Drug-Porous Silicon Dual Luminescent System for Monitoring and

Jul 12, 2017 - The free-standing porous silicon film was immersed in PBS solution (0.1 M, pH = 7.4) for 12 h to activate the luminescence emission, th...
109 downloads 12 Views 7MB Size
Download PDF
Drug-Porous Silicon Dual Luminescent System for Monitoring and Inhibition of Wound Infection Xisheng Chen,†,§ Fangjie Wo,†,§ Yao Jin,† Jie Tan,† Yan Lai,‡ and Jianmin Wu*,† †

Institute of Analytical Chemistry, Department of Chemistry, Zhejiang University, Hangzhou 310058, China Hangzhou GSPMED Medical Appliances Co. Ltd., Hangzhou 311401, China

‡

S Supporting Information *

ABSTRACT: Wound monitoring and curing is of great importance in biomedical research. This work created a smart bandage that can simultaneously monitor and inhibit wound infection. The main components of the smart bandage are luminescent porous silicon (LuPSi) particles loaded with ciprofloxacin (CIP). This dual luminescent system can undergo accelerated fluorescent color change from red to blue upon the stimulation of reactive oxygen species (ROS) and elevated pH, which are main biomarkers in the infected wound. The mechanism behind the chemicaltriggered fluorescent color change was studied in detail. In vitro experiment showed that the ratiometric fluorescent intensity (IRed/IBlue) of CIP-LuPSi particles decreased from 10 to 0.03 at pH 7.5 after 24 h, while the value deceased from 10 to 2.15 at pH 7.0. Strong correlation can be also found between the IRed/IBlue value and ROS concentration ranging from 0.1 to 10 mM. In addition, the oxidation of LuPSi also simultaneously triggered the release of CIP molecules, which exhibited bacterial inhibition activity. Therefore, the ratiometric fluorescent intensity change at red and blue channels can indicate not only the wound infection status but also the release of antibiotics. In vivo test proved that the smart bandage could distinguish infected wounds from acute wounds, just relying on the naked eyes or a cell phone camera. On the basis of the Si nanotechnology established in this work, theranostic wound care will be realized in future. KEYWORDS: luminescent porous silicon, ratiometric fluorescence, reactive oxygen species, smart bandage, wound monitoring healing process.3,4 The normal pH value of skin is slightly acidic (5.5−6), but an elevated pH will be found either in acute or chronic wounds at initial stage. However, the acute wound will return to normal acidic pH quickly, while the chronic wounds remain in alkaline (above 7.4) condition for a prolonged period spanning over several months.3,5 Besides pH, the reactive oxygen species (ROS) also play an important role in wound healing.6,7 Oxidative stress level is higher than normal skin when the wound is in a persistent inflammation state, because the immune cells such as neutrophils and macrophages invade into the wound. In the meantime, these immune cells phagocytose bacteria, produce and secret large amounts of ROS.8 A significant increase of ROS level in chronic wounds compared to acute wounds has been proved.9−12 Sustained elevation and survival of ROS in chronic wounds may cause tissue damage and impair wound. Therefore, the elevated pH

W

ound healing of injured skin involves in a series of complex biochemical processes. These physiological processes usually contain three major steps, including inflammation, new tissue formation and remodeling.1,2 These orderly processes occur in acute wounds but may be disrupted in chronic wounds, which usually exhibit a persistent pro-inflammatory state and result in poor wound healing. The high bacterial counts in the wound impede the normal progression to the proliferative phase of healing, thereby preventing the restoration of tissue integrity. Clinical assessment of chronic wounds is still fraught with ambiguity and heavily reliant on a clinician experience, causing increased financial burden and delay of proper treatment. Therefore, wound care products that can monitor the infection of wounds and aid clinical decision-making are urgently needed. Wound monitoring can take a number of forms by examining both physical and biochemical markers. Among them, detection of biomarkers in wound development is a more reasonable approach because the dynamic change of particular biochemical species can give insights into different wound status. For example, pH value displays a strong relationship with wound © 2017 American Chemical Society

Received: April 10, 2017 Accepted: July 12, 2017 Published: July 12, 2017 7938

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

www.acsnano.org

Article

ACS Nano

LuPSi due to the FRET effect.34 Exposing the CIP-LuPSi particles in aqueous medium containing oxidant species will oxidize the surface of LuPSi and eventually diminish its red fluorescence, especially with the assistance of alkaline ions, such as PO43− or OH−. In the meantime, the increasing thickness of oxide layer can reduce the efficiency of FRET effect between PSi and CIP, leading to the recovery of CIP fluorescence. By monitoring the ratiometric intensity of blue and red fluorescence, the overall oxidation ability of aqueous medium can be evaluated. For wound care application, the CIP-LuPSi particles were embedded in polyurethane (PU) membrane and chitosan film as a smart bandage. In vivo test indicated that the change of luminescent color from red to blue within skin wound area can be clearly observed by naked eyes and cell phone camera. By extracting RGB value from luminescent image, the wound infection can be monitored because of different oxidative environment between acute and chronic wound. Moreover, the accelerated oxidation of LuPSi caused by wound ROS and alkaline species will result in instant release of CIP from the host LuPSi particle, leading to the inhibition of wound infection. Therefore, theranostic wound care can be realized by the drug-loaded LuPSi smart bandage.

and ROS value constitute two major biomarkers in chronic or infected wounds. Electrochemical pH sensors have been adapted for the measurement of pH in wounds based on the pH dependence of quinone-hydroquinone redox peak.13,14 Sridhar and Takahata have developed a microfabricated wireless pH sensor, which uses a pair of wire coils sandwiching a pH-sensitive hydrogel.15 However, electrochemical method for wound monitoring still face big challenges since electrodes are usually vulnerable to be deteriorated in complex wound environment. In addition, electrochemical method is not convenient and cost-effective since it needs electronic device to generate and transmit signal during detection. In contrast, visualized sensing strategies are low-cost and convenient. Organic dyes or coordination compounds that can change their luminescence color or lifetime in response to pH or ROS change are usually employed.16−19 To improve the accuracy of fluorescent detection, ratiometric approaches are usually adopted. The ratiometric fluorescent detection consists of two luminescent dyes, one of which is sensitive to target analytes and the other has a stable luminescence. For example, ratiometric optical oxygen sensor is composed of an oxygen-independent reference dye and an oxygen-sensitive probe, whose luminescent intensity will change in different concentrations of oxygen or ROS.17,18,20,21 The quantitative analysis is based on the ratio of luminescent intensity between the oxygen-sensitive probe and reference probe. Fluorescence lifetime imaging for ROS measurement has also been reported. The photoluminescence of peptide-bridged dinuclear Ru(II) complex can be quenched by ROS through dynamic quenching mechanism, which results in the decrease of fluorescence lifetime.22 However, most of luminescent organic dyes or coordination compounds may have potential toxicity. In addition, drug loading for wound curing is not possible when these luminescent indicators are employed. Luminescent porous silicon (LuPSi) is a promising material for biomedical applications owing to its tunable porosity and surface chemistry, as well as its biocompatibility and degradability. These properties have led to the application of LuPSi as optical biosensors for bioimaging23−25 and drug delivery.26−33 Several works have revealed that the luminescent property of LuPSi is highly correlated with its surface chemistry, especially the oxidation of PSi surface. For example, oxidation-triggered fluorescence recovery from dye-modified PSi was used as a sensing strategy to indicate the ROS in biological samples.27 In that approach, the loaded luminescent dye was quenched by “dark” porous Si particles due to the fluorescence resonance energy transfer (FRET) effect. In the presence of ROS, the surface of porous Si particles was oxidized into an insulating oxide layer, which blocked the energy transfer between Si and dye molecules. As a result, the fluorescence of dye molecules recovered in response to ROS presented in sample solution. In the meantime, the surface oxidation can trigger the release of loaded drug, demonstrating that the oxidation-triggering approach can be potentially used in stimuliresponse drug release system. However, for nanosensor application, the measurement of fluorescence intensity in biological sample is not accurate, owing to several uncertain factors that may influence the absolute fluorescence intensity. Herein, drug loaded-LuPSi is introduced as a dual luminescent nanosystem for simultaneously wound monitoring and infection prevention. Ciprofloxacin (CIP) with blue luminescence was loaded in red luminescent PSi particles. At initial stage, the photoluminescence of CIP was quenched by the

RESULTS AND DISCUSSION Luminescence Activation of Porous Silicon Particles. The porosity and thickness of fresh etched PSi measured by spectroscopic liquid infiltration method (SLIM)35 were ∼64.2% and 22.9 μm, respectively. BET analysis showed that the specific surface area of the pSi is ∼561 m2/g with averaged pore diameter of 17.3 nm (see Figure S1a,c in Supporting Information). The newly etched porous silicon needed to be activated in aqueous environment to eventually display luminescence.36 Scanning electron microscopy (SEM) indicated that porous structure of PSi still remained after the luminescence activation (see Figure S2 in Supporting Information). However, the specific surface area of LuPSi reduced to ∼47.8 m2/g and 10.5 nm (see Figure S1b,d in Supporting Information), probably due to the partial dissolution of rough pSi surface and sealing of pore structure caused by oxidative corrosion. During activation process, the luminescent intensity of PSi gradually increased because of the passivation of nonradiative surface traps and emerging of surface state.36 The influence of pH on PSi oxidative degradation has been revealed by previous literature.37 At present work, we found that, the speed of luminescence activation was highly correlated with ion species of aqueous solution (see Figure S3 in Supporting Information). Even at the same pH value, the solution containing phosphate ions (PBS, pH = 7.0) can significantly accelerate the luminescence activation compared with NaCl solution (pH = 7.0). After 1 day, the PBS activated PSi displayed strong luminescent, inferring that phosphate ions may have stronger ability to attack the surface terminal of PSi. In addition, aqueous medium with higher pH value also tended to activate luminescence more quickly. We speculated that both phosphate ions and hydroxyl ions can accelerate the oxidation (or corrosion) of PSi by nucleophilic attacking to silicon atoms. However, the luminescence of PSi decreased when it was further oxidized, owing to the emerging of surface defects and eventual elimination of quantum-confined Si domains. Luminescent Behavior of CIP-LuPSi Particles in Different Oxidation Mediums. PSi is a promising material for bioimaging and drug delivery owing to its porous matrix, 7939

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano Scheme 1. Mechanism of Oxidation-Induced Luminescent Color Change of CIP-LuPSi

Figure 1. (a) Luminescent images of CIP-LuPSi particles in different oxidation mediums recorded by CCD detector. λex = 375 nm. (b) The luminescent spectra of CIP-LuPSi particles under different oxidative state. The inserted pictures were luminescent images of CIP-LuPSi particles.

nontoxic and biodegradable properties.23 For wound care purpose, PSi is also a good candidate since the porous matrix can not only load anti-infection drug but also monitor wound infection by measuring the change of optical signal caused by chemical stimulation.33 For example, a conceptus wound monitoring film derived from PSi microcavity supported on hydrogel has been proposed.38 Bacterial infection may cause the change of chemical composition in wound area, leading to the sensitive change in refractive index of PSi microcavity. However, measuring the reflectance peak shift of PSi microcavity still requires a spectroscopic device, because a small change in refractive index cannot be seen by naked eyes. The purpose of present work is to incorporate antibacterial agent into the LuPSi and monitor the luminescent intensity change of the dual luminescent system for wound care application.

The sensing strategy employed in this work is illustrated in Scheme 1. At initial stage, LuPSi with thin oxide layer still retained high quenching ability because of its high density of electronic states and the large surface area of the porous structure. When the antibiotics CIP was loaded in the LuPSi matrix, the blue luminescence of CIP would be quenched by FRET effect, in which the CIP molecules adsorbed on LuPSi surface acted as the donor while LuPSi played as the energy acceptor. Although the quenching of fluorescent molecules by “dark” PSi has been reported before,27,34 the situation present in this work is significantly different as the system contains two luminescent emitters, which provide a possibility to measure the change of ratiometric fluorescent intensity. When the LuPSi was further oxidized by oxidation agents, the LuPSi surface converted into a thicker porous SiO2 insulator, elongating the distance between LuPSi and CIP, thereby reducing the free 7940

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano

Figure 2. (a) Luminescent spectra of CIP-LuPSi particles after incubating in the blank and ROS solution with different concentrations for 1 day. The inserted pictures were luminescent images of CIP-LuPSi particles corresponding to the FL spectra and the scale bar length was 200 μm. The excitation wavelength λex = 375 nm, the spectra and picture was simultaneously acquired by a fluorescent microscopy coupled with CCD camera and fiber spectrometer. (b) Quantitative relationship between ratiometric FL intensity (I700 nm/I450 nm) and ROS concentration.

Figure 3. (a) Absorption spectra of porous silicon nanoparticles in 10 mM PBS (pH = 7.0) solution. The intensity of absorbance decreased during oxidation process. The blue trace represents the FL emission spectra of CIP. (b,c) Optical microscopic images of porous silicon microscale particles before and after oxidized by oxidation agents (Scale bar length is 200 μm). (d) Change of relative absorbance of PSi in different mediums.

carrier concentration and density of free electronic states.34 As a result, the FRET effect was weakened and the luminescent intensity of CIP gradually recovered. In the meantime, the LuPSi particles underwent an overoxidation process, resulting in the decrease of luminescent intensity of LuPSi due to the corrosion of quantum-confined Si domains and generation of defect energy level. Overall, during the oxidation process, the fluorescent (FL) intensity of CIP (Blue luminescence) steadily recovered whereas the FL intensity of LuPSi (Red luminescence) gradually decreased. By detecting ratiometric FL intensity, the kinetic process of surface oxidation of CIPLuPSi can be precisely measured. It is well-known that ratiometric approach is more accurate than single-wavelength detection since the absolute fluorescent intensity at single wavelength is vulnerable to be affected by subtle change of environment. In addition, the FL color change from red to blue can be easily discriminated by naked eyes. To investigate the effect of pH, ion species and ROS on the kinetic process of surface oxidation, CIP-LuPSi particles were incubated in different types of aqueous mediums, all of which contain 0.9% NaCl solution to ensure the same ionic strength.

The FL color change at single particle level was acquired by a fluorescent microscopy equipped with high sensitive CCD camera. As shown in Figure 1a, the speed of FL color change of CIP-LuPSi was obviously affected by ion species even at the same pH (7.0). In neutral salt solution (0.9% NaCl), it took ∼4 days for the particles to change from red to blue. In contrast, both ROS (1O2)39 and phosphate ion (PO43−) can accelerate the oxidization LuPSi particles, resulting in the FL color change from red to blue within 1 day. The pH value also exerted profound effect on the surface oxidation. The luminescence of particles was more stable in acidic medium (0.9% NaCl, pH = 5.5), while the change of FL color quickly proceeded in alkaline solution (0.9% NaCl pH = 7.5). The results indicated that the hydroxyl ions also catalyze the oxidation of PSi surface by H2O molecules which acted as mild oxidizing agent.23 As shown in Scheme 1, the catalysis oxidation of PSi involved a nucleophilic attack mechanism. Both hydroxyl and phosphate ions can effectively attack the silicon atom on the PSi surface, accelerating the oxidation and corrosion of silicon core. Among various types of phosphate ion species, the PO43− tends to attack silicon atoms more easily. Therefore, the 7941

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano

Figure 4. (a) The schematic illustration of smart bandage structure and its luminescence color change after mild oxidation (scale bar length is 0.5 cm). (b) The change of ratiometric intensity (I700 nm/I450 nm) as a function of time in different types of aqueous mediums.

kinetics of FL color change in the presence of PBS is also pHdependent since the degree of H3PO4 dissociation depends on pH value. As shown in Figure 1a, the FL color change in PBS solution at pH 5.5 was slightly slower than that at pH 7.5. The luminescent spectra acquired from the CIP-LuPSi particles further confirmed that the FL color change was ascribe to the change of ratiometric intensity at red and blue wavelength (Figure 1b). The double peaks shown in FL spectra of CIPLuPSi particle were identical to the peak positions corresponding to CIP and LuPSi (see Figure S4 in Supporting Information), whose ratiometric FL intensity at 700 and 450 nm decreased during the mild oxidation process. To quantitatively study the relationship between oxidative species and the ratiometric FL intensity of CIP-PSi, the particles were incubated in solutions with controlled ROS and OH − concentrations. After 1 day incubation, the ratiometric FL intensity at 700 and 450 nm decreased with the increase of ROS concentration (Figure 2a) and displayed a negative correlation with the ROS concentration ranging from 0.1 to 10 mM (Figure 2b). Although, the limit of detection (LOD) for ROS detection is above 0.1 mM in the present case, higher sensitivity could be expected if the CIP-LuPSi particles reacted with ROS species for longer time. When the particles were immersed at pH value of 5.5, 7.0, and 7.5 for 24 h, the ratiometric fluorescent intensity of CIP-LuPSi particles decreased from 10 to 2.66, 2.15 and 0.03, respectively, displaying linear relationship with the concentration of hydroxyl ions (see Figure S5 in Supporting Information). Mechanism Study on the FL Change of CIP-LuPSi. The FL spectra of CIP overlapped with the absorbance spectra of LuPSi particles, confirming that the quenching of CIP fluorescence by LuPSi follow a FRET mechanism (Figure 3a). We have proved that this FRET effect pervasively existed in the LuPSi loaded with different luminescent emitters. For example, the photoluminescence of fluorescein and graphene quantum dots hosted in the pore structure of LuPSi can be also quenched effectively (see Figure S6 in Supporting Information). The results inferred that various types of ratiometric sensing systems can be constructed. To further confirm the oxidation of Si nanocrystal core in aqueous medium, the UV− vis absorbance spectra of LuPSi nanoparticles was measured in different interval times. The utilization of nanoscale LuPSi particles (∼200 nm) in this experiment is to avoid the strong light scattering caused by the microscale LuPSi particles. Figure

3a showed that, in 10 mM PBS (pH = 7.0) solution, the intensity of absorptive spectral peak of LuPSi decreased steadily from 1.00 to 0.08 in 2 h. Compared to the microscale particles, it took less time to oxidize the nanoscale LuPSi particles in aqueous medium. ICP-AES analysis showed that the amount of silicon dissolved into the aqueous medium quickly increased within 10 h (see Figure S7 in Supporting Information). The results confirmed that the reduction of UV−vis absorbance is caused by the oxidation or corrosion of silicon core in aqueous medium. Optical microscopic image showed that the color of microscale PSi particles also changed from brownish to colorless during the mild oxidation process, indicating that Si nanocrystal core was finally oxidized (Figure 3b,c). Dynamic change of UV−vis spectra of LuPSi in different mediums also indicated that ROS solution (10 mM, pH 7.0) displayed the highest oxidation capacity, followed by the order of PBS and alkaline solution (Figure 3d). The Si nanocrystal can be relatively stable in acidic (NaCl, pH 5.5) or neutral mediums (NaCl, pH 7.0). The intensity of the IR peak at 2101 and 2258 cm−1 assigning to the Si−Hx and O−Si−Hx bonds were observed by diffuse reflective-FTIR (DR-FTIR).40,41 The peak intensity ratio at 2258 cm−1 vs 2101 cm−1 increased from 0.36 to 1.00 during the incubation of PSi particles in aqueous medium (see Figure S8 in Supporting Information), further proving the oxidation of silicon nanocrystal in aqueous solution. Smart Bandage Constructed from Polymer Embedded PU-CIP-LuPSi. In order to facilitate wound monitoring and wound cure, CIP-LuPSi particles were sandwiched into two layers of polymeric film. Polyurethane membrane (PU, Tegaderm Film, 3M, USA) was used as a substrate to support CIP-LuPSi particles, which were uniformly distributed and attached onto the surface of PU membrane with adhesive ability. However, direct contact of the particles to wound area may cause the quick degradation of CIP-LuPSi. Accordingly, chitosan film cross-linked with 3-Glycidyloxyproyl-trimethoxysilane42 was coated on the sample surface as a protecting layer to form a sandwich structure as depicted in Figure 4a. The reason to choose chitosan film is ascribed to its high biocompatibility and good permeation ability to tissue fluid. In addition, chitosan has been proved to be able to promote wound healing owing to its bacteriostatic property.43,44 Under ultraviolet LED irradiation (λex = 377 nm), red luminescence emitted from smart bandage can be observed at initial stage. After oxidation in aqueous medium, the bandage emitted blue 7942

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano

Figure 5. (a) Drug releasing rate of CIP-LuPSi in different mediums at the initial 12 h. The number of 1, 2, 3, 4 represent the CIP-LuPSi samples incubated in 0.9% NaCl (pH = 5.5), 0.9% NaCl (pH = 7.0), 10 mM PBS (pH = 7.0), 10 mM ROS (pH = 7.0), respectively. The number 5, 6, 7, 8 represent the polymer protected samples incubated in 0.9% NaCl (pH = 5.5), 0.9% NaCl (pH = 7.0), 10 mM PBS (pH = 7.0) and 10 mM ROS (pH = 7.0), respectively. (b) Inhibition zone of bacterial growth of S. aureus (Scale bar length is 1 cm). The released CIP refers to CIP released from CIP-LuPSi particles; the positive control was the standard CIP solution with the same concentration as the released CIP. The negative control (Blank) was the solution incubated with bare LuPSi particles at the same experimental condition. The inserted picture was the image of inhibition zone assay. The number 1, 2, 3 represent the inhibition zone of blank group, standard CIP and released CIP, respectively. Each sample was tested for three times. (c−f) Cell cytotoxicity of LuPSi particles and CIP-LuPSi particles. Viabilities of HaCaT cells incubated with (c) LuPSi particles; (d) CIP-LuPSi particles and NIH3T3 incubated with (e) LuPSi particles; (f) CIP-LuPSi particles (CellTiter-Glo assay) for 24 and 48 h. Each value is an average of five samples.

FL color (Figure 4a). Due to its large color contrast, this FL color change can be observed by naked eyes or smartphone camera without optical filter. In different aqueous mediums, the overall behaviors of FL color change observed in smart bandage were almost the same as those on bare CIP-LuPSi particles (see Figure S9 in Supporting Information), but the rates of color

change were slower because the aqueous solution needs to diffuse through the protective layer before mild oxidation. The kinetics of ratiometric intensity (I700 nm/I450 nm) change measured on smart bandage was also strongly correlated with the type of aqueous medium (Figure 4b). In accordance with the results obtained on bare CIP-LuPSi, the FL color change of 7943

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano Scheme 2. Smart Bandage for In Vivo Wound Monitoring and Healinga

a

The FL color of smart bandage covered on wound area will undergo an accelerated change from red to blue, which can be observed by the naked eye. Semiquantitative evaluation of wound infection can be achieved by cell phone camera with color analysis software.

the smart bandage can simultaneously act as dual functional biomaterial for monitoring wound infection and drug release. Antibacterial Effect and Cytotoxicity. The bactericidal activity of CIP released from CIP-LuPSi particles was evaluated via measuring the inhibition zone of bacterial growth of S. aureus (ATCC 25923).45 Experimental group was the solution containing CIP with concentration of 0.155 mg/mL released from CIP-LuPSi particles. The positive control group was the standard CIP solution whose concentration was adjusted to 0.165 mg/mL, which was almost equal to the CIP concentration released from the particles. The negative control group was the supernatant solution incubated with bare LuPSi particles under the same experimental condition. CIP diffused from filter paper to Luria−Bertani (LB)-agar plate can inhibit the growth of bacteria and form the round inhibition zone. As shown in Figure 5b, the inhibition zone area of negative control (No. 1), positive control (No. 2) and experimental group (No. 3) were 1.73 cm2, 13.20 cm2 and 12.55 cm2, respectively, indicating that the bactericidal activity of released CIP was almost the same as that of standard CIP. In contrast, supernatant of bare LuPSi particles displayed little antibacterial activity. These data verified that the chemical structure or bioactivity change of released drug did not change. Cytotoxicity experiment was also conducted to testify the biocompatibility of the CIP-LuPSi. HaCaT and NIH3T3 cells were incubated with LuPSi and CIP-LuPSi for 24 and 48 h, respectively. In all tested groups, the cell viabilities exceed 75%, even when the concentration of sample increased up to 1 mg/mL (Figure 5c−f). The in vitro experiments indicated that both LuPSi and CIP-LuPSi displayed low toxicity to both types of model cells, which are typical cells for repairing skin tissue. Smart Bandage for In Vivo Wound Monitoring and Inhibition of Wound Infection. Acute wounds usually repair orderly and go through a quick inflammatory response with lower oxidative stress level and pH value. Infection of bacteria is one of the most common complications preventing wound healing. Herein, we chose infected wound as a chronic wound model, which tended to stay in a persistent pro-inflammatory state with high oxidative stress level and alkaline pH value.3,5,9−12 Visualized wound monitoring and instant curing using the smart bandage is illustrated in Scheme 2. The smart bandage will undergo a luminescent color change from red to blue because of the wound-induced oxidative stress and

the smart bandage also accelerated in alkaline and ROS solution. For practical application, high stability during long-term storage is required. We have stored the smart bandage in ambient air and ethanol for 4 months and found that the intensity of red FL almost kept constant (see Figure S10 in Supporting Information), indicating that the spontaneous aging on FL change can be neglected. Stimulus-Response of Drug Release Kinetics in Different Oxidation Mediums. We anticipated that the smart bandage can not only monitor the wound condition, but also cure the infected wound by inhibiting bacterial growth. To evaluate the stimulus-response release of antibiotics, initial release rate of antibiotics within 12 h under different conditions were measured (Figure 5a). CIP-LuPSi particles and smart bandage were immersed in different types of aqueous solution to simulate the wound environment. Comparing to the abrupt release in bare CIP-LuPSi particles, the polymer protected samples displayed much lower initial release rates in different types of solution, indicating that the polymer protection can efficiently control the drug release. Overall, the drug release rate was significantly correlated with the oxidative ability of solution. The order of releasing rates in different mediums was almost the same as that found in FL color change. Among them, aqueous solution containing ROS showed the highest initial drug releasing rate. The concept of oxidation-triggered drug release from porous silicon has been proposed in previous literature.27 The situation observed in this work also followed the same mechanism. Even though the release of CIP may affect the intensity of blue fluorescence, we still found the intensity of blue FL color increased steadily during the mild oxidation process. As observed in Figure 1a and Figure S6 in Supporting Information, the oxidation process proceeded from the edge to the center and then spread over the entire particle. This phenomenon indicated that the reactivity of Si nanocrystals in different regions of the particle might be different. Consequently, the oxidation of Si nanocrystals did not take place simultaneously and the FL intensity of CIP increased when more and more Si nanocrystals were oxidized. On the other hands, a portion of CIP molecules may be irreversibly adsorbed on the PSi surface45 even after the surface of Si nanocrystal was oxidized. Due to the high synchronization between oxidation-triggered drug release and FL color change, 7944

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano

Figure 6. (a) Visual wound monitoring by smart bandages. The smart bandages were illuminated by ultraviolet LED (λex = 377 nm). The bright field images (left) and luminescent images (right) were recorded by mobile phone camera every day, respectively. Each luminescent image has the same scale bar of 0.5 cm as shown in right bottom image. (b) The ratiometric R/B value of luminescent images shown in (a). The RGB value was analyzed by Photoshop CS5. Each value is an average of five samples.

Figure 7. Comparison of bacterial density between the area with and without the covering of smart bandage. (a) Bright field image of infected wound covered with bandage. The area 1 represented the wound area covered with blank PU membrane, the Area 2 represented the wound area covered with smart bandage. (b) Bright field image of infected wound after removal of bandage. (c) Dark field image of wound covered with bandage. (d,e) Wound tissue sections of Area 1 (d) and Area 2 (e). The microscopic image of sided tissue sections with gram stained and HE stained.

elevated pH value. Under UV LED irradiation (λex = 377 nm), the image of FL color change can be simply acquired by a smartphone camera. With the assistance of color analysis software (Digital Media interactive LLC) installed in smartphone, the ratiometric intensity of red vs blue channel can be conveniently obtained. S. aureus and P. aeruginosa are the most prevalent pathogenic organisms found in chronic wounds.46 Infected wound models were established by inoculating S. aureus (ATCC 25923) on the wound area of mice. For the acute wound model, the rats were treated with the same operation without bacterial inoculation. Smart bandages were applied onto the wound area of different groups. Experimental results showed that the change of luminescent color in the group of infected wound was faster than that in the group of acute wound (Figure 6a). The FL color of the bandage applied onto the infected wound began to change from red to blue at the second day. In contrast, the FL color of smart bandage applied onto acute wound group still appeared to be red even after 4 days. Quantitative RGB analysis of luminescent images by Photoshop CS5 or smartphone app was also performed. The ratio of gray value at R/B channels tended to change rapidly in infected wounds and showed large

discrimination between the infected wound and acute wound after 2 days (Figure 6b). To confirm whether the FL color change of the smart bandage was caused by the elevated oxidation stress level and pH value, critical biomarkers in different in vivo wound models were measured. The blood and wound tissue samples were collected at 8 d after injured. The lipid peroxidation product, malondialdehyde (MDA), was measured to determine the oxidative stress level (see Figure S11a in Supporting Information). The MDA contents in both serum and tissue homogenate of infected group were significantly higher than those in acute group, confirming the high oxidative stress level occurred in the infected wounds. The percentage of neutrophil in infected wounds was significantly lower than that in acute wound, while the proportion of lymphocyte in infected wound increased obviously (see Figure S11b in Supporting Information). The reason can be ascribed to the damage of neutrophil and the compensatory increase of lymphocyte in a long-term “chronic effect”. In another word, infected wounds still persisted in a pro-inflammatory state. The initial pH values of both acute and infected wounds were ∼7.5. However, comparing to the acute wounds, the pH of infected wounds slightly increased to 7.9 and kept in the alkaline 7945

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano

camera and color analysis software. Owing to the high biocompatibility and drug loading ability of LuPSi, practical application of silicon-based smart bandage in wound care and bioimaging will be expected. Furthermore, we found that the FRET effect between LuPSi and luminescent dyes or quantum dots pervasively existed. Therefore, versatile nanosensors and drug delivery system based on dual luminescent nanomaterials can be constructed.

condition for a long-time. This phenomenon is in accordance with the results reported in previous literature.3 In addition, we also found that the content of PO43− in infected wounds were ∼2.16 mg/g tissue, which increased almost 10% compared to the value in acute wounds. The infection response may generate exudate and cause tissue breakdown47 as the necrosis of histocyte and immune cells, resulting the release of phosphate ions and the hydrolysis of phosphate ions-containing substance such as ATP, ADP and DNA. These phosphate ions and derivatives have also been used as indicators to diagnose a number of diseases.48−50 The overall increase in ROS, pH and PO43− in the infected wounds contributed to the accelerated FL color change of smart bandage, although it could not discriminate which was a major factor. Morphology of bacteria scraped from wound area appeared as a large white to golden colony with smooth surface, which matched the characteristic morphology of S. aureus (see Figure S11c in Supporting Information). Gram stain and HE stain on the tissue slide were also performed to confirm the existence of bacteria. Dye staining resulted in purple S. aureus in the section edge, red cytoplasm and hyacinthine nuclei (see Figure S11d in Supporting Information). These results indicated that infected wound model was successfully established. The in vivo test also verified that the smart bandage displayed antibacterial effect. The bright field image indicated that the wound area contacted with polymer protected CIP-LuPSi had lower bacterial density compared to the wound edge (Figure 7a−c). Since the size of smart bandage (∼1.3 × 1.3 cm2) was smaller than the circular wound with diameter of 1.8−2.0 cm, the wound edge was only covered by blank PU membrane and consequently regarded as the untreated control. The stained sections of wound tissue also confirmed that bacterial density on the center area was significantly lower than that on the margin area (Figure 7d, e). Both evidence further proved that the reproduction of bacteria was inhibited by the released CIP.

METHODS Preparation of CIP-LuPSi Particles and Smart Bandage (PUCIP-LuPSi). Preparation of Luminescent Porous Silicon Microparticles. Porous Si was obtained by anodization of a boron-doped silicon wafer (resistivity: 0.0005−0.0012 Ω/cm) with a [100] crystal orientation in an electrolyte mixture of aqueous hydrofluoric acid (48% by mass, Alatin Corp.) and ethanol in a volume ratio of 4:1. The etch cell with 63.6 cm2 exposed etching area was used. The Si wafer was etched at a constant current density of 77 mA/cm2 for 600 s. The porous Si film was then removed from the crystalline Si substrate by application of a current density of 22 mA/cm2 for 180 s in a solution of 3.3% aqueous hydrofluoric acid in ethanol. The free-standing porous silicon film was immersed in PBS solution (0.1 M, pH = 7.4) for 12 h to activate the luminescence emission, then it was washed with deionized water for several times. The luminescent porous silicon film was placed in deionized water and fractured into microparticles by ultrasonication (100 W) for 5 min. Preparation of CIP Loaded Luminescent Porous Silicon Microparticles (CIP-LuPSi). Ciprofloxacin (CIP) was purchased from Bio Basic Inc. Around 70 mg of luminescent PSi microparticles were incubated in 40 mL of CIP solution (1.5 mg/mL) and shaken in a vibrator at room temperature for 20 h. The CIP-LiPSi particles were isolated from solution by centrifugation at 8000 rpm for 10 min and rinsed with deionized water for several times to remove free CIP. Preparation of Smart Bandage PU-CIP-LuPSi. The polyurethane (PU) membrane is a product of 3 M (Tegaderm Film, 3M, USA). Before usage, a 1.3 × 1.3 cm2 paper substrate adhered on the sticky surface of PU film was removed. 4.0 mg of CIP-LuPSi particles were distributed uniformly on the surface of PU membrane. The particles can be strongly attached because of the adhesive coating on one side of PU membrane. In order to protect the CIP-LuPSi particles, chitosan solution (1% by mass dissolved in acetic acids) mixed with (3glycidyloxypropyl)trimethoxysilane at a volume ratio of 100:3 was coated on the CIP-LuPSi particles. After drying at 37 °C for 3 h, the smart bandage with sandwich structure was formed. Characterization of Luminescent PSi Particles. Characterization of Surface Morphology. The surface morphology of luminescent PSi particles was characterized by a field emission scanning electron microscope using an accelerating voltage of 10 keV (SEM, Model Utral 55, CorlaeisD, Germany). Porosity and Film Thickness Measurements. The porosity and thickness of porous films were determined from reflectance spectra using the spectroscopic liquid infiltration method (SLIM).35 Briefly, reflectance spectra were obtained on a USB 2000+ miniaturized fiber spectrophotometer (Ocean Optics, USA). An LS-1 halogen tungsten lamp was used as the light source to illuminate the surface of PSi through one arm of a bifurcated fiber optic cable at normal incidence. The quantity 2nL, commonly referred to as the effective optical thickness (EOT), was determined from the Fabry−Perot relationship:

CONCLUSION CIP-LuPSi smart bandage for visualized wounds monitoring and curing has been demonstrated. The newly etched “dark” PSi can be quickly activated in mild oxidative condition to generate bright red photoluminescence. More importantly, we found the LuPSi with thinner oxide layer can still quench fluorescent molecules adsorbed on its surface owing to the FRET effect. Further oxidation of LuPSi increased the thickness of oxidative layer, leading to the blockage of FRET effect and recovery of luminescence emitted from the adsorbed molecules. In the meantime, the introduction of surface defect and corrosion of quantum-confined Si domains during mild oxidation decreased the luminescence intensity of LuPSi. The rate of oxidation-triggered FL color change was significantly correlated with solution chemistry. Generally, aqueous solution with higher pH value and ROS level can accelerate the change of ratiometric FL intensity at 700 nm vs 450 nm. On the basis of this finding, smart bandage embedded with CIP-LuPSi particles was created to distinguish infected wounds from acute wounds, because higher ROS and pH usually occurred in the infected wounds. In addition, oxidation of LuPSi can simultaneously trigger drug release process, which was well synchronous with the FL color change. Due to the large color contrast between red and blue color, the FL color change of smart bandage can be clearly observed by naked eyes. More accurate detection of ratiometric intensity between R and B channels can be realized with the assistance of mobile phone

mλ = 2nL where λ is the wavelength of maximum constructive interference for spectral fringe of order m, n is the averaged refraction index of the porous layer and L is the thickness of the porous layer. The infiltration of ethanol will displace the air in the porous structure, causing the change of n value. The thickness and porosity of the porous Si film can be calculated by Bruggman dielectric constant approximation. BET Measurements. The specific surface area of newly etched porous silicon and activated porous silicon was measured by specific 7946

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano surface and pore analyzer (ASAP2020HD88, America) using N2 adsorption method. Methods for Mechanism Study. UV−Vis Absorbance Measurements. The LuPSi was fractured into nanoparticles by ultrasonic cell disruption system (NingBo SCIENTZ, China) for 1 h, the power used here was 600 W. Then the nanoparticles were soaked in different mediums (0.9% NaCl pH = 5.5, 0.9% NaCl pH = 7.0, 10 mM PBS pH = 7.0, 10 mM ROS pH = 7.0 and 0.9% NaCl pH = 7.5). The UV−vis absorbance spectra of LuPSi nanoparticles were measured with a fiber spectrometer (MAYA, Ocean optics, USA) at different time interval. Detection of Si Concentration by Inductive Coupled Plasma (ICP). The CIP-LuPSi sample with the concentration of 15 mg/mL was incubated in 6 mL 10 mM PBS solution (pH = 7.0) at 37 °C. An aliquot (200 μL) of the degradation supernatant was taken from the solution described above at different time points in 50 h. Then the supernatant was diluted with deionized water to 15 mL and the pH value was adjusted to 5.0 by 0.1 M HCl. The amount of Si elemental was measured by inductively coupled plasma optical emission spectrometry (ICP6000, Thermo SCIENTIFIC Co.) Diffuse Reflection-Infrared Fourier Transform Spectroscopy. The surface group of Freeze-dried samples including nonactivated PSi, activated PSi and color changed PSi particles were detected by diffuse reflection-infrared Fourier transform spectroscopy (NICOLET iS10, Thermo SCIENTIFIC). The resolution of FTIR spectrometer was set at 4.000 cm−1. Observation of Photoluminescence Emitted from CIP-LuPSi Particles or Smart Bandage in Different Types of Aqueous Mediums. Aqueous mediums including ROS, phosphate ions (PBS) and hydroxyl ions were prepared, respectively. Each solution was spiked with 0.9% NaCl solution to keep the ionic strength. The ROS (1O2) solution was prepared by mixing H2O2 with NaClO.39 The concentrations of ROS and phosphate ions (PBS) were 10 mM, respectively. The concentration of hydroxyl ions was adjusted by 0.1 M NaOH or 0.1 M HCl solution. The CIP-LuPSi particles and smart bandage were soaked in above-mentioned aqueous mediums at 37 °C. Under UV excitation at wavelength of 375 nm, the luminescence image of CIP-LuPSi particles and smart bandage PU-CIP-LuPSi were acquired by fluorescence microscope (NE950, Nexcope, China) attached with a high sensitive CCD detector (MC20-C, Mshot, China). The R/B value was analyzed by Photoshop CS5 software (Adobe Inc.) The fluorescent spectra were acquired by a fiber spectrometer (QE pro, Ocean Optics, USA) coupled to the fluorescence microscopy. The resolution of fiber spectrometer was 6.5 nm (fwhm). Drug Release Experiment in Different Conditions. The quantitative determination of CIP concentration released from CIPLuPSi particles in different mediums were detected by HPLC (SIL20A, SHIMADSU) system equipped with a Hypersil ODS2 column (4.6 mm × 150 mm, 5 μm) which was kept at 40 °C. Mobile phase was a mixture of phosphoric acid (0.05 M) and acetonitrile at volume ratio of 82:18 (v/v). The flow rate of mobile phased was set at 1.0 mL/min. The eluted CIP was detected by a diode array detector at wavelength of 271 nm. The CIP-LuPSi particles and smart bandages were dispersed in 6 mL different types of mediums (0.9% NaCl, 10 mM PBS, 10 mM ROS solution, pH = 7.0) at 37 °C, respectively. Then 0.7 mL of supernatant was collected at time intervals of 1, 6, 12 and 24 h, respectively. Each sample was placed in glass vials, which was later put into a tray of automatic sampler, whose injection volume was set as 20 μL. The concentrations of CIP released at different times were calculated according to the calibration curve of standard CIP solution. Antibacterial Effect and Cytotoxicity. Antibacterial Effect. The bacterial inhibition activity of the CIP released from CIP-LuPSi particles was detected via inhibition zone assay with filter paper. Around 10.27 mg of CIP-LuPSi particles were incubated in 4.0 mL of PBS solution in a 5 mL tube at 37 °C. After 24 h, the supernatant was transferred to a new tube and the concentration was determined by HPLC as described above. The positive control group was prepared from standard CIP solution, whose concentration was almost equal to that of released CIP. The negative control group was the supernatant

solution incubated with bare LuPSi particles under the same experimental condition. Then a 20 μL of each sample was spotted on a piece of round filter paper with a diameter of 5 mm. After drying, the filter paper was placed on the top of a Luria−Bertani (LB)-agar plate homogeneously colonized with S. aureus bacteria (ATCC 25923), which was then incubated at 37 °C for 24 h. The bacterial inhibition activity of each sample was evaluated by measuring the dimeter of the inhibition zone. Cytotoxicity. The cytotoxicity experiments were conducted on human immortal keratinocyte (HaCaT) and mouse embryonic fibroblast (NIH3T3) lines. Briefly, HaCaT and NIH3T3 cells were seeded in 96-well plates at a density of about 1 × 104 cells/well and cultured in 37 °C. After 24 h, the medium was removed and 100 μL of DMEM medium (containing 10% of fetal bovine serum) dispersed with LuPSi or CIP-LuPSi was added. The concentrations of LuPSi or CIP-LuPSi were 0.25, 0.5, and 1 mg/mL, respectively. After incubated in 37 °C for 24 or 48 h, the cell viabilities were measured by Infinite M200 Pro microplate reader (Tecan, Swiss) using CellTiter-Glo reagent (Promega, USA). Wound Model Creating and Monitoring. Creation of Acute and Infected Wound Models. To establish animal wound model, Sprague Dawely (SD) male rats (200−300 g) were obtained from Zhejiang Academy of Medical Sciences and fed for 1 week before the experiment. After anaesthetized by xylazine hydrochloride injection, the skin hair was shaved off and the surface was sterilized with 70% alcohol prior to the creation of wounds. Two transdermal wounds with 1.8−2.0 cm of diameter were made on dorsal surface of each rat. In the experimental group, the wounds were inoculated with S. aureus (ATCC 25923) with initial density of 106 cfu/mL. For bacteria reproduction, the wounds were covered with a bare PU membrane (Tegaderm Film, 3M, America) for 24 h. Then the PU membrane was removed and replaced with a smart bandage. As to the control group, the rats were treated with same operation but without bacterial inoculation. All of the rats were housed individually in sterile rearing cages, maintained at standard animal house conditions, and had free access to diet and water. Wound Monitoring. The smart bandages applied onto the infected or acute wounds were illuminated by UV LED (λex = 377 nm). Luminescent color changes of smart bandages during in vivo study were acquired by a cell phone camera (iPhone 6S, Apple Inc. USA). The gray values at red and blue channels were analyzed by Color Companion iOS App (Digital Media Interactive). Biochemical Analysis. Blood samples were collected at 8 d after wounding and the serum was obtained by centrifugation. The granulation tissues were also obtained from the wound area at the same time. Then tissues were treated by ultrasonic cell bath (JY88-IIN Xinyi, Ningbo, China) in NETN buffer (containing 150 mM of NaCl, 50 mM of Tris and 1% of NP-40) and centrifuged. The supernatant was collected for further experiments. All of the steps were performed at 4 °C. MDA levels were measured using the Lipid Peroxidation MDA Assay Kit (Beyotime, Jiangsu, China), which is based on the reaction of MDA and thiobarbituric acid (TBA). The absorptions were measured at 532 nm. MDA levels were calculated according to the established standard curve using the reference standards in the kit under the same condition. The concentration of phosphate ions in the tissue sample was measured by phosphorus molybdenum blue method. The neutrophil and lymphocyte in blood were analyzed by volume, conductivity and scatter (VCS) technology. Wound Bacteria Culture. The bacteria were scraped from wound surface and cultured in Luria−Bertani (LB) liquid medium. After 24 h, 10 μL of the bacterial suspension was incubated in LB-agar plate for 24 h. Then the bacteria were identified by observing the bacterial morphology. Wound Tissue Section and Staining. Paraformaldehyde-fixed wound tissues were embedded in paraffin and sectioned with freezingmicrotome (CryoStar NX50, Thermo, America). After removing paraffin film, the tissue sections were gram stained and HE stained. 7947

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano

(15) Sridhar, V.; Takahata, K. A Hydrogel-Based Passive Wireless Sensor Using a Flex-Circuit Inductive Transducer. Sens. Actuators, A 2009, 155, 58−65. (16) Bartsch, I.; Willbold, E.; Rosenhahn, B.; Witte, F. Non-Invasive pH Determination Adjacent to Degradable Biomaterials In Vivo. Acta Biomater. 2014, 10, 34−39. (17) Meier, R. J.; Schreml, S.; Wang, X. D.; Landthaler, M.; Babilas, P.; Wolfbeis, O. S. Simultaneous Photographing of Oxygen and pH In Vivo Using Sensor Films. Angew. Chem., Int. Ed. 2011, 50, 10893− 10896. (18) Schreml, S.; Meier, R. J.; Kirschbaum, M.; Kong, S. C.; Gehmert, S.; Felthaus, O.; Kuchler, S.; Sharpe, J. R.; Woltje, K.; Weiss, K. T.; et al. Luminescent Dual Sensors Reveal Extracellular pH-Gradients and Hypoxia on Chronic Wounds That Disrupt Epidermal Repair. Theranostics 2014, 4, 721−735. (19) Schreml, S.; Meier, R. J.; Weiss, K. T.; Cattani, J.; Flittner, D.; Gehmert, S.; Wolfbeis, O. S.; Landthaler, M.; Babilas, P. A Sprayable Luminescent pH Sensor and Its Use for Wound Imaging In Vivo. Exp. Dermatol. 2012, 21, 951−953. (20) Gao, X.; Ding, C. Q.; Zhu, A. W.; Tian, Y. Carbon-Dot-Based Ratiometric Fluorescent Probe for Imaging and Biosensing of Superoxide Anion in Live Cells. Anal. Chem. 2014, 86, 7071−7078. (21) Zhuang, M.; Ding, C. Q.; Zhu, A. W.; Tian, Y. Ratiometric Fluorescence Probe for Monitoring Hydroxyl Radical in Live Cells Based on Gold Nanoclusters. Anal. Chem. 2014, 86, 1829−1836. (22) Martin, A.; Byrne, A.; Burke, C. S.; Forster, R. J.; Keyes, T. E. Peptide-Bridged Dinuclear Ru(II) Complex for Mitochondrial Targeted Monitoring of Dynamic Changes to Oxygen Concentration and ROS Generation in Live Mammalian Cells. J. Am. Chem. Soc. 2014, 136, 15300−15309. (23) Park, J. H.; Gu, L.; Maltzahn, G. V.; Ruoslahti, E.; Bhatia, S. N.; Sailor, M. J. Biodegradable Luminescent Porous Silicon Nanoparticles for In Vivo Applications. Nat. Mater. 2009, 8, 331−336. (24) Gu, L.; Hall, D. J.; Qin, Z.; Anglin, E.; Joo, J.; Mooney, D. J.; Howell, S. B.; Sailor, M. J. In Vivo Time-Gated Fluorescence Imaging with Biodegradable Luminescent Porous Silicon Nanoparticles. Nat. Commun. 2013, 4, 2326. (25) Joo, J.; Liu, X.; Kotamraju, V. R.; Ruoslahti, E.; Nam, Y.; Sailor, M. J. Gated Luminescence Imaging of Silicon Nanoparticles. ACS Nano 2015, 9, 6233−6241. (26) Nieto, A.; Hou, H.; Moon, S. W.; Sailor, M. J.; Freeman, W. R.; Cheng, L. Surface Engineering of Porous Silicon Microparticles for Intravitreal Sustained Delivery of Rapamycin. Invest. Ophthalmol. Vis. Sci. 2015, 56, 1070−1080. (27) Tzur-Balter, A.; Shatsberg, Z.; Beckerman, M.; Segal, E.; Artzi, N. Mechanism of Erosion of Nanostructured Porous Silicon Drug Carriers in Neoplastic Tissues. Nat. Commun. 2015, 6, 6208. (28) Wang, M.; Hartman, P. S.; Loni, A.; Canham, L. T.; Bodiford, N.; Coffer, J. L. Influence of Surface Chemistry on the Release of an Antibacterial Drug from Nanostructured Porous Silicon. Langmuir 2015, 31, 6179−6185. (29) Xia, B.; Wang, B.; Zhang, W.; Shi, J. High Loading of Doxorubicin into Styrene-Terminated Porous Silicon Nanoparticles via π-Stacking for Cancer Treatments. RSC Adv. 2015, 5, 44660− 44665. (30) Xia, B.; Wang, B.; Chen, Z.; Zhang, Q.; Shi, J. Near-Infrared Light-Triggered Intracellular Delivery of Anticancer Drugs Using Porous Silicon Nanoparticles Conjugated with IR820 Dyes. Adv. Mater. Interfaces 2016, 3, 1500715. (31) Kafshgari, M. H.; Cavallaro, A.; Delalat, B.; Harding, F. J.; McInnes, S. J.; Mäkilä, E.; Salonen, J.; Vasilev, K.; Voelcker, N. H. Nitric Oxide-Releasing Porous Silicon Nanoparticles. Nanoscale Res. Lett. 2014, 9, 333. (32) McInnes, S. J. P.; Turner, C. T.; Al-Bataineh, S. A.; Airaghi Leccardi, M. J. I.; Irani, Y.; Williams, K. A.; Cowin, A. J.; Voelcker, N. H. Surface Engineering of Porous Silicon to Optimise Therapeutic Antibody Loading and Release. J. Mater. Chem. B 2015, 3, 4123−4133.

ASSOCIATED CONTENT S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.7b02471. Figures S1−S11 (PDF)

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. ORCID

Jianmin Wu: 0000-0002-0999-9194 Author Contributions §

X. Chen and F. Wo contributed equally to this work.

Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS The authors acknowledge the financial support from National Natural Science Foundation of China (Grant No. 21575127) and Hangzhou GSPMED Medical Appliances Co. Ltd. REFERENCES (1) Gurtner, G. C.; Werner, S.; Barrandon, Y.; Longaker, M. T. Wound Repair and Regeneration. Nature 2008, 453, 314−321. (2) Martin, P.; Nunan, R. Cellular and Molecular Mechanisms of Repair in Acute and Chronic Wound Healing. Br. J. Dermatol. 2015, 173, 370−378. (3) Schneider, L. A.; Korber, A.; Grabbe, S.; Dissemond, J. Influence of pH on Wound-Healing: a New Perspective for Wound-Therapy? Arch. Dermatol. Res. 2007, 298, 413−420. (4) Schreml, S.; Szeimies, R. M.; Karrer, S.; Heinlin, J.; Landthaler, M.; Babilas, P. The Impact of the pH Value on Skin Integrity and Cutaneous Wound Healing. J. Eur. Acad. Dermatol. Venereol. 2010, 24, 373−378. (5) Ono, S.; Imai, R.; Ida, Y.; Shibata, D.; Komiya, T.; Matsumura, H. Increased Wound pH as an Indicator of Local Wound Infection in Second Degree Burns. Burns 2015, 41, 820−824. (6) Schäfer, M.; Werner, S. Oxidative Stress in Normal and Impaired Wound Repair. Pharmacol. Res. 2008, 58, 165−171. (7) Soneja, A.; Drews, M.; Malinski, T. Role of Nitric Oxide, Nitroxidative and Oxidative Stress in Wound Healing. Pharmacol. Rep. 2005, 57, 108−119. (8) Dröge, W. Free Radicals in the Physiological Control of Cell Function. Physiol. Rev. 2002, 82, 47−95. (9) James, T. J.; Hughes, M. A.; Cherry, G. W.; Taylor, R. P. Evidence of Oxidative Stress in Chronic Venous Ulcers. Wound Repair Regen. 2003, 11, 172−176. (10) Yeoh-Ellerton, S.; Stacey, M. C. Iron and 8-Isoprostane Levels in Acute and Chronic Wounds. J. Invest. Dermatol. 2003, 121, 918− 925. (11) Abd-El-Aleem, S. A.; Ferguson, M. W. J.; Appleton, I.; Kairsingh, S.; Jude, E. B.; Jones, K.; McCollum, C. N.; Ireland, G. W. Expression of Nitric Oxide Synthase Isoforms and Arginase in Normal Human Skin and Chronic Venous Leg Ulcers. J. Pathol. 2000, 191, 434−442. (12) James, T. J.; Hughes, M. A.; Hofman, D.; Cherry, G. W.; Taylor, R. P. Antioxidant Characteristics of Chronic Wound Fluid. Br. J. Dermatol. 2001, 145, 185−186. (13) Enache, T. A.; Oliveira-Brett, A. M. Pathways of Electrochemical Oxidation of Indolic Compounds. Electroanalysis 2011, 23, 1337− 1344. (14) McLister, A.; Davis, J. Molecular Wiring in Smart Dressings: Opening a New Route to Monitoring Wound pH. Healthcare (Basel) 2015, 3, 466−477. 7948

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949

Article

ACS Nano (33) Dardano, P.; Calio, A.; Politi, J.; Rea, I.; Rendina, I.; De Stefano, L. Optically Monitored Drug Delivery Patch Based on Porous Silicon and Polymer Microneedles. Biomed. Opt. Express 2016, 7, 1645−1655. (34) Gu, L.; Orosco, M.; Sailor, M. J. Detection of Protease Activity by FRET Using Porous Silicon as an Energy Acceptor. Phys. Status Solidi A 2009, 206, 1374−1376. (35) Li, X.; Tan, J.; Yu, J.; Feng, J.; Pan, A.; Zheng, S.; Wu, J. Use of a Porous Silicon-Gold Plasmonic Nanostructure to Enhance Serum Peptide Signals in MALDI-TOF Analysis. Anal. Chim. Acta 2014, 849, 27−35. (36) Joo, J.; Cruz, J. F.; Vijayakumar, S.; Grondek, J.; Sailor, M. J. Photoluminescent Porous Si/SiO2 Core/Shell Nanoparticles Prepared by Borate Oxidation. Adv. Funct. Mater. 2014, 24, 5688−5694. (37) 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. Status Solidi A-Appl. Mater. 2003, 197, 331−335. (38) DeLouise, L. A.; Fauchet, P. M.; Miller, B. L.; Pentland, A. A. Hydrogel-Supported Optical-Microcavity Sensors. Adv. Mater. 2005, 17, 2199−2203. (39) Zhuang, M.; Ding, C.; Zhu, A.; Tian, Y. Ratiometric Fluorescence Probe for Monitoring Hydroxyl Radical in Live Cells Based on Gold Nanoclusters. Anal. Chem. 2014, 86, 1829−1836. (40) Anderson, R. C.; Muller, R. S.; Tobias, C. W. Chemical Surface Modification of Porous Silicon. J. Electrochem. Soc. 1993, 140, 1393− 1396. (41) Lavine, J. M.; Sawan, S. P.; Shieh, Y. T.; Bellezza, A. J. Role of SiH and Si-H2 in the Photoluminescence of Porous Si. Appl. Phys. Lett. 1993, 62, 1099−1101. (42) Wu, J.; Sailor, M. J. Chitosan Hydrogel-Capped Porous SiO2 as a pH Responsive Nano-Valve for Triggered Release of Insulin. Adv. Funct. Mater. 2009, 19, 733−741. (43) Agnihotri, S. A.; Mallikarjuna, N. N.; Aminabhavi, T. M. Recent Advances on Chitosan-Based Micro- and Nanoparticles in Drug Delivery. J. Controlled Release 2004, 100, 5−28. (44) Krajewska, B. Diffusional Properties of Chitosan Hydrogel Membranes. J. Chem. Technol. Biotechnol. 2001, 76, 636−642. (45) Kaasalainen, M.; Rytkönen, J.; Mäkilä, E.; Närvänen, A.; Salonen, J. Electrostatic Interaction on Loading of Therapeutic Peptide GLP-1 into Porous Silicon Nanoparticles. Langmuir 2015, 31, 1722− 1729. (46) Wong, S. Y.; Manikam, R.; Muniandy, S. Prevalence and Antibiotic Susceptibility of Bacteria from Acute and Chronic Wounds in Malaysian Subjects. J. Infect. Dev. Countries 2015, 9, 936−944. (47) Healy, B.; Freedman, A. ABC of Wound Healing Infections. Br. Med. J. 2006, 332, 838−841. (48) Xu, S.; He, M.; Yu, H.; Cai, X.; Tan, X.; Lu, B.; Shu, B. A Quantitative Method to Measure Telomerase Activity by Bioluminescence Connected with Telomeric Repeat Amplification Protocol. Anal. Biochem. 2001, 299, 188−193. (49) Timms, A. E.; Zhang, Y.; Russell, R. G. G.; Brown, M. A. Genetic Studies of Disorders of Calcium Crystal Deposition. Rheumatology 2002, 41, 725−729. (50) Shaikh, A.; Berndt, T.; Kumar, R. Regulation of Phosphate Homeostasis by the Phosphatonins and Other Novel Mediators. Pediatr. Nephrol. 2008, 23, 1203−1210.

7949

DOI: 10.1021/acsnano.7b02471 ACS Nano 2017, 11, 7938−7949