Influence of Surface Chemistry on the Release of an Antibacterial

Subscriber access provided by NEW YORK UNIV

Article

The Influence of Surface Chemistry on the Release of an Antibacterial Drug from Nanostructured Porous Silicon Mengjia Wang, Philip Hartman, Armando Loni, Leigh T. Canham, Nelli Bodiford, and Jeffery Lee Coffer Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.5b01372 • Publication Date (Web): 13 May 2015 Downloaded from http://pubs.acs.org on May 20, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The Influence of Surface Chemistry on the Release of an Antibacterial Drug from Nanostructured Porous Silicon Mengjia Wang,1 Philip S. Hartman,2 Armando Loni,3 Leigh T. Canham,3 Nelli Bodiford,1 and Jeffery L. Coffer1*

1

Department of Chemistry, Texas Christian University, Fort Worth, TX 76129

2

Department of Biology, Texas Christian University, Fort Worth, TX 76129

3

pSiMedica Ltd, Malvern Hills Science Park, Geraldine Road, Malvern, Worcestershire, WR14 3SZ, UK

Abstract Nanostructured mesoporous silicon possesses important properties advantageous to drug loading and delivery. For controlled release of the antibacterial drug triclosan, and its associated activity versus Staphylococcus aureus, previous studies investigated the influence of porosity of the silicon matrix. In this work, we focus on the complementary issue of the influence of surface chemistry on such properties, with particular regard to drug loading and release kinetics that can be ideally adjusted by surface modification. Comparison between drug release from as-anodized, hydride-terminated hydrophobic porous silicon and the oxidized hydrophilic counterpart is complicated due to the instability bio-resorption of the former; hence, a hydrophobic interface with long-term biostability is desired, such as can be provided by a relatively long chain octyl moiety. To minimize possible thermal degradation of the surfaces or drug activity during loading of molten drug species, a solution loading method has been investigated. Such studies demonstrate that the ability of porous silicon to act as an effective carrier for sustained delivery of antibacterial agents can be sensitively altered by surface functionalization.

1 Introduction Mesoporous silicon (pSi) has been shown to be a versatile platform for drug delivery,1-4 biosensing,5-7 and tissue engineering.8-9 Its fundamental ability to be

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

resorbed in physiological environments in a biocompatible manner,10-11 coupled with the option of fabrication with a tunable range of pore morphologies and surface chemistries are especially appealing in the area of drug delivery. Previous investigations have demonstrated the utility of porous silicon as a therapeutic carrier in a broad range of chemotherapeutic agents,12 analgesics,13 and antibacterial14 & antifungal15 compounds. For the latter, we have previously evaluated14 the role of silicon porosity (macroporous versus mesoporous material) on the released concentration and associated antibacterial activity of the hydrophobic drug triclosan initially loaded into pSi via a simple melt-loading process; initial comparisons between as-anodized (hydrideterminated) and thermally oxidized surfaces indicated that surface oxidation (with an associated hydrophilic character) exerted a significant diminution in terms of released triclosan concentration and associated antibacterial activity. Given the known differences in the dissolution behavior between hydride terminated and oxide-terminated pSi surfaces,16 the latter observation raises intriguing questions with regard to the dominant influence controlling drug release in this type of system: (a) carrier dissolution, or (b) differences in intermolecular interactions between a hydrophobic drug and hydrophobic surface or alternatively, a hydrophilic surface, each independent of the influence of Si dissolution. In this work, for a pSi matrix of fixed porosity, we focus on a detailed analysis of several additional properties. Significantly, the role of surface chemistry is analyzed using

hydrophobic

as-anodized

(hydride-terminated)

pSi,

hydrophobic

octyl-

functionalized pSi, and hydrophilic surface oxidized pSi. In contrast to the highlyresorbable as-anodized pSi, octyl-functionalized pSi provides a hydrophobic surface


2

Page 3 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

that resists degradation in vitro and in vivo. These different surfaces provide multiple points of comparison for analysis of the hydrophobic active triclosan and either hydrophobic or hydrophilic pSi surfaces. As described previously,14 melt methods are the most suitable loading method for drugs with good thermal stability and low aqueous solubility (such as triclosan). However, one challenge of the melt method is the uniformity of blending pSi powder with a given drug. The pSi particle size distribution, drug density and flowability, and the relative amounts used are all important variables. Melt-loading is also impractical for less thermally stable drugs. In contrast, a solution-based system is able to offer a homogenous environment for facile blending with the pSi particles. Hence, we employ a solution method in order to avoid any of the above possible complications in loading of this compound. Multiple assays and analytical methods are employed to correlate the role of the above parameters on released drug concentrations and antibacterial activity duration. For fundamental structural characterization of pSi, these include scanning (SEM) and transmission (TEM) electron microscopy, energy dispersive x-ray analysis (EDX), and FT IR vibrational spectroscopy. The extent of drug loading was evaluated quantitatively by thermogravimetric analysis (TGA), and the relative crystallinity of the loaded triclosan assessed by x-ray diffraction (XRD). For biological assays, we utilize disk diffusion assays of the released active versus Staphococcus aureus (S. aureus); released triclosan concentrations are evaluated spectrophotometrically; Si dissolution is readily quantified by established molybdate complexation methods.17-19 The combined series of


3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 21

measurements provide a detailed picture of the tunable range of antibacterial function associated with this nanostructured porous carrier.

2 Experimental Methods and Materials 2.1 Fabrication and Functionalization of pSi. pSi possessing a mesoporous morphology was prepared by electrochemical anodization of Si substrates in methanoic HF electrolyte to create 150 µm thick membranes of 81% porosity. The membranes were mechanically ground by hand and sieved to a particle size range of 38 ‒ 75 µm. For use in selected experiments, pSi particles were surface oxidized by a 60 min anneal in air at 600°C in order to convert the pSi surface from a hydrophobic to hydrophilic one. In order to create pSi particles with a relatively non-resorbable hydrophobic surface, as-anodized pSi particles (again with a size range of 38 ‒ 75 µm) were functionalized with octyl groups by a known thermal hydrosilylation method20 involving heating the pSi in a 1M 1-octene solution (in toluene) at reflux (113 °C) under argon for 20 hours, followed by centrifugation, washing, and drying the sample in vacuum overnight. The functionalized pSi materials were characterized by Fourier Transform Infrared (FT IR) Spectroscopy, in the form of KBr pellets (typically in a ratio of 1 mg pSi to 100 mg of dry KBr). 2.2 pSi Dissolution Assays. The measurement of pSi dissolution rate was achieved by spectrophotometric monitoring of silicic acid concentration. Samples were obtained by soaking 5 mg of each type of unloaded pSi (as-anodized pSi, oxidized pSi and octylfunctionalized pSi) in 1 mL of sterile de-ionized water in a microcentrifuge tube; after the tubes were shaken at 37°C for 24 h, as much as possible of the supernatant containing


4

Page 5 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

the dissolved Si was removed; a fresh 1 mL aliquot of sterile water was added to the original microcentrifuge tube to let the pSi dissolution continue. These steps were repeated in 24 h intervals up to 15 days. The concentration of silicic acid in dissolved solution was determined by the Molybdenum Blue method.17-19 2.3 Triclosan Solution Loading Method. Since triclosan has a high solubility in ethanol, and water can reduce the evaporation of ethanol and maintain the volume of solvent in a certain range under heating, 2 mL of an ethanol/water solution (1:1) was used to dissolve 1.25 g of triclosan; then 10 mg of pSi particles were added to the triclosan solution and the mixture subsequently heated in a water bath at 70 °C for 1 h, followed by centrifugation and drying in vacuum. All loaded samples were characterized by SEM, TEM, XRD, and TGA. Triclosan loading was also independently measured for the case of the octyl-functionalized pSi using an elevated temperature (65oC) extraction using ethanol for a total of 3 hrs, along with a spectrophotometric determination of total triclosan concentration. For this assay, absorbance values at 282 nm were monitored, with such values converted to triclosan concentrations via use of a standard curve generated independently from standard ethanolic solutions of known concentration (Supplementary Information). 2.4 Antibacterial and Concomitant Drug Release Assays. The activity of the triclosan released from the different pSi materials described above was evaluated as a function of time via disk diffusion assays. The aqueous supernatant into which the triclosan diffused from a given mesoporous Si powder sample was monitored continuously in 24 h intervals by soaking onto filter disks and testing for antibacterial activity. The activity was specifically evaluated by measuring the inhibition zone of bacterial growth of S.


5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 21

aureus (ATCC 25923) and the overall duration of inhibition activity for triclosan released from pSi carriers; prior to antibacterial assay, each triclosan-loaded pSi sample was sterilized via autoclaving at 120oC for 20 min; for a given triclosan-loaded mesoporous Si sample, 5 mg of loaded pSi powder was soaked in 1.0 mL of sterile water in a 1.5 mL microcentrifuge tube; the tube was continuously rotated using a LabQuake agitating apparatus at 37 °C; after 24 h, the supernatant was drawn to a new microcentrifuge tube; a 20 µL fraction was spotted onto a paper disk already placed on the top of a LBagar plate containing 106 bacteria of S. aureus for 24 h incubation at 37 °C; another fresh 1.0 mL aliquot of sterile DI water was added again into the microcentrifuge tube containing the pSi, with active agitation of the sample in the LabQuake instrument for 24 h. The above procedure was repeated until no inhibition activity was detected from the supernatant exposed to the pSi powder. It should also be pointed out here that unloaded pSi control samples did not demonstrate any antibacterial activity versus S. aureus. Quantitative evaluation of triclosan concentration released from a given porous Si matrix was achieved by spectrophotometric determination of the triclosan-containing supernatant using absorbance values at 280 nm; such values were converted to triclosan concentrations via use of a standard curve generated independently from standard aqueous solutions of known concentration (similar to that described above in section 2.3).


6

Page 7 of 21

3 Results Overall, three groups of anodized mesoporous pSi samples loaded with triclosan by a solution method are involved in this study: as-anodized pSi (pSi); oxidized pSi (OXpSi); and octyl-functionalized pSi (OTpSi). 3.1 Functional pSi Characterization by FTIR Measurements. Figure 1 shows the surface bonding characteristics of as-anodized pSi, oxidized pSi, and octylfunctionalized pSi. Figures 1a and 1b show the key vibrations of pSi samples before and after octyl-functionalization; Figure 1c represents the features associated with a surface-oxidized pSi sample, and Figure 1d, a control sample comprised of pure (>99%) 1-octene. Freshly-etched pSi (Figure 1a) shows the main absorption bands associated with vibrations of υ(Si-H) at 2114 cm-1, υ(Si-H2) at 2090 cm-1, υ(Si-O-Si) at 1095 cm-1, δ(SiH) at 630-670 cm-1, and υ (Si-Si) at 628 cm.-1 Compared with the freshly-etched pSi, a

1.2

0.65

a)

0.60

1.0

0.50

4000

3500

3000

2500

2000

1500

1000

500

c)

1.0 0.8 0.6 0.4 0.2 4000

3500

3000

2500

2000

Wavenumber, cm

1500 -1

1000

500

Absorbance

0.45

0.6

1.2

b)

0.55

0.8

Absorbance

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

0.40 0.35 4000

3500

3000

0.45 d) 0.40 0.35 0.30 0.25 0.20 0.15 0.10 4000

3500

3000

2500

2000

2500

2000

Wavenumber, cm

1500

1000

500

1500

1000

500

-1

Figure 1. FT IR spectra of (a) as-anodized pSi; (b) octyl-functionalized pSi (OTpSi); (c) oxidized pSi (OXpSi); and (d) a 1-octene control sample.


7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 21

vibration around 2800-3000 cm-1 for υ(C-H) in the spectra of the octyl-functionalized pSi samples (Figure 1b) is consistent with the presence of an alkyl chain (Figure 1d), and confirms that the octyl group is successfully attached to the pSi surface. In the spectrum of oxidized pSi (Figure 1c), the absorption by Si-O at 1100 cm-1 (Si-O-Si) and by O-H peaks (υ(OH) at 3500 cm-1) indicate the surface of the pSi has indeed been oxidized. These changes in surface functionality can also be correlated with visual changes in the relative hydrophilicity of pSi microparticles before and after surface modification. Addition of a drop of water onto the surface of an as-anodized or octylfunctionalized pSi sample results in particle repulsion to the edges, showing its hydrophobicity; in contrast, oxidized pSi is clearly hydrophilic, as the pSi particles are dispersed in a relatively uniform way across the droplet (see Supplementary Fig. 1). 3.2 pSi Characterization by TEM/EDX. Typical TEM analysis of the as-anodized pSi microparticles reveal the expected deep and ordered porous channels of a columnar nature (Supplementary Fig. 2). Covalent attachment of the octyl species tends to obscure the lattice imaging features readily apparent in the as-anodized material. Concomitant energy dispersive x-ray analysis (EDX) and associated EDX mapping (Supplementary Fig. 3) of a typical triclosan-loaded as-anodized pSi particle reveal measureable chlorine intensity, qualitatively confirming the presence of triclosan (average weight percent of chlorine in this image is 1.9%, corresponding to an approximate triclosan loading ~ 6%). It should be emphasized that this localized ev


8

Page 9 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

aluation of drug content in an individual microparticle is not expected to necessarily correlate with the macroscopic measurements provided by TGA given below. 3.3 TGA Measurements of Triclosan-loaded pSi. Figure 2 shows typical TGA measurements of selected loaded pSi samples, which includes as-anodized pSi loaded with triclosan (Figure 2a) and octyl-functionalized pSi (Figure 2b). The TGA measurements for these representative types of loaded pSi, as well as others evaluated in this study, reveal that all pSi samples have triclosan loading values in the range of

(a)

(b)

Figure 2. TGA measurements of triclosan-loaded (a) as-anodized pSi; (b) octylfunctionalized pSi (OTpSi). 42-59%, with the precise value dependent on surface chemistry (Table 1). For a given sample, the weight loss starts from around 120°C, consistent with the boiling point of triclosan, and the entire process takes about 20 minutes to evaporate the entire payload, finally reaching an equilibrium mass value. While the loading conditions for all three pSi surface types were identical, the elevated loading percentage of the oxidized pSi (59.3%) was the largest, followed by the as-anodized material (51.8%), and finally, the octyl-functionalized pSi (42.1%).


9

Langmuir

3.4 X-Ray Diffraction Analysis of Loaded pSi Structures. All loaded pSi samples were characterized by X-ray diffraction (XRD) in order to search for the possible presence of crystalline triclosan on the pSi surface. In the 20-30 degree range, crystalline Si possesses a strong peak associated with the reflection, with triclosan alone also 6000 5000

750 600 450 300 150 0

(a)

4000 3000

1000 0 20

22

24

26

28

30

Intensity

2000

Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 21

(b)

As-anodized PSi + soln-loaded triclosan

200 150 100 50

4000

0

3000

600

OXpSi + soln-loaded triclosan

450

2000

300

1000

150 0

0 20

22

24

26

28

30

OTpSi + soln-loaded triclosan 20

22

24

26

28

30

2θ

2θ

Figure 3. (a) X-ray powder diffraction patterns of crystalline triclosan (black, bottom) along with the unloaded pSi control (red, top) used in this work; (b) comparative XRD patterns of the three types of pSi surfaces loaded with triclosan. showing multiple peaks, most prominently in the 24-25 degree vicinity (Figure 3a). Prior investigations have demonstrated that infiltration of triclosan into the narrow mesoporous pSi results in the nanostructuring or amorphization of the triclosan, with a corresponding diminution and broadening of the x-ray features.14 For these high porosity samples loaded by a solution method, all x-ray patterns show no detectable features associated with crystalline triclosan, irrespective of surface functionalization (Figure 3b). Thus the solution loading process, in conjunction with the narrow confinement of the mesoporous matrix, strongly induces the formation of an amorphous or nanostructured phase for the infiltrated active drug in this case.


10

Page 11 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

3.5 Antibacterial Assays. Figure 4 illustrates the antibacterial activity (inhibition zone) as a function of time for 5 mg samples of each type of triclosan-loaded pSi, along with the concentrations of active drug released per 24 h period. All release activity was monitored up to 90 days, and the bacterial growth inhibition zones were kept in the range of 10 ‒ 15 mm. Details regarding the maximum and average concentration, total amount of triclosan released from each sample, as well as the inhibition zone sizes are compiled in Table 1.


11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 21

(a) (a)

(b)

(b)

(c)

Figure 4. Combined zone inhibition assay/triclosan concentration release assays for (a) as-anodized pSi; (b) surface-oxidized pSi (OXpSi); (c) octyl-functionalized pSi (OTpSi).


12

Page 13 of 21

3.6 Dissolution of As-anodized, Oxidized, and Octyl-functionalized pSi. Figure 5 shows the dissolution behavior of as-anodized pSi, oxidized pSi, and octyl-functionalized pSi. It is clear that as-anodized pSi yields the highest concentration of silicic acid, with an average concentration of 266 µg/mL. It should be noted that this value reflects likely supersaturation and/or silica oligomer formation and may be subject to deviation from the true value of the monomer in solution, but nevertheless shows significantly larger dissolution than those of the other Si surfaces studied here. The extent of dissolution of octyl-functionalized and oxidized pSi is similar (and significantly smaller than asanodized), but octyl-functionalized pSi is slightly higher than oxidized pSi (average level: 98 µg for octyl-functionalized pSi versus 62 µg for oxidized pSi), likely because of the incomplete surface functionalization in the octyl case (see FT IR spectra of Figs 1b, d and e). These observations are consistent with our a priori expectations, which are that oxidation and octyl-functionalization indeed decrease the rate of pSi dissolution, and the drug release from the pSi carrier will be strongly influenced consequently by the low dissolution rate.

450

conc. of silicic acid (μg/ml)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

400

350 300 250

200 150 100

50 0 1

2

3

4

5

6

7

as-‐prepared anodized As-anodized pSi PSi (45-‐75 μm)

8

days

9

10

11

12

13

14

15

octyl-‐functionalized PSi (45-‐75 μm) OTpSi

OXpSi oxidized PSi (anodized, 45-‐75 μm)

Figure 5. The degradation of as-anodized PSi, oxidized PSi and octyl-functionalized pSi in water, described by the concentration of silicic acid dissolved after each interval.


13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 21

4 Discussion In this section, the properties of the loaded pSi samples are compared first in terms of differences in drug loading for a given pSi surface functionality, followed by a focus on individual specific factors for each type of pSi studied here that affect drug delivery. Table 1 summarizes the main characterization data and highlights some key points in the release behaviors of these triclosan-loaded pSi samples. 4.1 Comparison of Triclosan Loading as a Function of pSi Surface Termination As shown in Table 1, from TGA measurements, there is a loading difference between as-anodized and oxidized pSi (51.8% vs. 59.3%). Relative to the as-anodized pSi loaded with triclosan, the loading percentage for the oxidized pSi sample is somewhat elevated due presumably to the reactivity of selective surface sites of this sample type with water present in the loading solution (presumably resulting in residual silicon hydride elimination, oxidation, and possible etching). Control experiments involving the reaction of this type of pSi surface with a methanol/water mixture at 70oC confirm that mass changes as much as 7.7% can take place in this system (with contributions from residual solvent possible). The measurably lower value for the octylfunctionalized pSi (OTpSi) presumably arises due to the lower accessible surface volume dictated by the long chain octyl moieties in this case. In any event, it should be recalled from the X-ray diffraction data that all samples have no significant triclosan features associated with the presence of significant crystalline loaded material.


14

Page 15 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Table 1. Comparison of triclosan loading and release between as-anodized, oxidized pSi (OXpSi), and octyl-functionalized pSi (OTpSi)

Sample

Triclosan loading % (TGA)

Signal of crystalline triclosan on surface (XRD)

pSi OXpSi OTpSi

51.8 (+0.5) 59.3 (+0.7) 42.1(+0.2)

None None None

Triclosan release in first 15 days

Theoretical loading of triclosan in 5 mg loaded pSi sample (mg) 2.59 2.97 2.11

Triclosan release in 90 days

Max conc. (µg/mL)

Average conc. (µg/mL)

Max inhibition zone (mm)

Total released amount (µg)

Triclosan release % (relative to theoretical loading)

pSi

4.40 (+0.17)

3.09 (+0.17)

31 (+1.2)

87

3.4

OXpSi

2.49 (+0.11)

0.89 (+0.11)

25 (+0.89)

61

2.1

OTpSi

7.83 (+0.14)

4.9 (+0.14)

27 (+0.94)

140

6.6

Sample

Note: Theoretical loading of triclosan is defined as the maximum amount of this compound present in a 5 mg sample of loaded pSi as gauged by the corresponding TGA value.

4.2 Comparison of Triclosan Release from As-Anodized pSi, Oxidized pSi, and OctylFunctionalized pSi. The most significant differences between these three pSi materials with different surface functionalities are with regard to the triclosan release behaviors. We make such comparisons in terms of the maximum triclosan concentration value (in a given 24h interval over a 15d observation period) and average concentrations (over the 15 d period) of released triclosan, with the order of release as follows: OTpSi> pSi >> OXpSi. For example, the average concentration of triclosan released from as-prepared pSi is 3.09 µg/mL, as compared with 0.89 µg/mL for oxidized pSi. One principal reason for this


15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 21

difference is presumably due to the different dissolution rates of the pSi materials. As shown in Fig. 5, the dissolution rate of as-anodized pSi is ~ 4-fold higher than that of oxidized pSi. Such a dissolution difference presumably leads ultimately to the triclosan release from these two carriers with different kinetics. However, this difference in carrier dissolution behavior does not explain the difference in release behavior between octyl-functionalized pSi (OTpSi) and asanodized pSi, as the OTpSi releases 4.9 µg/mL of triclosan on average, compared to 3.09 µg/mL for as-anodized pSi. It is possible that the flexible hydrocarbon chain for the octyl-functionalized surface can induce slightly more disorder in the loaded triclosan upon its solidification within the porous framework; a higher surface area of the amorphous/nanostructured drug will be associated with higher concentrations of released triclosan over time (in terms of both maximum and average values). For the drug carriers alone, both octyl-functionalized pSi and oxidized pSi have lower dissolution rates than as-anodized pSi, as shown in Fig. 5. The average dissolution rate of octyl-functionalized pSi is 98 µg/mL (per 24 h period), 2.7-fold lower than that of as-anodized pSi (266 µg/mL). However, recall that triclosan release from octyl-functionalized pSi is actually greater than that of the as-anodized pSi samples, quite unlike the difference between oxidized pSi and as-anodized pSi (vide supra). This indicates that the main factor affecting the triclosan release from octyl-functionalized pSi is not the dissolution rate; rather, it is suggested that the most likely factor is the type of interaction between the pSi surface and triclosan: a hydrophobic triclosan-hydrophobic pSi surface (as-anodized pSi and OTpSi) versus a hydrophobic triclosan-hydrophilic pSi surface (OXpSi). During the triclosan release, the water diffusing into the pores is


16

Page 17 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

repelled by the hydrophobic surfaces of as-anodized pSi and octyl-functionalized pSi, but wets the hydrophilic surface of oxidized pSi. Accordingly, the interaction between pSi surface and triclosan plays a leading role in the case of loaded octyl-functionalized pSi samples. As mentioned earlier, drug release is achieved by the synergistic cooperation of drug diffusion from pores and pSi degradation over time. However, the use of octyl-functionalized pSi surfaces provides a useful probe of decoupling Si dissolution effects from hydrophobic/hydrophilic surface contributions, as the alkyl species inhibit rapid pSi degradation while providing a nanostructured hydrophobic interface for the triclosan. Both hydrophobic surfaces strongly inhibit drug recrystallization during triclosan solidification. These smaller domains and corresponding higher surface area of the encapsulated drug allow it to dissolve somewhat more readily in water at such interfaces. Upon close scrutiny, it is expected that with the pSi dissolution rates observed with these systems, a higher release percentage of triclosan is anticipated. One possible reason for this low triclosan release percentage is in part associated with the triclosan (and its low aqueous solubility) likely trapped within any precipitated silica condensation products.

5 Summary In this study, three very different types of triclosan loaded pSi samples (pSi, OXpSi, and OTpSi) were prepared and analyzed in terms of release of the active antibacterial compound and its associated activity. Comparisons among these groups provide the following conclusions:


17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 21

In the comparison between as-anodized and oxidized pSi, as-anodized pSi samples always provide a higher level of triclosan release than oxidized samples, mainly a consequence of the differences in dissolution rate.

For the differences

between as-anodized pSi and octyl-functionalized pSi, triclosan release is not significantly affected by the lower dissolution rate of the octyl-functionalized pSi, but rather dominated by the hydrophobic surface of the porous matrix and its interfacial interactions. Therefore, the use of octyl-functionalized pSi surfaces provides a useful model for decoupling Si dissolution effects from hyrophobic/hydrophilic surface contributions, as the alkyl species inhibit rapid pSi degradation while providing a nanostructured hydrophobic interface for the triclosan. Overall, this work provides a comprehensive perspective on the relative interplay between drug properties, carrier surface chemistry, and carrier degradability in a loaded mesoporous matrix – in the case of a hydrophobic drug. Tzur-Balter et al. have previously examined the low level loading and release of the hydrophilic drug mitoxantrone dihydrochloride (MTX) into porous silicon grafted with relatively long chain alkyl species, with sustained long term delivery of the active species possible.21 Comparable influences in additional hydrophilic, biologic-based drugs loaded into pSi need to be investigated systematically.

Acknowledgement Financial support of this work by the NIH (award number R21EY021583) as well as the Robert A. Welch Foundation (Grant P-1212) is gratefully acknowledged.


18

Page 19 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Supplementary Information Optical images of as-anodized, oxidized, and octyl-functionalized pSi particles in the presence of a water droplet; selected TEM images of as-anodized and octylfunctionalized pSi particles; EDX analysis of Si and Cl content in pSi particles; experimental details of the molybdate-based silicic acid assays, and standard curve for spectrophotometric analysis of triclosan. This material is available free of charge via the Internet at http://pubs.acs.org.

References 1. Salonen, J.; Kaukonen, A. M.; Hirvonen, J.; Lehto, V. Mesoporous silicon in drug delivery applications J. Pharm. Sci. 2008, 97, 632– 651. 2. Anglin, E. J.; Cheng, L.; Freeman, W. R.; Sailor, M. J. Porous silicon in drug delivery devices and materials Adv. Drug Delivery Rev. 2008, 60, 1266– 1277. 3. Lehto, V-P.; Riikonen, J. Drug Loading and Characterization of Porous Silicon Materials, in Porous Silicon for Biomedical Applications, Santos, H. Ed. Cambridge: Woodhead Publishing, 2014, pp 337-355. 4. Tasciotti E, Liu X, Bhavane R, Plant K, Leonard AD, Price BK, Cheng MM, Decuzzi P, Tour JM, Robertson F, Ferrari M. Mesoporous silicon particles as a multistage delivery system for imaging and therapeutic applications. Nat. Nanotech., 2008, 3, 151 – 157. 5. Chan, S.; Fauchet, P. M.; Li, Y.; Rothberg, L. J.; Miller, B. L., Porous silicon microcavities for biosensing applications. Phys. Status Solidi A 2000, 182, 541546. 6. Orosco, M.M., Pacholski, C.; Sailor, M.J., Real-time monitoring of enzyme activity in a mesoporous silicon double layer. Nature Nanotech. 2009, 4, 255 - 258. 7. Bonanno, L; Segal, E. Nanostructured porous silicon-polymer-based hybrids: from biosensing to drug delivery, Nanomedicine, 2011, 6, 1755-1770. 8. Coffer, J. L.; Whitehead, M. A.; Nagesha, D. K.; Mukherjee, P.; Akkaraju, G.; Totolici, M.; Saffie, R. S.; Canham, L. T. Porous Silicon-Based Scaffolds for Tissue


19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 21

Engineering and Other Biomedical Applications Phys. Status Solidi A 2005, 202, 1451– 1455. 9. McInnes, S.J.P.; Voelcker, N.H. Porous Silicon-Polymer Composites for Cell Culture and Tissue Engineering Applications, Porous Silicon for Biomedical Applications, Santos, H. Ed. Cambridge: Woodhead Publishing, 2014, pp 420-469. 10. Bowditch, A. P.; Waters, K.; Gale, H.; Rice, P.; Scott, E. A. M.; Canham, L. T.; Reeves, C. L.; Loni, A.; Cox, T. I. In vivo assessment of tissue compatibility and calcification of bulk and porous silicon. Mater. Res. Soc. Symp. Proc. 1999, 536, 149- 154. 11. Anderson, S. H. C.; Elliott, H.; Wallis, D. J.; Canham, L. T.; Powell, J. J. Dissolution of different forms of partially porous silicon wafers under simulated physiological conditions. Phys. Stat. Sol. (a) 2003, 197, 331-335. 12. Gu, L.; Park, J.-H.; Duong, K.H.; Ruoslahti, E.; Sailor, M.J., Magnetic Luminescent Porous Silicon Microparticles for Localized Delivery of Molecular Drug Payloads, Small 2010, 6, 2546-2552. 13. Salonen, J; Laitinen, L.; Kaukonen, A. M.; Tuuraa, J.; Björkqvista, M.; Heikkiläa, T.; Vähä-Heikkiläa, K.; Hirvonen, J.; Lehto, V.-P. Mesoporous silicon microparticles for oral drug delivery: Loading and release of five model drugs J. Controlled Release 2005, 108, 362–374. 14. Wang, M.; Coffer, J.L.; Dorraj, K.; Hartman, P.S.; Loni, A.; and Canham, L.T. Sustained Antibacterial Activity From Triclosan-Loaded Nanostructured Mesoporous Silicon. Molecular Pharmaceutics, 2010, 7, 2232–2239. 15. Tang, L.; Saharay, A.; Fleischer, W.; Hartman, P.; Loni, A.; Canham, L.T.; Coffer, J.L. Sustained Antifungal Activity from a Ketoconazole-Loaded Nanostructured Mesoporous Silicon Platform. Silicon, 2013, 5, 213-217. 16. Limnell, T.; Riikonen, J.; Salonen, J.; Kaukonen, A.M.; Laitinen, L.; Hirvonen, J.; Lehto, V.-P. Surface chemistry and pore size affect carrier properties of mesoporous silicon microparticles, Int. J. Pharm., 2007, 343, 141-147. 17. Armstrong, F. A. J. The Determination of Silicate in Seawater. J. Mar. Biol. Assoc. U. K. 1951, 30, 149-160. 18. Tůma, J. Optimum conditions for the colorimetric microdetermination of silicon. Microchimica Acta, 1961, 50, 513-523. 19. Smith AL. The analytical chemistry of silicones. Series: chemical analysis, vol. 112. New York: Wiley; 1991, pp 182-183. 20. Boukherroub, R.; Wojtyk, J.; Wayner, D. ; Lockwood, D. Thermal hydrosilylation of undecylenic acid with porous silicon, J .Electrochem. Soc, 2002, 149, H59-H63.


20

Page 21 of 21

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

21. Tzur-Balter, A.; Rubinski, A.; Segal, E. Designing porous silicon-based microparticles as carriers for controlled delivery of mitoxantrone dihydrochloride. J. Mater. Res. 2013, 28, 231-239.

TOC Graphic

pSi

into

yields s. aureus inhibition

or

or


21

Influence of Surface Chemistry on the Release of an Antibacterial

Recommend Documents