How to eliminate MYCN-positive hepatic cancer stem cells to prevent the recurrence?

I read with great interest the work of Qin et al. (1), in which the authors reveal that acyclic retinoid (ACR) selectively eliminates cancer stem cells (CSCs) of hepatocellular carcinoma (HCC), which highly express MYCN with enhanced Wnt/β-catenin signal pathway. There exist several molecules that are highly expressed in HCC-CSCs, such as CD90, CD133, EpCAM, GPC3, etc. (2). Among them, MYCN is not only a promising biomarker of HCC-CSCs but also a novel therapeutic target of ACR in preventing de novo HCC recurrence (1).

Although NMYC gene amplification is frequently recognized in neuroendocrine tumors characterized by small-cell lung cancer and neuroblastoma (3), the restricted expression of MYCN in HCC-CSCs is expected to reflect that only 2.7% of HCC patients showed MYCN mRNA up-regulation according to The Cancer Genome Atlas database (1). Sp1 is likely to promote the transcription of MYCN gene in cooperation with E2F, while ACR-induced nuclear translocation of transglutaminase 2 is considered to inhibit MYCN expression with the inactivation of Sp1 (1, 4). However, there is also a possibility of enhanced stability of MYCN protein by the phosphorylation of GSK3β. NCYM, which was initially regarded to be a long noncoding RNA derived from a cis-antisense gene of MYCN, promotes the phosphorylation of GSK3β at serine 9 (5).

Given the susceptibility of EpCAM-negative HCC to ACR, HCC-CSCs negative for EpCAM and positive for MYCN are highly likely to exist. Indeed, there is an overlap between MYCN-positive and EpCAM-positive HCC-CSCs in vitro; about one-half of MYCN-positive cells express EpCAM, whereas ∼20–50% of EpCAM-positive cells express MYCN (1). This fact strongly supports the idea that the substantial entity of CSCs is obscure, depending on which molecule is used for the definition (6). Furthermore, the proportion of EpCAM-negative JHH7 cells increased after ACR treatment. ACR-induced selective elimination of HCC-CSCs with high EpCAM expression level resulted in the relatively increased ratio of EpCAM-negative cells (1). Remarkably, the differentiation with retinoic acid and all-trans retinoic acid (also known as tretinoin) is therapeutically effective in neuroblastoma and acute promyelocytic leukemia, respectively (7, 8). That is why it is also feasible that ACR-induced differentiation of HCC-CSCs causes the negative conversion of EpCAM expression.

I do agree with the authors’ description that the treatment with z-IETD-FMK (caspase 8-inhibitor) significantly attenuated the degree of ACR-induced suppression of HCC proliferation, but the simultaneous treatment with ACR and z-IETD-FMK led to the recovery only up to 80% compared with control (1). That is why the effect of caspase-independent cell death may occur upon the treatment with ACR or siRNA-mediated depletion of MYCN expression. NMYC induces up-regulation of PTEN, a critical negative regulator of the PI3K/Akt/mTOR pathway, which promotes the Warburg effect. Furthermore, lipid biosynthesis and glutaminolysis can be significantly activated in MYCN-driven HCC-CSCs (9, 10). ATF4 is a regulator that coordinates with MYCN to directly up-regulate ASCT2 to maintain the sufficient amount of glutamine essential for TCA cycle (10). Therefore, the disruption of MYCN-induced metabolic reprogramming can be responsible for the specific toxicity to HCC-CSCs.