Physically cross-linked pH-responsive hydrogels with tunable formulations for controlled drug delivery

Abstract

A variety of pH responsive hydrogels possessing macroporous interiors resembling a honey-comb framework with a continuous skin on the surface have been developed by free radical aqueous copolymerization of acrylic acid (AAc) and 2-(dimethylamino)ethyl methacrylate (DMAEMA) (poly(AAc-co-DMAEMA) (PAD) hydrogels). This one step polymerization process makes scaling-up easier for mass production. Our formulations, being devoid of any chemical cross-linkers, remained dimensionally stable in buffer solutions of pH 1.2–7.4 with interlocked nanogels being identified as the building blocks of the network structures. Fourier transform infrared (FTIR) spectroscopy, differential scanning calorimetry (DSC), uniaxial compression testing and scanning electron microscopy (SEM) were used to characterize the hydrogels (PADs). Compressive elastic modulus and compressive strength of the swollen hydrogels at pH 7 were found to vary with composition from ∼3 to ∼11 kPa and ∼178 to ∼206 kPa, respectively. The swollen gels showed fairly strong viscoelastic behaviour and underwent deformation from ∼70% to 85% before failure, indicating the formation of robust 3D structures of PADs. Preliminary investigation into the biocompatibility of our hydrogels done by cytotoxicity assays using HeLa and McCoy mouse fibroblast cell lines have revealed that they are non-cytotoxic, paving the way for further biomedical applications. Swelling behaviour and release kinetics of bovine serum albumin (BSA) were investigated in various buffer solutions that mimic the pH-metric hierarchy in the gastrointestinal (GI) tract. Equilibrium swelling ratio was found to vary from 171% (mass) to 2027% (mass) depending on the pH and composition of hydrogels. Different compositions of PAD systems were investigated to verify the possibility of tailor-making the drug release behaviour of PAD formations.

---

Introduction

Protein drugs, also referred to as biologics, have become increasingly important over the past few years with the advent of high throughput screening and breakthroughs in genetics and proteomics. Protein drugs are pharmaceuticals based on proteins or peptides, such as modified and engineered antibodies, hormones, interferons/interleukins, enzymes and growth factors.^1,2 Many of the most prevalent diseases, such as diabetes, cancer and cardiovascular disease are particularly amenable to treatment with protein drugs.¹

Conventionally, delivery of protein drugs has been through the parenteral route of administration despite poor patient compliance due to frequent administration, varying drug levels in blood stream, painful procedure and risk of local infections.^2,3 Oral route is the most desirable pathway for drug administration, particularly for therapeutic agents in which several doses are necessary, however; protein and peptide drugs cannot be administered through the oral route because of susceptibility to acidic pH of stomach, enzymatic degradation in the upper portion of small intestine and poor permeability across intestinal epithelium due to its high molecular weight and lack of lipophilicity.^4–6 The oral bioavailability of most peptides and proteins therefore is less than 1%.^7,8 Thus, improving oral protein bioavailability to anywhere between 30–50% is one of the most challenging goals in pharmaceutics and in drug delivery technology.⁷ A major approach, among attempts underway to face this challenge, is to formulate protein and peptide drugs in a carrier that will endure the adverse gastro-intestinal tract (GIT) conditions.³

Various carrier systems including hydrogels, microspheres, micro emulsions, nanoparticles, liposomes and micelles have been designed in an attempt to facilitate drug permeation, inhibit proteolytic hydrolysis and prolong intestinal retention, thereby enhancing the oral bioavailability.^3,4,9–11 Attempts have also been made to deliver protein and peptide drugs to the colon due to low activity of proteolytic enzymes in the colon.¹² Many systems such as time-dependent systems, pH-sensitive coatings and enzymatically degradable systems have been tested for colon specific delivery.¹²

Hydrogels, being three-dimensional, hydrophilic, and polymeric compounds have the ability to imbibe substantial amounts of water, therapeutic agents and other solvents.^13–15 They are also capable of simulating some physical characteristics of natural living tissues more than any other synthetic class of biomaterials.^16–19 Natural polymers have certain drawbacks, like uncontrolled rate of hydration, microbial contamination, and drop in viscosity on storing.²⁰ The hydrophilic networks and high water content contributes to the biocompatibility of the material. Therefore, hydrogels are an important class of materials that can be used for the controlled release of pharmaceutical proteins.

Formation of hydrogels can be achieved by both chemical and physical crosslinking.²¹ However, chemical cross-linkers are inherently toxic and can possibly damage the entrapped bioactive substance leading to a loss of activity; moreover their complete removal needs to be ensured before in vivo application. In contrast, physically cross-linked hydrogels have non-permanent bonds, based on physical interactions between the polymer chains.^22,23 These hydrogels eliminate the use of toxic cross-linkers.

To the best of our knowledge, there have been no reports on the synthesis of a stable hydrogel via aqueous copolymerization of acrylic acid (AAc) and 2-(dimethylamino)ethyl methacrylate (DMAEMA). Traditionally, scientists have attempted to create hydrogels by copolymerization via chemical cross-linking²⁴ or radiation techniques,²⁵ but these techniques have limitations such as toxicity of the chemical cross-linkers and inhomogeneous cross-linking.²⁶ A more recent approach has been the use of multi-step polymerization processes. Sherif et al.^22,27 have reported the successful immobilization of β-galactosidase enzyme in such hydrogels synthesized by a multi-step polymerization process called template polymerization.

Earlier attempts have been made for the synthesis of pH-sensitive hydrogels using AAc and N-[3-(dimethylamino)propyl]-methacrylamide. However, the hydrogels produced with this reported formulation requires a greater concentration of accelerators and initiators (10 fold and 20 fold respectively). They also exhibit lesser average glass transition temperature (T_g) which influences the cross-linking, swelling behavior and drug loading capacity of the hydrogels.²⁸

DMAEMA was deemed a suitable co-monomer as literature survey offered a plethora of applications involving poly-DMAEMA or various polymers with DMAEMA as one of its key constituents. DMAEMA has been used in formulations for development of drug delivery of therapeutic agents.²⁹ Owing to its pH responsive characteristics, poly-DMAEMA microgels have been synthesized with polystyrene,³⁰ acrylamide,³¹ and pyrrolidone³² for drug delivery. A self regulating insulin delivery system with poly-DMAEMA as one of its co-monomer, has been investigated by Satish and Shivakumar.³³ The applications of poly-DMAEMA are also extended to gene delivery systems,³⁴ use in cosmetics, demulsifiers, flocculants etc. DMAEMA has Food and Drug Administration (FDA) clearance for use as basic components of single and repeat use food contact surfaces (refer to US Code of Federal Regulations Title 21: 21CFR 117.1010).

Drug release can be triggered by various stimuli, including redox potential, light, temperature or pH.^35–45 pH-responsive hydrogels such as poly(acrylic acid) (PAAc) hydrogels are a class of smart hydrogels and have been shown to have potential use in site-specific delivery of drugs to the GIT due to their muco-adhesive properties.^46–48 These hydrogels are also capable of chelating divalent ions (e.g. calcium and zinc), which in turn affect the thermodynamic stability and the biological activity of calcium-dependent proteolytic enzymes like trypsin and chymotrypsin.^47,48 A reduction in extracellular divalent ion concentration can also result in the opening of epithelial tight junctions, allowing paracellular drug transport across the intestinal epithelium.⁴⁸

Superabsorbent hydrogels of PAAc and its copolymers can swell tens to hundred times the original weight upon contact with the intestinal fluids, and can thereby tightly attach to the gut wall via mechanical fixation. This will also facilitate opening of the epithelial tight junctions for enhanced paracellular drug transport and prolonged intestinal retention, irrespective of mucin turn-over.⁴

The present study reports a novel, facile and cost effective technique of single step aqueous copolymerization to develop a series of physically cross-linked, stable, superabsorbent and pH-responsive hydrogels from AAc and DMAEMA by varying the monomer feed ratios. These poly-(AAc-co-DMAEMA) (PAD) hydrogels designated as PAD50, PAD70, PAD80 and PAD90, where the numbers indicate the mol% of AAc in the monomer feed ratio, are composed of interlocked nanogels and non-porous external walls, which develop in situ during polymerization. Characterization of the PAD matrices has been established in terms of chemical composition, T_g, mechanical properties (viz. compressive modulus, compressive strength and deformation at failure), morphology and pH-dependent swelling property. Cytotoxic effects of the hydrogels were also examined in HeLa cells and in McCoy mouse fibroblasts cells via ‘indirect contact’ cytotoxicity assay and general morphological assessment by ‘direct contact’ test.

Experimental

Materials and methods

Acrylic acid (AAc) of purity 99%, sodium persulphate (SPS) and 2-butanone were purchased from Qualigen fine chemicals, India. 2-(Dimethylamino)ethyl methacrylate (DMAEMA) of purity 98% was purchased from Aldrich (St. Louis, MO, U.S.A). Bovine serum albumin (BSA) (molecular weight 66 kDa, purity 96%) and Bradford reagent were purchased from Sigma Life Science (St. Louis, MO, U.S.A). N,N,N′,N′-Tetramethyl ethylenediamine (TEMED) was purchased from SRL Pvt. Ltd., India. All chemicals were used without further purification if not otherwise mentioned. Buffer solution of pH 1.2, 3.5, 5.0 and 7.0 were prepared according to Indian Pharmacopoeia, 1996, II. Water for all the reactions, solution preparation and polymer purification was triple deionized water.

Synthesis of PAD hydrogels

Five different hydrogels, varying in their monomer feed ratios, were synthesized by aqueous free-radical polymerization using AAc, DMAEMA, SPS and TEMED, as shown in Scheme 1. AAc and DMAEMA monomers were distilled prior to use. Briefly, each hydrogel was prepared by first mixing AAc and DMAEMA at ∼0 °C (ice-bath) in a two-necked round bottom flask for about 5 min. Distilled water, concentrated aqueous solution of SPS (0.025 mol% of monomers) and TEMED (0.3 mol% of monomers) were then added to the monomer mixture and mixed along with constant nitrogen purging for about 15 minutes.

Scheme 1 Synthesis of poly(AAc-co-DMAEMA) (PAD) hydrogels.

The reaction mixture was sonicated for 2 minutes to expel the entrapped air. The prepared reaction mixture was then poured in cylindrical poly(vinyl chloride) (PVC) moulds (internal diameter 3 mm and length 80 mm) and sealed with Teflon™ tape in a constant-temperature water bath, the reaction mixture was incubated for 24 h at 41 ± 1 °C. The synthesized PAD hydrogel was removed from the moulds after the reaction and cut into cylindrical pieces of ∼8 mm length. Unreacted water-soluble constituents were removed from the hydrogel pieces by immersing them in distilled water for five consecutive days. Water was replaced after every hour on the first day and twice a day for the remaining period. The gels were dried to a constant weight by drying in air for 24 h, followed by drying in vacuum oven for 5 h at 40 °C. Dried gels were stored in a vacuum desiccator until further used. Feed composition of all the PAD hydrogels are listed in ESI 1.† Homopolymers of AAc and DMAEMA (PAAc and poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA), respectively) were also synthesized using a similar method. After the reaction, PAAc was obtained as a cylindrical solid, whereas, PDMAEMA was precipitated from 2-butanone.

Fourier-transform infrared (FTIR) measurements

The FTIR measurements were done on an FTIR-ATR model Alpha-P, Bruker. Dry PAD hydrogels of each composition were dipped in liquid nitrogen and crushed in a mortar pestle to get fine powder. The powdered samples were stored in vacuum desiccators for further use. The IR spectrum of each hydrogel was obtained by scanning at a resolution of 4 cm⁻¹ and by averaging 24 scans.

Differential scanning calorimetry (DSC)

The DSC measurements were carried out on a Pyris 6 DSC instrument (PerkinElmer) containing refrigerator-cooling system. T_g is the temperature range where a polymer changes from a hard, rigid or glassy state to a more pliable, compliant or rubbery state. For determining the T_g, powdered PAD hydrogels were subjected to two consecutive heating cycles, under nitrogen atmosphere. About 8–10 mg of dry powdered sample was placed in a DSC aluminium pan and sealed thereafter. An empty-sealed aluminium pan was used as the reference cell. In the first heating cycle, the sample was heated from room temperature (30 °C) to 175 °C at 10 °C minute⁻¹, for removing its thermal history. T_g was obtained from the thermogram of the second heating cycle (30–230 °C, 10 °C min⁻¹), wherein the midpoint value of the special heat increment was assigned as T_g. Averages of three determinations were made for each hydrogel.

Swelling study

The swelling studies were carried out in 100 ml of pH 1.2, 3.5, 5.0 and 7.0 buffer solutions at 37 ± 1 °C. Ionic strength of all the buffer solutions was kept roughly constant by adding 1 M KCl salt. The initial weight (M₀) of dry PAD hydrogels (∼8 mm in length) was noted and finally submerged in respective buffer solutions. The gels were taken out of the buffer solution after every predetermined time period and their swollen weight (M_t) was noted. A moist filter paper was used to soak the surface water of the swollen gels before weighing.

The swelling study was performed until the gels reached their equilibrium weight. The swelling ratio (Q_t), at any particular time (t), was calculated by correlating the weight of water (M_t − M₀) within the hydrogel, at t, to the weight of dry hydrogel (M₀), as shown in eqn (1). All the experiments were performed in triplicate and the average value was used for the calculation of Q_t

The swelling kinetics of the hydrogels was investigated by using eqn (2):

where, F represents the fractional water uptake, M_∞ is the weight of the hydrogel at equilibrium swelling, K is related to diffusion constant (a constant characteristic of the hydrogel) and n is the diffusional exponent, that takes mode of transport of water into account. By taking natural log of eqn (2), we get:

ln(F) = ln(K) + n ln(t) (3)

The values of n and K were calculated from the slope and intercept, respectively, of the plot of ln(F) against ln(t). Values of t were to be expressed in seconds.^49,50

Morphology of PAD hydrogels

The surface morphology of swollen PAD hydrogels was observed by using a scanning electron microscope (SEM) ZEISS EVO 50. Briefly, samples of PAD hydrogel were swollen to equilibrium buffer solutions of pH 1.2 and 7.4 phosphate buffer solution (PBS). The swollen gels were then frozen in a −80 °C deep-freezer (Ilshin Lab Co., Ltd., Korea) and finally lyophilized in a freeze drier (Ilshin Lab Co., Ltd., Korea) for 24 h. The freeze-dried samples were stored in a vacuum desiccator until further use. The samples were fixed on an aluminium stub by means of carbon tape (double-sided adhesive and electrically conductive) and subjected to gold coating in a sputter coater (PolaronE 5100 Gold Sputter Coating unit) prior to SEM examinations. SEM images were taken at magnifications of 100× and 50,000× magnifications.

Mechanical properties

Dried cylindrical disc of approximately 8 mm height were taken from each composition (PAD70, PAD80 and PAD90). The dry weight (W_d) of the gel discs were measured and then they were dipped in 50 ml buffer solution of pH 7.0 (ionic strength adjusted by addition of 1 M potassium chloride salt). The samples were allowed to swell till equilibrium at room temperature (20 ± 1 °C). Universal Testing Machine (UTM) model H5KS (Tinius Olsen, England) with QMAT 5.37 professional software was used to carry out the compression experiments. All the experiments were performed at room temperature (20 ± 1 °C). Briefly, for each swollen sample the surface water was wiped off by a moist filter paper and its weight and dimensions (height and diameter) were recorded.

A Vernier caliper was used to measure the height and diameter of the gels. The gels were replaced into the buffer solution until the compression experiments, to avoid loss of water. For carrying out the compression experiments, each gel was placed between a pair of compression plate of 50 mm diameter after wiping off the surface water by a moist filter paper. A load cell of 1000 N was used to compress the samples up to the required strain. The compressive modulus (M) was measured by subjecting the gel to 25% deformation. During this measurement, the test speed was set at 1 mm minute⁻¹.

Compressive modulus was calculated from the slope of the initial linear part of the generated stress–strain curve, using the QMAT software. In a similar experiment, the sample was compressed till 90% deformation (test speed = 2 mm minute⁻¹) to find the compressive strength (σ_max) and deformation at failure (% D). The highest point in the graph, before failure, was used to calculate both σ_max and % D. All the tests were started within 1 minute of removing the gels from buffer solution to avoid water loss and surface hardening. All the experiments were carried out in triplicates.

Loading of BSA in PAD hydrogels and determination of loading efficiency

BSA was weighed accurately and dissolved in phosphate buffered saline (PBS) of pH 7.4 to make a 2 mg ml⁻¹ solution. A series of standard BSA solutions were prepared by serial dilution of the 2 mg ml⁻¹ solution. The absorbance (A) of the standards with different BSA concentration (C) was determined at 580 nm with MICRO SCANTM MS5608 microplate reader (ECIL, India) using Bradford method (96 well plate assay protocol, Sigma). A standard curve of BSA was drawn on the graph of A and C and the standard equation was obtained. Dry, cylindrical PAD hydrogels of ∼8 mm length were weighed accurately and dipped separately in 50 ml of 20 mg ml⁻¹ BSA solution in PBS (pH 7.4) at 4 °C for 8 days. After the soaking period, the swollen gels were rinsed carefully with PBS to wash off proteins adsorbed on the surface of the hydrogel. The gels were then dried till constant weight by first drying at room temperature (∼30 °C) for 24 h followed by vacuum drying at 40 °C for 5 h. The amount of BSA loaded in each gel was estimated by Bradford method from the difference of concentration (mg ml⁻¹) in the original and residual BSA solution.

In vitro release of BSA from PAD hydrogels

Dry PAD hydrogels loaded with BSA were dipped first in 25 ml of pH 1.2 buffer (simulated gastric fluid) for 3.5 h followed by dipping in 25 ml of PBS, pH 7.4 (simulated intestinal fluid). During the release study, the samples were shaken in an orbital incubator shaker at 100 rpm and 37 ± 1 °C. After every predetermined interval of time, 1 ml of aliquot was withdrawn and the amount of BSA released was determined by Bradford method. After every withdrawal, 1 ml of fresh medium, kept at 37 °C, was replaced to maintain the sink condition. The release study was carried out for 30 h (minimum retention time in the intestinal tract).

Cell culture

HeLa human cervical cancer cells (ATCC CCL-2) and McCoy mouse fibroblast cells (ATCC CRL 1696) were used for cytotoxicity test. The cells were routinely cultured and maintained in 25 cm² tissue culture flask (Corning) in Dulbecco's Modified Eagle's medium (DMEM) (Sigma) containing 10% fetal bovine serum (FBS) (Sigma), 10 μg ml⁻¹ ciprofloxacin, and 5% CO₂.

PAD hydrogel extract preparation

The cytotoxicity assay of the gels was carried out according to ISO 10993–5 (1992). The cylindrical gels, equilibrated in PBS of pH 7.4, were cut into uniform pieces. The samples were kept in PBS and then sterilized by autoclaving at 120 °C for 20 minutes. One millilitre of culture media was added to two pieces of each kind of the samples in a 24-well plate (Corning), and incubated at 37 °C for 48 h with 5% CO₂. The extract thus obtained was used for the cytotoxicity test.

Cytotoxicity assay

The hydrogel extracts (50 μl per well in quadruplicates) of each type were added for the preparation of 96-well tissue culture plates (Corning). The cells were trypsinized and resuspended in the culture media. A haemocytometer was used to determine the cell density. In all wells containing the hydrogel extract, 50 μl per well of cell suspension (∼2800 cells) was added. Culture media without cells was taken as blank and the culture media with cells (in quadruplicates) was taken as the negative control. After a period of 72 h, cytotoxicity was assessed using CellTiter 96VR Aqueous One Solution Reagent (Promega), as per the protocol provided by the manufacturer. The readings were taken using ELISA plate reader (Anthos HT 1) at 492 nm after 4 h.

Morphology of HeLa and McCoy cells in direct contact of the gels was examined after 72 h and compared with that of the cells in negative control.

The percentage survival was calculated as:

T-test was used to perform statistical analysis. A P-value of <0.05 was considered significant.

Results and discussion

FTIR measurement

The spectra of PAD50 copolymer displayed a broad absorption peak at ∼3371 cm⁻¹ (Fig. 1a), which may be attributed to the O–H stretching (of carboxylic acids) or due to moisture. However, presence of moisture in PAD samples is unlikely as the samples were thoroughly dried and kept in desiccator until further use. Therefore, it indicates the presence of PAAc units in the copolymer. The copolymer shows a single peak at 1721 cm⁻¹, which may be due to carbonyl stretch of PDMAEMA and not PAAc. This may be attributed to the conversion of PAAc into carboxylate anions because of the presence of tertiary amine groups of PDMAEMA. The carboxylate ions thus formed exhibited a strong symmetrical stretching band near 1650–1550 cm⁻¹.⁵¹ Hence, we believe that in PAD50 hydrogel all the carbonyl groups of PAAc are in the form of carboxylate anions (–COO⁻) and the peak at 1579 cm⁻¹ is a confirmatory proof for the same. This reaffirmed that the PAD copolymer contains both carboxylic acid (PAAc) and tertiary amine (PDMAEMA).

Fig. 1 FTIR-ATR spectra of (a) PAAC, PDMAEMA and PAD50 hydrogel, and (b) PAD50, PAD70, PAD80, PAD90 hydrogels.

A peak at 1579 cm⁻¹ of relatively small intensity was also seen in the spectra of PDMAEMA. This peak was unanticipated in PDMAEMA but its appearance may be presumed as a result of cleavage of the dimethyl amino ethyl group [–(CH₂)₂N(CH₃)₂] from the parent PDMAEMA, during purification of the homopolymers, resulting in the formation carboxylate anions. Fig. 1b shows that the peak intensity for the carboxylate anion at 1579 cm⁻¹ decreases from PAD50 to PAD90 copolymer. This may be due to the decrease in carboxylate anion formation with the decrease in proportion of PDMAEMA from PAD50 to PAD90.

DSC measurement

The thermal behaviour of a polymer is important in relation to its properties for controlling the release rate in order to have a suitable drug dosage form.⁵² The T_g of the copolymer gels as seen in Fig. 2a increases from PAD 70 to PAD 90. This trend was seen because of the increase in strongly polar carboxylic acid groups of PAAc. These strongly polar carboxylic acid groups can increase the T_g value due to the formation of internal hydrogen bonds between the polymer chains.⁵²

Fig. 2 (a) Variation of glass transition temperature (T_g) with hydrogel composition, (b) equilibrium swelling of different PAD hydrogels in buffered solution of various pH at 37 ± 1 °C; each point represents the mean ± SD.

Swelling behaviour

The PAD hydrogels displayed a pH-dependent swelling behaviour, as shown in Fig. 2b. At pH ∼3.5 (isoelectric point, pI) minimum swelling in the hydrogels was observed, which is a characteristic behaviour of polyampholytes.⁴⁷ At pI, the network is collapsed due to the formation of interpolymer complexes. As the pH is increased or decreased, the interpolymer complexes dissociate due to ionization of the pendant groups. This electrostatic repulsion causes the network to swell, and water is allowed to enter.⁵³ Swelling of PAD hydrogels at pH < pI is mainly due to the ionization (protonation) of the DMAEMA units [–N(CH_s)₂H⁺, pK_a ≈ 8], whereas, ionization (deprotonation) of the AAc units (–COO⁻, pK_a ≈ 4.5) at pH > pI accounts for the swelling at higher pH. Thus, increasing DMAEMA content increases the swelling ratio of PAD gels at lower pH, and increasing AAc content does the same at higher pH.

The study of swelling kinetics or swelling mechanism gives a good prediction of the nature of water's diffusion towards the inside of the gel and the ability of the hydrogel to be used as a drug delivery system. In highly swellable gels the initial phase of volume variation follows zero-order swelling kinetics; hence eqn (3) is applied to derive the mechanism of volume variation at the initial phase (when the density of the device does not vary) of any polymer-penetrant system whatever the temperature and the penetrant activity.^22,47,54 Eqn (3) is applicable for F ≤ 0.6 and the plots of ln(F) against ln(t) yield straight lines indicating up to almost 60% increase in the weight of the hydrogel. Eqn (2) is termed as the Ritger–Peppas model and has been used frequently in literature. It is important to note that the n is a parameter depending on the geometrical shape of the hydrogel matrix.⁵⁴ Cylinder-shaped samples were used in this study, this implies that the values of the parameter n have the following conceptual meanings: (i) n = 0.5 for Fickian diffusion (case I), here water diffusion is mainly controlled by chemical potential gradient, (ii) 0.5 < n < 1.0 for anomalous or non-Fickian transport, contribution of Fickian diffusion and controlled-relaxation, (iii) n = 1.0 for zero order (case II), the solutes migration occurs at constant speeds due to controlled-relaxation of the polymer chains, and (iv) n > 1.0 for super case II, contribution of the macromolecular relaxation of the polymer chains.^54–56

The ln(F)–ln(t) relationship during water diffusion into PAD hydrogels at pH 1.2 and 7.0 are shown in Fig. 3a and b, respectively. The swelling exponents (n) and swelling constants (K) were calculated from the slopes and intercepts of the curves, and are listed in ESI 2.† In pH 1.2, the n values were on the magnitude of order < 0.34. This indicated that water absorption mechanism of the hydrogels at pH 1.2 is governed by pseudo-Fickian diffusion only. The n values decreased from PAD70 to PAD90 in pH 1.2 due to decrease in PDMAEMA amount in the PAD matrices. Also in pH 7.0, the n values were on the magnitude of order n < 0.5, indicating that water absorption mechanism of the hydrogels is governed by pseudo-Fickian diffusion. A pseudo-Fickian diffusion behaviour is observed with the sorption curves resembling Fickian curves but with a slower approach to the equilibrium. Pseudo-Fickian diffusion is also known as “two-stage sorption”. Deviation from Fickian behaviour generally arises as a consequence of finite rates by which changes in the polymer structure occur in response to stresses during the sorption diffusion process. Possibly, the water permeating into our material occupies the free volume between the polymer chains. As a function of time, the volume fraction of water increases and the volume fraction of free space decreases.⁵⁷ However, a proper understanding of the thermodynamics of this process is required to give further insight into the water transport mechanism. An increase of n with an increase in pH of the surrounding liquid demonstrates that the water absorption profile becomes more dependent on the polymer relaxation at higher pH. This effect was attributed to the increased ionization of –COOH groups on PAAc segments of the hydrogel at pH 7. The unexpected slow swelling of PAD90 may be due to the formation of high molecular weight copolymer chains. During synthesis of the PAD matrices, gelation of the monomer mixture started very late for PAD90 compared to that of PAD70 and PAD80.

Fig. 3 (a) Fractional swelling of individual PAD gels against time in buffered solutions of pH 1.2 at 37 ± 1 °C, and (b) fractional swelling of individual PAD gels against time in buffered solutions of pH 7.0 at 37 ± 1 °C.

Morphology of PAD hydrogels

SEM micrographs of the swollen PAD hydrogels are shown in Fig. 4. Fig. 4a and c shows cross-section of PAD70, PAD80 and PAD90, respectively, after swelling in pH 1.2. In acidic pH the PAD hydrogels are relatively non-porous. Few pores formed are due to ionization of DMAEMA units as discussed earlier, in swelling behaviour. Fig. 4d–f indicates that the interior of the gels is macroporous at pH 7.4. The PAD70 hydrogel exhibited an average pore size of ∼100 μm which increased to ∼250 μm and ∼400 μm in case of PAD80 and PAD90 respectively. At pH 7.0 the swelling of PAD hydrogels is mainly due to the ionization (deprotonation) of the AAc units (–COO⁻). The PAAc chains dissociate due to the ionization of the pendant groups and this electrostatic repulsion causes the network to swell and water is allowed to enter.⁵³ The higher is the amount of water entrapped in between the polymer networks, the greater will be the pore size formed upon lyophilisation of the swollen gel. Thus, increasing AAc content increases the average pore size of the PAD hydrogels at pH 7.4.

Fig. 4 SEM micrographs of swollen PAD hydrogels: (a–c) cross-section of PAD70, PAD80 and PAD90 after swelling in pH 1.2 at 100× magnification; (d–f) cross-section of PAD70, PAD80 and PAD90 after swelling in PBS (pH 7.4) at 100× magnification; (g–i) external surface of PAD70, PAD80 and PAD90 after swelling in PBS (pH 7.4) at 100× magnification; (j–l) interlocked nanogels in PAD70, PAD80 and PAD90 after swelling in PBS (pH 7.4) at 50 000× magnification.

Morphology of the external surface of the lateral wall in cylindrical PAD hydrogels was unlike the interior [Fig. 4g–i]. The wall was nonporous and covered the entire gel like a continuous outer skin (with few holes which appeared as defects in it) whereas the macroporous interiors bore a resemblance to honeycomb type structure. Fig. 4j–l is the image of PAD70, PAD80 and PAD90 matrix swollen in PBS (pH 7.4) at 50 000× magnification.

All the PAD matrices are seen to be made-up of interlocked nanogels of ∼200 nm size. This finding can be a plausible explanation to the stability of the PAD hydrogels in pH 7.4 buffers.

Mechanical properties

Mechanical properties of hydrogels are very important for pharmaceutical applications. A drug delivery system designed to protect a sensitive therapeutic agent, such as protein, must maintain its integrity to be able to protect the protein until it is released out of the system.⁵⁸ Table 1 shows the values of compressive strength (σ_max), compressive modulus (M) and deformation at failure (%) for all the composition of PAD hydrogels. Both σ_max and M showed a decreasing trend with an increase in PAAc amount. The effect may be explained on the basis of difference in their equilibrium-swelling ratio (ESR), as shown in Fig. 5a. The ESR of PAD hydrogels in pH 7.0 buffers increases from PAD70 (2296%) to PAD90 (4027%) due to the increase in acrylic acid content (as discussed earlier). Higher acrylic acid content resulted in larger pore size and decreased pore density as observed through SEM analysis (Fig. 4). Pore size is an important factor governing mechanical strength, as in case of matrices with smaller pore size, the load gets distributed evenly throughout the surface. Well-connected small pores form a barrier and avert catastrophic crack propagation.⁵⁹ Fig. 5b shows the stress–strain graph for all the compositions. It can be observed from the graph that although the PAD90 hydrogel failed at lower stress, but it underwent highest deformation (85%) before failure. On the other hand, all other hydrogels deformed less than 80%. This can be explained on the basis of high swelling ability (high water content) and larger pore size of PAD90 hydrogel. The imbibed water molecules migrate from the regions under load towards the unloaded regions thus allowing extended deformation of the hydrogel as observed.⁵⁹ The shape of the gel is maintained by the strong bonds which serve as permanent crosslinks; whereas, other mechanical functions such as, fracture resistance by bond rupturing, is maintained by weaker bonds.⁶⁰ In order to be used for biomedical applications, it is important to have the ability to modify the structural and mechanical properties of the hydrogel in a controlled manner.⁶¹

Table 1 Mechanical properties of PAD hydrogels as a function of composition

Sample code	%Swelling (mass)	Compressive strength (kPa)	Deformation at failure (%)	Compressive modulus (kPa)
PAD70	2296 ± 210	205.8 ± 1.3	69.0 ± 1.4	10.8 ± 1.4
PAD80	313 ± 215	199.7 ± 1.5	76.5 ± 0.7	7.6 ± 0.9
PAD90	4027 ± 576	178.4 ± 14.6	85.0 ± 0.8	3.0 ± 0.6

---

Fig. 5 (a) Comparison of compressive modulus (M) and equilibrium swelling ratio (ESR) of PAD hydrogels, (b) stress–strain curves of different PAD gels equilibrated in buffer of pH 7.0.

Loading of BSA in PAD hydrogels

The loading of BSA (mg g⁻¹ dry gel) in the PAD hydrogels was appreciable. Loading was observed to be 273.5 ± 12.1 mg g⁻¹ dry gel for PAD70, 824.5 ± 17.5 mg g⁻¹ dry gel for PAD80 and 1226.7 ± 213.2 mg g⁻¹ dry gel for PAD90. Thus, the order of loading was PAD90 > PAD80 > PAD70. The equilibrium swelling ratio of PAD90 is highest, thereby implying that the swollen PAD90 encompasses highest amount of water inside its network, thus leading to higher loading of BSA (Fig. 6a). Consequently, the cumulative release (%) of BSA should be highest in PAD90 and it should be in descending order from PAD90 to PAD70. But surprisingly, the cumulative release (%) pattern is in the reverse order i.e. after 30 h the cumulative release (%) of BSA from PAD90 was the lowest and it followed the order PAD70 > PAD80 > PAD90 (Fig. 6a). The reason behind this trend is that the rate of swelling of PAD90 is slowest and that of PAD70 is fastest as explained in the section of swelling behaviour. After 30 h the swelling ratio of PAD90 is just 121% (i.e. ∼3% of its ESR) whereas for PAD80 it is 1010% (i.e. ∼32% of its ESR) and for PAD70 it is 1391% (i.e. ∼60% of its ESR).

Fig. 6 (a) Cumulative release in (mg g⁻¹) and (b) in (%) of PAD70, PAD80, PAD90.

In vitro BSA release

The cumulative release (%) of BSA in pH 1.2 is minimal i.e. less than 6% for all the PAD samples. However, in pH 7.4 it is much higher and varies from ∼24–73% with the change in gel composition. The amount of BSA (mg g⁻¹ dry gel) released can be tuned by changing the composition of the hydrogel (please see ESI 3†). The hydrodynamic diameter of BSA is 7.7 nm.⁶²

Cytotoxicity assay

Cytotoxicity of biomaterials is an important aspect for its future application. They should neither release toxic products nor produce adverse reactions. These can be evaluated through in vitro cytotoxic tests. The cytotoxicity of PAD with different monomer feed ratio towards HeLa and McCoy cell lines were determined by MTT and XTT assay, respectively. The absorbance of Formazan crystals at 560 nm has a direct proportional relationship with the number of living cells in a reasonable linear range.⁶³ In this study, the cell only group was used as a control. The results demonstrate that there are no significant differences among the PADs with different monomer feed ratios found in cytotoxicity against HeLa cells (Fig. 7). Similar results were also found in McCoy mouse fibroblast cell systems. These results gathered from both kinds of cell lines show that the PADs have no apparent cytotoxicity (Fig. 8).

Fig. 7 Confocal microscopy images of HeLa Human Cervical Cancer cells (ATCC CCl-2) cultured for 72 h with (a) PAD50, (b) PAD70, (c) PAD80, and (d) PAD90 after MTT assay, and (e) control (HeLa human cervical cancer cells (ATCC CCl-2) cultured for 72 h).

Fig. 8 Cytotoxicity of PAD hydrogels to HeLa human cervical cancer cell line and McCoy mouse fibroblast cell line, after incubation for 72 h. Cell viability was determined by MTT assay.

Conclusions

A series of stable, physically cross-linked hydrogels with tunable mechanical properties and pH-sensitive swellability were prepared by single step aqueous copolymerization of AAc and DMAEMA. The presence of two monomer units with their complexation in the membranes was ensured by FTIR spectroscopy. It has been found that monomer feed ratio, composition and water content in feed plays a crucial role in the development of stable PADs without any active cross-linking agents. Investigation into the morphology unveiled that the hydrogel matrices consist of unique nano to macro scale hierarchical patterns which elucidated the self-assembly behaviour of copolymers composed of two hydrophilic monomer units. Experimental results indicate that these interlocked nanogels act as stable building blocks in the monoliths. The swollen gels had appreciable compressive strength and remained rigid. These smart hydrogels were successful in their ability to house appreciable amount of BSA and sustainably release it in various buffer solutions mimicking the pH range of GIT. The degree of swelling as well as the pore size increases with AAc content in the PAD hydrogels. The PAD hydrogels have intrinsically low cytotoxicity to HeLa and McCoy fibroblast cell lines, envisioning its role in controlled drug delivery and prospective utilization in the biomedical industry.

Acknowledgements

This work was supported by Department of Science and Technology (DST), Government of India (SR/FT/CS-006/2010, SERB).

References

1. G. Talukder, The market for bioengineered protein drugs, in Market research reports and technical publications.bbcresearch, 2007, http://www.bccresearch.com/report/BIO009E.html, accessed 15 Aug 2010.
2. N. Rasool, T. Yasin, J. Y. Y. Heng and Z. Akhter, Polymer, 2010, 51, 1687.
3. Y. Dekel, Y. Glucksam and R. Margalit, J. Controlled Release, 2010, 143, 128.
4. L. Yin, J. Ding, J. Zhang, C. He and C. T. C. Yin, Biomaterials, 2010, 31, 3347.
5. K. Nakamura, R. J. Murray, J. I. Joseph, N. A. Peppas, M. Morishita and A. M. Lowman, J. Controlled Release, 2004, 95, 589.
6. D. Jain, A. K. Panda and D. K. Majumdar, AAPS PharmSciTech, 2005, 6, E100.
7. V. H. Lee, Oral route of peptide and drug delivery, in Peptide and Protein Drug Delivery, Marcel Dekker Inc., New York, 1991, ch. 16, pp. 691–738.
8. R. Langer and N. A. Peppas, AIChE J., 2003, 49, 2990.
9. R. Langer, Science, 1990, 249, 1527.
10. M. A. C. Stuart, W. T. S. Huck, J. Genzer, M. Muller, C. Ober, M. Stamm, G. B. Sukhorukov, I. Szleifer, V. V. Tsukruk, M. Urban, F. Winnik, S. Zauscher, I. Luzinov and S. Minko, Nat. Mater., 2010, 9, 101.
11. S. W. Song, K. Hidajat and S. Kawi, Chem. Commun., 2007, 4396.
12. S. Li and X. Liu, Polym. Adv. Technol., 2008, 19, 1536.
13. S. K. Bajpai and S. Singh, React. Funct. Polym., 2006, 66, 431.
14. F. J. Xu, E. T. Kang and K. G. Neoh, Biomaterials, 2006, 27, 2787.
15. B. Singh, G. S. Chauhan, S. Kumar and N. Chauhan, Carbohydr. Polym., 2007, 67, 190.
16. G. Chao, H. Deng, Q. Huang, W. Jia, W. Huang, Y. Gu, H. Tan, L. Fan, C. Liu, A. Huang, K. Lei, C. Gong, M. Tu and Z. Qian, J. Polym. Res., 2006, 13, 349.
17. G. T. Chao, Z. Y. Qian, M. J. Huang, B. Kan, Y. C. Gu, C. Y. Gong, J. L. Yang, K. Wang, M. Dai, X. Y. Li, M. L. Gou, M. J. Tu and Y. Q. Wei, J. Biomed. Mater. Res., Part A, 2008, 85, 36.
18. C. Liu, C. Gong, Y. Pan, Y. Zhang, J. Wang, M. Huang, Y. Wang, K. Wang, M. Gou, M. Tu, Y. Wei and Z. Qian, Colloids Surf., A, 2007, 302, 430.
19. R. Zhang, M. Tang, A. Bowyer, R. Eisenthal and J. Hubble, Biomaterials, 2005, 26, 4677.
20. D. Das, R. Das, P. Ghosh, S. Dhara, A. B. Panda and S. Pal, RSC Adv., 2013, 3, 25340.
21. W. E. Hennink and C. F. V. Nostrum, Adv. Drug Delivery Rev., 2012, 64, 223.
22. H. El-Sherif and M. El-Masry, Polym. Adv. Technol., 2008, 19, 342.
23. R. Sophie, J. Mies, C. Stefaan, F. Cornelus and E. Wim, Biomaterials, 2005, 26, 2129.
24. M. Zhai, Y. Chen, M. Yi and H. Ha, Polym. Int., 2004, 53, 33.
25. A. S. A. El-Hag, Biomaterials, 2005, 26, 2733.
26. A. Bin Imran, K. Esaki, H. Gotoh, T. Seki, K. Ito, Y. Sakai and Y. Takeoka, Nat. Commun., 2014, 5, 5124.
27. M. M. El-Masry, M. M. M. Elnashar and H. M. El-Sherif, J. Appl. Polym. Sci., 2007, 106, 3571.
28. A. Das, S. Ghosh and A. R. Ray, Polymer, 2011, 52, 3800.
29. S. Brahim, D. Narinesingh and A. Guiseppi-Elie, Biomacromolecules, 2003, 4, 1224.
30. G. H. Ma, M. Nagai and S. Omi, J. Appl. Polym. Sci., 2001, 79, 2408.
31. B. Yildiz, B. Isik, M. Kis and O. Birgul, J. Appl. Polym. Sci., 2003, 88, 2028.
32. M. Sen and M. Sari, Eur. Polym. J., 2005, 41, 1304.
33. C. S. Satish and H. G. Shivakumar, J. Macromol. Sci., Part A: Pure Appl.Chem., 2007, 44, 379.
34. P. van de Wetering, J. Y. Cherng, H. Talsma, D. J. H. Crommelin and W. E. Hennink, J. Controlled Release, 1998, 53, 145.
35. S. J. Lee, Y. Bae, K. Kataoka, D. Kim, D. S. Lee and S. C. Kim, Polym. J., 2007, 40, 171.
36. S. C. Alley, N. M. Okeley and P. D. Senter, Curr. Opin. Chem. Biol., 2010, 14, 529.
37. J. Chen, M. Shi, P. Liu, A. Ko, W. Zhong, W. Liao and M. Xing, Biomaterials, 2014, 35, 1240.
38. J. Chen, X. Qiu, J. Ouyang, J. Kong, W. Zhong and M. M. Q. Xing, Biomacromolecules, 2011, 12, 3601.
39. S. Liu, A. Ko, W. Li, W. Zhong and M. Xing, J. Mater. Chem. B, 2014, 2, 1125.
40. D. Oupický, A. L. Parker and L. W. Seymour, J. Am. Chem. Soc., 2001, 124, 8.
41. S. Dirk, Adv. Drug Delivery Rev., 2006, 58, 1655.
42. J. Chen, M. M. Q. Xing and W. Zhong, Polymer, 2011, 52, 933.
43. Y. Tian, A. Glogowska, W. Zhong, T. Klonisch and M. Xing, J. Mater. Chem. B, 2013, 1, 5264.
44. J. Chen, F. Zehtabi, J. Ouyang, J. M. Kong, W. Zhong and M. M. Q. Xing, J. Mater. Chem., 2012, 22, 7121.
45. J. Shi, J. Ouyang, Q. Li, L. Wang, J. Wu, W. Zhong and M. M. Q. Xing, J. Mater. Chem., 2012, 22, 23952.
46. M. I. El-Sherbiny, Carbohydr. Polym., 2010, 80, 1125.
47. A. Das, M. Mehndiratta, P. Chattopadhyay and A. R. Ray, J. Appl. Polym. Sci., 2010, 115, 393.
48. S. Sajeesh, C. Vauthier, C. Gueutin, G. Ponchel and C. P. Sharma, Acta Biomater., 2010, 6, 3072.
49. C. Yanfeng and Y. Min, Radiat. Phys. Chem., 2001, 61, 65.
50. P. D. Sawant, D. Luu, R. Ye and R. Buchta, Int. J. Pharm., 2010, 396, 45.
51. R. M. Silverstein, G. C. Bassler and T. C. Morrill, in Spectroscopic identification of organic compounds, John Wiley & Sons, Inc., 4th edn, 1981, p. 121.
52. M. Mahkam, R. Mohammadi, S. O. R. Siadat and S. E. Ranaei-siadat, E Polymer, 2006, 5, 1.
53. M. Torres-Lugo and N. A. Peppas, Macromolecules, 1999, 32, 6646.
54. N. Martínez-Vázquez, R. del C. Antonio-Cruz, A. Álvarez-Castillo, A. M. Mendoza-Martinez and A. B. Morales-Cepeda, Rev. Mex. Ing. Quim., 2007, 6, 337.
55. M. R. Guilherme, T. A Moia, A. V. Reis, A. T. Paulino, A. F. Rubira, L. H. C. Mattoso, E. C. Muniz and E. B. Tambourgi, Biomacromolecules, 2009, 10, 190.
56. P. L. Ritger and N. A. Peppas, J. Controlled Release, 1987, 5, 23.
57. M. M. Wind and H. J. W. Lenderink, Prog. Org. Coat., 1996, 28, 239.
58. N. A. Peppas, P. Buresa, W. Leobandunga and H. Ichikawa, Eur. J. Pharm. Biopharm., 2000, 50, 27.
59. B. B. Mandal, S. Kapoor and S. C. Kundu, Biomaterials, 2009, 30, 2826.
60. T. L. Sun, T. Kurokawa, S. Kuroda, A. B. Ihsan, T. Akasaki, K. Sato, M. A. Haque, T. Nakajima and J. P. Gong, Nat. Mater., 2014, 12, 932.
61. M. Malkoch, R. Vestberg, N. Gupta, L. Mespouille, P. Dubois, A. F. Mason, J. L. Hedrick, Q. Liao, C. W. Frank, K. Kingsbury and C. J. Hawker, Chem. Commun., 2006, 14, 2774.
62. D. Imren, M. Gumusderelioglu and A. Guner, J. Appl. Polym. Sci., 2006, 102, 4213.
63. J. A. Plumb, R. Milroy and S. B. Kaye, Cancer Res., 1989, 49, 4435.

---

Footnotes

† Electronic supplementary information (ESI) available: (1) Feed composition of PAD gels, (2) diffusion parameters during swelling of gels in buffered solutions, (3) in vitro BSA release behaviour of hydrogels. See DOI: 10.1039/c5ra07424j

‡ These authors contributed equally to this work.

---