Potent anti-proliferative actions of a non-diuretic glucosamine derivative of ethacrynic acid

Article abstract of DOI:10.1016/j.bmcl.2016.04.062

Ethacrynic acid (EA), a known inhibitor of the neoplastic marker glutathione S-transferase P1 and other GSTs, exerts a weak antiproliferative activity against human cancer cells. The clinical use of EA (Edecrin) as an anticancer drug is limited by its pot

Full text of DOI:10.1016/j.bmcl.2016.04.062

Bioorganic & Medicinal Chemistry Letters 26 (2016) 2829–2833
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Potent anti-proliferative actions of a non-diuretic glucosamine
derivative of ethacrynic acid
Surendra R. Punganuru, A. G. M. Mostofa, Hanumantha Rao Madala, Debasish Basak,
⇑
Kalkunte S. Srivenugopal
Department of Biomedical Sciences and Cancer Biology Center, School of Pharmacy, Texas Tech University Health Sciences Center, 1406 S. Coulter Dr., Amarillo, TX 79106, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Ethacrynic acid (EA), a known inhibitor of the neoplastic marker glutathione S-transferase P1 and other
GSTs, exerts a weak antiproliferative activity against human cancer cells. The clinical use of EA (Edecrin)
as an anticancer drug is limited by its potent loop diuretic activity. In this study, we developed a
Received 7 March 2016
Accepted 21 April 2016
Available online 22 April 2016
non-diuretic 2-amino-2-deoxy-D-glucose conjugated EA (EAG) to target tumors cells via the highly
expressed glucose transporter 1 (GLUT1). Cell survival assays revealed that EAG had little effect on
normal cells, but was cytotoxic 3 to 4.5-fold greater than EA. Mechanistically, the EAG induced selective
cell death in cancer cells by inhibiting GSTP1 and generating abundant reactive oxygen species.
Furthermore, EAG induced p21cip1 expression and a G2/M cell cycle block irrespective of the p53 gene
status in tumor cells. These data encourage the development of new EA analogs.
Ó 2016 Published by Elsevier Ltd.
Keywords:
Ethacrynic acid
2-Amino-2-deoxy-D-glucose
Reactive oxygen species
GSTP1
Cell cycle arrest
Redox-stress
Low levels of reactive oxygen species (ROS) generated as a
result of mitochondrial electron transport activity, NADH oxidase
and cytochrome p450 enzyme activities in normal cells function
in cell physiological signaling, mitogenesis and angiogenesis.
Human cancers, however, harbor higher levels of ROS, including
H2O2 due to their increased metabolic activity, cytokine production
and various gene alterations.1 Because the ROS are capable of dam-
aging crucial cellular macromolecules, including DNA, cancers
adapt to oxidative stress by upregulating the antioxidant systems
such as glutathione to counteract the detrimental effects of the
oxygen free radicals, H2O2, nitrosative and other stresses.2 This
dependency might not be shared by many nontransformed cells,
whose lower basal ROS levels and/or elevated antioxidant capacity
could provide resistance to treatments that impair ROS metabo-
lism. Therefore, the elevation of oxidative stress preferentially in
cancer cells by depleting glutathione or generating ROS is a logical
therapeutic strategy for the development of anticancer drugs.3
Consistent on this hypothesis, various small molecules having
(a Michael acceptor) and can be attacked by nucleophiles (e.g., sul-
fur-atom containing glutathione) at the b-carbon, thereby induces
oxidative stress in the cells.6 The
a,b-unsaturated carbonyl moiety
of EA also participates in the inhibition of glutathione S-transferase
P1-1 (GSTP1) by binding to the cysteinyl residue in the active site
via a Michael-like addition.7 GSTP1 is highly expressed in human
malignancies and plays both catalytic and non-catalytic roles as a
major determinant of tumor resistance to alkylating agents,
cisplatin and other anticancer drugs.8 While the conjugation of
the electrophilic drugs with GSH by GSTs directly inactivates the
drugs, the ability of GSTP1 to form protein complexes with the
stress-related and apoptosis-inducing kinases (JNK, ASK1 etc.)
restrains the cell signaling in non-stressed cells.9 Along this lines,
it has been demonstrated that EA exerts anti-proliferative effects
against tumor cells, albeit at higher concentrations, but enhances
the therapeutic efficacies of several anticancer agents. Neverthe-
less, the relative lack of potency and the potent diuretic properties
diminish the use of EA as a chemotherapeutic agent.10 Structure
and activity relationship studies have revealed that the ability of
disulfide,
a,b-unsaturated carbonyl, sulfonate, or other elec-
trophilic functional groups, have previously been reported to
elevate ROS levels and induce cancer cell death by depleting
glutathione levels.4
EA to inhibit GSTP1 activity results from its a,b-unsaturated
carbonyl group and that the main diuretic side effect of EA results
from its carboxyl group.11 Thus, several groups developed EA
analogs by modifying acid group into ester, amide, oxazole,
thiazole and others.12
Ethacrynic acid (EA) is a loop diuretic used to treat hypertension
treatment.5 EA possesses an
a,b-unsaturated carbonyl unit
Cancers of diverse origins exhibit a marked demand for glucose
for sustaining elevated rates of glycolysis and metabolite require-
ments. Glucose transporter 1 (GLUT1) is a major mediator of
⇑
Corresponding author. Tel.: +1 806 414 921.
E-mail address: Kalkunte.Srivenugopal@ttuihsc.edu (K.S. Srivenugopal).
http://dx.doi.org/10.1016/j.bmcl.2016.04.062
0960-894X/Ó 2016 Published by Elsevier Ltd.

2830
S. R. Punganuru et al. / Bioorg. Med. Chem. Lett. 26 (2016) 2829–2833
glucose uptake into the cancers through a facilitative diffusion.13
Consequently, an increased supply for glucose in tumors translates
to an enhanced expression and activity of GLUT1, which can be
exploited as a targeting strategy. Thus, glycoconjugation, the link-
activity by targeting the oxidative stress response along with
GSTP1 inhibition without the unwanted diuretic activity.
The preparation of ethacrynic acid-glucosamine conjugate
(EAG) was accomplished by a three-step reaction as presented in
Scheme 1. Commercially available EA was first converted in to
NH-ester with N-hydroxy succinamide using DCC and activated
ing of a drug to glucose, particularly, 2-amino-2-deoxy-D-glucose
(2DG), which is recognized by and transported into the cells by
GLUT1 has emerged as a viable approach for tumor targeting.
The above strategy has been exploited in preclinical and clinical
imaging for tumor diagnosis, staging, and monitoring of therapeu-
EA 2 was reacted with 2,3,4,6-tetra-O-acetyl-beta-D-glucosamine
(2-DG) hydrochloride in the presence of base diisopropylethy-
lamine to yield acetyl protected EAG intermediate 3. Deprotection
of acetyl groups by using sodium methoxide gave the final product
EAG in good yield.
tic response.14 In the present study, we developed
a new
glucosamine conjugated EA, which shows excellent anticancer
O
O
OAc
NH2
AcO
AcO
Cl
O
N
OH
Cl
Cl
O
O
OAc
Cl
O
O
N
O
HO
NEt3, CH2Cl2
rt, 3h
O
DCC, CH2Cl2
rt, 12h
80 %
O
O
O
1
82 %
EA
Cl
O
Cl
O
Cl
OH
Cl
H
N
OAc
H
N
CH3ONa
O
O
O
O
HO
O
CH3OH, rt, 2h
75 %
OH
AcO
O
OAc
OH
OAc
2
EAG
Scheme 1. Synthesis of EAG.
Figure 1. (A) Time course of diuresis in mice treated with EA or EAG at 10 mg/kg. Control group received the vehicle alone. Mice were placed on an elevated grid floor and
isolated. Narrow graduated cylinders were placed underneath each mouse to collect the voided urine. (B) Cytotoxicity of EAG against MCF10A normal breast epithelial cells
and cancerous MDA-MB-231 cells. (C) Cytotoxic effect of 2-DG against MCF10A and MDA-MB-231 cells. MTT assays were used. (D) Thiol depletion in cancer cell lines by EAG.
The H1299, HCT116 and MDA-MB-231cells were exposed to EAG at concentrations shown for 3 h, trypsinized the total thiol levels in cell extracts were quantified using the
Ellman’s reagent. (E) Measurement of GSTP1 activity in H1299, HCT116 and MDA-MB-231cells after EAG treatment. Cells were exposed to EAG at concentrations specified for
3 h. The resulting cell extracts were used for the spectrophotometric assay using CDNB as the substrate. The kinetics of GSH-CDNB conjugation reactions were followed over
5 min and linear parts of the curves were used for quantitation.

S. R. Punganuru et al. / Bioorg. Med. Chem. Lett. 26 (2016) 2829–2833
2831
As a first step of biological evaluation, we tested the effect of
EAG on toxicity in mice. The animals given single doses of EAG at
120 mg/kg per week for two weeks were devoid of any ill effects.
No tremors, lethargy, paralysis, stress, weight loss or diarrhea were
observed. Next, we examined the diuretic activity of EAG and EA on
female CD-1 mice, using the procedure of Hailu et al.15 The mice
were deprived of food for 12 h prior to drug administrations. Test
compounds were prepared in PEG/EtOH/H2O (54:14:32) and
injected IP at 10 mg/kg. The animals were then divided into three
groups, namely, the vehicle control, EAG- treated and EA-treated.
The mice were placed in separate metabolic cages with free access
to water. Urine was collected 0–6 h after drug injections in gradu-
ated tubes kept directly under the cage manifolds. The urine
volumes shown in Fig. 1A indicate that EAG did not induce any
diuretic effect over a 6 h test period compared to the progressive
increase (up to 6-fold) of urine production seen with the EA.
Furthermore, the urine output during the 6–24 h period was simi-
lar between the EAG and untreated controls (not shown). The
results confirm that the side chain acid group of EA is indeed a
major contributor to the diuretic property of the molecule.
Next, we evaluated the antiproliferative activity of EAG against
a panel of ten different human cancer cell lines from lung (H1299),
brain (SF188, GBM10), breast (SKBR-3, MDA-MB-231, MDA-MB-
468, MCF7), colon (HT29 and HCT116) and fibrosarcoma
(HT1080) along with the normal breast epithelial cells (MCF10a)
using cell survival assays. The data shown in Table 1 indicates that
EAG inhibited the growth of all cancer cell lines more potently (3 to
4.5-fold) than EA. Among the different cell lines tested, EAG
showed
a greater potency toward SF188 glioblastoma and
HCT116 colon cancer cells. Overall, the EAG was equally potent
Table 1
In vitro cell growth inhibitory effects of EA and EAG on human cancer cell linesa
Number
Tumor type
Cell lines
Cytotoxicity (IC50),
EAG EA
lM
1
2
3
4
5
6
7
8
9
10
Lung
Brain
Brain
Breast
Breast
Breast
Breast
Colon
H1299
GBM10
SF188
SKBR-3
MDA-MB-231
MDA-MB-468
MCF7
HT29
HCT116
HT1080
11.53 2.51
15.2 1.23
44.29 1.79
29.88 2.12
31.65 1.26
34.21 1.6
45.24 4.88
81.45 4.22
28.92 3.25
62.39 3.67
30.22 4.62
54.53 4.26
12.29 0.98
16.09 1.68
12.17 1.63
18.37 1.92
10.12 1.12
21.73 2.98
10.29 1.79
16.34 2.19
Colon
Fibrosarcoma
a
Figure 3. The effect of EAG on p53, p21, PUMA and PARP proteins was determined
by western blot analysis in H1299 (p53-null), HCT116 (p53 wt) and MDA-MB-231
(p53 mutant) cells. b-Actin was used as a loading control.
IC50: compound concentration required to inhibit tumor cell proliferation by
50%. Data is expressed as mean SD from the cell survival curves of three inde-
pendent experiments.
Figure 2. (A) Flow cytometric analyses of EAG-induced ROS elevation and its reversal by N-acetyl cysteine (NAC) in cancer cells. H1299, HCT116 and MDA-MB-231 cells were
treated with or without EAG or 25 M H2O2 for 3 h. In some cases, cells wert exposed to 3 mM NAC for 1 h prior to EAG treatment. The cell pellets were stained with 20
DCF-DA for 30 min prior to flow cytometry using a BD-Biosciences FACSVERSE instrument. (B) Representative Fluorescence microscopy images of H1299, HCT116 and MDA-
l
lM
MB-231cells treated for 3 h with different concentrations of EAG. H2O2 (25
EAG exposure.
lM) was used as positive control. In the last panel, cells were pretreated with 3 mM NAC before

2832
S. R. Punganuru et al. / Bioorg. Med. Chem. Lett. 26 (2016) 2829–2833
against all the cancer cell lines irrespective of their p53 gene
status; the p53-null H1299 cells were also sensitive to the drug.
EAG inhibited the proliferation of the normal breast epithelial cells
(MCF10A) very slightly compared with the breast cancer MDA-MB-
231 cells (Fig. 1B); the free 2DG ligand had no negative effect on
cell growth on either cancerous or normal cells (Fig. 1C).
Structurally, the EAG shares characteristics of EA to conjugate
with GSH and inhibit the GSTP1. EA can conjugate with GSH
non-enzymically or enzymically and this conjugate is a more
potent inhibitor of GSTP1.16 Therefore, we measured the EAG-
induced depletion of total thiols in three cell lines (H1299,
HCT116, MDA-MB-231) using the Ellman’s reagent. Fig. 1D shows
(Fig. 2A). EAG-induced ROS generation was also confirmed by
fluorescence microscopy (Fig. 2B). Again, 40 M EAG induced
intense green fluorescence at similar levels produced by 25
H2O2 in all three cell lines; as in the flow cytometry experiments,
preincubation with N-acetylcysteine eliminated the EAG-gener-
ated fluorescence (Fig. 2B). Collectively, the observations made in
Figs. 1 and 2 strongly suggest that EAG effectively depletes GSH
to suppress the antioxidant systems and promote an oxidatively
stressed cellular milieu.
To investigate the mechanistic basis of EAG’s antiproliferative
effects and consequences of the redox stress created, we deter-
mined the levels of p53 and related apoptotic protein regulators.
As shown in Figure 3, western blot analyses revealed that wt p53
expression along with p21waf1/cip1 was significantly enhanced in
HCT116 cells by EAG. Further, the p21cip1 and PUMA, a p53
proapoptotic target, were induced in response to EAG, even in
p53 null (H1299) and p53 mutant (MDA-MB-231) cells (Fig. 3).
EAG treatment greatly enhanced the levels of the cleaved PARP
(poly ADP-ribose polymerase) in all three cell lines. Overall, the
results suggest that the redox imbalance triggered by EAG func-
tions to prime the tumor cells toward apoptotic cell death. Further
insight into the cytotoxic mechanisms of EAG was obtained by flow
cytometric analyses of cell cycle progression. Accumulating data
suggest that oxidative stress induced p21cip1 inhibits cell cycle reg-
ulating cyclin-dependent kinases and orchestrates a G2/M phase
arrest in cancer cells.17 Consistent with this notion, the histograms
displayed in Figure 4 show that EAG consistently induced a G2/M
cell cycle arrest in a concentration dependent manner in HCT116,
H1299 and MDA-MB-231cells.
l
lM
that cells incubated with 0–40
lM EAG for 3 h lost the thiol con-
tent in a concentration-dependent manner, with 40
l
M achieving
a maximal 58% depletion. Extracts prepared from the EAG-treated
cells were also subjected to GST activity using CDNB (1-chloro-2,4-
dinitrobenzene) as the substrate. The catalytic activity of GSTs in
these cells, again showed a significant and EAG-dose dependent
inhibition with a maximal 60% inhibition in HCT116 cells following
50 lM drug treatment for 3 h (Fig. 1E). Consistent with this obser-
vation, the HCT116 cells were also more vulnerable to thiol deple-
tion (Fig. 1D).
The ability of EAG to trigger ROS generation in cancer cells was
studied by flow cytometry and confocal fluorescence microscopy
(Fig. 2). Cells were treated with EAG (20 and 40 lM) and the level
of ROS was measured using DCFH-DA (dichlorofluorescein-diac-
etate) as a fluorescent probe for intracellular ROS. The cell suspen-
sions were analyzed using a BD Biosciences flow cytometer. EAG
treatment for 3 h in H1299, HCT116 and MDA-MA-231 cells
induced a significant accumulation of ROS as evident from the
fluorescence shift compared with the untreated controls (Fig. 2A).
In conclusion, we designed and synthesized a non-diuretic
ethacrynic acid analog EAG with significant anti-cancer activity.
EAG selectively targeted the cancer cells by downregulating the
GSH/GST coupled antioxidant pathway and generating abundant
ROS. The glucosamine linkage in EAG appeared to serve a dual pur-
pose in blocking the diuretic activity and increasing the tumor
Cells treated with 25 lM H2O2 served as positive control. Pretreat-
ment of cells with N-acetylcysteine prior to EAG resulted in a
fluorescence shift and signal attenuation to control levels, thus
verifying the ROS production by EAG and its quenching by NAC
Figure 4. Induction of G2/M cell cycle arrest by EAG is not related to the p53 gene status. H1299, HCT116 and MDA-MB-231 cells were treated with EAG at concentrations
shown for 24 h and subjected to flow cytometry by standard procedures described in Supporting information.

S. R. Punganuru et al. / Bioorg. Med. Chem. Lett. 26 (2016) 2829–2833
2833
selectivity. The EAG-induced p21cip1expression and the accompa-
nying G2/M blockade occurred irrespective of the p53 gene status.
Our findings encourage a preclinical evaluation of EAG and synthe-
sis of new non-diuretic EA analogs for cancer treatment.
Cell 2006, 10, 241; (c) Shaw, A. T.; Winslow, M. M.; Magendantz, M.; Ouyang, C.;
Dowdle, J.; Subramanian, A.; Lewis, T. A.; Maglathin, R. L.; Tolliday, N.; Jacks, T.
Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 8773; (d) Raj, L.; Ide, Takao; Gurkar, A. U.;
Foley, M.; Schenone, M.; Li, X.; Tolliday, N. J.; Golub, T. R.; Carr, S. A.; Shamji, A.
F.; Stern, A. M.; Mandinova, A.; Schreiber, S. L.; Lee, S. W. Nature 2011, 475, 231;
(e) Akladios, F. N.; Andrew, S. D.; Parkinson, C. J. Bioorg. Med. Chem. 2015, 23,
3097.
Acknowledgements
5. Tew, K. D.; Dutta, S.; Schultz, M. Adv. Drug Delivery Rev. 1997, 26, 91.
6. Zhao, G.; Wang, X. Curr. Med. Chem. 2006, 13, 1461.
7. Ploemen, J. H. T. M.; Van Schanke, A.; Van Ommen, B.; Van Bladeren, P. J. Cancer
Res. 1994, 54, 915.
8. Townsend, D. M.; Tew, K. D. Oncogene 2003, 22, 7369.
9. Gate, L.; Tew, K. D. Expert Opin. Ther. Targets 2001, 5, 477.
10. Jin, G.; Lu, D.; Yao, S.; Wu, C. C. N.; Liu, J. X.; Carson, D. A.; Cottam, H. B. Bioorg.
Med. Chem. 2009, 17, 606.
This work was supported by grants from the Cancer Prevention
Research Institute of Texas (RP130266), the Carson-Leslie Founda-
tion and the Association for Research of Childhood Cancer, all to K.
S.S.
11. Van Iersel, M. L.; van Lipzig, M. M.; Rietjens, I. M.; Vervoort, J.; van Bladeren, P.
J. FEBS Lett. 1998, 441, 153.
Supplementary data
12. (a) Zhao, G.; Yu, T.; Wang, R.; Wang, X.; Jing, Y. Bioorg. Med. Chem. 2005, 13,
4056; (b) Yang, X.; Liu, G.; Li, G.; Zhang, Y.; Song, D.; Li, C.; Wang, R.; Liu, B.;
Liang, W.; Jing, Y.; Zhao, G. J. Med. Chem. 2010, 53, 1015.
13. Hediger, M. A.; Rhoads, D. B. Physiol. Rev. 1994, 74, 993.
14. (a) Gallagher, B. M.; Fowler, J. S.; Gutterson, N. I.; MacGregor, R. R.; Wan, C. N.;
Wolf, A. P. J. Nucl. Med. 1978, 19, 1154; (b) Song, Y. S.; Lee, W. W.; Chung, J. H.;
Park, S. Y.; Kim, Y. K.; Kim, S. E. Lung Cancer 2008, 61, 54; (c) van Baardwijk, A.;
Dooms, C.; van Suylen, R. J.; Verbeken, E.; Hochstenbag, M.; Dehing-Oberije, C.
Eur. J. Cancer 2007, 43, 1392.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.04.
060.
References and notes
15. Hailu, W.; Engidawork, E. BMC Complement. Altern. Med. 2014, 14, 135.
16. Shi, B.; Stevenson, R.; Campopiano, D. J.; Greaney, M. F. J. Am. Chem. Soc. 2006,
128, 8459.
17. Seomun, Y.; Kim, J. K.; Kim, K. Y.; Kim, H. S.; Park, Y. Y.; Joo, C. K. Mol. Vision
2005, 11, 764.
1. Szatrowski, T. P.; Nathan, C. F. Cancer Res. 1991, 51, 794.
2. Halliwell, B. Biochem. J. 2007, 401, 1.
3. Trachootham, D.; Alexandre, J.; Huang, P. Nat. Rev. Drug Disc. 2009, 8, 579.
4. (a) Huang, P.; Feng, L.; Oldham, E. A.; Keating, M. J.; Plunkett, W. Nature 2000,
407, 390; (b) Trachootham, D.; Zhou, Y.; Zhang, H.; Demizu, Y.; Chen, Z.;
Pelicano, H.; Chiao, P. J.; Achanta, G.; Arlinghaus, R. B.; Liu, J.; Huang, P. Cancer