J Mater Sci: Mater Med (2015) 26:117 DOI 10.1007/s10856-015-5444-0
DELIVERY SYSTEMS
Controlling antibiotic release from mesoporous silica nano drug carriers via self-assembled polyelectrolyte coating Tasnuva Tamanna • Jurgen B. Bulitta Aimin Yu
•
Received: 2 August 2014 / Accepted: 21 November 2014 Ó Springer Science+Business Media New York 2015
Abstract Mesoporous silica nanoparticles (MSNs) have been explored as controlled drug delivery systems since the early 2000s, but many fundamental questions remain for this important application. We sought to design a pH controlled delivery system of gentamicin, an aminoglycoside antibiotic, based on MSNs. Under optimal conditions, MSN was able to load 219 lg gentamicin per mg MSNs. Polymeric networks encompassing gentamicin loaded MSNs were then established to tune the release kinetics. Embedding of drug pre-loaded MSNs was performed by an efficient layer-by-layer (LbL) self-assemble strategy using polystyrene sulfonate (PSS) and poly (allylamine hydrochloride) (PAH). We characterised the release kinetics by nonlinear mixed-effects modelling in the S-ADAPT software. The mean release time from uncoated MSNs was 3.6 days at pH 7.4 and 0.4 days at pH 1.4. A further slower release was achieved by diffusion through one or two PSS/PAH bilayer(s) which had a mean transit time of 6.0 days at pH 7.4 and 3.5 days at pH 1.4. The number of bilayers affected the shape of the release profile. The developed nano-drug carriers combined with the self-assembled polyelectrolyte coating allowed us to tune the release kinetics by pH and the number of bilayers.
T. Tamanna A. Yu (&) Faculty of Science, Engineering and Technology, Swinburne University of Technology, Melbourne, Australia e-mail: [email protected] J. B. Bulitta Faculty of Pharmacy and Pharmaceutical Sciences, Monash University, Melbourne, Australia
1 Introduction Controlling the kinetics of drug release from pharmaceutical formulations and surface coatings is important to achieve the targeted drug effects. Delivery systems with a high drug loading capacity and controllable release kinetics are promising for many applications [1, 2] such as antibiotic coated thin films that protect implanted devices from bacterial infections. To provide a protective effect, these thin films should maintain effective and safe antibiotic concentrations over a prolonged period of time and without an early burst release [3, 4]. Various types of nanoparticles were explored as drug carriers for antibiotics and other drugs [5]. Both organic and inorganic nanostructures were developed for different biomedical applications [1, 2]. Among all nano-carriers, mesoporous silica nanoparticles (MSNs) are widely used due to favourable inherent properties [6]. The ‘unique’ textural features of MSNs include high surface area, high pore volume, a homogenously ordered pore network, and silanol-containing surface. The latter provides a platform for surface-functionalization to maximize drug loading and control the release kinetics depending on the physicochemical properties of the loaded drug(s) [7–11]. Controlled drug delivery from MSNs has been developed in response to diverse stimuli such as pH, magnetic field, enzyme, light, or temperature [12–18]. Lui et al. developed a pH-responsive polymer coating based on poly(4-vinyl pyridine) (PVP) [19] and loaded MSNs using a dye as a model drug. This polymer provided a pH sensitive swelling and de-swelling mechanism to control the release kinetics. The PVP polymer usually undergoes shrinkage at pH 7.4 which inhibits drug release. However, decrease in pH led to the ionization of the pyridine group of PVP and promoted swelling of the polymer
123
117
Page 2 of 7
J Mater Sci: Mater Med (2015) 26:117
coating. Popat et al. used chitosan as pH sensitive coating material and developed polymer coated MSNs as nanocarriers for the controlled release of ibuprofen [20]. The present study aims to regulate the release of gentamicin from MSNs. Gentamicin is an aminoglycoside antibiotic which has rapid bacterial killing effect on most gram-positive and gram-negative bacterial pathogens that commonly cause device-related infections such as Staphylococcus aureus, S. epidermidis, Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis, and Pseudomonas aeruginosa [21]. Gentamicin also possesses good chemical stability and gets solubilized easily in aqueous medium. The five amino groups on its structure (Fig. 1) could undergo protonation to bring the positive charges to the molecule which could facilitate the drug attachment onto negatively charged silica via electrostatic interaction. We firstly optimize the loading capacity of gentamicin in the drug carrier via adjusting the pH of the drug loading medium. We then coat the MSNs with a multilayer film of polyelectrolytes namely sodium polystyrene sulfonate (PSS) and poly (allylamine hydrochloride) (PAH) (Fig. 1). The tailored kinetics of gentamicin release from MSNs by the swelling and de-swelling mechanism of the polymeric network are characterised by monitoring the release profiles via time-course modelling.
9.2 were prepared by mixing 0.1 M potassium dihydrogen phosphate and 0.1 M dipotassium hydrogen phosphate and the final pH was adjusted by using 1 M hydhochloric acid (HCl) or 1 M sodium hydroxide (NaOH). Ultrapure water was obtained from a Millipore water purification system (Milli-Q) and used throughout all experiments. Simulated body fluid (SBF, pH 7.4) was prepared according to the Ref. [22]. 2.2 Synthesis of MSNs
2 Experimental part
MSNs with an average pore diameter of *2.8 nm were synthesised based on the modified Sto¨ber method [23]. This synthesis used a sol–gel process with a binary surfactant system including triblock copolymer pluronic F127 and CTAB. The CTAB and pluronic F127 were used as templating and co-templating agents and TEOS served as the silica precursor. Briefly, 0.5 g CTAB and 2.05 g F127 were dissolved in 96 mL Milli Q water along with 43.1 mL ethanol (EtOH), and 11.2 mL concentrated ammonium hydroxide solution (28 wt%) at room temperature. After obtaining a homogenous solution, 1.93 mL of TEOS was added quickly into the mixture under vigorous stirring which was continued for 1.5 min. For further condensation, the mixture was kept static for next 24 h at room temperature. Then, the white precipitate was filtered followed by three times washing with water. Finally, MSNs were obtained by burning off surfactants via calcination at 550° for 6 h.
2.1 Materials and reagents
2.3 Drug loading of MSNs
Tetraethylorthosilicate (TEOS), cetyltrimethyl-ammonium bromide (CTAB), triblock copolymer F127, ammonium hydroxide (28 % NH3 basis), PSS (average Mw *70,000), and PAH (average Mw *58,000) were purchased from Sigma-Aldrich. Gentamicin sulphate (active fraction: 590 lg gentamicin/mg) was obtained from AK Scientific Inc. Phosphate buffer solutions (PBS) of pH 5.0, 7.4 and
Loading of MSNs with gentamicin was carried out by direct impregnation (‘soaking’). Drug loading was performed at different pHs to maximize the extent of loading. MSNs (60 mg) were dispersed in 1 mL of 20 mg/mL gentamicin in PBS at pH 9.2/7.4/5.0 or milli Q water (pH *5.8). The extent of drug loading was determined by measuring the absorbance of gentamicin in the medium before and after
HO H3C HN H3C
O OH
H2N HO O H2N
O
O NH2
Gentamicin
CH3
CH3 n
H3C
NH2
123
H3C
n NH2 PAH
O S O ONa PSS
Fig. 1 Chemical structures of gentamicin, PSS and PAH
CH3
J Mater Sci: Mater Med (2015) 26:117
immobilization via UV-Vis spectrophotometry at 247 nm. We observed no degradation of gentamicin over 15 days in PBS. The loading efficiency was calculated as mass (lg) of drug loaded per mg of MSNs. 2.4 Coating of polymer over MSN-G via LbL technique Gentamicin loaded MSN (MSN-G) were coated with polyelectrolyte multilayer films via the LbL self-assembly technique. We coated MSN-G with either one bilayer [MSN-G-/(PSS/PAH)] and two bilayers [MSN-G-/(PSS/ PAH)2]. For this coating procedure, 1 mL of 5 mg/mL PSS solution (prepared in 0.5 M NaCl water solution) was added into 15 mg MSN-G and mixed for 5 min. Excess polyelectrolyte was removed by three times washing with water. The PAH layer was adsorbed by the same procedure as for PSS. For MSN-G-/(PSS/PAH)2, this process was repeated to obtain two bilayers of PSS/PAH. 2.5 Drug release from MSN-G Gentamicin release studies were carried out in SBF at 37 °C. Initially, release profiles of uncoated MSN-G were studied at different SBF volumes (1, 2 and 5 mL) at pH 7.4 to assess the impact of the concentration gradient as driving force for drug release. Then the release kinetics of polyelectrolyte coated MSN-G in 1 mL SBF were studied at both physiologic pH 7.4 and gastric pH 1.4 to assess the pH controlled release. For each release study, liquid supernatant (0.5 mL after centrifugation) was extracted at multiple time points to measure the gentamicin concentration via a UV–Vis assay. The sample was subsequently added again to the release medium to maintain a constant volume. 2.6 Material characterization
Page 3 of 7
117
PAH bilayers were simultaneously modelled. Estimation was performed via nonlinear mixed-effects modelling in the S-ADAPT software [24] supported by the SADAPTTRAN facilitator tool [25]. Standard estimation settings as implemented in SADAPT-TRAN were utilized [26]. We used a standard additive plus proportional residual error model to fit the fractions of gentamicin released. Models for uncoated and coated MSNs contained up to four compartments for the amount of gentamicin in the respective system component (Fig. 2). The first compartment represented gentamicin in MSNs. Release from MSN-G was considered to follow either firstorder or Michaelis–Menten kinetics. The differential equation for the amount of gentamicin in MSN-G (AMSN-G) released via a first-order release rate constant (krel) was: dAMSNG ¼ krel AMSNG dt CSBF 1 Initial condition ðICÞ CSol : A0
ð1Þ
The CSBF represents the concentration of gentamicin in SBF and was calculated as the amount of gentamicin in SBF (ASBF) divided by the experimental SBF volume. The CSol is the solubility limit of gentamicin in SBF and A0 the total amount of loaded gentamicin in MSN-G. The differential equation for the amount of gentamicin (ASBF) in SBF was: dASBF CSBF ¼ krel AMSNG 1 IC : 0 ð2Þ dt CSol For MSN-G coated with one bilayer [MSN-G-/(PSS/ PAH)], the differential equations for AMSN-G, the amount of gentamicin within the single bilayer (A1BL) and ASBF were: dAMSNG ¼ krel AMSNG dt
IC : A0
ð3Þ
Scanning electronic microscope (SEM) images were obtained on a field emission scanning electron microscope (FeSEM, ZEISS SUPRA 40VP, Germany) at an acceleration voltage of 3 kV. The UV–Vis absorbance spectra were measured with a Halo RB-10 UV–Vis spectrophotometer (Australia). FTIR was employed to confirm the successful entrapment of gentamicin in MSN-G. The FTIR spectra were recorded at a Thermo Scientific Nicolet iD5 spectrometer (USA). The zeta potential of the particles was determined by a particle size analyzer (Brookhaven 90Plus Nanoparticle Size Analyzer, USA). 2.7 Modelling of gentamicin release kinetics The gentamicin release profiles at pH 7.4 and 1.4 from uncoated MSNs and MSNs embedded into one or two PSS/
Fig. 2 Model structure for the release of gentamicin from uncoated and coated MSN-G
123
117
Page 4 of 7
J Mater Sci: Mater Med (2015) 26:117
dABilayer1 CSBF = krel AMSNG - k1BL ABilayer1 1 dt CSol IC : 0 ð4Þ dASBF dt
CSBF = k1BL ABilayer1 1 SSol
IC : 0
ð5Þ
For MSN-G coated with two bilayers [MSN-G-/(PSS/ PAH)2], one differential equation each described the amount of gentamicin in bilayer1 (ABilayer1) and bilayer 2 (ABilayer2): dAMSNG ¼ - krel AMSNG IC : A0 dt dABilayer1 ¼ krel AMSNG k2BL ABilayer1 IC : 0 dt dABilayer2 ¼ k2BL ABilayer1 k2BL ABilayer2 dt CSBF 1 IC CSol : 0 dASBF CSBF ¼ k2BL ABilayer2 1 IC : 0 dt CSol
ð6Þ ð7Þ
ð8Þ ð9Þ
Depending on the type of uncoated or coated MSN-G, the respective set of differential equations was applied.
3 Results and discussion 3.1 Drug loading in MSNs MSNs with an average pore diameter of *2.8 nm were synthesised via a sol–gel process using a binary surfactant system including triblock copolymer pluronic F127 and CTAB. The SEM image of the as synthesized MSNs (Fig. 3a) showed the particles have a uniform spherical shape with an average diameter of 192 nm (Fig. 3b). This synthesis method was reproducible and efficient. Fig. 3 SEM image of (a) MSNs and (b) the particle size distribution curve of the as synthesized MSNs
123
When immersed in water, silica surfaces acquire a negative surface charge density, primarily through the dissociation of terminal silanol groups. The negative surface charge increases with increasing pH [27, 28]. On the other hand, the pKa of five amino groups of gentamicin are in the range of 6.3–9.6 [29]. The amino groups could ionise to introduce the positively charge to the drug molecule when the pH is below 9.6. The positively charged gentamicin could be directly adsorbed onto negatively charged MSNs via electrostatic interactions. As the pH affects the surface charge density of both gentamicin and MSNs, the pH of the loading medium was studied to optimise the gentamicin loading. As shown in Fig. 4, pH values of 7.4 and 9.2 yielded a much higher loading amount of gentamicin compared to lower pHs. Because the initial drug attachment was via the electrostatic interaction with the silica surface, the greater amount of drug loading at higher pH was likely caused by a higher negative charge density of silica at increasing pHs which facilitated the electrostatic interactions with positively charged gentamicin. The maximum loading of gentamicin was 219 lg per mg MSNs at pH 9.2 and 211 lg per mg MSNs at pH 7.4. At both pHs, the maximum loading was achieved within 6 h. These loading amounts are relatively higher than the other types of mesoporous silica nano drug carriers with smaller pore sizes, such as MCM-41 and MCM-48 which had a gentamicin loading of 79.4–120 lg per mg of silica [30]. Our achieved loading of 219 lg gentamicin/mg MSNs is approximately ten-fold higher than gentamicin loading on PLGA nanoparticles (22.4 lg/mg PLGA) [31]. Entrapment of gentamicin in MSNs was further confirmed by FTIR analysis of bare MSN and MSN-G. The bare MSNs showed an intense peak at 1,046 cm¯ 1 due to the vibration of the siloxane (Si–O–Si) bond (Fig. 5a). Gentamicin possesses five amino groups, three methyl groups and three hydroxyl groups (Fig. 1). In Fig. 5b, a strong spike was observed at 1,055 cm-1 which can be due to the C–N stretching vibration of amino groups of gentamicin. Bending of the N–H bond provided a peak at 1,628 cm-1
Drug loading (µg gentamicin/mg MSNs)
J Mater Sci: Mater Med (2015) 26:117
240
Page 5 of 7
pH 9.2 pH 7.4
180
120
60
pH 5.8 pH 5.0
0 0
1
2
3
Time (day)
Fig. 4 Gentamicin loading efficiency at different pH
a
Transmittance (%)
b
1450 1628 C− H
3326
N− H
O−H Si − O − Si
3000
2500
gradient was the principal driving force for gentamicin release from MSN-G. For uncoated MSN-G at pH 1.4, approximately 70 % of gentamicin was released after 24 h and almost 100 % at 48 h (Fig. 6, top). While the release was much slower at pH 7.4 (e.g., being 48 % at the third day, Fig. 6, bottom), we sought to control the rapid release rate by coating MSNG with a protecting polymer film to achieve a sustained release over longer times. Uniform coating of polyelectrolyte multilayer film over MSN-G was performed using the layer-by-layer (LbL) self-assembly technique. PSS is a strong polyelectrolyte and pH does not affect its ionisation and dissolution in the whole pH range of 1–14. However, PAH is a weak polyelectrolyte; therefore pH does influence its ionization and dissolution. The pKa of PAH is about pH 10 [32]. When pH is below 10, the amino groups of PAH ionised to bring the positive charge to the polymer chain; while above pH 9, PAH has very little charge. During the coating of polyelectrolyte bilayer(s) over MSN-G, polymeric solutions (either PSS or PAH) were prepared in
1046
C− N 3500
117
1055
2000
1500
1000
−1
Wavenumber (cm )
Fig. 5 FTIR spectra of (a) MSNs and (b) MSN-G
for MSN-G (Fig. 5b) but not for MSN (Fig. 5a). Both MSNs and gentamicin contain hydroxyl groups, a broad band at 3,550–3,200 cm-1 (Fig. 5b) was prominent for MSN-G but not for MSN. This band may have resulted from the stretching vibration of O–H. Moreover, a small signal at 1,450 cm-1 represented the bending vibration of the C–H bonds from methyl groups in gentamicin. These results confirmed the presence and entrapment of gentamicin in the MSN-G. 3.2 Drug release from uncoated and coated MSNs The release behaviour of gentamicin from uncoated MSN in the simulated body fluid (SBF, pH 7.4) was examined and the amount of drug released was monitored by UV–Vis absorbance. This was performed by adding the gentamicin loaded MSN into SBF and the mixture was incubated at 37 °C. Almost 100 % of gentamicin was released in 5 mL SBF at 5 days, whereas 70.4 % and 42.2 % of gentamicin were released in 2 and 1 mL SBF respectively at 8 days. This suggested that the release of gentamicin became saturated for low SBF volumes and that the concentration
Fig. 6 Observed and fitted gentamicin release from MSN-G, MSNG-/(PSS/PAH), MSN-G-/(PSS/PAH)2 at pH 1.4 (top) and 7.4 (bottom)
123
117
Page 6 of 7
J Mater Sci: Mater Med (2015) 26:117
Table 1 Mean parameter estimates (relative standard errors) for the release kinetics of gentamicin from uncoated and coated MSN Parameter
Symbol
Unit
pH 1.4
pH 7.4
Mean release time from MSN-G into SBF or into the first bilayer
MRTaRel
day
0.400 (25.6 %)
3.62 (2.85 %)
Mean transit time through a single bilayer
MTTa1BL
day
3.25 (7.13 %)
6.32 (18.2 %)
Mean transit time through two bilayers (i.e. total transit time through both bilayers)
MTTa2BL
day
3.77 (4.71 %)
5.58 (5.87 %)
Solubility of gentamicin in SBF
CSol
mg/mL
4.75 (11.9 %)
3.38 (13.4 %)
a
The first-order rate constant krel was calculated as 1/MRTRel and the k1BL was calculated as 1/MTT1BL. To reflect the two subsequent first-order transfer processes through the two bilayers, the k2BLwas calculated as 2/MTT2BL
0.5 M NaCl at neutral pH. Hence PSS is negatively charged while PAH is positively charged, leading to the electrostatic interaction between oppositely charged polymers to form a bilayer. The coating process was monitored by measuring the Zeta potentials of the particles which were -18.71 mV for MSN, -1.12 mV for MSN-G, and 3.44 mV for MSN-G-/(PSS/PAH)2. The loading of positively charged gentamicin (MSN-G) reduces the negative surface charge of MSN. Further coating of the positively charged PAH over MSN-G resulted in a positive zeta potential value of MSN-G-/(PSS/PAH)2. During the LbL polyelectrolyte coating, approximate 3 % gentamicin was found to desorb from MSNs. The majority lost occurred during the first polyelectrolyte (PSS) layer coating around MSN-G. However, the coating of MSN-G did yield steady and slow release of gentamicin both at pH 1.4 (Fig. 6, top) and 7.4 (Fig. 6, bottom). For example, the total release amount of gentamicin after three days incubation at physiologic pH was much less than uncoated MSN-G, from 48 % down to 9.35 % for MSNG-/(PSS/PAH) and 2.03 % for MSN-G-/(PSS/PAH)2. After the slow initial release phase for the MSN-G-/(PSS/PAH)2, the release profiles of both coated MSNs was comparable. This showed that the second bilayer slowed down the release rate during the first 2–3 days, but not thereafter. Overall, these data demonstrate that the release rate could be effectively controlled by pH and coating of MSNs. Time-course modelling estimated a mean release time from MSN-G of 0.4 day at pH 1.4 and 3.6 days at pH 7.4 (Table 1). This release time applies to both coated and uncoated MSN-G (Fig. 2); however, the coating caused an additional delay for the appearance of gentamicin in the medium. The mean transit time through one or two bilayers was 3.2 or 3.8 days at pH 1.4 and 6.3 or 5.6 days at pH 7.4. This shows that the transfer of gentamicin through the bilayer(s) was slower than the drug release from MSN-G and suggests that the overall rate of drug release could be effectively controlled by optimising the coating and pH. While the data and associated modelling showed a much slower release at pH 7.4 compared to pH 1.4, it is important
123
to understand the mechanisms for this pH dependent behaviour. The PSS/PAH multilayer was assembled in a neutral pH condition. When exposed to a strong acidic solution (pH 1.4), there were likely changes in the charge density, leading to the swelling of the polyelectrolyte layer or reversible pore formation. These changes greatly increased the film permeability to allow the transfer of gentamicin across the PSS/PAH bilayers [23, 33]. While at pH 7.4, the pH condition was close to the film formation. The induced change on the permeability of the film was much less than at pH 1.4. Thus the overall release of gentamicin was slow. Overall, this yielded a pH-responsive and controllable release of gentamicin due to the swelling and deswelling mechanism encompassing a polymeric network of PSS/PAH bilayer(s) around the drug loaded MSN-G.
4 Conclusion A high gentamicin loading capacity of 211 lg per mg MSNs was achieved at pH 7.4 condition. The drug release kinetics from MSNs were controllable via varying the pH of the release media and the coating of MSN-G with PSS/ PAH bilayers which resulted in a slow release rate without initial burst release. Changes in pH caused swelling and deswelling of PSS/PAH bilayers and therefore affected the rate of gentamicin transfer through the coating. As PSS and PAH polyelectrolytes are economic, this strategy to control the release kinetics is cost effective. The sustained release over more than 7 days at physiological pH 7.4 is highly promising for future applications to coat biomedical device surfaces such as pacemakers and other implanted devices. In a broader context, this technique can be applied to other fields such as agriculture, cosmetics, personal care, and food science to provide a high loading capacity and controlled release of compounds from thin films. Acknowledgments J. B. B. acknowledged the Australian Research Council for his DECRA Fellowship (DE120103084).
J Mater Sci: Mater Med (2015) 26:117
Page 7 of 7
References 1. Carbone C, Campisi A, Musumeci T, Raciti G, Bonfanti R, Puglisi G. FA-loaded lipid drug delivery systems: preparation, characterization and biological studies. Eur J Pharm Sci. 2014;52(1):12–20. 2. Ariga K, Ji Q, McShane MJ, Lvov YM, Vinu A, Hill JP. Inorganic nanoarchitectonics for biological applications. Chem Mater. 2012;24(5):728–37. 3. Hasan AS, Socha M, Lamprecht A, Ghazouani FE, Sapin A, Hoffman M, et al. Effect of the microencapsulation of nanoparticles on the reduction of burst release. Int J Pharm. 2007;344(1–2):53–61. 4. Patel BK, Parikh RH, Aboti PS. Development of oral sustained release rifampicin loaded chitosan nanoparticles by design of experiment. J Drug Deliv. 2013;2013:370938. 5. Zhang L, Pornpattananangkul D, Hu CMJ, Huang CM. Development of nanoparticles for antimicrobial drug delivery. Curr Med Chem. 2010;17(6):585–94. 6. Kwon S, Singh RK, Perez RA, Abou Neel EA, Kim H-W, Chrzanowski W. Silica-based mesoporous nanoparticles for controlled drug delivery. J Tissue Eng. 2013;4(1):1–18. 7. Vallet-Regı´ M, Balas F, Arcos D. Mesoporous materials for drug delivery. Angew Chem Int Ed. 2007;46(40):7548–58. 8. Trewyn BG, Slowing II, Giri S, Chen HT, Lin VSY. Synthesis and functionalization of a mesoporous silica nanoparticle based on the sol-gel process and applications in controlled release. Acc Chem Res. 2007;40(9):846–53. 9. Xu W, Gao Q, Xu Y, Wu D, Sun Y, Shen W, et al. Controllable release of ibuprofen from size-adjustable and surface hydrophobic mesoporous silica spheres. Powder Technol. 2009;191(1–2):13–20. 10. Izquierdo-Barba I, Colilla M, Vallet-Regi M. Nanostructured mesoporous silicas for bone tissue regeneration. J Nanomater. 2008;2008:106970. 11. Strømme M, Brohede U, Atluri R, Garcia-Bennett AE. Mesoporous silica-based nanomaterials for drug delivery: evaluation of structural properties associated with release rate. Wiley Interdiscip Rev. 2009;1(1):140–8. 12. Sun JT, Hong CY, Pan CY. Fabrication of PDEAEMA-coated mesoporous silica nanoparticles and pH-responsive controlled release. J Phys Chem C. 2010;114(29):12481–6. 13. Hu X, Wang Y, Peng B. Chitosan-capped mesoporous silica nanoparticles as pH-responsive nanocarriers for controlled drug release. Chem Asian J. 2014;9(1):319–27. 14. Yanes RE, Tamanoi F. Development of mesoporous silica nanomaterials as a vehicle for anticancer drug delivery. Therapeutic Deliv. 2012;3(3):389–404. 15. Chen Z, Li Z, Lin Y, Yin M, Ren J, Qu X. Bioresponsive hyaluronic acid-capped mesoporous silica nanoparticles for targeted drug delivery. Chemistry. 2013;19(5):1778–83. 16. Nadrah P, Maver U, Jemec A, Tisˇler T, Bele M, Drazˇic´ G, et al. Hindered disulfide bonds to regulate release rate of model drug from mesoporous silica. ACS Appl Mater Interfaces. 2013;5(9):3908–15. 17. Yan H, Teh C, Sreejith S, Zhu L, Kwok A, Fang W, et al. Functional mesoporous silica nanoparticles for photothermal-
18.
19.
20.
21. 22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
117
controlled drug delivery in vivo. Angew Chem Int Ed. 2012;51(33):8373–7. Aznar E, Mondrago˜n L, Ros-Lis JV, Sanceno˜n F, Marcos MD, Martı´nez-Ma´n˜ez R, et al. Finely tuned temperature-controlled cargo release using paraffin-capped mesoporous silica nanoparticles. Angew Chem Int Ed. 2011;50(47):11172–5. Liu R, Liao P, Liu J, Feng P. Responsive polymer-coated mesoporous silica as a pH-sensitive nanocarrier for controlled release. Langmuir. 2011;27(6):3095–9. Popat A, Liu J, Lu GQ, Qiao SZ. A pH-responsive drug delivery system based on chitosan coated mesoporous silica nanoparticles. J Mater Chem. 2012;22(22):11173–8. Donlan RM. Biofilms and device-associated infections. Emerg Infect Dis. 2001;7(2):277. Marques MR, Loebenberg R, Almukainzi M. Simulated biological fluids with possible application in dissolution testing. Dissolution Technol. 2011;18(3):15–28. Kim TW, Chung PW, Lin VSY. Facile synthesis of monodisperse spherical MCM-48 mesoporous silica nanoparticles with controlled particle size. Chem Mater. 2010;22(17):5093–104. Bauer RJ, Guzy S, Ng C. A survey of population analysis methods and software for complex pharmacokinetic and pharmacodynamic models with examples. AAPS J. 2007;9(1):7. Bulitta JB, Bingo¨lbali A, Shin BS, Landersdorfer CB. Development of a new pre-and post-processing tool (SADAPT-TRAN) for nonlinear mixed-effects modeling in S-ADAPT. AAPS J. 2011;13(2):201–11. Bulitta JB, Landersdorfer CB. Performance and robustness of the Monte Carlo importance sampling algorithm using parallelized S-ADAPT for basic and complex mechanistic models. AAPS J. 2011;13(2):212–26. Liu Y, Tourbin M, Lachaize S, Guiraud P. Silica nanoparticle separation from water by aggregation with AlCl3. Ind Eng Chem Res. 2011;51(4):1853–63. Schwarz S, Lunkwitz K, Keßler B, Spiegler U, Killmann E, Jaeger W. Adsorption and stability of colloidal silica. Colloids Surf A. 2000;163(1):17–27. Lee HY, Choi KH, Bank N. Protective effect of urinary alkalinization on gentamicin nephrotoxicity in rats. Yonsei Med J. 1988;29(3):225–32. Zhu YF, Shi JL, Li YS, Chen HR, Shen WH, Dong XP. Hollow mesoporous spheres with cubic pore network as a potential carrier for drug storage and its in vitro release kinetics. J Mater Res. 2005;20(1):54–61. Abdelghany SM, Quinn DJ, Ingram RJ, Gilmore BF, Donnelly RF, Taggart CC, et al. Gentamicin-loaded nanoparticles show improved antimicrobial effects towards Pseudomonas aeruginosa infection. Int J Nanomed. 2012;7:4053–63. Lewis SR, Datta S, Gui M, Coker EL, Huggins FE, Daunert S, et al. Reactive nanostructured membranes for water purification. Proc Natl Acad Sci. 2011;108(21):8577–82. De Geest BG, Sanders NN, Sukhorukov GB, Demeester J, De Smedt SC. Release mechanisms for polyelectrolyte capsules. Chem Soc Rev. 2007;36(4):636–49.
123
DELIVERY SYSTEMS
Controlling antibiotic release from mesoporous silica nano drug carriers via self-assembled polyelectrolyte coating Tasnuva Tamanna • Jurgen B. Bulitta Aimin Yu
•
Received: 2 August 2014 / Accepted: 21 November 2014 Ó Springer Science+Business Media New York 2015
Abstract Mesoporous silica nanoparticles (MSNs) have been explored as controlled drug delivery systems since the early 2000s, but many fundamental questions remain for this important application. We sought to design a pH controlled delivery system of gentamicin, an aminoglycoside antibiotic, based on MSNs. Under optimal conditions, MSN was able to load 219 lg gentamicin per mg MSNs. Polymeric networks encompassing gentamicin loaded MSNs were then established to tune the release kinetics. Embedding of drug pre-loaded MSNs was performed by an efficient layer-by-layer (LbL) self-assemble strategy using polystyrene sulfonate (PSS) and poly (allylamine hydrochloride) (PAH). We characterised the release kinetics by nonlinear mixed-effects modelling in the S-ADAPT software. The mean release time from uncoated MSNs was 3.6 days at pH 7.4 and 0.4 days at pH 1.4. A further slower release was achieved by diffusion through one or two PSS/PAH bilayer(s) which had a mean transit time of 6.0 days at pH 7.4 and 3.5 days at pH 1.4. The number of bilayers affected the shape of the release profile. The developed nano-drug carriers combined with the self-assembled polyelectrolyte coating allowed us to tune the release kinetics by pH and the number of bilayers.
T. Tamanna A. Yu (&) Faculty of Science, Engineering and Technology, Swinburne University of Technology, Melbourne, Australia e-mail: [email protected] J. B. Bulitta Faculty of Pharmacy and Pharmaceutical Sciences, Monash University, Melbourne, Australia
1 Introduction Controlling the kinetics of drug release from pharmaceutical formulations and surface coatings is important to achieve the targeted drug effects. Delivery systems with a high drug loading capacity and controllable release kinetics are promising for many applications [1, 2] such as antibiotic coated thin films that protect implanted devices from bacterial infections. To provide a protective effect, these thin films should maintain effective and safe antibiotic concentrations over a prolonged period of time and without an early burst release [3, 4]. Various types of nanoparticles were explored as drug carriers for antibiotics and other drugs [5]. Both organic and inorganic nanostructures were developed for different biomedical applications [1, 2]. Among all nano-carriers, mesoporous silica nanoparticles (MSNs) are widely used due to favourable inherent properties [6]. The ‘unique’ textural features of MSNs include high surface area, high pore volume, a homogenously ordered pore network, and silanol-containing surface. The latter provides a platform for surface-functionalization to maximize drug loading and control the release kinetics depending on the physicochemical properties of the loaded drug(s) [7–11]. Controlled drug delivery from MSNs has been developed in response to diverse stimuli such as pH, magnetic field, enzyme, light, or temperature [12–18]. Lui et al. developed a pH-responsive polymer coating based on poly(4-vinyl pyridine) (PVP) [19] and loaded MSNs using a dye as a model drug. This polymer provided a pH sensitive swelling and de-swelling mechanism to control the release kinetics. The PVP polymer usually undergoes shrinkage at pH 7.4 which inhibits drug release. However, decrease in pH led to the ionization of the pyridine group of PVP and promoted swelling of the polymer
123
117
Page 2 of 7
J Mater Sci: Mater Med (2015) 26:117
coating. Popat et al. used chitosan as pH sensitive coating material and developed polymer coated MSNs as nanocarriers for the controlled release of ibuprofen [20]. The present study aims to regulate the release of gentamicin from MSNs. Gentamicin is an aminoglycoside antibiotic which has rapid bacterial killing effect on most gram-positive and gram-negative bacterial pathogens that commonly cause device-related infections such as Staphylococcus aureus, S. epidermidis, Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis, and Pseudomonas aeruginosa [21]. Gentamicin also possesses good chemical stability and gets solubilized easily in aqueous medium. The five amino groups on its structure (Fig. 1) could undergo protonation to bring the positive charges to the molecule which could facilitate the drug attachment onto negatively charged silica via electrostatic interaction. We firstly optimize the loading capacity of gentamicin in the drug carrier via adjusting the pH of the drug loading medium. We then coat the MSNs with a multilayer film of polyelectrolytes namely sodium polystyrene sulfonate (PSS) and poly (allylamine hydrochloride) (PAH) (Fig. 1). The tailored kinetics of gentamicin release from MSNs by the swelling and de-swelling mechanism of the polymeric network are characterised by monitoring the release profiles via time-course modelling.
9.2 were prepared by mixing 0.1 M potassium dihydrogen phosphate and 0.1 M dipotassium hydrogen phosphate and the final pH was adjusted by using 1 M hydhochloric acid (HCl) or 1 M sodium hydroxide (NaOH). Ultrapure water was obtained from a Millipore water purification system (Milli-Q) and used throughout all experiments. Simulated body fluid (SBF, pH 7.4) was prepared according to the Ref. [22]. 2.2 Synthesis of MSNs
2 Experimental part
MSNs with an average pore diameter of *2.8 nm were synthesised based on the modified Sto¨ber method [23]. This synthesis used a sol–gel process with a binary surfactant system including triblock copolymer pluronic F127 and CTAB. The CTAB and pluronic F127 were used as templating and co-templating agents and TEOS served as the silica precursor. Briefly, 0.5 g CTAB and 2.05 g F127 were dissolved in 96 mL Milli Q water along with 43.1 mL ethanol (EtOH), and 11.2 mL concentrated ammonium hydroxide solution (28 wt%) at room temperature. After obtaining a homogenous solution, 1.93 mL of TEOS was added quickly into the mixture under vigorous stirring which was continued for 1.5 min. For further condensation, the mixture was kept static for next 24 h at room temperature. Then, the white precipitate was filtered followed by three times washing with water. Finally, MSNs were obtained by burning off surfactants via calcination at 550° for 6 h.
2.1 Materials and reagents
2.3 Drug loading of MSNs
Tetraethylorthosilicate (TEOS), cetyltrimethyl-ammonium bromide (CTAB), triblock copolymer F127, ammonium hydroxide (28 % NH3 basis), PSS (average Mw *70,000), and PAH (average Mw *58,000) were purchased from Sigma-Aldrich. Gentamicin sulphate (active fraction: 590 lg gentamicin/mg) was obtained from AK Scientific Inc. Phosphate buffer solutions (PBS) of pH 5.0, 7.4 and
Loading of MSNs with gentamicin was carried out by direct impregnation (‘soaking’). Drug loading was performed at different pHs to maximize the extent of loading. MSNs (60 mg) were dispersed in 1 mL of 20 mg/mL gentamicin in PBS at pH 9.2/7.4/5.0 or milli Q water (pH *5.8). The extent of drug loading was determined by measuring the absorbance of gentamicin in the medium before and after
HO H3C HN H3C
O OH
H2N HO O H2N
O
O NH2
Gentamicin
CH3
CH3 n
H3C
NH2
123
H3C
n NH2 PAH
O S O ONa PSS
Fig. 1 Chemical structures of gentamicin, PSS and PAH
CH3
J Mater Sci: Mater Med (2015) 26:117
immobilization via UV-Vis spectrophotometry at 247 nm. We observed no degradation of gentamicin over 15 days in PBS. The loading efficiency was calculated as mass (lg) of drug loaded per mg of MSNs. 2.4 Coating of polymer over MSN-G via LbL technique Gentamicin loaded MSN (MSN-G) were coated with polyelectrolyte multilayer films via the LbL self-assembly technique. We coated MSN-G with either one bilayer [MSN-G-/(PSS/PAH)] and two bilayers [MSN-G-/(PSS/ PAH)2]. For this coating procedure, 1 mL of 5 mg/mL PSS solution (prepared in 0.5 M NaCl water solution) was added into 15 mg MSN-G and mixed for 5 min. Excess polyelectrolyte was removed by three times washing with water. The PAH layer was adsorbed by the same procedure as for PSS. For MSN-G-/(PSS/PAH)2, this process was repeated to obtain two bilayers of PSS/PAH. 2.5 Drug release from MSN-G Gentamicin release studies were carried out in SBF at 37 °C. Initially, release profiles of uncoated MSN-G were studied at different SBF volumes (1, 2 and 5 mL) at pH 7.4 to assess the impact of the concentration gradient as driving force for drug release. Then the release kinetics of polyelectrolyte coated MSN-G in 1 mL SBF were studied at both physiologic pH 7.4 and gastric pH 1.4 to assess the pH controlled release. For each release study, liquid supernatant (0.5 mL after centrifugation) was extracted at multiple time points to measure the gentamicin concentration via a UV–Vis assay. The sample was subsequently added again to the release medium to maintain a constant volume. 2.6 Material characterization
Page 3 of 7
117
PAH bilayers were simultaneously modelled. Estimation was performed via nonlinear mixed-effects modelling in the S-ADAPT software [24] supported by the SADAPTTRAN facilitator tool [25]. Standard estimation settings as implemented in SADAPT-TRAN were utilized [26]. We used a standard additive plus proportional residual error model to fit the fractions of gentamicin released. Models for uncoated and coated MSNs contained up to four compartments for the amount of gentamicin in the respective system component (Fig. 2). The first compartment represented gentamicin in MSNs. Release from MSN-G was considered to follow either firstorder or Michaelis–Menten kinetics. The differential equation for the amount of gentamicin in MSN-G (AMSN-G) released via a first-order release rate constant (krel) was: dAMSNG ¼ krel AMSNG dt CSBF 1 Initial condition ðICÞ CSol : A0
ð1Þ
The CSBF represents the concentration of gentamicin in SBF and was calculated as the amount of gentamicin in SBF (ASBF) divided by the experimental SBF volume. The CSol is the solubility limit of gentamicin in SBF and A0 the total amount of loaded gentamicin in MSN-G. The differential equation for the amount of gentamicin (ASBF) in SBF was: dASBF CSBF ¼ krel AMSNG 1 IC : 0 ð2Þ dt CSol For MSN-G coated with one bilayer [MSN-G-/(PSS/ PAH)], the differential equations for AMSN-G, the amount of gentamicin within the single bilayer (A1BL) and ASBF were: dAMSNG ¼ krel AMSNG dt
IC : A0
ð3Þ
Scanning electronic microscope (SEM) images were obtained on a field emission scanning electron microscope (FeSEM, ZEISS SUPRA 40VP, Germany) at an acceleration voltage of 3 kV. The UV–Vis absorbance spectra were measured with a Halo RB-10 UV–Vis spectrophotometer (Australia). FTIR was employed to confirm the successful entrapment of gentamicin in MSN-G. The FTIR spectra were recorded at a Thermo Scientific Nicolet iD5 spectrometer (USA). The zeta potential of the particles was determined by a particle size analyzer (Brookhaven 90Plus Nanoparticle Size Analyzer, USA). 2.7 Modelling of gentamicin release kinetics The gentamicin release profiles at pH 7.4 and 1.4 from uncoated MSNs and MSNs embedded into one or two PSS/
Fig. 2 Model structure for the release of gentamicin from uncoated and coated MSN-G
123
117
Page 4 of 7
J Mater Sci: Mater Med (2015) 26:117
dABilayer1 CSBF = krel AMSNG - k1BL ABilayer1 1 dt CSol IC : 0 ð4Þ dASBF dt
CSBF = k1BL ABilayer1 1 SSol
IC : 0
ð5Þ
For MSN-G coated with two bilayers [MSN-G-/(PSS/ PAH)2], one differential equation each described the amount of gentamicin in bilayer1 (ABilayer1) and bilayer 2 (ABilayer2): dAMSNG ¼ - krel AMSNG IC : A0 dt dABilayer1 ¼ krel AMSNG k2BL ABilayer1 IC : 0 dt dABilayer2 ¼ k2BL ABilayer1 k2BL ABilayer2 dt CSBF 1 IC CSol : 0 dASBF CSBF ¼ k2BL ABilayer2 1 IC : 0 dt CSol
ð6Þ ð7Þ
ð8Þ ð9Þ
Depending on the type of uncoated or coated MSN-G, the respective set of differential equations was applied.
3 Results and discussion 3.1 Drug loading in MSNs MSNs with an average pore diameter of *2.8 nm were synthesised via a sol–gel process using a binary surfactant system including triblock copolymer pluronic F127 and CTAB. The SEM image of the as synthesized MSNs (Fig. 3a) showed the particles have a uniform spherical shape with an average diameter of 192 nm (Fig. 3b). This synthesis method was reproducible and efficient. Fig. 3 SEM image of (a) MSNs and (b) the particle size distribution curve of the as synthesized MSNs
123
When immersed in water, silica surfaces acquire a negative surface charge density, primarily through the dissociation of terminal silanol groups. The negative surface charge increases with increasing pH [27, 28]. On the other hand, the pKa of five amino groups of gentamicin are in the range of 6.3–9.6 [29]. The amino groups could ionise to introduce the positively charge to the drug molecule when the pH is below 9.6. The positively charged gentamicin could be directly adsorbed onto negatively charged MSNs via electrostatic interactions. As the pH affects the surface charge density of both gentamicin and MSNs, the pH of the loading medium was studied to optimise the gentamicin loading. As shown in Fig. 4, pH values of 7.4 and 9.2 yielded a much higher loading amount of gentamicin compared to lower pHs. Because the initial drug attachment was via the electrostatic interaction with the silica surface, the greater amount of drug loading at higher pH was likely caused by a higher negative charge density of silica at increasing pHs which facilitated the electrostatic interactions with positively charged gentamicin. The maximum loading of gentamicin was 219 lg per mg MSNs at pH 9.2 and 211 lg per mg MSNs at pH 7.4. At both pHs, the maximum loading was achieved within 6 h. These loading amounts are relatively higher than the other types of mesoporous silica nano drug carriers with smaller pore sizes, such as MCM-41 and MCM-48 which had a gentamicin loading of 79.4–120 lg per mg of silica [30]. Our achieved loading of 219 lg gentamicin/mg MSNs is approximately ten-fold higher than gentamicin loading on PLGA nanoparticles (22.4 lg/mg PLGA) [31]. Entrapment of gentamicin in MSNs was further confirmed by FTIR analysis of bare MSN and MSN-G. The bare MSNs showed an intense peak at 1,046 cm¯ 1 due to the vibration of the siloxane (Si–O–Si) bond (Fig. 5a). Gentamicin possesses five amino groups, three methyl groups and three hydroxyl groups (Fig. 1). In Fig. 5b, a strong spike was observed at 1,055 cm-1 which can be due to the C–N stretching vibration of amino groups of gentamicin. Bending of the N–H bond provided a peak at 1,628 cm-1
Drug loading (µg gentamicin/mg MSNs)
J Mater Sci: Mater Med (2015) 26:117
240
Page 5 of 7
pH 9.2 pH 7.4
180
120
60
pH 5.8 pH 5.0
0 0
1
2
3
Time (day)
Fig. 4 Gentamicin loading efficiency at different pH
a
Transmittance (%)
b
1450 1628 C− H
3326
N− H
O−H Si − O − Si
3000
2500
gradient was the principal driving force for gentamicin release from MSN-G. For uncoated MSN-G at pH 1.4, approximately 70 % of gentamicin was released after 24 h and almost 100 % at 48 h (Fig. 6, top). While the release was much slower at pH 7.4 (e.g., being 48 % at the third day, Fig. 6, bottom), we sought to control the rapid release rate by coating MSNG with a protecting polymer film to achieve a sustained release over longer times. Uniform coating of polyelectrolyte multilayer film over MSN-G was performed using the layer-by-layer (LbL) self-assembly technique. PSS is a strong polyelectrolyte and pH does not affect its ionisation and dissolution in the whole pH range of 1–14. However, PAH is a weak polyelectrolyte; therefore pH does influence its ionization and dissolution. The pKa of PAH is about pH 10 [32]. When pH is below 10, the amino groups of PAH ionised to bring the positive charge to the polymer chain; while above pH 9, PAH has very little charge. During the coating of polyelectrolyte bilayer(s) over MSN-G, polymeric solutions (either PSS or PAH) were prepared in
1046
C− N 3500
117
1055
2000
1500
1000
−1
Wavenumber (cm )
Fig. 5 FTIR spectra of (a) MSNs and (b) MSN-G
for MSN-G (Fig. 5b) but not for MSN (Fig. 5a). Both MSNs and gentamicin contain hydroxyl groups, a broad band at 3,550–3,200 cm-1 (Fig. 5b) was prominent for MSN-G but not for MSN. This band may have resulted from the stretching vibration of O–H. Moreover, a small signal at 1,450 cm-1 represented the bending vibration of the C–H bonds from methyl groups in gentamicin. These results confirmed the presence and entrapment of gentamicin in the MSN-G. 3.2 Drug release from uncoated and coated MSNs The release behaviour of gentamicin from uncoated MSN in the simulated body fluid (SBF, pH 7.4) was examined and the amount of drug released was monitored by UV–Vis absorbance. This was performed by adding the gentamicin loaded MSN into SBF and the mixture was incubated at 37 °C. Almost 100 % of gentamicin was released in 5 mL SBF at 5 days, whereas 70.4 % and 42.2 % of gentamicin were released in 2 and 1 mL SBF respectively at 8 days. This suggested that the release of gentamicin became saturated for low SBF volumes and that the concentration
Fig. 6 Observed and fitted gentamicin release from MSN-G, MSNG-/(PSS/PAH), MSN-G-/(PSS/PAH)2 at pH 1.4 (top) and 7.4 (bottom)
123
117
Page 6 of 7
J Mater Sci: Mater Med (2015) 26:117
Table 1 Mean parameter estimates (relative standard errors) for the release kinetics of gentamicin from uncoated and coated MSN Parameter
Symbol
Unit
pH 1.4
pH 7.4
Mean release time from MSN-G into SBF or into the first bilayer
MRTaRel
day
0.400 (25.6 %)
3.62 (2.85 %)
Mean transit time through a single bilayer
MTTa1BL
day
3.25 (7.13 %)
6.32 (18.2 %)
Mean transit time through two bilayers (i.e. total transit time through both bilayers)
MTTa2BL
day
3.77 (4.71 %)
5.58 (5.87 %)
Solubility of gentamicin in SBF
CSol
mg/mL
4.75 (11.9 %)
3.38 (13.4 %)
a
The first-order rate constant krel was calculated as 1/MRTRel and the k1BL was calculated as 1/MTT1BL. To reflect the two subsequent first-order transfer processes through the two bilayers, the k2BLwas calculated as 2/MTT2BL
0.5 M NaCl at neutral pH. Hence PSS is negatively charged while PAH is positively charged, leading to the electrostatic interaction between oppositely charged polymers to form a bilayer. The coating process was monitored by measuring the Zeta potentials of the particles which were -18.71 mV for MSN, -1.12 mV for MSN-G, and 3.44 mV for MSN-G-/(PSS/PAH)2. The loading of positively charged gentamicin (MSN-G) reduces the negative surface charge of MSN. Further coating of the positively charged PAH over MSN-G resulted in a positive zeta potential value of MSN-G-/(PSS/PAH)2. During the LbL polyelectrolyte coating, approximate 3 % gentamicin was found to desorb from MSNs. The majority lost occurred during the first polyelectrolyte (PSS) layer coating around MSN-G. However, the coating of MSN-G did yield steady and slow release of gentamicin both at pH 1.4 (Fig. 6, top) and 7.4 (Fig. 6, bottom). For example, the total release amount of gentamicin after three days incubation at physiologic pH was much less than uncoated MSN-G, from 48 % down to 9.35 % for MSNG-/(PSS/PAH) and 2.03 % for MSN-G-/(PSS/PAH)2. After the slow initial release phase for the MSN-G-/(PSS/PAH)2, the release profiles of both coated MSNs was comparable. This showed that the second bilayer slowed down the release rate during the first 2–3 days, but not thereafter. Overall, these data demonstrate that the release rate could be effectively controlled by pH and coating of MSNs. Time-course modelling estimated a mean release time from MSN-G of 0.4 day at pH 1.4 and 3.6 days at pH 7.4 (Table 1). This release time applies to both coated and uncoated MSN-G (Fig. 2); however, the coating caused an additional delay for the appearance of gentamicin in the medium. The mean transit time through one or two bilayers was 3.2 or 3.8 days at pH 1.4 and 6.3 or 5.6 days at pH 7.4. This shows that the transfer of gentamicin through the bilayer(s) was slower than the drug release from MSN-G and suggests that the overall rate of drug release could be effectively controlled by optimising the coating and pH. While the data and associated modelling showed a much slower release at pH 7.4 compared to pH 1.4, it is important
123
to understand the mechanisms for this pH dependent behaviour. The PSS/PAH multilayer was assembled in a neutral pH condition. When exposed to a strong acidic solution (pH 1.4), there were likely changes in the charge density, leading to the swelling of the polyelectrolyte layer or reversible pore formation. These changes greatly increased the film permeability to allow the transfer of gentamicin across the PSS/PAH bilayers [23, 33]. While at pH 7.4, the pH condition was close to the film formation. The induced change on the permeability of the film was much less than at pH 1.4. Thus the overall release of gentamicin was slow. Overall, this yielded a pH-responsive and controllable release of gentamicin due to the swelling and deswelling mechanism encompassing a polymeric network of PSS/PAH bilayer(s) around the drug loaded MSN-G.
4 Conclusion A high gentamicin loading capacity of 211 lg per mg MSNs was achieved at pH 7.4 condition. The drug release kinetics from MSNs were controllable via varying the pH of the release media and the coating of MSN-G with PSS/ PAH bilayers which resulted in a slow release rate without initial burst release. Changes in pH caused swelling and deswelling of PSS/PAH bilayers and therefore affected the rate of gentamicin transfer through the coating. As PSS and PAH polyelectrolytes are economic, this strategy to control the release kinetics is cost effective. The sustained release over more than 7 days at physiological pH 7.4 is highly promising for future applications to coat biomedical device surfaces such as pacemakers and other implanted devices. In a broader context, this technique can be applied to other fields such as agriculture, cosmetics, personal care, and food science to provide a high loading capacity and controlled release of compounds from thin films. Acknowledgments J. B. B. acknowledged the Australian Research Council for his DECRA Fellowship (DE120103084).
J Mater Sci: Mater Med (2015) 26:117
Page 7 of 7
References 1. Carbone C, Campisi A, Musumeci T, Raciti G, Bonfanti R, Puglisi G. FA-loaded lipid drug delivery systems: preparation, characterization and biological studies. Eur J Pharm Sci. 2014;52(1):12–20. 2. Ariga K, Ji Q, McShane MJ, Lvov YM, Vinu A, Hill JP. Inorganic nanoarchitectonics for biological applications. Chem Mater. 2012;24(5):728–37. 3. Hasan AS, Socha M, Lamprecht A, Ghazouani FE, Sapin A, Hoffman M, et al. Effect of the microencapsulation of nanoparticles on the reduction of burst release. Int J Pharm. 2007;344(1–2):53–61. 4. Patel BK, Parikh RH, Aboti PS. Development of oral sustained release rifampicin loaded chitosan nanoparticles by design of experiment. J Drug Deliv. 2013;2013:370938. 5. Zhang L, Pornpattananangkul D, Hu CMJ, Huang CM. Development of nanoparticles for antimicrobial drug delivery. Curr Med Chem. 2010;17(6):585–94. 6. Kwon S, Singh RK, Perez RA, Abou Neel EA, Kim H-W, Chrzanowski W. Silica-based mesoporous nanoparticles for controlled drug delivery. J Tissue Eng. 2013;4(1):1–18. 7. Vallet-Regı´ M, Balas F, Arcos D. Mesoporous materials for drug delivery. Angew Chem Int Ed. 2007;46(40):7548–58. 8. Trewyn BG, Slowing II, Giri S, Chen HT, Lin VSY. Synthesis and functionalization of a mesoporous silica nanoparticle based on the sol-gel process and applications in controlled release. Acc Chem Res. 2007;40(9):846–53. 9. Xu W, Gao Q, Xu Y, Wu D, Sun Y, Shen W, et al. Controllable release of ibuprofen from size-adjustable and surface hydrophobic mesoporous silica spheres. Powder Technol. 2009;191(1–2):13–20. 10. Izquierdo-Barba I, Colilla M, Vallet-Regi M. Nanostructured mesoporous silicas for bone tissue regeneration. J Nanomater. 2008;2008:106970. 11. Strømme M, Brohede U, Atluri R, Garcia-Bennett AE. Mesoporous silica-based nanomaterials for drug delivery: evaluation of structural properties associated with release rate. Wiley Interdiscip Rev. 2009;1(1):140–8. 12. Sun JT, Hong CY, Pan CY. Fabrication of PDEAEMA-coated mesoporous silica nanoparticles and pH-responsive controlled release. J Phys Chem C. 2010;114(29):12481–6. 13. Hu X, Wang Y, Peng B. Chitosan-capped mesoporous silica nanoparticles as pH-responsive nanocarriers for controlled drug release. Chem Asian J. 2014;9(1):319–27. 14. Yanes RE, Tamanoi F. Development of mesoporous silica nanomaterials as a vehicle for anticancer drug delivery. Therapeutic Deliv. 2012;3(3):389–404. 15. Chen Z, Li Z, Lin Y, Yin M, Ren J, Qu X. Bioresponsive hyaluronic acid-capped mesoporous silica nanoparticles for targeted drug delivery. Chemistry. 2013;19(5):1778–83. 16. Nadrah P, Maver U, Jemec A, Tisˇler T, Bele M, Drazˇic´ G, et al. Hindered disulfide bonds to regulate release rate of model drug from mesoporous silica. ACS Appl Mater Interfaces. 2013;5(9):3908–15. 17. Yan H, Teh C, Sreejith S, Zhu L, Kwok A, Fang W, et al. Functional mesoporous silica nanoparticles for photothermal-
18.
19.
20.
21. 22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
117
controlled drug delivery in vivo. Angew Chem Int Ed. 2012;51(33):8373–7. Aznar E, Mondrago˜n L, Ros-Lis JV, Sanceno˜n F, Marcos MD, Martı´nez-Ma´n˜ez R, et al. Finely tuned temperature-controlled cargo release using paraffin-capped mesoporous silica nanoparticles. Angew Chem Int Ed. 2011;50(47):11172–5. Liu R, Liao P, Liu J, Feng P. Responsive polymer-coated mesoporous silica as a pH-sensitive nanocarrier for controlled release. Langmuir. 2011;27(6):3095–9. Popat A, Liu J, Lu GQ, Qiao SZ. A pH-responsive drug delivery system based on chitosan coated mesoporous silica nanoparticles. J Mater Chem. 2012;22(22):11173–8. Donlan RM. Biofilms and device-associated infections. Emerg Infect Dis. 2001;7(2):277. Marques MR, Loebenberg R, Almukainzi M. Simulated biological fluids with possible application in dissolution testing. Dissolution Technol. 2011;18(3):15–28. Kim TW, Chung PW, Lin VSY. Facile synthesis of monodisperse spherical MCM-48 mesoporous silica nanoparticles with controlled particle size. Chem Mater. 2010;22(17):5093–104. Bauer RJ, Guzy S, Ng C. A survey of population analysis methods and software for complex pharmacokinetic and pharmacodynamic models with examples. AAPS J. 2007;9(1):7. Bulitta JB, Bingo¨lbali A, Shin BS, Landersdorfer CB. Development of a new pre-and post-processing tool (SADAPT-TRAN) for nonlinear mixed-effects modeling in S-ADAPT. AAPS J. 2011;13(2):201–11. Bulitta JB, Landersdorfer CB. Performance and robustness of the Monte Carlo importance sampling algorithm using parallelized S-ADAPT for basic and complex mechanistic models. AAPS J. 2011;13(2):212–26. Liu Y, Tourbin M, Lachaize S, Guiraud P. Silica nanoparticle separation from water by aggregation with AlCl3. Ind Eng Chem Res. 2011;51(4):1853–63. Schwarz S, Lunkwitz K, Keßler B, Spiegler U, Killmann E, Jaeger W. Adsorption and stability of colloidal silica. Colloids Surf A. 2000;163(1):17–27. Lee HY, Choi KH, Bank N. Protective effect of urinary alkalinization on gentamicin nephrotoxicity in rats. Yonsei Med J. 1988;29(3):225–32. Zhu YF, Shi JL, Li YS, Chen HR, Shen WH, Dong XP. Hollow mesoporous spheres with cubic pore network as a potential carrier for drug storage and its in vitro release kinetics. J Mater Res. 2005;20(1):54–61. Abdelghany SM, Quinn DJ, Ingram RJ, Gilmore BF, Donnelly RF, Taggart CC, et al. Gentamicin-loaded nanoparticles show improved antimicrobial effects towards Pseudomonas aeruginosa infection. Int J Nanomed. 2012;7:4053–63. Lewis SR, Datta S, Gui M, Coker EL, Huggins FE, Daunert S, et al. Reactive nanostructured membranes for water purification. Proc Natl Acad Sci. 2011;108(21):8577–82. De Geest BG, Sanders NN, Sukhorukov GB, Demeester J, De Smedt SC. Release mechanisms for polyelectrolyte capsules. Chem Soc Rev. 2007;36(4):636–49.
123
