Synthesis and biological evaluation of glyco-GA compounds as anticancer agents

Article abstract of DOI:10.1016/j.cclet.2012.01.019

A series of novel glyco-gambogic acid (GA) compounds were synthesized and evaluated for their in vitro anti-proliferative activity against human hepatocellular carcinoma (HCC) cells. All compounds showed much better aqueous solubility (0.92-1.89 mg/mL) than GA (0.013 mg/mL), and displayed potent inhibition on HCC cells (IC₅₀: 0.21-12.23 μmol/L) and little affects on non-tumor liver cells (IC₅₀: 42.56-86.43 μmol/L), suggesting that glyco-GA compounds selectively inhibit HCC proliferation, and may be promising candidates for further intensive study.

Full text of DOI:10.1016/j.cclet.2012.01.019

Available online at www.sciencedirect.com
Chinese Chemical Letters 23 (2012) 383–386
www.elsevier.com/locate/cclet
Synthesis and biological evaluation of glyco-GA
compounds as anticancer agents
Li Qin He a,b, Zhong Hu b, La Mei Yuan b, Xiao Shan Wang b, Yi Hua Zhang a,
*
a Center of Drug Discovery, China Pharmaceutical University, Nanjing 210009, China
b Anhui University of Traditional Chinese Medicine, Anhui Key Laboratory of Modernized Chinese Material Medical, Hefei 230031, China
Received 29 September 2011
Available online 3 March 2012
Abstract
A series of novel glyco-gambogic acid (GA) compounds were synthesized and evaluated for their in vitro anti-proliferative
activity against human hepatocellular carcinoma (HCC) cells. All compounds showed much better aqueous solubility (0.92–
1.89 mg/mL) than GA (0.013 mg/mL), and displayed potent inhibition on HCC cells (IC50: 0.21–12.23 mmol/L) and little affects on
non-tumor liver cells (IC50: 42.56–86.43 mmol/L), suggesting that glyco-GA compounds selectively inhibit HCC proliferation, and
may be promising candidates for further intensive study.
# 2012 Yi Hua Zhang. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
Keywords: Gambogic acid; Glyco-GA compounds; Anti-hepatocellular carcinoma activity
Gambogic acid (GA, 1) is one of the major active ingredient of gamboge [1,2], which was used as a dye or folk
medicine for an internal purgative and externally infected wounds [3]. Recent studies have demonstrated that GA
possesses cytotoxicity against various human cancer cell lines, such as gastric carcinoma, hepatoma, and lung cancer
cells, and displays inhibitory activity in tumor-bearing mouse models [4–7]. In addition, GA is not able to adversely
affect the number of white blood cells (WBC) in blood and akaryote in marrow of rats [8]. Unfortunately, the aqueous
solubility (0.013 mg/mL) of GA is very low, thus limiting its clinical application.
Structure-activity relationships (SARs) studies have shown that 30-carboxy group of GA can tolerate a variety of
modifications with no or little effects on its bioactivity [9]. Additionally, a large variety of studies have demonstrated
that introduction of carbohydrate moiety to a molecule usually improves its water solubility and cell penetrations, and
enhances selectivity and bioactivity through intra/intercellular carbohydrate-protein interaction [10–13]. Accordingly,
a series of novel glycol-derivatives of GA were therefore designed and synthesized by coupling various hydrophilic
glycosyl groups to the 30-carboxyl of GA. After structural characterization, these target compounds were evaluated for
their water solubility and inhibitory activity against HCC cell proliferation.
The starting material GAwas obtained by isolation from gamboges and further purification, as described previously
[1]. Coupling of 30-COOH of GA with O-acetylated glycosyl bromides, which were generated by treatment of
corresponding mono- or di-saccharides with Ac2O–CH3COBr–MeOH [14], in the presence of K2CO3 and cetyl
* Corresponding author.
E-mail address: zyh@sohu.com (Y.H. Zhang).
1001-8417/$ – see front matter # 2012 Yi Hua Zhang. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
doi:10.1016/j.cclet.2012.01.019

384
L.Q. He et al. / Chinese Chemical Letters 23 (2012) 383–386
COOR
1
COOR
O
2
COOH
O
O
O
O
O
O
ii
i
O
O
O
O
O
O
O
OH
O
OH
OH
1
3a-d
2a-d
HO
AcO
HO
AcO
O
OH
OH
OH
O
OR =
2
OAc
OAc
OAc
3a
3b
3c
OR =
1
O
2a
O
HO
AcO
HO
O
AcO
OR =
2
OR =
1
2b
2c
O
O
O
OH
OAc
HO
HO
AcO
HO
O
OH
AcO
AcO
O
HO
AcO
OH
OH
OAc
OAc
O
O
O
OR =
2
O
O
OR =
O
1
OAc
OH
OAc
HO
HO
AcO
HO
O
OH
AcO
AcO
O
OAc
HO
AcO
O
OR =
2
3d
O
O
2d OR =
O
O
1
O
OH
OAc
Scheme 1. The synthetic routes of 3a–d. Regents and conditions: (i) Tetra-O-acetyl-a-D-glucopyranosyl bromide, tetra-O-acetyl-a-D-galactopyr-
anosyl bromide, hepta-O-acetyl-a-D-maltosyl bromide or hepta-O-acetyl-a-D-lactosyl bromide, K2CO3, CTAB, H2O–CH2Cl2, r.t., 48 h, in 70–80%
yields; (ii) 25% NaOH, CH3COCH3, 0–10 8C, 20–30 min, in 50–72% yields.
trimethyl ammonium bromide (CTAB) in H2O and CH2Cl2 gave acetylated esters 2a–d in 70ꢀ80% yields. Attempted
deacetylation of 2a–d with sodium methoxide in anhydrous methanol failed [15]. However, modified procedure using
25% sodium hydroxide in acetone successfully provided the target compounds 3a–d in 50–72% yields, as shown in
Scheme 1.
1
The structures of 3a–d were characterized by spectra of IR, MS, H NMR, and elemental analyses [16].
The aqueous solubility of glyco-GA compounds was determined as reported previously [17]. As shown in Table 1,
all GA derivatives 3a–d showed an increase in water solubility ranging from 0.92 to 1.89 mg/mL, which are 70- to
145-fold more soluble than the parent GA.
All target compounds were evaluated for their anti-proliferative effects on both human HCC cells and non-tumor
cells in vitro by MTT assay [18], using GA as control (Table 2). The results indicated that anti-proliferative activity of
3a–d against five HCC cell lines was largely comparable to or even stronger than that of GA. In sharp contrast, all of
Table 1
Solubility of the target compounds in water.
Compound
3a
3b
3c
3d
GA
Solubility (mg/mL)
0.92
0.95
1.78
1.89
0.013
Table 2
Anti-proliferative effects of GA and 3a–d on both human HCC cells and normal liver cells HL-7702.
Compound
IC50 (mmol/L)a
HL-7702
BEL-7402
SMMC-7721
Bel-7404
QGY-7701
HepG2
GA
3a
3b
3c
3.42
86.43
79.65
44.83
42.56
0.75
1.12
1.13
2.43
4.70
3.02
3.60
7.85
8.03
0.24
0.21
0.28
4.89
5.13
1.17
2.01
1.50
2.84
2.21
10.68
12.23
10.41
10.92
9.86
3d
10.52
a
IC50: a concentration required to inhibit tumor cell proliferation by 50%. Data are expressed as the mean IC50 from the dose-response curves of at
least three independent experiments.

L.Q. He et al. / Chinese Chemical Letters 23 (2012) 383–386
385
them, especially for the most potent compounds 3a and 3b, displayed only little inhibition on proliferation of human
normal liver cells HL-7702 (IC50: 86.43 and 79.65 mmol/L vs 3.421 mmol/L), indicating that glyco-GA compounds
selectively inhibited HCC cells.
The above data suggest that the introduction of glycosyl group(s) to the carboxyl group of GA may improve
aqueous solubility and selective anti-proliferative effect of GA on human HCC cells. It was observed that the
monosaccharide derivatives (3a and 3b) had stronger inhibitory activity on human HCC cells and much lower
inhibitory effects on non-liver cancer cell than disaccharide derivatives (3c and 3d), probably due to that high levels of
glucose transporter proteins and asialoglycoprotein receptors (ASGPR) are expressed in human HCC cells [11,19] and
they may interact with those glucosyl and galactosyl derivatives by recognization and endocytosis, leading to
preferable entry of 3a and 3b into HCC cells.
In summary, a new class of glyco-GA compounds were synthesized, and all of them exhibited better aqueous
solubility than GA, and showed potent inhibitory effects on proliferation of human HCC cells. More importantly, they
displayed little inhibition on non-HCC human liver cells. The most potent compounds 3a and 3b may be promising
candidates for further intensive study. As a result, our novel findings provide a novel framework for the design of new
glyco-GA compounds for the intervention of human HCC cells.
Acknowledgment
This study was supported by a grant from the Nature and Science Foundation of Department of Education, Anhui
province (No. KJ2010A204).
References
[1] W.D. Ollis, M.V.J. Ramsay, I.O. Sutherland, Tetrahedron 21 (1965) 1453.
[2] S.A. Ahmad, W. Rigby, R.B. Taylor, J. Chem. Soc. C (1966) 772.
[3] K. Venkataraman, Proc. Ind. Natl. Acad. Sci. U.S.A. 39A (1973) 365.
[4] W. Liu, Q.L. Guo, Q.D. You, et al. World J. Gastroenterol. 11 (2005) 3655.
[5] Z.Q. Wu, Q.L. Guo, Q.D. You, et al. Chin. J. Nat. Med. 1 (2003) 99.
[6] Q.L. Guo, Q.D. You, Z.Q. Wu, et al. Acta Pharmacol. Sin. 25 (2004) 769.
[7] Z.Q. Wu, Q.L. Guo, Q.D. You, et al. Biol. Pharm. Bull. 27 (2004) 1769.
[8] Z.L. Guo, Q.L. Wu, Z.Q. You, et al. Chin. J. Nat. Med. 1 (2003) 229.
[9] H.Z. Zhang, S. Kasibhatla, Y. Wang, et al. Bioorg. Med. Chem. 12 (2004) 309.
[10] J. Ramirez, L.B. Yu, J. Li, et al. Med. Chem. Lett. 6 (1996) 2575.
[11] Y. Hou, X.J. Wu, W.H. Xie, et al. Tetrahedron Lett. 42 (2001) 825.
[12] G. Cantuaria, A. Magalhaes, R. Angioli, et al. Cancer 88 (2000) 381.
[13] A. Wong, I. Toth, Curr. Med. Chem. 8 (2001) 1123.
[14] T.J. Ritchie, S.J.F. Macdonald, R.J. Young, et al. Drug Discov. Today 16 (2011) 164.
[15] A.K. Ghose, T. Herbertz, J.M. Salvino, et al. Drug Discov. Today 11 (2006) 1107.
[16] General procedure for the synthesis of the target compounds 3a–d: compound 1 (628 mg, 1 mmol) was dissolved in the mixture of 15 mL
CH2Cl2 and 20 mL H2O in the presence of K2CO3 (207 mg, 1.5 mmol) and CTAB (91 mg, 0.25 mmol). The mixture was vigorously stirred and
reacted with the corresponding O-acetylated glycosyl bromides (1.2 mmol) in several fractions at RT for 48 h. The organic layer was harvested
and the remaining organic solvents in the aqueous layer were extracted with CH2Cl2 (10 mL Â 3), dried over sodium sulfate, and concentrated
in vacuo to obtain oil-like materials, which was subsequently purified by column chromatography using (PE/EtOAc = 3:1) to give pure 2a–c in
70–80% yields. The resulting materials were dissolved in acetone on ice and its pH was adjusted to 9.0–10.0 with 25% sodium hydroxide. The
deacetylation was monitored by TLC (1: 6, v/v, MeOH–CH2Cl2) and its pH was then adjusted to 5.0 with 10% HCl. After filtration, the filtrate
was evaporated in vacuo and the resulting residue was extracted with EtOAc (10 mL Â 3), dried over sodium sulfate, purified by column
chromatography (EtOAc to MeOH–CH2Cl2 1:10, v/v) to give the title compounds (50–72%). Analytical data for compounds 3a–d. Compound
3a: yield 72%, yellow solid, mp: 134.8–135.0 8C. IR (KBr): 3304, 2930, 2866, 1735, 1715, 1626, 1596, 1456, 1070 cmÀ1; MS (ESI, m/z):
789[MÀH]À; 1H NMR (500 MHz, DMSO-d6): d 11.93 (s, C6-OH), 7.62 (d, 1H, J = 6.7 Hz, C10H), 6.68 (d, 1H, J = 9.88 Hz, C4H), 6.57 (s, 1H,
C27H), 5.66 (d, 1H, J = 7.82 Hz, sugar-H), 5.49 (d, 1H, J = 10 Hz, C3H), 5.32 (m, 1H), 5.06 (m, 4H), 4.39 (s, 1H), 3.83 (m, 1H, sugar-H), 3.67
(d, 1H, J = 11.6 Hz, sugar-H), 3.39 (m, 2H, sugar-H), 3.20 (m, 4H), 2.96 (m, 1H), 2.90 (m, 1H), 2.52 (m, 1H), 2.30 (m, 1H), 2.08–2.01 (m, 2H),
2.00 (s, 3H), 1.90 (m, 1H), 1.81 (m, 1H), 1.75 (s, 3H), 1.68 (s, 3H), 1.58 (s, 3H), 1.43 (m, 1H), 1.38 (s, 3H), 1.31 (s, 3H), 1.27 (m, 1H), 1.17 (s,
3H). Anal. Calcd. for C44H54O13: C 66.84, H 6.84; found: C 66.78, H 6.91. Compound 3b: yield 70%, yellow solid, mp: 137.0–138.4.0 8C. IR
(KBr): 3283, 2925, 2848, 1732, 1711, 1628, 1586, 1452, 1069 cmÀ1; MS (ESI, m/z): 789[MÀH]À; 1H NMR(500 MHz, DMSO-d6): d 11.93 (s,
C6-OH), 7.60 (d, 1H, J = 6.8 Hz, C10H), 6.61 (d, 1H, J = 9.82 Hz, C4H), 6.55 (s, 1H, C27H), 5.61 (d, 1H, J = 8.12 Hz, sugar-H), 5.48 (d, 1H,
J = 10 Hz, C3H), 5.10 (m, 1H), 4.92 (m, 1H), 4.39 (s, 2H), 4.10 (s, 1H), 4,05 (s, 1H), 3.95 (m, 1H, sugar-H), 3.73 (m, 1H, sugar-H), 3.65(m, 1H,
sugar-H), 3.47(m, 2H, sugar-H), 3.36 (m, 1H, sugar-H), 3.20 (m, 2H), 2.91 (m, 2H), 2.64 (m, 1H), 2.51 (d, 1H), 2.35(m, 1H), 2.19 (s, 3H), 2.01

386
L.Q. He et al. / Chinese Chemical Letters 23 (2012) 383–386
(m, 1H), 1.90 (m, 1H), 1.83 (s, 3H), 1.75 (s, 3H), 1.68 (s, 3H), 1.57 (m, 1H), 1.45 (s, 3H), 1.42 (m, 1H), 1.38 (s, 3H), 1.29 (s, 3H), 1.26 (m, 1H),
1.14 (s, 3H). Anal. Calcd. for C44H54O13: C 66.84, H 6.84; found: C 66.82, H 6.86. Compound 3c: yield 54%, yellow solid, mp: 164.5–
166.1 8C. IR (KBr): 3307, 2925, 2876, 1735, 1715, 1627, 1593, 1456, 1174, 1045 cmÀ1; MS (ESI, m/z): 951 [MÀH]À; 1H NMR (500 MHz,
DMSO-d6): d 11.93 (s, C6-OH), 7.62 (d, 1H, J = 6.7 Hz, C10H), 6.62 (d, 1H, J = 10 Hz, C4H), 6.41 (s, 1H, C27H), 5.74 (d, 1H, J = 7.82 Hz,
sugar-H), 5.58 (d, 1H, J = 10 Hz, C3H), 5.42 (d, 1H, J = 9.1 Hz, sugar-H), 5.35 (d, 1H, J = 9.8 Hz, C37H), 5.11 (d, 1H, J = 4.5 Hz), 5.06 (d, 1H,
J = 5.5 Hz), 4.97 (d, 1H, J = 4.7 Hz), 4.80–4.89 (m, 3H), 4.39 (m, 1H), 4.26 (m, 1H, sugar-H), 4.18 (d, 1H, J = 11.6 Hz, sugar-H), 4.06 (m, 1H,
sugar-H), 3.76 (m, 1H, sugar-H), 3.46 (m, 1H, sugar-H), 3.16 (m, 4H), 2.76 (m, 1H), 2.52 (m, 1H), 2.30 (m, 1H), 2.08–2.01 (m, 2H), 1.96 (s,
3H), 1.84 (m, 1H), 1.80 (m, 1H), 1.73 (s, 3H), 1.65 (s, 3H), 1.54 (s, 3H), 1.41 (m, 1H), 1.32 (s, 3H), 1.29 (s, 3H), 1.25 (m, 1H), 1.16 (s, 3H). Anal.
Calcd. for C50H64O18: C 63.03, H 6.72; found: C 62.91, H 6.85. Compound 3d: yield 50%, yellow solid, mp: 174.2–175.5 8C. IR (KBr): 3306,
2932, 2878, 1732, 1713, 1627, 1594, 1457, 1072 cmÀ1; MS (ESI, m/z): 951 [MÀH]À; 1H NMR (500 MHz, DMSO-d6): d 11.95 (s, C6-OH), 7.64
(d, 1H, J = 6.8 Hz, C10H), 6.60 (d, 1H, J = 9.8 Hz, C4H), 6.52 (s, 1H, C27H), 5.49 (d, 1H, J = 8.12 Hz, sugar-H), 5.45 (d, 1H, J = 10 Hz, C3H),
5.35 (d, 1H, J = 9.1 Hz, sugar-H), 5.32 (d, 1H, J = 9.8 Hz, C37H), 5.12 (d, 1H, J = 4.3 Hz), 5.06 (d, 1H, J = 5.6 Hz), 4.83 (d, 1H, J = 4.8 Hz),
4.65 (m, 1H), 4.52 (d, 1H, J = 4.6 Hz), 4.47 (t, 1H, J = 5.8 Hz), 4.28 (m, 2H), 4.17 (m, 2H), 3.95 (m, 2H, sugar-H), 3.72 (m, 3H, sugar-H),
3.65(m, 2H, sugar-H), 3.47(m, 2H, sugar-H), 3.36 (m, 2H, sugar-H), 3.30 (m, 1H), 3.10 (m, 2H), 2.64 (m, 1H), 2.51 (d, 1H), 2.27(m, 1H), 2.16
(s, 3H), 2.01 (m, 1H), 1.96 (m, 1H), 1.83 (s, 3H), 1.77 (s, 3H), 1.69 (s, 3H), 1.58 (m, 1H), 1.46(s, 3H), 1.43(m, 1H), 1.37 (s, 3H), 1.29 (s, 3H),
1.26 (m, 1H), 1.14 (s, 3H). Anal. Calcd. for C50H64O18: C 63.03, H 6.72; found: C 62.94, H 6.79.
[17] S. Hess, M.A. Akermann, S. Wnendt, et al. Bioorg. Med. Chem. 9 (2001) 1279.
[18] C.S. Cheung, K.K. Chung, J.C. Lui, et al. Cancer Lett. 253 (2007) 224.
[19] S.L. Wang, F.B. Yu, T.Y. Jiang, et al. Drug Target. 16 (2008) 233.