A novel metformin derivative, HL010183, inhibits proliferation and invasion of triple-negative breast cancer cells

Article abstract of DOI:10.1016/j.bmc.2013.02.015

Mounting evidence suggests that metformin (N,N-dimethylbiguanide), a widely prescribed drug for the treatment of type II diabetes, exerts an anti-tumor effect on several cancers including breast cancer. Breast cancer has been estimated as one of the most commonly diagnosed types of cancer among women. In particular, triple-negative breast cancers are associated with poor prognosis and metastatic growth. In the present study, we synthesized a novel metformin derivative 5 (HL010183) and metformin salts, 9a, 9b, and 9c (metformin gamma-aminobutyric acid (GABA) salt, metformin pregabalin salt and metformin gabapentin salt), which exerted more potent inhibitory effects on the proliferation and invasiveness of Hs578T triple-negative breast carcinoma cells than metformin. Importantly, 5 showed approximately 100-fold more potent effects compared to metformin. In a triple-negative breast cancer xenograft model, 5 showed a comparable degree of inhibitory effect on in vivo tumor growth at the 100 mg/kg dose to that of metformin at 500 mg/kg. Our results clearly demonstrate that 5 exerts a potent anti-tumor effect both in vitro and in vivo, paving the way for a strategy for treatment of triple-negative breast cancer.

Full text of DOI:10.1016/j.bmc.2013.02.015

Bioorganic & Medicinal Chemistry 21 (2013) 2305–2313
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
A novel metformin derivative, HL010183, inhibits proliferation and
invasion of triple-negative breast cancer cells
Minsoo Koh a, Jong-Cheol Lee b, Changhee Min c, Aree Moon a,
⇑
a College of Pharmacy, Duksung Women’s University, Seoul 132-714, Republic of Korea
b Energy Materials Research Center, Korea Research Institute of Chemical Technology, Daejeon, Republic of Korea
c HanAll Biopharma Co., Ltd, Central Research Institute, Daejeon, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Mounting evidence suggests that metformin (N,N-dimethylbiguanide), a widely prescribed drug for the
treatment of type II diabetes, exerts an anti-tumor effect on several cancers including breast cancer.
Breast cancer has been estimated as one of the most commonly diagnosed types of cancer among women.
In particular, triple-negative breast cancers are associated with poor prognosis and metastatic growth. In
the present study, we synthesized a novel metformin derivative 5 (HL010183) and metformin salts, 9a,
9b, and 9c (metformin gamma-aminobutyric acid (GABA) salt, metformin pregabalin salt and metformin
gabapentin salt), which exerted more potent inhibitory effects on the proliferation and invasiveness of
Hs578T triple-negative breast carcinoma cells than metformin. Importantly, 5 showed approximately
100-fold more potent effects compared to metformin. In a triple-negative breast cancer xenograft model,
5 showed a comparable degree of inhibitory effect on in vivo tumor growth at the 100 mg/kg dose to that
of metformin at 500 mg/kg. Our results clearly demonstrate that 5 exerts a potent anti-tumor effect both
in vitro and in vivo, paving the way for a strategy for treatment of triple-negative breast cancer.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 18 December 2012
Revised 5 February 2013
Accepted 7 February 2013
Available online 22 February 2013
Keywords:
Metformin
Triple-negative breast cancer
Anti-cancer agent
Invasion
Xenograft model
1. Introduction
breast cells requires a matrix-degrading activity which is exerted
by matrix metalloproteinases (MMPs), especially MMP-2 which is
The metabolic abnormalities associated with diabetes mellitus
type II have been linked to an increased cancer risk. A central en-
ergy sensor, AMP-activated protein kinase (AMPK) which inhibits
the mammalian target of rapamycin (mTOR), is known as a crucial
factor in the interaction between metabolism and cancer.1 An
AMPK activator metformin (N,N-dimethylbiguanide) is a widely
prescribed oral medication used for the treatment of type II diabe-
tes which causes reversal of metabolic disturbances such as hyper-
glycemia and insulin resistance.2 Several concordant series of
epidemiological studies indicate that metformin has a high poten-
tial efficacy as an anti-cancer drug.3–6 Animal and cellular studies
support that metformin has a strong anti-proliferative effect on
various cancers.7–11 The anti-tumor effect of metformin shown
in vitro and in vivo, and its low toxicity favor the use of this com-
pound for cancer treatment.
shown to be associated with breast cell invasion.15,16 Metformin
suppressed cancer metastasis in vivo as well as the invasion and
migration of cancer cells in vitro.11,17,18 In addition, metformin in-
duced apoptosis of breast cancer cells in vitro and reduced the tu-
mor growth in a breast cancer xenograft mouse model.8 The anti-
tumor effect of metformin on breast cancer has been demonstrated
in clinical studies.1,4,19
A biguanide analogue, phenformin (phenethyl biguanide) was
used as an anti-diabetic drug, but was withdrawn from clinical
use due to the association with lactic acidosis.2 Interestingly, re-
cent reports demonstrated that phenformin exerted more potent
anti-tumoral activity than metformin in colon cancer cells and in
a triple-negative breast cancer xenografts model,20,21 suggesting
a promising application of this agent as an anti-cancer drug. Gam-
ma-aminobutyric acid (GABA) and its structural analogues, gaba-
pentin and pregabalin, are FDA-approved drugs for the treatment
of anxiety and convulsions.22 GABA has emerged as a regulator of
cancer cell proliferation including colon, gastric, ovarian and breast
cancers.23
Breast cancer has been estimated as the second most common
cause of cancer death among women.12 Especially, triple-negative
breast cancers, which do not express the genes for estrogen and
progesterone receptors (ER/PR) and HER2, are associated with poor
prognosis and metastatic growth.13 Major cause of death from
breast cancer is the metastatic spread of the cancer.14 Invasion of
Our current study aimed to synthesize a metformin derivative
and metformin salts that may have more potent inhibitory effects
on the proliferation and invasive phenotype of triple-negative
breast cancer cells than metformin. Here, we present a novel tri-
azine compound 5 which exerted a stronger potency to inhibit pro-
⇑
Corresponding author. Tel.: +82 2 901 8394; fax: +82 2 901 8386.
E-mail address: armoon@duksung.ac.kr (A. Moon).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.02.015

2306
M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
NH2
N
NH2
N
NH2
N
HCl
NH2
N
CN
d
b
N
N
a
c
N
1
N
N
N
N
N
H
N
N
N
H
MeS
SMe
N
N
S
O
N
N
S
4
5
2
3
Scheme 1. Synthesis of a metformin derivative, 5. Reagents and conditions: (a) N,N-dimethylguanidine sulfate, 40% K2CO3 solution, DMSO, 120 °C, 2 h; (b) m-CPBA, CH2Cl2,
0 °C to room temp, 12 h; (c) 2-propylaniline, 1,4-dioxane, reflux, 15 h; (d) 4 M HCl in dioxane, room temp, 2 h.
NH NH
O
O
NH2
NH2
N
N
N
H
NH2
HO
HO
9a
9b
NH NH
NH NH
HCl
NH NH
a
b
H
N
N
H
NH2
N
N
H
NH2
N
H
NH2
8a
8b
8c
S
7
6
NH NH
O
NH2
N
N
H
NH2
HO
9c
Scheme 2. Synthesis of metformin salts, 9a, 9b, and 9c. Reagents and conditions: (a) KOH, i-PrOH, 50 °C, 2 h; (b) acid (8a: GABA, 8b: pregabalin, 8c: gabapentin), i-PrOH,
50 °C, 2 h.
NH NH
NH NH
NH NH
b
a
HCl
AcOH
NH2
N
H
N
H
NH2
N
H
N
H
NH2
N
H
N
H
10
11
12
Scheme 3. Synthesis of a phenformin acetic acid salt, 12. Reagents and conditions: (a) KOH, i-PrOH, 50 °C, 2 h; (b) acetic acid, i-PrOH, 50 °C, 2 h.
liferation and invasion of a triple-negative breast cancer cells than
metformin. Metformin salts, 9a, 9b, and 9c, also showed higher
anti-proliferative and anti-invasive activities than metformin.
formin salts, 9a, 9b, and 9c, as shown in Scheme 2. The transforma-
tion of compound 6 to free base (compound 7) was achieved using
potassium hydroxide in isopropyl alcohol. Then, treatment of the
free base with organic acids resulted in the production of 9a, 9b
and 9c.
2. Chemistry
2.3. Chemical synthesis of phenformin acetic acid salt, 12
2.1. Chemical synthesis of a metformin derivative, 5 (HL010183)
Compound 12 was synthesized as shown in Scheme 3. To a stir-
red suspension of phenformin hydrochloride (compound 10) in i-
PrOH was added potassium hydroxide. The filtrate was concen-
trated under reduced pressure and dried in vacuo to provide a
phenformin free base (compound 11). Acetic acid was added to
the compound 11 in i-PrOH to give compound 12. Structures of
metformin, 5 and 12 were depicted in Figure 1.
Based on the structures of metformin and phenformin, we syn-
thesized a novel metformin derivative 5 which contains a rigid and
symmetrical scaffold as shown in Scheme 1. The triazine formation
reaction by treating dimethyl-N-cyanodithioiminocarbonate (com-
pound 1) with N,N-dimethylguanidine sulfate in the presence of
potassium carbonate (K2CO3) afforded the triazine compound 2.
Next, the subsequent oxidation of the triazine compound 2 was
carried out with m-chloroperbenzoic acid (m-CPBA) at room tem-
perature to give the sulfinyl compound 3. Then, treatment of com-
3. Results and discussion
pound
3 with 2-propylaniline in 1,4-dioxane under reflux
condition, provided compound 4. Finally, compound 4 was treated
with hydrochloric acid (HCl) to give compound 5, denominated as
HL010183, in a quantitative yield.
3.1. Compound 5, 9a, 9b, and 9c showed more potent anti-
proliferative effects than metformin on triple-negative breast
cancer cells
2.2. Chemical synthesis of metformin salts, 9a, 9b, and 9c
To examine the effects of 5, 9a, 9b, and 9c on the growth of
breast cancer cells, we performed the MTT assay on Hs578T and
MDA-MB-231 human breast carcinoma cells which were derived
from triple-negative breast cancer.24 As shown in Figure 2A, 5
Because GABA was shown to be involved in the regulation of
several cancer cells including breast cancer,23 we synthesized met-

M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
2307
NH NH
metformin (IC50 value of 58.4 mM) (Table 1). Compounds 9a, 9b,
and 9c exerted more potent cytotoxic effects against Hs578T with
IC50 values of 9.1, 7.2 and 4 mM, respectively, than metformin (IC50
value of 26.8 mM). Compound 9a, 9b and 9c also inhibited the pro-
liferation of MDA-MB-231 cells with IC50 values of 41.5, 17.7 and
16.1 mM, respectively (Fig. 2B and Table 1). Treatment with GABA
alone did not markedly affect the growth of these cells. Compound
5 showed stronger cytotoxic activities than 9a, 9b, and 9c, showing
the most potent cytotoxic effect on the growth of Hs578T cells.
Next, we examined whether the most potent 5 can induce apop-
tosis stronger than metformin by flow cytometric analysis. As
shown in Figure 3, treatment with 0.1 mM of 5 and 10 mM of met-
formin showed a similar effect on the induction of apoptosis in
MDA-MB-231 cells.
HCl
N
N
H
NH2
6
Metformin
NH2
N
HCl
N
N
N
N
H
5
HL010183
3.2. Compound 5 inhibited invasive phenotype of Hs578T cells
NH NH
AcOH
We investigated the effects of 5, 9a, 9b, and 9c on invasive phe-
notype of Hs578T cells which are highly invasive.13,24 As shown in
Figure 4A, treatment with 0.2 mM of 5 inhibited the invasiveness
of Hs578T cells by 36.6%, which was comparable to the inhibiting
effect exerted by 20 mM of Metformin. These data implicate that
about 100-fold more potent activity was exhibited by 5 compared
to metformin.
N
H
N
H
NH2
12
Phenformin acetic acid salt
Figure 1. Chemical structures of metformin, 12 and 5.
Neither metformin nor GABA affected the invasive phenotype at
10 mM concentration (Fig. 4A). Treatment with metformin GABA
salts, however, significantly inhibited the invasiveness in a dose-
dependent manner. Compounds 9a, 9b, and 9c at 5 mM concentra-
tion reduced the invasiveness of Hs578T cells by 43.6%, 64.6%, and
73.6%, respectively. GABA has been shown to reduce norepineph-
rine-induced migration of colon carcinoma cells and inhibit isopro-
and 12 showed stronger cytotoxic activities than metformin in
both cell lines. A novel compound 5 exerted approximately 100-
fold more potent cytotoxic effect on the growth of Hs578T cells
(IC50 value of 0.28 mM) compared to metformin (IC50 value of
26.8 mM). Compound 5 inhibited the growth of MDA-MB-231 cells
with an IC50 value of 0.28 mM which was much lower than that by
Hs578T
MDA-MB-231
A
125
**
100
**
**
*
*
**
**
*
**
75
50
25
0
**
**
**
**
**
**
**
**
(mM)
0 10 20 30
Metformin
0 0.1 0.2 0.3
0 0.1 0.51.0
0 20 40 60
Metformin
0 0.10.20.3
0 0.5 1.0 2.0
5
12
5
12
Hs578T
MDA-MB-231
B
125
100
75
50
25
0
*
*
*
*
**
**
**
*
**
**
**
**
**
**
**
**
**
**
**
**
0 10 2030
GABA
(mM)
0 2.5 5 10
0 2.5 5 10 0 2.5 5 10
9b 9c
0 20 40 60
GABA
0 10 20 40
0 5 10 20
0 5 10 20
9b
9c
9a
9a
Figure 2. Compound 5 shows a potent anti-proliferative effect on Hs578T and MDA-MB-231 triple-negative breast cancer cells. Cells were treated with various
concentrations of compounds for 24 h and subjected to MTT assay. The results represent means SE of triplicates. ⁄
<0.01, respectively.
,
⁄⁄ Statistically different from the control at p <0.05 and p

2308
M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
Table 1
5 exerted approximately 100-fold higher anti-invasive activity
compared to metformin, suggesting anti-invasive activity of 5 is
associated with down-regulation of MMP-2 in Hs578T triple-nega-
tive human breast cancer cells.
IC50 values for metformin, 5, 9a, 9b, 9c and 12 in Hs578T and MDA-MB-231 cells
Compound
IC50 (mM)
Hs578T
MDA-MB-231
Metformin reduced phorbol-12-myristate-13-acetate-induced
migration and invasion as well as the expression of MMP-2 and
MMP-9 in HT-1080 fibrosarcoma cells at 5 mM concentration.18
In this study, we showed that 5 mM of metformin did not inhibit
the invasion of Hs578T cells, suggesting that the in vitro potency
of metformin on cell invasion may depend on cell type. Recent
studies showed that phenformin showed a more potent anti-tu-
moral activity than metformin in vitro and in vivo.20,21 Consistent
with these findings, our data showed that the anti-proliferative
and anti-invasive effects of phenformin salts, 12, on breast cancer
cells were higher than those of metformin. Because phenformin
has been known to be less suitable for clinical use due to the higher
rates of lactic acidosis,2 further investigations would be needed to
evaluate the efficacy and/or safety of 12 in breast cancer therapy.
Metformin
5
9a
9b
9c
12
26.8
0.28
9.1
7.2
4.0
58.4
0.28
41.5
17.7
16.1
1.68
0.9
terenol-induced migration of pancreatic cancer cells.25,26 Our data
showed that GABA exerted neither anti-proliferative nor anti-inva-
sive activities in Hs578T cells. Interestingly, the GABA salts of met-
formin, 9a, 9b, and 9c, showed strong inhibitory effects on
proliferation and invasion, implying that the combinatorial synthe-
sis of metformin with GABA or its analogues exerted synergistic
anti-tumoral and anti-invasive activities than the use of metformin
or GABA alone.
Among the compounds tested, we selected 5 for subsequent
studies since it showed the most potent inhibitory effect on prolif-
eration and invasive phenotype of Hs578T human breast carci-
noma cells. We examined whether 5 affects the expression of
MMP-2 or MMP-9 in Hs578T cells. The expression level of MMP-
2, but not that of MMP-9, was dose-dependently reduced in condi-
tioned media of Hs578T cells treated with metformin and 5 as evi-
denced by gelatin zymogram assay (Fig. 4B, top) and immunoblot
analysis (Fig. 4B, bottom). A comparable degree of inhibition was
observed by treatment with 10 mM of metformin and 0.1 mM of
5. These results demonstrated that a novel metformin derivative
3.3. Compound 5 showed about 100-fold more potent effect on
AMPK activation and mTOR inhibition compared to metformin
Metformin is a known activator of AMPK which inhibits phos-
phorylation of mTOR.1 To compare the effects of metformin and
5 on the activation of AMPK, we performed immunoblot analysis.
As shown in Figure 5A, 20 mM of metformin and 0.2 mM of 5 ex-
erted a comparable degree of activation of AMPK in Hs578T cells.
A similar level of inhibitory effect on mTOR phosphorylation was
observed by 20 mM of metformin and 0.2 mM of 5 (Fig. 5B). These
data indicate that 5 exerts about 100-fold more potent effect on
the activation of AMPK and inhibition of mTOR phosphorylation
A
control
5 (0.1mM)
Apoptosis
Apoptosis
Dead
Dead
9.7%
2.8%
17.0%
1.6%
1.5%
86.1%
78.4%
3.1%
Early
Early
apoptosis
apoptosis
Metformin (10mM)
control
B
Apoptosis
Dead
Apoptosis
Dead
6.4%
0.5%
15.3%
1.8%
2.1%
91%
0.5%
82.4%
Early
Early
apoptosis
apoptosis
Figure 3. Compound 5 induces apoptosis of MDA-MB-231 triple-negative breast cancer cells. MDA-MB-231 breast cancer cells were treated with metformin (A) and 5 (B) for
24 h. Flow cytometric analysis was performed using annexin V-FITC and PI stain.

M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
2309
A
120
100
80
60
40
20
0
*
**
**
**
**
**
**
**
**
**
**
(mM)
0 5 10
0
5 10 20
0
5
10
0
5
10
0 0.1 0.2 0.5
0 0.10.2 0.5
0
5 10
Metformin
GABA
5
9a
9b
9c
12
Metformin
5
B
(mM)
0
5
10 20
0
0.05 0.1 0.2
MMP-9 (92kDa)
MMP-2 (72kDa)
←
←
1.25
1.00
0.75
0.5
*
*
*
*
0.25
0
5
Metformin
10 20 (mM)
← MMP-2
0
0.05 0.1 0.2
0
5
1.25
1.00
0.75
0.5
*
*
*
*
*
0.25
0
Figure 4. Compound 5 inhibits invasive phenotype and MMP-2 in Hs578T cells. (A) Cells were treated with various concentrations of compounds and subjected to in vitro
invasion assay. The number of invaded cells per field was counted (Â400) in 13 arbitrary visual fields. (B) The conditioned media were analyzed for MMP-2 by gelatin
⁄
⁄⁄
Statistically different from the control at p <0.05 and p <0.01, respectively.
zymogram assay (top) and immunoblot analysis (bottom).
,
compared to metformin, suggesting that the rigid and symmetrical
scaffold structure of 5 may potentiate the pharmacological activi-
ties of metformin.
could be a promising therapeutic drug against triple-negative
breast cancer progression.
Liu et al. (2009) reported that metformin treatment (2 mg/ml in
drinking water) significantly slowed tumor growth in a xenograft
tumor model.8 In the present study, a similar degree of growth
inhibition was observed by 5 at the dose of 100 mg/kg to that by
metformin at the dose of 500 mg/kg. Compound 5 showed about
fivefold stronger in vivo anti-tumor efficacy than metformin, while
5 exerted approximately 100-fold more potent in vitro inhibitory
effects on proliferation and invasion of breast cancer cells. A plau-
sible explanation would be unfavorable in vivo pharmacokinetic
properties of 5.27 It is important to understand the underlying
mechanisms responsible for the differences between in vitro and
in vivo efficacies of 5, which may lead to a successful and rational
drug development.
3.4. Compound 5 showed a more potent anti-tumor effect than
metformin in vivo
Next, we investigated the inhibitory effect of 5 on the in vivo tu-
mor growth in a xenograft mouse model using a MDA-MB-231 tri-
ple-negative breast cancer cells. The MDA-MB-231 cells were
injected subcutaneously into mice. The tumor-bearing mice were
orally administered with 5 (25, 50 and 100 mg/kg) and metformin
(500 mg/kg) once a day for 35 days. The changes in body weight
between the vehicle and the 5 or metformin-treated mice were
not remarkably different during the experiment (data not shown).
Tumors were excised and tumor volume was measured. Tumor
growth was time-dependently slowed in the 5 or metformin-trea-
ted mice (Fig. 6). The mean volume of tumor tissues from mice
treated with 5 for 35 days at 50 and 100 mg/kg doses was signifi-
cantly reduced by 17.4% and 27.9%, respectively, compared to vehi-
cle group. The anti-tumor effect of 5 at the dose of 100 mg/kg
which caused 27.9% growth inhibition was comparable to that of
metformin at the dose of 500 mg/kg (28% growth inhibition) (Ta-
ble 2). These results demonstrate that 5 inhibits breast cancer cell
growth more effectively than metformin in vivo, suggesting that it
4. Conclusions
In the present study, we synthesized a novel metformin deriva-
tive 5 with a rigid and symmetrical scaffold on the basis of bigua-
nide structure. Here, we report that 5 exerts a stronger anti-
proliferative and anti-invasive activities than a phenformin salt,
12 and metformin against triple-negative breast cancer cells. Since
metastasis is the major cause of death from breast cancer, the po-
tent anti-invasive activity of 5 against triple-negative breast cancer

2310
M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
Table 2
Metformin
10 20 (mM)
← pAMPK
5
A
The anti-tumor effects of 5 and metformin in a MDA-MB-231 breast cancer xenografts
model
0
0.05 0.1 0.2
0
5
Compound
Tumor volume (mm3)
Tumor growth inhibition (%)
Vehicle
2005.7 246.7
0.0
← AMPK
5
5
4
3
2
1
0
25 mg/kg
50 mg/kg
100 mg/kg
1881.2 271.7
1656.5 161.0
1445.4 106.1
6.2
17.4
27.9
*
*
*
*
*
Metformin
500 mg/kg
*
1444.1 169.0
28.0
Metformin
5
B
current drugs may accelerate the successful development for
breast cancer therapeutics.
(mM)
0
0.05 0.1 0.2
0
5
10 20
← p mTOR
5. Experimental section
5.1. Chemistry
← β-actin
1.25
1.00
0.75
0.5
Melting points were determined on an Electrothemal No.9201
and were uncorrected. Proton and carbon nuclear magnetic reso-
nance (1H and 13C NMR) spectra were recorded on the following
instruments: Bruker model Avence-500 (500 MHz) and Varian
model Inova-600 (600 MHz). Chemical shifts were reported in
ppm with residual undeuterated solvent peaks as internal refer-
ence for 1H NMR: CDCl3 (7.24 ppm), DMSO (2.50 ppm), or D2O
(4.80 ppm). Multiplicities were reported as singlet (s), doublet
(d), triplet (t), or multiplet (m), and coupling constants (J) are given
in Hz. Mass spectra were obtained on Schumadzu GCMSQP 1000
mass spectrometer at 70 eV and recorded herein (relative intensity
and assignment). Flash chromatography was carried out using sil-
ica gel (F60, 230–400 mesh) and thin layer chromatography (TLC)
was conducted on silica gel 60F-254, 0.25 mm pre-coated TLC
plates purchased from E. Merck. TLC plates were visualized using
UV254 and ninhydrin (1.5 g ninhydrin, 100 mL n-butanol, 3 mL
acetic acid) with charring. A purity of more than 95% was deter-
mined for all compounds by NMR using the conditions described
above. Reagents were purchased from commercial sources.
*
*
*
*
0.25
0
Figure 5. Compound 5 activates AMPK and inhibits phosphorylation of mTOR in
Hs578T cells. Hs578T cells were treated with various concentrations of metformin
and 5 for 24 h. Immunoblot analysis was performed on cell lysate⁄s using anti-
phospho AMPK antibody (A) and anti-phospho mTOR antibody (B). Statistically
different from the control at p <0.05.
2500
2000
*
**
1500
1000
500
0
5.1.1. N2,N2-Dimethyl-6-methylthio-1,3,5-triazine-2,4-diamine
(compound 2)
To a stirred solution of dimethyl-N-cyano-dithioiminocarbonate
(compound 1) (5.0 g, 34.2 mmol) in 40% K2CO3 aqueous solution
(35 mL) at room temperature was added a solution of N,N-dimeth-
ylguanidine sulfate (4.6 g, 17.1 mmol) in DMSO (40 mL). After stir-
ring for 2 h at 120 °C, the cooled reaction mixture was poured into
H2O. The white precipitate was filtered and then washed with
MeOH and H2O to provide a compound 2, which was used for next
step without further purification. Yield: 67% (2.1 g).
1H NMR (500 MHz, DMSO-d6) d 6.75 (s, 2H), 3.03 (br s, 6H), 2.37
(s, 3H); 13C NMR (125 MHz, DMSO-d6) d 178.9, 165.5, 164.1, 35.9,
12.6; EIMS (70 eV) m/z (rel intensity): 185 (M+, 100), 170 (20),
139 (20).
0
5
10
15
20
25
30
35 40
Administration days
Figure 6. Compound 5 shows an anti-tumor effect in vivo. Metformin and 5 were
administrated orally to MDA-MB-231 cell-bearing xenograft mice for 35 days.
⁄
⁄⁄
Statistically
Tumor volumes (n = 6) were measured three times
different from the control at p <0.05 and p <0.01, respectively.
a
week.
,
5.1.2. N2,N2-Dimethyl-6-methylsulfinyl-1,3,5-triazine-2,4-
diamine (compound 3)
cells suggests a promising application of this compound for ame-
liorating the progression of breast cancer. Compounds 9a, 9b and
9c, GABA salts of metformin, showed strong inhibitory effects on
proliferation and invasion, implying that the combinatorial synthe-
sis of metformin with GABA or its analogues exerted synergistic
anti-tumoral and anti-invasive activities than the use of metformin
or GABA alone. Taken together, a novel metformin derivative syn-
thesized from this study and metformin combination therapy with
To a solution of compound 2 (2.0 g, 10.8 mmol) in CH2Cl2
(50 mL) at 0 °C was added m-CPBA (2.2 g, 12.9 mmol), and the
reaction mixture was stirred for 12 h at room temperature. The
mixture was diluted with CH2Cl2 and washed with ice water and
5% NaHCO3 aqueous solution, and the organic layer was dried over
anhydrous Na2SO4. The solvent was removed under vacuum to af-
ford the crude product. Further purification by column chromatog-

M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
2311
raphy (MeOH/CH2Cl2 = 5/95 (v/v)) provided a compound 3 as a
white solid. Yield: 77% (1.7 g).
(150 MHz, D2O) d 183.15, 160.22, 158.62, 40.72, 37.60, 35.33,
29.07. Anal (C8H20N6O2) C, 41.4; H, 8.80; N, 36.5; O, 13.8.
1H NMR (500 MHz, DMSO-d6) d 7.28–7.37 (m, 2H), 3.09 (s, 3H),
3.07 (s, 3H), 2.77 (s, 3H); 13C NMR (125 MHz, DMSO-d6) d 181.8,
166.5, 164.6, 36.2; EIMS (70 eV) m/z (rel intensity): 201 (M+, 60),
156 (30), 138 (68), 96 (100).
5.1.7. Metformin pregabalin ((S)-3-(aminomethyl)-5-
methylhexanoic acid) salt (compound 9b)
The title compound was prepared according to the same proce-
dure for compound 9a, using (S)-3-(aminomethyl)-5-methylhexa-
noic acid (compound 8b). Yield: 63% (11.3 g). Mp: 147–148 °C;
1H NMR (600 MHz, D2O) d 2.85 (s, 6H), 2.34 (m, 2H), 1.92 (m,
2H), 1.70 (m, 1H), 1.43 (m, 1H), 1.03 (m, 1H), 0.89 (m, 1H), 0.70
(d, J = 6.6 Hz, 6H); 13C NMR (150 MHz, D2O) d 182.7, 160.2, 158.6,
44.76, 41.55, 41.27, 37.63, 36.03, 24.82, 22.46. Anal (C12H28N6O2)
C, 48.0; H, 9.30; N, 30.2; O, 13.20.
5.1.3. N2,N2-Dimethyl-N4-(2-propylphenyl)-1,3,5-triazine-2,4,6-
triamine (compound 4)
To a solution of compound 3 (2.4 g, 11.9 mmol) in 1,4-dioxane
(30 mL) was added 2-propylaniline (2.5 mL, 17.9 mmol), and the
mixture was stirred at reflux for 15 h. The resulting solution was
evaporated, diluted with CH2Cl2, and washed with 5% NaHCO3
aqueous solution, and the organic layer was dried over anhydrous
Na2SO4 and evaporated under reduced pressure. The crude product
was purified by column chromatography (MeOH/CH2Cl2 = 10/90
(v/v)) to afford compound 4 as a white solid. Yield: 72% (2.3 g).
Mp: 134.8–137.2 °C; 1H NMR (500 MHz, CDCl3) d 7.90–7.92 (m,
1H), 7.18–7.24 (m, 2H), 7.05–7.10 (m, 1H), 6.87 (br s, 1H), 5.33 (br
s, 2H), 3.18 (s, 3H), 3.09 (s, 3H), 2.59 (t, J = 7.4, 2H), 1.63 (q, J = 7.4,
2H), 0.97 (t, J = 7.3, 3H); 13C NMR (125 MHz, CDCl3) d 166.9, 165.8,
164.8, 136.4, 134.0, 129.2, 126.0, 124.2, 123.9, 36.0, 33.3, 22.7,
13.9; EIMS (70 eV) m/z (rel intensity): 272 (M+, 60), 257 (90), 229
(20), 84 (100); HRMS (EI) calcd for C14H20N6: 272.1749, found:
272.1751.
5.1.8. Metformin gabapentin (2-[1-
(aminomethyl)cyclohexyl]acetic acid) salt (compound 9c)
The title compound was prepared according to the same proce-
dure for compound 9a, using 2-[1-(aminomethyl)cyclohexyl]acetic
acid (compound 8c). Yield: 47% (8.81 g). Mp: 122–123 °C; 1H NMR
(600 MHz, D2O) d 2.90 (s, 6H), 2.45 (s, 2H), 2.03 (s, 2H), 1.32–1.20
(m, 10H); 13C NMR (150 MHz, D2O) d 181.68, 160.17, 158.63, 49.02,
48.16, 45.08, 37.61, 36.59, 33.58, 25.97, 21.41. Anal (C13H28N6O2) C,
51.01; H, 9.27; N, 25.77; O, 12.33.
5.1.9. Phenformin acetic acid salt (compound 12)
To a stirred suspension of phenformin hydrochloride (com-
pound 10) (10.2 g, 42.1 mmol) in i-PrOH (50 mL) was added potas-
sium hydroxide (2.54 g, 42.1 mmol), and the reaction mixture was
heated to 50 °C. After stirring at 50 °C for 2 h, the resulting mixture
was filtered and the filter-cake was washed with i-PrOH (250 mL).
The filtrate was concentrated under reduced pressure and dried in
vacuo to provide a phenformin free base (compound 11) as a white
solid. Yield: 87%, (7.48 g).
To a stirred suspension of the compound 11 (7.48 g, 36.5 mmol)
in i-PrOH (50 mL) was added acetic acid (6.25 mL, 109.2 mmol),
and the reaction mixture was heated to 50 °C. After stirring at
50 °C for 2 h, ethyl acetate (50 mL) was added dropwise to the mix-
ture. The resulting precipitate was filtered, and the filter-cake was
washed with i-PrOH (10 mL) and ethyl acetate (50 mL). The solid
residue was dried under vacuum to afford phenformin acetic acid
salt (compound 12) as a white solid. Yield: 80%, (7.75 g).
Mp: 130–131 °C; 1H NMR (600 MHz, D2O) d 7.34 (m, 2H), 7.26
(m, 3H), 3.45 (t, J = 6.6 Hz, 2H), 2.83 (m, 2H), 1.86 (s, 3H); 13C
NMR (150 MHz, D2O) d 181.25, 159.84, 159.72, 138.92, 129.08,
128.90, 126.85, 42.80, 34.78, 23.73. Anal (C12H19N5O2) C, 54.51;
H, 7.07; N, 25.65; O, 10.23.
5.1.4. N2,N2-Dimethyl-N4-(2-propylphenyl)-1,3,5-triazine-2,4,6-
triamine HCl (compound 5)
To a solution of compound 4 (2.0 g, 0.73 mmol) in CHCl3
(20 mL) was added 4 M HCl in dioxane (0.22 mL, 0.88 mmol), and
the mixture was stirred at room temperature for 2 h. The resulting
solution was evaporated and dried under vacuum to afford com-
pound 5 as a white solid. Yield: 95% (2.2 g).
Mp: 145–148 °C; 1H NMR (500 MHz, DMSO-d6) d 9.92 (br s, 1H),
8.06 (br s, 2H), 7.22–7.39 (m, 4H), 3.11 (s, 6H), 2.50 (s, 2H), 1.51 (s,
2H), 0.87 (s, 3H). MS (ESI) m/z 273.7 ([MÀCl]+).
5.1.5. Preparation of metformin free base (compound 7)
A two-necked 500 mL round-bottom flask equipped with a
mechanical stirrer was charged with metformin hydrochloride
(compound 6) (49.8 g, 300 mmol) and i-PrOH (200 mL). Potassium
hydroxide (18 g, 321 mmol) was added to the stirred solution, at
50 °C. The white slurry was stirred at 50 °C for 2 h, and then the
reaction mixture was cooled to room temperature. The resulting
mixture was filtered and the filter-cake was washed with i-PrOH
(30 mL) and acetone (2 Â 50 mL). The combined filtrates were con-
centrated under reduced pressure affording a white solid. The solid
was again dissolved in acetone (300 mL), the insoluble material
was filtered, and the filtrate was concentrate under reduced pres-
sure. The residue was dried in vacuo to provide metformin free
base (compound 7) as a white solid. Yield: 98% (38.2 g).
5.2. Cell culture and preparation of reagents
Hs578T breast cancer cells were purchased from the Korean Cell
Line Bank (KCLB). MDA-MB-231 breast cancer cells were kindly
provided by Dr. Dong Young Noh (Seoul National University, Seoul,
Korea). Hs578T and MDA-MB-231 cells were cultured in DMEM
Mp: 119–120 °C; 1H NMR (600 MHz, D2O) d 3.07 (s, 6H); 13C
NMR (150 MHz, D2O) d 161.1, 158.5, 37.35.
supplemented with 10% FBS and 100 lg/ml penicillin–streptomy-
5.1.6. Metformin GABA salt (compound 9a)
cin. Cells were maintained in humidified atmosphere with 95%
To a stirred suspension of metformin free base (compound 7)
(8.00 g, 61.9 mmol) in i-PrOH (30 mL) was added GABA (compound
8a) (7.02 g, 68.1 mmol), and the reaction mixture was heated to
50 °C. After stirring at 50 °C for 2 h, ethyl acetate (200 mL) was
added dropwise to the mixture. The resulting precipitate was fil-
tered, and the filter-cake was washed with i-PrOH (20 mL) and ace-
tone (50 mL). The solid residue was dried under vacuum to provide
metformin GABA salt (compound 9a) as a white solid. Yield: 76%
(10.9 g).
air and 5% CO2 at 37 °C.
Compounds 5 and 12 were dissolved in DMSO. Metformin,
GABA, 9a, 9b, and 9c were dissolved in distilled water. Preparation
of each reagent is listed in Table 3. Stock solutions were stored at
À20 °C.
5.3. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-
tetrazolium bromide) assay
Mp: 127–128 °C; 1H NMR (600 MHz, D2O) d 2.99 (s, 6H), 2,55 (t,
J = 7.2 Hz, 2H), 2.14 (t, J = 7.8 Hz, 2H), 1.50 (m, 2H); 13C NMR
Cells (5 Â 103) cultured in a 96-well plate were treated with the
indicated drugs for 24 h. After 24 h of incubation, 25 ll of 0.5 mg/

2312
M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
Table 3
metformin and 5 for 24 h with cultured in serum-free media. The
cells (1 Â 106 cells/ml) were added 5
ll of FITC annexin V and/or
Preparation of reagents used for in vitro and in vivo experiments
propidium iodide and analyzed by flow cytometry (Beckman Coul-
ter, UK). The results are represented in the form of dot plots di-
vided into four quadrants. Lower left quadrant of the dot plots
shows viable cells. Lower right quadrant shows early apoptotic
cells with preserved plasma membrane integrity. Upper right/left
quadrants show late apoptotic/necrotic cells which have lost their
plasma membrane integrity.
Compound
In vitro
In vivo
Final Dosage
Vehicle Final Treatment
conc vol (v/v) (%)
(M)
Vehicle
conc vol
(mg/ (ml/kg)
ml)
Metformin D.W.
5
1.5
1
4
0.1
20%
PEG400, 5%
100
20
5
5
DMSO
DMSO in D.W.
GABA
9a
9b
9c
12
D.W.
D.W.
D.W
D.W.
DMSO
1.5
1
1
1
2
4
4
4
4
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
5.8. Anti-tumor activity of 5 in vivo a xenograft tumor model
Four-week-old female BALB/c athymic nude mice were pur-
chased from Orient Bio Co. (Seongnam, Korea). All experiments
were approved and carried out according to the Guide for Care
and Use of Animals (Chungbuk National University Animal Care
Committee, Korea). MDA-MB-231 cells were injected subcutane-
ously (5 Â 106 cells in 0.1 mL serum free medium per mouse) into
the right-lower flanks of the mice. When the tumors had reached
an average volume of 100–130 mm3, the tumor-bearing nude mice
(n = 6) were administered orally with 5 (25, 50 and 100 mg/kg) and
metformin (500 mg/kg) once a day via oral gavage needle. The
preparation of each reagent is listed in Table 3.
0.1
ml of MTT was added and incubated for 4 h. The crystals of forma-
zan which are converted from MTT were dissolved with 100 l of
DMSO. The optical density was measured at 540 nm using a mi-
cro-ELISA reader (Molecular Devices, Sunnyvale, CA) for quantifica-
tion of cell viability. Assays were performed triplicate.
l
5.4. In vitro invasion assay
The body weights and tumor volumes of mice were monitored
three times a week. The tumor volumes were measured with ver-
nier calipers and calculated by the following formula: (A Â B2)/2,
where A is the length and B is the width of the two dimension
tumor.
In vitro invasion assay was performed using 24-well transwell
unit with polycarbonate filters (Corning Costar, Cambridge, MA)
as previously described.28 The lower compartment was filled with
serum-free media containing 0.1% BSA with the indicated drugs.
Cells (5 Â 104 cells) were placed in the upper compartment with
the indicated drugs. After 17 h of incubation, the invasive pheno-
types were determined by counting the cells that had invaded
the lower side of the filter using light microscopy (Olympus
CKX31, Tokyo, Japan) at 400Â optical resolution. Thirteen fields
were counted for each filter and each sample was assayed in
triplicate.
Acknowledgments
This work was supported by the National Research Foundation
of Korea (NRF) Grants funded by the Korean government (Nos.
ROA-2012-0006262 and R11-2007-0056820). This work was partly
supported by HanAll Biopharma Co., Ltd. We thank Dr. Sang Hyup
Lee (Duksung Women’s University, Seoul, Korea) for helpful
discussion.
5.5. Gelatin zymogram assay
Cells were cultured in serum-free media containing metformin
and 5 for 24 h. Gelatinolytic activity of the conditioned medium
was determined by gelatin zymogram assay as described previ-
ously.28 Relative band intensities were determined by quantitation
of each band with a Gel Logic 200 Imaging System (Kodak, Roches-
ter, USA).
References and notes
1. Jalving, M.; Gietema, J. A.; Lefrandt, J. D.; de Jong, S.; Reyners, A. K.; Gans, R. O.;
de Vries, E. G. Eur. J. Cancer 2010, 46, 2369.
2. Lalau, J. D.; Race, J. M. Drugs 1999, 58, 55.
3. Wright, J. L.; Stanford, J. L. Cancer Causes Control 2009, 20, 1617.
4. Jiralerspong, S.; Palla, S. L.; Giordano, S. H.; Meric-Bernstam, F.; Liedtke, C.;
Barnett, C. M.; Hsu, L.; Hung, M. C.; Hortobagyi, G. N.; Gonzalez-Angulo, A. M. J.
Clin. Oncol. 2009, 27, 3297.
5.6. Immunoblot analysis
5. Li, D.; Yeung, S. C.; Hassan, M. M.; Konopleva, M.; Abbruzzese, J. L.
Gastroenterology 2009, 137, 482.
6. Currie, C. J.; Poole, C. D.; Gale, E. A. Diabetologia 2009, 52, 1766.
7. Ben Sahra, I.; Laurent, K.; Loubat, A.; Giorgetti-Peraldi, S.; Colosetti, P.;
Auberger, P.; Tanti, J. F.; Le Marchand-Brustel, Y.; Bost, F. Oncogene 2008, 27,
35761.
Immumoblot analysis was performed as previously described.29
Protein extracts in lysis buffer (0.5% Triton X-100, 0.15 M NaCl,
50 mM Tris–HCl, pH 7.4, 25 mM NaF, 20 mM EGTA, 1 mM DTT,
1 mM Na3VO4) containing protease inhibitor cocktail (Roche,
Mannheim, Germany) were subjected to 10% SDS–PAGE analysis.
8. Liu, B.; Fan, Z.; Edgerton, S. M.; Deng, X. S.; Alimova, I. N.; Lind, S. E.; Thor, A. D.
Cell Cycle 2009, 8, 2031.
Anti-phospho-AMPKa, anti-AMPKa, and anti-phospho-mTOR anti-
9. Kisfalvi, K.; Eibl, G.; Sinnett-Smith, J.; Rozengurt, E. Cancer Res. 2009, 69, 6539.
10. Buzzai, M.; Jones, R. G.; Amaravadi, R. K.; Lum, J. J.; DeBerardinis, R. J.; Zhao, F.;
Viollet, B.; Thompson, C. B. Cancer Res. 2007, 67, 6745.
11. Rattan, R.; Graham, R. P.; Maguire, J. L.; Giri, S.; Shridhar, V. Neoplasia 2011, 13,
483.
12. Jemal, A.; Ward, E.; Thun, M. J. Breast Cancer Res. 2007, 9, R28.
13. Sotiriou, C.; Neo, S. Y.; McShane, L. M.; Korn, E. L.; Long, P. M.; Jazaeri, A.;
Martiat, P.; Fox, S. B.; Harris, A. L.; Liu, E. T. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
10393.
14. Jemal, A.; Siegel, R.; Xu, J.; Ward, E. CA Cancer J. Clin. 2010, 60, 277.
15. Moon, A.; Kim, M. S.; Kim, T. G.; Kim, S. H.; Kim, H. E.; Chen, Y. Q.; Kim, H. R. Int.
J. Cancer 2000, 85, 176.
16. Song, H.; Ki, S. H.; Kim, S. G.; Moon, A. Cancer Res. 2006, 66, 10487.
17. Tan, B. K.; Adya, R.; Chen, J.; Lehnert, H.; Sant Cassia, L. J.; Randeva, H. S. J. Clin.
Endocrinol. Metab. 2011, 96, 808.
bodies were purchased from Cell Signaling Technology, Inc. (Bev-
erly, MA, USA). Anti-MMP-2 antibodies were purchased from
R&D system Inc. (Minneapolis, MN, USA). The enhanced chemilu-
minescence system (ECL, Amersham-Pharmacia, UK) was used for
detection. Relative band intensities were determined by quantita-
tion of each band with a Gel Logic 200 Imaging System (Kodak,
Rochester, USA).
5.7. Flow cytometric analysis for apoptosis
Flow cytometric analysis was performed using FITC annexin V
apoptosis detection kit (BD Biosciences, San Diego, CA) as manufac-
ture’s protocol. MDA-MB-231 breast cancer cells were exposed to
18. Hwang, Y. P.; Jeong, H. G. Br. J. Pharmacol. 2010, 160, 1195.
19. Vazquez-Martin, A.; Oliveras-Ferraros, C.; del Barco, S.; Martin-Castillo, B.;
Menendez, J. A. Ann. Oncol. 2009, 20, 592.

M. Koh et al. / Bioorg. Med. Chem. 21 (2013) 2305–2313
2313
20. Lea, M. A.; Chacko, J.; Bolikal, S.; Hong, J. Y.; Chung, R.; Ortega, A.; Desbordes, C.
Anticancer Res. 2011, 31, 421.
F. M.; McCormick, F.; Dickson, R. B.; Johnson, M. D.; Lippman, M.; Ethier, S.;
Gazdar, A.; Gray, J. W. Cancer Cell 2006, 10, 515.
21. Appleyard, M. V.; Murray, K. E.; Coates, P. J.; Wullschleger, S.; Bray, S. E.;
Kernohan, N. M.; Fleming, S.; Alessi, D. R.; Thompson, A. M. Br. J. Cancer 2012,
106, 1117.
22. Bryans, J. S.; Wustrow, D. J. Med. Res. Rev. 1999, 19, 149.
23. Young, S. Z.; Bordey, A. Physiology (Bethesda) 2009, 24, 171.
24. Neve, R. M.; Chin, K.; Fridlyand, J.; Yeh, J.; Baehner, F. L.; Fevr, T.; Clark, L.;
Bayani, N.; Coppe, J. P.; Tong, F.; Speed, T.; Spellman, P. T.; DeVries, S.; Lapuk,
A.; Wang, N. J.; Kuo, W. L.; Stilwell, J. L.; Pinkel, D.; Albertson, D. G.; Waldman,
25. Joseph, J.; Niggemann, B.; Zaenker, K. S.; Entschladen, F. Cancer Res. 2002, 62,
6467.
26. Schuller, H. M.; Al-Wadei, H. A.; Majidi, M. Cancer 2008, 112, 767.
27. Lin, J. H.; Lu, A. Y. Pharmacol. Rev. 1997, 49, 403.
28. Kim, M. S.; Lee, E. J.; Choi Kim, H. R.; Moon, A. Cancer Res. 2003, 63, 5454.
29. Shin, I. C.; Kim, S. H.; Song, H.; Choi Kim, H. R.; Moon, A. J. Biol. Chem. 2005, 280,
14675.