Chapter 5. Drug release behavior of chitosan–montmorillonite
5.5 Repeating electrical-stimulation on release behavior
nanostructural enlargement will increase the passage of ionic species and accelerate the drug release (and the rate) from the nanohydrogel, resulting in a pseudo-zero-order release under electric-stimulation. On the contrary, higher cross-linking density as a result of higher MMT addition retards the diffusion of the drug molecules across the nanohydrogel.
An interesting finding revealed that the effect of MMT increment on the diffusion exponent n is very similar to that on the voltage-induced swelling ratio.
As observed in Figure 5-4(b), both parameters show exactly the same dependence of the MMT concentration. Under applied voltage of 5V, the diffusion exponent n exhibits a linear correlation with the voltage-induced swelling ratio with a R2 as high as 0.99, indicating that the degree of cross-linking in the nanohydrogels due to the incorporation of MMT profoundly affects the drug diffusion behavior. This provides powerful evidence that electric-stimuli release of drug is strongly determined by the free volume in the matrix as the diffusion passage of molecules.
5.5 Repeating electrical-stimulation on release behavior
A change in the electrical field has been realized to cause a change in drug release profile. It is therefore to use a switch of the electrical field from
“on” to “off” or vise versa to create pulsatile patterns of drug release. Figure 5-5 shows the resulting variation in the release profile of pure CS where a pulsatile release appeared immediately right after an electric field was applied and followed by little or negligible release of drug when the applying electric field was removed, i.e., switch “off”. After several on-off switching operations, pure CS (solid curve) exhibited a decrease of responsiveness to the stimulation and cumulative release amount, compared to the ideal case (which is able to
maintain identical release rate as first-time release behavior after repeated electric-stimulation, i.e., dotted curve) which is believed due to fast swelling of the polymeric matrix to the stimulus.
0 20 40 60 80 100
0 5 10 15 20
Time (min.) Cum u la ti ve r el ease amount ( m g)
Real release curve Ideal release curve
Figure 5-5 Real (…) and the ideal (—) pulsatile drug release profiles of pure CS hydrogels as an electric field was switched “on” and
“off”.
56
In order to understand the mechanism that led to the decrease of electric-stimuli responsivity after several on-off switching operations, the standard release rates (Rsd) as defined in previous section of the nanohydrogels with different CMMT upon stimulations is plotted against the numbers of on-off switching cycles, which is shown in Figure 5-6(a). Under an applied voltage, a noticeable shrinkage of CS at the cathode occurred, presumably due to bulk solvent flow towards the anode. The H+ ions formed by the electrolysis of water are subjected to intra- and intermolecular electro-repulsive forces arising from their interactions with adjacent H+ ions and the positively-charged CS network. These electro-repulsive interactions will lead to an osmotic pressure gradient, which facilitates CS syneresis and
57
ultimately results in gel shrinkage. Water together with the dissolved drug was expelled out of the hydrogel due to this contractile deformation. Moreover, the voltage-induced collapse after on-off switching operations leads to a decrease of ionizable groups (i.e., lower charged density) on pure CS, resulting in a decrease of the amount of the ions transported through the gel. As a result, for pure CS, it exhibited a substantial decrease of release rate with increasing number of switching operations. In order to improve the anti-fatigue property of pure CS, the incorporation of MMT into the CS matrix, i.e., the nanohydrogel, effectively maintained the same capability of deswelling-swelling behavior after several cyclic switching operations, compared the pure CS hydrogel [12].
For the nanohydrogels with 1 wt % MMT addition, it was observed that it has greater pulsatile release rate in the beginning stage (i.e., first two cycles) of operation but after several on-off cycles (until ten cycles), its release rate becomes smaller than that of pure CS. As mentioned above, the nanohydrogels with lower MMT addition (1 wt %) would deteriorate the crystallinity of CS and increase the passage of charge, which enhanced the release rate of model drug. Also, this deterioration of crystallinity weakened the strength of the nanohydrogels, which reduced the electric-stimuli responsiveness (release rate) after cyclic on-off operations as well. However, with higher MMT concentrations (exceed 1 wt % addition), stronger cross-linking in the nanohydrogels effectively improved the on-off cyclic electric-stimuli responsiveness, as shown in Fig. 5-6(a). The release rate under electric stimulation is reduced because of the decrease of the molecular mobility in the nanohydrogels with addition of MMT. Hence, it also suggested that loose structure of the nanohydrogels (i.e., pure CS and the nanohydrogels with 1 wt % MMT concentration) can hardly control the release upon
electric-stimulation protocol in an effective manner.
Standard Release rate [(M i-1-M i)/(M i-1]/t (10-2 /min)
On-Off Cyclic number
Standard Release rate [(M i-1-M i)/(M i-1]/t (10-2 /min)
MMT content (wt %)
A n ti -fa tig u e c o ef ficie n t
Figure 5-6 (a) Standard release rates of drug from the nanohydrogels (CS, CS-MMT1, CS-MMT2, and CS-MMT4) after cyclic on-off switching operations. (b) Dependences of the loaded MMT contents on the drug standard release rate in the first on-off cycle (filled boxes) and anti-fatigue coefficient of the nanohydrogels (empty circles).
58
However, this electrically-stimuli control can be effectively improved in the nanohydrogels with higher density of crosslink, i.e., higher MMT concentrations. Both initial release rate (R1) and relative anti-fatigue dependence of MMT addition can be further plotted in Figure 5-6(b) in which anti-fatigue coefficient was expressed with R10/R1; R1 and R10 indicate the standard release rate under on-off cycle number of 1 and 10, respectively. A higher R10/R1 ratio means that the nanohydrogels displays a better anti-fatigue property. Therefore, the optimal range of MMT contents, typically around 2 wt
%, in the nanohydrogels is expected where the resulting nanostructure of the nanohydrogels can be well manipulated with optimal cross-linking density to keep the pulsatile release profile relatively constant even after numerous on-off switching operations. Figure 5-7 shows a typical comparison of repeated on-off operation for pure CS and CS-MMT2 compositions, the anti-fatigue property of the CS can be largely improved with the incorporation of the MMT nanoplates.
2 4 6 8 10
0.0 0.5 1.0
Standard Release rate [(M i-1-M i)/(M i-1]/t (10-2 /min)
OFF ON
On-Off Cyclic number
pure CS CS-MMT2
Figure 5-7 Standard release rates of drug from pure CS and CS-MMT2 under cyclic on-off switching of electric-stimuli
59
60
Such an improvement in the anti-fatigue behavior can be explicitly observable from a reversible mechanical action of the nanohydrogels upon cyclic switching operation under a given electrical stimulation. This constantly reversible action of the nanohydrogels also ensures a constant rate of pulsatile release of the drug, as experimentally evidenced, and this allows a more reliable performance of a consecutive pulsatile-type drug release profile to be achieved from such nanohydrogels.
61
Chapter 6
Electrical-Sensitive Nanoparticle Composed of Chitosan and TEOS for Controlled Drug Release
6.1 Introduction
It is one of the most challenging works to develop new multi-functional materials that possess smart functionality capable of monitoring the changes of environment, followed by a corresponding actuation according to the stimuli.
A number of different polymer hydrogels that are able to mechanically respond to stimuli, which has been named as smart hydrogel and has been used as drug delivery vehicles. Environmental stimuli such as pH, pressure, temperature, light, magnetic fields and electric fields, would cause those smart polymer gels to undergo macroscopic deformation and produce contractile force. Numerous smart or responsive hydrogels have been studied with regard to their applications in biomedical fields, e.g. controlled drug delivery systems [68], muscle-like actuators [66], and bio-separation [67]. One most significant drawback of such smart hydrogels is their slow stimuli-to-response activity.
Thus, hydorgels with fast responsivity to its environmental stimuli are practically desirable, and a simple and straightforward method to enhance its responsivity is to make such responsive hydrogels with thinner and smaller dimensions.
In recent years, chitosan (CS) nanoparticles (NPs) have been widely studied for the controlled release of drugs included antibiotics, catalysts, proteins, and peptides [69, 90, 91]. Chitosan is a biodegradable, biocompatible, and non-toxic polycationic polymer with low immunogenicity. It has been
62
extensively investigated for formulating carrier and delivery systems for therapeutic macrosolutes, particularly genes and protein molecules because positively charged chitosan can be easily combined with negatively charged DNAs and proteins [92, 93]. To date, there have been a variety of reports in the preparation of CS NPs by ionic cross-linking. Alonso et al. [94] reported the use of an ionic gelation method to prepare CS NPs. In this method, positively charged CS NPs were formed through the inter- and intra-cross-linking of the amino groups of CS with the negatively charged phosphate groups of tripolyposphate (TPP). They demonstrated the encapsulation of bovine serum albumin (BSA) by the CS NPs with maximum loading capacity (LC) of 51 wt %.
Zhang et al. [95] reported the preparation of CS NPs in the size range of 90-200 nm achieved by combining post-deacetylation and fractionation of commercially available CS. They showed that the BSA could be loaded into the CS NPs at protein-to-material weight ratios of up to 40%. In addition, the CS NPs were found to provide sustained release of this protein in simulated intestinal fluid (pH=7.5) over 6-day period. Physical crosslinking by electrostatic interaction is simple and mild. However, CS NPs crosslinked by TPP have poor mechanical strength thus, limiting their applications in drug delivery.
During the last decade, considerable attention has been paid to inorganic-organic hybrid materials and interpenetrating polymer hybrid networks because it is possible to tailor their solid-state properties in relation to the nature and relative content of constitutive components [96]. Sol-gel process is known as a very straightforward way to prepare hybrid materials that effectively combine inorganic oxides and organic polymers. Seong-Bae et al. [97] developed a novel organic-inorganic composite membrane via a sol-gel
63
process, using tetraethyl orthosilicate (TEOS) as an inorganic material and CS as an organic compound. In the study, TEOS was hydrolyzed and condensed to form network structure and CS was incorporated into TEOS network structure to prepare pH-sensitive membrane. Some metal oxides are highly biocompatible because of their surfaces decorated with hydroxyl groups that render them intrinsically hydrophilic, as demonstrated by their numerous applications as implants or coatings. Most specifically, amorphous silica particles in contrast to crystalline silica are not toxic and are regularly used as food additives and components of vitamin supplements. It was also found that the shelf life of bioactive molecular encapsulated in silica could be prolonged [98, 99]. Even though all of these intrinsic advantages, metal oxide particles largely remain an untapped resource for the drug delivery system. It is due to the high processing temperature in traditional technique (>1000oC) and the difficulty of manipulating the internal microstructure of particles. However, sol-gel technology can easily overcome both problems above. Furthermore, the addition of organic molecules during the formation of the oxide backbone facilitates their encapsulation within the evolving oxide matrix and then forms a composite gel with homogeneously distributing throughout the resulting gel.
In this study, the NPs with particle size of 50-130 nm composed of CS and TEOS were obtained through emulsion and sol-gel process. Here, TEOS that hydrolyzed and condensed to form network structure with the CS improved the drug loading capacity (LC) and encapsulation efficiency (AE) of the process, compared to the pure CS NPs. Besides, the drug release mechanism was changed with increase of TEOS contents. Interpenetration network of CS/TEOS was expected to provide the mechanical enhancement of the nano sphere to restrict drug release before each electric-stimuli application
64
of cyclic operations. Therefore, the drug release from the CS/TEOS NPs effectively controlled by electric-stimuli operations was demonstrated in this work.
6.2 Fabraction of drug-loaded and non-loaded CS/TEOS nanoparticles The NPs composed of CS and TEOS were prepared by hydrolysis of TEOS in reverse microemulsion and ionic gelation between CS and sodium tripolyposphate (TPP). Typically, a mixture of 12g Triton X-100, 9.6 ml hexanol, and 20.4 ml H2O was ultrasonicated for 30 minutes to generate the microemulsion. After a pre-calculated amounts of TEOS were added to the mixture and mixed for six hours, 0.03 ml HCl solution was introduced to initiate TEOS hydrolysis. Next, 0.06g CS was dissolved in 10 ml 0.1 wt % acetic acid solutions which was then added to the TEOS solutions until the homogeneous solution was obtained. These two components were mixed to yield CS-TEOS with weight ratios of 10:1, 5:1, 10:3, and 2:1. After 24 h, 4mL TPP solution (10 wt %) was added to the mixed solution and then vibrated for additional 30 minutes by ultrasonic processor. Ethanol was used to wash the NPs to remove the surfactant and un-reacted chemicals. Finally, the CS/TEOS NPs were dispersed in the water solution.
Myoglobin protein was added to the mixture of TEOS and CS solution prior to the addition of TPP solution. The encapsulation efficiency and loading capacity of NPs with the different TEOS contents were determined by ultra-centrifugation of samples at 20,000g and 15oC for 30 min. The amount of free myoglobin protein was determined in clear supernatant by UV-visible spectrophotometer at 410 nm. The loading capacity (LC) of NPs and the encapsulation efficiency (AE) of the process were calculated from Eqs (6-1)
65
and (6-2) indicated below:
LC=(A-B)/C×100 (6-1) AE=(A-B)/A×100 (6-2)
where A is the total amount of myoglobin added during preparation, B is the amount of myoglobin remaining in the supernatant and C is the weight of the NPs.
The in vitro release of myoglobin from the NPs was evaluated over a 10-day period in pH=7.4 phosphate buffer solution (PBS). A total of 5 ml of release buffer was added to each of a series of 5mg of protein-loaded NPs.
The vials were incubated at 37oC without stirring. At specific time points, the NPs were isolated from buffer by ultracentrifugation. The concentration of myoglobin in the supernatant was measured by UV-visible spectrophotometer (Shimatzu, Model UV-2101PC) at a wavelength of 410 nm. Each batch was analyzed in triplicate. The drug release percent was determined using Eq (6-3) [100]:
Drug release (%) = Rt/L × 100% (6-3)
where L and Rt represent the initial amount of drug loaded and the cumulative amount of drug released at time t.
For electrical-induced drug release, drug-loaded NPs were placed in the middle of two platinum electrodes in phosphate buffer solution (PBS) (7.4).
Then, the amount of myoglobin protein released into medium was monitored and analyzed periodically by a UV spectrophotometer (Shimatzu, Model UV-2101PC). The UV absorbance of myoglobin protein was measured at λmax= 410 nm.
66
6.3 Results and Discussion
The CS-TEOS hybrid NPs were prepared using ionic gelation between CS and TEOS by TTP. In this sol-gel synthesis, polymer CS was encapsulated within the TEOS matrix resulting in a composite with homogeneous microstructure (or nanostructure). At first, TEOS was hydrolyzed and condensed to form meshed matrix or network structure wherein the silica nanophase and porous phase formed a dual continuous network throughout the materials (Fig. 6-1(a)). Then, CS was incorporated into the TEOS IPN structure homogeneously (Fig. 6-1(b)). When it was completely ionically crosslinked by P3O103- ions in an acidic TPP solution, CS could almost consume most of the binding sites and interact with the negatively charged phosphate groups of TPP (Fig 6-1(c)). In acidic condition, only TPP ions exist in the medium and then, CS NPs gelled in the TPP solution are completely ionically-crosslinked.
67
Figure 6-1. The illustrations of the formation of CS and TEOS NPs. (a) the formation of TEOS network structure, (b) the original CS polymer entwined into the network, (c) the formation of NPs by ionic gelation of TPP.
The SEM image (Fig. 6-2(a)) of pure CS NPs ionically crosslinked by TPP shows the nanoparticle having a size of 20-50 nm. However, with the TEOS content of 23%, as shown in Fig 6-2(b), hybrid NPs shows large aggregates. The CS and TEOS-IPN NPs reveal that the NP has a size range between 50 and 130 nm and shows an oval-like morphology. After swelling in the solution, the NP has a size about 150-230 nm.
100
100 nm
(b)
50 nm
(a)
100 nm
Figure 6-2 (a) SEM image of pure CS NPs. The (b) lower- magnification and higher-magnification TEM image of the CS and TEOS NPs (23%).
The model protein, myoglobin, was encapsulated into the hybrid NPs to evaluate their suitability as a delivery system It was found that the AE (the encapsulation efficiency of the process) of the CS NPs was improved from 62.38 to 91.53% with the increase of TEOS content up to 33%.The LC (the loading capacity of NPs) of CS NPs was enhanced greatly from 32.18 to 70.24%, as shown in Fig. 6-3. The LC and AE of the hybrid NPs for myoglobin were relatively high and comparable to that obtained from pure CS NPs.
Furthermore, the LC and AE can be improved by increasing the TEOS content.
The improvement of TEOS interaction with CS can be explainable that a higher TEOS concentration enhances the entwining degree of the interpenetration network and reduced the mesh sizes between the network structures. Hence, it is reasonable to believe that the encapsulation efficiency of myoglobin can be improved by increasing TEOS.
68
0 5 10 15 20 25 30 35
Figure 6-3 AE and LC of the CS and TEOS NPs with different TEOS contents.
69
Christophe et al. [101] pointed out that the physical characteristics of the oxides produced by the sol-gel processing can be tailored by controlling the sol-gel reaction kinetics, and in particular, the relative rates of hydrolysis and condensation. Hence, the ability of manipulating the gel microstructure has a very important consequence for design of a drug controlled release system.
For example, in sol-gel silica chemistry, introduction of acids or bases during the sol-gel process results in significantly different polymeric structures. In our work, acid catalysis promoted hydrolysis and end-of-chain condensation, leading to the production of small linear polymeric entities. A further crosslink between these linear polymers leads to the formation of microporous gels during gelation. As expected, the diffusion of molecules inside a microporous solid is much slower than that inside a mesoporous gel if a basic-catalyzed synthesis is carried out. This leads to significantly lower release rate for the gels synthesized using acid catalysis. As shown in Fig 6-4, the in vitro release of myoglobin from the NPs was measured over a 10-day period in PBS
(pH=7.4). It was found that the release of myoglobin from the pure CS NPs is much faster than that from the hybrid nanoparticles. In fact, 80% of the total myoglobin loaded in the pure CS NPs was released in the first 5 days.
0 2 4 6 8 10 12
Figure 6-4 In vitro release of myoglobin from the CS and TEOS NPs with different TEOS contents.
However, what is more than surprising is that the loaded myoglobin can still stay in the hybrid NPs even over 12 days. The released amount of myglobin from the hybrid NPs was less than 60% for 12 days. As mentioned above, the TEOS network structure could effectively stop the model protein myoglobin from diffusing outward from the hybrid NPs. In order to investigate the diffusion mechanism in the NPs, the data were further characterized with Eq (6-4):
t n
M Kt
M = (6-4)
where Mt is the amount of drug released at time t, M is the amount of drug released at equilibrium state, K is a constant and n is diffusion exponent related to the diffusion mechanism. Lee et al. [102] distinguished three classes
70
71
of diffusion according to the relative rates of diffusion and polymer relaxation.
The first is Fickian diffusion (n=0.5), in which the rate of diffusion is much smaller than the rate of relaxation. When the exponent n takes a value of 1.0,
The first is Fickian diffusion (n=0.5), in which the rate of diffusion is much smaller than the rate of relaxation. When the exponent n takes a value of 1.0,