Poly(ethylene glycol) grafted chitosan as new copolymer material for oral delivery of insulin

In disease medical treatment, oral route is always preferred to injection. However, it is not always possible to deliver every therapeutic agent through the oral route including vaccines, genes, proteins and peptides [1]. A common example is insulin, a peptide hormone that has to be daily administered by injection, normally before meals, to treat diabetes. If administered through gastrointestinal (GI) pathway, it will be chemically and enzymatically inactivated by the high acidity and the secreted enzymes of the stomach. Study showed that less than 0.1% of any oral insulin dose reaches the bloodstream intact [2]. Developing an efficient delivery agent that can protect insulin through the GI tract remains a persistent challenge, hence a race between big pharmaceutical companies. Chemical modification and use of mucoadhesive polymeric system, for example chitosan derivatives, seem to be promising methods [3].

Chitosan (CS) is deacetylation product from chitin, the most abundant natural polysaccharide after cellulose found in crustaceous shells. This material has clear advantages of being natural origin is that it possesses good biocompatibility and very good mucoadhesive properties [4]. For these characteristics, CS is being investigated for many applications, especially in medicine, cosmetics and nutrition [5]. One difficulty that usually comes across in exploiting CS is its insolubility in most organic solvents [6]. This makes CS a tough substance to engineer in, for example, chemical modifications or nanoparticles preparations, but at the same time gives a very interesting challenge.

Among ideas for chemically modifying CS structure, pegylation emerged to be very competent method to obtain higher soluble capacity of CS. Grafting of methoxy poly(ethylene glycol) amine (PEG) molecule onto CS backbone resulted in improving water solubility of CS [7, 8], thermal stability [9] and could be used as pH dependent drug release system for protein delivery [10–12]. In two separate studies, insulin loaded PEG-g-CS nanoparticles had 150–400 nm in size, good encapsulation capacity but unsustained release of active agents were obtained [13, 14]. Another study showed that pegylated CS could reduce cytotoxicity of the nanocapsules hence increased cellular viability of the caco-2 cells [15].

In this study the purchased CS was first deacetylated by alkali method. Studies showed that degree of deacetylation (DD) affect CS's physiochemical properties in such a way that high DD possessed more favorable behaviors for drug delivery [16, 17]. Pegylation of CS was performed in three steps: (1) N-phthaloylation protection, (2) O-pegylation and (3) N-phthaloylated CS deprotection (figure 1). During synthesis, solubility of the reagents was carefully examined and optimised since it was the key challenge that decided the success and yield of the outputs. The synthesis products were throughly monitored by Fourier transform infrared spectroscopy (FTIR) and nuclear magnetic resonance (NMR) analysis to validate their structures as well as their purity after each of the synthesis steps. Insulin loaded nanoparticles (NPs) were prepared by ionic-gelation method using tripolyphosphate (TPP) as CS cross linker and characterized by transmission electron microscopy (TEM), dialysis and UV–vis methods.

Zoom In Zoom Out Reset image size

2.1. Materials

2.2. Deacetylation of chitosan

Chitosan (CS) powder was refluxed at 120 °C in 50% KOH solution of 1:20 (w/v) ratio under nitrogen atmosphere for 6 h. After the reaction, the slurries was filtered and washed with distilled water until pH neutral then dried at 40 °C and stored under vacuum. Deacetylation degree (DD) of CS was measured by reserve acid-base titration and FTIR methods as fully explained by Renata et al [18] expressed by following formulae

$$\begin{eqnarray}&&DD(\%)=2.03\frac{{{V}_{HCl}}-{{V}_{NaOH}}}{0.0042\left( {{V}_{HCl}}-{{V}_{NaOH}} \right)+{{m}_{CS}}},\end{eqnarray} \tag{ 1 }$$

$$\begin{eqnarray}&&DD(\%)=100-115\frac{{{A}_{1655}}}{{{A}_{3450}}},\end{eqnarray} \tag{ 2 }$$

where m_CS is the weight in g of CS, V stands for volume in ml of (0.1 N) solutions used for titration, A₁₆₅₅ and A₃₄₅₀ are values of FTIR absorbance from baseline.

2.3. Grafting of PEG onto chitosan

2.3.1. N-phthaloylation of chitosan

Deacetylated CS (5.00 g, DD = 90% so 25.0 mmol of glucosamine units) and PhA (12.1 g, 81.8 mmol) in DMF:H₂O = 95:5 (100 ml) was refluxed at 120 °C under nitrogen atmosphere. After 5–6 h of reaction, the cooled mixture was precipitated in ice-water, collected by filtration, washed overnight with MeOH, and air dried to obtain a pale tan powdery material N-phthaloylchitosan (PhACS) (yield: 81%). IR (KBr): ν 3450 (O–H), 2902 (C–H, pyranose), 1773 and 1716 (C=O, imide), 1394 (C=C, PhA), 1150–1010 (C–O, pyranose), and 721 cm⁻¹ (arom, phthaloyl) (figure 2).

Zoom In Zoom Out Reset image size

2.3.2. O-pegylation of chitosan

PhACS (1 g, 3.19 mmol) was dispersed in 30 ml of DCM:DMSO 50:50. The mixture was cooled to 0 °C then TsCl (1.5 eq) and Et₃N (1.5 eq) were gradually added. The reaction was kept at 0 °C for 1 h then 15 °C for 5 h and room temperature overnight. The mixture was then precipitated in ice-water, filtered, washed with MeOH. The powder was then redispersed in DCM together with PEG and the mixture was gently stirred at room temperature for 5 days. The slurries were then washed with EtOH and air dried. IR (KBr): ν 3440 (O–H), 2905 (C–H, pyranose), 1654 (amide I, N-PEG), 703 cm⁻¹.

2.3.3. N-phthaloylated chitosan deprotection

PEG-g-PhACS from previous (3.0 mmol of PhACS units), NaBH₄ (5 eq) were dissolved in 30 ml of 2-propanol:H₂O 6:1 and stirred overnight. Acetic acid (18 eq) was then added and the mixture was air closed and heated to 80 °C in 2 h. After the reaction, the slurries was filtered and washed with EtOH, air dried to obtain the final product PEG-g-CS dried powder. IR (KBr): 3432 (broad, OH, NH and NH₂), 2900 (C–H, pyranose), 1100 cm⁻¹ (C–O, pyranose). ¹H NMR: δ 5–5.2 (1H, br, H-1 of GlcN), 3.6–3.9 (m, H-3, H-4, H-5, H-6, H-6', N–CH_2b– of N-alkylated PEG), 3.1–3.4 (singlet of –OCH2– of PEG), 3.5 (–OCH3), 3.4–3.5 (br s, H-2).

2.4. Preparation and characterization of PEG-g-CS nanoparticles loaded with insulin

3.1. Deacetylation of chitosan

Deacetylation degree (DD) was measured before and after deacetylation (DA) reaction and by both methods to determine the effectiveness of the deacetylation process. The idea behind using two different methods was that to increase the precise of results while DD determination by NMR method was temporally unavailable. After deacetylation, DD was increased and reached over 90% (table 1).

Table 1.  Degrees of deacetylation (n = 3).

Chitosan	Titration (%)	FTIR(%)
Before DA	80.1 ± 4.0	75.1 ± 3.1
After DA	90.0 ± 3.4	92.2 ± 2.9

Figure 3 shows absorbance values interpretation at 3450 and 1655 cm⁻¹ for the DD calculations. It can be seen that after deacetylation reaction, the band at 1655 cm⁻¹ (C=O, acetyl groups) decreases as compared to the band at 3450 cm⁻¹ (O–H groups). Following the formula (2), this would resulted in increasing deacetylation degree value which was confirmed in table 1 results.

Zoom In Zoom Out Reset image size

3.2. Grafting of PEG onto chitosan

Chemical modifications of CS were carried out under heterogeneous conditions. Special care must be paid to various problems during synthesis such as the limit choice of reaction, low level of reagents' solubility, control of regioselective substitution, purification of the products. Besides, because CS is thermo instable, all heat assist reactions had to be taken place under nitrogen atmosphere or air sealed to avoid CS molecules degradation.

In figure 4 we see that the appearance of stronger IR absorptions at 2882 (CH₂) and 1171 cm⁻¹ (C–O–C) showed the existence of PEG chains. All of the obtained PEG-g-CS was soluble in DMSO as a sign of solubility improving of grafting copolymers over the origin CS.

Zoom In Zoom Out Reset image size

The final purified product PEG-g-CS was analyzed by ¹H NMR (figure 5). Peaks spectrum in the range of 3.0–4.0 ppm were multiples and not well separated due to the overlapping of peak of PEG methylene group and peaks of saccharine units of CS. Methyl group of PEG was clearly observed at 3.3 ppm.

Zoom In Zoom Out Reset image size

3.3. Nanoparticles preparation

Nanoparticles preparation with CS and PEG-g-CS were realized using TPP as crosslinking agent. Results in TEM visualization (figure 6) showed that NPs prepared from CS were more gel-like. They were also thermo instable and hard to centrifuge. NPs prepared from PEG-g-CS were larger in size but much more rigid and stable. Loading tests showed less insulin was encapsulated by PEG-g-CS (5.2% compared to 60% of CS NPs).

Zoom In Zoom Out Reset image size

PEG-g-CS copolymers were prepared through a new protection–graft–deprotection route with phthaloylchitosan as an intermediate. PEG branches were grafted regioselectively at the hydroxyl groups of phthaloylchitosan, and then phthaloyl groups were deprotected to regenerate the free amino groups. The graft copolymers obtained presented improved thermo stability and less gel-like.

However, more investigation was needed to carry out to improve the synthesis yield and productivities and the insulin loading capacity.

This research was funded by Vietnam National University Ho Chi Minh City (VNU-HCM) under grant number C2013-32-02 and was realized at the Laboratory for Nanotechnology (LNT).