European Journal of Medicinal Chemistry 85 (2014) 95e106

Contents lists available at ScienceDirect

European Journal of Medicinal Chemistry journal homepage: http://www.elsevier.com/locate/ejmech

Original article

Synthesis and biological evaluation of andrographolide analogues as anti-cancer agents* Ranjan Preet c, 1, Biswajit Chakraborty a, 1, Sumit Siddharth c, Purusottam Mohapatra c, Dipon Das c, Shakti Ranjan Satapathy c, Supriya Das b, Nakul C. Maiti b, Prakas R. Maulik a, Chanakya Nath Kundu c, *, Chinmay Chowdhury a, * a b c

Chemistry Division, CSIR-Indian Institute of Chemical Biology, 4 Raja S. C. Mullick Road, Kolkata 700032, India Structural Biology and Bioinformatics Division, CSIR-Indian Institute of Chemical Biology, 4 Raja S. C. Mullick Road, Kolkata 700032, India Cancer Biology Division, KIIT School of Biotechnology, KIIT University, Campus-11, Patia, Bhubaneswar, Odisha 751024, India

a r t i c l e i n f o

a b s t r a c t

Article history: Received 21 October 2013 Received in revised form 23 July 2014 Accepted 24 July 2014 Available online 24 July 2014

A new family of andrographolide analogues were synthesized and screened in vitro against kidney (HEK293) and breast (MCF-7) cancer cells. The anti-cancer effects of the active analogues (2b, 2c and 4c) were determined by multiple cell based assays such as MTT, immunostaining, FACS, western blotting and transcriptional inhibition of NF-kB activity. Importantly, these compounds were found to possess higher anti-cancer potency than andrographolide and low toxicity to normal (VERO and MCF-10A) cells. Increased level of Bax/Bcl-xL ratio, caspase 3, and sub G1 population, higher expression level of tumor suppressor protein p53 and lower expression level of NF-kB suggested potent apoptotic property of the active analogues. Data revealed that the andrographolide derivative-mediated cell death in cancer cells was p53 dependent. © 2014 Elsevier Masson SAS. All rights reserved.

Keywords: Andrographolide C14 ester analogues Epoxy diastereomers Anti-cancer HEK-293 MCF-7 Normal cells Apoptosis

1. Introduction Andrographis paniculata Nees (Acanthaceae) is considered as one of the most important medicinal plants in India, China and other Asian countries due to its popular use in traditional systems of medicines [1]. Andrographolide 1 (Fig. 1), a major phytoconstituent of the plant, has been recognized as an important pharmacophore because of its key role as inducer of apoptosis against different types of cancers [2] in addition to other pharmacological effects [3] (e.g., anti-viral [3a], anti-inflammatory [3b], antimalarial [3c], anti-hyperglycemic [3d], immunostimulatory [3e] etc.).

Abbreviations: AG, andrographolide; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide; DAPI, 4,6-diamidino-2-phenylindole; PARP, poly(ADP-ribose)polymerase. * This paper is dedicated to Dr. Pradeep Kumar Dutta, a former scientist and head of Chemistry Division, Indian Institute of Chemical Biology, Kolkata 700032, India. * Corresponding authors. E-mail addresses: [email protected] (C.N. Kundu), [email protected] (C. Chowdhury). 1 Contributed equally. http://dx.doi.org/10.1016/j.ejmech.2014.07.088 0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.

However, despite its impressive biological activities, the major drawback of andrographolide is poor oral bioavailability [4] making it difficult to prepare formulations for clinical use. Thus, only welldesigned derivatives of andrographolide might have the potential to be developed as anti-cancer chemotherapeutic agents. Indeed, a growing interest has been observed in recent times for designing, synthesizing and subsequently screening different analogues of andrographolide in order to discover lead(s) having better pharmacological profile than the parent compound. Towards this endeavor, few promising compounds having ester functionality at C14 of 1 have recently been identified by Stanslas [5a] (14acetylandrographolide), Nanduri [5b] (14-cinnamoyl-8,17-epoxyandrographolide), Rajagopal [5c] (DRF 3188), and us [5d] (14succinylandrographolide). In our previous report [5d], we also established the important role of the a-alkylidene-g-butyrolactone moiety of andrographolide to account for its cytotoxicity in human leukemic cell lines (U937, K562 and THP1). In continuation of our work [5d,6] in lead identification from bioactive natural products, we therefore became interested to check the anti-cancer potential of the ester analogs of 2 and their epoxy derivatives 3e4 as well

96

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

O

O HO

14

X

O

O O

O O

X

O

O O

O

X

O O

O

12

O

3

HO

H

HO

19

1 Andrographolide

HO

H HO 2 (X=Cl, Br, I)

HO

H

HO 3 (X=Cl, Br, I)

O HO HO

H 4 (X=Cl, Br, I)

Fig. 1. Andrographolide and its designed analogues.

(Fig. 1). The rationale for choosing the aforesaid compounds was primarily the following: (a) an ester moiety at C14 of andrographolide has been shown to be important for cytotoxicity [5]; (b) the ester moiety (or the halogen in the side chain of the ester moiety) of compounds (2e4) may serve as good leaving group to facilitate the binding with intracellular glutathione (GSH) [7], thereby triggering suicide of the cell leading to apoptosis; (c) esters might act like prodrugs, being hydrolyzed by esterases or under physiological pH to release andrographolide 1 (or its epoxy derivatives) in vivo [8]. Besides, compounds 3e4 differing in their epoxide configurations could be important for structureeactivity relationship (SAR) studies [9]. Despite several reports on the structural modifications of andrographolide [10], chemo-selective functionalizations at C14 hydroxyl are limited in number [5,11]. In this report, we describe the synthesis of novel C14-ester analogues 2e4 and their cytotoxic effects (in vitro) on kidney (HEK-293) and breast (MCF-7) cancer cells. We have also demonstrated the apoptotic properties of the most active analogues and compared their cytotoxicity with that in normal cells. 2. Results and discussion 2.1. Chemistry Andrographolide (~100 g) was isolated from the leaves of A. paniculata and used as starting material for derivatization. The synthetic pathways used in the present work are outlined in Scheme 1. We chose to carry out chemoselective esterification of andrographolide at C14 hydroxy, which is allylic in nature. Towards this objective, the other hydroxyls were converted (Scheme 1) into 3,19-isopropylidene derivative 5 by treatment with 2,2dimethoxypropane and p-toluenesulfonic acid (cat.). Compound 5 was initially reacted with chloroacetyl chloride in dry THF in the presence of pyridine (1.5 eqv.) and 4-dimethylaminopyridine (cat.) to afford the intermediate ester 6a with 66% yield. The corresponding bromo-ester 6b was obtained (63% yield) by reacting bromoacetyl bromide with 5, while the iodo derivative 6c was prepared (65% yield) by treating 6a with sodium iodide in acetone at rt for 3 h. The isopropylidene moiety of products 6aec was then removed by exposing the products to aqueous acetic acid (3:7), affording the targeted compounds 2aec which were then purified (>95%) by HPLC separations. Thereafter, we directed our efforts for chemoselective epoxidation of the exocyclic double bond (D8(17)) of intermediates 6aec. Towards this end, compound 6a was treated with m-chloroperbenzoic acid (m-CPBA) in dry dichloromethane to furnish a diastereomeric mixture of epoxides 7a and 8a with 51% yield. Pleasingly, this mixture was separated through silica gel (100e200 mesh) column chromatography (7a:8a ¼ 2:3). This

reaction protocol was subsequently applied on compounds 6b and 6c, which afforded the corresponding diastereomeric epoxides 7b/ 8b (54%, 7b:8b ¼ 1:2) and 7c/8c (52%, 7c:8c ¼ 2:5), respectively (Scheme 1). Thereafter, the isopropylidene moiety of the epoxides (7aec and 8aec) was removed using aqueous acetic acid (3:7) leading to the formation of the targeted products 3aec and 4aec as shown in Scheme 1. These compounds were finally purified (>95%) using HPLC separations. As epoxidation of intermediate 6 resulted in a diastereomeric mixture, we tried to make this reaction stereoselective by changing the strategy. Pleasingly, replacing intermediate 6 with 5 for epoxidation using m-CPBA ensured a totally stereoselective formation of epoxide 9 with 76% yield (Scheme 2). The absolute stereochemistry at C8 of product 9 was found to be S by single crystal X-ray analysis (see Fig. S1 of the Supplementary material). Notably, a previous study [12] on this reaction failed to get the expected epoxide 9; instead, a deprotected product was isolated. The facial selectivity operating in b-epoxidation of compound 5 is possibly influenced by the hydroxyl group which takes up b-orientation by rotation around the C11eC12 single bond (for energy minimized conformations see Fig. S2 of the Supplementary Material) to become proximate to the reactive double bond and directs (through hydrogen bonding [13]) the approach of m-chloroperbenzoic acid towards it. However, attempted esterification at C14 hydroxy of 9 employing chloroacetyl chloride, pyridine and DMAP (cat.), as shown in Scheme 2, did not furnish ester 8a; instead, an intractable solid resulted. After having this disappointing result, we performed this reaction using bromoacetyl bromide; this resulted in the formation of the expected product 8b, but with a discouraging yield of 35% only (Scheme 2). Due to the poor yield and lack of the consistency of the reaction for esterification of epoxide 9, the former strategy (Scheme 1) seemed to be a better option for the generation of such analogues. However, the structure determination of epoxide 8b from the latter study (Scheme 2) helped us to identify the stereochemical outcomes of the conversion of intermediate 6 described in Scheme 1 unambiguously. In 1H NMR, the olefinic proton (C12eH) signal appeared as triplet (t) at around d 6.99 ppm for diastereomers 7, but d 7.15 ppm in other diastereomers 8. The stereochemical assignments were further supported by NOESY analysis. 2.2. Biology 2.2.1. Anticancer properties of the synthesized compounds To check the anticancer effects of the synthesized compounds, we initially performed an MTT assay in human embryonic kidney cancer cells (HEK-293) and compared its effect with the normal monkey kidney cells (VERO). VERO cells are non cancerous in nature and are derived from the normal kidney epithelial cells. Cells

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

97

Andrographolide (1) (a) O

O

X

O

HO

O

O X

O

O

(b)

O

O 14

O

X O

(e)

O H O

5

H O

6a,7a,8a: X=Cl 6b,7b,8b: X=Br 6c ,7c,8c : X=I

(c)

12

O

6a-b

X

6a

O

O 8a-c

(d)

O X

O

O

O

X

HO

H

O

O HO

H

H

HO

HO

4a- c

3a- c

2a-c

O

O

O

O

O

HO

H O

(d)

2a, 3a,4a : X=Cl 2b,3b,4b: X=Br 2c,3c,4c: X=I HO

O

6c

O O

O

+

H 7a- c

(d)

O

14

12

O O

O O

Scheme 1. Synthesis of andrographolide analogues 2e4. Reagents and conditions: (a) 2,2-dimethoxypropane, p-TsOH (cat.), acetone, reflux, 2 h, 70%; (b) chloroacetyl chloride/ bromoacetyl bromide, pyridine, DMAP(cat.), dry THF, 0  C, 30 min, 66% for 6a, 63% for 6b; (c) NaI, acetone, rt, 3 h, 65%; (d) acetic acid/water (7:3), rt, 30 min, 67e75%; (e) m-CPBA, 0  C, DCM, 3 h, 51e54%.

were treated with increasing concentrations of the compounds for 48 h and cell viability was measured. Among the derivatives screened, only three compounds (2b, 2c, 4c) were found to be significantly active (LC50 ~ 8e12 mM), while others caused 50% cell (HEK-293) death at concentrations only above 20 mM. The LC50 of the remaining compounds has been provided in supplementary material (Table S2). Compounds 2b, 2c and 4c displayed LC50 of 12, 8, and 12 mM, respectively, in HEK-293 cells, but were not significantly toxic in normal (VERO) cells (Fig. 2A) as 50% cell death was not reached upto tested concentrations (40 mM). Andrographolide (AG) kills 50% of HEK-293 cells at ~25 mM but is noncytotoxic to VERO cells (Fig. 2A) at this range. The relative anticancer potency of the investigational compounds (2b, 2c and 4c) in HEK-293 and VERO cells with some of the already known anticancer agents like andrographolide, etoposide and 5-fluorouracil is provided in Table S3 of the Supplementary material. Additionally, we checked the cell viability of these compounds (2b, 2c and 4c) in MCF-7 (breast cancer) and MCF-10A cells (normal breast epithelial cells) as well. The tested compounds displayed LC50 of 8.4, 6, and 7 mM, respectively against MCF-7 cells, while in normal cells (MCF-10A) 50% cell death was not observed up to 40 mM (Fig. 2B). Andrographolide (AG) caused 50% cell death in MCF-7 cells at 16 mM but not in MCF-10A even up to 40 mM. Though 14-acetylandrographolide was reported [5a] to be an active compound against MCF-7 cell lines (GI50 ¼ 5.49 ± 1.5 mM), there is no report about its activity against kidney cancer cells (HEK-293). In the present study, 14-(2-bromo/iodoacetyl)andrographolide (2b/2c) displayed comparable results in MCF-7 cells and significant cytotoxicity in HEK-293 cells. On the other hand, 14acetyl-8,17-b-epoxyandrographolide was previously found [5b] to be inactive in most of cancer cells except few (ovarian/renal); we however observed that 14-(2-iodoacetyl)-8,17-b-epoxyandrographolide (4c) was active against kidney and breast cancer as well. Surprisingly, compound 3c, the other diastereomer, did not

appear as an active compound. Detailed study about the role of epoxide configuration in the activity profile is currently in progress. 2.2.2. Investigational compounds cause apoptosis in kidney cancer cells without affecting the phases of cell cycle To analyze the effect of 2b, 2c and 4c on the cell cycle regulation and apoptosis, HEK-293 cells were treated with these compounds in a dose dependent manner for 48 h, followed by analysis of the cells by flow cytometry after propidium iodide staining (Table 1) (see also Fig. S3 in the Supplementary material). It was noted that exposure of increasing concentrations (0e20 mM) of the investigational compounds led to an increase in the Sub G1 phase population, representing apoptosis without arresting the cells in any phase of cell cycle. A greater than four-fold increase in apoptosis in comparison to the untreated cells was observed at 20 mM concentration. 2.2.3. Immunocytochemistry experiment of caspase 3 For further studies on apoptosis caused by the test compounds, we performed an immunocytochemistry experiment of O 14

HO

11

O O

12

O

HO

b

8a

a O

O

O

H O

5

H O

c

8b

9

Scheme 2. Approach for stereoselective epoxidation and subsequent esterifications. Reagents and conditions: (a) m-CPBA, dry DCM, 0  C to rt, 3 h, 76%; (b) chloroacetyl chloride, pyridine, DMAP (cat.), dry THF, 0  C to rt, 2 h, 0%; (c) bromoacetyl bromide, pyridine, DMAP (cat.), dry THF, 0  C to rt, 2 h, 35%.

98

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

Fig. 2. A. MTT cell viability assay. Compound 2b/2c/4c and andrographolide (AG) cause a decrease in cell (HEK-293) survival. The sign , , and represent HEK-293/ investigational compound, VERO/investigational compound, HEK-293/AG, and VERO/AG, respectively. Data represented here is the mean of three independent experiments and the values are mean ± SD. Statistical significance was determined by paired t-test. *p < 0.05. B. MTT cell viability assay. Compounds 2b/2c/4c trigger decrease in cell (MCF-7) survival. The signs and represent MCF-7/2b, MCF-7/2c, MCF-7/4c, respectively, while the signs , , and represent MCF-10A/2b, MCF-10A/2c, and MCF-10A/4c, , respectively. However, MCF-7/AG and MCF-10A/AG are represented by and , respectively. Data represented are the mean of three independent experiments and the values are mean ± SD. Statistical significance was determined by paired t-test. *p < 0.05.

caspase 3 in HEK-293 cells. An increased expression of caspase 3 was observed when HEK-293 cells were treated with the fixed concentration (respective LC50 value) of 2b/2c/4c for 48 h (Fig. 3). Higher accumulation of caspase 3 was observed in treated cells in comparison to untreated and andrographolide (AG) treated cells.

2.2.4. Transcriptional inhibition of NF-kB activity by the investigational compounds The overexpression of NF-kB transcription factor activity is considered as the hallmark in initiation, promotion and growth of cancer, and thus inhibition of this activity of NF-kB is an important approach in the discovery of chemotherapeutic agents. To

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

determine the role of the test compounds on transcriptional activity of NF-kB gene in HEK-293 cells, a luciferase based assay was used following our earlier protocol [14e18]. We observed that the relative luciferase activity of the NF-kB gene decreased in a dose dependent manner after treatment of the cells with the investigational compounds (Fig. 4). Approximately five-fold reduction of luciferase activity was observed at 5 mM concentration of 2c, but at 10 mM concentration for both 2b and 4c. Additionally, in order to determine the effect of the investigational compounds on apoptosis in HEK-293 cells, a DAPI (40 ,6diamidino-2-phenylindole) nuclear staining method and determination of anti-cell migratory activity were also used (see Fig. S4 in the Supplementary material for details). 2.2.5. Mechanism of action To further confirm the apoptotic effect of investigational compounds and study the mechanism of anti-cancer action, we have measured the expression of different pro-apoptotic and antiapoptotic protein markers in HEK-293 (expressed wt p53) and HEK-293 p53 / cells after exposure to these agents (Fig. 5A). The left panel of Fig. 5A demonstrates the increase in expression of Bax protein, while the level of Bcl-xL was reduced significantly with simultaneous increase in the PARP cleaved product (86 kDa) in treated HEK-293 cells. An increase in the tumor suppressor protein p53 with a significant decrease in the level of NF-kB was also noted when HEK-293 cells were treated with the investigational compounds. Interestingly, no significant alteration was noted in the expression level of the above mentioned protein biomarkers in the treated HEK-293 p53 / cells (right panel of Fig. 5A). Thus it appeared that investigational compound mediated apoptosis in HEK-293 cells is p53 dependent. Based on above result, we have drawn a scheme representing the mechanism of action of the compounds on kidney cancer cells (Fig. 5B). We propose that compounds (2b, 2c and 4c) promote the activation of p53, leading to a downregulation of NF-kB with simultaneous upregulation of Bax/Bcl-xL ratio that triggers cytochrome C release followed by upregulation of caspase 3 and thereby causing apoptosis. 3. Biophysical studies 3.1. Binding efficacy and mode of binding of compound 2c Among the active analogues (2b, 2c and 4c), compound 2c appeared to be the best; we, therefore, decided to study the binding efficacy and mode of binding on this compound. DNA shows an intense absorption band at 260 nm and the intensity of this absorption band was reduced significantly in the presence of 2c with increasing concentration (Fig. 6A). These indicated that this compound interacts with DNA effectively and further investigations were carried out focusing on the binding efficacy, mode of binding and the interaction pattern. The binding isotherm (Fig. 6B) for DNA - 2c interaction was obtained by plotting fraction (q) of DNA bound to 2c versus total compound concentration [19]. The fraction (q) was calculated as q ¼ (A0 e Ai)/(A0 e Aa), where, A0 is the absorbance of DNA in the absence of compound at 260 nm (or slightly blue or red shifted), Ai is the absorbance of DNA at any compound concentration, and Aa is the absorbance of DNA at compound concentration for which maximum binding took place. Kb was obtained from the reciprocal of the ligand (2c) concentration corresponding to the half-saturation value of the binding isotherm. The measured binding constant was 2.5  105 M1. In addition, circular dichroism (CD) spectral analysis was carried out to measure the effect of drug molecule binding to conformational changes in DNA duplex structure (Fig. 7). The CD spectra of the DNA solution (500 mg/mL) in the absence and

99

Table 1 Quantitative analysis of cell cycle profile.a Apoptosis 2b Control 5 mM 10 mM 15 mM 20 mM 2c 5 mM 10 mM 15 mM 20 mM 4c 5 mM 10 mM 15 mM 20 mM

GI

S

G2/M

24.26 60.44 67.33 76.62 88.84

± ± ± ± ±

0.8 0.7 0.6 0.9 1.1

53.21 25.16 19.57 14.98 9.01

± ± ± ± ±

0.8 0.8 0.5 0.6 1.2

19.9 10.44 10.28 6.69 1.57

± ± ± ± ±

0.5 0.5 0.7 0.6* 1.1*

2.63 3.96 2.82 1.71 0.58

± ± ± ± ±

0.7 0.7 0.9 0.8 1.3

43.09 74.03 78.50 89.69

± ± ± ±

0.5* 0.7* 0.6 0.8

24.77 15.60 14.80 7.96

± ± ± ±

0.5* 0.8* 0.6 0.7

9.21 7.57 5.67 1.66

± ± ± ±

0.6* 0.6 0.9 0.7

2.14 2.74 1.28 0.69

± ± ± ±

1.1 0.8 0.7 0.9

62.05 77.99 79.55 92.93

± ± ± ±

0.7 0.8* 0.6 1.1*

25.38 15.61 14.98 5.37

± ± ± ±

0.7* 0.9* 0.7 1.1

9.87 4.98 4.63 1.18

± ± ± ±

0.6 0.8* 0.9 1.2

2.70 1.41 1.31 0.52

± ± ± ±

0.8 0.9 0.8 1.3

Data represent the mean ± SD of experiments in triplicate. Statistical significance was determined by paired t-test. *p < 0.05. a Quantitative analysis of cell cycle profile was done using histogram data (provided in Fig. S3 in the Supplementary material) analyzed by Cell Quest Pro software (Becton Dickinson, CA).

presence of 2c (100 mM) were recorded after 2 h incubation at 37  C. The solution was half diluted with the same buffer for CD measurement. It was observed that the CD band at 275 nm (due to base stacking) and the absorbance at 245 nm (due to the Bconformation of DNA) did not change significantly in the presence of 2c, while EtBr (a well known DNA intercalator) which is often used as positive control [20] showed changes in the absorption spectra (shape and absorbance). Earlier report [21] suggested that simple electrostatic interaction or groove binding of small molecules to DNA did not severely affect the absorbance at 275 nm band of the DNA. Similarly, changes in absorption of the band at 245 nm implied the change of B-DNA conformation or the formation of a different conformation of DNA. Our results showed that the CD spectra of the DNA in the presence and absence of the compound were very similar; indicating that compound 2c caused no detectable conformational change of DNA and was not intercalated like EtBr. We further investigated the mode of binding of 2c to DNA using DAPI (40 ,6-diamidino-2-phenylindole)-DNA complex. DAPI has a high affinity for the minor groove of A/T-rich binding sites [22] and binds to double stranded DNA. It forms a complex that gives high fluorescence with a band maximum at 460 nm. However, DAPI alone is weakly fluorescent in water. When compound 2c was added to DAPI-DNA complex, the fluorescence of the solution was reduced (Fig. S6 in Supplementary material) due to release of DAPI and incorporation of compound 2c to DNA. The result indicated that the compound was capable of binding to the minor groove of DNA where DAPI also preferred to bind. Finally, a molecular docking analysis was performed to further characterize the binding interactions of compound 2c to DNA (see Table S5 and Fig. S7 in Supplementary material). Out of the 100 docking runs performed, favorable binding poses were obtained with those presenting the most negative binding free energies. Molecular docking results showed that compound 2c binds favorably to DNA minor groove and the calculated binding energy was 8.47 kcal/mol. The docking of netropsin, a well known minor grove binder with DNA [23], showed binding energy 7.02 kcal/ mol. The minor groove bound state of 2c resulted in a close fit of the molecule along the wall of the minor groove (Fig. S7 in Supplementary material). The results of docking analysis with other compounds (2b, 4c) are also shown in Supplementary material (Table S5 in Supplementary material). The docking

100

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

Fig. 3. Caspase 3 staining: HEK-293 cells treated with andrographolide (AG) and 2b/2c/4c were stained with anti-caspase-3 antibody and DAPI. Photographs were taken at 40 magnification. The fields of Caspase 3 staining were merged with respective DAPI stained nucleus with the help of IMAGE J software. Photographs presented here is the best of three independent experiments.

results thus indicated that 2c can act as a minor groove binder like netropsin. Thus all the experiments and computation analysis favored the notion that compound 2c effectively binds to the minor grove of

DNA with significant affinity. We therefore became interested to check the potentiality of 2c for DNA adduct formation. For this purpose, we performed a model reaction between guanosine and 2c in glacial acetic acid at 75  C following the literature procedure

Fig. 4. Relative luciferase gene reporter assay of p5XIP10 B plasmid containing NF-kB construct after transfection into HEK-293 cells and treatment with indicated concentrations of compounds andrographolide (AG), 2b, 2c and 4c, respectively. Data represent the mean ± SD of three independent experiments. Statistical significance was determined by paired ttest. *p < 0.05.

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

101

Fig. 5. A: Western blot of HEK-293 and HEK-293 p53 / cells. Western blot analysis of the investigational compounds (2b/2c/4c) and andrographolide (AG) for apoptotic markers in HEK-293 and HEK-293 p53 / cells after treatment for 48 h. The lower panel shows GAPDH as the loading control. B: Analogs (2b, 2c and 4c) of andrographolide cause apoptosis in kidney cancer cells by upregulating p53. Upward arrow indicates the upregulation and downward arrow indicates the downregulation.

[24]. However, even after heating the reaction mixture for 12 h, no such adduct formation was observed by TLC and this was further confirmed from LCMS analysis. 3.2. Hydrolysis of C14 ester moiety of compound 2c and determination of the half life period (t1/2) We also planned to study the hydrolysis of the C14 ester moiety of 2c at physiological pH (7.4) using buffer and employing suitable enzyme (esterase type). Before performing this experiment, we attempted to construct the requisite calibration curve using HPLC employing different concentrations of 2c in PBS buffer (pH ¼ 7.4) without adding any enzyme. Surprisingly, it was observed that 2c was slowly hydrolyzed to andrographolide in PBS buffer only. It is worthwhile to mention here that the prodrug isotaxel, an analogue of taxol, was converted to taxol at physiological pH and this facilitated the delivery (in vivo) of taxol [25]. This gave us the impetus to calculate the half life period of 2c in PBS buffer to check the possibility of using this compound as a prodrug. Accordingly, the hydrolysis experiment of compound 2c was carried out in PBS buffer

(at pH 7.4) and reaction kinetics was followed by HPLC through monitoring the amount of unreacted 2c at different time intervals. We observed an exponential decrease of the amount of 2c with time and the reaction kinetics was found to be of first order. Thus, fitting to a first order rate equation the measured rate constant of 2c was 0.02 min1 and corresponding half-life (t1/2) was 35 min (see Fig. S8 in Supplementary material). This experiment indicates that 2c could serve as a prodrug under physiological pH. Besides, there is the other possibility that esterases may also cause this hydrolysis (in vivo). 3.3. GSH binding of compound 2c Next, we investigated the binding efficacy of 2c with GSH. As ester moiety of 2c is susceptible to hydrolysis in buffer of pH 7.4, we decided to conduct this study in high dilution of the buffer compared to the substrate. Accordingly, ethanolic solutions of both 2c and GSH were added to the PBS buffer where final concentrations of each substrate were adjusted to 10 mM and that of buffer to 2 mM. After heating this reaction mixture at 50  C for 8 h under

102

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

Fig. 6. (A) Ultravioletevisible (UVevis) absorption spectra of DNA (100 mg/mL) after incubation with 2c in different concentrations in buffer (50 mM TriseHCl/NaCl, pH 7.5) at 37  C for 2 h. Spectra were taken after six dilutions with the same buffer in which the final concentrations of 2c were calculated as 0 (a), 3.33 (b), 6.66 (c), 9.99 (d), 13.33 (e), 16.66 (f), and 20.00 (g) mM. The absorption spectra were normalized at 230 nm to avoid contribution from turbidity due to the formation of complex. (B) shows the fraction (q) of DNA binding site involved in complexation with the ligand. Details about the calculation of fraction of DNA occupied by the ligand are provided in the result section. The data were fitted to nonlinear Boltzman curve fit using Origin software with the equation, y ¼ A2 þ (A1  A2)/(1 þ exp((x  x0)/dx)), where, x0 ¼ half saturation point where the concentration of ligand equals Kd, i.e., ~4 mM, the reciprocal of which provides the binding constant Kb ¼ 2.5  105 M1.

argon atmosphere, two major products were found to be formed; these were separated by reverse phase HPLC and characterized using ESI mass and NMR spectroscopy (see Supplementary material for details). Both products were identified to be 14-deoxy-12(glutathione-S-yl)-andrographolide (Fig. 8) but must differ in C12 stereochemistry (diastereomeric at C12). The crucial evidence in favor of linkage through sulfur atom of glutathione (GSH) came from clear HMBC correlations observed between the signals for CH2S- protons and C12 carbon in both the isomers. Incidentally, earlier workers [7] reported formations of two products from the reaction of GSH with andrographolide or a different C14 ester and

identified them as 14-deoxy-12-(glutathione-S-yl)andrographolide and 14-deoxy-12-(glutathione-amino)andrographolide. But in our case, there was no formation of any 14-deoxy-12-(glutathioneamino)andrographolide. However, formations of our products can be explained through a possible mechanism where sulfhydryl (SH) group of GSH undergo Michael addition to the C12eC13 double bond of 2c followed by elimination of its ester group at C14. This study demonstrates that 2c makes covalent binding with GSH and is capable of reducing the level of the intracellular GSH augmenting its cytotoxic activity.

4. Conclusion DNA DNA + 2C DNA + EtBr

8

mdeg

4

0

-4

-8

220

240

260

280

300

320

Wavelength (nm) Fig. 7. CD spectra of DNA in the presence of different compounds in aqueous buffer: black, red and blue line represent the spectra of only DNA (500 mg/mL), DNA in the presence of compound 2c (50 mM) and DNA in the presence of EtBr (150 mM), respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

A new family of C14-ester analogs of andrographolide and their

a and b diastereomeric epoxy derivatives were synthesized [26]. The anti-cancer activity of these compounds was tested in kidney (HEK-293) and breast cancer cells (MCF-7) and compared with the results in corresponding normal (VERO and MCF-10A) cells. MTT cell viability assay revealed that analogs 2b, 2c and 4c exerted significant cytotoxicity (8e12 mM in HEK-293 cells and 6e8.4 mM in MCF-7 cells) in cancer cells and low toxicity towards normal cells. Importantly, these analogues showed more potency than andrographolide. Increase in apoptotic nuclei after DAPI staining, sub G1 population after FACS analysis, and Bax/Bcl-xL ratio, cleavage of poly (ADP) polymerase, caspase 3, and p53, and decreased level of NF-kB promoter activity pointed to the significant apoptotic property of these compounds. Compound 2c turned out to be the most active in both kidney and breast cancer cells suggesting its potential to be utilized in drug discovery through further optimization. A scheme representing the mechanism of action of the investigational compounds (2b, 2c and 4c) on kidney cancer cells is shown (Fig. 5B). We propose that the compounds activate p53, a tumor suppressor protein, leading to the downregulation of NF-kB with simultaneous upregulation of Bax/Bcl-xL ratio, thereby causing apoptosis. These compounds were shown to act as minor groove binder of DNA. The study will aid in the development of novel anti-

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

OH

O 9' O 14

1 9

3 HO

5

HO 19

O

O 12

NH 2'

1'

S 17

NH2

O

HN 4' O OH

Fig. 8. 14-Deoxy-12-(glutathione-S-yl)-andrographolide.

cancer agents based on andrographolide and provide insight into anti-cancer mechanism. 5. Experimental 5.1. Chemistry 5.1.1. General methods Andrographolide 1 was isolated (~100 g, 0.54% yield) from dried and powdered leaves of A. paniculata following the reported procedure [27]. All solvents were distilled prior to use. All the reactions were performed under argon atmosphere and anhydrous conditions unless otherwise noted. Reactions were monitored by thinlayer chromatography (TLC) on TLC aluminium sheets 20  20 cm silica gel 60F254 and silica gel 60 RP-18 F254s. Visualization of the developed chromatogram was performed by UV (254 nm, 365 nm) light or iodine or Liebermann solution. The analytical and semi preparative HPLC was performed using X-Bridge columns (C18, 5 mm, 4.6  250 mm and C18, 5 mm, 10  250 mm). 1H and 13C NMR spectra were recorded in 300 or 600 MHz spectrometer. Chemical shifts (d) were reported in parts per million (ppm) downfield from tetramethylsilane (d ¼ 0.00) with the residual protons of deuterated solvents used [CDCl3: 1H NMR, d ¼ 7.26 ppm (s), 13C NMR, d ¼ 77.0 ppm (t); CD3OD : 1H NMR, d ¼ 4.86 ppm (s), 3.32 ppm (s), 13 C NMR, d ¼ 49.0 ppm (multiplet)]. Coupling constants (J) were expressed in hertz (Hz) and spin multiplicities are given as ‘s’ (singlet), ‘d’ (doublet), ‘dd’ (double doublet), ‘t’ (triplet), ‘dt’ (doublet of triplet), ‘m’ (multiplet) ‘br’ (broad) and brs (broad singlet). 5.1.2. Typical procedure for the synthesis of the intermediate chloroester 6a Initially, the hydroxyl groups at C3 and C19 of andrographolide 1 were protected according to our earlier protocol [5d] resulting in the synthesis of 3,19-isopropylideneandrographolide 5. To a well stirred solution of compound 5 (250 mg, 0.64 mmol) in dry THF (5 mL) maintained at 0  C, chloroacetyl chloride (0.06 mL, 0.67 mmol), pyridine (0.07 mL, 0.96 mmol), and a catalytic amount of DMAP (4-dimethylaminopyridine) were added sequentially keeping the pH of the reaction mixture neutral or just basic. The reaction mixture was then stirred for 30 min under argon atmosphere. After completion of the reaction (monitored by TLC), the solvent was evaporated under reduced pressure at 0  C; the resulting residue was then mixed with water (10 mL) and ether (10 mL). The mixture was neutralized using 1(N) acetic acid and washed with saturated Na2CO3 solution (10 mL). The organic extracts were dried over anhydrous Na2SO4, filtered and concentrated

103

in vacuo. The residue was purified by silica gel (100e200 mesh) column chromatography using 20% ethyl acetate in petroleum ether (v/v) as eluent to afford the product 6a (66%). The same procedure was adopted for the synthesis of product 6b using bromoacetyl bromide. 5.1.3. Procedure for the synthesis of 6c Compound 6a (200 mg, 0.42 mmol) was dissolved in dry acetone and the mixture was stirred with NaI (128 mg, 0.85 mmol) for 4 h at rt under argon atmosphere. After completion of the reaction (TLC), the solvent was evaporated under reduced pressure and the residue was extracted with EtOAc (3  10 mL). The organic extracts were dried over anhydrous Na2SO4 and concentrated in vacuo; the residue was purified by silica gel (100e200 mesh) column chromatography using 20% ethyl acetate in petroleum ether (v/v) as eluent to furnish the product 6c (65%). 5.1.4. Typical procedure for the synthesis of the epoxides 7a and 8a To an ice cooled solution of 6a (300 mg, 0.64 mmol) in dry dichloromethane under argon atmosphere, m-chloroperbenzoic acid (143 mg, 0.83 mmol) was added portion wise over a period of 30 min; the reaction mixture was then allowed to come to rt over a period of 40 min and stirred for another two and half hr at rt. After completion of the reaction (TLC), it was extracted with EtOAc (2  20 mL), and washed with saturated Na2CO3 solution (15 mL), and brine (10 mL), respectively. The organic extracts were dried over anhydrous Na2SO4, filtered and evaporated in vacuo. The residue obtained was purified over silica gel (100e200 mesh) using 22% ethyl acetate in petroleum ether (v/v) resulting in the separation of the diastereomeric epoxides 7a and 8a (2:3) in 51% yield. The same procedure was adopted for the preparations of other epoxides 7b/8b and 7c/8c using substrates 6b and 6c, respectively and these diastereomeric mixtures were also separated successfully using silica gel (100e200 mesh) chromatography leading to isolation of 7b/8b (54%, 1:2) and 7c/8c (52%, 2:5). The spectral data of 6aec, 7a/8a, 7b/8b and 7c/8c have been provided under “Supplementary material”. 5.1.5. General procedure for the deprotection of isopropylidene moiety of intermediate 6 To a well stirred solution of compound 6 (0.05 mmol) in 1,4dioxan (3 mL), 1.5 mL acetic acid-water (7:3, v/v) was added. The reaction mixture was allowed to stir at room temperature for 30 min. After completion of the reaction (TLC), the mixture was evaporated under reduced pressure. The residue was mixed with water (10 mL) and extracted with ethyl acetate (3  20 mL). The combined organic extracts were dried over anhydrous Na2SO4, filtered and concentrated under reduced pressure. The residue obtained was purified through semi preparative HPLC using a mixture of water, methanol, acetic acid as mobile phase to get the desired product 2 (69e74%). The same procedure was adopted for the deprotection of other intermediates 7aec, 8aec resulting in the formations of 3aec and 4aec, respectively. The spectral data of compounds 2a, 3aec, 4aeb have been provided under “Supplementary material”. 5.1.6. Andrographolide-14a-O-bromoacetate (2b) Yellow gum, yield 70%, IR (KBr) nmax : 3291, 2927, 1759, 1268, 1194, 1081, 1029 cm1; 1H NMR (CDCl3, 600 MHz): d 7.07 (1H, td, J ¼ 6, 1.2 Hz), 5.97 (1H, d, J ¼ 6 Hz), 4.89 (1H, s), 4.57 (1H, dd, J ¼ 11.4, 6 Hz), 4.49 (1H, s), 4.26 (1H, dd, J ¼ 11.4, 1.8 Hz), 4.17 (1H, d, J ¼ 10.8 Hz), 3.86 (1H, d, J ¼ 12 Hz), 3.82 (1H, d, J ¼ 12 Hz), 3.48e3.46 (1H, m), 3.32 (1H, d, J ¼ 10.8 Hz), 2.85 (1H, brd), 2.65 (1H, brd), 2.47e2.41 (3H, m), 1.97 (1H, td, J ¼ 12.6, 4.2 Hz), 1.84e1.80 (4H,

104

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

m), 1.74e1.72 (1H, m), 1.63 (1H, brs), 1.24 (3H, s), 1.31e1.17 (2H, m), 0.66 (3H, s); 13C NMR (CDCl3, 75 MHz): d 168.6, 167.1, 151.7, 146.5, 123.1, 109.1, 80.4, 71.1, 69.5, 64.1, 55.7, 55.1, 42.8, 38.8, 37.6, 36.9, 28.1, 25.4, 24.7, 23.6, 22.7, 15.1; MS (ESI) 495.21 [MþNa]þ. HRMS (ESI) m/z calculated for C22H32BrO6 [MþH]þ 471.1382, found 471.1380. 5.1.7. Andrographolide-14a-O-iodoacetate (2c) Yellow solid, yield 69%, m.p. 148e150  C. IR (KBr) nmax: 3275, 2923, 1753, 1250, 1073, 1028 cm1; 1H NMR (CDCl3, 300 MHz): d 7.08 (1H, t, J ¼ 6.15 Hz), 5.94 (1H, d, J ¼ 5.4 Hz), 4.91 (1H, s), 4.59e4.53 (2H, m), 4.25 (1H, d, J ¼ 10.5 Hz), 4.17 (1H, d, J ¼ 11.1 Hz), 3.76 (1H, d, J ¼ 9.6 Hz), 3.69 (1H, d, J ¼ 9.6 Hz), 3.49e3.47 (1H, m), 3.33e3.29 (1H, m), 2.83e2.70 (1H, m), 2.50e2.41 (4H, m), 2.04e1.92 (2H, m), 1.83e1.58 (6H, m), 1.25 (3H, s), 0.68 (3H, s); 13C NMR (CDCl3, 75 MHz): d 168.8, 168.7, 151.6, 146.4, 123.2, 109.1, 80.3, 70.9, 69.2, 64.05, 55.7, 56.0, 42.8, 38.8, 37.6, 36.9, 29.0, 25.5, 23.6, 22.6, 15.2, 7.0; MS (ESI) 541.49 [MþNa]þ. HRMS (ESI) m/z calculated for C22H31INaO6 [MþNa]þ 541.1063, found 541.1065. 5.1.8. Andrographolide-8,17-b-epoxide-14a-O-iodoacetate (4c) Colorless gum (yield 67%) IR (Neat) nmax : 3435, 2939, 1755, 1032, 736 cm1; 1H NMR (CDCl3, 600 MHz): d 7.13 (1H, t, J ¼ 6.9 Hz), 5.92 (1H, d, J ¼ 4.8 Hz), 4.55 (1H, dd, J ¼ 11.1, 6.3 Hz), 4.23 (1H, d, J ¼ 11.4 Hz), 4.18 (1H, d, J ¼ 10.8 Hz), 3.76 (1H, d, J ¼ 10.2 Hz), 3.72 (1H, d, J ¼ 10.2 Hz), 3.49e3.47 (1H, m), 3.35 (1H, d, J ¼ 10.8 Hz), 2.65 (1H, d, J ¼ 1.8 Hz), 2.54 (1H, brs), 2.09e2.04 (2H, m), 1.99e1.80 (3H, m), 1.78e1.67 (2H, m), 1.63e1.57 (2H, m), 1.48e1.43 (2H, m), 1.27 (3H, s), 1.22e1.17 (3H, m), 0.83 (3H, s); 13C NMR (CDCl3, 150 MHz): d 168.7, 152.1, 122.5, 80.1, 70.9, 69.0, 63.9, 58.1, 54.6, 53.9, 50.1, 42.6, 39.4, 37.3, 35.7, 27.4, 23.4, 22.7, 21.2, 15.3, 6.2; (ESI) 557.18 [MþNa]þ. HRMS (ESI) m/z calculated for C22H31NaIO7 [MþNa]þ 557.1012, found 557.1015. 5.2. Biology 5.2.1. Maintenance of cell lines The HEK-293 (human kidney cancer cells), HEK-293 p53 /, VERO (Kidney epithelial cells of African Green monkey) and MCF7 cells (breast cancer cells) were maintained in DMEM supplemented with 10% FBS, 1.5 mM L-Glutamine and 1% antibiotic (100 U of Penicillin and 10 mg of streptomycin per ml in 0.9% normal saline) under 5% CO2 at 37  C in a humidified CO2 incubator. MCF-10A cells (normal breast epithelial cells) was grown in DMEM/F-12 (50:50, v/v) medium supplemented with 10% (v/v) FBS, 100 U/mL of penicillin, 100 mg/mL of streptomycin, 0.5 mg/mL of hydrocortisone, 100 ng/mL of cholera toxin, 10 ng/mL of epidermal growth factor and 1% (w/v) L-Glutamine in 5% CO2 at 37  C in a humidified CO2 incubator. Cell culture reagents and other growth supplements were procured from HiMedia, India; all the antibodies were purchased from Cell Signaling Technology, USA. SiRNA targeting p53 (catalog no. # sc-37459) was purchased from Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA. 5.2.2. Cell viability assay The anchorage dependent viability of the cells after treatment with the synthesized andrographolide analogues was measured using MTT [3-(4,5-dimethylthiazol-2yl-)-2,5-diphenyl tetrazolium bromide] cell proliferation assay according to our earlier protocol [14e18]. Cytotoxicity of the investigational compounds was measured by MTT assay. Cells were plated in 96-well plates at a density of 10,000 cells per well/200 mL of the medium. Cultures were treated with different concentrations of the investigational compounds for 48 h. The cells were washed with 1 PBS and then MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium

bromide) (2.5 mg/mL) was added to each well. It was then incubated at 37  C for 6 h. Detergent solution (10% NP-40 with 4 mM HCL) was added to each well of culture plate and the color intensity was measured at 570 nm using microplate reader (Berthold, Germany). 5.2.3. NF-kB luciferase assay Luciferase was used as a reporter gene to measure the activity of promoters and to study the transfection efficiency. The Luciferase assay was carried out according to the protocol mentioned earlier [14]. HEK-293 cells were seeded in six well plates. After reaching 70% confluency, the cells were transfected with two plasmids, NFkB (2.0 mg) and b-Gal (1.0 mg) by Calcium Phosphate method and incubated for 6e8 h. After incubation, the transfection media were replaced with fresh serum containing media and incubated for 12 h. Then the cells were treated with the analogues (2b/2c/4c) and andrographolide (25 mM) for 48 h. Finally, the cells were harvested and washed twice with PBS, then lysed with lysis buffer and the efficiency of transfection was normalized to b-galactosidase activity. Luciferase activity was measured by microplate reader (Multimode ELISA Reader, Tristar, Berthold technologies, Germany). 5.2.4. Analysis of cell cycle by flow cytometry A FACS analysis was carried out to confirm the apoptosis of cells after 2b/2c/4c treatment for 48 h. After the end of the treatment, cells were harvested, washed with PBS containing RNase-A, and fixed with 70% ethanol. Later they were stained with 0.1 ml of PI (50 mg/ml) and finally analyzed by Flow Cytometry (Becton and Dickinson, CA) using Cell Quest Pro Software (Becton and Dickinson, CA). 5.2.5. Western blot analysis For western blot analysis, approximately 5  105 cells/plate of HEK-293 and HEK-293 p53/ were seeded on 60 mm tissue culture discs, treated with the LC50 concentration of 2b/2c/4c and andrographolide and incubated for 48 h. Total cellular lysates were prepared using modified RIPA lysis buffer (50 mM tris, 150 mM NaCl, 0.5 mM deoxycholate, 1% NP 40, 0.1% SDS, 1 mM Na3VO4, 5 mM EDTA, 1 mM PMSE, 2 mM DTT, 10 mM b-glycerophosphate, 50 mM NAF, 0.5% triton X-100, protease inhibitor cocktail), 80 mg of protein was loaded, and separated on an SDS-PAGE gel electrophoresis apparatus. Proteins were transferred on to PVDF membranes and western blotting was performed by using anti-p53, anti-Bax, anti-Bcl-xL, anti-PARP, anti-NF-kB and anti-GAPDH antibodies. 5.2.6. Knockdown of p53 in HEK-293 cells The wild type expression level of p53 was knocked down in HEK-293 cells according to the protocol referred earlier [18]. In brief, HEK-293 cells were grown on 60 mm tissue culture dishes. On attaining 70% confluency, the cells were transfected with 4 mg of p53 SiRNA (catalog no. # sc-37459, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) using Lipofectamine 2000® as the transfection reagent in serum and antibiotic free media. The cells were then incubated for 8 h, following which the transfection media was replaced with normal serum containing media. The cells were then treated with the LC50 concentrations of the investigational compounds (2b, 2c and 4c) along with andrographolide and incubated for 48 h. Total cellular lysates were prepared using modified RIPA lysis buffer and western blot analysis was performed as mentioned above. 5.2.7. Immunocytochemistry analysis of caspase-3 expression HEK-293 cells were seeded in a 96 well plate and incubated at 37  C in 5% CO2 incubator for 24 h. After attaining 70% confluence,

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106

the cells were treated with the compounds at LC50 value of the 2b/ 2c/4c and incubated for 48 h. After incubation, the media was removed and washed with 1 PBS. The cells were fixed with acetone:methanol in 1:1 ratio and the plate was kept for 20 min at 20  C followed by blocking with 2% BSA and triton X-100 in 1 PBS. Anti-caspase 3 antibody (cat #9662 from Cell Signalling Technology) was added, and incubated for 2 h at 37  C, then secondary antibody conjugated to FITC was added and incubated for 1 h at 37  C. DAPI was used as a nuclear stain and fluorescent microscope (Nikon, Japan) photographs were taken at 40 magnification. 5.2.7.1. Statistical analysis. A two-tailed Student's t-test was employed and P < 0.05 was considered to be statistically significant. 5.3. Biophysical studies 5.3.1. UVevis absorption spectroscopic study The UVevis absorption spectrophotometric studies were carried out with 1 cm path length quartz cuvette using UVevis spectrophotometer; (UV-2401PC, Shimadzu). The spectra were scanned in the wavelength range 200e400 nm. Stock solution of each of the compounds having different concentration was prepared in 60% ethanol in 50 mM TriseHCl/NaCl buffer (pH 7.5) and then diluted to varied concentration with the buffer. Then the absorption spectra for different concentrations were taken to obtain the molar extinction coefficient which was used to determine the actual concentration of the compound throughout the experiment. CTDNA concentration was kept fixed at 50 mg/mL and the concentration of the compound was varied from 0 to a higher concentration as required for individual compounds. The DNA-compound complex was incubated at 37  C for 2 h and the absorption spectra were taken with proper dilution in the range 200e400 nm. 5.3.2. Circular dichroism (CD) spectroscopic study CD spectroscopic measurements were performed at 25  C with cuvette having path length 1 mm using CD spectrometer (Jasco J815) with the sensitivity of 100 millidegree and in continuous scanning mode with a speed of 100 nm/min; band-width was 1 nm and the number of accumulations was three. Scans were taken from 190 to 250 nm. CT-DNA (1000 mg/mL) was used for recording spectra in the presence and absence of compound 2c (100 mM). EtBr (300 mM) was used as positive control. Spectra were recorded after 2 h of incubation at 37  C. Acknowledgments This work is financially supported by Council for Scientific and Industrial Research network project (CSC0108). Partial financial support from Department of Science and Technology, Ministry of Science and Technology, New Delhi (project no: SR/S1/OC-42/2009) is gratefully acknowledged. PRM thanks CSIR (New Delhi) for an EMS grant. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ejmech.2014.07.088. References [1] (a) D. Bensky, A. Gamble, Chinese Herbal Medicine: Materia Medica, Eastland Press, CA, USA, 1993; (b) D. Chakraborty, R.N. Chakravarti, 314 Andrographolide. Part 1, J. Chem. Soc. (1952) 1697e1700.

105

[2] (a) For a recent review article, see: J.C.W. Lim, T.K. Chan, D.S. Ng, S.R. Sagineedu, J. Stanslas, W.F. Wong, Andrographolide and its analogues: versatile bioactive molecules for combating inflammation and cancer Clin. Exp. Pharmacol. Physiol. 39 (2012) 300e310; (b) J. Zhou, S. Zhang, O. Choon-Nam, H.-M. Shen, Critical role of pro-apoptotic Bcl-2 family members in andrographolide-induced apoptosis in human cancer cells, Biochem. Pharmacol. 72 (2006) 132e144; (c) S. Rajagopal, R.A. Kumar, D.S. Deevi, C. Satyanarayana, R. Rajagopalan, Andrographolide, a potential cancer therapeutic agent isolated from Andrographis paniculata, J. Exp. Ther. Oncol. 3 (2003) 147e158; (d) J. Stanslas, P.S. Liew, N. Iftikhar, C.P. Lee, S. Saad, N. Lajis, Potential of andrographolide in the treatment of breast cancer, Eur. J. Cancer 37 (Suppl. 6) (2001) s169; (e) F.-P. Liang, C.-H. Lin, C.-D. Kuo, H.-P. Chao, S.-L. Fu, Suppression of v-Src transformation by andrographolide via degradation of the v-Src protein and attenuation of the Erk signaling pathway, J. Biol. Chem. 283 (2008) 5023e5033. [3] (a) J.-X. Chen, H.-J. Xue, W.-C. Ye, B.-H. Fang, Y.-H. Liu, S.-H. Yuan, P. Yu, Y.Q. Wang, Activity of andrographolide and its derivatives against influenza virus in vivo and in vitro, Biol. Pharm. Bull. 32 (2009) 1385e1391; (b) T. Wang, B. Liu, W. Zhang, B. Wilson, J.-S. Hong, Andrographolide reduces inflammation-mediated dopaminergic neurodegeneration in mesencephalic neuron-Gila cultures by inhibiting microglial activation, J. Pharmacol. Exp. Ther. 308 (2004) 975e983; (c) P. Misra, N.L. Pal, P.Y. Guru, J.C. Katiyar, V. Srivastava, J.S. Tandon, Antimalarial activity of Andrographis paniculata (kalmegh) against Plasmodium berghei NK 65 in Mastomys natalensis, Int. J. Pharmacogn. 30 (1992) 263e274; (d) B.-C. Yu, C.-R. Hung, W.-C. Chen, J.-T. Cheng, Antihyperglycemic effect of andrographolide in streptozotocin-induced diabetic rats, Planta Med. 69 (2003) 1075e1079; (e) A. Puri, R. Saxena, R.P. Saxena, K.C. Saxena, V. Srivastava, J.S. Tandon, Immunostimulant agents from Andrographis paniculata, J. Nat. Prod. 56 (1993) 995e999. [4] K. Maiti, K. Mukherjee, V. Murugan, B.P. Saha, P.K. Mukherjee, Enhancing bioavailability and hepatoprotective activity of andrographolide from Andrographis paniculata, a well-known medicinal food, through its herbosome, J. Sci. Food Agric. 90 (2010) 43e51. [5] (a) S.R. Jada, G.S. Subur, C. Matthews, A.S. Hamzah, N.H. Lajis, M.S. Saad, M.F.G. Stevens, J. Stanslas, Semisynthesis and in vitro anticancer activities of andrographolide analogues, Phytochemistry 68 (2007) 904e912; (b) S. Nanduri, V.K.N. Nyavanandi, S.S.R. Thunuguntla, S. Kasu, M.K. Pallerla, P.S. Ram, S. Rajagopal, R.A. Kumar, R. Ramanujam, J.M. Babu, K. Vyas, A.S. Devi, G.O. Reddy, V. Akella, Synthesis and structureeactivity relationships of andrographolide analogues as novel cytotoxic agents, Bioorg. Med. Chem. Lett. 14 (2004) 4711e4717; (c) C. Satyanarayana, D.S. Deevi, R. Rajagopalan, N. Srinivas, S. Rajagopal, DRF 3188 a novel semi-synthetic analog of andrographolide: cellular response to MCF 7 breast cancer cells, BMC Cancer 4 (2004) 26 article no 26; (d) B. Das, C. Chowdhury, D. Kumar, R. Sen, R. Roy, P. Das, M. Chatterjee, Synthesis, cytotoxicity and structureeactivity relationship (SAR) studies of andrographolide analogues as anti-cancer agent, Bioorg. Med. Chem. Lett. 20 (2010) 6947e6950. [6] (a) R. Roy, D. Kumar, B. Chakraborty, C. Chowdhury, P. Das, Apoptotic and autophagic effects of Sesbania grandiflora flowers in human leukemic cells, PLoS One 8 (2013) e71672; (b) P. Das, D. Kumar, R. Roy, C. Chowdhury, M. Chatterjee, Andrographolide analogue induces apoptosis and autophasy mediated cell death in U937 cells, Eur. J. Cancer 48 (Suppl. 5) (2012) S156; (c) V. Srivastava, M.P. Darokar, A. Fatima, J.K. Kumar, C. Chowdhury, H.O. Saxsena, G.R. Dwivedi, K. Shrivastava, V. Gupta, S.K. Chattopadhya, S. Luqman, M.M. Gupta, A.S. Negi, P.S. Khanuja, Synthesis of diverse analogues of oenostacin and their antibacterial activities, Bioorg. Med. Chem. 15 (2007) 518e525. [7] (a) Z. Zhang, G.H.-L. Chan, J. Li, W.-F. Fong, H.-Y. Cheung, Molecular interaction between andrographolide and glutathione follows second order kinetics, Chem. Pharm. Bull. 56 (2008) 1229e1233; (b) H. Yao, S. Li, P. Yu, X. Tang, J. Jiang, Y. Wang, Reaction characteristics of andrographolide and its analogue AL-1 with GSH, as a simple chemical simulation of NF-kB inhibition, Molecules 17 (2012) 728e739. [8] (a) M. Deshmukh, P. Chao, H.L. Kutscher, D. Gao, P.J. Sinko, A series of a-amino acid ester prodrugs of camptothecin: in vitro hydrolysis and A549 human lung carcinoma cell cytotoxicity, J. Med. Chem. 53 (2010) 1038e1047; (b) S.B. Singh, D. Rindgen, P. Bradley, L. Cama, W. Sun, M.J. Hafey, T. Suzuki, N. Wang, H. Wu, B. Zhang, L. Wang, C. Ji, H. Yu, R. Soll, D.B. Olsen, P.T. Meinke, D.A. Nicoll-Griffith, Design, synthesis, and evaluation of prodrugs of Ertapenem, ACS Med. Chem. Lett. 4 (2013) 715e719. [9] The idea of the introduction of the epoxide moiety at C8/C17 double bond of compound 2 leading to the formations of compounds 3e4 stems from the report of Nanduri (see Ref. [5b]) and co-workers who observed significant anti-cancer potency in a number of andrographolide derivatives having epoxide ring at C8/C17 and ester moiety at C14. [10] (a) H.J. Hocker, K.-J. Cho, C.-Y.K. Chen, N. Rambahal, S.R. Sagineedu, K. Shaarl, J. Stanslas, J.F. Hancock, A.A. Gorfe, Andrographolide derivatives inhibit guanine nucleotide exchange and abrogate oncogenic Ras function, Proc. Natl.

106

[11]

[12]

[13]

[14]

[15]

[16]

R. Preet et al. / European Journal of Medicinal Chemistry 85 (2014) 95e106 Acad. Sci. U. S. A. 110 (2013) 10201e10206; (b) D. Chen, Y. Song, Y. Lu, X. Xue, Synthesis and in vitro cytotoxicity of andrographolide-19-oic acid analogues as anti-cancer agents, Bioorg. Med. Chem. Lett. 23 (2013) 3166e3169; (c) S.K. Dey, D. Bose, A. Hazra, S. Naskar, A. Nandy, R.N. Munda, S. Das, N. Chatterjee, N.B. Mondal, S. Banerjee, K.D. Saha, Cytotoxic activity and apoptosis including potential of di-spiropyrrolidino and di-spiropyrrolizidino oxindole andrographolide derivatives, PLoS One 8 (3) (2013) e58055; (d) U. Sirion, S. Kasemsook, K. Suksen, P. Piyachaturawat, A. Suksamrarn, R. Saeeng, New substituted C-19-andrographolide analogues with potent cytotoxic activities, Bioorg. Med. Chem. Lett. 22 (2012) 49e52; (e) M.M. Uttekar, T. Das, R.S. Pawar, B. Bhandari, V. Menon, Nutan, S.K. Gupta, S.V. Bhat, Anti-HIV activity of semisynthetic derivatives of andrographolide and computational study of HIV-1, gp120 protein binding, Eur. J. Med. Chem. 56 (2012) 368e374; (f) X. Haiwei, J. Pingjuan, L. Weiyi, W. Junfeng, L. Hongmin, Synthesis and bioactivities of andrographolide derivatives: new (-)-limonidilactone analogues, Chin. J. Chem. 29 (2011) 2114e2118; (g) C. Mang, S. Jakupovic, S. Schunk, H.-D. Ambrosi, O. Schwarz, J. Jakupovic, Natural products in combinatorial chemistry: an andrographolide-based library, J. Comb. Chem. 8 (2006) 268e274. (a) X. Jiang, P. Yu, J. Jiang, Z. Zhang, Z. Wang, Z. Yang, Z. Tian, S.C. Wright, J.W. Larrick, Y. Wang, Synthesis and evaluation of antibacterial activities of andrographolide analogues, Eur. J. Med. Chem. 44 (2009) 2936e2943; (b) Z. Wang, P. Yu, G. Zhang, L. Xu, D. Wang, L. Wang, X. Zeng, Y. Wang, Design, synthesis and antibacterial activity of novel andrographolide derivatives, Bioorg. Med. Chem. 18 (2010) 4269e4274. X. Hai-Wei, D. Gui-Fu, L. Gai-Zhi, W. Jun-Feng, L. Hong-Min, Synthesis of andrographolide derivatives: a new family of a-glucosidase inhibitors, Bioorg. Med. Chem. 15 (12) (2007) 4247e4255. A. Robinson, V.K. Aggarwal, Stereocontrolled asymmetric synthesis of syn-E1,4-diol-2-enes using allyl boronates and its application in the total synthesis of solandelactone F, Org. Biomol. Chem. 10 (2012) 1795e1801. R. Preet, P. Mohapatra, S. Mohanty, S.K. Sahu, T. Choudhuri, M.D. Wyatt, C.N. Kundu, Quinacrine has anticancer activity in breast cancer cells through inhibition of topoisomerase, Int. J. Cancer 130 (2012) 1660e1670. P. Mohapatra, R. Preet, M. Choudhuri, T. Choudhuri, C.N. Kundu, 5-Fluorouracil increases the chemopreventive potentials of resveratrol through DNA damage and MAPK signaling pathway in human colorectal cancer cells, Oncol. Res. 19 (2011) 311e321. A.T. Baviskar, C. Madaan, R. Preet, P. Mohapatra, V. Jain, A. Agarwal, S.K. Guchhait, C.N. Kundu, U.C. Banerjee, P.V. Bharatam, N-fused imidazoles as novel anticancer agents that inhibit catalytic activity of topoisomerase IIa and induce apoptosis in G1/S phase, J. Med. Chem. 54 (2011) 5013e5030.

[17] M. Kashyap, D. Das, R. Preet, P. Mohapatra, S.R. Satapathy, S. Siddharth, C.N. Kundu, S.K. Guchhait, Scaffold hybridization in generation of indenoindolones as anticancer agents that induce apoptosis with cell cycle arrest at G2/M phase, Bioorg. Med. Chem. Lett. 22 (2012) 2474e2479. [18] S. Siddharth, P. Mohapatra, R. Preet, D. Das, S.R. Satapathy, T. Choudhuri, C.N. Kundu, Induction of apoptosis by 4-(3-(tert-butylamino)imidazo[1,2-a] pyridine-2-yl) benzoic acid in breast cancer cells via upregulation of PTEN, Oncol. Res. 21 (2013) 1e13. [19] R. Karsten, Analysis of proteinDNA binding at equilibrium, B. I. F. Futura 12 (1997) 20e26. [20] S.K. Samanta, D. Dutta, S. Roy, K. Bhattacharya, S. Sarkar, A.K. Dasgupta, B.C. Pal, C. Mandal, C. Mandal, Mahanine a DNA minor groove binding agent exerts cellular cytotoxicity with involvement of C-7-OH and eNH functional groups, J. Med. Chem. 56 (2013) 5709e5721. [21] S. Kashanian, A.T. Ghobadi, DNA interaction studies of ethylene diamine tetraacetic acid food additive and selenium effect in DNA cleavage inhibition, DNA Cell. Biol. 30 (12) (2011) 1085e1090. [22] S.Y. Breusegem, R.M. Clegg, F.G. Loontiens, Base sequence specificity of Hoechst 33258 and DAPI binding to five (A/T)4 DNA sites with kinetic evidence for more than one high-affinity Hoechst 33258-AATT complex, J. Mol. Biol. 315 (2002) 1049e1061. [23] (a) K.V. Hecke, P.C. Nam, M.T. Nguyen, L.V. Meervelt, Netropsin interactions in the minor groove of d(GGCCAATTGG) studied by a combination of resolution enhancement and ab initio calculations, FEBS J. 272 (2005) 3531e3541; (b) A.Y. Chen, C. Yu, B. Gatto, L.F. Liu, DNA minor groove-binding ligands: a different class of mammalian DNA topoisomerase I inhibitors, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 8131e8135. €hm, J. Feld, D. Scherbl, F. Berger, K.H. Merz, A. Lampen, [24] N. Watzek, N. Bo T. Reemtsma, S.R. Tannenbaum, P.L. Skipper, M. Baum, E. Richling, G. Eisenbrand, N7-Glycidamide-guanine DNA adduct by orally ingested acrylamide in rats: a doseeresponse study encompassing human diet-related exposure levels, Chem. Res. Toxicol. 25 (2012) 381e390. [25] Y. Hayashi, M. Skwarczynski, Y. Hamada, Y. Sohma, T. Kimura, Y. Kiso, A novel approach of water-soluble prodrug with no auxiliary and no byproduct: design and synthesis of Isotaxel, J. Med. Chem. 46 (2003) 3782e3784. [26] Though epoxides are considered as structural alert in drug discovery, we chose epoxidation in the derivatizations of the andrographolide; the idea is that once the lead compound having epoxide moiety is identified, the epoxy group could be used as synthetic handle for further lead optimization. [27] M. Rajani, N. Shrivastava, M.N. Ravishankara, A rapid method for isolation of andrographolide from Andrographis paniculata Nees (kalmegh), Pharm. Biol. 38 (2000) 204e209.