Poly (methacrylic acid)-based molecularly imprinted

0 downloads 0 Views 660KB Size Report
May 13, 2018 - Neda Madadian‐Bozorg1. | Payam Zahedi2 ...... Abdouss M, Asadi E, Azodi‐Deilami S, Beik‐Mohammadi N, Aslanzadeh. SA. Development ...
Received: 27 March 2018

Revised: 13 May 2018

Accepted: 13 May 2018

DOI: 10.1002/pat.4353

RESEARCH ARTICLE

Poly (methacrylic acid)‐based molecularly imprinted polymer nanoparticles containing 5‐fluourouracil used in colon cancer therapy potentially Neda Madadian‐Bozorg1

|

Payam Zahedi2

|

Mohammad Shamsi2

|

Shahrokh Safarian3

1

Department of Life Science Engineering, Faculty of New Science and Technology, University of Tehran, Tehran, Iran

2

Nano‐Biopolymers Research Laboratory, School of Chemical Engineering, College of Engineering, University of Tehran, PO Box: 11155‐4563 Tehran, Iran

3

Department of Cell and Molecular Biology, School of Biology, College of Science, University of Tehran, 1417614411 Tehran, Iran

Correspondence Payam Zahedi, Nano‐Biopolymers Research Laboratory, School of Chemical Engineering, College of Engineering, University of Tehran, PO Box: 11155‐4563, Tehran, Iran. Email: [email protected]

The objective of this work was to synthesize molecularly imprinted polymer (MIP) nanoparticles based on methacrylic acid (MAA) monomer with a high selectivity against an anti‐cancer drug, 5‐fluorouracil (5‐FU), as a template. In this case, the nanoparticles were prepared via precipitation polymerization in the presence of ethylene glycol dimethacrylate as cross‐linker and azobisisobutyronitrile as initiator. Besides, 3 independent variables including MAA: 5‐FU molar ratio (X1), temperature (X2), and time (X3) were investigated utilizing response surface methodology. The scanning electron microscopy and dynamic light scattering resulted the average diameter of approximately 65 nm, and the MIP nanoparticle sample with the imprinting factor of 1.57 was polymerized in optimized conditions as follows: X1 = 6: 1, X2 = 60°C, and X3 = 3 days in acetonitrile as porogenic solvent. Also, Fourier transform infrared spectroscopy (FTIR) and thermogravimetric analysis confirmed the formation of MAA/5‐FU complex and lower thermal stability of the washed MIP sample than the unwashed MIP and non‐imprinted polymer (NIP) samples, respectively. Moreover, the optimized MIP nanoparticles have more controlled release of 5‐FU rather than the NIP sample. Finally, the flow cytometry showed that 5‐FU‐loaded MIP sample has the highest apoptosis of human colon cancer cell line, HCT‐116, after 3 days compared with NIP sample and also the exclusive use of drug. KEY W ORDS

5‐fluorouracil, cancer, drug delivery systems, molecular imprinting, nanoparticles

1

|

I N T RO D U CT I O N

Amid the extensive polymer carriers, molecularly imprinted polymers (MIPs) are promising candidates for biomedical and pharma-

In recent years, researchers have clearly shown that nanoscale medical

ceutical applications3,9-12 owing to a series of exceptional proper-

devices are capable of performing very complex tasks, ie, enhanced

ties.13-15 They are extremely networked polymer materials and are

drug delivery within biological tissues owing to their unique proper-

characterized by their high selectivity against a specific drug molecule

1,2

ties.

Many studies have been conducted to synthesize and develop

because of the presence of specific recognition sites within their

nanomaterials used for advanced drug delivery systems, especially

cavities.3,7 Briefly, the MIP synthesis is made by the following proce-

hydrophobic medicines which cannot be prescribed without a

dure: (1) the formation of a complex between a functional monomer

controlled dose program.3,4 In this regard, the application of polymer

and template (drug molecule) by means of non‐covalent interactions,

materials is strongly recommended due to their versatility, suitable

which are often hydrogen bonds, (2) reacting the provided complex

physical chemistry characteristics, and ability to conjugate with drug

through a radical polymerization by using a suitable crosslinking

molecules as well as acceptable rate of degradation under appropriate

agent,13,14,16 and (3) the template removal to form specific binding

5

biological conditions, preserving the concentration of drugs in their 6-8

therapeutic window.

Polym Adv Technol. 2018;1–9.

sites corresponding to the size, shape, and orientation of the targeted drug molecule.10,17,18

wileyonlinelibrary.com/journal/pat

Copyright © 2018 John Wiley & Sons, Ltd.

1

2

MADADIAN‐BOZORG

ET AL.

In the past half century, 5‐fluorouracil (5‐FU) has found a distinct

of nanoparticles, binding experiment between the MIPs and NIPs, the

place for chemotherapy in treatment of different cancer cells occurred

drug release kinetic, and 5‐FU‐loaded MIP nanoparticles performance

at liver, chest, neck and colon.19-22 The action mechanism of this drug

on programmed cancer cell death is obvious.

includes the inhibition of thymidylate synthase enzyme which inter-

This current work is aimed to synthesize MAA‐based MIP nano-

feres with the synthesis of nucleic acid and deoxyribonucleic acid;

particles containing 5‐FU by using precipitation polymerization. A

consequently, it can prevent the growth of the cancer cells.19,23

series of main independent variables of polymerization process such

Recent studies showed that the drug response rate particularly on

as MAA:5‐FU molar ratio, temperature, and time are optimized using

advanced colon cancer cells was less than 15%19 in spite of dissolving

D‐optimal design method to attain the sample with a minimum size

the drug in both aquatic and blood fluids. The most important reason

of diameter. Afterward, the optimized MIP and NIP samples are

for this low therapeutic level is the rapid metabolism of 5‐FU while

characterized in viewpoint of morphology, binding capacity, the drug

passing through the gastrointestinal track and its short half‐life in

release along with its kinetic, and induction of colon cancer cell line

the body (10‐20 min), limiting its availability in the human metabolism

apoptosis.

3,15,24

system.

Moreover, the continuous use of 5‐FU to compensate

the concentration fluctuations and to achieve the effective therapeutic doses can have chronic toxic effects on gastrointestinal pathway,

2

E X P E R I M E N T A L P RO C E D U R E

|

bone marrow, and nervous system as well as skin reactions during the course of the therapy. Therefore, the drug concentration not only

2.1

must be in its therapeutic dosage, but it should be also below the

Methacrylic acid (MAA, functional monomer, synthesis grade, purity of

toxicity threshold.5,21,22,24 By considering these shortcomings, it

98.5%), azobisisobutyronitrile (AIBN, thermal initiator, 98% purity),

seems that choosing an appropriate polymer carrier for transferring

and EGDMA (crosslinking agent, technical grade) were purchased from

5‐FU will be an inspiring solution for reducing the systemic adverse

Merck Co. (Darmstadt, Germany). Prior to use MAA, inhibitors like

side effects and achieving safer treatment processes.23,25-29

hydroquinone were removed via vacuum distillation. In addition, AIBN

|

Materials

Tummala et al30 synthesized chitosan nanoparticles to minimize

has been re‐crystallized by ethanol before use. 5‐Fluorouracil [5‐FU

the adverse side effects of 5‐FU. Their results exhibited that the drug

(C4H3FN2O2) with molecular weight of 130.077 g/mol, anti‐cancer

molecule was able to accumulate in intestine along with a sustained

drug, white, odourless and crystalline powder, and melting point of

release overnight. In addition, Sutar and co‐workers31 prepared poly

282°C] was provided from Merck Co. (Darmstadt, Germany). The cell

(lactic‐glycolic) acid nanoparticles containing 5‐FU using emulsion

line related to human colon cancer, HCT‐116, was obtained from the

droplet coalescence method in the presence of eudragit s‐100. The

cell bank of Pasteur Institute of Iran (Tehran, Iran). The other

in vitro results exhibited that 5‐FU nanoparticles had an acceptable

chemicals were technical reagent grades and used without further

potential for treating colon cancer and could kill approximately 80%

purification.

of the cancer cell line, HT‐29. Hosseinifar et al32 reported a newly published article on the preparation of pressure sensitive 5‐FU‐loaded nanogels based on alginate modified with β‐cyclodextrin as cross‐ linker through emulsification method. Their obtained results showed

2.2 | Synthesis of 5‐FU template‐based MIP and NIP nanoparticle samples

that the nanogels containing the drug could be employed as an

The overall schematic of 5‐FU template‐based MIP nanoparticles syn-

excellent candidate to overcome the inefficiency of the drug using

thesis via precipitation polymerization is shown in Figure 1. In this line,

33

exclusively in cancer therapy. Recently, Raza et al

synthesized chon-

5‐FU drug as template, MAA as functional monomer, EGDMA as

droitin sulphate‐poly (vinyl alcohol) cross‐linked microcapsules

crosslinking agent, and acetonitrile as porogenic solvent were used.

(miCAPs) using emulsion procedure for controlled delivery of 5‐FU.

In brief, the polymerization process for the formation of MIP nanopar-

Their results showed that the loading capacity of the drug was

ticle samples based on 5‐FU template was as follows: first, due to

75.3% and the drug release mechanism followed Fickian diffusion. In

MAA/5‐FU complex formation, the predetermined amount ratios of

another work, Jalalvandi et al34 designed dextran/polyhydrazide‐

MAA to 5‐FU (2, 4, and 6) were dissolved in 8 mL of acetonitrile in

based hydrogel network samples to investigate their capability for 5‐

100 mL of glass vessels and were then placed into an ultrasonic bath

FU release. The obtained outcomes revealed that these samples

with 80‐MHz frequency for 6 minutes. In this step, owing to providing

enabled to prolong the drug release for almost 2 days.

more stable hydrogen bonds between the monomer and template

By following up the literature, rare reports have been found on

groups, the resulting solutions were chilled in an ice and water bath

the MIPs containing 5‐FU. Instantly, Puoci and co‐workers22 synthe-

and stirred for 6 hours at 30 rpm. Subsequently, EGDMA and AIBN

sized methacrylic acid (MAA)‐based MIP nanoparticles containing 5‐

and the remaining solvent were added to the mixtures with the

FU using ethylene glycol dimethacrylate (EGDMA) as cross‐linker to

amounts of 16 mmol, 0.019 g, and 32 mL, respectively, and then they

investigate controlled release of the drug in a biological fluid. On

were placed again into the ultrasonic bath for 6 minutes. Thereafter,

one hand, their observations resulted in that the drug adsorption

the vessels were degassed by purging nitrogen gas for 20 minutes to

capacity of the MIP sample was higher than the non‐imprinted

ensure the oxygen molecules removal from the polymerization media.

polymer (NIP) ones. On the other hand, the MIPs gradually released

After carefully sealing the vessels, the prepared samples were slowly

5‐FU during 30 hours. By recapitulating this unique work, lack of

stirred at 20 rpm in an oil bath at 60°C until the polymerization

information about the effect of polymerization parameters on the size

reaction was started. It is worth noting that the polymerization

MADADIAN‐BOZORG

3

ET AL.

FIGURE 1 Overall schematic of precipitation polymerization of MAA‐based MIP nanoparticles containing 5‐FU as template [Colour figure can be viewed at wileyonlinelibrary.com]

from ambient to the desired temperature. After the polymerization

TABLE 1 design

reaction was accomplished, the synthesized MIP nanoparticle samples

Symbol

Variables

were washed with 100 mL of distilled water and 100 mL of warm

X1

MAA: 5‐FU (molar ratio)

2: 1

6: 1

methanol. The final step, the removal of the template molecules from

X2

Temperature (°C)

60

76

the polymeric networks, was carried out using a Soxhlet apparatus

X3

Time (day)

1

3

temperature was slowly increased over a period of 2 hours starting

The range of independent variables for the experimental Low Level

High Level

containing the washing mixture of methanol:acetic acid (9: 1 v/v) during 24 hours, and then pure methanol. Accordingly, a series of active sites through the MIP cavities were created. To ensure the template removal completely, the concentration of 5‐FU in the supernatant was measured using an ultraviolet‐visible (UV‐vis) spectrophotometer (2800, UNICO, United States) at maximum wavelength of

TABLE 2 D‐optimal design to study the effects of MAA: 5‐FU molar ratio (X1), polymerization temperature (X2), and polymerization time (X3) on the MIP average diameter (Y) Run

X1: MAA: 5‐FU (molar ratio)

X2: Temperature (°C)

X3: Time (day)

Y: Diameter (nm)

1

4

60

1

90

2

6

76

1

91

3

6

68

2

76

4

2

60

1

82

5

6

60

3

62

6

4

76

3

85

7

2

76

1

119

8

2

60

3

9

2

64

2.5

10

2

60

3

71

11

2

76

1

55

12

6

60

1

70

13

4

68

2

92

14

6

76

3

110

266 nm. Eventually, to eliminate the remaining solvent and unreacted monomer, the samples were washed with distilled water several times. In order to dry the samples, they were washed with acetone and placed into a vacuum chamber at temperature of 40°C for 24 hours. Furthermore, the NIP samples were prepared similar to the protocol for MIPs synthesis, exceptionally without 5‐FU as template followed by complex formation step.

2.3

|

D‐optimal design

D‐optimal design is a subset of response surface methodology and has received a great deal of attention to optimize the process conditions with different independent variables. Herein, by the use of this method, 3 numerical factors in 3 levels including MAA:5‐FU (X1), polymerization temperature (X2), and polymerization time (X3) were

81 107

selected. As is seen in Table 1, the ranges of X1, X2, and X3 were considered 2 to 6, 60°C to 76°C, and 1 to 3 days. Table 2 shows the

response (Y). It should be noted that by varying the amounts of X1

total runs of 14 recommended by D‐optimal design method, thereby

to X3 the amounts of other components including porogenic solvent,

providing the optimum condition for the minimal size of MIPs as a

initiator, and cross‐linker were kept constant.

4

2.4

MADADIAN‐BOZORG

|

Fourier transform infrared spectroscopy

factor (IF) as the main parameter in the molecular imprinted technique

Fourier transform infrared (FTIR) spectroscopy (model Spectrum

was calculated according to Equation (3).

Two, PerkinElmer Inc., United States) was utilized to evaluate the

IF ¼

formation of complex between the monomer and template. The −1

wavenumber range was considered 400 to 4000 cm

ET AL.

QMIP ;; QNIP

(3)

at resolution

of 4 cm−1.

2.8 2.5

|

Thermogravimetric analysis

|

In vitro drug release

To evaluate the 5‐FU release and its trend from the MIP and NIP

The thermal behavior of the samples including washed and unwashed

nanoparticles samples, 50 mg of each sample containing the drug

MIPs as well as NIP sample was assessed using thermogravimetric

was dispersed in 5 mL of phosphate buffer solution (PBS, pH 7.4)

analysis (TGA, model Q50, TA Instrument, United States). By consider-

and put into a dialysis bag (molecular weight cut off 12 kD, Sigma‐

ing initial weight around 5 mg for the 3 samples, the weight loss

Aldrich Co., Pilsburg, The Netherlands) sealed and soaked in 120 mL

percentage was studied. This experiment was carried out in tempera-

of PBS stirring at 50 rpm at 37°C. The samplings with volume of

ture ranging from 25°C to 600°C at the rate of 20°C/min under argon

3 mL were withdrawn from the medium at predetermined time inter-

atmosphere with flow rate of 10 mL/min.

vals; instead, an equivalent volume of the fresh PBS was added to keep the volume of the medium at 120 mL. This trend was done up

2.6 | Scanning electron microscopy and dynamic light scattering

to 96 hours, and the drug concentration in the supernatant was

The morphology observation of PMAA‐based MIP nanoparticles was

samples as a function of time. Moreover, to analyze the data on drug

evaluated using a scanning electron microscopy (SEM, model AIS

release from the samples, the release kinetic was studied using

2100, Seron Technology, South Korea) at 30 000 × magnification.

Korsmeyer‐Peppas equation (Equation (4)).

measured by the UV‐vis spectrophotometer at 266 nm. Finally, the cumulative release profile of 5‐FU was plotted for MIP and NIP

The dynamic light scattering (DLS, model ZEN 3600, Malvern Mt ¼ ktn M0

Instruments, Worcestershire, UK) was used to evaluate the average diameter and the size distribution of the nanoparticles.

2.7

|



 Mt < 0:6 ; M0

(4)

where “Mt” is the concentration of drug at time “t”, and “M0” is the

Binding capacity experiments

total concentration of drug loaded. Also, “k” is a constant coefficient

The absorption efficiency of 5‐FU drug in the polymer matrix and

of the drug‐polymer system, which depends on the release of the drug

drug loading on MIP and NIP nanoparticles dispersed into acetonitrile

from the polymeric network. The propagation power or characteristic

were investigated at pH of 7.4 and ambient temperature. In summary,

of the discharge mechanism is showed by “n”, which illuminates the

100 mg of the MIP and NIP nanoparticles was weighed and

mechanism of its release process. For further information, when “n”

suspended in 30 mL of acetonitrile containing 5‐FU with concentra-

is less than 0.5, the release of the drug in the system follows a Fickian

tion of 25 μg/mL. The mixture was stirred at 300 rpm for 6 hours

diffusion or propagation phenomenon. In the case where 0.5 < n < 1,

in an ice and water bath, and then it was centrifuged at

its release mechanism is in accordance with non‐Fickian diffusion. If

12 000 rpm for 30 minutes. Consequently, the absorption amount

“n” is 1, the release mechanism is time independent and the model is

of 5‐FU in the supernatant was measured using the UV‐vis spectro-

governed by zero‐order equation.

photometer at 266 nm, and its concentration was calculated using the calibration curve (Equation (1)). A ¼ 0:0887 × C;

2.9 (1)

|

MTT assay

The cell cytotoxicity (ISO 10993‐12) of the MIP nanoparticles containing different concentrations of 5‐FU (10, 25, and 50 μg/mL) was

where “A” and “C” are the absorbance intensity and concentration of

characterized at overnight by using 3‐[4‐dimethyl thiazol‐2‐yl]‐2, 5‐

the 5‐FU in distilled water, respectively.

diphenyl tetrazolium bromide, thaizolyl blue (MTT) according to the

The amount of 5‐FU bound to the polymer matrix was obtained

following steps: HCT‐116 cells were cultivated at a density of

by comparing the concentration of drug in MIP and NIP samples.

1 × 104 cells/well onto 96‐well plates (Nunc, Denmark) in Roswell

The capacity of absorption (Q) as the amount of bonded template

Park Memorial Institute medium containing 10% fetal bovine serum

(mg) to the amount of dried nanoparticles (mg) was obtained according

and 1% antibiotics (penicillin/amphotericin) for 24 hours in the incuba-

to Equation (2).

tor (in moist atmosphere with 5% CO2 at 37°C). On the other hand, Q¼

ðC0 −CÞ × V ;; W

the MIPs were put in an autoclave (15 minutes at pressure of (2)

1.5 bar and temperature of 121°C), then relocated in extraction medium (a Roswell Park Memorial Institute medium with 10% fetal

where “C0” is the initial drug concentration, “C” is the concentration of

bovine serum) for 5 days. The extract was collected and added to

drug in the supernatant, “V” (mL) is the volume of adsorption medium,

the full‐growth media of confluent HCT‐116 cells to obtain the

and “W” (mg) is the mass of the nanoparticles. Moreover, imprinting

predetermined drug concentrations and incubated for 24 hours, then

MADADIAN‐BOZORG

5

ET AL.

100‐μL PBS containing 0.5 mg mL−1 MTT was added to each cell, and cells were once more incubated for 4 hours at 37°C in CO2

3.2 | Optimizing the different variables of polymerization

atmosphere. Then, the resultant formazan crystals were suspended into solubilization buffer. The optical density of the solution was measured at a wavelength of 545 nm by ELISA reader device.

According to the Table 2, D‐optimal design gave 14 runs to evaluate and optimize the involved 3 variables (X1‐X3) as a function of average diameter of nanoparticles (Y). The obtained experimental data were evaluated by analysis of variance (ANOVA) and a generalized 2‐factor interaction (2FI) equation was estimated using the software as follows:

2.10

|

Flow cytometry analysis

Apoptosis (programmed cell death) illustrates a chief causative parameter for the anticancer activity of a drug.35,36 Flow cytometry analysis

Y ¼ 151:7 þ 23:7X1 − 0:23X 2 −96:7X 3 − 0:45X1 X 2 þ 0:1X 1 X3 þ 1:3X2 X3 ;

(5)

is a powerful tool to investigate whether apoptosis can result in the cell death induced by drug‐loaded system. In this work, human colon cancer cell line, HCT‐116, was pre‐ treated with 5‐FU‐loaded MIP, 5‐FU‐loaded NIP, washed MIP, and 5‐FU drug. In order to determine the percentage of apoptosis and necrosis, the cells were stained with Annexin V/PI (eBioscience detection kit FITC). Then, they were cultured in a 6‐well plate and allowed to reach 50% confluence before incubation with the nanoparticles samples to induce cell death. After 72 hours, first, the cells were removed from the plate using trypsin, and the precipitates were collected by the use of centrifuge at 3500 rpm. The precipitated cells with approximately 1 to 1.5 × 106 cells were suspended in 200 μL of binding buffer. Then, the cells contained with Annexin‐FITC dye (5 μL) and PI (5 μL) were incubated at suitable time and analyzed using a flow cytometer device. The data analysis was performed using FLowJo software. The determination of cell population and distribution was performed using 2‐dimensional plots. The areas of the 4 regions (Q1‐Q4) were determined using the software which are defined as the percentage of necrotic cells, old apoptotic cells, young apoptotic cells, and natural cell, respectively.

In Equation (5), in order to get the minimum diameter for MAA‐ based MIP sample which was approximately 65 nm, the optimized conditions for X1, X2, and X3 parameters were as follows: MAA:5‐FU (X1) = 6, the polymerization temperature (X2) = 60°C, and the polymerization time (X3) = 3 days. The regression coefficient (r2) for the Equation (5) was calculated 0.92. Generally, every parameter eliminated in this equation was because of their unimportance effects on the answer, had a P‐value > 0.05. Supplementary 2 (a‐f) [S. 2 (a‐f)] depicts the 3‐dimensional (3‐D) and interactions of 3 parameters on the answer in the optimum conditions. As is seen in S. 2 (a‐c), it was clear that by increasing the MAA:5‐ FU ratio and time and decreasing the temperature, the average diameter of the MIP sample was decreased. Also, the interaction between the time and temperature has an antagonistic property on the answer [S. 2 (f)]. Moreover, as is evident from S. 2 (d), the interaction of 2 parameters namely MAA: 5‐FU and temperature has a synergistic property, thereby, increasing the temperature led to decrease the average diameter of MIP nanoparticles. At last, S. 2 (e) revealed that the interaction of MAA:5‐FU and time has no effect on each other. It could be concluded that amid the 3 variables, temperature has more effect on the average diameter of nanoparticles sample compared with

3

|

RESULTS AND DISCUSSION

the other parameters. Regarding the monomer concentration, it is worth noting that in higher monomeric ranges, due to the dimerization

3.1 | Formation of the complex between the MAA and 5‐FU Supplementary 1 (S. 1) shows FTIR spectra of the functional groups between MAA and 5‐FU before and after pre‐polymerization to study the formation of complex. As is seen from S. 1, the locations of

of the MAA, the size of the nanoparticles has increased, in which the MAA:5‐FU ratios higher than 6 have not been considered.

3.3 | Morphology and size distribution of the optimized MIP nanoparticles sample

characteristic peaks were identical before and after the formation of

According to Figure 2A,B, it is clear that the MIP nanoparticles synthe-

complex. After complex formation, 4 characteristic peaks could be

sized in optimal conditions have a uniform shape with narrow size dis-

observed at locations of 1100 cm−1, 1451 cm−1, 1650 cm−1, and

tribution. Besides the effective parameters, in general, the choice of

−1

1718 cm . These characteristic peaks in the order mentioned above

suitable solvent, plays a significant role in the size of synthesized

were attributed to the stretching vibration of C―O, bending motion

MIP nanoparticles and their distribution as well. Therefore, MIP syn-

of ―CH2―, N―H bond, and vibration of ―C═O. Furthermore, the

thesis in the presence of acetonitrile, the solvent used in this work,

presence of H―F bond through the complex formation could not be

seemed more effective. Because acetonitrile is subset of aprotic polar

‐1 37

Although, O―H

solvents, it has good compatibility with the all components involved in

bond was observed around 3500 cm−1 before complex formation, its

the polymerization process, leading to faster deposition of polymer-

intensity was increased after complex happening due to strong

ized precipitates, as a result, the size of the MAA‐based MIP nanopar-

interactions between MAA and 5‐FU functional groups. Therefore,

ticles underwent smaller diameter. By considering the obtained results

the formation of complex between the functional groups on the struc-

from SEM, the average diameter for optimized MIP sample was

tures of MAA (monomer) and 5‐FU (template) was the fundamental

around 65 nm. Nevertheless, the DLS outcome showed greater aver-

procedure to create active sites in the cavities of the MIP samples.

age diameter size for the sample approximately 100 nm because of

seen owing to its wavenumber around 4200 cm .

6

MADADIAN‐BOZORG

ET AL.

FIGURE 3 TGA curves of washed MIP, unwashed MIP, and NIP [Colour figure can be viewed at wileyonlinelibrary.com]

5‐FU as template. Secondly, the second weight loss was observed at temperature of around 320°C to 340°C, which was related to the degradation of the polymeric network. In addition, by comparing the thermal curves of the NIP and unwashed MIP samples, they indicated that the stability of the unwashed MIP sample was a little higher than the NIP sample due to the network structure in MIPs. Eventually, all the polymer nanoparticles were completely decomposed at 470°C and converted to ashes.

3.5 | Binding capacity and the drug loading experiments In general, the potential of polymer matrices (MIP and NIP) for the detection and absorption of drug is influenced by various factors such as pH, monomer to drug ratio, type of absorption medium, and drug concentration, etc. According to the results of the previous study,4 FIGURE 2 A, SEM micrograph image of the optimized MIP sample (the scale bar is 1 micron); B, Gaussian profile of nanoparticles size for the optimal MIP; and C, DLS results of optimal MIP sample [Colour figure can be viewed at wileyonlinelibrary.com]

the absorption of drug in acidic and non‐acidic environments was not clear very well, but at pH 7.4, there was the greatest difference between MIP and NIP samples, because the concentration of OH− and H+ was low, and the drug and monomer were in the best condition for the formation of a bond. In the acidic pH, both the drug and

the swelling of the nanoparticles in the solvent medium during the

the monomer were protonated, and the monomer was ionized and

sample preparation for the test (Figure 2C).

degraded in the basic pH; consequently, there was no possibility of effective hydrogen bonding between the monomer and the drug. For

3.4 | Thermal behaviors of the optimized MIP and NIP nanoparticles samples

this reason; herein, the adsorption test has been investigated at pH 7.4. Therefore, by adjusting the conditions for obtaining the binding capacity (Q) of the optimized MIP sample and NIP ones, this

Figure 3 illustrates the thermal behaviors of the unwashed MIP,

value was calculated 21.4 and 13.6, respectively. In conclusion, by

washed MIP, and NIP samples. In this test, the variation of weight loss

dividing these values, the IF was obtained 1.57 which showed the

of the samples as a function of temperature was studied. From these

remarkable efficiency of the optimized MIP nanoparticles sample to

TGA curves, 2 notifications could be considered: firstly, the thermal

select the 5‐FU molecules as template.

stability of the samples, in which the initial weight loss of the washed MIP sample was occurred at around 200°C, while this trend was happened for unwashed MIP and NIP samples at around 340°C. It was due to the recognition active sites in the cavities of washed

3.6 | In vitro drug release of the optimized MIP and NIP nanoparticles samples

MIP sample which were highly sensitive against higher temperatures

5‐FU release studies related to the optimal MIP and NIP samples were

whereas in the unwashed MIP sample the presence of hydrogen

simultaneously performed in PBS at pH 7.4 and 37°C. Figure 4 clearly

bonds led to higher thermal stability for those provided sites. Also,

shows that the drug release from NIP sample was significantly faster

in the NIP sample, there was no active sites because of the lack of

than MIP ones. In other words, because of the presence of special

MADADIAN‐BOZORG

7

ET AL.

and the accumulative release was calculated around 60%. Whereas, in the similar conditions mentioned in this work, the prepared nanoparticles showed higher extent of release time (96 hours) with the total accumulative drug release of 80%. In order to analyze the release kinetic of 5‐FU from the samples, the well‐known Korsmeyer‐Peppas equation was fitted on the experimental data and exponent value of “n” was calculated approximately 0.43. As mentioned before, this value revealed that the main mechanism of the drug release was governed by Fick's law.

3.7 | Cell cytotoxicity of the optimized MIP nanoparticles sample containing 5‐FU FIGURE 4 The release profile of 5‐FU from the MIP and NIP samples in PBS at 37°C and pH 7.4 during 96 hours [Colour figure can be viewed at wileyonlinelibrary.com]

Cytotoxicity of 5‐FU‐loaded MIP nanoparticles sample on HCT‐116 cells was studied using MTT assay [Supplementary 3 (S. 3)]. As is shown, a high viability (80%‐95% as compared with control) of HCT‐

recognition sites through the MIP nanoparticles structure, they were

116 cells was remarkable when exposed to 10, 25, and 50 μg mL−1

able to control the release of 5‐FU rather than NIP sample. Further-

of the sample. These results exhibited that 5‐FU‐loaded MIP sample

more, the initial burst release for both samples was occurred before

in this concentration range has a non‐toxic effect on HCT‐116 cells;

10 hours which was associated with the adsorption and poor non‐spe-

as a result, it could be considered biocompatible.

cific interactions between the polymer structure and template. After 10 hours, it could be seen that the MIP sample has a gradually release profile with a gentle slope, thereby diffusing out 80% of the drug released until 96 hours. To compare the obtained results with those

3.8 | Induction of cell apoptosis for the optimized MIP and NIP samples

reported by Puoci and coworkers,22 it is worth noting that their MIP

Figure 5A‐D demonstrates the cell apoptosis patterns for the

samples based on PMAA could diffuse out 5‐FU during 30 hours,

samples including washed MIP, NIP, 5‐FU‐loaded MIP and exclusive

FIGURE 5 Apoptosis assay by Annexin V‐FITC/PI, A, exposed cells to 25 μg/mL of the exclusive use of 5‐FU; B, exposed cells to NIP sample; C, exposed cells to the washed MIP sample; and D, exposed cells to 25 μg/mL of 5‐FU‐loaded MIP sample [Colour figure can be viewed at wileyonlinelibrary.com]

8

MADADIAN‐BOZORG

use of 5‐FU (control) after 3 days. By comparing Figure 5A‐D, it could be observed that the exclusive use of 5‐FU has the lowest cell death (23.92%) (Figure 5A) compared with NIP (39.22%) (Figure 5B), washed MIP (51.39%) (Figure 5C) and 5‐FU‐loaded MIP (89.64%) (Figure 5D) samples. It could be referred to the short half‐life this drug (10‐20 min) which significantly reduced its efficiency. Regarding the cell apoptosis window related to washed MIP, the programmed cell line death was also minimal because of MAA in the MIP sample in the absence of drug (owing to washing) did not have any influence on cancer cell line apoptosis. However, the amount of cell apoptosis of washed MIP was higher than the exclusive use of 5‐FU, because of the residual drug might be remained in the MIP's cavities during the washing. Interesting results were considered in the NIP and 5‐FU‐ loaded MIP samples for the induction of HCT‐116 cell death (Figure 5 B,D). As is evident, the cell apoptosis of 5‐FU‐loaded MIP sample was higher than NIP ones because of sustained control release of the drug from MIP nanoparticles. However, the NIP results on this test were considerable as well owing to the presence of the drug molecules adsorbed onto the surface of the sample.

4

|

C O N CL U S I O N S

To synthesize MIP based on MAA for higher efficiency to select 5‐FU molecules and having great binding capacity along with minimum average diameter of the nanoparticles, a series of formulations were designed using D‐optimal method and were then evaluated. By considering and optimizing 3 independent parameters including monomer:template molar ratio of 6, polymerization temperature of 60°C, and polymerization time of 3 days, the minimum size of nanoparticles sample was obtained around 65 nm in the presence of acetonitrile as porogenic solvent. To study the thermal stability of the samples before and after washing and also NIP sample, they showed that active sites thermal sensitivity through the sample removed the template. On the other hand, the MIP based on MAA monomer had a controlled release of 5‐FU compared with NIP sample due to the recognition sites formed in the cavities. Also, the release kinetic followed Fickian diffusion theory in the PBS. Finally, the MIP sample showed a high cytocompatibility and could effectively improve cell line (HCT‐116) apoptosis. In summary, the different experiments confirmed that the MIP based on MAA was the promising candidate for adsorption and controlled release of 5‐FU for using in colon cancer therapy potentially. ORCID Payam Zahedi

http://orcid.org/0000-0001-6636-4534

RE FE R ENC E S

ET AL.

4. Aeinehvand R, Zahedi P, Kashani‐Rahimi S, Fallah‐Darrehchi M, Shamsi M. Synthesis of poly (2‐hydroxyethyl methacrylate)‐based molecularly imprinted polymer nanoparticles containing timolol maleate: morphological, thermal, and drug release along with cell biocompatibility studies. Polym Adv Technol. 2017;28(7):828‐841. 5. Banerjee A, Pathak S, Vasan D, Murugesan R, Verma RS. Strategies for targeted drug delivery in treatment of colon cancer: current trends and future perspectives. Drug Discov Today. 2017;22(8):1224‐1232. 6. Shamsi M, Zahedi P, Ghourchian H, Minaeian S. Microfluidic‐aided fabrication of nanoparticles blend based on chitosan for a transdermal multidrug delivery application. Int J Biol Macromol. 2017;99:433‐442. 7. Cirillo G, Parisi OI, Curcio M, et al. Molecularly imprinted polymers as drug delivery systems for the sustained release of glycyrrhizic acid. J Pharm Pharmacol. 2010;62(5):577‐582. 8. Kryscio DR, Peppas NA. Mimicking biological delivery through feedback‐controlled drug release systems based on molecular imprinting. AIChE J. 2009;55(6):1311‐1324. 9. Oliveira D, Gomes CP, Dias RCS, Costa MRPFN. Molecular imprinting of 5‐fluorouracil in particles with surface RAFT grafted functional brushes. React Funct Polym. 2016;107:35‐45. 10. Sinha VR, Mittal BR, Bhutani KK, Kumria R. Colonic drug delivery of 5‐ fluorouracil: an in vitro evaluation. Int J Pharm. 2004;269(1):101‐108. 11. Zaidi SA. Molecular imprinted polymers as drug delivery vehicles. Drug Deliv. 2016;23(7):2262‐2271. 12. Madikizela L, Tavengwa N, Pakade V. Molecularly imprinted polymers for pharmaceutical compounds: synthetic procedures and analytical applications. In: Recent Research in Polymerization. London: InTech; 2018:47‐67. 13. Abdouss M, Asadi E, Azodi‐Deilami S, Beik‐Mohammadi N, Aslanzadeh SA. Development and characterization of molecularly imprinted polymers for controlled release of citalopram. J Mater Sci‐Mater M. 2011;22(10):2273‐2281. 14. Asadi E, Abdouss M, Leblanc RM, Ezzati N, Wilson JN, Azodi‐Deilami S. In vitro/in vivo study of novel anti‐cancer, biodegradable cross‐linked tannic acid for fabrication of 5‐fluorouracil‐targeting drug delivery nano‐device based on a molecular imprinted polymer. RSC Adv. 2016;6(43):37308‐37318. 15. Mohebali A, Abdouss M, Mazinani S, Zahedi P. Synthesis and characterization of poly (methacrylic acid)‐based molecularly imprinted polymer nanoparticles for controlled release of trinitroglycerin. Polym Adv Technol. 2016;27(9):1164‐1171. 16. Zaidi SA. Latest trends in molecular imprinted polymer based drug delivery systems. RSC Adv. 2016;6(91):88807‐88819. 17. Xiao Y, Xiao R, Tang J, et al. Preparation and adsorption properties of molecularly imprinted polymer via RAFT precipitation polymerization for selective removal of aristolochic acid I. Talanta. 2017;162:415‐422. 18. Varaprasad K, Vimala K, Ravindra S, Reddy NN, Reddy GSM, Raju KM. Biodegradable chitosan hydrogels for in vitro drug release studies of 5‐fluorouracil an anticancer drug. J Polym Environ. 2012;20(2):573‐582. 19. Chandran SP, Natarajan SB, Chandraseharan S, Shahimi MSBM. Nano drug delivery strategy of 5‐fluorouracil for the treatment of colorectal cancer. J Cancer Res Pract. 2017;4(2):45‐48.

1. Zhao J, Mo Z, Guo F, Shi D, Han QQ, Liu Q. Drug loaded nanoparticle coating on totally bioresorbable PLLA stents to prevent in‐stent restenosis. J Biomed Mater Res B Appl Biomater. 2018;106(1):88‐95.

20. Sinkel C, Greiner A, Agarwal S. A polymeric drug depot based on 7‐(2′‐ methacryloyloxyethoxy)‐4‐methylcoumarin copolymers for photoinduced release of 5‐fluorouracil designed for the treatment of secondary cataracts. Macromol Chem Phys. 2010;211(17):1857‐1867.

2. Shamsi M, Zahedi P. On‐chip preparation of streptokinase entrapped in chitosan nanoparticles used in thrombolytic therapy potentially. J Pharm Sci. 2017;106(12):3623‐3630.

21. Cirillo G, Iemma F, Puoci F, et al. Imprinted hydrophilic nanospheres as drug delivery systems for 5‐fluorouracil sustained release. J Drug Target. 2009;17(1):72‐77.

3. Luliński P. Molecularly imprinted polymers based drug delivery devices: a way to application in modern pharmacotherapy. A review. Mat Sci Eng C‐Mater. 2017;76:1344‐1353.

22. Puoci F, Iemma F, Cirillo G, Picci N, Matricardi P, Alhaique F. Molecularly imprinted polymers for 5‐fluorouracil release in biological fluids. Molecules. 2007;12(4):805‐814.

MADADIAN‐BOZORG

9

ET AL.

23. Rahman Z, Kohli K, Khar RK, Ali M, Charoo NA, Shamsher AAA. Characterization of 5‐fluorouracil microspheres for colonic delivery. AAPS PharmSciTech. 2006;7(2):113‐121. 24. Minhas MU, Ahmad M, Anwar J, Khan S. Synthesis and characterization of biodegradable hydrogels for oral delivery of 5‐fluorouracil targeted to colon: screening with preliminary in vivo studies. Adv Polym Tech. 2018;37(1):221‐229. 25. Cetin K, Denizli A. 5‐Fluorouracil delivery from metal‐ion mediated molecularly imprinted cryogel discs. Colloid Surface B. 2015;126:401‐406. 26. Huynh TP, Pieta P, D'Souza F, Kutner W. Molecularly imprinted polymer for recognition of 5‐fluorouracil by RNA‐type nucleobase pairing. Anal Chem. 2013;85(17):8304‐8312. 27. Men JY, Gao BJ, Chen ZP, Yao L. Preparation of 5‐fluorouracil‐surface imprinted microspheres MIP‐PSSS/CPVA and studies on its character of colon‐specific drug delivery. Acta Chim Sinica. 2012;70(21):2273‐2280. 28. Singh B, Chauhan N. Preliminary evaluation of molecular imprinting of 5‐fluorouracil within hydrogels for use as drug delivery systems. Acta Biomater. 2008;4(5):1244‐1254. 29. Kavousipour S, Khademi F, Zamani M, Vakili B, Mokarram P. Novel biotechnology approaches in colorectal cancer diagnosis and therapy. Biotechnol Lett. 2017;39(6):785‐803. 30. Tummala S, Kumar MNS, Prakash A. Formulation and characterization of 5‐fluorouracil enteric coated nanoparticles for sustained and localized release in treating colorectal cancer. Saudi Pharm J. 2015;23(3):308‐314. 31. Sutar PS, Joshi VG. Preparation and characterisation of 5‐fluorouracil loaded PLGA nanoparticles for colorectal cancer therapy. UJPBS. 2013;1(2):52‐58. 32. Hosseinifar T, Sheybani S, Abdouss M, Najafabadi SAH, Ardestani MS. Pressure responsive nanogel base on alginate‐cyclodextrin with

enhanced apoptosis mechanism for colon cancer delivery. J Biomed Mater Res. 2018;106(2):349‐359. 33. Raza H, Ranjha NM, Mahmood A, Azam F, Sarfraz RM, Rashid Z. Development and in vitro characterization of 5‐fluorouracil loaded, colon‐ targeted drug delivery system. Trop J Pharm Res. 2018;17(2):195‐204. 34. Jalalvandi E, Hanton LR, Moratti SC. Preparation of a pH sensitive hydrogel based on dextran and polyhydrazide for release of 5‐fluorouracil, an anticancer drug. J Drug Deliv Sci Tec. 2018;44:146‐152. 35. Zheng JS, Zheng SY, Zhang YB, et al. Sialic acid surface decoration enhances cellular uptake and apoptosis‐inducing activity of selenium nanoparticles. Colloid Surface B. 2011;83(1):183‐187. 36. Kiho T, Sakushima M, Wang S, Nagal K, Ukai S. Polysaccharides in fungi. XXVI. Two branched (1→ 3)‐β‐D‐glucans from hot water extract of Yu er. Chem Pharm Bull. 1991;39(3):798‐800. 37. Grayson M. The infrared and Raman spectroscopic signals of HF, HCl, HBr and HI. Internet Electron J Mol Des. 2005;4(11):786‐792.

SUPPORTI NG INFORMATION Additional supporting information may be found online in the Supporting Information section at the end of the article.

How to cite this article: Madadian‐Bozorg NM, Zahedi P, Shamsi M, Safarian S. Poly (methacrylic acid)‐based molecularly imprinted polymer nanoparticles containing 5‐fluourouracil used in colon cancer therapy potentially. Polym Adv Technol. 2018;1–9. https://doi.org/10.1002/pat.4353