Electrically Actuatable Smart Nanoporous Membrane for Pulsatile

Jan 31, 2011 - Biranche Tandon , Adrián Magaz , Richard Balint , Jonny J. Blaker , Sarah H. .... Baljeet Singh , Tarun Garg , Amit K. Goyal , Goutam ...
0 downloads 0 Views 5MB Size
Download PDF
LETTER pubs.acs.org/NanoLett

Electrically Actuatable Smart Nanoporous Membrane for Pulsatile Drug Release Gumhye Jeon, Seung Yun Yang, Jinseok Byun, and Jin Kon Kim* National Creative Research Initiative Center for Block Copolymer Self-Assembly, Department of Chemical Engineering, Pohang University of Science and Technology, Kyungbuk 790-784, Korea

bS Supporting Information ABSTRACT: We report on the fabrication of electrically responsive nanoporous membrane based on polypyrrole doped with dodecylbenzenesulfonate anion (PPy/DBS) that was electropolymerized on the upper part of anodized aluminum oxide membrane. The membrane has regular pore size and very high pore density. Utilizing a large volume change of PPy/DBS depending on electrochemical state, the pore size was acutated electrically. The actuation of the pores was experimentally confirmed by in situ atomic force microscopy and in situ flux measurement. We also demonstrated successfully pulsatile (or on-demand) drug release by using fluorescently labeled protein as a model drug. Because of a fast switching time (less than 10 s) and high flux of the drugs, this membrane could be used for emergency therapy of angina pectoris and migraine, which requires acute and on-demand drug delivery, and hormonerelated disease and metabolic syndrome. KEYWORDS: Pulsatile drug release, responsive membrane, electric stimulus, polypyrrole, anodic aluminum oxide membrane

C

ontrolled and long-term drug release of protein therapeutics as well as small-sized drugs has been recognized as one of the most promising biomedical technologies for the treatment of chronic diseases requiring daily dosing.1,2 Among many different types of the controlled release, the pulsatile drug release has gained much attention.3,4 This is because drug delivery with pulsatile manner is required for treatment of hormone-related diseases that need frequently painful injections, local pain relief of chronic diseases, post surgical pain control, and precise dose control of highly toxic medication such as localized chemotherapy. To achieve pulsatile drug delivery, many research groups have used various responsive materials5 based on external stimuli, such as pH,6,7 temperature,8,9 biological agent,10,11and degradation rate control.12 However, the above-mentioned stimuli cannot be controlled artificially. Among many actively (or on-demand) controlled stimuli (ultrasound,13 light,14 magnetic field,15,16 and electrical field3,17-27), the electrical signal would be the best source because it is portable and does not need large or special equipment to trigger it. The signal can also be easily and ondemand controllable, and long cycles are possible. Furthermore, when a sensor or microchip system is combined, the feedback and remote control outside the body is possible.4,20 Langer and co-workers3,18 fabricated the microchip containing individually openable reservoirs capped with a thin gold membrane. An applied electric potential dissolved gold film electrochemically, resulting in the ejection of the chemicals (or drugs) in each reservoir artificially (or on-demand). Later, Santini and coworkers19,20 extended this microchip, and the cap made of gold or multiple metal layers were removed electrothermally by r 2011 American Chemical Society

applying electric current. The pulsatile release of the nonapeptide leuprolide acetate was carried out in vivo inside dogs by telemetry. However, the fabrication of the microchip is time-consuming and expensive due to multisteps such as lithography, chemical vapor deposition, and reactive ion etching. Furthermore, when frequent release is required, the number of reservoir should be increased dramatically. Martin and co-workers21 prepared poly(3,4-ethylene dioxythiopene) (PEDOT) nanotubes by electropolymerization outside biodegradable poly(lactide-co-glycolide) (PLGA) matrix prepared by electrospinning. A drug (dexamethasone) was loaded in PLGA matrix. The contraction of PEDOT nanotubes at the oxidation state created a mechanical force resulting in openings or cracks on the surface of nanotubes, and thus the drug could be released. However, they could not achieve complete pulsatile release, since a small amount of dexamethasone could be released without the electrochemical stimulation. Also, the tube diameter and openings were not uniform, which makes difficult to control drug dosage precisely. The best way to achieve pulsatile release of drugs based on electric stimulus for an extended period is to release drugs through the nanoporous membrane with uniform pore sizes from a depot system containing large amounts of drugs for the release during extended periods.23 This is because the pore size

Received: December 13, 2010 Revised: January 18, 2011 Published: January 31, 2011 1284

dx.doi.org/10.1021/nl104329y | Nano Lett. 2011, 11, 1284–1288

Nano Letters

LETTER

Figure 1. Schematic to fabricate electrically responsive nanoporous membrane. (a) Fabrication of AAO membrane. (b) A thin gold layer was thermally deposited on top and side wall of the membrane. (c) Polypyrrole was electropolymerized on gold layer, and the electrically responsive nanoporous membrane was fabricated at the oxidation state. (d) Reversible change of pore size (and the drug release rate) between oxidation and reduction states.

could be actuatable with electrical signal when the nanoporous membrane made of electrically responsive polymers is used. In this study, we fabricated electrically responsive nanoporous membrane based on polypyrrole doped with the dodecylbenzenesulfonate anions (PPy/DBS) that was electropolymerized onto the top and upper side wall of anodized aluminum oxide (AAO) membrane. The rationale for choosing PPy/DBS as the electrically responsive material is that it exhibits a very large volume change (up to 35 vol %) depending on the electrochemical state24-26 and excellent biocompatibility.27 The driving potential is less than 1.1 V, which is sufficiently lower than that (3 V) used for the operation of an artificial heart. The actuation of the pore size was successfully realized by changing the electrochemical state, and it was experimentally confirmed by in situ atomic force microscopy (AFM) and in situ flux measurement. The pore size decreased at the reduction state while it increased at the oxidation state. When an appropriate pore size was selected, pulsatile drug release was successfully demonstrated using fluorescein isothiocyanate-labeled bovine serum albumin (FITC-BSA) as a model protein drug. Figure 1 shows a scheme to fabricate an electrically responsive nanoporous membrane. First, AAO membrane was prepared by two-step anodization (Figure 1a).28-33 A thin gold layer was thermally evaporated on the top of the AAO membrane (Figure 1b). During the evaporation, some gold was also deposited on the upper parts of the pore walls. The pore size after gold deposition was slightly smaller than that of original AAO membrane. Then, PPy/DBS was electropolymerized on the gold layer. The thickness of PPy/DBS layer, which decides the initial pore size, was controlled by electropolymerization time (Figure 1c). The height of PPy/DBS layer was very short (∼1.5 μm) compared with the entire thickness of the AAO (60 μm) due to the limited penetration of gold into the pore during deposition. This gives another advantage to allow higher flux come from the short channel length. The pore size of the

responsive membrane could be actuated electrochemically due to the difference of PPy/DBS volume depending on electrochemical state (oxidation vs reduction) (Figure 1 d). Figure 2 gives top and cross-sectional field emission-scanning electron microscopy (FE-SEM) images of the nanoporous membranes at various fabrication steps. The AAO membrane has hexagonally packed cylindrical microdomain with a centerto-center distance between two neighboring pores of 500 nm and uniform pore size with a diameter of 410 nm (Figure 2a,b). When a 40 nm thick Au layer was thermally deposited, the average pore diameter decreased to 380 nm (Figure 2c,d). The pore diameter of the membrane was easily tuned because the thickness of polypyrrole increases almost linearly with electropolymerization time (Supporting Information Figure S1). At an electropolymerization for 60 s, the pore diameter became 200 nm (Figure 2e,f). To confirm pore size actuation of the membrane with initial pore diameter of 200 nm by the electrical stimulus, in situ flux and AFM measurement were carried out. For this purpose, we prepared a homemade flux cell with three electrodes (Figure S2 in the Supporting Information). The effective membrane area was 28.3 mm2 and a constant pressure of 0.1 bar was used. Detailed experimental information was presented in Section 2 of Supporting Information. Figure 3a shows in situ flux measurement of 0.1 M NaDBS aqueous solution at two different electrochemical states (oxidation vs reduction states). The average flux at the oxidation state was 730 L/m2h, while it decreased to 250 L/m2h at the reduction state. Interestingly, the switching time upon the change of the electrical stimulus was very fast (less than 10 s) and good recyclability (more than 13) was obtained. To test recyclability and stability of this nanoporous film at very long oxidation/ reduction repeated cycles, we performed cyclovoltammetry test. We found that even after 1000 repeated cycles, the characteristics of PPy did not change. Also, we did not observe any detachment of PPy/DBS layer from the gold layer. 1285

dx.doi.org/10.1021/nl104329y |Nano Lett. 2011, 11, 1284–1288

Nano Letters

Figure 2. Top (left panel) and cross-sectional (right panel) SEM images of the membranes at various fabrication steps. (a,b) AAO membrane. (c,d) Gold deposited membrane. (e,f) After electropolymerization of PPy/DBS for 60 s.

Finally, the good adhesion between PPy/DBS layer from gold layer was maintained even when a peel test was performed by using 3M Scotch tape (Section 3 and Figure S3 of Supporting Information). The enhanced adhesion can be explained by two factors. First, since the volume change of PPy/DBS depending on electrochemical states occurs within nanometer scale, the mechanical stress resulting from volume change would be small. Second, compared with a conventional PPy actuator where the interface between PPy and gold is very smooth, nanoporous membrane has large interfacial area because of very high density of nanopores. Figure 3b,c gives in situ AFM height images at the oxidation and reduction states, respectively. Both AFM images were obtained in the presence of the solution. The pore size change depending on electrochemical state is well consistent with flux results. The average pore diameter at the oxidation and reduction states was 190 and 140 nm, respectively. The pore diameter measured by AFM is slightly smaller than that (200 nm) measured by SEM (Figure 2e). This is because the sample for SEM was prepared after drying, while AFM images were directly measured in the presence of the solution. The mechanism of pore size actuation depending on electrochemical state was explained as follows.22,24 PPy was initially electropolymerized at an oxidation state in the presence of DBSdopant ions. In this situation, the negatively charged DBSshould be located near the positively charged PPy chains. If a negative potential is applied (reduction state), the PPy chains should be converted to the electrically neutral state. One possible situation to make the neutrality is that DBS- ions should expel from the PPy chains. However, because of the large size of the DBS- ions, these ions could not be removed from the PPy chains. Thus, the hydrated positive counterions (Naþ3H2O) should enter the PPy chains to make the neutrality, resulting in the volume expansion of the PPy/DBS thin film at the reduction state. At an oxidation state, Naþ3H2O could expel from the PPy chains, which causes volume shrinkage. The pore size at the

LETTER

Figure 3. The in situ flux and AFM results of a membrane with initial pore diameter of 200 nm at two different electrochemical states. (a) In situ flux versus time. Data points were taken every 15 s. Open (blue) and closed (magenta) circles indicate the oxidization and reduction states. (b,c) The in situ AFM height images corresponding to the oxidization and reduction states.

reduction state is smaller than the oxidation state since the pore size of the membrane decreases with increasing the PPy/DBS volume. Now, we correlate the change between flux (J) and the pore diameter (d). According to the Hagen-Poiseuille equation, J through a membrane with cylindrical straight pores is given by J¼

2 dpore ε ΔP 32ητ h

ð1Þ

in which dpore is the diameter of pores, ε is the porosity of the film (proportional to the square of dpore at the same number of pores), ΔP is the pressure difference, η is the viscosity of water solution with 0.1 M NaDBS, τ is the tortuosity of the pore, and h is the thickness of the responsive membrane. Although AAO has a long cylindrical channel (60 μm), we only consider the thickness of the PPy/DBS layer (1.5 μm) because the flux was mainly affected by small-sized pores. Then, at a given ΔP, the ratio of the flux at different electrochemical states is given by J2 h1 d24 ¼ ¼ J1 h2 d14

h1 d24  ðd1 −d2 Þ 4 h1 þ d1 2

ð2Þ

Here, the subscripts 1 and 2 represent the oxidation and reduction states, respectively. We assumed that the expansion (or reduction) of the PPy/DBS chains toward pore radial direction is very similar to that toward the thickness direction on the film surface. This is because PPy/DBS chains could not expand toward the rigid Au layer (Supporting Information Figure S4). By using d1 = 190 nm and h1 = 1500 nm, measured by AFM and SEM images (Figure 3 and Supporting Information Figure S1f) and J2/J1 = 730/250 (Figure 3a), d2 is calculated as 146 nm, which is similar to the measured value (140 nm) by in situ AFM 1286

dx.doi.org/10.1021/nl104329y |Nano Lett. 2011, 11, 1284–1288

Nano Letters

LETTER

Figure 5. The cumulative concentration of FITC-BSA with time. The membrane with initial pore diameter of 110 nm was prepared by electropolymerization at 80 s. Open (blue) and closed (magenta) circles represent the oxidation and reduction states. Figure 4. (a) Flux versus time for a membrane with a pore diameter of 110 nm at the oxidation (blue open circles) and the reduction state (magenta closed circles). A data point was collected every 10 s. (b,c) Insitu AFM height images corresponding to the oxidation and the reduction states, respectively.

(Figure 3c). The volume change between two states is calculated as 24% (Section 4 of Supporting Information), which is a reasonable value, because the reported values of the volume change of PPy between two electrochemical states are from (0.5 to 35)% depending on polymerization methods and types of dopant ion.24,25 For pulsatile (or on-demand) durg delivery, on-off control of the drug release is critical. To check whether the pore is completely closed at the off state, we fabricated another membrane with initial pore diameter of 110 nm, which was prepared by electropolymerization for 80 s. Figure 4a shows in situ flux measurement of 0.1 M NaDBS aqueous solution at two different electrochemical states (oxidation vs reduction states). The flux of 60 L/m2h at the oxidation state (onstate) decreased to zero at the reduction state (off-state), suggesting that the pores were completely closed at the reduction state. This is consistent with in situ hight AFM images (Figure 4b,c). Using this membrane, we performed pulsatile release by using 0.3 mg/mL FITC-BSA in 0.1 M NaDBS aqueous solution in the reservoir as a model protein drug. FITC is easily detected even at very low concentration (0.01 μg/mL). The concentration of released amount of FITC through the membrane was determined by using a fluorometer (the details are given in Section 5 of Supporting Information). Figure 5 shows the change of the culmulative concentration of FITC-BSA penetrated through the membrane at the oxidation and reduction states with time. At the oxidation state, the concentration of FITC-BSA increased with time because the pores were open. However, at the reduction state the concentration does not change with time, indicating that no FITC-BSA penetrated through the membrane because of the completely closed pores. The repeated experiment by changing oxidation and reduction states showed that the on-off control is reversible; thus the pulsatile release of FITC-BSA is successfully achieved by using the responsive membrane.

In conclusion, we have shown that pore size of the PPy/DBS nanoporous membrane was reversibly actuated by the electrochemical state. As a result, a pulsatile release of the model drug was demonstrated. The response time is very quick (less than a few seconds) and on-demand stimulus is available due to using electrical stimulus. Also, when a pore is open state, a high flux of the drugs could be obtained from very high pore density (2.5  109/ in2) and short effective channel length. Furthermore, the size of the pores was very uniform through the entire membrane during the actuation. Considering these results, drug release of accurate dosage control depending on a patient state would be possible. Also, the denaturation or damage of drugs could be effectively prevented from surrounding environments due to using the depot system. This responsive membrane could be used for emergency therapy of angina pectoris and migraine, which requires acute and on-demand drug delivery, hormone-related disease, and metabolic syndrome.

’ ASSOCIATED CONTENT

bS

Supporting Information. Materials and method for fabrication of responsive membranes, flux and drug release experiments, and Figures S1-S5. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by the National Creative Research Initiative Program of the National Research Foundation of Korea (NRF). ’ REFERENCES (1) Desai, T. A.; Hansford, D. J.; Ferrari, M. Biomol. Eng. 2000, 17, 23–36. (2) Peng, L.; Mendelsohn, A. D.; LaTempa, T. J.; Yoriya, S.; Grimes, C. A.; Desai, T. A. Nano Lett. 2009, 9, 1932–1936. 1287

dx.doi.org/10.1021/nl104329y |Nano Lett. 2011, 11, 1284–1288

Nano Letters

LETTER

(3) Santini, J. T., Jr; Richards, A. C.; Scheidt, R.; Cima, M. J.; Langer, R. Angew. Chem., Int. Ed. 2000, 39, 2396–2407. (4) Timko, B. P.; Dvir, T.; Kohane, D. S. Adv. Mater. 2010, 22, 4925– 4943. (5) Stuart, M.; Huck, W.; Genzer, J.; Muller, M.; Ober, C.; Stamm, M.; Sukhorukov, G.; Szleifer, I.; Tsukruk, V.; Urban, M.; Winnik, F.; Zauscher, S.; Luzinv, I.; Minko, S. Nat. Mater. 2010, 9, 101–113. (6) Richardson, T. P.; Peters, M. C.; Ennett, A. B.; Mooney, D. J. Nat. Biotechnol. 2001, 19, 1029–1034. (7) Binder, W. H. Angew. Chem., Int. Ed. 2008, 47, 3092–3095. (8) Chu, L.-Y.; Li, Y.; Zhu, J.-H.; Chen, W.-M. Angew. Chem., Int. Ed. 2005, 44, 2124–2127. (9) Keurentjes, J. T. F.; Kemmere, M. F.; Bruinewoud, H.; Vertommen, M. A. M. E.; Rovers, S. A.; Hoogenboom, R.; Stemkens, L. F. S.; Peters, F. L.; Tielen, N. J. C.; Asseldonk, D. T. A. V. Angew. Chem., Int. Ed. 2009, 48, 9867–9870. (10) Miyata, T.; Asami, N.; Uragami, T. Nature 1999, 399, 766–769. (11) Ulijn, R. V.; Bibi, N.; Jayawarna, V.; Thornton, P. D.; Todd, S. J.; Mart, R. J.; Smith, A. M.; Gough, J. E. Mater. Today 2007, 10, 40–48. (12) Grayson, A. C. R.; Choi, I. S.; Tyler, B. M.; Wang, P. P.; Brem, H.; Cima, M. J.; Langer, R. Nat. Mater. 2003, 2, 767–772. (13) Kost, J.; Leong, K.; Langer, R. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 7663–7666. (14) Kloxin, A. M.; Kasko, A. M.; Salinas, C. N.; Anseth, K. S. Science 2009, 324, 59–63. (15) Hoare, T.; Santamaria, J.; Goya, G. F.; Irusta, S.; Lin, D.; Lau, S.; Padera, R.; Langer, R.; Kohane, D. S. Nano Lett. 2009, 9, 3651–3657. (16) Cai, K.; Luo, Z.; Hu, Y.; Chen, X.; Liao, Y.; Yang, L.; Deng, L. Adv. Mater. 2009, 21, 4045–4049. (17) Kwon, I. C.; Bae, Y. H.; Kim, S. W. Nature 1991, 354, 291–293. (18) Santini, J. T., Jr; Cima, M. J.; Langer, R. Nature 1999, 397, 335– 338. (19) Maloney, J. M.; Uhland, S. A.; Polito, B. F.; Sheppard, N. F., Jr; Pelta, C. M.; Santini, J. T., Jr. J. Controlled Release 2005, 109, 244–255. (20) Prescott, J. H.; Lipka, S.; Baldwin, S.; Sheppard, N. F., Jr; Maloney, J. M.; Coppeta, J.; Yomtov, B.; Staples, M. A.; Santini, J. T., Jr. Nat. Biotechnol. 2006, 24, 437–438. (21) Abidian, M. R.; Kim, D. H.; Martin, D. C. Adv. Mater. 2006, 18, 405–409. (22) Wood, K. C.; Zacharia, N. S.; Schmidt, D. J.; Wrightman, S. N.; Andaya, B. J.; Hammond, P. T. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 2280–2285. (23) Yang, S. Y.; Yang, J.-A.; Kim, E.-S.; Jeon, G.; Oh, E. J.; Choi, K. Y.; Hahn, S. K.; Kim, J. K. ACS Nano 2010, 4, 3817–3822. (24) Smela, E.; Gadegaard, N. Adv. Mater. 1999, 11, 953–957. (25) Smela, E. J. Micromech. Microeng. 1999, 9, 1–18. (26) Smela, E.; Gadegaard, N. J. Phys. Chem. B 2001, 105, 9395– 9405. (27) Smela, E. Adv. Mater. 2003, 15, 481–494. (28) Masuda, H.; Yada, K.; Osaka, A. Jpn. J. Appl. Phys., Part 2 1998, 37, L1340–L1342. (29) Li, A.-P.; Muller, F.; Birner, A.; Nielsch, K.; Gosele, U. Adv. Mater. 1999, 11, 483–487. (30) Grimm, S.; Giesa, R.; Sklarek, K.; Langner, A.; Go sele, U.; Schmidt, H. W.; Steinhart, M. Nano Lett. 2008, 8, 1954–1959. (31) Lim, J. K.; Ciszek, J. W.; Huo, F.; Jang, J. W.; Hwang, S.; Mirkin, C. A. Nano Lett. 2008, 8, 4441–4445. (32) Byun, J.; Kim, N.-H.; Lee, D. H.; Lee, K.-H.; Kim, J. K. Soft Matter 2009, 5, 3835–3838. (33) Byun, J.; Lee, J. I.; Kwon, S.; Jeon, G.; Kim, J. K. Adv. Mater. 2010, 22, 2028–2032.

1288

dx.doi.org/10.1021/nl104329y |Nano Lett. 2011, 11, 1284–1288