A novel pH-responsive hydrogel based on natural polysaccharides for controlled release of protein drugs

Abstract

A biocompatible and biodegradable hydrogel based on a natural polysaccharide was prepared and characterized to evaluate its applicability as an effective carrier for controlled protein delivery. The hydrogel exhibited significant pH-sensitivity, which was favorable for protein release in a simulated intestinal medium. It is capable of incorporating considerable amounts of protein drugs (encapsulation efficiency up to 97.6 wt%) following a protein-friendly preparation procedure. It was shown that the hydrogel is able to release two entrapped model protein drugs (bovine serum albumin and lysozyme) in a controlled manner with full preservation of protein stability and enzymatic activity for lysozyme. Moreover, the insulin-loaded hydrogel was effective in reducing blood glucose level in diabetic animal models. Importantly, the hydrogel showed no evidence of cytotoxicity in vitro and in vivo, rather, it is biodegradable. The synthesized hydrogel shows favorable features as a promising delivery carrier candidate for targeted delivery of protein drugs to specific sites.

---

1. Introduction

Polysaccharides have gained increasing attention in drug delivery recently due to their favorable features. They are naturally abundant, biocompatible, biodegradable and inexpensive. Also, the other characteristics of polysaccharides, such as their high water-affinity and ease of chemical modification, facilitate the versatile applications of polysaccharide-based materials in the field of drug delivery.^1–11

Pachyman, a naturally occurring β-(1,3)-D-glucan, is the key component of the sclerotium of Poria cocos (P. cocos), one of the most important and time-honored herbal medicines in China and Japan.^12–15 Several polysaccharides have shown obvious anti-cancer activities, such as Panax ginseng and Ganoderma lucidum. However, pachyman fails to show any anti-tumor effect despite the great efforts made in developing pachyman as an antitumor agent. This has been attributed to its insolubility in water.^16–21 To date, pachyman has been mainly applied in nutrition supplements and cosmetics additives.^22–26 Further studies are still necessary to explore the possible applications of pachyman in drug therapy to take full advantage of this natural resource. Previously, our group has investigated the applications of pachyman in pharmaceuticals for the first time.^27,28 In this study, another pachyman-based hydrogel system with favourable properties was developed and evaluated for use in drug delivery.

Currently, protein drugs are becoming quite an important class of therapeutic agents along with the rapid development of biotechnology.^29–32 Therapeutic proteins usually have low stability, short circulation half-life and large molecular size. Therefore, suitable delivery systems are needed to guarantee the safe and controlled delivery of protein drugs.^32,33 Also, protein drugs are commonly administrated parenterally and repeated injections are required. To increase patient compliance and to minimize possible health hazards caused by frequent injections, oral delivery is an ideal alternative route for protein drug administration.^34–36

Among all the delivery systems for protein drugs, “smart” hydrogels have attracted considerable attention because of their perceived “intelligence”. Hydrogels can effectively protect protein drugs from degradation in the gastrointestinal (GI) tract.^29,37,38 Moreover, hydrogels are able to retain a significant amount of water to resemble natural living tissues. They are also inherently biocompatible, offering them good opportunities as protein delivery systems.^{29,33,39–41} Amongst these, sugar-based hydrogels are of particular interest as they show minimal toxicity compared to conventional hydrogels and can potentially serve as a targeting ligand in the human body.^4,42

In this study, we aim to explore the possible application of pachyman as a safe and effective drug carrier for the controlled release of protein drugs. The structure of the hydrogel was characterized by infrared spectroscopy and the morphology was observed under a scanning electron microscope (SEM). Bovine serum albumin (BSA) and lysozyme were used as two different model protein drugs to study the encapsulation efficiency and drug release pattern in simulated gastric and intestinal buffers. In addition, the pH-sensitivity, biodegradability and biocompatibility of the hydrogel were all examined. Finally, the performance of the hydrogel system to orally deliver active insulin in vivo was evaluated on model diabetic mice.

Compared to conventional hydrogel systems, the pachyman-based hydrogel showed obvious pH-sensitivity, which could protect the loaded protein drugs from degradation by the harsh environment of the stomach. Importantly, since a natural polysaccharide was used as the starting material, the hydrogel offers superior polyfunctionality, biocompatibility, biodegradability and low cost. The experimental results presented indicate the great potential of the pachyman-based hydrogel as a safe, pH-responsive and controllable protein drug delivery vehicle.

2. Materials and methods

2.1 Materials and animals

Fresh sclerotium of P. cocos was naturally planted in Luotian county and supplied by Hubei Hongyuan Medicine Company (Hubei, China). The BCA Protein Assay Kit was purchased from Pierce Biotechnology (USA). Bovine serum albumin (BSA), epichlorohydrin (ECH), Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS) and penicillin–streptomycin solution were purchased from HyClone (USA). Dimethyl sulfoxide (DMSO) (Sigma, USA) was distilled under vacuum before use. All other chemicals were of analytical or better grade and used as received from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA) unless otherwise noted.

BALB/c mice of both sexes weighing 20 ± 2 g were purchased from the Animal Centre of Wuhan University (China). Animal experiments were performed according to the Guiding Principles for the Care and Use of Laboratory Animals in Wuhan University and in compliance with the local ethics committee.

2.2 Hydrogel synthesis

Epichlorohydrin (ECH) crosslinked hydroxypropyl pachyman (ECH–HPP) hydrogel was prepared via a two-step reaction (Scheme 1). Firstly, a series of hydroxypropyl pachymans (HPPs) with different degrees of substitution (DS) were prepared according to our earlier report.²⁸ ECH–HPP was then synthesized by crosslinking HPP with epichlorohydrin (ECH) as the crosslinking agent. In order to get the desired ECH–HPP hydrogel with the proper swelling capacity for protein delivery, different amounts of ECH were added to optimize the crosslinking reaction. The influence of the DS on the water solubility of the products as well as the correlation between swelling ratio (SR) and crosslinking density (CLD) were investigated in detail by varying the reaction parameters, such as amounts of reactants, reaction time and reaction temperature.

Scheme 1 Synthetic procedure of the ECH–HPP polymer.

2.3 Characterization

The chemical structures of HPP and ECH–HPP were identified by infrared spectroscopy (IR, Spectrum One, Perkin Elmer, USA). The DS values of HPP were estimated based on the principles given by Jones and Riddick.⁴³ Size exclusion chromatography combined with multi-angle laser light scattering (SEC-LLS, Viscotek, USA) measurements was carried out to measure the molecular weight distribution of HPP.

Cross-linking densities (CLD) of the ECH–HPP hydrogels were measured by determining the modulus of elasticity in compression as reported previously.⁴⁴

2.4 SEM morphology investigation

The surface morphology of the hydrogel was determined using a scanning electron microscope (JEOL JSM-5600 LV, Japan). The hydrogel samples were first equilibrated in distilled water at room temperature, then quickly frozen in liquid nitrogen. Afterwards, the samples were freeze-dried under vacuum at −48 °C for 3 days until all the solvent was sublimed.

2.5 Swelling characterization

The swelling profiles of the hydrogels with different crosslinking densities were determined by both one-step and two-step swelling characterization. For one-step swelling characterization, the swelling ratio (SR) was measured gravimetrically by immersing the gels in two different buffer solutions (pH 1.2 HCl–KCl buffer or pH 7.4 phosphate buffer) in sealed containers at 37 °C, respectively, simulating gastrointestinal (GI) tract environments.⁴⁵ Afterwards, the immersed gel was periodically removed from the buffer solution, blotted with filter paper to wipe off the excess surface water, weighed and returned to the same container until swelling equilibrium was observed. This procedure was performed in triplicate at all of the time points. The SR was calculated by the dynamic weight change of the hydrogels in terms of time by the following eqn (1):where W_s is the weight of the gel in the equilibrium-swollen state and W_d is the initial weight of the dried test sample.

Also, a two-step swelling characterization of the hydrogel formed by 1.75% ECH was performed by immersing the dried gel in 5 mL of a solution at pH 1.2 (simulated gastric fluid) for 2 h and subsequently transferring it into a solution of pH 7.4 (simulating intestinal tract medium) for an additional 14 h. The temperature was kept at 37 °C throughout the measurement. At specific time intervals, samples were removed, blotted and their swelling ratios (SR) were calculated by the aforementioned eqn (1).

2.6 Biocompatibility test

In vitro cytotoxicity and in vivo acute oral toxicity tests were conducted to evaluate the biocompatibility of the hydrogel.

The in vitro cytotoxicity of the hydrogel was tested by MTT assay against HUH 7, Hacat and L02 cell lines and the cell morphology was recorded by microscopic examination. In brief, cells (1 × 10⁴) were seeded in 96-well plates and incubated overnight in DMEM. The media was replaced with fresh media containing ECH–HPP polymer solutions with different concentrations (0.01, 0.1, 1, 5, 10, 50, 100 mg mL⁻¹) and incubated for 24 h in a humidified atmosphere of 5% CO₂ at 37 °C. Cells grown in pure culture medium served as a 100% cell viability control. Afterwards, cells were washed twice with PBS and 0.2 mL of MTT solution was added to each well. The plate was further incubated for 4 h to allow the metabolism of MTT. Later, 150 μL of DMSO was added to each well and the plate was incubated for 30 min at room temperature. Finally, the cytotoxic activity of the polymer was quantitatively analyzed using the MTT assay to measure the metabolic reduction of 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide to formazan by viable cells.

Acute oral toxicity was carried out to test the in vivo biocompatibility of the hydrogel. All animal studies were conducted under a protocol approved by the Wuhan University Animal Care and Use Committee. Twenty mice of both sexes were fasted overnight with free access to water prior to the experiments. The mice were orally injected with the ECH–HPP hydrogel suspension twice (4 h intervals) at a dose of 7.5 g kg⁻¹ body weight. So, the total dose given to each animal was up to 15 g kg⁻¹ b.w. The same volume of 0.9% aqueous NaCl was injected orally into the negative control mice. Subsequently, all animals were observed for 14 days and the mortality and their general conditions including the energy, activity, hair, feces, behavior pattern, and other clinical signs were recorded. On the 14th day, all animals were sacrificed and gross histological examinations of the major organs were performed. Finally, the acute toxic classifications of the ECH–HPP hydrogel were evaluated according to China’s criteria of acute classification (Ministry of Health, People’s Republic of China).

2.7 Hydrogel degradation

The in vitro biodegradability of the ECH–HPP hydrogel was determined by both chemical hydrolysis and enzymatic hydrolysis. Chemical hydrolysis of the hydrogel was performed in a pH 7.4 phosphate buffer solution. In particular, exactly weighed aliquots of the hydrogel (30 mg) were dispersed in the hydrolysis medium and incubated at 37 °C under continuous orbital shaking (100 rpm). At predetermined time intervals, the gels were separated out from the release medium, washed with DI water and gently blotted. Then, they were freeze-dried for 48 h and weighed again.

Enzymatic hydrolysis was conducted by incubating aliquots of the hydrogel for 24 h under continuous stirring (100 rpm) at 37 °C in PBS (pH 7.4) containing dextranase (final enzyme concentration 15 U mL⁻¹) and esterase (final enzyme concentration 125 U mL⁻¹). The enzyme solutions were prepared just before use. The activities of dextranase and esterase were confirmed by performing the assays reported in the literature.⁴⁶ After the treatment, the hydrogel was treated with the same procedure used for the samples recovered after the chemical hydrolysis. Each experiment was performed in triplicate at all of the time points for both chemical hydrolysis and enzymatic hydrolysis.

2.8 Drug loading and release studies

The drug loading and release profiles from the ECH–HPP hydrogels were studied in simulated GI medium by employing BSA and lysozyme as two model protein drugs. First, the hydrogels were swelled in BSA or lysozyme solution with a final concentration of 1 wt%. After reaching swelling equilibrium, the gels were taken out, dried and reweighed. For confirmation of the percentage of drug loaded in the hydrogels, the amount of protein solution left in the loading medium was determined by the Micro-BCA assay as described by the manufacturer. The entrapment efficiency (ee) was defined as the percentage of protein mass loaded in the hydrogel compared to the originally provided protein mass, by means of the following eqn (2):

The protein release from the test hydrogels were evaluated by both one-step and two-step releasing studies. For the one-step releasing study, the dried, drug loaded gel was immersed in solutions of pH 1.2 (simulated gastric fluid), pH 6.8 (simulated intestinal fluid) or pH 7.4 (simulated colonic fluid). For the two-step releasing study, the gel was first immersed in a pH 1.2 solution at 37 °C for 2 h and subsequently transferred into a solution of pH 7.4 for an additional 8 h. Specifically, 20 mL of the solutions were taken in a conical flask and were placed in a rotary water bath shaker at 100 rpm at physiological temperature (37 °C). Samples were withdrawn at regular intervals and the release of BSA was calculated by the Micro-BCA assay. With each sampling, the solution was replenished with fresh medium which was pre-equilibrated at 37 °C, keeping the total volume constant. This procedure was performed in triplicate at all of the time points studied. The percentage cumulative amount of the released protein drugs (BSA or lysozyme) was calculated and plotted against time.

2.9 Stability of the released BSA

The stability of the released BSA was determined by analyzing the conformation of the BSA using a Jasco J-810 spectropolarimeter (Jasco, Japan).⁴⁷ In addition, SDS-polyacrylamide gel electrophoresis (PAGE) was conducted to study the structural integrity of BSA in the supernatant.

2.10 Determination of the enzymatic activity of lysozyme

In order to test whether the activity of the lysozyme loaded in the hydrogel had been kept after release, the enzymatic activity of the lysozyme before and after release from the hydrogel was measured according to a literature reported method (these are referred to as “initial lysozyme” and “lysozyme released from hydrogel”, respectively).⁴⁸ Briefly, 200 μL of sample solution containing the released lysozyme was added to 2.0 mL of phosphate buffer solution (pH 7.4) containing 0.6 mg of Micrococcus lysodeikticus suspension as the substrate. A decrease in suspension turbidity which was attributed to the degradation of the substrate was recorded with a spectrophotometer in terms of the incubation time. The enzymatic activity was calculated from the initial slope of the course of turbidity vs. time. The activity of lysozyme which had been kept dissolved in PBS (pH 7.4) at 37 °C for various periods of time was also measured for comparison (referred to as “lysozyme free in solution”).

2.11 In vivo pharmacological response of insulin-loaded hydrogel

A diabetic animal model was created by injecting streptozotocin (STZ) intraperitoneally in male BALB/c mice. After 3 days of injections, diabetic values were checked. The diabetic mice were fasted overnight prior to treatment and remained fasting for another 12 h during the experiment, but were allowed water ad libitum. Mice were divided into four groups each containing a minimum of 6 mice. The first two groups of diabetic animals were treated with insulin-loaded ECH–HPP hydrogel orally at two different doses of insulin (50 and 100 IU kg⁻¹ b.w., respectively). The next two groups of diabetic mice were used for the comparison study, and were treated with either subcutaneous injection of insulin solution at a dose of 5.0 IU kg⁻¹ b.w. or oral gavage of insulin solution at a dose of 100 IU kg⁻¹ b.w. Blood samples from the tail vein of the mice were collected in a microcentrifuge tube at specific times after administration and the blood glucose level (BGL) was checked by a glucose meter (Roche).

2.12 Statistical analysis

All the experiments were repeated at least three times and the data were presented as means ± standard deviation. Statistical analysis was performed in Origin 7.0. In all the tests, the statistical significance was set at p < 0.05.

3. Results and discussion

3.1 Synthesis of ECH–HPP

Chemical modification of a natural polysaccharide gives the opportunity to combine the typical biocompatibility of natural polymers with the advantages of synthetic polymers such as chemical versatility, better mechanical properties, etc. Among all of the chemical modifications, chemical crosslinking is a commonly used approach for producing hydrogels. In this study, ECH–HPP hydrogel was prepared by HPP crosslinking with epichlorohydrin (ECH) in a basic medium (Scheme 1). ECH was employed as a convenient base-catalyzed cross-linking agent. It could be claimed that each ECH molecule could react with two hydroxyl groups from different HPP molecules, yielding one hydroxyl group, in this manner, two neighboring polysaccharides chains were attached together to form a network.

First, the hydroxypropyl group substitution reaction of pachyman was conducted based on a typical etherification of polysaccharides.⁴⁹ Comparing the FT-IR spectrum of the ECH–HPP to that of pachyman (Fig. 1) shows that the FT-IR spectrum of pachyman exhibited bands at 890 cm⁻¹, 1260 cm⁻¹ and 3400 cm⁻¹ which could be attributed to the saccharide structure. These bands did not change when HPP was cross-linked by ECH. In addition, the spectrum of ECH–HPP exhibited a relatively low intensity –OH vibration band at 3400 cm⁻¹ and much higher intensity –CH₃ or –CH₂– vibration bands at 2891 cm⁻¹, due to the introduction of isopropyl groups. Moreover, stronger peaks from 1030 cm⁻¹ to 1110 cm⁻¹ appeared, corresponding to the increase in –C–O–C– bands and –CHOH– groups after crosslinking.

Fig. 1 FT-IR spectra of HPP and ECH–HPP.

It is important to assess the reaction conditions for the etherification of the polysaccharides because the degree of substitution (DS) of the polysaccharides plays a crucial role on the physical properties such as the molecular weight (M_w), solubility and swelling properties of the products. As shown in Table 1, optimization of the hydroxypropylation of pachyman for obtaining HPP with good water solubility and moderate molecular weight was achieved by varying the reaction conditions including the amounts of reactants, reaction time and reaction temperature. Table 2 reveals the correlations between the DS and the solubility of HPP. Overall, it can be concluded that the solubility of HPP improves gradually with the increase of the DS, mainly due to the introduction of hydrophilic hydroxypropyl groups into the macromolecule chain of pachyman. When the DS reached 0.62, the product was able to dissolve in water and gave a clear transparent solution, with a M_w of 181 000 and a low molecular polydispersity (M_w/M_n = 1.12).

Table 1 Optimized reaction conditions for the preparation of HPP

Pachyman : propylene oxide ratio (g mL^−1)	Pachyman : water ratio (g mL^−1)	Quantity of NaOH (g)	Reaction temperature (°C)	Reaction time (h)	DS^a	Molar mass distributions^b		Solubility^c
						Mw (g mol^−1)	Mw : Mn	Water	DMSO
a DS means the degree of substitution of hydroxypropyl groups to monosaccharide residue of pachyman.b Determined by SEC-LLS chromatograms.c Determined by naked eyes.d +: soluble.
1 : 8	1 : 12	12	40	8	0.62	181 000	1.12	+^d	+

---

Table 2 Correlation between the DS and the water solubility of HPP

DS of HPP^a	Solubility^b^,^c
a Data are expressed as mean ± S.D. of four experiments.b Data were determined by the naked eye.c −: insoluble in water; ±: swelling in water; +: soluble in water.
0.1053 ± 0.0003	−
0.1356 ± 0.0022	±
0.3412 ± 0.0019	±
0.4672 ± 0.0031	±
0.4966 ± 0.0040	±
0.5190 ± 0.0032	±
0.5708 ± 0.0028	±
0.6233 ± 0.0031	+
0.6275 ± 0.0021	+
0.6502 ± 0.0029	+

---

The major drawback restricting the application of pachyman is the water insolubility. The reaction process optimized in this study for obtaining water soluble pachyman derivatives may be of great significance in extending the applications of pachyman.

3.2 Hydrogel characterization

It is well known that the swelling ability of hydrogels plays an important role in regulating many of their properties, such as permeability to hydrophilic or hydrophobic drugs, rates of enzymatic or hydrolytic degradation, and mechanical properties.⁵⁰ Thus, the swelling behavior of the hydrogel was studied in different media at pH 1.2 and pH 7.4 mimicking the gastric and intestinal liquids, respectively.

As shown in Fig. 2A and B, in the beginning, the swelling ratio was very small at the ECH amounts of 0.4% and 0.5% (vol%), probably because a relatively loose crosslinking network was formed. Later, the swelling ratio (SR) generally showed a tendency to increase with increasing amounts of ECH. However, when the amount of ECH increased to 1.75 vol%, a decrease in the SR of the hydrogel occurred. Although this result is consistent with an earlier report on composite hydrogels based on xanthan and poly(vinyl alcohol),⁵¹ it contradicts the literature data which predicted a decrease in the swelling degree of a hydrogel when raising the quantity of the cross-linking agent. This phenomenon is probably caused by the destruction of the helicoidal structure of pachyman in the basic medium and under heated conditions during the cross-linking reaction. As a result, a portion of the hydroxyl groups used to stabilize the helicoidal structure by hydrogen bonds were involved in the cross-linking reaction while the rest of the hydroxyl groups were able to interact with water in the swelling process. Thus, compared to the initial helicoidal structure of pachyman, a less compact network with greater net-blanks formed, which allowed the absorption of more “unbound” water. However, when the cross-linking density increases to a higher extent, swelling becomes restrained as a more compact network was constructed, which led to the reduction of the free volume entered into the network and the expandability and penetrability of the hydrogel network was restricted.

Fig. 2 Swelling profiles of the ECH–HPP hydrogel: (A) one-step swelling characteristics of hydrogels with different amounts of ECH at pH 1.2; (B) one-step swelling characteristics of hydrogels with different amounts of ECH at pH 7.4 and (C) two-step swelling characteristics of the hydrogel at pH 1.2 for 2 h and subsequently at pH 7.4 for 14 h.

Comparison of the SEM images of the HPP and ECH–HPP hydrogel (Fig. 3B and C) also revealed a continuous and porous structure of the ECH–HPP hydrogel, which suggests the formation of a polymer network. The porous nature of the hydrogel makes it less dense and provides more surface area, where the capillary forces help the diffusion of water into the hydrogel. Upon the hydroxypropylation and the crosslinking reaction, it can be clearly seen from Fig. 3A that the starting material, pachyman which was neither water-soluble nor swellable (tube 1) was first changed to a transparent water-soluble solution of HPP (tube 2), and finally turned into a highly swellable ECH–HPP hydrogel (tube 3).

Fig. 3 The surface morphology and photographs of the ECH–HPP hydrogel: (A) photograph of the pachyman in water (tube 1), HPP in water (tube 2) and ECH–HPP hydrogel (tube 3); (B) SEM image of HPP and (C) SEM image of the hydrogel.

From the swelling ratio studies with different ECH and HPP combinations (Fig. 2A and B), crosslinking with 1.75 vol% of ECH was found to be the most favourable amount for crosslinking, as the ECH–HPP hydrogel synthesized with 1.75 vol% of ECH showed a suitable SR at both pH 1.2 and pH 7.4 (∼2.5 at pH 1.2 and ∼8.6 at pH 7.4). However, the SR of the ECH–HPP hydrogels of all other groups were not desirable for protein drug delivery since some of them showed too much swelling at both pH 1.2 and 7.4 while other groups showed too little swelling at pH 7.4. The rapid swelling of the hydrogel is not favorable for controlled drug release due to intestinal denaturation by proteolytic enzymes. Besides, with the increasing SR of the hydrogel, gel disintegration occurred at pH 7.4.

Moreover, the entrapment efficiency study (Table 3) also confirmed 1.75 vol% ECH to be the best, as it showed the highest entrapment efficiency (97.6%). Thus, the hydrogel formed by 1.75 vol% of ECH was selected for further pH-sensitivity and protein release studies.

Table 3 Percentage entrapment efficiency of hydrogels with different ECH amounts

ECH (%)	Solubility characteristics^a	Cross-linking density^b (mol m^−3)	% Entrapment efficiency (ee)
a Determined by the naked eye.b Data are expressed as mean ± S.D. of four experiments.
0.3	Soluble	—	—
0.4	Partially soluble	5.6 ± 0.3	10.6
0.5	Partially soluble	8.5 ± 1.2	12.3
0.75	Nonsoluble and swollen	9.6 ± 2.2	28.4
1.0	Nonsoluble and swollen	10.2 ± 2.1	47.9
1.25	Nonsoluble and swollen	12.6 ± 3.6	59.8
1.5	Nonsoluble and swollen	15.3 ± 3.2	70.2
1.75	Nonsoluble and swollen	19.6 ± 1.8	97.6
2.0	Nonsoluble and swollen	20.3 ± 2.7	53.7
2.5	Nonsoluble and swollen	19.2 ± 1.9	20.2

---

3.3 pH-sensitivity

Generally, the swelling performance of a hydrogel in gastric and intestinal fluids determines its relevance for drug release. The tested hydrogel presented similar swelling patterns between one-step swelling profiles (Fig. 2A and B) and two-step (Fig. 2C) swelling profiles in simulated physiological environments. As expected, the ECH–HPP hydrogel showed obvious differences in swelling that were dominated by external pH changes. Specifically, the swelling ratio of the hydrogel was very low at pH 1.2 (gastric environment) while the swelling ratio was moderate at pH 7.4 (intestinal environment). The swelling ratio of the hydrogel was low at pH 1.2, probably because of the formation of intermolecular hydrogen bonds. However, at pH 7.4, the hydroxy groups on the hydrogel were likely to be progressively ionized (–ONa). Therefore, the swelling of the hydrogel was probably highly increased due to the large swelling force created by the electrostatic repulsion between the ionized groups. Thus, the ECH–HPP hydrogel could protect the loaded protein drug from degradation by the low pH environment of the stomach.

3.4 Biocompatibility of the hydrogel

One of the key factors for the application of polymeric vectors in drug delivery is the absence of cytotoxicity. The MTT assay has proven to be a sensitive and reliable method for evaluating the biocompatibility of polymers.⁵²

In this study, an in vitro MTT assay with the HUH 7 cell line, Hacat cell line and human normal hepatocyte cell line L02 (Fig. 4) was performed to test the biocompatibility of the ECH–HPP hydrogel. As compared to the negative control, for all concentrations studied the viabilities of all three cell lines were almost 100% and were dose dependent. Moreover, as shown by the cell morphological analyses in Fig. 4A–F, the cells attached and spread regularly compared to the negative control, there were no morphological alterations such as retraction or detachment of the cell monolayer. Thus, it can be deduced that the ECH–HPP hydrogel was well tolerated by cells.

Fig. 4 In vitro biocompatibility test of the ECH–HPP hydrogel. Microphotographs of HUH7 cells, Hacat cells and L02 cells after three days in culture on ECH–HPP hydrogel: (A) the control HUH7 cells; (B) HUH7 cells cultured on the hydrogel; (C) the control Hacat cells; (D) Hacat cells cultured on the hydrogel; (E) the control L02 cells; (F) L02 cells cultured on the hydrogel and (G) in vitro cytotoxicity of the ECH–HPP as a function of polymer concentration towards HUH7, Hacat and L02 cell lines, respectively.

Furthermore, the in vivo biocompatibility test also confirmed that oral administration of mice with ECH–HPP hydrogel up to 15 g kg⁻¹ b.w. had no toxic effect. There was no death and no toxic response in all the mice studied during the 14 day study period. The animals displayed full energy, normal behavior, and free movement. No macroscopic pathological alterations were found in all mice at necropsy. Since the highest tested dose (15 g kg⁻¹ b.w.) caused no mortality, it could be concluded that the maximal tolerance dose (MTD) of ECH–HPP hydrogel was higher than 15 g kg⁻¹ b.w. in BALB/c mice.

Thereby, the synthesized ECH–HPP hydrogel presented excellent biocompatibility, which was expected since the starting material, pachyman, is a biocompatible natural polysaccharide. It could also be confirmed that all the residual reagents and cross-linkers which might lead to the cytotoxicity of the products have been removed completely after the reaction.

3.5 Hydrogel biodegradation

The ability of synthetic biomaterials to degrade in biological environments is a crucial property for a wealth of applications.⁵³ The presence of potentially degradable linkages in the structure of ECH–HPP hydrogels, such as glucoside and ether, confer a possible biodegradability to these systems. In order to evaluate whether degradation could occur, chemical and enzymatic degradation studies were performed in simulated physiological conditions. Fig. 5 shows that the chemical hydrolysis of the hydrogel was negligible since only a small dry weight was lost. However, approximately 15% of the hydrogel was degraded after 24 h of incubation with dextranase and esterase. Moreover, the degradation rate of the hydrogel relied on the swelling ratio (SR). It can be deduced from Fig. 5 that the degradation rate decreased with decreasing SR. The reason that the hydrogels with a higher swelling ratio had a faster degradation rate is probably due to the formation of a less compact network when the swelling ratio of the hydrogel is high. It is speculated that after an initial surface erosion of the hydrogel, enzymes diffused into the polymeric network and then more internal sites were degraded.

Fig. 5 Weight loss of the ECH–HPP hydrogel as a function of immersion time in PBS (pH = 7.4) at 37 °C: (A) chemical hydrolysis and (B) enzymatic hydrolysis.

3.6 In vitro drug loading and release profile

One of the most challenging obstacles in developing oral protein formulations is their inactivation or rapid enzymatic degradation in the harsh environment of the stomach. Thus, pH-sensitive drug delivery systems are considered a superior strategy in oral protein delivery as they can protect the protein drugs from proteolytic degradation in the stomach.^54,55 From the drug release profile in a simulated GI tract shown in Fig. 6, it can be deduced that the protein release from the ECH–HPP hydrogel was responsive to external pH changes under the simulated GI conditions. Lower degree of swelling and slower protein release was observed under low pH conditions (pH 1.2, gastric environment); however, under high pH conditions (pH 7.4, intestinal environment), the hydrogels swelled rapidly and showed a quick release of BSA from the hydrogels.

Fig. 6 Drug release profile of the ECH–HPP hydrogel: (A) one-step BSA release profile of the hydrogel at pH 1.2, pH 6.8 or pH 7.4; (B) two-step BSA release profile of the hydrogel at pH 1.2 for 2 h and subsequently at pH 7.4 for 8 h and (C) two-step lysozyme release profile of the hydrogel at pH 1.2 for 2 h and subsequently at pH 7.4 for 8 h.

Specifically, at pH 1.2, only about 12% of the encapsulated BSA was released from the test hydrogel in the one-step release study, however, at both pH 6.8 and pH 7.4 the protein release was significantly higher (89.3% and 96.8%, respectively). So, at the pH of both the duodenum and the ileum, the protein was released in a controlled manner over a period of 10 h. As for the two-step release profile (Fig. 6B), protein release at pH 1.2 was similar to that obtained in the one-step study. But when the hydrogel was transferred to simulated intestinal fluid (SIF, pH 7.4), the protein release was found to be much higher (about 89%) and a sustained and prolonged release profile was observed. Thus, the ECH–HPP hydrogel was found to be effective in releasing protein in a sustained fashion in the intestine, while protecting it from the harsh stomach environment. The favorable BSA release performance from the ECH–HPP hydrogel could be attributed to the pH-sensitivity of the hydrogel. It could be deduced that the protein release from the hydrogel matrix is mainly due to the diffusion of the drug through the pores of the swollen matrix at intestinal pH. Since the hydrogel is highly porous at pH 7.4, capillary forces could facilitate the diffusion of solvents into the hydrogel.

It was predicted that the release mechanism of BSA was caused first by diffusion and further accelerated by the weight loss of the gel. However, this hypothesis cannot be fully established in the early stage of the drug release when considering the weight loss profile (Fig. 5) and the drug release kinetics (Fig. 6). In the first 6 h of the two-step BSA release study, about 57% of the BSA was released from the hydrogel, but only a negligible amount of the gel matrix (∼5%) was lost. This result suggests that during the first stage of the release, the main driving force for BSA release was not gel degradation but diffusion, which might be attributed to the large amount of water entering the gel. After the first 6 h, the weight loss of the hydrogel increased to 15% and the swollen gels began to be degraded from the outside and then from the interior. Therefore, it can be presumed that after the first stage of the drug release, the drug molecules were released from the swollen hydrogel on account of both erosion of the polymeric matrix and diffusion.

Moreover, the ECH–HPP hydrogel could also release the other model drug, lysozyme, in a controlled release manner and the enzymatic activity of the loaded lysozyme was fully preserved (Fig. 6C and Table 4). The two-step release profile of lysozyme shown in Fig. 6C reveals that the release pattern of lysozyme was similar to that of BSA. Only about 7.8% of the lysozyme was released at pH 1.2, however, the amount of lysozyme released increased significantly to 65% at pH 7.4. This favorable controlled release performance of lysozyme could be also attributed to the pH-sensitivity of the hydrogel.

Table 4 Enzymatic activity of the released and free lysozyme

Lysozyme treatment	Remaining activity^a (units per mg lysozyme)
a Data are expressed as mean ± S.D. of three experiments.
Initial lysozyme	890 ± 90
Released from hydrogel	906 ± 39
Lysozyme free in solution	910 ± 69

---

3.7 BSA structural integrity

Exposure of BSA to the ionic solution or cross-linking agents could lead to possible detrimental effects on its protein structure and stability, such as protein denaturation, aggregation, hydrolysis, and reaction with the cross-linking agents. Therefore, the influence of the hydrogel environment on the integrity of the encapsulated BSA was investigated using both SDS-PAGE and circular dichroism (CD) spectroscopy. Comparing the BSA released from the ECH–HPP hydrogel and the BSA standard in solution (i.e., a BSA standard), the CD spectra (Fig. 7A) show that there was a slight change in the conformation of the loaded BSA when released at pH 1.2. This was probably due to the poor stability of proteins at pH 1.2.⁵⁶ However, no significant conformation change was observed for the BSA released from the two-step release study (at pH 1.2 for 2 h first and then at pH 7.4 for 8 h). This result suggests that the secondary structure of the loaded BSA was preserved during the processes of loading, prolonged contact of BSA with the ECH–HPP hydrogel, and release into a buffer solution at pH 1.2 for 2 h and subsequently at pH 7.4 for 8 h.

Fig. 7 The stability test of the released BSA: (A) CD spectra of standard BSA, BSA released at pH 1.2 and BSA released at pH 7.4 (at pH 1.2 for 2 h and then at pH 7.4 for 8 h) and (B) Coomassie-stained SDS-PAGE gel of the released BSA at pH 7.4. Lanes 1, 2 and 3 are, respectively, the molecular weight markers, BSA standard, and BSA released from the ECH–HPP hydrogel at pH 7.4.

Moreover, the SDS-PAGE results of the BSA released from the two-step release study also revealed that the integrity of the released protein was well maintained as the released BSA solution had a clear band present at 66 kDa which is consistent with the commercial standard BSA (Fig. 7B). Besides, there were no bands representing fragments of lower molecular weights detected, suggesting that the released BSA did not undergo hydrolysis.

Therefore, the ECH–HPP hydrogel has potential as a safe carrier for the delivery of protein drugs because it could fully maintain the stability of the protein. This is mainly attributed to the benign environment for the protein encapsulation process and the shielding effect on proteins susceptible to proteolytic attack by pH and enzymes.

3.8 In vivo evaluation on diabetic animals

In vivo oral hypoglycemic effects of the insulin-loaded ECH–HPP hydrogel were evaluated in diabetic mice in comparison to the control groups, i.e. diabetic mice treated with subcutaneous injection of free insulin solution at a dose of 5.0 IU kg⁻¹ b.w. or oral injection of free insulin solution at a dose of 100 IU kg⁻¹ b.w. As shown in Fig. 8, no significant reduction in glycemia was observed with the group treated with oral injection of free insulin solution at a dose of 100 IU kg⁻¹ b.w., which was probably due to the enzymatic degradation of the insulin in the GI tract. However, the diabetic mice treated with the insulin-loaded ECH–HPP hydrogel showed an obvious reduction in glycemia in a dose dependent manner (50 IU kg⁻¹ b.w. and 100 IU kg⁻¹ b.w.). Notably, the insulin-loaded ECH–HPP hydrogel at a higher dose (100 IU kg⁻¹ b.w.) could reduce blood glucose level (BGL) to 6 mM after oral administration and this was sustained for at least up to 9 h. In addition, subcutaneous administration of the free insulin solution at a dose of 5 IU kg⁻¹ b.w. also decreased the BGL significantly within 30 min, but a rapid increase in BGL was observed 2 h after the injection. Thus, the synthesized ECH–HPP hydrogel could be considered as a suitable system for oral insulin delivery since it could protect insulin from degradation in the GI tract and reduced glycemia in a sustained manner in a diabetic model through peroral delivery.

Fig. 8 In vivo hypoglycemic effects of the ECH–HPP hydrogel.

4. Conclusions

Protein drug delivery systems which allow proteins to retain their integrity and biological activity are one of the key components in the development of therapeutic proteins and peptides. Polysaccharides have great structural diversity and are inexpensive, naturally abundant, nontoxic and easy to chemically modify. These superior properties make polysaccharides quite promising base structures for drug carriers. The present study aims to develop a new polysaccharide-based hydrogel for safe and effective protein delivery. The synthesized ECH–HPP hydrogel offers the following advantages and has been shown to be quite promising as a drug carrier for oral protein drug delivery: (1) biodegradability and biocompatibility; (2) pH-sensitivity which provides desirable protective effects on protein drugs in the low pH environment of the stomach; (3) efficient loading and controlled release of the entrapped proteins without any loss of their stability and biological activity; (4) protection of insulin from degradation by the GI tract and prolonged hypoglycemic effects in a diabetic model with peroral delivery and (5) low expense, and ease of production and structural modification.

Pachyman is the key component of many herbal medicines in China and Japan. As a polysaccharide, it exhibits many favorable features as a drug carrier. However, its application in the pharmaceutical area has not been explored extensively. The present study has provided some initial effort in assessing the application of pachyman as a protein drug carrier. More work in investigating the use of pachyman in a nano-system in anticancer drug delivery is currently ongoing (unpublished results).

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21204069), the Research Fund for the Doctoral Program of Higher Education (20100141120013), the Natural Science Foundation of Hubei Province (2012FFB04429), the Chen-Guang Foundation of Scientific and Technologic Council of Wuhan, Hubei Province (2014070404010200), National Mega Project on Major Drug Development (2011ZX09401-302) and the Scientific and Technological innovative Research Team of Wuhan (2013070204020048).

Notes and references

1. X. Wang, Y. Du, J. Luo, B. Lin and J. F. Kennedy, Carbohydr. Polym., 2007, 69, 41–49.
2. Q. Wang, X. Hu, Y. Du and J. F. Kennedy, Carbohydr. Polym., 2010, 82, 842–847.
3. J. Wang, C. Guo, T. Yue, Y. Yuan, X. Liu and J. F. Kennedy, Carbohydr. Polym., 2012, 88, 966–972.
4. Y. Tang, X. Wang, Y. Li, M. Lei, Y. Du, J. F. Kennedy and C. J. Knill, Carbohydr. Polym., 2010, 82, 833–841.
5. V. Rana, P. Rai, A. K. Tiwary, R. S. Singh, J. F. Kennedy and C. J. Knill, Carbohydr. Polym., 2011, 83, 1031–1047.
6. Z. Yu, R. M. Schmaltz, T. C. Bozeman, R. Paul, M. J. Rishel, K. S. Tsosie and S. M. Hecht, J. Am. Chem. Soc., 2013, 135, 2883–2886.
7. C. Bhattacharya, Z. Yu, M. J. Rishel and S. M. Hecht, Biochemistry, 2014, 53, 3264–3266.
8. U. Sridhar, P. S. Pramod and M. Jayakannan, RSC Adv., 2013, 3, 21237–21241.
9. A. R. Fajardo, M. B. Silva, L. C. Lopes, J. F. Piai, A. F. Rubira and E. C. Muniz, RSC Adv., 2012, 2, 11095–11103.
10. D. Das, R. Das, P. Ghosh, S. Dhara, A. B. Panda and S. Pal, RSC Adv., 2013, 3, 25340–25350.
11. P. L. Lam, K. K. H. Lee, Y. W. Ho, R. S. M. Wong, S. W. Tong, C. H. Cheng, K. H. Lam, J. C. O. Tang, Z. X. Bian, R. Gambari, S. H. L. Kok and C. H. Chui, RSC Adv., 2014, 4, 14109–14114.
12. X. Chen, X. Xu, L. Zhang and J. F. Kennedy, Carbohydr. Polym., 2009, 75, 586–591.
13. G. Chihara, J. Hamuro, Y. Maeda, Y. Arai and F. Fukuoka, Nature, 1970, 225, 943–944.
14. Y. Jin, L. Zhang, Y. Tao, C. Zeng, Y. Chen and P. C. K. Cheung, Carbohydr. Polym., 2004, 57, 205–209.
15. Q. Huang, Y. Jin, L. Zhang, P. C. K. Cheung and J. F. Kennedy, Carbohydr. Polym., 2007, 70, 324–333.
16. Q. Huang and L. Zhang, Carbohydr. Polym., 2011, 83, 1363–1369.
17. X. Chen, X. Xu, L. Zhang and F. Zeng, Carbohydr. Polym., 2009, 78, 581–587.
18. Q. Huang, L. Zhang, P. C. K. Cheung and X. Tan, Carbohydr. Polym., 2006, 64, 337–344.
19. Y. Wang, L. Zhang, Y. Li, X. Hou and F. Zeng, Carbohydr. Res., 2004, 339, 2567–2574.
20. Y. Wang and L. Zhang, Carbohydr. Polym., 2006, 65, 504–509.
21. X. Chen, L. Zhang and P. C. K. Cheung, Int. Immunopharmacol., 2010, 10, 398–405.
22. Y. Zhao, Q. Ma, S. Huang, H. Dai and H. Wang, China Pat., 102 764 340, 2012.
23. L. Guo, China Pat., 102 764 407, 2012.
24. J. Wang, China Pat., 102 766 561, 2012.
25. J. W. Choi, S. K. Jung and K. T. Moon, Korea Pat., 200 003 3289, 2000.
26. G. Chai, China Pat., 187 9796, 2006.
27. Y. Xiao, W. Xu, Q. Zhu, B. Yan, D. Yang, J. Yang, X. He, S. Liang and X. Hu, Carbohydr. Polym., 2009, 77, 612–620.
28. Y. Xiao, S. Liang, G. Qiu, J. Wu, J. Zhang and X. Hu, Polym. Adv. Technol., 2007, 18, 268–274.
29. L. Chen, Z. Tian and Y. Du, Biomaterials, 2004, 25, 3725–3732.
30. L. Yang, J. S. Chu and J. A. Fix, Int. J. Pharm., 2002, 235, 1–15.
31. G. H. Gao, M. J. Park, Y. Li, G. H. Im, J.-H. Kim, H. N. Kim, J. W. Lee, P. Jeon, O. Y. Bang, J. H. Lee and D. S. Lee, Biomaterials, 2012, 33, 9157–9164.
32. C. E. Bobst, S. Wang, W.-C. Shen and I. A. Kaltashov, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 13544–13548.
33. T. Vermonden, R. Censi and W. E. Hennink, Chem. Rev., 2012, 112, 2853–2888.
34. D. S. Pisal, M. P. Kosloski and S. V. Balu-Iyer, J. Pharm. Sci., 2010, 99, 2557–2575.
35. X.-W. Shi, Y.-M. Du, L.-P. Sun, J.-H. Yang, X.-H. Wang and X.-L. Su, Macromol. Biosci., 2005, 5, 881–889.
36. S. Shu, X. Zhang, Z. Wu, Z. Wang and C. Li, Biomaterials, 2010, 31, 6039–6049.
37. E. Moreno, J. Schwartz, E. Larrañeta, P. A. Nguewa, C. Sanmartín, M. Agüeros, J. M. Irache and S. Espuelas, Int. J. Pharm., 2014, 459, 1–9.
38. M. George and T. E. Abraham, Int. J. Pharm., 2007, 335, 123–129.
39. J. Shi, Y. Gao, Y. Zhang, Y. Pan and B. Xu, Langmuir, 2011, 27, 14425–14431.
40. Y. Pan, Y. Gao, J. F. Shi, L. Wang and B. Xu, J. Mater. Chem., 2011, 21, 6804–6806.
41. Y. P. J. Shi, Y. Gao and B. Xu, MRS Proceedings, 2011, 1418.
42. K. Van der Heyden, K. Babooram, M. Ahmed and R. Narain, Eur. Polym. J., 2009, 45, 1689–1697.
43. L. R. Jones and J. A. Riddick, Anal. Chem., 1957, 29, 1214–1216.
44. K. Ulbrich, K. Dušek, M. Ilavský and J. Kopeček, Eur. Polym. J., 1978, 14, 45–49.
45. M. V. Risbud, A. A. Hardikar, S. V. Bhat and R. R. Bhonde, J. Controlled Release, 2000, 68, 23–30.
46. A. J. Adler and G. B. Kistiakowsky, J. Am. Chem. Soc., 1962, 84, 695–703.
47. C. Ramkissoon-Ganorkar, F. Liu, M. Baudyš and S. W. Kim, J. Controlled Release, 1999, 59, 287–298.
48. K. Burczak, T. Fujisato, M. Hatada and Y. Ikada, Biomaterials, 1994, 15, 231–238.
49. C. P. Iovine and D. K. Ray-Chaudhuri, US Pat., 412 9722, 1978.
50. G. Pitarresi, F. S. Palumbo, G. Giammona, M. A. Casadei and F. Micheletti Moracci, Biomaterials, 2003, 24, 4301–4313.
51. I. C. Alupei, M. Popa, M. Hamcerencu and M. J. M. Abadie, Eur. Polym. J., 2002, 38, 2313–2320.
52. Y. Tang and J. Singh, Int. J. Pharm., 2009, 365, 34–43.
53. M. P. Lutolf and J. A. Hubbell, Nat. Biotechnol., 2005, 23, 47–55.
54. J. Yang, J. Chen, D. Pan, Y. Wan and Z. Wang, Carbohydr. Polym., 2013, 92, 719–725.
55. M. George and T. E. Abraham, J. Controlled Release, 2006, 114, 1–14.
56. N. A. Peppas, Y. Huang, M. Torres-Lugo, J. H. Ward and J. Zhang, Annu. Rev. Biomed. Eng., 2000, 2, 9–29.

---