D. Kanabar, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127372
Scheme 2. Reagents: (a) Biphenyl-4-sulfonyl
chloride or 4-iodobenzenesulfonyl chloride, DMAP,
TEA, CH2Cl2.
with I79 (Fig. 7) as well as an optimized π-π interaction with W46
suggests a change in binding mode causing each aryl substitution to no
longer optimally interact with gankyrin. Future derivatives will utilize
these findings to help guide our design.
We next turned our attention to substituting the phenyl ring at-
tached to the triazole with various heteroaromatic systems. Compounds
14a-d demonstrated modest but not significant improvements in gan-
kyrin binding as compared to cjoc42. Substituting the phenyl ring of
cjoc42 with the 2- and 4-pyridine rings (14a and 14c, respectively)
exhibited similar gankyrin binding affinities, whereas 3-pyridine (14b)
displayed a slightly better binding affinity than cjoc42, 14a, and 14c.
Substituting the phenyl ring with a thiazole (14d) ring also resulted in a
modest but not significant improvement in gankyrin binding affinity as
compared to cjoc42. Since compounds 14a-d did not show a significant
improvement in gankyrin binding affinity as compared to cjoc42, they
were not further evaluated in cell-based assays.
The 5 most potent binders of gankyrin (11a, 11e, 11f, 12c, and 12g)
had their anti-proliferative activity assessed in four gankyrin-over-
liferation in these four cell lines when evaluated up to 100 µM. How-
ever, all 5 cjoc42 derivatives (11a, 11e, 11f, 12c, and 12g) demon-
strated an ability to inhibit cell proliferation at 100 µM, resulting in IC50
values less than 100 µM. Specifically, 11a exhibited the greatest anti-
proliferative effect in MDA-MB-231 cells with an IC50 value of 36.6 µM.
Compound 11e demonstrated the greatest ability to inhibit prolifera-
tion in both HepG2 and Hep3B cells with IC50 values of 42.6 µM and
42.7 µM, respectively. Compound 11f then displayed the best ability to
inhibit A549 cell proliferation resulting in an IC50 value of 74.6 µM.
While compounds 12c and 12 g also exhibited an ability to inhibit
Hep3B and HepG2 cells, neither proved more potent than compound
11e. Interestingly, compounds 11a, 11e, 11f, 12c, and 12 g did de-
monstrate similar anti-proliferative activity to cjoc42 in certain cell
lines despite their enhanced gankyrin binding ability. Specifically, each
of these compounds exhibited an IC50 value greater than 100 µM in at
least one of the cell lines they were evaluated in (Table 1). Although the
reason for this is currently unknown, it could be due to poor membrane
permeability and/or a lack of metabolic stability (i.e., sulfonate ester)
and will be the focus of future studies.
As we previously reported, substituting the p-tosyl group of cjoc42
with a biphenyl-4-sulfonate ester group or 4-iodobenzene sulfonate
ester group resulted in a ~2-fold improvement in gankyrin binding
(Fig. 1, AFM-1-2 and JA-1-38, respectively). Armed with this knowl-
edge, we sought to combine these features with our most potent aryl-
triazole substitutions. Therefore, we synthesized cjoc42 derivatives
15a-d to combine the best features of the R1 and R2 substitutions to
cjoc42 (Scheme 2). The synthesis relied on constructing key triazole
underwent
a nucleophilic substitution with biphenyl-4-sulfonyl
chloride or 4-iodobenzenesulfonyl chloride to afford compounds 15a-d
in good yield (57–71%).
Cjoc42 derivatives 15a-d presented only 2 derivatives, 15a and
15d, which exhibited improved gankyrin binding affinity as compared
to cjoc42 (Fig. 6). However, all four derivatives demonstrated de-
creased ΔTm values as compared to AFM-1-2, JA-1-38, 11f, and 12e
In summary, a novel series of cjoc42 derivatives with a modified Ar
group (Fig. 1) were designed, synthesized, and evaluated for their
ability to bind gankyrin and inhibit liver, lung and breast cancer cell
proliferation. Modifications to the aryl triazole moiety resulted in 3 new
derivatives of cjoc42 (11f, 12g, and 13e) with superior anti-cancer
activity in Hep3B, HepG2, A549, and MDA-MB-231 as compared to
cjoc42. In terms of gankyrin binding, an extensive SAR (Fig. 8) revealed
that electron donating substituents were favored at the R = 4-position
(11e), while at the R = 3-position the electron withdrawing CN group
was favored (12g), and at the R = 2-position the electron withdrawing
CF3 group was preferred (13e). These second generation cjoc42 deri-
vatives also exhibited enhanced gankyrin binding compared to AFM-1-
2 (Fig. 1), our most potent first generation cjoc42 derivative.27 How-
ever, AFM-1-2 demonstrated similar or superior anti-proliferative ac-
tivity against Hep3B and HepG2 cells than our most potent gankyrin-
binding second generation cjoc42 derivatives 11a, 11e, 11f, 12c, and
12g. In addition, second generation cjoc42 derivatives 11f, 12g, and
13e all exhibited improved anti-proliferative activity against Hep3B,
HepG2, A549 and MDA-MB-231 cells compared to cjoc42. These find-
ings mark a significant advance over cjoc42 in terms of both gankyrin
binding as well as anti-cancer activity in liver, lung, and breast cancers.
(ΔTm
< 1.0). Although disappointing, this loss in binding affinity
Declaration of Competing Interest
Fig. 6. Evaluation of cjoc42 derivatives 15a-15d for relative gankyrin binding.
Compounds 15a-15d (300 µM) and gankyrin (20 µM) were incubated together
in a thermal shift assay and their respective ΔTm values were determined.
DMSO (5%) served as a negative control (vehicle), and cjoc42 (300 µM) served
as a positive control. All compounds were tested in triplicate.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
4