Bioorganic & Medicinal Chemistry 22 (2014) 1104–1114

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry journal homepage: www.elsevier.com/locate/bmc

Design, synthesis and ex-vivo release studies of colon-specific polyphosphazene–anticancer drug conjugates Rajiv Sharma a,b,⇑, Ravindra K. Rawal a,c, Manav Malhotra a, A. K. Sharma d, T. R. Bhardwaj a a Polymer Chemistry and Technology Research Laboratory, Department of Pharmaceutical Chemistry, Indo-Soviet Friendship (I.S.F), College of Pharmacy, Ferozepur Road, Moga 142 001, India b Research Scholar, Uttarakhand Technical University, Dehradun 248 007, India c The University of Georgia, College of Pharmacy, GA 30602, USA d School of Pharmacy, Asmara College of Health Sciences, Asmara, Eritrea

a r t i c l e

i n f o

Article history: Received 9 August 2013 Revised 14 December 2013 Accepted 16 December 2013 Available online 25 December 2013 Keywords: Azoreductase Polyphosphazene Colon-specific Methotrexate Gemcitabine

a b s t r a c t Colon-specific azo based polyphosphazene–anticancer drug conjugates (11–18) have been synthesized and evaluated by ex-vivo release studies. The prepared polyphosphazene drug conjugates (11–18) are stable in acidic (pH = 1.2) buffer which showed that these polymer drug conjugates are protected from acidic environment which is the primary requirement of colon specific targeted drug delivery. The ex-vivo release profiles of polyphosphazene drug conjugates (11–18) have been performed in the presence as well as in the absence of rat cecal content. The results showed that more than 89% of parent drugs (methotrexate and gemcitabine) are released from polymeric backbone of polyphosphazene drug conjugates (14 and 18) having n-butanol (lipophilic moiety). The in-vitro cytotoxicity assay has also been performed which clearly indicated that these polymeric drug conjugates are active against human colorectal cancer cell lines (HT-29 and COLO 320 DM). The drug release kinetic study demonstrated that Higuchi’s equation is found to be best fitted equation which showed that release of drug from polymeric backbone as square root of time dependent process based on non-fickian diffusion. Therefore, the synthesized polyphosphazene azo based drug conjugates of methotrexate and gemcitabine are the potential candidates for colon targeted drug delivery system with minimal undesirable side effects. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Cancer is a group of diseases characterized by uncontrolled growth and spread of abnormal cells. If the spread is not controlled, it can result in death. Cancer is the second leading cause of deaths worldwide. In 2013, about 580,350 deaths are expected due to cancer; almost 1600 people died per day in USA.1 Cancers are of many types like breast cancer, prostate cancer, lung cancer, colorectal cancer and leukemia etc. Among cancers, colorectal cancer is the third most common cause of deaths in both men and women in USA. According to the American Cancer Society, about 141,210 people were diagnosed with colorectal cancer resulted to 49,380 people death. The colon specific targeted drug delivery has been developed as one of the most successful approach.2 Colon has become the center of attraction for the treatment of colonic diseases such as crohn’s diseases, ulcerative colitis and inflammatory bowel diseases and colon cancer.3–5 The prodrug also has the potential for the delivery of proteins and peptides which are sensitive to the enzymes both

⇑ Corresponding author. Tel.: +91 9988080822. E-mail address: [email protected] (R. Sharma). 0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bmc.2013.12.037

in stomach and intestine.6 There are mainly two types of approaches to deliver a drug to colon: (i) covalent linkage of drug with a carrier; (ii) delivery of intact drug to the colon. The covalent linkage of drug with carrier includes prodrug approach, azo bond conjugates, glycoside conjugates, glucuronide conjugates, cyclodextrin conjugates, amino acid conjugates and polymeric prodrug approach. The delivery of intact form of drug includes coating with pH sensitive polymers, biodegradable polymers, redox-sensitive polymers and embedding in biodegradable matrices and hydrogels etc.7–9 In the prodrug approach, drug is covalently bound to the carrier that improve the physicochemical properties by increasing the drug concentration at the target site, decrease toxicity and undesirable side effects.2 To achieve the successful delivery of drugs to the target site i.e. colon drug needs to be protected from the gastrointestinal tract (GIT) environment or absorption in the upper GIT and then releases in the target site.10 It is well known that colonic microflora has large number of anaerobic bacteria’s (about 108–109 bacterial count/g gut contents in rats and 1010–1012 bacterial count/g gut contents in humans) which is not present in the rest part of GIT such as b-glycosidase, b-glucuronidase, nitroreductase, nitrate reductase and azoreductase.11,12 There are large numbers of enzyme based prodrugs available but none of

1105

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

Figure 1. Structures of azo based prodrugs of 5-aminosalicylic acid (5-ASA) (1): sulfasalazine (2); balsalazide (3); ipsalazine (4); olsalazine (5).

Polymeric Backbone NH2 N

N

N

NH2

NH2

N

N

Linker

N

N

CH3

O

HO O

COOH

OH

F F

CONHCHCH2CH2COOH

N

(6)

(7)

N N N

Drug

N N

Figure 2. Technical approach for the design of polymer linked azo based prodrugs of anticancer agents.

OH

N

N N N N

OH

N N

N

CH3

O

HO O

them reached to clinically useful as azo prodrug of 5-aminosalicyclic acid (5-ASA) (1) such as sulphasalazine (2)13 and its analogues balsalazide (3),14 ipsalazine (4)14 and olsalazine (5)15 which were used for the treatment of inflammatory bowel disease (Fig. 1). The reason behind the success of 5-ASA derivatives is due to the presence of azo bond in these molecules and it has been found that the azo bond was reduced specifically to amine by the colonic microflora which is present in cecum and feces of human and rats.16 Biodegradable polymers have been widely used in the area of biomedical applications such as polylactides (PLA), polyglycolides (PGA), poly (lactide-co-glycolides) (PLGA), polyanhydrides, polyorthoesters, poly (2-hydroxy ethyl methacrylate), poly (methyl methacrylate), poly (vinyl alcohol), polyacrylamide, poly (ethylene glycol), poly (methacrylic acid),17 chitosan, guargum, dextran, cyclodextrins8 and polyphosphazene18 etc. Among them, polyphosphazene (10) is a class of inorganic biocompatible and biodegradable polymer which has alternative nitrogen and phosphorus atom attached by single and double bond. Polyphosphazene polymer has versatile nature of polymer because of its two chlorine atoms attached on both side of phosphorus atom which can be easily replaced by nucleophilic substitution reactions. The substitution attached on both side of phosphorus in place of chlorine plays an important role in the physicochemical properties of the polymer. A plenty of research work has been done to explore the biomedical applications of polyphosphazene, for example, in vaccine delivery and immunomodulation,19–21 tissue engineering,22 polyphosphazene drug conjugates in anticancer chemotherapy like

OH

N

COOCH3

F

OH F

CONHCHCH2CH2COOCH3 (8)

(9)

Figure 3. Chemical structures of methotrexate (6), gemcitabine (7), azo prodrugs of methotrexate (8) and gemcitabine (9).

polyphosphazene–paclitaxel conjugate,23 polyphosphazene– 24 doxorubucin conjugate, polyphosphazene–platinum(II) conjugates25–29 and polyphosphazene–camptothecin conjugates30 etc. Polymeric azo based prodrugs approach is one of the most important approach for the targeted delivery of anticancer drugs to the colon. In this approach, drug molecules are covalently bounded with polymeric backbone with an azo based drug carrier (Fig. 2). The targeted delivery of drug to the colon depends upon various factors like nature of polymeric backbone and drug carrier which protect the drug from upper GIT environment and also maintain the physicochemical properties of the polymer drug conjugate.31 In this present work, methotrexate (6) and gemcitabine (7) (Fig. 3) were used as model drugs. Methotrexate and gemcitabine are potent anticancer agents which have been used in the treatment of colorectal cancer alone or in combination with other drugs like 5-Flurouracil (5-FU)32–34 but due to their absorption in the upper GIT and less oral bioavailability, these drugs cannot be considered as safer drugs for patients suffering from colorectal cancer. Therefore, colon-specific azo based prodrugs of anticancer agents like methotrexate, gemcitabine have been synthesized and

1106

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

coupled with phenol.18 The low molecular weight (Mw = 104–105) poly(dichlorophosphazene) was obtained as per the literature procedure36 and it was used without any purification. The two chlorine atoms attached on the polyphosphazene backbone was replaced by equimolar substitution with different less lipophilic (methanol and methylamine) and more lipophilic groups (p-aminobenzoate and n-butanol) and second chlorine was replaced with azo based prodrugs of methotrexate (8) or gemcitabine (9) to get the desired polymer drug conjugates (11–18). The synthesized polyphosphazene drug conjugates (11–18) were characterized by modern analytical techniques such as FTIR, 1H NMR, 31P NMR and Gel Permeation Chromatography (GPC). IR spectroscopy showed characteristics stretching near 1589– 1640 cm 1 (N@N), 1228-1235 cm 1 (P@N) which demonstrate the formation of drug conjugates with the substituted polymer. In 1H NMR, the characteristics chemical shifts observed for conjugates (11–18) were the protons of aromatic region d (6.54–7.89), pteridine nucleus d (8.12–8.23), pyrimidine d (4.43), ribose d (3.93) which were the main key features of conjugation. The 31P NMR observed the peaks for phosphorus atom, which were near d ( 3.2 to 9.5) and further confirmed the attachment of polyphosphazene with drug conjugates. GPC was used to determine the

evaluated by our research group. The synthesized compounds (8, 9) (Fig. 3) showed better stability in the upper GIT and have satisfactory release behavior (60–70%) in the rat fecal content.35 In continuation of our research work, we thought to explore the concept of polymeric drug conjugate with azo based linkage between the drug and polymer to have targeted delivery of these potent anticancer agents to the colon site. Therefore, in this communication, we have synthesized the appropriately substituted polymeric azo based prodrugs of methotrexate and gemcitabine (Fig. 2) to have targeted drug delivery to the colon and also maintain its physiochemical properties. 2. Results and discussion 2.1. Chemistry Polyphosphazene drug conjugates (11–18) were synthesized as shown in Schemes 1 and 2. [(2S)-2-[(4-{[(2,4-bis(4-hydroxyphen1yldiazo)]pteridin-6-yl)methyl](methyl)amino}benzoyl)amino]-1, 5-dimethylpentonoate (8) and [4-(4-hydroxyphen-1-yldiazo)-1(3,3-difluoro-tetrahydro-4-hydroxy-5-(hydroxymethyl)furan-2-yl)pyrimidin-2-one] (9) were synthesized by diazotization and

N

N

O

N

N CH 3OH

N N

Cl

N N

H3CN

O

(8)

N P OCH 3 N P Cl N P OCH 3

COOCH 3 n

CONHCHCH 2CH 2COOCH 3

(11)

N

*

Cl N P Cl

N

* n

O

N

N NH 2CH 3

N N

Cl

N N

H3CN

O

(8)

N P NHCH 3 N P Cl N P NHCH 3

COOCH 3

(10)

N

N

n

CONHCHCH 2CH 2COOCH 3

(12)

N N

O

N

N

Cl

N N

H3CN

O

p-NH 2-C 6H4COOCH 3

COOCH 3

Cl NH

COOCH 3

N N H3CN

N

n

CONHCHCH 2CH 2COOCH 3

(8)

H N

COOCH 3

(8)

C4H9OH

N P N P N P

(13)

N N

O

N

Cl

N N O

N P OCH 2CH 2CH 2CH 3 N P Cl N P OCH 2CH 2CH 2CH 3

COOCH 3 n

CONHCHCH 2CH 2COOCH 3

(14)

Scheme 1. Synthesis of polyphosphazene azo based drug conjugates of methotrexate (11–14).

1107

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

N O P OCH3

N N

n

O

HO

N

(15)

N

n

N N

N O P NHCH3

N N

O

(16)

HO

O F

O F

OH F CH3OH

NH2CH3

(9)

*

Cl N P Cl

OH F (9)

* n

(10) p-NH2-C6H4COOCH3

C4H9OH

N O P OCH2CH2CH2CH3

N N COOCH3

N

n

N

N

O

HO

O

O F

HO O F

(9)

n

N O P NH

N N

N

(9)

(17)

OH F

(18)

OH F

Scheme 2. Synthesis of polyphosphazene azo based drug conjugates of gemcitabine (15–-18).

distribution of molecular weight of polymer. All the synthesized drug conjugates were in the range of low molecular weight distribution, that is, Mw = 2.0–2.5  105 daltons with polydispersity of 1.0. 2.2. Physiological stability ex-vivo There is a great challenge for colon targeting azo prodrugs to protect themselves while passing from stomach and small intestine as an intact form to reach the colon site where polymeric prodrugs have to be degraded by colonic microfloral enzyme, that is, azoreductase. Therefore, to explore the behavior of polymeric prodrugs in acidic and basic environment, stability studies was carried out in acidic buffer pH = 1.2 (0.1 N HCl) and sorenson’s buffer having pH = 7.4 (SIF = simulated intestinal fluid). The sorenson’s buffer contains dibasic sodium phosphate heptahydrate (Na2HPO4
7H2O) and monobasic potassium phosphate (KH2PO4) with the ratio of 4.1:1.0 respectively. Ex-vivo stability studies showed that polyphosphazene linked azo prodrugs (11, 12, 15 and 16) having less lipophilic substitution (methanol and methylamine) at polymeric backbone showed their release between (3 to 7%) in simulated gastric fluid (SGF) (pH = 1.2; 0.1 N HCl) at first 2 h as shown in Figures 4a and 5a whereas the release of drug from polyphosphazene drug conjugates (13, 14, 17 and 18) having more lipophilic substitution (p-aminobenzoate and n-butanol) were 10–14% at SIF at pH = 7.4 (sorenson’s buffer after 12 h as shown in Figures 4b and 5b. As we know that the azo bond are synthesized in the acidic conditions and are stable to hydrolysis as compared to basic medium.38–40 In case of in-vitro % of released drug to check the stability of polymer drug conjugates in acidic medium, the study was carried out for 12 h. Out of 12 h, our concern is upto 2 h because the transit time of stomach is about 2 h. The colon targeted drug should be protected from the acidic environment. We observed that our polymer drug conjugates (11–18) substituted with different substitutions (lipophilic or less lipophilic) showed only 3–7% after 2 h of study and this release was because of 2 h exposure of drug conjugates with the acidic environment because in

Figure 4. (a) The % of released drug methotrexate (6) from polyphosphazene drug conjugates of methotrexate (11–14) in acidic pH = 1.2 (0.1 N HCl) (b) The % of released drug methotrexate (6) from polyphosphazene drug conjugates of methotrexate (11–14) in alkaline pH = 7.4 sorenson’s buffer at 37 °C. All values are expressed as mean ± S.D (n = 6).

this environment the nitrogen atoms of the azo bond (N@N) get protonated which weakens the azo bond. The 3–7% of released

1108

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

dipotassium hydrogen phosphate (K2HPO4) in a ratio of 1:4.3 respectively) for 19 h with or without rat cecal content (4% w/w). This was done for mimicking the conditions from mouth to colonic transit time. The release pattern of polyphosphazene drug conjugates having less lipophilic substitution (methanol (CH3OH), methylamine (NH2CH3)) at polymeric backbone were as follows: (11 (4.65% (2 h), 15.34% (3 h), 33.54% (24 h)), (12 (3.67% (2 h), 19.86% (3 h), 43.45% (24 h)), (15 (6.32% (2 h), 15.18% (3 h), 34.21% (24 h) and (16 (8.35% (2 h), 16.38% (3 h), 46.32% (24 h)) in the absence of rat cecal content (Figs. 6a and 7a) whereas in the presence of 4% w/w rat cecal content the decreasing order of their release pattern was, (16) (81.27%) > (15) (77.24%) > (12) (60.26%) > (11) (53.28%) as shown in Figures 6b and 7b. In case of more lipophilic substitution (p-aminobenzoate and n-butanol) at polymeric backbone of methotrexate and gemcitabine conjugates, the drug release pattern in the absence of rat cecal content was as follows: (13 (2.67% (2 h), 28.48% (3 h), 58.87% (24 h)), (14 (3.27% (2 h), 18.25% (3 h), 58.87% (24 h)), (17 (3.26% (2 h), 18.25% (3 h), 53.25% (24 h)) and (18 (2.54% (2 h), 19.27% (3 h), 62.15% (24 h)) (Figs. 6a and 7a). In the presence of 4% w/w rat cecal content the release pattern of polymeric drug conjugates (13, 14, 17 and 18 were 85.23%, 91.48%, 84.36%, 89.76% respectively as shown in Figures 6b and 7b. The ex-vivo evaluation of different polymeric azo based drug conjugates having different substituents were

Figure 5. (a) The % of released drug gemcitabine (7) from polyphosphazene drug conjugates of gemcitabine (15–18) in acidic pH = 1.2 (0.1 N HCl) (b) The % of released drug gemcitabine (7) from polyphosphazene drug conjugates of gemcitabine (15–18) in alkaline pH = 7.4 sorenson’s buffer at 37 °C. All values are expressed as mean ± S.D (n = 6).

drug is acceptable because of very less amount of drug being released in this environment. In case of pH 7.4, the experiment was carried out for 12 h, but our concern is upto 5 h because of the transit time for small intestine is about 5 h and it should also be protected or less released of drug in the small intestine. The observed % of released drug was about 10–14% after 5 h, this is higher than acidic environment due to unstability of azo bond in basic environment because of hydrolysis of azo bond in this medium. After 5 h of exposure to this medium the drug release was 10–14% which is less only due to attached different polymeric backbone, otherwise it may increase the release pattern of azo compounds upto 40–60% after 5 h. Therefore, the observed decomposition of azo bond in acidic and basic medium is depends on the time of exposure of these polymeric drug conjugates in acidic and basic medium and also on the properties of polymeric backbone attached with it. 2.3. Ex-vivo drug release 2.3.1. Ex-vivo drug release with and without rat cecal content In order to estimate the actual release behavior of polymeric drug conjugates to the targeted site, that is, colon, an ex-vivo release studies was carried out with or without rat cecal content. The polymeric azo based drug conjugates of methotrexate (11–14) and gemcitabine (15–18) were carried out in SGF, that is, at pH = 1.2 (0.1 N HCl) for 2 h, in SIF at pH = 7.4 (sorenson’s buffer; dibasic sodium phosphate heptahydrate (Na2HPO4
7H2O) and monobasic potassium phosphate (KH2PO4) with the ratio of 4.1:1.0 respectively) for 3 h and then at pH = 6.8 (phosphate buffer saline; potassium dihydrogen phosphate (KH2PO4) and

Figure 6. (a) Cumulative % of released methotrexate (6) from polyphosphazene drug conjugates (11–14) during incubation at 37 °C at pH = 1.2 (0.1 N HCl) for 2 h in simulated gastric fluid (SGF), at pH = 7.4 (sorenson’s buffer) for 3 h in simulated intestinal fluid (SIF) and pH = 6.8 (phosphate buffer saline, PBS) for 19 h without rat cecal content; (b) cumulative % release of methotrexate (6) from polyphosphazene drug conjugates (11–14) during incubation at 37 °C at pH = 1.2 (0.1 N HCl) for 2 h in simulated gastric fluid (SGF) at pH = 7.4 (sorenson’s buffer) for 3 h in simulated intestinal fluid (SIF) and pH = 6.8 (phosphate buffer saline, PBS) for 19 h with rat cecal content. All values are expressed as mean ± S.D (n = 6).

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

1109

that the maximum release was observed only in rat cecal content, other release was due to exposure to acidic medium for 2 h where azo bond get protonated and weaken. Whereas in case of pH 6.8 (without rat cecal content) more release was observed because of long exposure, that is, 19 h to this medium which is just control test. Further, the actual and appropriate release will be observed only after in-vivo study. 2.3.2. Ex-vivo drug release in different intestinal contents The different substituted polyphosphazene drug conjugates (11–18) were incubated with SI mucosa, SI content, LI mucosa and rat cecal content at 37 °C for 12 h. The release pattern of different substituted polymeric drug conjugates in intestinal contents was as shown in Figures 8 and 9. The results of conjugates showed that less than 6% of release was in SI content, up to 8% in SI mucosa and less than 10% of release was observed in LI mucosa. The increase release of drug in case of LI mucosa may be due to fecal contamination. In the presence of rat cecal content the release was more than 80% to 95% which showed the specificity of enzymes azoreductase present in the cecal content which reduced the azo bond and released the drug. The rat SI content and SI mucosa was used to simulate upper GI tract conditions and LI mucosa and cecal content was used to simulate the colonic conditions. The ex-vivo release study in different intestinal contents of rats confirmed that the polyphosphazene drug conjugates were stable in upper GIT. The maximum release of these drug conjugates was only in rat cecal content, which showed that these polymer drug conjugates are colon specific. 2.4. In-vitro drug release kinetics Figure 7. (a) Cumulative % of released gemcitabine (7) from polyphosphazene drug conjugates (15–18) during incubation at 37 °C at pH = 1.2 (0.1 N HCl) for 2 h in simulated gastric fluid (SGF) at pH = 7.4 (sorenson’s buffer) for 3 h in simulated intestinal fluid (SIF) and pH = 6.8 (phosphate buffer saline, PBS) for 19 h without rat cecal content; (b) cumulative % of released gemcitabine (7) from polyphosphazene drug conjugates (15–18) during incubation at 37 °C at pH = 1.2 (0.1 N HCl) for 2 h in simulated gastric fluid (SGF) at pH = 7.4 (sorenson’s buffer) for 3 h in simulated intestinal fluid (SIF) and pH = 6.8 (phosphate buffer saline, PBS) for 19 h with rat cecal content. All values are expressed as mean ± S.D (n = 6).

evaluated which showed that all conjugates (11–18) have released a lesser amount of drug in the stomach and small intestine but release was only in the colonic pH, that is, 6.8 either in absence or presence of rat cecal content. In the presence of 4% w/w cecal content the drug release was maximum in case of polymeric drug conjugates which was due to the presence of enzymes azoreductase in the rat cecal content. The substitutions at polymeric backbone also have great influence on their release pattern as mentioned in the above section. Among all, polymer drug conjugates having n-butanol as more lipophilic nature (14 and 18) attached to the polymeric backbone are better candidates for the colon targeted delivery system. The release of polymer drug conjugates (11–18) in the colonic pH = 6.8 is only to mimic the conditions of colon, the study was carried out upto 24 h, after 24 h of exposure to the medium of pH = 6.8, the observed release was (11 (33.54%), 12 (43.45%), 13 (58.87%), 14 (58.87%), 15 (34.21%), 16 (46.32%) , 17 (53.25%), 18 (62.15%) where as in the presence of rat cecal content the release was (11 (53.28%), 12 (60.26%), 13 (85.23%), 14 (91.48%), 15 (77.24%), 16 (81.27%), 17 (84.27%) and 18 (89.76%). The % of released drug was compared with and without rat cecal content. The release in case of without rat cecal content was only for comparison point of view, but our main aim was to deliver the drug at colon by passing the stomach. We also checked the release of drug in different intestinal contents to clarify the comparison of cecum with other intestinal contents (LI mucosa, SI content, SI mucosa) as shown in Figures 8 and 9. Therefore, it is cleared from the results

In order to explain the site specific nature of polymer drug conjugates, there is essential to study the drug release kinetics of synthesized polyphosphazene azo based drug conjugates of methotrexate and gemcitabine. It has been seen that the polymeric drug conjugates were fitted to various kinetics models such as zero order, first order, Higuchi, Korsmeyer and peppas. The zero-order rate shows that the release of drug conjugates is independent of its concentration. The first order describes that the release is concentration dependent whereas Higuchi model shows the release of drug conjugates is square root of time dependent process based on non-fickian diffusion. The release constant was calculated from respective graphs where (r2) is the regression coefficient. The best linearity was found in Higuchi model having (r2) value was between 0.967–0.998, that is, close to one. The release kinetics indicated that release of drug from polymeric backbone as square root of time dependent process based on non-fickian diffusion as shown in Figure 10. 2.5. An azoreductase assay An azoreductase assay was performed on all the synthesized polyphosphazene drug conjugates. As we know that azo dye are reduced to amines by the action of enzyme azoreductase on the azo bond. More than 70% of fresh human faeces contain colonic microflora having anaerobic bacteria’s such as azoreductases. Therefore, fresh human faeces were used as a source of azoreductases. Gram anaerobic medium (GAM) was prepared according to the literature procedure.35 The prepared culture of GAM with fresh human faeces was assayed by incubation the azo dye amaranth which was reduced at kmax 520 nm but it was absent in flora free medium. Further, all the synthesized polyphosphazene azo based drug conjugates were tested on prepared intestinal flora medium and analysed by U.V. spectrophotometer. Polymer azo based drug conjugates of methotrexate (11–14) were reduced at kmax 256 nm whereas conjugates of gemcitabine (15–18) were reduced

1110

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

Figure 8. The % of released methotrexate (6) from polyphosphazene drug conjugates. (a) 11; (b) 12; (c) 13 and (d) 14 in different intestinal contents (cecal contents, LI mucosa, SI mucosa and SI content).

at kmax 284 nm. The reduction of azo bond to release drug from polymeric conjugate system confirmed that azo bond was reduced only by the action of azoreductases enzyme present in intestinal microflora. Therefore these polymer azo based drug conjugates will be beneficial for colon targeting.

same level or more potent of cytotoxic activities with methotrexate and gemcitabine (Tables 1 and 2) without activation of cecal content.

2.6. In-vitro cytotoxicity assay

In conclusion, polymeric drug conjugate is one of the most important approach for the targeting the drugs to colon in which drug is covalently bound to the polymer (carrier) via a linker that is susceptible to cleavage in the colon with the improvement of the physicochemical properties by increasing the drug concentration at the target site, decrease toxicity and undesirable side effects. Therefore, different substituted polyphosphazene drug conjugates of anticancer agents like methotrexate, gemcitabine have been synthesized and characterized by modern analytical techniques like FT-IR, 1H NMR, 31P NMR and GPC. The prepared polymeric drug conjugates were stable in acidic (pH = 1.2) and basic (pH = 7.4) buffers which showed their stability in upper GIT environment. Further, ex vivo release profile of polyphosphazene drug conjugates (11–18) was performed with or without rat cecal content and other parts of intestinal content (SI mucosa, SI content and LI mucosa). The release behavior of polymeric drug conjugates of methotrexate and gemcitabine were more than 85–90% after 24 h incubation at 37 °C in which polymer having more lipophilic backbone such as n-butanol and p-aminobenzoate in the replacement of chlorine atoms of polyphosphazene by nucleophilic substitution reaction. The drug release kinetics of polyphosphazene linked azo based drug conjugates were also calculated which showed that the release of drug conjugates is square root of time dependent process based on non-fickian diffusion in Higuchi model. The in-vitro cytotoxicity assay was performed which clearly indicated that these polymeric drug conjugates with more lipophilic were active against colorectal cancer cell lines (HT-29

In order to examine the in vitro cytotoxicity on the different substituted polyphosphazene drug conjugates, human colorectal cancer cell lines (HT-29 and COLO 320 DM) has been used for the MTT assay. The results are summarized in Tables 1 and 2. The polyphosphazene polymeric conjugates of methotrexate prodrug (13 & 14) showed greater cytotoxicity against cancer cell lines (HT-29 and COLO-320 DM) as compared to methotrexate prodrug (8) and methotrexate (6) as shown in Table 1. But the less lipophilic polymeric conjugates of methotrexate prodrug (11 & 12) were having almost same cytotoxicity as methotrexate prodrug (8) and methotrexate (6). Similarly, the polymeric conjugates of gemcitabine prodrug (17 & 18) showed better cytotoxicity as compared to gemcitabine prodrug (9) and gemcitabine (7) as mentioned in Table 2. So on the basis of in-vitro cytotoxicity studies of polyphosphazene drug conjugates of methotrexate (11–14) and gemcitabine (15–18), it may be concluded that more lipophilic groups such as p-aminobenzoate and n-butanol group attached to the polyphosphazene polymeric backbone exhibited greater cytotoxicity as compared to less lipophilic moiety like methylamine and methoxy (OCH3). These lipophilic groups may enhance the penetration in the cell line which in results helps to increase the potency of these polymeric conjugates (11–14 & 15–18) of methotrexate and gemcitabine prodrugs (8 & 9). So, this may be the reason that polyphosphazene drug conjugates (11–18) as well as prodrugs (8) and (9) demonstrated the

3. Conclusions

1111

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

Figure 9. The % of released gemcitabine (7) from polyphosphazene drug conjugates (a) 15; (b) 16; (c) 17 and (d) 18 in different intestinal contents (cecal contents, LI mucosa, SI mucosa and SI content).

% cumulative release

Higuchi Graph Between % release vs Sqroot Time 60 40

(15) (16) (17) (18) (9) Gemcitabine (7)

30 20 10 0 0.000

y = 10.175x + 3.4542 R² = 0.999 2.000 4.000 Square root of Time (hrs)

6.000

Table 1 In-vitro cytotoxicity of polyphosphazene drug conjugates of methotrexate (11–14) on human colorectal cancer cell lines (HT-29 and COLO-320 DM)

a

HT-29

COLO-320 DM

2.79 ± 0.15 2.62 ± 0.87 0.83 ± 0.13 1.26 ± 0.08 2.61 ± 0.25 1.95 ± 0.04

4.08 ± 0.93 3.82 ± 0.72 2.51 ± 0.09 2.66 ± 0.17 3.72 ± 0.15 3.92 ± 0.05

Cytotoxicity represented as inhibitory concentration in 50% of cell growth.

and COLO 320 DM) as compared to their parent molecules. Therefore, the synthesized polyphosphazene drug conjugates of anticancer agents are the potential candidates for the targeted drug delivery system. However, further in-vivo experiments will be carried out soon to confirm the results obtained by ex-vivo studies on polymeric drug conjugates. 4. Experimental

IC50a (lM)

Compound

(11) (12) (13) (14) (8) Methotrexate (6)

IC50a (lM)

Compound

50

Figure 10. Higuchi release model of polyphosphazene drug conjugates.

a

Table 2 In-vitro cytotoxicity of polyphosphazene drug conjugates of gemcitabine (15–18) on human colorectal cancer cell lines (HT-29 and COLO-320 DM)

HT-29

COLO-320 DM

3.92 ± 0.18 3.72 ± 0.11 2.45 ± 0.75 2.79 ± 0.13 3.92 ± 0.10 3.45 ± 0.27

6.76 ± 0.75 5.15 ± 0.70 3.92 ± 0.56 4.04 ± 0.03 5.74 ± 0.21 5.25 ± 0.43

Cytotoxicity represented as inhibitory concentration in 50% of cell growth.

All the reagents and chemicals were purchased from Sigma Aldrich, Loba and CDH, India and used after redistillation. The IR spectra (KBr) cm 1 was obtained with a Perkin–Elmer 1600 FTIR spectrometer in KBr pellets. 1H NMR spectra (d, ppm) were recorded in DMSO-d6 solutions on Bruker Avance II 400 spectrometer at 400 MHz using tetramethylsilane as the internal reference. Molecular weight distribution as well as determination of polymer (polyphosphazene) was monitored with Water’s GPC with THF

1112

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

Columns. Methotrexate and gemcitabine were procured from Panchsheel Organics Ltd, Indore and Arch Pharmalabs Ltd Thane, Maharashtra respectively. 4.1. Typical experimental procedure for the synthesis of substituted polyphosphazene drug conjugates (11–18)

7.75 (s, 2H, Ar-H, J = 8.1), 7.12 (m, 4H, Ar-H), 6.78 (d, 2H, Ar-H, J = 8.1), 6.91 (m, 4H, Ar-H), 4.78 (s, 2H, CH2), 4.45 (q, 1H, C-H), 3.61 (s, 10H, CH2, CH3), 2.34–2.38 (m, 12H, CH2), 2.03 (s, 3H, N-CH3), 1.98 (m ,6H, CH3); 31P NMR (400 MHz, DMSO-d6, d ppm): 8.7 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.4  105 daltons. 4.2. General procedure for the synthesis of substituted polyphosphazene linked azo prodrugs of gemcitabine (15–18)

General procedure for the synthesis of substituted polyphosphazene linked azo prodrugs of methotrexate (11–14) Poly(dichloro)phosphazene (10) (1.0 g, 1.45 mmol) was dissolved in tetrahydrofuran (20.0 mL) and was added to a solution of (2S)-2-[(4-{[(2,4-bis(4-hydroxyphenyl-1yldiazo)]pteridin-6-yl) methyl](methyl)amino}benzoyl)amino]-1,5-dimethylpentonoate (8) (1.0 g, 1.45 mmol) in tetrahydrofuran (20.0 mL). To this methanol (1.45 mmol) for (11), methylamine solution (1.45 mmol) for (12), p-aminobenzoate (1.45 mmol) for (13) and n-butanol (1.45 mmol) for (14), triethylamine (0.17 mL) in tetrahydrofuran (5.0 mL) was added under the atmosphere of nitrogen over a period of 30 min. The reaction mixture was refluxed for 48 h and allowed to stand at 25 °C for 3 days. The precipitated material was filtered; filtrate was concentrated under reduced pressure and precipitated with petroleum ether to obtain the polyphosphazene azo based drug conjugates of methotrexate (11–14).

Poly(dichloro)phosphazene (10) (1.0 g, 2.72 mmol) was dissolved in tetrahydrofuran (20.0 mL) and was added to a solution of 4-(4-hydroxyphen-1-yldiazo)-1-(3,3-difluoro-tetrahydro-4-hydroxy-5-(hydroxymethyl)furan-2-yl)-pyrimidin-2-one (9) (1.0 g, 2.72 mmol) in tetrahydrofuran (20.0 mL). To this methanol (2.72 mmol) for (15), methylamine (2.72 mmol) for (16), p-aminobenzoate (2.72 mmol) for (17) and n-butanol (2.72 mmol) for (18), triethylamine (0.17 mL) in tetrahydrofuran (5.0 mL) was added under the atmosphere of nitrogen over a period of 30 min. The reaction mixture was refluxed for 48 h and allowed to stand at 25 °C for 3 days. The precipitated material was filtered; filtrate was concentrated under reduced pressure and precipitated with petroleum ether to obtain the polyphosphazene linked azo prodrugs of gemcitabine (15–18).

4.1.1. Poly(methoxy)[(2S)-2-[(4-{[(2,4-bis(4-hydroxyphen1yldiazo)]pteridin-6-yl)methyl](methyl)amino}benzoyl) amino]-1,5-dimethylpentonoate]phosphazene (11) Light brown powder; yield 57%; IR (KBr, m cm 1): 1675, 1125, 1472, 1590, 1732, 1251, 1635, 1235; 1H NMR (400 MHz, DMSOd6, J = Hz, d ppm): 8.67 (s, 1H, pteridine), 8.07 (s, 1H, NH), 7.76 (d, 2H, Ar-H, J = 7.2), 7.12 (m, 4H, Ar-H), 6.78 (d, 2H, Ar-H, J = 7.2), 6.92 (m, 4H, Ar-H), 4.79 (s, 2H, CH2), 4.46 (q, 1H, C-H), 3.23 (s, 6H, OCH3), 3.61 (s, 6H, CH3), 2.35 (t, 2H, CH2), 2.12 (s, 3H, N-CH3), 2.02 (q, 2H, CH2); 31P NMR (400 MHz, DMSO-d6, d ppm): 7.5 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.4  105 daltons.

4.2.1. Poly(methoxy)[4-(4-hydroxyphen-1-yldiazo)-1-(3,3difluoro-tetrahydro-4-hydroxy-5-(hydroxymethyl)furan-2-yl)pyrimidin-2-one]phosphazene (15) Brown coloured powder; yield 63%; IR (KBr, m cm 1): 3610, 1601, 1652, 1638, 1132, 1310, 1370, 1465, 1238; 1H NMR (400 MHz, DMSO-d6, J = Hz, d ppm): 8.09 (d, 2H, Ar-H, J = 8.1), 7.63 (d, 1H, pyrimidine, J = 8.2), 6.95 (d, 2H, 1H, Ar-H, J = 8.1), 6.69 (s, 1H, ribose), 4.23 (t, 2H, ribose), 3.57 (s, (br), 2H, OH), 3.24 (s, 1H, OCH3), 3.05 (d, 1H, pyrimidine, J = 8.2), 2.12 (m, 2H, furan); 31P NMR (400 MHz, DMSO-d6, d ppm): 7.5 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.5  105 daltons.

4.1.2. Poly(methylamino)[(2S)-2-[(4-{[(2,4-bis(4-hydroxyphen1yldiazo)]pteridin-6-yl)methyl](methyl)amino}benzoyl) amino]-1,5-dimethylpentanoate]phosphazene (12) Light brown powder; yield 65%; IR (KBr, m cm 1): 3240, 1524, 1626, 1156, 1471, 1596, 1731, 1253, 1632, 1228; 1H NMR (400 MHz, DMSO-d6, J = Hz, d ppm): 8.56 (s, 1H, pteridine), 8.07 (s, 1H, NH), 7.76 (d, 2H, Ar-H, J = 8.0), 7.12 (m, 4H, Ar-H), 6.79 (d, 2H, Ar-H, J = 8.0), 6.90 (m, 4H, Ar-H), 4.73 (s, 2H, CH2), 4.45 (q, 1H, C–H), 3.62 (s, 6H, CH3), 2.37 (t, 2H, CH2), 2.05 (s, 9H, CH3), 1.98(m, 4H, CH2, NH); 31P NMR (400 MHz, DMSO-d6, d ppm): 9.5 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.0  105 daltons.

4.2.2. Poly(methylamino)[4-(4-hydroxyphen-1-yldiazo)-1-(3,3difluoro-tetrahydro-4-hydroxy-5-(hydroxymethyl)furan-2-yl)pyrimidin-2-one]phosphazene (16) Brown powder; yield 61%; IR (KBr, m cm 1): 3630, 1597, 1636, 1632, 1138, 1210, 1350, 1453, 1235; 1H NMR (400 MHz, DMSOd6, J = Hz, d ppm): 8.10 (d, 2H, Ar-H, J = 8.1), 7.62 (d, 1H, pyrimidine, J = 8.2), 6.94 (d, 2H, Ar-H, J = 8.1), 6.61 (s, 1H, ribose), 4.25 (t, 2H, ribose), 3.58 (s (br), 2H, OH), 3.04 (d, 1H, pyrimidine, J = 8.2), 2.71 (s, 1H, NH), 2.13 (m, 2H, ribose), 1.21 (s, 3H, CH3); 31P NMR (400 MHz, DMSO-d6, d ppm): 6.9 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.7  105 daltons.

4.1.3. Poly(p-aminobenzoate)[(2S)-2-[(4-{[(2,4-bis(4hydroxyphen-1yldiazo)]pteridin-6-yl)methyl](methyl)amino} benzoyl)amino]-1,5-dimethylpentanoate]phosphazene (13) Blackish brown powder; yield 57%; IR (KBr, m cm 1): 1685, 1128, 1474, 1605, 1737, 1255, 1630, 1232; 1H NMR (400 MHz, DMSO-d6, J = Hz, d ppm): 8.55 (s, 1H, Ar-H, J = 8.1), 8.05 (s, 3H, NH), 7.75 (m, 6H, Ar-H), 7.17 (m, 4H, Ar-H), 6.79 (m, 6H, Ar-H), 6.93 (m, 4H, Ar-H), 4.79 (s, 2H, CH2), 4.47 (q, 1H, C–H), 3.60 (s, 12H, CH3), 2.35 (t, 2H, CH2), 2.13 (s, 3H, N-CH3), 2.01 (q, 2H, CH2); 31P NMR (400 MHz, DMSO-d6, d ppm): 8.2 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.9  105 daltons.

4.2.3. Poly(p-aminobenzoate)[4-(4-hydroxyphen-1-yldiazo)-1(3,3-difluoro-tetrahydro-4-hydroxy-5-(hydroxymethyl)furan-2yl)-pyrimidin-2-one]phosphazene (17) Dark brown powder; yield 59%; IR (KBr, m cm 1): 3610, 1604, 1631, 1131, 1368, 1453, 1238; 1H NMR (400 MHz, DMSO-d6, J = Hz, d ppm): 8.12 (d, 2H, Ar-H, J = 8.2), 8.03 (d, 2H, Ar-H, J = 8.3), 7.80 (d, 1H, pyrimidine, J = 8.1), 7.21 (s, 1H, ribose), 7.08 (d, 2H, Ar-H, J = 8.2), 6.71 (d, 2H, Ar-H, J = 8.3), 4.16 (t, 1H, CH2), 4.10 (t, 1H, CH2), 3.67 (s, (br), NH), 3.84 (s, 1H, ribose), 3.52 (s, 3H, OCH3), 2.97 (s (br), 2H, OH), 2.32 (d, 1H, pyrimidine, J = 8.1), 2.05 (s, 1H, ribose); 31P NMR (400 MHz, DMSO-d6, d ppm): d 3.9 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.8  105 daltons.

4.1.4. Poly(butoxy)[(2S)-2-[(4-{[(2,4-bis(4-hydroxyphen1yldiazo)]pteridin-6-yl)methyl](methyl)amino}benzoyl) amino]-1,5-dimethylpentanoate]phosphazene (14) Dark brown powder; yield 59%; IR (KBr, m cm 1): 3613, 1646, 1228, 1475, 1600, 1732, 1245, 1630, 1236; 1H NMR (400 MHz, DMSO-d6, J = Hz, d ppm): 8.63 (s, 1H, pteridine), 8.06 (s, 3H, NH),

4.2.4. Poly(butoxy)[4-(4-hydroxyphen-1-yldiazo)-1-(3,3difluoro-tetrahydro-4-hydroxy-5-(hydroxymethyl)furan-2-yl)pyrimidin-2-one]phosphazene (18) Brown powder; yield 69%; IR (KBr, m cm 1): 3630, 1610, 1636, 1146, 1375, 1438, 1234; 1H NMR (400 MHz, DMSO-d6, J = Hz, d

R. Sharma et al. / Bioorg. Med. Chem. 22 (2014) 1104–1114

ppm): 8.12 (d, 2H, Ar-H, J = 8.2), 7.39 (d, 1H, pyrimidine, J = 8.2), 6.95 (d, 2H, Ar-H, J = 8.1), 6.76 (s, 1H, ribose), 4.24 (t, 1H, CH2 ribose), 4.13 (t, 1H, ribose), 3.98 (s , 1H, ribose), 3.58 ((br), s, 2H, OH), 3.05 (d, 1H, pyrimidine, J = 8.2), 2.84 (s, 1H, ribose), 2.14 (quintet, 2H, CH2), 1.72 (sextet, 2H, CH2), 1.19 (t, 3H, CH3); 31P NMR (400 MHz, DMSO-d6, d ppm): 4.54 ppm [s; rel to (Ph)3P]; GPC: Mw = 2.4  105 daltons.

1113

4.4.3. In-vitro drug release kinetics In order to explain the site specific nature of polymer drug conjugates, there is essential to study the drug release kinetics of synthesized polyphosphazene azo based drug conjugates of methotrexate and gemcitabine. The polymeric drug conjugates were fitted to various kinetics models such as zero order, first order, Higuchi, Korsmeyer and peppas in order to find the best equation to fit the above mentioned model equations.

4.3. Physiological stability ex-vivo 4.5. An azoreductase assay The stability of polymer drug conjugates were checked in the stimulated acidic (pH = 1.2) and alkaline (pH = 7.4) media (sorenson’s phosphate buffer). The weighed amount (10 mg) of each polymer linked conjugates was placed in a membrane dialysis bag (MWCO 14000; Sigma Aldrich) which was closed and placed in the beaker having 50 mL of buffer solution. The solution was maintained at 37 °C with continues stirring. At appropriate interval of time, sample (2 mL) was withdrawn from the beaker and was estimated on UV spectrophotometer (Shimadu, UV-1700 pharmaspec) at kmax 306 nm, 256 nm (0.1 N HCl, pH = 7.4) for (8) and kmax 274 nm and 284 nm (0.1 N HCl, pH = 7.4) for (9) for the estimation of decrease in concentration of polymer drug conjugates in the upper GIT medium. 4.4. Ex-vivo drug release 4.4.1. Ex-vivo drug release with and without rat cecal content The ex-vivo release studies of polyphosphazene linked anticancer agents were carried out with or without rat cecal content. The ex-vivo release studies were carried out in physiological environment of stomach and small intestine by maintaining the conditions mimicking mouth to colon transit time. The experiment was carried out using USP dissolution rate test apparatus of basket type (100 rpm, 37 ± 0.5 °C). The weighed amount of polyphosphazene drug conjugates (11–18) were pooled to dialysis bag (MWCO 14000; Sigma Aldrich) and sealed properly with thread. The release profile was checked initially at pH = 1.2 for 2 h in 900 mL of SGF (simulated gastric fluid (0.1 N HCl)) as gastric empting time is about 2 h. After that medium was changed with pH = 7.4 Sorensen’s phosphate buffer (contains dibasic sodium phosphate heptahydrate (Na2HPO4
7H2O) and monobasic potassium phosphate (KH2PO4) with the ratio of 4.1:1.0 respectively) 900 mL and study was performed for 3 h as the drug is supposed to be reach the small intestine. The dissolution medium was then replaced with phosphate saline buffer, PBS pH = 6.8 (potassium dihydrogen phosphate (KH2PO4) and dipotassium hydrogen phosphate (K2HPO4) in a ratio of 1:4.3 respectively) and study was carried out for upto 24 h. Similar experimental protocol was performed upto pH 7.4 afterwards 4% w/w of fresh rat cecal content was taken in 150 mL of beaker having pH = 6.8 PBS and experiment was carried out upto 24 h. At different intervals of time, 10 mL of sample was withdrawn each time and analysed by UV spectrophotometer at kmax 306 nm, 256 nm (0.1 N HCl, pH = 7.4) for (8) and kmax 274 nm and 284 nm (0.1 N HCl, pH = 7.4) for (9) for the estimation of released drug from polyphosphazene drug conjugates. 4.4.2. Ex-vivo drug release in different intestinal contents (SI contents, SI mucosa and LI mucosa) The SI content were removed from wistar rats and suspended in cold PBS, pH = 7.4 and 10% (w/v) slurry was maintained. The mucosa of SI and LI were scraped gently from the lumen of intestines using glass slides, dispersed in cold PBS, pH = 7.0 and homogenized using a blender. The final concentrations 2% (w/w) was maintained and incubated 37 °C for 12 h in shaking water bath. After incubation, 0.5 mL of sample was withdrawn and estimated on UV spectrophotometer.

The degradation pattern of polyphosphazene azo based drug conjugates of methotrexate (11–14) and gemcitabine (15–18) were evaluated in the presence of azoreductase which was cultured after the preparation of Gram anaerobic medium (GAM).34,35 More than 70% of human faeces37 and rat faeces36 contain colonic microflora. Therefore, fresh human fecal matter was used as a source of azoreductase enzyme in colonic microflora. Human faeces (1 g) were poured into 9 mL of GAM which was further diluted with GAM to get the standard solution of microflora. A few mL of this solution was transferred to the cultured tube which was diluted 200 times with the gram anaerobic medium and kept aside under anaerobic conditions at 37 °C for a day. The pH (7.2–7.4) was maintained throughout the experiment. The reductive behavior of azoreductase was confirmed by incubating the amaranth (an azo dye) and analyzed by UV at kmax 520 nm. The azo dye was reduced with colonic microflora due to the presence of azoreductase enzyme which was absent in the flora free medium. Therefore, similar experimental protocol was followed with newly synthesized azo based polyphosphazene drug conjugates of methotrexate (11–14) and gemcitabine (15–18) instead of amaranth and analyzed by UV Spectrophotometer at kmax 256 nm (8), 284 nm (9). This experiment clearly defined the mechanism of azo based drugs that the azo bond was only reduced in the presence of specific azoreductase enzyme which is present the human fecal material (intestinal microflora) and releases the drug molecules to the colon site. 4.6. In-vitro cytotoxicity assay The in-vitro cytotoxicity was performed on different human colon cancer cell lines. The cell lines (HT-29 and COLO 320 DM) were purchased from NCCS, Pune. The cell lines were cultured in different media. To examine the effects of different substituted polyphosphazene drug conjugates on human colon cancer cell lines, the cells were treated with different concentrations. 4.6.1. MTT assay The cytotoxicity of synthesized prodrugs were determined by means of MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide) colorimeteric microculture assay as per standard procedure.27 The cell lines COLO 320 DM and HT-29 were chosen for the cytotoxicity study. Cells were plated in (5  105/ well) in 96 well plates. After 48 h, cells were incubated in 0.1% DMSO for 37 °C for 3 h. After the removal of sample solution, it was washed properly with phosphate saline buffer (pH = 7.4) and then cells were treated with MTT solution. 0.04 M of isopropranol solution was added after the 4 h of incubation. The viability of cells was determined by the absorbance at 595 nm with Elisa microplate reader (Bio Rad). All the measurements were performed in triplicates and IC50 were expressed in mean ± S.D. and p