Regioselective synthesis and initial evaluation of a folate receptor targeted rhaponticin prodrug

Article abstract of DOI:10.1016/j.cclet.2012.08.006

To improve the therapeutic effect of rhaponticin (RHA), a folate receptor (FR) targeted RHA prodrug was designed and regioselectively synthesized by utilizing a hydrophilic peptide spacer linked to folic acid (FA) via a releasable disulfide linker. A series of biological evaluation was investigated in vitro and in vivo. The positive results of biological investigations warrant further preclinical study before this novel targeted chemotherapeutic is considered for clinical investigation.

Full text of DOI:10.1016/j.cclet.2012.08.006

Available online at www.sciencedirect.com
Chinese Chemical Letters 23 (2012) 1133–1136
www.elsevier.com/locate/cclet
Regioselective synthesis and initial evaluation of a folate receptor
targeted rhaponticin prodrug
Xu Hua Liang a, Yang Sun b, Lu Sha Liu a, Ying Yong Zhao c, Xiao Yun Hu d, Jun Fan a,
*
a School of Chemical Engineering, Northwest University, Xi’an 710069, China
b Department of Chemistry & Chemical Engineering, Xi’an University of Arts and Science, Xi’an 710065, China
c Biomedicine Key Laboratory of Shaanxi Province, Northwest University, Xi’an 710069, China
d Department of Physics, Northwest University, Xi’an 710069, China
Received 20 June 2012
Available online 21 September 2012
Abstract
To improve the therapeutic effect of rhaponticin (RHA), a folate receptor (FR) targeted RHA prodrug was designed and
regioselectively synthesized by utilizing a hydrophilic peptide spacer linked to folic acid (FA) via a releasable disulfide linker. A
series of biological evaluation was investigated in vitro and in vivo. The positive results of biological investigations warrant further
preclinical study before this novel targeted chemotherapeutic is considered for clinical investigation.
# 2012 Jun Fan. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
Keywords: Rhaponticin (RHA); Targeted therapeutics; Folate; Prodrug; Anticancer
Rhaponticin (RHA), a major representative of the stilbene glucoside compounds, exists widely in medicinal plant
of Rheum L. such as Rheum officinale, Rheum undulatum, and Rheum palmatum [1]. Previous studies revealed that
RHA showed potent antitumor and antitumor-promoting, antithrombotic, and antioxidant activities [2]. However, as
an anticancer drug, it kills healthy cells along with cancerous ones. As a consequence, severe side effects were
observed, and therapeutic doses become limited due to the risk of severe toxicity. In addition, RHA was rapidly
eliminated from the systemic circulation [3]. Considering these problems, there is an urgent need to design it into novel
anticancer prodrugs that concentrate their lethal activities specifically in cancer tissues. Receptor-specific targeting, as
one approach, can potentially satisfy the selective delivery criteria for toxic agents to pathologic cells. The folate
receptor (FR) has been considered as a very attractive molecular target for tumor selective drug delivery, since it is
over-expressed on a wide variety of tumor cells, but highly restricted in most normal human tissues [4,5]. Moreover,
the vitamin folic acid (FA) binds to FR with high affinity (Kd = 10À10 mol/L) [6].
In this paper, we report the design and regioselective synthesis of a FA–RHA prodrug 1 (Scheme 1). As indicated in
Scheme 1, conjugate 1 can be assembled by tethering the corresponding thiol-containing FA-spacer 2 to the thiol-
reactive derivative of RHA 3. As has been reported before [7], the disulfide bond is important for drug delivery
application since reduction-mediated release of the drug cargo from a disulfide linked FA-conjugate efficiently occurs
within the endosomes of cancer cells.
* Corresponding author.
E-mail address: fanjun@nwu.edu.cn (J. Fan).
1001-8417/$ – see front matter # 2012 Jun Fan. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
http://dx.doi.org/10.1016/j.cclet.2012.08.006

1134
X.H. Liang et al. / Chinese Chemical Letters 23 (2012) 1133–1136
H2N NH
HN
O
OH
O
OH
S
O
O
O
H
N
H
N
H
N
O
O
N
N
H
N
H
N
H
S
OH
N
N
O
O
O
HN
H2N
N
H
O
OH
HO
HO
OH
O
OH
O
O
OH
1
H2N NH
HN
H
O
OH
O
OH
O
N
O
O
H
O
PySS
OH
H
N
N
N
SH
O
O
N
N
H
N
H
O
HO
HO
H
OH
O
N
N
O
O
+
OH
O
HN
H2N
N
H
OH
O
OH
3
2
Fmoc SPPS
H2N CO2Resin
SMtt
H2N
OH
N
HO
N
N
O
O
O
OH
+
OH
SSPy
HO
O
O
O
4
HO
OH
I
OH
HO
O
HO
HO
O
OH
HO
OH
O
OH
O
O
O
OH
Scheme 1. The retrosynthetic analysis of the FR targeted RHA prodrug (conjugate 1).
The peptide-based derivative FA-spacer 2 was designed as a novel bifunctional molecular spacer unit containing
both acidic (Asp) and basic (Arg) amino acids to provide the best water-solubility of the FA–RHA conjugates under
physiological conditions. Pteroic acid was to serve as N-terminus, whereas the thiol group of cysteine was to serve as
the attachment site for the cleavable linker. These concepts allowed for the assembly of the spacer unit using standard
fluorenylmethyloxycarbonyl-based solid phase peptide synthesis (Fmoc SPPS) as described in previous reports [7,8].
Activated carbonate 4 served as an important heterobifunctional crosslinker for the releasable drug conjugate synthesis
[9], and can be reacted under mild conditions with many N-and O-nucleophiles according to Vlahov [7]. Thiol-reactive
derivative of RHA 3 was prepared in three steps from RHA which was isolated by our laboratory. Briefly, as shown in
Scheme 2, RHA was treated with (PhO)3P-CH3I in DMF, followed by ammonolysis in an autoclave to yield amido-
reactive derivative of RHA 5. Compound 5 was treated with activated carbonate 4 and diisopropylethylamine (DIPEA)
in DCM to yield 3, followed by chromatographic purification on silica gel, 3 was isolated in 80% yield. Treatment of a
suspension of FA-spacer 2 in PBS under argon with NaHCO3 resulted in a clear yellow solution at pH > 6.5. This
mixture was added at once under vigorous stirring to a solution of 3 in THF. The HPLC purification [7] gave pure final
1
conjugate 1. H NMR, 13C NMR and HRMS signals were in agreement with the expected structure [10].
Treatmentofconjugate1withareducingagentdemonstratedthereleaseoffreeRHA.Briefly, a50 mmol/Lsolutionof
conjugate 1 in PBS (pH 7.4) was treated with 2 mmol/L L-glutathione (GSH) at 37 8C. The HPLC-MS profile (UV
detection at 324 nm) (Fig. 1) showed cleavage of the disulfide bond with concomitant release of RHA derivative 3.
Subsequently, 3 fragments resolved into free RHA [ESI (M+H)+: 421], thus demonstrating self-immolative nature of the
linker system [11]. Efficient liberation of parental drug upon exposure to GSH was encouraging, indicating that the
release of RHA should proceed spontaneously following disulfide bond reduction within the endosomes.
In this study, HeLa cells were used to investigate the cytotoxicity of conjugates 1 in comparison with the parental
RHA by utilizing the MTT proliferation assay [12]. HeLa cells were incubated at 37 8C for 48 h with increasing
concentration of conjugate 1 or parental RHA in the presence or absence of 0.1 mmol/L FA (as a competitor). As
shown in Fig. 2, it was obvious that the cell viability decreased with increasing of drug concentration. It was also
showed that conjugate 1 were much more toxic against HeLa cells with an IC50 of about 9.7 nmol/L against 34 nmol/L
for parental RHA. While in the presence of excess free FA, the activity of conjugate 1 was effectively blocked but kept

X.H. Liang et al. / Chinese Chemical Letters 23 (2012) 1133–1136
1135
HO
HO
OH
H2N
HO
HO
OH
i, (PhO)3P-CH3I
18h, 25 , DMF
O
O
OH
O
HO
O
OH
O
O
OH
ii, NH3H2O
DMF, 12h
OH
5
4, DCM
2, H2O, THF
pH 6.9, RT
15min, 70%
1
3
DIPEA, RT
1h, 80%
Scheme 2. The preparation of conjugate 1 from RHA.
500
1
1 in PBS
treated with GSH
t = 0 min
400
300
200
100
0
3
1 h
4 h
RHA
0
2
4
6
8
10
Time / min
Fig. 1. Conjugate 1 in PBS buffer treated with GSH at 37 8C.
almost the same for parental RHA. These results indicated from one certain aspect that the observed activity of
conjugate 1 was folate-mediated but the activity of RHA had no relationship with FA.
In this study, human serum albumin was utilized to evaluate the in vitro protein binding properties of conjugate 1.
As a result, when the concentration of conjugate 1 was 8.0 mmol/L, the binding rate achieved maximum 79.3 Æ 2.5%.
The stability of conjugate 1 and parental RHA in human serum and the blood clearance were also investigated. Fig. 3
revealed that conjugate 1 decreased with time in the serum matrix with an estimated t1/2 of ꢀ10 h against ꢀ3.5 h for
parental RHA. While as shown in Fig. 4, after intravenous injection to mice (Balb/c mice, treated with FR-positive KB
cells), compound 1 was rapidly removed from systemic circulation. Taken together, the data in Figs. 2–4 suggested
that conjugate 1 is possible to reach an FR-positive tumor in an intact form within minutes following an intravenous
injection, and the further research is ongoing.
In this study, a FR targeted RHA prodrug was successfully regioselectively synthesized by utilizing a hydrophilic
peptide spacer linked to FA via a releasable disulfide linker. Following positive in vitro and in vivo results and
toxicological evaluation, compound 1 was selected as a candidate for further preclinical or clinical development and
the further research is ongoing.
1.0
0.8
0.6
0.4
conjugate 1
free RHA
0.2
0.0
conjugate 1 +free FA
free RHA +free FA
-10
-9
-8
-7
-6
Concentration (logM)
Fig. 2. Viability of HeLa cells after exposure to conjugate 1 and parental RHA in the presence or absence of FA (as a competitor) at 37 8C.

1136
X.H. Liang et al. / Chinese Chemical Letters 23 (2012) 1133–1136
100
80
60
40
20
Conjugate 1
Parental RHA
0
0
5
10
15
20
25
Serum incubation time /h
Fig. 3. Serum stability of conjugate 1 in human serum at 37 8C (n = 3 with %RSD < 2%).
100
Conjugate 1
80
Parental RHA
60
40
1.5
1.0
0.5
0.0
0
100
200
300
400
Time post injection /min
Fig. 4. Blood clearance of conjugate 1 following an 1800 nmol/kg intravenous injection dose to treated Balb/c mice (n = 3).
Acknowledgment
This work was supported by the National Scientific Foundation of China (No. 81001622).
References
[1] R. Zhang, K.A. Kang, M.J. Piao, et al. J. Toxicol. Environ. Health 70 (2007) 1155.
[2] H. Hibasami, K. Takagi, T. Ishii, et al. Oncol. Rep. 18 (2007) 347.
[3] Y.Y. Zhao, L.L. Wang, H. Miao, et al. Chromatographia 73 (2011) 599.
[4] H. Elnakat, M. Ratnam, Adv. Drug Deliv. Rev. 56 (2004) 1067.
[5] H. Shmeeda, L. Mak, D. Tzemach, et al. Mol. Cancer Ther. 5 (2006) 818.
[6] C.P. Leamon, J.A. Reddy, I.R. Vlahov, et al. Bioconjugate Chem. 16 (2005) 803.
[7] I.R. Vlahov, H.K.R. Santhapuram, P.J. Kleindl, et al. Bioorg. Med. Chem. Lett. 16 (2006) 5093.
[8] J.A. Reddy, E. Westrick, C.P. Leamon, Cancer Chemother. Pharmacol. 58 (2006) 229.
[9] I.R. Vlahov, H.K. Santhapuram, F. You, et al. J. Org. Chem. 75 (2010) 3685.
[10] Spectra data for compound 1: 1H NMR (300 MHz, D2O): d 8.67 (s, 1H), 7.93–8.23 (m, 6H), 7.35 (d, 2H, J = 7.8 Hz), 6.96 (d, 2H, J = 6.78 Hz),
6.88 (d, 1H, J = 8.31 Hz), 6.75 (s, 1H), 6.64 (d, 1H, J = 7.58 Hz), 6.51 (d, 2H, J = 8.7 Hz), 6.45 (s, 2H), 6.13 (s, 1H), 5.88 (d, 1H, J = 5.69 Hz),
4.8–4.82 (m, 3H), 4.47 (t, 2H, J = 5.37 Hz), 4.35 (t, 2H, J = 7.21 Hz), 4.30 (s, 2H), 4.25 (m, 1H), 4.16 (s, 1H), 3.94 (d, 2H, J = 6.28 Hz), 3.85 (t,
1H, J = 5.83 Hz), 3.73 (s, 3H), 3.43 (d, 2H, J = 5.90 Hz), 3.17 (d, 2H, J = 7.86 Hz), 3.10 (d, 2H, J = 7.83 Hz), 3.02 (m, 3H), 2.93 (t, 2H,
J = 7.21 Hz), 2.88 (d, 2H, J = 7.03 Hz), 2.72 (m, 3H), 2.63 (m, 2H), 2.41 (m, 4H), 2.00–2.30 (m, 10H), 1.69 (m, 4H); 13C NMR (300 MHz,
D2O): d 176.8, 175.2, 174.1, 167.3, 166.3, 161.7, 160.2, 158.8, 153.9, 150.7, 149.5, 148.4, 146.6, 142.8, 138.2, 136.7, 130.0, 127.9, 124.8,
121.9, 119.8, 115.4, 114.3, 111.2, 105.3, 104.1, 100.6, 98.9, 71.4, 69.5, 68.7, 67.3, 65.0, 63.8, 58.6, 56.5, 54.9, 50.4, 44.1, 39.8, 37.6, 36.0, 34.4,
29.2, 28.2, 26.9, 25.7; HRMS (ESI): (M+Na)+ calcd. for C59H73N15O23S2Na 1446.4343, found 1446.4348.
[11] I.R. Vlahov, F. You, H.K.R. Santhapuram, et al. Bioorg. Med. Chem. Lett. 21 (2011) 1202.
[12] K. Ulbrich, M. Michaelis, F. Rothweiler, et al. Int. J. Pharm. 406 (2011) 128.