized and caspase-9 was activated, indicating that the phosphorylation of p38 MAPK stimulated by VK2 possibly induces apoptosis by initiating the mitochondrial pathway. In addition, these authors found that ROS generation stimulated by VK2 is associated with this apoptosis (41). Sibayama-Imazu et al (25) investigated the apoptosis induction of PA-1 ovarian cancer cells by VK2 and concluded that the increase of synthesis and accumulation in the mitochondria of TR3, also known as Nur77 and neuron growth factor inducible factor I-B, highly expressed in various tumor cell lines, is possibly associated with the mitochondrial apoptosis pathway induced by VK2. In addition, Sibayama-Imazu et al (25) deduced that VK2 may activate JNK to phosphorylate TR3 and increase TR3 levels in the mitochondria. VK2 is reported to inhibit ERK phosphorylation by suppressing Ras activation and subsequently suppressing the activation of MAPK kinase (MEK), which causes the apoptosis of HCC cells (43). Conversely, research investigating the inhibitory role of VK2 on pancreatic cancer demonstrated that apoptosis of cancer cells treated with VK2 was primarily associated with an increase in phosphorylated ERK (26). This contradiction may be due to differences in the type of cancer cells, and requires further investigation. Another unsolved problem is that not all pancreatic cancer cell lines are sensitive to VK2 treatment. Hence, the effects of VK2 on pancreatic cancer cells warrant further investigation.

In addition to the mitochondrial pathway, described as an intrinsic pathway, the extrinsic apoptosis pathway also participates in the mechanism of VK2-dependent induction of cell death. Evidence indicates that VK2 can induce apoptosis in cancer cells by activating p53 and initiating the extrinsic apoptosis pathway (45). Notably, p53 is a multi-faceted tumor-repressor gene capable of inducing cell-cycle arrest, cell differentiation or apoptosis in reaction to oncogenic stress (46). The extrinsic apoptosis pathway is depedent on the death receptors binding to ligands to form the death-inducing signaling complex, which then contributes to caspase-8 activation and further activates caspase-3 (47). A study investigating the antitumor effects of VK2 in Smmc-7721 HCC cells established that VK2 stimulated the extrinsic apoptosis pathway by increasing p53 phosphorylation and then activating caspase-8 (Fig. 2) (45).

Another previous study (6) reported that a large dose of VK2 could induce apoptosis of Hep40 HCC cells by increasing the expression of c-JUN and c-MYC; however, it did not identify the detailed apoptosis pathway.

In different cancer cell lines, VK2 can inhibit the growth of cancer cells by evoking autophagy. Autophagy is a mechanism that aids cell survival in response to stresses, such as nutrition starvation. It has been reported that autophagy is essential for the inhibition of certain antitumor agents in cancer (27,48). Owing to a lack of clarity regarding the molecular mechanism that lead to autophagy, whether autophagy protects or promotes cell death is debatable. Yokoyama et al (48) reported that VK2 could stimulate apoptosis and autophagy in leukemic cells simultaneously, but autophagy is more dominant when Bcl-2 is highly expressed, restraining apoptosis. Hence, Yokoyama et al (48) suggested that autophagy may act as an alternative inducer of apoptosis. Another study that exposed cholangiocellular carcinoma cell lines to VK2 identified the effects of inducing apoptosis and cell-cycle arrest to be inconspicuous; however, the autophagy induction exerted a maximal effect in VK2 inhibiting cholangiocellular carcinoma cells (27). Among these cells lines, the TFK-1 cell line exhibited a higher expression of Bcl-2. On the basis of these two studies, it can be inferred that the molecular mechanism of apoptosis may lead to autophagy, which could be regarded as the cell state prior to apoptosis. In addition, at least in cholangiocellular carcinoma cells, Bcl-2 may be involved in the determination of whether cells eventually undergo apoptosis or autophagy. When Bcl-2 is highly expressed, apoptosis induction may change to autophagy. In short, inducing autophagy is a significant part in the mechanism of VK2-dependent inhibition of cancer cells, although further details remain to be investigated.

On the basis of the study conducted by Ide et al (49), VK2 can restrain the invasion of tumor cells mainly by downregulating the expression of matrix metalloproteinases (MMPs) at the transcriptional level. MMPs are a group of proteinases that degrade extracellular matrix proteins, which are reportedly linked to tumor invasion and metastasis; AP-1 and NF-κB are the common transcription factors of the promoter regions of MMPs. AP-1 is the transcription complex comprising proto-oncogene proteins c-FOS and c-JUN, and activity of the MAPK pathway leads to the phosphorylation of specific threonine and tyrosine residues of c-FOS and c-JUN (50). Ide et al (49) revealed that VK2 could inhibit MMP expression by mediating NF-κB inhibition and downregulating AP-1 by suppressing the ERK and JNK pathways, which restrained the invasion of HCC cells.

In several cases, the combination of VK2 with other chemotherapy agents can produce stronger effects than the use of either alone. The mechanisms associated with the synergistic effects can also be classified into induction of the cell-cycle arrest, differentiation, and apoptosis in cancer cells; however, the specific details may slightly vary from the mechanism of action of VK2 alone. VK2 pretreatment can restrain NF-kB activation and increase cyclin D1 expression caused by 5-fluorouracil (5-FU), promoting cell-cycle arrest and improving the 5-FU-dependent inhibition of HCC cell growth (51). In addition, VK2 can enhance the ability of cotylenin A to induce cell differentiation in HL-60 leukemia cells, as VK2 can enhance the increase of cyclin G2 expression stimulated by cotylenin A (35). Cyclin G2 plays a beneficial role in promoting and maintaining cell-cycle arrest, but treatment with VK2 alone induces a non-significant cyclin G2 expression in cancer cells (35). When treating with VK2 and sorafenib together in HCC cells, VK2 can counteract the decrease of p21 level and improve the inhibition of ERK phosphorylation induced by sorafenib, which promotes cell-cycle arrest and apoptosis (21).

VK2 can augment the efficacy of retinoids in cancer cells (9,43). Retinoids suppress cell growth through nuclear receptor retinoid X receptor (RXR), which can bind to the RXR-responsive element located in the promoter regions of retinoid-target genes. In HCC cells, the Ras/ERK pathway is aberrantly activated, which causes accumulation of phosphorylated RXRα and weakens the effects of retinoids (43). However, treatment with VK2 plus retinoid can apparently decrease the level of phosphorylated ERK and phosphorylated RXRα by cooperatively inhibiting the Ras/ERK pathway, which contributes to retinoid binding of functional RXRα and to the apoptosis of HCC cells (43). This mechanism is likely to be the reason why the combination of acyclic retinoid and VK2 could decrease the recurrence rate of HCC in clincal trials. Therefore, VK2 alone and the combination of VK2 with other antitumor agents requires further investigation.

Vitamin D3 is another micronutrient capable of restricting the growth of cancer cells by inhibiting proliferation and stimulating differentiation (52). However, an adverse effect of vitamin D3 treatment is hypercalcemia, which can lead to vascular calcification. VK2 regulates the calcium deposition between bone tissue and other tissues, and inhibits the formation of vascular calcified foci. The combination of vitamin D3 with VK2 on cancer cells can synergistically improve the induction of cellular differentiation and also significantly reduces the risk of hypercalcemia and vascular calcification (52,53).