Over the past few years, compelling evidence has emerged in support of the hierarchic
cancer stem cell (CSC) model for many solid tumors, including hepatocellular cancers.
We have recently identified a CSC gene signature that exhibited a pernicious interaction
with a variety of key oncogenic and common stemness pathways. Among these common pathways,
activation of NF-kB signaling was consistently observed in human liver CSCs. Based
on these data, we hypothesized that NF-kB may be a specific therapeutic target against
CSC. In this study, we evaluated whether the inhibition of NF-kB would selectively
deplete liver CSC.
Inhibition of NF-kB signaling was performed using (i) curcumin, an effective IKK inhibitor,
(ii) siRNA against p65 and (iii) the specific inhibitory peptide SN50. Anti-proliferative
and pro-apoptotic capacity of the compounds was evaluated in different liver cancer
cell lines. The effect on CSC was assessed by the Side Population (SP) approach, and
expression levels of selected targets determined by RT-qPCR, gene expression microarray,
EMSA, and Western blotting.
Curcumin treatment caused anti-proliferative and pro-apoptotic responses in liver
cancer cells which were directly related to the extent of NF-kB inhibition. In curcumin-sensitive
cell lines, the treatment led to a selective depletion of CSCs as determined by a
significant reduction in the SP population, sphere formation and tumorigenicity as
well as down-regulation of the CSC markers CD133, EpCAM, NANOG and c-Kit. In contrast,
curcumin-resistant cells, including hepatoblastoma WRL68 and a freshly isolated human
HCC cell line, exhibited a paradoxical increase in proliferation and activation of
the CSC markers. Specific inhibition of NF-kB signaling by SN50 and siRNA led to a
general suppression of cell growth accompanied by a drastic reduction in the size
of SP fraction. Mechanistically, CSC-depleting activity of curcumin was exerted by
NF-kB mediated HDAC inhibition leading to down-regulation of c-MYC and other key oncogenic
targets. Co-administration of a class I and II HDAC inhibitor sensitized the curcumin-resistant
cells to curcumin treatment.
These data demonstrate that NF-kB inhibtion can specifically target CSC populations.
Future investigations will determine the potential of combined inhibition of NF-kB
signaling and HDAC for CSC-directed HCC therapy.
