Compound C was even used to rescue the effects of AMPK activa-
tors, although the effects of Compound C alone were not categori-
cally examined in some studies. For example, the compound
AICAR and the biguanide metformin activate AMPK through
disparate AMPK-dependent and AMPK-independent mechanisms
[32]. Strangely, while Compound C was used to rescue the anti-
proliferative effects of AICAR and metformin [33, 34], we and
others have found that each of the three reagents, AICAR, metfor-
min, and Compound C, inhibits cancer cell proliferation in vitro
and tumor growth in vivo through AMPK-independent mechan-
isms [32, 35–37, 38]. In the study by Tang et al. [34], 5 and
10 mM Compound C was used for 72 h to rescue the in vitro
antiproliferative effects of the non-specific AMPK activator AICAR
in mouse embryonic fibroblasts (MEFs). In contrast to this study,
we found that in 72 h, Compound C alone significantly inhibited
viability of multiple human glioma cell lines at all concentrations
between 1 and 10μM[ 35]. Even at 24 h, 10μM, Compound C
caused significant cell cycle arrest at G2M in these glioma cells. All
these effects were independent of AMPK. Ironically, we found that
AMPK depletion by silencing RNA increased cell death by Com-
pound C indicating not only a promiscuous mode of action of this
reagent but also that AMPK protects cells from insults by xenobio-
tics such as Compound C. We also tested the effect of Compound
C in WT and AMPK-null MEFs in the presence and absence of the
AMPK activators AICAR and metformin. While metformin
(5 mM), AICAR (0.5 mM), and Compound C (5 μM) each
reduced viability of both WT- and AMPK-null MEFs following
72 h of treatment, Compound C did not have any protective effects
against AICAR or metformin. In fact, Compound C alone killed
significantly more MEFs (about 90%) regardless of the presence or
absence of AICAR or metformin (our unpublished observations).
AMPK has been shown to play a role in the migration of normal
as well as cancer cells. Because we and others have found many
AMPK-independent effects of compound C, we also tested the
AMPK dependency of this phenotype. Again, Compound C
strongly inhibited migration of cancer cells independent of AMPK
and at doses (1 and 2.5μM) where it showed little effect on inhibi-
tion of AMPK activity in these cells [35]. In this study, we also
reported that Compound C significantly inhibits Akt phosphoryla-
tion (at both T308 and S473) and mTOR activity in glioma cells
independent of AMPK and induces apoptosis and destructive
autophagy, all independent of AMPK. One surprising observation
was that while Compound C effectively inhibited Akt phosphoryla-
tion in both glioma cells and MEFs, mTOR inhibition was observed
specifically in glioma cells but not in MEF = mouse embryo fibro-
blast (unpublished). The reasons for this discrepancy are unknown
to us. Erk1/2 phosphorylation remained unaffected by Compound
C (unpublished). We observed that blocking apoptosis was
Compound C: Its Use and Misuse 199