Oral delivery of nanoparticles containing anticancer SN38 and hSET1 antisense for dual therapy of colon cancer

International Journal of Biological Macromolecules 78 (2015) 112–121

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac

Oral delivery of nanoparticles containing anticancer SN38 and hSET1 antisense for dual therapy of colon cancer M. Dinarvand a , M. Kiani a , F. Mirzazadeh a , A. Esmaeili a , Z. Mirzaie b , M. Soleimani c , R. Dinarvand a,b , F. Atyabi a,b,∗ a

Department of Pharmaceutical Nanotechnology, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran 1714614411, Iran Nanotechnology Research Centre, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran c Department of Hematology, School of Medical Sciences, Tarbiat Modares University, Tehran, Iran b

a r t i c l e

i n f o

Article history: Received 11 January 2015 Received in revised form 18 March 2015 Accepted 31 March 2015 Available online 6 April 2015 Keywords: Nanoparticles Oral drug delivery Chitosan

a b s t r a c t An oral delivery system intended for treatment of colon cancer in HT29 cancerous cells was investigated by encapsulating hSET1 antisense and SN38 anticancer in nanoparticles based on cysteine trimethyl chitosan (cysTMC) and carboxymethyl dextran (CMD). Studies have shown hSET1 as the main type of histone methyltransferase (HMT) complex, is significantly overexpressed in malignant cells. In this study, hSET1 antisense was employed to inhibit gene expression. Additionally, SN38 was incorporated into nanoparticles to enhance the efficiency of the system by inhibition of topoisomerase 1. CysTMC was synthetized and characterized by 1 H NMR and FTIR. Nanoparticles were prepared through complexation of CMD and cysTMC. Particle size and surface charge was 100–150 nm and 17–21 mV respectively with drug content of around 2.6%. Gel electrophoresis assay proved the stability of antisense in simulated gastric and intestinal fluids. Nanoparticles showed high mucoadhesion and glutathione responsive release. Cellular uptake was observed by confocal microscopy and quantified by flow cytometry. Cytotoxicity of NPs was assessed using MTT assay. Results showed hSET1/SN38 nanoparticles had significantly higher cytotoxicity against HT29 cells compared with nanoparticles containing SN38, free SN38 or naked hSET1. Therefore, present system could be considered an effective combination therapy of highly hydrophobic SN38 and hSET1. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Recent studies have highlighted the importance of epigenetic changes in formation of cancer cells. These epigenetic changes may contribute to unregulated gene expression, mutations and more importantly the silencing of tumor suppressors, genes responsible for apoptosis, or other DNA repair pathways [1]. Histone methyltransferase (HMT) enzymes as post-translational histone modifiers, are mostly responsible for epigenetic regulations [2]. It has been discovered that hSET1 as the main type of HMT complex is significantly over-expressed in malignant cells [3]. HMT enzymes are responsible for histone methylation, specifically at the position of lysine 4 on histone 3 (H3K4) [4]. Yadav et al. [3] in a study suggested that inhibition of hSET1 caused apoptosis of tumor-cells in vitro.

∗ Corresponding author at: Faculty of Pharmacy, Tehran University of Medical Sciences, Nanomedicine, Enghelab St., Teheran, Iran. Tel.: +98 21 66959052; fax: +98 21 66959052. E-mail address: [email protected] (F. Atyabi). http://dx.doi.org/10.1016/j.ijbiomac.2015.03.066 0141-8130/© 2015 Elsevier B.V. All rights reserved.

In vivo studies also showed remission of implanted colon cancer xenografts. Gene expression could be successfully manipulated at the post translational level by degradation of mRNA. Because of its encouraging therapeutic features, mRNA silencing has been extensively researched in recent years. The major ways to promote gene silencing are the induction of siRNA, ␮RNA, shRNA or antisense into the cells [5]. In this method a single strand oligonucleotide specifically binds to its complementary mRNA, prevents translation of a specific gene and later promotes mRNA degradation [6]. Antisense therapy is particularly attractive in treatment of cancers, neurodegenerative and autoimmune diseases. The use of nucleic acid macromolecules to promote gene silencing although very attractive, poses many obstacles for gene therapy [7]. First, a nucleic acid macromolecule is highly unstable in physiological media, rapidly degrades by endonucleases of plasma and is quickly eliminated by liver. Moreover, as a negatively charged macromolecule, it has minimal uptake by the negatively charged cellular membrane. Designing a proper carrier based delivery system is required to overcome these hurdles. Among proposed carriers, the application of positively charged polymers is widely studied because of

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

their superior properties such as higher uptake levels as a result of electrostatic attraction between cellular membrane and such polymers. In this category chitosan, a natural polysaccharide has drawn much interest since chitosan possesses many favorable properties such as biodegradability, biocompatibility, non-immunogenicity and availability. Overall, antisense therapy has low efficiency and high production costs [8]. To overcome these problems, in this study a nano-based oral delivery system was employed and the anticancer drug SN38 was also incorporated in the system to elevate therapeutic effects. As a result, the nanoparticles promote anticancer effects by two distinct mechanisms, gene silencing and inhibition of topoisomerase 1. The present delivery system benefits from synergistic effects of antisense and SN38. SN38, the active metabolite of irinotecan, is up to 1000 times more potent than its prodrug [9]. However, extremely low solubility and high toxicity of SN38, has rendered its application clinically impractical. Many studies have attempted to overcome these issues. In an effort in our lab [10] we designed an efficient delivery system based on a chitosan conjugated system that increased SN38 bioavailability which could minimize its toxicity by reducing the consumption of active ingredient. In that study MUC1 DNA aptamer was used for active targeting. Targeting didn’t affect the toxicity of SN38 while it increased the efficiency of NPs. In this study we endeavored to design an effective oral delivery system in order to maximize the stability and delivery of antisense in physiological media and also increase its cellular uptake. Second, we attempted to overcome shortcomings of SN38 such as low bioavailability and high toxicity by increasing its solubility and intestinal absorption. Oral drug delivery is the most preferred administration form since it has higher patient compliance. However, there are significant barriers to oral drug administration. (a) Hydrophobic drugs such as SN38 have low solubility and bioavailability when administered orally [11] and the abundance of enzymes degrade the drug molecules. (b) The mocus layer of the GI tract develops a strong barrier to drug absorption. The constant secretion of mucus contributes to rapid clearance of drugs from the GI tract [12]. As a result new drug delivery methods have been investigated to increase residence time of drugs in the GI tract. The use of positively charged polymers such as chitosan as a natural mucoadhesive has gained much interest. Chitosan directly targets the GI tract by attachment to the GI surface and promoting cellular uptake on site. The electrostatic interaction of negatively charged mocusa and positively charged chitosan leads to strong mucoadhesion [13]. To increase the mucoadhesion properties of the nanoparticles, thiomers (thiolated polymer) can be used. The free thiol groups on thiomers can create disulfide bonds within the chains of the polymer and substantially increase its adhesiveness [14,15]. Accordingly, we designed nanoparticles containing cysteine-trimethyl chitosan (cysTMC) and carboxymethyl dextran (CMD) to effectively encapsulate the active drug substances and benefit from superior mucoadhesion properties.

2. Materials and methods 2.1. Materials Chitosan (MW 400 kDa) was purchased from Primex (Siglufjörður, Norway). Carboxymethyl dextran (CMD), l-cysteine hydrochloride monohydrate, N-(3-dimethylaminopropyl)-N ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), Methyl iodide were supplied from Sigma-Aldrich (MO, USA). All other reagents were of analytic grade. Oligonucleotides antisense (5 -AAGGGGGTTCCTTGGGA-3 ) and Cy5 labeled oligonucleotides (5 -CATCGAAATCGCAGTTAC-3 ) were obtained from

113

Macrogene (Seoul, Korea). HT29 cell line was obtained from Pasteur Institute of Iran (Tehran, Iran). 2.2. Methods 2.2.1. cysTMC synthesis First, trimethyl chitosan (TMC) was prepared according to the method previously described by our group with slight modification [16]. Briefly 0.5 g (MW 50 kDa) chitosan was dispersed in 60 mL NMP and left under stirring for 30 min, then 1.5 g sodium azide, 5 mL sodium hydroxide (15% w/v) and 3.8 mL methyl iodide was added to the above mixture. The mixture was stirred at 60 ◦ C for 3 h. Afterwards, 200 mL deionized water (DW) was added and stirred for 2 h. The mixture was dialyzed against DW using dialysis tube (12 kDa) for 24 h. The mixture was precipitated by addition of an excess amount of acetone. The product was washed with sodium hydroxide (15% w/v) thoroughly. The precipitate was suspended in water and dialyzed against DW for 48 h followed by freeze drying. TMC-cysteine (cysTMC) was conjugated by a previously described method [17]. Briefly 200 mg TMC and 400 mg cysteine was dissolved in 20 mL DW following addition of EDC and NHS to the mixture to reach the final concentration of 200 mM. Afterward, the pH was adjusted to 5 and left for 5 h in a dark place. The solution was dialyzed against hydrochloric acid (pH 5) for 5 days. The solution was freeze dried and stored at 4 ◦ C for further investigation. 2.2.2. cysTMC characterization The structure of TMC was confirmed by Fourier Transform Infrared (FTIR) spectroscopy (Nicolet Magna 550-FT, Spectra Lab Scientific Inc., Canada).1 H NMR spectrum of TMC in D2 O was studied to determine the degree of quaternization of TMC, (Varian 400 Unityplus spectrometer, USA). The degree of quaternization was measured using the following equation:



%DQ = where

TMC

 

H



×

1 × 100 9

TMC is the integral of the quaternary amino peak at



3.4 ppm on the 1 H NMR spectrum and H is the integral of the 1 H peaks between 4.7 and 5.7 ppm. To calculate the degree of thiolation of TMC, Ellman’s test was employed. Briefly 1.3 mg of TMC was dissolved in 0.1 M sodium phosphate buffer (pH 8.0) and added to a test tube containing 50 ␮L of Ellman’s reagent (4 mg/mL) anda 2.5 mL of reaction buffer. The sample was incubated at room temperature for 15 min. The UV absorptions of the samples were measured at 412 nm by UV/vis spectrophotometer (Cecil-CE7500, Cambrige, UK). A standard curve was plotted from 6 known concentrations of cysteine and the sample concentration was determined by this plot. 2.2.3. cysTMC/CMD/SN38/hSET1 nanoparticle preparation Preparation of the nanoparticles (NPs) was carried out by simple complexation method. Positively charged cysTMC forms electrostatic interaction with negatively charged CMD and hSET1 oligonucleotide to produce stable polyelectrolyte complexes. Briefly, 100 ␮L CMD (1 mg/mL in DW) and 50 ␮L SN38 (6 mg/mL in DMSO) were mixed together and sonicated for 10 min in 45 mHz and added drop-wise to a 1 mL cysTMC (10 mg/mL in DW) mixture and stirred on a vortex for 1 min (1500 rpm). Subsequently the mixture was freeze dried. For further studies the freeze dried powder was dissolved in 1 mL DW and filtered through a 0.45 ␮m membrane. For preparation of the nanoparticles containing antisense, 10 ␮L of antisense (100 pm/␮L) was added to CMD, SN38 mixture and thoroughly mixed by means of a vortex. The mixture was then added to the solution of cysTMC as described above.

114

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

2.2.4. NP characterizations NP size and surface charge was measured by dynamic light scattering microscopy (Zetasizer Nano ZS, Malvern, Worcestershire, UK). The morphology of the nanoparticles was observed through SEM imaging. 2.2.5. Structural stability of NPs To investigate stability of the NPs against ionic strength, dilution, harsh pH conditions and physiological fluid exposure, the suspensions containing the NPs were diluted up to 20 times with Fasted State Simulated Gastric Fluid (FaSSGF) (pH 1.6) and Fasted State Simulated Intestinal Fluid (FaSSIF) (pH 6.8). In another attempt the particles were diluted in FaSSGF solution. After 30 min of incubation, pH of the medium was raised to 8 to investigate the effects of pH alterations on stability of the nanoparticles. Structural stability of NPs was observed through measurement of particle size and zeta potential of the nanoparticles. 2.2.6. Antisense stability in simulated physiological fluids To analyze the stability of the antisense and capability of NPs to protect the antisense from intense ionic and pH alteration in physiological fluids, NPs were incubated in FaSSIF and FaSSGF at 37 ◦ C for 2 h. Stability of the antisense was examined by gel electrophoresis assay on 4% (w/v) agarose gel at 56 V for 15 min. 2.2.7. In vitro mucoadhesion study Mucoadhesion properties of the NPs were measured according to a method previously reported [18]. Briefly 2 mg/mL mucin was mixed with either TMC/CMD NPs or cysTMC/CMD NPs and incubated in 37 ◦ C for 5 h, then the mixture was centrifuged at 20,000 rpm for 30 min. 200 ␮L of 10 mg/mL periodic acid was added to the supernatant containing non absorbed mucin and incubated for another 2 h. Afterward 200 ␮L of schiff solution was added at room temperature and the mixture was left for 30 min. The UV absorption of the free mucin was measured at wavelength of 555 nm thru UV/vis spectrophotometry. 2.2.8. SN38 loading measurement Drug quantity and encapsulation efficiency was measured through UV absorbance of SN38 at 370 nm applying the following equations. drug loading (DL%) =

Wmass of drug in NPs Wmass of polymer loading

encapsulation efficiency (EE%) =

× 100

Wmass of drug in NPs Wmass of drug in loading

× 100

2.2.9. Drug release study In vitro drug release from NPs was performed in simulated gastric and intestinal fluids. To mimic the gastric condition, 50 mL of FaSSGF (0.01 mol/L, pH 1.6) was used as a dialysis medium at 37 ± 0.5 ◦ C under constant stirring. A 2 mL solution of nanoparticles containing 0.3 mg SN 38 was charged into a dialysis tube (3500 kD) and after 40 min [19], 1 mL of the medium was withdrawn. The sample was analyzed by UV/vis absorption at 370 nm. To mimic the intestinal condition, 50 mL of FaSSIF was used as the medium at 37 ± 0.5 ◦ C under constant stirring. Samples were withdrawn at 0.5, 1, 2, 3, 4 h and analyzed by UV/vis absorption at 370 nm. 2.2.10. Glutathione-responsive release NPs containing SN38 were incubated at 37 ◦ C in 1 mL of 0.2 M PBS (pH 7.4). Each sample contained 0, 4.5 ␮M or 10 mM glutathione. At time intervals of 0, 2, 4, 6 h the suspension was centrifuged at 16,000 rpm for 30 min and 500 mL of the supernatant

was quantified for the contents of SN38 by UV/vis spectrophotometry at 370 nm. The precipitate was resuspended with 0.2 M PBS containing the same glutathione concentration. 2.2.11. Cellular uptake of NPs HT29 cells were seeded on 6-well plates at 1 × 105 cells/well density and incubated for 24 h. Following treatment with NPs containing 2 ␮L (100 pmol/␮L) of Cy5 labeled oligonucleotides for 2 h; cells were washed with 0.2 M PBS (pH 7.4) three times and observed with confocal microscopy (Nikon confocal microscope A1, Nikon Inc., Switzerland). To evaluate the uptake mechanism of NPs in HT29 cells, first, cells were treated with endocytic inhibitors [20] including sodium azide (5 mM), methyl ß cyclodextrin (25 ␮M), chlorpromazine (25 ␮M), amiloride 250 ␮g/mL and incubated for 30 min. Afterward the NPs were applied and the 2 h uptake was studied. The amount of cellular uptake was quantified by flow cytometry assay. 2.2.12. Cell growth inhibition Cytotoxicity of NPs containing both SN38 and hSET1 was assessed through calculation of viability of human colon cancer cell line (HT-29) by MTT test. Cells were cultured in 98 well plates with the density of 2 × 104 with growth media, at 37 ◦ C in 5% CO2 humidified incubator. 100 ␮L of growth media containing different concentrations of SN38 and constant concentration of hSET1were replaced. The cytotoxicity of cysTMC/CMD NPs containing both hSET1 and SN38 (SN38/hSET1 NPs), NPs containing only SN38 (SN38 NPs), free SN38 and naked hSET1 antisense was assessed. SN38 concentrations of 30, 25, 20, 15, 10 ␮g/mL and hSET1 concentration of (1 ␮M) was used. Cytotoxicity was measured 24 and 48 h after treatment. The cell viability was measured through the following equation: cell viability =

OD of samples × 100 OD of controls

where OD of samples denotes the mean value of 4 wells treated with samples and OD of controls denotes the mean value of 4 well treated with PBS buffer. Each experiment was done in triplicate. Results resemble mean value of three samples. 2.2.13. Statistical data analysis Statistical data analysis was performed using the Student’s t-test using p < 0.05 as the level of significance. 3. Results and discussion In this study we attempted to design a nanoparticle based system for combination therapy of an anticancer drug and an antisense to effectively treat colorectal cancerous cells. SN38, a highly potent anticancer drug, never reached its potential activity despite its pharmaceutical advantage over irinotecan as a result of its high toxicity and very low solubility in biological media. Therefore, many researches have been conducted aiming to bypass these hindrances [21]. The clinical application of gene silencing in cancer therapy makes slow progress despite its theoretical advantages. This is due to complications in developing efficient nucleic acid delivery systems where RNA/DNA is protected from degradation and retains its biological activity plus ensuring efficient cellular internalization. One delivery strategy which has recently gained prominence is embedding nucleic acids in nanoparticles. To achieve this in present study, chitosan polysaccharide was chosen as the backbone of the nanoparticle system. Chitosan possesses valuable features for delivery such as being biodegradable, biocompatible, nontoxic and abundantly available.

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

115

Fig. 1. The FTIR spectrum of trimethyl chitosan, the dotted line denotes the FT IR spectrum of chitosan and the line represents the FT IR spectrum of TMC.

3.1. Polymer synthesis and characterization To improve chitosan’s cationic and solubility features, trimethyl chitosan was synthesized [22]. Trimethyl chitosan is soluble in a wider range of pH. TMC was synthesized according to a method previously reported by our group. This method yields a high methylation degree. The presence of methylated amine was verified by FTIR spectroscopy. Accordingly, a TMC potassium bromide disk was prepared and the peaks were observed by FTIR. Results showed the appearance of an absorption band at 1469 cm−1 which is attributed to the stretch of C H bond. The absorption band belonging to the stretch of N H in amino group occurring in chitosan at 1567 cm−1 substantially weakened in TMC (occurring in 1638 cm−1 ) and new bands appeared at 1404 and 1469 cm−1 which belongs to the absorption of N C in amine methylation. The bands belonging to

alcohols from 1060 to 1151 cm−1 was not altered which verifies that no alkylation has taken place at the position of C3 and C6 in chitosan (Fig. 1). The amount of quaternary amines on chitosan was measured by 1 H NMR. A peak around 3.4 ppm confirms the presence of quaternary amines and the peaks between 4.7 and 5.7 denotes the H bonded to C1 of chitosan (Fig. 2). The degree of quaternization was measured to be 71%. The methylation degree at 71% was high enough to ensure satisfactory cationic effects while adequate primary amines remained to react with cysteine to form disulfide bond Thiolation of TMC was carried out to ensure higher mucoadhesion. The cysteine TMC conjugation reaction was carried out according to a simple method reported by Yin et al. [17]. The concentration of thiol groups was 1.56 ± 0.1 ␮M as measured by the Ellman’s test.

Fig. 2. The 1 H NMR spectrum of TMC.

116

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

3.2. Nanoparticle formation and characterization Nanoparticles were prepared by simple complexation through self-assembly owing to electrostatic interactions of polymers and active ingredients. This was particularly desirable since application of complicated methods in this study with two active agents and two polymers could make NP’s preparation rather difficult. TMC/CMD/SN38/hSET1 weight ratio was optimized at 100:10:3:1 to yield NPs with efficient encapsulating abilities for both SN38 and hSET1 antisense. Size and surface charge of NPs were measured by dynamic light scattering microscopy. As expected the use of two polymers and encapsulating antisense and SN38 slightly increased the diameter of nanoparticles. NPs retained particle diameter sizes of 120–150 nm with small size distribution. cysTMC/CMD NPs had slightly larger particle size. The surface charge of nanoparticles ranged from +15 in the case of TMC/CMD NPs to +19 in the case of cysTMC/CMD NPs. CysTMC/CMD NPs had a more positive surface charge as a result of the presents of thiol groups (Table 1). The relative positive surface charge is desirable in cellular uptake. Among many physicochemical parameters that affect systemic circulation and tumoral internalization of NPs, surface charge seems to play a critical role. Positive charge NPs reduces their plasma exposure [23]. This is especially important in delivery of the nucleic acids considering that nucleic acids are highly susceptible to endonucleases and ionic compounds in physiological fluids [5]. Thus, it is expected that our prepared NPs would have higher accumulation in tumors compared to neutral or negatively charged NPs. This is also confirmed in other studies where cationic liposomes showed higher accumulation in tumor tissue compared to non-cationic liposomes [24]. In present study the inclusion of CMD, a negatively charged polymer, in the NPs considerably reduced the surface charge of the NPs. The NPs containing high deacetylated chitosan yield very high positive charged surfaces because of the ionization of chitosan by its quaternary amines in the acidic extracellular and intracellular media. Results from studies performed on isolated mice ECM showed that electrostatically charged NPs (positive or negative) had hampering in their movements through the matrix [25]. They further concluded that intensive charges above +10 mV and below −30 mV stop the proper diffusion of NPs. Therefore, masking the intense positive surface charges with CMD produced better diffusible nanoparticles. Fig. 3 represents the SEM images of the prepared NPs which had relatively uniformed spherical shapes with smooth surfaces the SEM imaging shows particles that are slightly larger (150–200 nm) than what was measured by DLS method. This could be attributed to the aggregation that occurs in the process of drying the NPs for SEM imaging. The shape of NPs plays a pivotal role in regulating their distribution and accumulation in tumors [26]. In a study nanorods were compared with nanospheres with the same diameter size. In spite of comparable circulation kinetics, the results showed nanorodes diffused deeper and faster in tumors [27]. The results of several studies suggested that elongated shapes might possess better EPR effects [26]. In a study two particle shapes (nanorods of 360 × 80 nm and nanospheres 200 × 200 nm) were compared. Although the rod shaped particles had higher tumor exposure compared to spherical shaped particles, the spherical particles could increase drug release rates due to enhanced surface to volume ratio. In addition in the same study, the tumor drug levels for spherical particles in the first 24 h were higher than that of rod shaped nanoparticles. However the overall drug levels of nanorods in tumor were higher than that of nanospheres [28]. As for our prepared NPs, it seems an increase in release rate might be preferred since particles retained in the tumor diffuse back to the circulation system and a release time more than 24 h could increases systemic adverse effects of anticancer drugs.

Fig. 3. SEM image of cysteine trimethyl chitosan NPs.

3.3. Stability of NPs and encapsulated antisense While designing an orally administered nano-system, maintaining the stability and integrity of NPs in physiological media with vigorous pH changes, presence of ionic compounds and endonucleases is a big challenge, particularly when the nanoparticles carry nucleic acids. As naked nucleic acid is very sensitive to enzymatic degradation, successful encapsulation of antisense is essential in rendering it resistant to nuclease activity [8]. As represented in Table 1 NPs maintained their structural stability in terms of particle size and zeta potential after incubation in simulated gastric fluid (pH 1.6) for 30 min. No significant change was observed but increasing the pH of the same medium from 1.6 to 8 caused augmentation of particle size by more than 2 folds and decrease in zeta potential, suggesting the occurrence of slight aggregations between NPs. Therefore, drastic low pHs have no significant effect on the particle size of NPs and only renders a slight increase in their zeta potential. This could be explained by the fact that primary amines of TMC will become ionized in acidic media and this increases the overall positive charge of the surface of the NPs. The data obtained in present study were in contrast to the data stated by He et al. and Han et al. where after pH reduction, NPs aggregated and particle size increased [29,30]. In both studies cysteine conjugated TMC was synthesized as the backbone of the polymeric carrier with monosaccharide modifications (galactose and mannose). The fact that our nanoparticles do not undergo structural changes could be attributed to the supplying of CMD in addition to chitosan to the structure of nanoparticles. Furthermore, to investigate the effects of dilution, harsh pH and presence of ionic compounds in the intestinal fluids, nanoparticles were incubated in simulated intestinal fluid (20 times of their original volume). Consequently, surface charges of the NPs dramatically decreased to −18 and −20 mV and a growth in particle size (by 1.6 to 1.8 folds) was also observed. Interestingly polydispersity of the nanoparticles decreased from 0.2 to 0.1 suggesting the presence of uniformed nanoparticles despite the increase in particle size which confirms that nanoparticles were not aggregated. These phenomena could be explained by considering the water solubility of both applied polymers. Chitosan is insoluble in basic solutions since its primary amines are not ionizable in that media whereas CMD is freely soluble in aqueous media. Increasing the pH of the medium bring about the migration of insoluble

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

117

Table 1 TMC/CMD and cysTMC/CMD NPs characterization when incubated in DW, FaSSIF, FaSSGF and pH 8 media. Incubation media DW FaSSGF pH 1.6 pH 8 FaSSIF pH 6.8

TMC NPs Size 120 125 370 200

CysTMC NPs

Charge ± ± ± ±

6 2 5 2.4

17.6 19.1 −11 −20.5

± ± ± ±

0.3 0.3 0.5 0.2

PDI

Size

0.2 0.3 0.3 0.1

148 126 326 300

Charge ± ± ± ±

10 5 3 15

21.8 24.6 −7.12 −18.2

± ± ± ±

PDI 0.2 2 0.5 0.2

0.2 0.3 0.3 0.2

3.4. In vitro studies (mucoadhesion, drug loading, release profile, GSH release)

Fig. 4. Gel electrophoresis assay (a) cysTMC/CMD NPs in DW, (b) TMC/CMD NPS in DW, (c) Naked hSET1 in DW, (d) cysTMC/CMD in FaSSIF, (e) TMC/CMD in FaSSIF, (f) naked hSET1 in FsSSIF, (g) cysTMC/CMD in FaSSGF, (h) TMC/CMD in FaSSGF, (i) Naked hSET1 in FaSSGF, (h) Ladder.

chitosan chains to the core of the nanoparticle to limit their exposure, while negatively charged dextran polymer displays at the outer layer of nanoparticles. Gel electrophoresis study was conducted to determine whether nanoparticles had sufficient ability to encapsulate antisense in exposure to gastric and intestinal simulated fluids. The NPs containing hSET1 antisense were incubated in FaSSIF and FaSSGF and DW for 2 h. Then, the NPs were loaded into agarose 4% gel. The presence of shiny well shows that both cysTMC/CMD and TMC/CMD NPs successfully encapsulated antisense in DW and nucleic acids did not migrate in the gel (Fig. 4a and b); however, naked hSET1 migrated in the well as shown by the shiny band (fig. 4c). When NPs where incubated in FaSSIF and FaSSGF, a shiny band is visible in the gel which shows that NPs effectively protect antisense from degradation (Fig. 4d, e, g, h), but when naked antisense was incubated in simulated fluids and loaded on the gel no shiny band was visible which shows that the antisense was degraded (Fig. 4f and i).

Fig. 5. Drug releases from NPs in simulated intestinal fluid.

In an oral delivery system NPs with mucoadhesive properties allow fewer administrations with longer intervals by preserving the drug in contact with mucosa for longer periods [31]. Our NPs could be considered as a suitable mucoadhesive system since they possessed the following qualities: biodegradable, non-toxic, durable and prompt adhesion to mucosa. Both NPs had desirable mucoadhesion properties with cysTMC/CMD having stronger mucoadhesion. TMC/CMD and cysTMC/CMD NPs had 67% and 80% mucoadhesion respectively. The drug loading of NPs was obtained indirectly by measuring the amount of free SN38 remained in the supernatant and deducing it from the total drug applied. The concentration of SN38 was acquired from plotting a standard curve. The drug loading of cysTMC/CMD NPs and TMC/CMD NPs were 1.84 ± 0.01% and 2.6 ± 0.3% respectively. The encapsulation efficacies of cysTMC/CMD and TMC/CMD NPs were measured to be 68.33 ± 5% and 86.6 ± 1% respectively. The loading of SN38 in the NPs reported in our previous studies where typically in the range of 2–7% [10,32]. Considering that in this study we aimed to employ a

Fig. 6. Glutathione response release of TMC/CMD (a) and cysTMC/CMD (b) NPs.

118

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

Fig. 7. Confocal images showing cellular uptake of NPs containing CY5 labeled oligonucleotide, (a) TMC/CMD NPs, (b) cysTMC/CMD NPs (c) naked CY5 oligonucleotide as control, the nuclei are stained with DAPI.

dual therapy system that took benefit of therapeutic effects of both SN38 and hSET1 antisense, the drug loading was decreased so that superior therapeutic effects could be achieved while reducing toxic adverse effects of anticancer SN38. In vitro release profiles of SN38 from TMC/CMD showed a controlled release behavior while release profiles of cycTMC/CMD NPs showed an initial burst release following a controlled release manner. An initial burst of drug from NPs might be desirable to ensure that sufficient amount of drug reaches its target site and provide therapeutic effects. As depicted in Fig. 5, drug release from TMC/CMD NPs in FaSSIF was about 5% after 4 h. Conversely, the drug showed a burst release from cysTMC/CMD NPs of more than 6% in the first hour and a controlled release in the following 3 h. The cumulative release was about 14%. The release profile of NPs in simulated gastric fluid is measured after 40 min [19]. Release of SN38 from TMC/CMD NPs and cysTMC/CMD NPs in FaSSGF within 40 min was 2.1 ± 0.3 and 5.7 ± 0.5% respectively. As a general rule, the extracellular environment has low glutathione (GSH) concentrations of about 2–20 ␮M which provides minimal reducing ability. On the other hand the cytosol has high glutathione concentrations of about 2 to 10 mM which yields high reducing ability [33]. Interestingly the tumor tissue has high glutathione content as well [34]. This unique feature allows the

construction of active intracellular drug release by devising glutathione sensitive NPs [35]. Both our prepared NPs exhibited effective glutathione sensitivity although responsive release of cysTMC was significantly higher than that of TMC by 4.6 folds. As shown in Fig. 6a and b, when the GSH content was 4.5 ␮M (simulating the extracellular environment), NP’s drug release was rather similar to that of zero content GSH at these GSH concentrations the 8 h accumulative release of SN38 from TMC/CMD and cysTMC/CMD NPs was ≈9.3% and 20–25%, respectively. Comparatively, the maximum drug release was witnessed when the GSH content was 10 mM (representing the intracellular redox environment). The accumulative 6 h drug release from TMC/CMD NPs was approximately 13% while that for cysTMC/CMD NPs was about 61%. As expected, incorporating disulfide bonds in TMC chains meaningfully enhanced glutathione responsiveness of NPs. Cai et al. also designed NP micelles to enhance intracellular specific release of camptothecin. The NPs consisted of PEG-SS-PBLG and PEG-SS-poly (benzyloxycarbonyl-l-lysine), a hydrophobic and a hydrophilic polymer conjugated by a disulfide bond. Upon cellular uptake the reducing environment of cytosol destroys the disulfide bond and the drug is released [36]. The principal of designing a glutathione responsive system as reported by Meng et al. [37] is conjugating disulfide bonds in the main chain, side chain or as a cross-linker to the polymer.

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

119

Fig. 8. Flow cytometry results of uptake after treatment with endocytic inhibitors. After treatment with sodium azide (a), after treatment with amiloride (b), after treatment with methyl ß cyclodextrin (c), and after treatment with chlorpromazine (d).

3.5. Cell studies 2 h uptake of NPs was studied in HT29 (human colon cancer cell line). To visualize the potential uptake of NPs, CY5 labeled oligonucleotide was chosen. As shown in Fig. 7, confocal images of both NPs showed maximal uptake although in both cases cellular morphology was slightly affected. Naked oligonucleotide had no cellular uptake as shown in Fig. 7c. As a general rule, particles with positive surface charge demonstrate strong cellular uptake due to attraction of opposite electrostatic charges of cellular membrane and nanoparticles. In the case of CMD it seems that negative surface charge does not impact its cellular uptake in a harmful way. Remarkably in another study on magnetic nanoparticles coated with carboxymethyl dextran the authors concluded that nanoparticles were internalized by endocytic mechanisms. Interestingly it is believed that the uptake of CMD is cell-type specific. Tumor cells from various origins showed higher NP uptake than primary cells such as leukocytes [38]. In the same experiment, when the CMD shells of NPs were degraded, nanoparticles lost their celltype specificity. After reconstructing the CMD shells the tumor cells showed faster uptake than the leukocytes [39]. Therefore, the negative surface charge encountered after incubation in intestinal simulated fluid should not deleteriously affect cellular uptake.

To evaluate the uptake mechanism, 4 endocytic inhibitors were applied 30 min prior to addition of NPs to the cells. Sodium azide as a metabolic inhibitor of energy-mediated endocytosis, amiloride as a strong micropinocytic inhibitor, methyl ß-cyclodextrin as a liquid mediated endocytic inhibitor and chlorpromazine as a clathrin mediated endocytic inhibitor were employed. The degree of uptake was quantified by flow cytometry visualizing internalization of NPs encapsulating CY5 labeled oligonucleotides (Fig. 8). After treatment with sodium azide, uptake reduced to 9%. After treatment with amiloride 61–58% uptake was observed. After treatment with methyl ß cyclodextrin, 9.25% uptake was observed and after treatment with chlorpromazine 10.7% uptake was observed. Treatment with methyl ß cyclodextrin, chlorpromazine and sodium azide decreased uptake of NPs while treatment with amiloride did not affect the uptake. Therefore, it can be conclude that the NPs are internalized by several mechanisms except micropinocytosis. The cytotoxicity of NPs was measured through MTT assay. As shown in Fig. 9a the cell cytotoxicity after 24 h was directly influenced by the concentration of SN38. In this test NPs showed higher cell growth inhibition than free SN38. The cell toxicity of NPs containing both SN38 and hSET1 was significantly (p < 0.05) higher than SN38 NPs or free SN38. This was very significant when NPs with the lowest concentration of SN38 (5 ␮g/mL) was used

120

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121

effects. Free SN38 (5 ␮g/mL) had minimal cell cytotoxicity while encapsulating the drug in NPs significantly increased the cytotoxic effects from 14% to 78% cell death. Overall results show that combination therapy had significantly higher toxicity compared to either monotherapy, this was more evident in the case of NPs containing the lowest concentration SN38 (5 ␮g/mL). Since the aim of this study was to design NPs that encapsulate two active ingredients that decrease side effects of toxic SN38 and provides synergistic effects, it seems that the NPs containing 5 ␮g/mL SN38 and 1 ␮M antisense was very ideal. Because compared to free SN38, it had superior toxic effects. 4. Conclusion The results of our study showed that designing NPs for combination therapy of colon cancer was more effective than either therapy alone. Prepared NPs effectively encapsulated SN38, a highly potent anticancer whose usage has been greatly limited because of its toxic adverse effects and low bioavailability and also benefited from gene silencing effects by inhibiting hSET1 which is overexpressed in colon cancer cells. This way a single therapeutic NP exerted its effects by two distinct pathways to reduce the amount of active components and limit side effects of both methods and also provided synergistic effects. Acknowledgements The authors would like to thank the “Iran National Science Foundation” (INSF) for supporting this study. References

Fig. 9. Cytotoxicity of HT29 cells after 24 h (a) and 48 h (b) incubation with 5 concentrations of SN38 in SN38/hSET1 NPs, SN38 NPs and free SN38.

which rationalizes the use of combination therapy for colon cancer in order to lower the concentration of toxic SN38. Naked hSET1 had little cytotoxicity which is due to minimal cellular uptake. However, encapsulating antisense in NPs significantly (p < 0.05) increased their cytotoxicity this shows nanoparticles increased cellular uptake of antisense. After 48 h the cell viability dramatically decreased. However, different drug concentrations did not exhibit significant (p > 0.05) changes in cell cytotoxicity. All concentrations of NPs caused more than 70 cell deaths except for NPs containing 5 ␮g/mL SN38/hSET1. In the case of NPs containing 5 ␮g/mL SN38 the difference in cell cytotoxicity between SN38 NPs and SN38/hSET1 NPs was higher than the rest of concentrations. This formulation is ideal because it enables effective cancer therapy with SN38 while decreases the drug concentration and lowers side

[1] M. Esteller, J.G. Herman, J. Pathol. 196 (2002) 1–7. [2] I.M. Fingerman, C.-L. Wu, B.D. Wilson, S.D. Briggs, J. Biol. Chem. 280 (2005) 28761–28765. [3] S. Yadav, J. Singhal, S.S. Singhal, S. Awasthi, Biochem. Pharmacol. 77 (2009) 1635–1641. [4] R.J. Sims, D. Reinberg, Genes Dev. 20 (2006) 2779–2786. [5] M. Masiero, G. Nardo, S. Indraccolo, E. Favaro, Mol. Aspects Med. 28 (2007) 143–166. [6] R.I. Gregory, T.P. Chendrimada, N. Cooch, R. Shiekhattar, Cell 123 (2005) 631–640. [7] K.A. Whitehead, R. Langer, D.G. Anderson, Nat. Rev. Drug Discovery 8 (2009) 129–138. [8] Y. Shu, F. Pi, A. Sharma, M. Rajabi, F. Haque, D. Shu, M. Leggas, B.M. Evers, P. Guo, Adv. Drug Delivery Rev. 66 (2014) 74–89. [9] K. Duan, X. Zhang, X. Tang, J. Yu, S. Liu, D. Wang, Y. Li, J. Huang, Colloids Surf., B: Biointerfaces 76 (2010) 475–482. [10] E. Sayari, M. Dinarvand, M. Amini, M. Azhdarzadeh, E. Mollarazi, Z. Ghasemi, F. Atyabi, Int. J. Pharm. 473 (2014) 304–315. [11] H. Hosseinzadeh, F. Atyabi, R. Dinarvand, S.N. Ostad, Int. J. Nanomed. 7 (2012) 1851. [12] L.M. Ensign, R. Cone, J. Hanes, Adv. Drug Delivery Rev. 64 (2012) 557–570. [13] F.A. Moghaddam, F. Atyabi, R. Dinarvand, Nanomed. Nanotechnol. Biol. Med. 5 (2009) 208–215. [14] G. Mythri, K. Kavitha, M. Kumar, S.J. Singh, J. Appl. Pharm. Sci. 1 (2011) 37–42. [15] P. Yousefpour, F. Atyabi, R. Dinarvand, E. Vasheghani-Farahani, Daru 19 (2011) 367 (Journal of Faculty of Pharmacy, Tehran University of Medical Sciences). [16] F. Atyabi, S. Majzoob, M. Iman, M. Salehi, F. Dorkoosh, Carbohydr. Polymers 61 (2005) 39–51. [17] L. Yin, J. Ding, C. He, L. Cui, C. Tang, C. Yin, Biomaterials 30 (2009) 5691–5700. [18] F. Talaei, M. Azhdarzadeh, H.H. Nasel, M. Moosavi, A. Foroumadi, R. Dinarvand, F. Atyabi, Daru 19 (2011) 257 (Journal of Faculty of Pharmacy, Tehran University of Medical Sciences). [19] S. Hellmig, F. Von Schoning, C. Gadow, S. Katsoulis, J. Hedderich, U.R. Folsch, E. Stuber, J. Gastroenterol. Hepatol. 21 (2006) 1832–1838. [20] D. Dutta, J.G. Donaldson, Cell. Logist. 2 (2012) 203–208. [21] V. Bala, S. Rao, B.J. Boyd, C.A. Prestidge, J. Controlled Release 172 (2013) 48–61. [22] S. Majzoob, F. Atyabi, F. Dorkoosh, K. Kafedjiiski, B. Loretz, A. Bernkop-Schnürch, J. Pharm. Pharmacol. 58 (2006) 1601–1610. [23] C. He, Y. Hu, L. Yin, C. Tang, C. Yin, Biomaterials 31 (2010) 3657–3666. [24] M. Schmitt-Sody, S. Strieth, S. Krasnici, B. Sauer, B. Schulze, M. Teifel, U. Michaelis, K. Naujoks, M. Dellian, Clin. Cancer Res. 9 (2003) 2335–2341. [25] O. Lieleg, R.M. Baumgärtel, A.R. Bausch, Biophys. J. 97 (2009) 1569–1577. [26] N. Bertrand, J. Wu, X. Xu, N. Kamaly, O.C. Farokhzad, Adv. Drug Delivery Rev. 66 (2014) 2–25.

M. Dinarvand et al. / International Journal of Biological Macromolecules 78 (2015) 112–121 ´ O. Chen, J. Cui, D. Fukumura, M.G. Bawendi, R.K. Jain, [27] V.P. Chauhan, Z. Popovic, Angew. Chem. 123 (2011) 11619–11622. [28] K.S. Chu, W. Hasan, S. Rawal, M.D. Walsh, E.M. Enlow, J.C. Luft, A.S. Bridges, J.L. Kuijer, M.E. Napier, W.C. Zamboni, Nanomed. Nanotechnol. Biol. Med. 9 (2013) 686–693. [29] L. Han, C. Tang, C. Yin, Biomaterials 35 (2014) 4589–4600. [30] C. He, L. Yin, C. Tang, C. Yin, Biomaterials 34 (2013) 2843–2854. [31] V. Hearnden, V. Sankar, K. Hull, D.V. Juras, M. Greenberg, A.R. Kerr, P.B. Lockhart, L.L. Patton, S. Porter, M.H. Thornhill, Adv. Drug Delivery Rev. 64 (2012) 16–28. [32] N. Sepehri, H. Rouhani, F. Tavassolian, H. Montazeri, M.R. Khoshayand, M.H. Ghahremani, S.N. Ostad, F. Atyabi, R. Dinarvand, Int. J. Pharm. 471 (2014) 485–497.

121

[33] G. Wu, Y.-Z. Fang, S. Yang, J.R. Lupton, N.D. Turner, J. Nutr. 134 (2004) 489–492. [34] P. Kuppusamy, H. Li, G. Ilangovan, A.J. Cardounel, J.L. Zweier, K. Yamada, M.C. Krishna, J.B. Mitchell, Cancer Res. 62 (2002) 307–312. [35] F. Meng, W.E. Hennink, Z. Zhong, Biomaterials 30 (2009) 2180–2198. [36] X.-J. Cai, H.-Q. Dong, W.-J. Xia, H.-Y. Wen, X.-Q. Li, J.-H. Yu, Y.-Y. Li, D.-L. Shi, J. Mater. Chem. 21 (2011) 14639–14645. [37] F. Meng, R. Cheng, C. Deng, Z. Zhong, Mater. Today 15 (2012) 436–442. [38] J. Clement, M. Schwalbe, N. Buske, K. Wagner, M. Schnabelrauch, P. Görnert, K. Kliche, K. Pachmann, W. Weitschies, K. Höffken, J. Cancer Res. Clin. Oncol. 132 (2006) 287–292. [39] M. Schwalbe, N. Buske, M. Vetterlein, K. Höffken, K. Pachmann, J. Clement, Zeitschrift für Physikalische Chemie 220 (2006) 125–131.