Encapsulation of Uncharged Water-Insoluble Organic Substance in

Oct 1, 2008 - We report a general and versatile method for the encapsulation of electrically uncharged organic substance in polymeric capsules by usin...
0 downloads 0 Views 530KB Size
13258

J. Phys. Chem. B 2008, 112, 13258–13262

Encapsulation of Uncharged Water-Insoluble Organic Substance in Polymeric Membrane Capsules via Layer-by-Layer Approach Uttam Manna and Satish Patil* Solid State and Structural Chemistry Unit, Indian Institute of Science, Bangalore-560012 ReceiVed: July 11, 2008; ReVised Manuscript ReceiVed: August 13, 2008

We report a general and versatile method for the encapsulation of electrically uncharged organic substance in polymeric capsules by using a layer-by-layer (LbL) approach. Electrical charge was induced on the surface of pyrene (uncharged organic substance) with an amphiphilic surfactant (sodium dodecyl sulfate, SDS) by micellar solubilization. The SDS micellar solution of pyrene in water was then deposited on a flat substrate as well as colloidal particles with chitosan as an oppositely charged polyelectrolyte. Pyrene was used as a model drug because it displayed intrinsic fluorescence that allowed us to monitor LbL growth by fluorescence and under confocal laser scanning microscopy (CLSM). To examine the proof of concept, multilayers were coated on the planar support by the LbL method. UV-vis spectroscopy showed regular growth of each layer deposited. Thin film formation was evidenced by scanning electron microscopy. The LbL method was extended to particles where fluorescence spectroscopy revealed LbL growth and transmission electron microscopy (TEM) provided evidence of particle coating. The quantification of dye in each deposited layer further proved LbL growth. The removal of sacrificial core provided thin capsules. The capsules were characterized by TEM and CLSM. The capsules showed potential as a drug delivery system, which is suggested by the slow release of entrapped dye by concentration-dependent diffusion in isotonic saline solution. The kinetics of desorption of pyrene from this thin film was modeled by a pseudo-second-order model. Introduction Over the past decade, there has been growing interest in the fabrication of thin films by the layer-by-layer (LbL) approach, which was originally developed by Decher et al.1-3 It has been successfully applied to charged colloidal particles as a template (adsorbing substrate) to grow multilayer films.4-7 Polymer microcapsules with hollow microspheres have been fabricated by LbL assembly.8 This unique architecture has found enormous applications in the fields of medicine,9 drug and gene delivery,10 enhanced immunoessays,11 biomaterial engineering,12 and templates for various inorganic and organic nanoparticle synthesis.13 These vigorous activities have inspired the development of several approaches to achieve polymer microcapsules such as self-assembly of block copolymers,14 interfacial emulsion polymerization,15 and sacrificial core techniques.16 Despite these methods, there has been growing interest in new strategies to encapsulate and deliver water-insoluble drugs.17 Water-insoluble organic compounds are an important class of materials that are widely employed in pharmaceutics as drugs. The insolubility of medical drugs in aqueous solutions poses a serious problem for encapsulation as well as delivery.18 Most of the pharmaceutical drugs contain nonpolar aromatic, heterocyclic, and one or more polar functional groups such as hydroxyl, aldehyde, ketone, carboxylic, which is not enough to dissolve them in aqueous solutions.19 It affects their formulation and subsequent use in human body. Several common strategies have been developed to encapsulate water-insoluble drugs including incorporation of drugs with membrane lipids,20 polymer micelles,21 and model pyrene drug in sodium dodecylsulfate (SDS) as a templating agent.22 Another important issue for a successful drug delivery utility of the capsule is drug loading in terms of * To whom correspondence should be addressed. Phone: +91-8022932651. Fax: +91-80-23601310. E-mail: [email protected].

dose per unit volume in terms of formulation.20 Previous studies addressed this issue by using fourth-generation poly-(amidoamine) dendrimer (4G PAMAM) and poly(styrenesulfonate) (PSS) multilayer films on planar substrate as well as a charged template.23 But the stability of the capsule was a major issue due to different interactions between 4G PAMAM and PSS than those between two oppositely charged linear polyelectrolytes on solid substrate.24 In this study, we describe a method to encapsulate uncharged model drug (pyrene) in amphiphilic surfactant that would otherwise exhibit poor solubility in water by versatile LbL approach.25 We have used a micellar solution of pyrene in sodium dodecyl sulfate (SDS) as a model drug to exploit electrostatic interaction between chitosan and SDS. With this strategy it is possible to encapsulate a wide variety of waterinsoluble substances with high concentration. This method also allows for selective encapsulation of two different drugs: one preloaded into the capsule interior and other in the capsule core, thus providing a dual drug release system. Additionally, such capsular colloids also act as a pH-sensing system,26 if embedded with pH sensitive dyes. Such sensor capsules provide advantage of the functionalization inside the core as well as in the layer. Experimental Section Materials. Chitosan, Mw ≈ 200000, and pyrene were obtained from Aldrich. SDS was obtained from Spectrochem, India. Melamine formaldehyde (MF) particles, 3.3-3.5 µm (hydrochloric acid, HCl, soluble), were obtained from Microparticles GmbH, Berlin, Germany. HCl and AcOH were obtained from Qualigens, India. All materials were used without further purification. Ultrapure water (Millipore) with specific resistance around 18 MΩ cm was used. All experiments were carried out at room temperature.

10.1021/jp806140s CCC: $40.75  2008 American Chemical Society Published on Web 10/01/2008

Encapsulation of Water-Insoluble Organic Substance

J. Phys. Chem. B, Vol. 112, No. 42, 2008 13259

SCHEME 1: Schematic Illustration of the Capsule Formation

Solution Preparation. Chitosan was dissolved in Millipore water at pH 4.0 with a concentration of 1 mg/mL. Pyrene (0.0378 g) was dissolved in 50 mL of SDS solution (8.3 mmol dm-3) by using ultrasonication and heating to 80 °C for 1 h, and volume was made up to 100 mL in a volumetric flask to obtain a pyrene concentration of 378 ppm. Thin Film Preparation. The quartz substrates were cleaned in piranha (7:3 v/v H2SO4/H2O2) solution and then rinsed several times with Millipore water. Chitosan (1 mg/mL) solution was adsorbed on the surface of clean quartz glass slide for 2 h, carefully rinsed with water to remove unabsorbed chitosan, and dried under nitrogen gas flow. It was then placed in the micellar solution of pyrene and allowed to adsorb for 2 h, followed by three washing steps of water, carried out before depositing the next layer of chitosan. Multilayer films were prepared by this method. Each step was monitored with UV-vis spectroscopy. Calibration CurWe of Pyrene. For calibration, a solution of pyrene was prepared by dissolving 0.0405 g of pyrene in 100 mL of SDS (above critical micelle concentration) by successive ultrasonication and heating in a volumetric flask. Then these solutions were diluted and used for calibration at 335 nm. From this calibration curve, the amount of water-insoluble pyrene molecules in the micellar/chitosan membrane was calculated, as discussed later. Capsule Preparation. The coating of MF particles was achieved as shown in Scheme 1. The positively charged MF particles (0.1 mL of 10 wt % dispersion) were incubated with micellar solution of pyrene allowing adsorption for 1 h. The suspension was centrifuged, followed by four washings steps with water in order to remove nonadsorbed pyrene on MF particles. Then the chitosan solution (1 mg/mL) was added to the MF particles suspension and allowed to adsorb for 20 min, followed by three washing steps with water before depositing the next layer of micellar solution of pyrene. Alternatively, pyrene and chitosan steps were repeated to construct multilayers on MF particles. Hollow capsules were obtained by dissolving MF particles in 0.1 N HCl solution for 10 min and then washed three times with water. Characterization. The UV-vis spectra of chitosan/pyrene deposited quartz glass slide and MF particles were obtained with Perkin-Elmer (Lambda 35) spectrometer. Fluorescence spectra were recorded by using Perkin-Elmer, Fluorimeter (LS 50B) with excitation wavelength λmax ) 337 nm. All measurements were undertaken at room temperature. Microscopy. Transmission Electron Microscopy (TEM). A dilute solution of coated MF particle and capsule suspension

Figure 1. Multilayer film growth of chitosan and pyrene on a quartz substrate, as studied by UV-vis spectrophotometry. The inset shows the SEM image of a coated flat substrate with a scale bar of 10 µm.

was drop cast on the carbon-coated copper grid. The measurements were carried out at room temperature with a Tecnai F30 field-emission microscope. From the AFM image the thickness of the capsule was 90 nm for 5 bilayers (not shown). The thickness per bilayer was calculated by dividing total capsule thickness by number of deposited layers. Confocal Laser Scanning Microscopy (CLSM). CLSM measurements were performed using Leica (version-sp I) with a 40× objective. The excitation wavelength was 350 nm. The dilute solution of suspension on glass plate was allowed for natural air drying. Scanning Electron Microscopy (SEM). SEM measurements were performed at room temperature and at atmospheric pressure using JEOL, JSM-5600LV scanning electron microscope at 5 kV. Results and Discussion LbL on Planar Substrate. Pyrene was incorporated as a model hydrophobic drug because the steady state fluorescence and UV-vis measurement provides the information of its location in the multilayer growth and hence, it is easy to follow LbL growth on the flat substrate and particles. The multilayer growth of pyrene/chitosan on quartz substrate was followed by UV-vis, which showed the regular growth of the film (Figure 1). LbL coating of micelles followed by chitosan on the glass plate is further established by SEM, as shown in inset of the Figure 1. The film was scratched from the quartz plate and transferred to SEM grid for measurement. From the UV-vis spectra, the regular film growth was observed for the deposition of each layer of pyrene, and it also revealed adsorption/desorption trend. The characteristic peak of pyrene decreased (not shown here) after the chitosan adsorption step. This behavior has been previously observed by Caruso et al. for polyelectrolyte/dye multilayer film.27 It also provides additional evidence of chitosan adsorption. The linear increase of absorbance intensity is typical for a stepwise and regular growth of a thin film. The conditions employed for planar supports were extended to colloid particles. LbL on MF Particles. The micellar solution of pyrene/ chitosan multilayers were adsorbed onto MF particles by LbL approach. UV-vis experiments were conducted to follow multilayer growth on the surface of particles.

13260 J. Phys. Chem. B, Vol. 112, No. 42, 2008

Figure 2. LbL growth of pyrene on MF particles, as studied by UV-vis spectrophotometry. The inset shows linear growth of pyrene layer on MF particles.

Figure 3. LbL growth of pyrene on MF particle, as studied by fluorescence spectrophotometry.

UV-vis data with each coating step of pyrene revealed successful growth of multilayers occurring on the particles. Figure 2 shows the spectra of pyrene on colloid particles. The strongest π-π* transition absorption peak occurs at about 240, 273, and 336 nm. In addition to these peaks it also shows an additional peak at 367 nm, which may be brought about by aggregated pyrene molecule on the surface of the particles. Aggregation of pyrene molecule is induced by change in conformation and composition of the pyrene molecule in SDS upon adsorption on the surface of particles. In other words, the interaction between the pyrene molecule and the underlying polyelectrolyte (chitosan) leads to a reordering of the aggregates on the surface. LbL growth on particles was also confirmed by fluorescence as shown in Figure 3. The emission spectrum of monomer pyrene exhibits five vibronic bands between 370 and 400 nm due to π-π* transition. The vibration bands, I at 373 nm and III at 383 nm, were clearly resolved. Additionally, examining the vibronic band intensities of the fluorescence emission spectrum can elucidate the nature of the pyrene environment.28 It is known that the ratio of the emission intensities at 383 to 373 nm (I1/I3) is indicative of the polarity of the pyrene molecule′s environment. I1/I3 value for the fifth layer of

Manna and Patil

Figure 4. TEM micrograph of MF particle coated with pyrene and chitosan comprising 10 layers. The inset shows higher magnification of one of the spheres, showing the surface roughness due to the multilayers deposited and scale bar is 0.15 µm.

pyrene-SDS layer is 1.32 and for the fourth layer is 1.44. The I1/I3 ratio indicates that pyrene is in a hydrophobic environment and is at least as hydrophobic as the original micelles in solution.29 The broadband centered around 420 nm is due to pyrene excimers. This is in agreement with the broad absorption peak at 367 nm in UV-vis measurement. Similar to the UV-vis measurements, there was a linear relationship between the numbers of pyrene layers on the MF particles to the fluorescence intensity of the pyrene. The growth and morphology of pyrene/chitosan multilayer on MF particles were examined using TEM. Figure 4 shows the TEM micrograph of coated MF particles. The coated particles exhibited increased surface roughness, which clearly shows the coating of pyrene/chitosan on the surface of the particles. Coating of pyrene/chitosan layers also resulted in an increase in the overall diameter of the MF particles. From the TEM it can also be deduced that the thickness of the film was 120 nm for 5 bilayers. Thus thickness per bilayer was 24 nm. The obtained thickness for each bilayer is greater than those attributed to coating of conventional linear polyelectrolytes.30 This could be due to a different kind of an interaction between SDS and chitosan. The multilayer coating of pyrene/chitosan was further investigated by CLSM. The particles appeared as fluorescent ring due to presence of pyrene in the multilayer shell (not shown). Pyrene/Chitosan Hollow Capsule. The capsules were prepared by exposing pyrene/chitosan multilayer-coated MF particles to 0.1 M HCl solution. The resulting hollow capsules were characterized by TEM and CLSM. From parts a-c of Figure 5, it is evident that stable capsules have been formed. TEM images of pyrene/chitosan capsules are shown in parts a and b of Figure 5. The morphology of capsules is invariably different from the coated particle. The dissolution of MF particle induces the change in the shape of hollow capsules. Previous studies have also shown the change in the shape of hollow capsules induced by air-drying.31 The thickness of the capsule was 90 nm for a 5-bilayer capsule, which provided thickness per bilayer as 18 nm. This thickness is almost 2 times greater than the multilayers of conventional polyelectrolytes. This may be due to the shrinking

Encapsulation of Water-Insoluble Organic Substance

J. Phys. Chem. B, Vol. 112, No. 42, 2008 13261

Figure 5. (a and b) TEM image of hollow capsule and (c) CLSM image of hollow capsule. TABLE 1 no. of layers

amount of drug loaded (µg/15 µL MF particles)

2 3 4 5

0.488 0.525 0.986 1.173

and altogether different elastic characteristics of the surfactant/ PE multilayer film. Elastic characteristics of such films are under investigation in our laboratory. CLSM image show fluorescent capsules suggesting presence of pyrene in the capsular membrane (Figure 5c). Loading of Pyrene. The loading of encapsulated pyrene with respect to the number of layers was investigated by UV-vis spectroscopy. Pyrene loading was determined by taking UV absorption of pyrene at 335 nm. The particles were incubated with pyrene incorporated micelle solution and kept for adsorption of micelles on 15 µL MF particles for 2 h. After that we performed washing of coated particles by centrifugation, and completion of washing was confirmed by UV analysis of the wash fluid after centrifugation. After washing, 3 mL of milli-Q water was added to the coated particles, and the UV spectrum of that sample was taken. The same procedure was followed for other layers also. Thus we calculated the amount of encapsulated pyrene for different layers (as shown in Table 1) using the calibration curve. The amount of drug loading increases with increase in number of layers from 2 to 4 layers of pyrene. By use of the procedure outlined above, we encapsulated 1.17 µg of the model drug where 15 µL MF particles were taken for coating of pyrene incorporated SDS micelles followed by Chitosan. Pyrene Release. To demonstrate release of entrapped pyrene, the multilayer film of pyrene/chitosan on the quartz substrate was exposed to 40 mL of 0.155 M NaCl, pH 6.5 at room temperature. At regular intervals, the coated glass slide was removed from the solution, rinsed with water, and dried under nitrogen flow, and the absorbance spectra were recorded.

Figure 6. Release profile of pyrene/chitosan multilayers (10 layers) on a quartz slide in 0.155 M NaCl at room temperature.

The percentage release was calculated using following equation,

(C0 - Ct) × 100 C0 where C0 and Ct are concentration of pyrene in thin film at time zero and at time t, respectively, at room temperature. The intensity of pyrene in multilayer thin film followed a gradual decrease with time. Thus the amount of pyrene in saline solution was found to increase with increasing time. Figure 6 shows the release profile of pyrene entrapped in multilayer film by the LbL approach. It showed an initial burst release of 17-20%, followed by an additional release of more than 70% over a period of 4 h. This demonstrates the potential utility of such films for sustained release applications. It shows that high ionic

13262 J. Phys. Chem. B, Vol. 112, No. 42, 2008

Manna and Patil coating and capsule formation has been observed by TEM, AFM, and CLSM. We also showed control release of model drug in isotonic saline solution. This unique approach will provide dual drug release system. We are currently exploiting the possibility of loading and release of two drugs with this novel concept. Acknowledgment. Mr. K. R. Satyanarayana, and Mr. Amit Kumar Mondal are thanked for assistance with TEM and SEM measurements. We also thank Prof. Giridhar Madras and Dr. Ajay Khopade for useful discussion. Financial support from the Department of Science and Technology, India, is greatly acknowledged. References and Notes

Figure 7. Desorption kinetics study of pyrene from thin film in saline solution The value of qe obtained from the slope of the linear fit is 0.024. This is consistent with the equilibrium value of q obtained from the plot of qt with time (inset of Figure 7).

strength solutions cause the removal of dye molecules from multilayers composed of polyelectrolytes and micelle complexes assembled on planar supports. However, the factors influencing the encapsulation and release of pyrene in micelles/chitosan capsule, effect of ionic strength, and interaction of micelles with chitosan are under investigation in our laboratory. Desorption of Pyrene. To investigate the desorption process of model hydrophobic drug pyrene from pyrene incorporated micelles/chitosan thin films, a pseudo-second-order model was used32

qe )

(C0 - Ce)V W

where V is volume of saline solution, W is the weight of the thin film and, ks is the rate constant. In this study, V is 20 mL saline solution and W is weight of thin film (0.4 mg). C0 is the amount of pyrene incorporated into the thin film before starting the desorption study in saline solution. Intergration of the above differential equation with q ) qt at t yields

t 1 1 ) + t qt k q2 qe s e

Thus, a plot of t/qt with desorption time (t) would be linear with a slope of 1/qe. Figure 7 shows the experimental values of t/qt fitted linearly with desorption time t. Conclusion In conclusion, we have demonstrated the encapsulation of uncharged water-insoluble organic substances in polymeric membrane capsules. This novel concept has been successfully applied to a flat substrate as well as colloid particles. Subsequently, we have also shown a formation of stable capsules by dissolving the MF core by the addition of 0.1 N HCl. The

(1) Decher, G.; Hong, J. D. Makromol. Chem. Macromol. Symp. 1991, 46, 321. (2) Decher, G. Science 1997, 277, 1232. (3) Decher, G.; Eckle, M.; Schmitt, J.; Struth, B. Curr. Opin. Colloid Interface Sci. 1998, 3, 32. (4) Donath, E.; Sukhorukov, G, B.; Caruso, F.; Davis, S. A.; Mohwald, H. Angew. Chem., Int. Ed. 1998, 37, 2201. (5) Caruso, F.; Schuler, C.; Kurth, D. G. Chem. Mater. 1999, 11, 3394. (6) Schuler, C.; Caruso, F. Biomacromolecules 2001, 2, 921. (7) Pastoriza-Santos, I.; Schuler, B.; Caruso, F. AdV. Funct. Mater. 2001, 11, 122. (8) Khopade, A. J.; Caruso, F. Chem. Mater. 2004, 16, 2107. (9) Serizawa, T.; Yamaguchi, M.; Akashi, M. Biomacromolecules 2002, 3, 724. (10) Caruso, F.; Trau, D.; Mohwald, H.; Renneberg, R. Langmuir, 2000, 16, 1485. (11) Charych, D.; Reichert, A. U.S. Patent No. 6,180,135, 1997. (12) Decher, G.; Lehr, B.; Lowack, K.; Lvov, Y.; Schmitt, J. Biosens. Bioelectron. 1994, 9, 677. (13) Wang, Y.; Angelatos, A. S.; Caruso, F. Chem. Mater. 2008, 20, 848. (14) Huang, H.; Remsen, E. E.; Kowalewski, T.; Wooley, K. L. J. Am. Chem. Soc. 1999, 121, 3805. (15) Rmaile, H. H.; Schlenoff, J. B. J. Am. Chem. Soc. 2003, 125, 6602. (16) Peyratout, C. S.; Daehne, L. Angew. Chem., Int. Ed. 2004, 43, 3762. (17) Haynes, D. H. U.S. Patent No. 5,091,187, 1992. (18) Haynes, D. H. U.S. Patent No. 5,246,707, 1993. (19) Kirpotin, D.; Chan, D. C. F.; Bunn P. WO 98/14180, 1998. (20) Zhihua, A.; Mohwald, H.; Junbai, L. Biomacromolecules 2006, 7, 580. (21) Nguyen, P. M.; Zacharia, N. S.; Verploegen, E. V.; Hammond, P. T. Chem. Mater. 2007, 19, 5524. (22) Caruso, F.; Yang, W.; Trau, D.; Renneberg, R. Langmuir 2000, 16, 8932. (23) Khopade, A. J.; Caruso, F. Nano Lett. 2002, 2, 415. (24) Khopade, A. J.; Caruso, F. Biomacromolecules 2002, 3, 1154. (25) Humphry-Baker, R.; Gratzel, M.; Moroi, Y. Langmuir 2006, 22, 11205. (26) Kreft, O.; Javier, A. M.; Sukhorukov, B. G.; Parak, W. J. Mater. Chem. 2007, 17, 4471. (27) Tedeschi, C.; Caruso, F.; Mohwald, H.; Kirstein, S. J. Am. Chem. Soc. 2000, 122, 5841. (28) Ray, G. B.; Chakraborty, I.; Moulik, S. P. J. Colloid Interface Sci. 2006, 294, 248. (29) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039. (30) Dong, W.; Ferri, J. K.; Adalsteinsson, T.; Schonhoff, M.; Sukhorukov, G. B.; Mohwald, H. Chem. Mater. 2005, 17, 2603. (31) Caruso, F.; Donath, E.; Mohwald, H.; Georgieva, R. Macromolecules 1998, 31, 7365. (32) Ho, Y. S.; McKay, G. J. EnViron. Sci. Health 1999, A34, 1179.

JP806140S