L. N. Solano et al. / Bioorg. Med. Chem. Lett. 25 (2015) 5777–5780
5779
Scheme 3. Synthesis of allylic quaternary ammonium salts of N-methylmorpho-
line: (a) allylic bromide, Et2O, rt, 1 h.
Figure 3. In vivo tumor growth inhibition of 5a based on isolated tumor mass in
MIAPaCa-2 xenograft model (n = 6, *P 6 0.05).
treated animals survived and 5a was found to be well tolerated
based on observed normal body weights compared to the control
group. None of the animals exhibited any signs of stress during
the treatment.
We next carried out an in vivo tumor growth inhibition study
using flank based tumor xenografts in female athymic nude mice.
For this study, we utilized human pancreatic cancer cell line
MIAPaCa-2. We chose this tumor model due to the lack of effective
therapeutics for pancreatic cancer treatment. MIAPaCa-2 cells
(5 ꢁ 106) in 1:1 PBS–matrigel were inoculated onto the right flank
of mice. When the tumor volume reached ꢀ250 mm3, mice were
assigned into four groups (n = 6). Mice in group 1 were adminis-
tered with compound 5a (40 mg/Kg, ip, bid) whereas group 2 mice
were treated with gemcitabine (100 mg/Kg, ip, twice weekly).
Group 3 mice were given a combination of 5a (40 mg/Kg, ip, bid)
and gemcitabine (100 mg/Kg, ip, twice weekly) and group 4 was
designated as a control group (saline administration). At the end
of the study, mice were euthanized and tumors were resected
and weighed. Based on these studies, 42%, 47%, and 57% tumor
growth inhibition was observed in groups 1, 2 and 3, respectively,
compared to the control group (Fig. 3).
In conclusion, we have developed several BH bromide based
highly water soluble quaternary ammonium curcuminoids as
potential anti-cancer agents. Many of these novel derivatives have
been found to exhibit several fold higher activity than native cur-
cumin against three solid tumor cell lines. We have evaluated
the systemic toxicity of one of the derivatives (5a) in healthy
CD-1 mice. Compound 5a has been found to be well tolerated
based on normal body weight changes compared to control group.
Furthermore, we have carried out tumor growth inhibition studies
with 5a in a pancreatic cancer xenograft model. 5a has good tumor
growth inhibition as a single agent and also in combination with
gemcitabine. Future studies include determining the mechanism
of action, pharmacokinetics, pharmacodynamics, metabolic stabil-
ity and off-target effects of these novel derivatives.
Figure 2. In vivo toxicity study of 5a in CD-1 mice (n = 6).
low micro molar concentrations (Table 1). Compound 5b derived
from benzaldehyde BH bromide 1b also exhibited a similar type
of cytotoxicity against these three cell lines. To further understand
the structure activity relationship (SAR), we modified the methyl
ester to butyl 5c and hexyl 5d esters and also replaced the ester
moiety with N,N-dimethyl amide 5e. However, these modifications
did not result in enhanced activity compared to the methyl ester.
We then synthesized the quaternary ammonium curcuminoids
5f and 5g from simple allyl and benzyl bromides to understand the
role and importance of BH derived allyl bromides. These molecules
5f and 5g have good water solubility (ꢀ15 mg/mL) but did not
exhibit in vitro cytotoxicity against three cancer cell lines even at
higher concentrations of 100 lM. Quaternary ammonium curcum-
inoid salt derived from piperazinyl ethanol curcuminoid 6 and BH
bromide 1a was found to be highly deliquescent and did not show
any enhanced activity compared to 5a.
We also investigated the role of the curcuminoid template in
relation to the cytotoxic properties. In this regard, we have synthe-
sized quaternary ammonium salts 7a–7c derived from N-methyl-
morpholine 7 in place of curcuminoid unit (Scheme 3). Although
these salts 7a–7c were found to be highly water soluble
(>100 mg/mL), they did not exhibit any appreciable cytotoxic prop-
erties (IC50 > 100 lM, 4T1 cells). These results clearly illustrate the
0
importance of allylic rearrangement with S2N, SN2 , or 1,4-addition
mechanism in combination with curcuminoid template in provid-
ing the pharmacological activity.
To explore the translational potential of these quaternary
ammonium curcuminoid salts as anticancer agents, we carried
out systemic toxicity and in vivo tumor growth inhibition studies
in mice.44–47 As a representative example, compound 5a was cho-
sen for all the in vivo studies due to its higher water solubility
(>100 mg/mL) and slightly better potency than the other deriva-
tives. For systemic toxicity studies, we utilized healthy CD-1 mice.
Group 1 and 2 mice (n = 6) were administered twice daily (bid),
intraperitoneally (ip) with compound 5a (30 mg/kg) and vehicle
(saline), respectively. The dosage for group 1 was increased to
50 mg/kg after 7 days. In both cases, body weight gains were found
to be similar in treated group and control group (Fig. 2). All the
Acknowledgments
This work was supported by Whiteside Clinical Research
Institute, Duluth, Minnesota and University of Minnesota Duluth.
References and notes