BiomaterialsScience

PAPER

Cite this: Biomater. Sci., 2018, 6,2460

Received 30th May 2018,Accepted 17th July 2018

DOI: 10.1039/c8bm00602d

rsc.li/biomaterials-science

Construction of perfluorohexane/IR780@liposomecoating on Ti for rapid bacteria killing underpermeable near infrared light†

Xiuhua Wang,a Lei Tan,a Xiangmei Liu,*a Zhenduo Cui,b Xianjin Yang,b

Kelvin W. K. Yeung,c Paul K. Chud and Shuilin Wu *a,b

Near infrared (NIR) light induced photodynamic antibacterial therapy (PDAT) is a promising antibacterial

technique in rapid in situ disinfection of bacterially infected artificial implants due to its penetration ability

into tissues. However, the lower oxygen content in vivo may restrict the yields of reactive oxygen species

(ROS), thus reducing the antibacterial efficacy of PADT significantly. Herein, liposome encapsulated

photosensitizers (PS), IR780 and perfluorohexane (PFH), have been constructed on the surface of Ti

implants via a covalent linkage to overcome this issue. Thanks to the high oxygen capacity of PFH, more

ROS can be generated during NIR irradiation regardless of the low content of oxygen in vivo. As a result,

in vitro tests demonstrated that 15 minutes of 808 nm near-infrared irradiation could achieve a high anti-

bacterial efficacy of 99.62% and 99.63% on the implant surface against Escherichia coli and

Staphylococcus aureus, respectively. By contrast, the PDAT system without PFH modification shows a

lower antibacterial efficacy (only 66.54% and 48.04%, respectively). In addition, this enhanced PDAT

system also possesses great biocompatibility based on the in vitro and in vivo subcutaneous assays. This

surface system makes it possible for rapid bacteria-killing in artificial implants that have been implanted

in vivo under local conditions with lower oxygen content.

1. Introduction

Titanium (Ti) and its alloys are mainstream implant materialsin orthopedics and dentistry for their superior mechanical pro-perties and biocompatibility.1–4 However, delayed healing,implant failure, and repeated surgeries still occur due toimplant-associated bacterial infection and ensuing compli-cations.5,6 Therefore, an artificial implant with superior anti-microbial characteristics is urgently needed. The implantsurface can be endowed with superior antimicrobial propertiesby various surface modification and coating strategies includ-

ing direct deposition with antibiotics and encapsulation withinorganic bactericides like Ag, Cu, and ZnO or the introduc-tion of organic antibacterial polymers.7–11 However, themisuse/abuse of antibiotics will inevitably result in multidrug-resistant pathogens.12 Moreover, the toxicity of inorganic bac-tericides or organic antibacterial polymers hampers widercommercial applications,13,14 and it is necessary to developstrategies to control implant-associated bacterial infection.

A promising means to treat infection-related diseases isphotodynamic antibacterial therapy (PDAT), which can avoidthe emergence of drug-resistant bacteria.15,16 This strategy isbased on the application of photosensitizers (PS) to transferthe energy of the appropriate wavelength to the surroundingoxygen atoms to generate reactive oxygen species (ROS) such assinglet oxygen (1O2), which are cytotoxic and can destroymalignant tissues and pathogens.17 Compared with conven-tional therapies, PDAT possesses several advantages includingminimal invasiveness and fast recovery.18,19 Furthermore, thelight penetration depth and light toxicity decrease with longerwavelength; so near infrared (NIR) is generally the preferredexcitation light.20 However, several drawbacks related to con-ventional PDAT impeded their practical clinical appli-cation.21,22 Generally, the aqueous solubility of most PS is verylow, which hinders their practical applicability in biomedical

†Electronic supplementary information (ESI) available. See DOI: 10.1039/c8bm00602d

aHubei Collaborative Innovation Center for Advanced Organic Chemical Materials,

Ministry-of-Education Key Laboratory for the Green Preparation and Application of

Functional Materials, Hubei Key Laboratory of Polymer Materials, School of

Materials Science & Engineering, Hubei University, Wuhan 430062, ChinabSchool of Materials Science & Engineering, Tianjin University, Tianjin 300072,

China. E-mail: shuilin.wu@gmail.com, shuilinwu@tju.edu.cn,

liuxiangmei1978@163.comcDepartment of Orthopaedics & Traumatology, Li KaShing Faculty of Medicine,

The University of Hong Kong, Pokfulam, Hong Kong 999077, ChinadDepartment of Physics and Department of Materials Science and Engineering,

City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong 999077, China

2460 | Biomater. Sci., 2018, 6, 2460–2471 This journal is © The Royal Society of Chemistry 2018

www.rsc.li/biomaterials-sciencehttp://orcid.org/0000-0002-1270-1870http://crossmark.crossref.org/dialog/?doi=10.1039/c8bm00602d&domain=pdf&date_stamp=2018-08-15

fields.23,24 On the other hand, low ROS yields of PS also inevi-tably results in the finite efficacy of PDAT.25

In order to deal with these problems, the encapsulation ofPS into nanostructured drug emulsion systems, such as lipo-somes, micelles and mesoporous silica, is very promising.26–28

Although the water solubility of PS can be increased obviouslythrough these strategies, the incorporation of PS into a nano-emulsion system will result in the aggregation of PS; then self-quenching of PS aggregates occurs, which eventually leads tothe decline of ROS generation yield.29,30 Therefore, a novelstrategy to fabricate another PDAT system with superior watersolubility and enhanced ROS generation yield is urgentlyrequired. Several studies have developed supramolecular den-drimer PS31,32 or PS based metal organic frameworks (MOFs)33

to avoid the aggregation of PS. However, the complicated syn-thesis process, the biocompatibility and biosafety issuescannot be ignored in practical applications.

Perfluorohexane (PFH), as a kind of traditional oxygencarrier, has been approved by the U.S. Food and DrugAdministration (FDA) and it has been applied as artificialblood in clinical surgery.34 It has been reported that the ROSlifetime in the super-hydrophobic environment is longer thanthat in the hydrophilic environment.35 Due to the super-hydrophobic nature of PFH, the lifetime of ROS aroundPFH is extended significantly. Therefore, the design of thePDAT system involving PFH may improve the ROS generationyield.

Keeping all these issues in mind, a NIR excited PDATsystem with enhanced efficiency was fabricated by introducingPFH and IR780 (a commercial NIR hydrophobic dye) to thehydrophobic core of a liposome denoted as Lip(IR780 + PFH).The bare liposome and IR780 encapsulated liposome can bedenoted as Lip and Lip(IR780), respectively. Then this PDATsystem is further grafted on Ti implants through covalentlinkage between the phosphate radical of the liposome andthe hydroxyl group on Ti (indicated as the enhanced PDATsystem) (Scheme 1). Lip, Lip(IR780) and Lip(IR780 + PFH)modified Ti can be denoted as Lip-Ti, Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti, respectively.

It is widely accepted that PFH possesses superior oxygen-storage capacity, and is able to maintain a higher oxygencontent than other solvents, approximately 20 fold than that ofwater under the same oxygen partial pressure.36 Furthermore,tissue hypoxia is common under bacterial infections or inflam-mations, thus severely impairing the effectiveness of PDAT.37

This tissue hypoxia issue during bacterial infection andinflammation can also be solved by the presence of a PFHmodified PADT system. So in this work, we built a sub-cutaneous in vivo system as a hypoxia model to explore theROS generation ability and PDAT activity of different samplesof this novel system.38,39 Compared with traditional nano-PDAT systems and previous studies, an enhanced antibacterialeffect was achieved especially in the antibacterial field underthe condition of lower dosage and hypoxia environment. Abetter PDAT efficiency can be achieved through this enhancedPDAT platform.

The aims of the present work were: (1) to develop andcharacterize a stabilized and safe Lip(IR780 + PFH)-Ti modi-fied antibacterial implant system; (2) to assay the antibacterialability of IR780 under hypoxia conditions in the presence ofPFH and NIR irradiation in vitro; (3) to evaluate in vitro howMC3T3-E1 cells react with the liposome modified Ti implants;and (4) to evaluate the physiological responses of subdermaltissues upon different sample exposures in a rat model inin vivo subcutaneous tests.

2. Experimental section2.1 Preparation of samples

2.1.1 Preparation of Lip, Lip(IR780), and Lip(IR780 + PFH).The preparation of Lip, Lip(IR780) and Lip(IR780 + PFH) wascarried out according to the methods described previouslywith some modification.40 Briefly, the lipid films were formedby mixing 24.7 mg of lecithin (Aladdin Reagent Co., China),4.3 mg of cholesterol (Aladdin Reagent Co., China), 3.8 mg ofDSPE-PEG2000 (A.V.T. Pharm. Ltd Shanghai, China), and0.1 mL of IR780 (2 mg mL−1, dissolved in dichloromethane,Sigma Co., USA) in a 25 mL round bottom flask and evapor-ated to remove dichloromethane. Then 1.4 mL of pure waterwas added to peel off the lipid film in a water bath at 30 °C.The homogeneous liposome dispersion was formed with soni-cation for 10 min using an ultrasonic homogenizer (JY92-IIN,China) (15%) in an ice bath. Lip(IR780 + PFH) was prepared bygradually adding 0.6 mL PFH and sonicating for 3 min (50%)in an ice bath. Lip(IR780) was formed by the addition of purewater instead of PFH. The pure liposome was prepared bysimilar procedures in the absence of PFH and IR780. Lip(PFH)was formed without the addition of IR780. Finally, the

Scheme 1 Schematic illustration of the difference of ROS generationbetween Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti irradiated by NIR. ThePDAT mechanism of Lip(IR780)-Ti is associated with the disruption ofthe cell membrane integrity while that of Lip(IR780 + PFH)-Ti can beattributed to the serious cell shrinkage and appearance of holes (pits)(red arrow).

Biomaterials Science Paper

This journal is © The Royal Society of Chemistry 2018 Biomater. Sci., 2018, 6, 2460–2471 | 2461

dispersion was extruded for 15 times by 0.22 μm. The PESmembrane and the dispersions were stored at 4 °C in theabsence of light. The concentration of IR780 in the liposomewas determined with UV-Vis absorption spectrophotometryusing a standard curve method.

2.1.2 Preparation of Ti plates. Pure Ti plates (d = 6 mm,h = 2 mm) were mechanically polished according to standardprocedures (MP-Ti) and the MP-Ti was hydrothermally treatedat 80 °C in 2 M KOH for 4 h to obtain the alkali heat treatedTi (AHT), which was cleaned with deionized water and dried inair for further use.

2.1.3 Preparation of liposome-modified AHT Ti plates. Thesolutions with different liposomes were diluted with deionizedwater to a final concentration of IR780 of 40 µg mL−1; then20 µL of the diluted dispersion was dropped onto the AHT anddried at 4 °C in darkness. The liposome dropped AHT waswashed with deionized water and stored at 4 °C in a moist anddark environment for further use.

2.2 Sample characterization

The existence of PFH in Lip(IR780 + PFH) was confirmed withgas chromatography-mass spectrometry (GC-MS) and19F nuclear magnetic resonance (NMR). The morphology ofthe encapsulated liposome nanostructure was examined withtransmission electron microscopy (TEM, Tecnai G20, USA) andthe functionalized liposome modified Ti disks were character-ized with scanning electron microscopy (SEM). The particlesize and zeta potential were determined using a MalvernZetasizer ZS™ (Malvern Instruments, UK) and compositionalanalysis was conducted using X-ray photoelectron spectroscopy(XPS, Thermo Fisher Scientific 250Xi) and energy-dispersiveX-ray spectroscopy (EDS).

2.3 PFH loading and O2 release measurement

For oxygen loading, since the total volume of the freshly pre-pared liposome is 2 mL, 80 μL liposome (Lip; Lip(IR780) andLip(IR780 + PFH)) dispersion or 40 μL pure PFH wereimmersed in 200 μL deoxygenated water (using N2 as aremover of oxygen) in a 48-well plate and pure water was usedas the control group; the oxygen concentration in the waterwas detected by a dissolved oxygen assay kit immediately; thenan oxygen flow was fed into the water slowly to achieve oxygensaturation until the oxygen concentration reached an equili-brium, and the final oxygen saturation and oxygen loadingcapacity were calculated using a reported method.41,42

The PFH loading efficiency of Lip(IR780 + PFH) can be calcu-lated as this equation: the PFH loading efficiency (%) ={[Vpure PFH added × (CLip(IR780+PFH) increased − CLip(IR780) increased)]/Cpure PFH added} × [2000/80]/Vall PFH added × 100%.

To measure oxygen release, different samples (Lip-Ti, Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti) were placed in 200 μLdeoxygenated water (using N2 as a remover of oxygen) in48-well plates. The oxygen concentration was determinedusing a dissolved oxygen assay kit and the O2 concentration inreal time was recorded.

2.4 PDAT activity in vitro

2.4.1 Detection of ROS. ROS generation under NIRirradiation was detected by an ROS sensitive sensor, dichloro-fluorescein diacetate (DCFH-DA), using reported methods.43

Briefly, DCFH-DA was converted to dichlorofluorescein (DCF)by hydrolysis with NaOH and neutralized with PBS. 100 μL ofthe dispersions (10 μg mL−1) or modified Ti samples (immer-sing in 100 μL deionized water) were mixed with 20 μL of theabove DCFH (10 μM) and exposed to NIR light for differentintervals with a power of 0.5 W cm−2. The fluorescence inten-sity of highly fluorescent DCF converted by DCFH-DA wasdetermined using an excitation wavelength of 488 nm andemission wavelength of 525 nm on a microplate reader(SpectraMax I3, MD, USA).

2.4.2 Antibacterial activity in vitro. The antibacterialactivity against Gram-negative Escherichia coli (E. coli, ATCC8099) and Gram-positive Staphylococcus aureus (S. aureus,ATCC 25923) was evaluated by the spread plate method, a Live/Dead assay and bacteria morphology in Luria Bertani (LB)medium,44 and the LB agar plates were sterilized by autoclav-ing at 120 °C for 15 min. The samples were placed on a 96-wellplate and 200 μL of the diluted bacterial suspension (1 × 107

CFU mL−1)45,46 were added to each well. Part of the samples(MP-Ti, AHT, Lip-Ti, Lip(IR780)-Ti, and Lip(IR780 + PFH)-Ti)were covered with an aluminum foil to exclude the influenceof light. Two samples (Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti)were irradiated with NIR light (808 nm, 1 W cm−2)47,48 for15 min. The suspension of E. coli and S. aureus was diluted100 and 200 times, respectively, and 10 μL of the diluted sus-pension was spread uniformly on the agar medium and incu-bated in a bacteria incubator at 37 °C for 24 h.

The antibacterial efficiency was quantified by the numberof colonies on the plates by the following equation and theexperiment was performed in triplicate and independently:Antibacterial ratio (%) = (the number of CFUs in the controlgroup − the number of CFUs in the experimental group)/(thenumber of CFUs in the control group) × 100%.

The antibacterial activity was further evaluated using Live/Dead fluorescence staining. After the bacteria on the sampleswere treated according to the above procedures, the bacteriawere washed with PBS three times and stained with propidiumiodide (PI, dead cells, red fluorescence) and acridine orange(AO, live cells, green fluorescence) for fluorescence microscopyto distinguish the living and dead cells. Each sample wasstained for 15 min in the absence of light and observed undera fluorescence microscope (IX73, Olympus, USA). The livebacteria cells were stained green while dead ones were stainedred.

SEM was utilized to investigate the morphology and earlyadhesion behavior of E. coli and S. aureus on the Ti disks. Afterthe bacterial suspension was treated using the aboveprocedures, the bacteria on the Ti disks were fixed with 2.5%glutaraldehyde for 2 h, rinsed with PBS 3 times, dehydratedwith ethanol solutions (30, 50, 70, 90, and 100 v/v%) for15 min and dried in air for SEM observation.

Paper Biomaterials Science

2462 | Biomater. Sci., 2018, 6, 2460–2471 This journal is © The Royal Society of Chemistry 2018

2.5 Cytocompatibility assay

The cell line MC3T3-E1 was used in this work. The cells wereusually cultured in a Petri dish in a medium composed ofα-MEM (HyClone) supplemented with 1% penicillin–strepto-mycin (HyClone) and 10% fetal bovine serum under an atmo-sphere of 5% CO2 at 37 °C in an incubator. The cells weretreated with sterilized PBS and trypsin/ethylenediaminetetra-acetic acid (Sigma) to remove the dead ones and then dilutedto a density of 1 × 105 cells per mL for further use.

2.5.1 MTT assay. After incubating the samples in 96-wellplates with 200 μL diluted cell suspensions for differentperiods (1 day, 3 days and 7 days), 0.5 mg mL−1 3-(4,5-di-methylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT,Aladdin Reagent Co., China) were added and the cells wereincubated at 37 °C for 4 h and refreshed with 200 μL DMSO todissolve the purple precipitate. Dissolution was completed byvibrating the clusters for 10 min and the optical density at490 nm was recorded on a microplate reader. The optical pro-cessing MTT assay was conducted in the same manner asabove in addition to light irradiation. Briefly, at the beginningof each culturing period (1 day, 3 days and 7 days), the cellswere attached fully on the samples; then each sample was irra-diated for 15 min (808 nm, 1 W cm−2) and then treated withMTT after culturing. The cell viability was calculated by theratio of the experimental group to the control group. Theculture medium was refreshed every 3 days, and experimentswere done on three parallel groups.

2.5.2 Adhesion and spreading assay. The diluted cell sus-pension was seeded on different samples on a 48-well plate.After incubation for 24 h, the cells were attached on the surfaceof implants homogeneously and fixed with 4% formaldehyde(Sinopharm Chemical Reagent Co., China) for 10 min at roomtemperature after rinsing. The cells were stained withFITC-Phalloidin (YiSen, China) in darkness for 30 min and thenincubated with 4,6-diamidino-2-phenylindole dihydrochloride(DAPI, YiSen, China) for 30 s. The cell adhesion and spreadingbehavior was visualized using fluorescence microscopy.

2.5.3 Cell and spreading assay. The cell morphology in theearly stage was evaluated using SEM. After the samples wereco-cultured with different samples, the cells on the sampleswere fixed according to the above procedures for SEM.

2.6 In vivo studies

In vivo assays were implemented to evaluate the effectivenessof Lip(IR780 + PFH) on titanium surfaces against bacterialcolonisation49 and all the animal experiments and procedureswere approved by Hubei Provincial Centers for DiseasePrevention & Control. All animals were maintained and usedin accordance with the Animal Management Rules of theMinistry of Health of the People’s Republic of China and theGuidelines for the Care and Use of Laboratory Animals ofChina. The samples were subcutaneously implanted intopathogen-free male Wistar rats (200 g) obtained from HubeiProvincial Centers for Disease Prevention & Control. The ratswere individually raised in cages at a standardized temperature

for 2 days and randomly divided into three groups as follows:MP-Ti, Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti. Before implant-ing into the rats’ subcutaneous tissue, 4 wt% pentobarbitalsodium salt (1 mL kg−1) was used as an anesthetic. The hairon the right back was shaved and disinfected with 1% povi-done iodine, and sterilized surgical instruments were used tocut the epidermis of the rats’ back until the tissue wasexposed. The Ti plates that were seeded with 20 μL of S. aureus(108 CFU mL−1) on the surface were implanted and then theincision was stitched. The other rats underwent the same sur-gical protocol, and all rats were housed in individual sterilizedcages. The rats were raised for 1 day so that the surgical siteswere infected by S. aureus and the tissue related infection andthe inflammation induced tissue hypoxia appeared. Theinfected surgical sites of one group were illuminated with NIRlight (808 nm, 1 W cm−2) for 15 min and the rats were eutha-nized on the second day. The other groups in the dark wereeuthanized after 2 days. The implants were removed and 50 µLLB media was injected. After fully washing the surgical sites,10 µL tissue fluids were extracted. The number of bacteria inthe tissue fluids was quantified by diluting the tissue fluids400 times with LB media and spreading 10 μL onto an agarplate. The tissues surrounding the implants were extractedfrom each rat and fixed in 4% formalin for 24 h. The tissueswere subjected to a routine histological analysis and were cutinto 4 μm thick transverse sections. The sections were stainedwith hematoxylin–eosin (HE) and Giemsa.

3. Results and discussion3.1 Synthesis and characterization of liposomes

The liposome based drug delivery system is designed based onthe large loading capacity of hydrophobic molecules and excel-lent biocompatibility.50,51 The commercial NIR hydrophobicIR780 as the photosensitizer (Fig. 1a) and PFH approved by the

Fig. 1 (a) Chemical structure of IR780. (b) UV-Vis absorption spectra ofIR780 and different liposomes. (c) 19F-NMR spectra of pure PFH and Lip(IR780 + PFH). (d) DLS spectra of different liposomes.

Biomaterials Science Paper

This journal is © The Royal Society of Chemistry 2018 Biomater. Sci., 2018, 6, 2460–2471 | 2463

FDA as the oxygen reservoir are incorporated into thePEGylated liposome (Fig. S1†). Lip(IR780) is prepared byhydrating the dried lipid films formed by mixing lecithin,cholesterol, DSPE-PEG2000, and IR780 at a molar ratio of7 : 1 : 0.1 : 0.01 with deionized water followed by sonication(10 min, 15%). PFH is introduced to the liposome by addingPFH dropwise to the hydrated lipid film under sonication(3 min, 50%).

The encapsulation of IR780 is confirmed by the absorptionpeak at 780 nm close to the peak of bare IR780 (Fig. 1b).52 Thered-shift in the absorption peaks of IR780 in Lip(IR780) andLip(IR780 + PFH) stems from the hydrophobic interaction andchanges in the polarity in the hydrophobic core of the lipo-some and around PFH. Although IR780 is insoluble in water(water solubility equal to 0.4 μg mL−1), the water solubility canbe increased to 100 μg mL−1 after encapsulation. The color ofaqueous Lip(IR780) and Lip(IR780 + PFH) turns to greenwithout aggregation while the incorporation of PFH rendersthe solution slightly turbid (inset image in Fig. 1b).

The existence of PFH in Lip(IR780 + PFH) was confirmedusing GC-MS. The retention time of Lip(IR780 + PFH) (0.927 s)is very close to that of pure PFH (0.922 s) (Fig. S2†). Moreover,the MS spectra are similar and the molecular weights (MW)are equal to that of PFH. The existence of PFH in the liposomewas corroborated using 19F NMR. Intense signals are observedfrom the 19F NMR spectrum of Lip(IR780 + PFH) and thechemical shift is identical to that of pure PFH confirming theencapsulation of PFH in the liposome (Fig. 1c).

The hydrodynamic (HD) sizes of Lip, Lip(IR780) and Lip(IR780 + PFH) determined using DLS are 122.4, 190.1, and255.0 nm, respectively (Fig. 1d). The obvious increase in theHD size may be associated with the incorporation of IR780and PFH in the hydrophobic core of the liposome. Besides, thezeta potentials of pure Lip, Lip(IR780) and Lip(IR780 + PFH)are −36.7, −32.2 and −42 mV, respectively (Fig. S3†),suggesting minor changes to the surface charge of Lip, whichreveals little aggregation of the liposome.

The photostability was studied by acquiring the UV-Visabsorption spectra from IR780, Lip(IR780), and Lip(IR780 +PFH) right after the dilution process or storage for 1 day(under light illumination or in darkness) as shown in Fig. S4.†The maximum absorbance at 780 nm of IR780 decreases afterstorage at room temperature for 1 day and the degree of thedecline is greater under light illumination, implying thatIR780 is unstable in water and sensitive to light. However, themaximum absorbance of Lip(IR780) and Lip(IR780 + PFH) at780 nm is almost constant under light or in the dark,especially for Lip(IR780 + PFH), indicating that both the lipo-some and PFH improve the dispersity of Lip(IR780 + PFH) inwater and stability as well as reduce the self-quenching ofIR780, which is eligible for further biological experiments.

3.2 Synthesis and characterization of liposome modified Ti

As shown in Fig. 2a, the as-prepared liposome on Ti platesexhibits a uniform, quasispherical, monodisperse morphologyand the trend of size is in accordance with the DLS results. It

has been reported that the liposome absorbs onto titaniumoxide via a covalent linkage formed by the phosphate oxygenof the liposome attacking the surface hydroxyl group of Ti.53,54

In order to obtain superior PDAT activity, the mechanicallypolished Ti (MP-Ti) first undergoes an alkali heat treatment(sample labeled as AHT) to form a hydroxyl enriched surface55

and the suspension of a different liposome is then added ontothe surface of AHT and incubated for 12 h to form strongercovalent bonds between the liposome and AHT. The liposomefunctionalized Ti disks are characterized using SEM and asshown in Fig. 2b, compared to AHT with a polyporous net-likestructure, a solid lipid nanofilm is formed on AHT, which isconsistent with the lipid fusion mechanism.56 Indeed, IR780and PFH are co-encapsulated within the lipid bilayer of theliposome; so, the structures of Lip, Lip(IR780) and Lip(IR780 +PFH) show no obvious differences, which is also consistentwith Fig. S1† and Fig. 2a. Lipid fusion based on spontaneouslipid adsorption and rupture is widely exploited to fabricateplanar nanofilms efficiently and the covalent coupling mecha-nism is shown schematically in Fig. S5.†

The surface morphology is investigated using AFM(Fig. S6†). The surface of MP-Ti is very flat but that of AHT isrougher. With regard to the liposome modified Ti disks, someparts of the surface are higher but other parts are flatter thanthat of AHT as shown in Table S1.† The roughness of MP-Ti is14.31 nm and that of AHT is 71.41 nm in accordance with themorphological examination. The surface roughness of Lip-Ti,Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti decreases to less than10 nm (Table S1†), further proving the formation of planarnanofilms using the lipid fusion mechanism.

The surface chemical compositions are determined usingEDS and XPS. As shown in Fig. S7,† the signals of C, N, O, P,and Ti in Lip-Ti prove the absorption of the liposome on thesurface of AHT. Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti alsoshow C, N, O, P and Ti, confirming the presence of the lipo-some. Although Cl and I (composition of IR780) are notdetected with EDS due to being present in trace amounts, XPSconfirms their presence (Fig. S8†). For both Lip(IR780)-Ti andLip(IR780 + PFH)-Ti, the I 3d peaks at 618.3 and 629.8 eV indi-cate the presence of IR780 on AHT. The absence of Cl may becaused by the binding energy of Cl 2p overlapping that of the

Fig. 2 (a) TEM morphologies of different liposome nanospheres (scalebar = 0.5 μm). (b) SEM of AHT and different liposome modified Ti disks(scale bar = 1 μm).

Paper Biomaterials Science

2464 | Biomater. Sci., 2018, 6, 2460–2471 This journal is © The Royal Society of Chemistry 2018

stronger P 2p signal. The XPS spectra only reflect the surfacecomposition and so Ti cannot be detected and fluorine is notdetected with EDS and XPS due to the low boiling point ofPFH (57 °C). All in all, the liposome has successfully beendeposited onto the surface of the Ti plates.

3.3 Oxygen loading and releasing measurement of PFH

The oxygen loading of different liposomes under a slow oxygenflow in deoxygenated water is shown in Fig. S9a;† before theaddition of oxygen, Lip, Lip(IR780) and Lip(IR780 + PFH)exhibit a similar O2 concentration consistent with the controlgroup, and after the addition of oxygen, once the dissolvedoxygen concentration reaches equilibrium, the correspondingoxygen concentrations of Lip and Lip(IR780) change incon-spicuously, but it can be found that the initial O2 concen-tration of Lip(IR780 + PFH) is much greater than before, andthus we can calculate that the oxygen saturation of Lip(IR780 +PFH) is 8.67 mg L−1, and the oxygen loading capacity of purePFH is 9.67 mg L−1; besides, the PFH loading efficiency in Lip(IR780 + PFH) is 67.70%, indicating a fine encapsulatingability in the liposome and a good oxygen carrying capacity ofPFH.

For the oxygen releasing measurement (Fig. S8b†), aremarkable increase in oxygen content is shown in the Lip(IR780 + PFH)-Ti group owing to the oxygen enrichment ofPFH, even in an oxygen-deficient environment; compared withLip(IR780)-Ti, Lip(IR780 + PFH)-Ti possesses a higher and amore gradual oxygen releasing ability. Therefore, Lip(IR780 +PFH)-Ti can be a good candidate as an enhanced PDAT agent.

3.4 In vitro antibacterial activity

3.4.1 ROS detection in vitro. As far as we know, the activityof PDAT strongly depends on the generation of ROS to destroybacteria; thus the PDAT efficiency can be determined using theROS generation yield. So the ROS sensor, dichlorofluorescein(DCFH), is applied to determine the in vitro ROS generationability of a different liposome in hypoxia PBS. Oxygen in PBSis consumed by treating with sodium thiosulfate (20 μg mL−1)to obtain hypoxia PBS and in the presence of ROS,57 the oxi-dation of DCFH increases the fluorescence intensity andallows the determination of ROS generation.58 As shown inFig. 3a, before loading on Ti plates, after irradiation for 15 minwith 2 min intervals in hypoxia PBS, the ROS produced by Lipis in extremely low amounts and that generated by Lip(IR780)is a little higher. However, Lip(IR780 + PFH) shows the highestfluorescence intensity, which is much higher than that of Lip(IR780).

As for this modified PDAT system, the microenvironment ofIR780 can be altered obviously by PFH; then more ROS will beproduced through the photo-excitation reactions.59

Furthermore, compared with other solvent conditions, gener-ated ROS can be dissolved by PFH; then the lifetime of ROSwill be extended and it can also be prevented from quenchingby PFH.60 In the other aspects, it is widely accepted that PFHcan maintain a higher oxygen content in identical oxygenpartial pressure, regardless of pre-existing hypoxia, while the

effectiveness of PDAT relies largely on the local oxygencontent. Therefore, the reason for enhanced ROS generationcan be understood through ROS protective effects and thehigh oxygen capacity of PFH.

After loading on Ti plates, the ROS generation efficiencyunder hypoxia conditions is also determined. As shown inFig. 3b, negligible ROS production is observed from MP-Ti,AHT, and Lip-Ti. ROS production from Lip(IR780)-Ti is quiteweak but DCFH fluorescence is detected from Lip(IR780 +PFH)-Ti, indicating unaffected ROS production under hypoxiaconditions. The results indicate that Lip(IR780 + PFH)-Ti canenhance the ROS generation via the extended ROS lifetime andenriched oxygen around PFH (Scheme 1).

3.4.2 Antibacterial activity in vitro. The PDAT performanceagainst E. coli and S. aureus as model bacteria is assessed bystudying colony formation on LB agar plates. The bacterial sus-pension is diluted with deoxygenated LB medium and paraf-ilm is used to isolate hypoxia environment from air, followedby NIR irradiation (808 nm, 1 W cm−2, 15 min) and incubationfor 24 h on LB agar medium at 37 °C. It is confirmed that PFHitself encapsulated in the liposome has no antibacterialactivity with or without NIR irradiation (Fig. S9†). As shown inFig. S10,† the number of colonies for the bacterial suspensiontreated with NIR irradiation alone is very close to that forMP-Ti, suggesting that NIR irradiation by itself has no effectson the bacterial cells. Comparable viable colonies are observedfrom both AHT and Lip-Ti on the agar plates after co-culturingwith bacteria in the absence of NIR and the situation is similarfor Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti in the absence ofNIR, verifying that AHT, liposome, IR780, or PFH separatelyhave little negative effects on bacterial activity. When combin-ing Lip(IR780)-Ti with NIR illumination, an obvious decline inthe colony number is observed but not all the bacteria aredestroyed. However, when PFH is incorporated, the ROS yieldwas improved under NIR irradiation, and colony formation onLip(IR780 + PFH)-Ti is almost completely inhibited. The anti-bacterial ratio is quantified by counting the colony number onthe plate as shown in Fig. 3c. The bacterial killing efficacy ofLip(IR780)-Ti under NIR irradiation is 66.54% and 48.04% forE. coli and S. aureus, respectively. As a contrast, after PFHencapsulation, a better 1O2 yield was achieved owing to itshigh O2 content, and the corresponding values of Lip(IR780 +PFH)-Ti under the same condition are 99.62% and 99.63%,indicating that the addition of PFH can enhance the photo-dynamic antibacterial efficacy of Lip(IR780) under 808 nmlight irradiation.

The fluorescence based Live/Dead staining assay is adoptedto evaluate PDAT performance. The Live/Dead assay simul-taneously visualizes living and dead bacteria based on thedifferent penetrating abilities of two dyes in bacteria.61 Asshown in Fig. 3d, the red spots on the samples without NIRirradiation are negligible and most of the fluorescence imagesare decorated with green spots indicating the inability ofMP-Ti and different liposome capped Ti to kill bacteria in theabsence of NIR. However, most bacteria that adhered to theLip(IR780)-Ti irradiated with NIR are partially red stained and

Biomaterials Science Paper

This journal is © The Royal Society of Chemistry 2018 Biomater. Sci., 2018, 6, 2460–2471 | 2465

almost all the bacteria on Lip(IR780 + PFH)-Ti illuminated byNIR are completely red stained because the generation of ROSdisrupts cell membrane integrity. The Live/Dead assay is con-sistent with the spread plate method indicating that PFH canextend the lifetime of ROS and enrich oxygen to accelerateROS production from IR780, resulting in an obvious improve-ment of PDAT efficiency against both E. coli and S. aureus.

The results are also confirmed by the morphologicalchanges induced by ROS in the bacteria as examined usingSEM. As shown in Fig. 3e, both E. coli and S. aureus on MP-Tiwithout NIR have the typically rod and spherical shape,respectively, with smooth and integrated cell walls. A similarshape can be observed in both bacteria seeded on Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti in the absence of NIR. However,when Lip(IR780)-Ti is illuminated with NIR, the shape of thecells changes obviously as shown by the disruption of cellmembrane integrity (white arrow). As for the cells seeded on

Lip(IR780 + PFH)-Ti illuminated by NIR, a complete defor-mation of the cellular structure is visible. That is, increasedmembrane roughness, serious cell shrinkage, and appearanceof holes (pits) (black arrow) are observed from the cell mem-brane, indicative of cell membrane perforation.

It is likely that the different amounts of ROS generationinduced by a longer ROS lifetime and the oxygen self-enrichedbehavior of PFH may contribute to the significant difference inPDAT activity. It has been reported that ROS is the main factorin photo-cytotoxicity, which can damage the treated tissue,causing the oxidation and degradation of the cell membrane.62

So fewer ROS produced by Lip(IR780)-Ti can merely affect cellmembrane integrity while abundant ROS generated by Lip(IR780 + PFH)-Ti may cause serious cell shrinkage and theappearance of holes (pits) on the cell membranes through bio-membrane oxidation and degradation; this is illustrated inScheme 1.

Fig. 3 (a) 1O2 production from different liposome nanospheres in PBS and (b)1O2 production of MP-Ti, AHT and different liposome modified Ti

disks under NIR irradiation as determined using DCFH (mean ± SD, n = 3). The in vitro antibacterial assay of different Ti disks. (c) Antibacterial ratiohistogram of E. coli and S. aureus on different samples under irradiation of NIR for 15 min or in darkness determined by the number of colonies inthe plate samples (mean ± SD, n = 3, *P < 0.05, **P < 0.01, ***P < 0.001 (t-test)). (d) Live/Dead fluorescent images of E. coli and S. aureus incubatedwith different samples under NIR irradiation for 15 min or in the dark (scale bar = 50 μm). (e) Morphology of E. coli and S. aureus incubated withdifferent samples under NIR irradiation for 15 min or in darkness (scale bar = 1 μm).

Paper Biomaterials Science

2466 | Biomater. Sci., 2018, 6, 2460–2471 This journal is © The Royal Society of Chemistry 2018

3.5 In vitro biocompatibility evaluation

3.5.1 MTT assay. The cytotoxicity to the preosteoblastMC3T3-E1 cells of mice was investigated using the MTT assay.The MC3T3-E1 cells are seeded onto the samples and culturedfor 1, 3 and 7 days. As shown in Fig. 4a, compared to theuntreated samples (MP-Ti), the cell viability on AHT and Lip-Tiis comparable (94%–100% for AHT and 85%–92% for Lip-Ti),demonstrating the good biocompatibility of AHT and liposome.When the cells are seeded on Lip(IR780)-Ti and Lip(IR780 +PFH)-Ti, the corresponding cell viability remains almost the

same indicating no significant cytotoxicity of Lip(IR780)-Ti andLip(IR780 + PFH)-Ti towards MC3T3-E1 cells. In addition, thephototoxicity of different samples has been evaluated. As shownin Fig. 4b, compared with the untreated Ti, there is a slightdownside to cell viability in the first day due to the impact ofthe cytotoxicity of ROS induced by NIR irradiation, and soon anobvious rise in the next 3 and 7 days (over 100% for Lip(IR780)and Lip(IR780 + PFH)). The results suggest the appreciable cyto-compatibility of the Lip(IR780 + PFH) nanofilm.

3.5.2 Adhesion and spreading assay. As the initial celladhesion and spreading behavior of cells at the interface have

Fig. 4 In vitro assay: MTT assay of MC3T3-E1 cell cultured on five different samples without (a) and with (b) light irradiation (808 nm, 1 W cm−2,15 min) (mean ± SD, n = 3, *P < 0.05, **P < 0.01, ***P < 0.001 (t-test)). (c) Fluorescent images of cells (scale bar = 50 μm) after seeding on differentsamples for 24 h. (d) SEM images of MC3T3-E1 cultured on different samples for 24 h.

Biomaterials Science Paper

This journal is © The Royal Society of Chemistry 2018 Biomater. Sci., 2018, 6, 2460–2471 | 2467

a great influence on proliferation and differentiation,63 theeffects of the liposome modified Ti disks on cell attachmentwere investigated by visualizing the filamentous F-actin andcell nuclei under a fluorescence microscope. The cells on thesamples display superior adhesion with a spreading mor-phology after incubation for 24 h (Fig. 4c). The number of cellsseeded on Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti detected by

visual inspection is slightly less than MP-Ti due to the effectsof photosensitizer. Compared to MP-Ti, a similar expression ofF-actin filaments can be observed from AHT and Lip-Ti butthat on Lip(IR780)-Ti and Lip(IR780 + PFH)-Ti is a little lower.In addition, similar prominent filopodia and unidirectionallamellipodia extensions are apparent on the functionalizedliposome modified Ti and an elongated morphology is

Fig. 5 In vivo antibacterial and histological assay: (a) representative images of the plates for the colonies of S. aureus taken from the subcutaneousinfection mode. (b) Antibacterial ratio histogram of S. aureus on different samples determined by the number of colonies in the plate samples(mean ± SD, n = 3, **P < 0.01, ***P < 0.001 (t-test)). Representative histological images of tissues adhered to different samples after Giemsa (c) andH&E staining (d).

Paper Biomaterials Science

2468 | Biomater. Sci., 2018, 6, 2460–2471 This journal is © The Royal Society of Chemistry 2018

observed in all the samples. This result is consistent withFig. 4a indicating that the liposome modified sample favorsinitial adhesion, spreading, and growth of cells with noobvious cytotoxicity.

3.5.3 Cell and spreading assay. To study the effects of thenanofilm on cells, the morphology and pseudopodia are exam-ined using SEM (Fig. 4d). After incubation for 24 h, the cellsspread well in the five groups. However, polygonal cells appearon Lip(IR780 + PFH)-Ti and a larger spindle morphology isvisible on MP-Ti, AHT, Lip-Ti, and Lip(IR780)-Ti. As for MP-Ti,AHT, Lip-Ti and Lip(IR780)-Ti, clustered cells cover the surfacewith connections through threadlike filopodia and the cellconnections are conducive to proliferation.64 As shown in thehigh-magnification SEM images, filamentous pseudopodiawith different lengths are visible on the five groups, implyingthat liposome modification favors cell proliferation and differ-entiation consistent with fluorescence imaging.

3.6 In vivo experiments

To evaluate the in vivo histological response and anti-infectionperformance, the liposome modified Ti samples are implantedsubcutaneously into rats. The infection status of wounds canbe determined by quantifying the viable bacteria extractedfrom tissue fluids (Fig. 5a). The trend of the in vivo antibacter-ial results is similar to that of the in vitro assay. The number ofbacterial colonies is similar for the three different samples inthe absence of NIR. When MP-Ti is irradiated with NIR light,the number of the colonies is comparable to that of thecontrol group (MP-Ti in darkness). However, less viable bac-terial colonies are visible on Lip(IR780)-Ti after NIR irradiationand not all the bacteria can be killed under the conditions,indicating the inability of IR780 itself to destroy bacteria underhypoxia conditions induced by bacterial infection and inflam-mation. As shown in Fig. 5b, the antibacterial efficacy of Lip(IR780 + PFH)-Ti under the NIR irradiation is 99.61% againstS. aureus, while that of Lip(IR780)-Ti under NIR irradiation is68.75%. The significant improvement indicates that when PFHis introduced to the liposome film, bacterial colony formationis almost completely inhibited, suggesting that the incorpor-ation of PFH improves the PDAT efficiency by producing moreROS from IR780 owing to the oxygen enriching behavior ofPFH.

Giemsa staining is employed to identify the bacterial resi-dues65,66 (Fig. 5c). There are lots of bacteria (red arrow in theselected area) in the subcutaneous tissues for the differentsamples without NIR illumination as well as MP-Ti + NIR. Thenumber of bacteria in the subcutaneous tissues diminishobviously in Lip(IR780) + Ti under NIR irradiation and thereare no detectable signs of bacteria on Lip(IR780 + PFH) + Ti inthe presence of NIR.

The in vivo tissue response of PDAT is evaluated with hema-toxylin/eosin staining (Fig. 5d). As for the untreated group(MP-Ti), a large number of inflammatory cell infiltrations(mainly neutrophils) are visible in the section images implyingserious acute inflammation following implantation (blackarrow). In addition, vast inflammatory cell infiltration (mainly

neutrophils) is observed from Lip(IR780)-Ti and Lip(IR780 +PFH)-Ti without light irradiation (black arrow) similar to thehost reaction in the control group (MP-Ti). As cells gather onthe infected tissues, neutrophils in the circulating bloodmigrate rapidly to the infected sites to deal with the infectionstimuli and destroy invading bacteria.67,68 As shown inFig. S11 (ESI†), the inflammatory cell ratios of group MP-Tiwith or without NIR are 42.56% and 47.06%, respectively, indi-cating a serious infection, compared with group Lip(IR780) +dark (50.96%) and Lip(IR780 + PFH) + dark (46.00%); after NIRirradiation, Lip(IR780) shows 36.97% of the inflammatory cellratio, but the value for Lip(IR780 + PFH) is much lower(10.00%), indicating improved bacterial infection. We can con-clude that the bacterial infection can be reduced by ROS pro-duced by IR780 under irradiation of NIR but the efficiency isnot high enough to completely eliminate the bacteria in thehypoxia environment, when PFH is introduced. At the veryleast, neutrophils are visible and the majority of the tissues arenormal, suggesting that Lip(IR780 + PFH)-Ti after irradiationpossesses superior PDAT properties without inducing adverseforeign-body reactions. That is, the host tissue responseinduced by the bacterial infection can be totally eliminated byLip(IR780 + PFH)-Ti + NIR under hypoxia conditions.

4. Conclusions

A PFH@liposome modified surface system with enhancedPDAT efficacy is designed and constructed on Ti implants. Thegeneration efficiency of 1O2 under NIR light irradiation(808 nm, 1 W cm−2, 15 min) is obviously enhanced for thisenhanced PDAT system (Lip(IR780 + PFH)-Ti). Bacteria cannotbe totally eliminated by the normal PDAT system (Lip(IR780)-Ti), since the ROS generated by a photosensitizer cannot killbacteria thoroughly under the conditions with lower oxygencontent. In contrast, this enhanced PDAT system (Lip(IR780 +PFH)-Ti) can almost completely inhibit bacterial growth asdemonstrated by the in vitro and in vivo studies through theenhanced ROS yields of PFH. As seen from the results in therat subcutaneous infection model, this surface biofunctionali-zation strategy provides valuable insights into in vivo anti-inflammation and can help remove postoperative infection ofimplants under the conditions with lower oxygen content inclinical implantation.

Conflicts of interest

The authors report no conflicts of interest in this work.

Acknowledgements

This work is jointly supported by the Natural Science Fund ofHubei Province (2018CFA064), the National Key Research andDevelopment Program of China, no. 2016YFC1100600 (subpro-ject 2016YFC1100604), the National Natural Science

Biomaterials Science Paper

This journal is © The Royal Society of Chemistry 2018 Biomater. Sci., 2018, 6, 2460–2471 | 2469

Foundation of China, 51671081, and 51422102, the HongKong Research Grants Council (RGC) General Research Funds(GRF) No. CityU 11301215, and 11205617, the Natural ScienceFund of Hubei Province, 2018CFA064, the Hong Kong ITC(ITS/287/17, GHX/002/14SZ), the Health and Medical ResearchFund (03142446), Hong Kong RGC GRF (17214516) and RGC/NSFC (N_HKU725-16).

References

1 Z. Jia, P. Xiu, M. Li, X. Xu, Y. Shi, Y. Cheng, S. Wei,Y. Zheng, T. Xi and H. Cai, Biomaterials, 2016, 75, 203–222.

2 G. Pan, S. Sun, W. Zhang, R. Zhao, W. Cui, F. He, L. Huang,S. H. Lee, K. J. Shea, Q. Shi and H. Yang, J. Am. Chem. Soc.,2016, 138, 15078–15086.

3 K. G. Neoh, X. Hu, D. Zheng and E. T. Kang, Biomaterials,2012, 33, 2813–2822.

4 X. Y. Ma, Y. F. Feng, T. S. Wang, W. Lei, X. Li, D. P. Zhou,X. X. Wen, H. L. Yu, L. B. Xiang and L. Wang, Biomater.Sci., 2017, 6, 225–238.

5 L. Zhang, C. Ning, T. Zhou, X. Liu, K. W. Yeung, T. Zhang,Z. Xu, X. Wang, S. Wu and P. K. Chu, ACS Appl. Mater.Interfaces, 2014, 6, 17323–17345.

6 X. Khoo, G. A. O’Toole, S. A. Nair, B. D. Snyder, D. J. Kenanand M. W. Grinstaff, Biomaterials, 2010, 31, 9285–9292.

7 P. Zhu, Z. Weng, X. Li, X. Liu, S. Wu, K. W. K. Yeung,X. Wang, Z. Cui, X. Yang and P. K. Chu, Adv. Mater.Interfaces, 2016, 3, 1500494.

8 S. Wu, Z. Weng, X. Liu, K. W. K. Yeung and P. K. Chu, Adv.Funct. Mater., 2015, 24, 5464–5481.

9 S. Mei, H. Wang, W. Wang, L. Tong, H. Pan, C. Ruan,Q. Ma, M. Liu, H. Yang and L. Zhang, Biomaterials, 2014,35, 4255–4265.

10 G. Cado, R. Aslam, L. Séon, T. Garnier, R. Fabre, A. Parat,A. Chassepot, J. C. Voegel, B. Senger and F. Schneider, Adv.Funct. Mater., 2013, 23, 4801–4809.

11 X. Z. Xiang, W. Y. Gong, M. S. Kuang and L. Wang, RareMet., 2016, 35, 289–298.

12 M. M. Konai and J. Haldar, ACS Infect. Dis., 2015, 1, 469–478.

13 S. P. Mukherjee, M. Davoren and H. J. Byrne, Toxicol. InVitro, 2010, 24, 169–177.

14 N. R. Panyala, E. M. Penamendez and J. Havel, J. Appl.Biomed., 2008, 6, 117–129.

15 B. Wang, M. Wang, A. Mikhailovsky, S. Wang andG. C. Bazan, Angew. Chem., Int. Ed., 2017, 56, 5031–5034.

16 F. Xu, M. Hu, C. Liu and S. K. Choi, Biomater. Sci., 2017, 5,678–685.

17 J. Gehring, B. Trepka, N. Klinkenberg, H. Bronner,D. Schleheck and S. Polarz, J. Am. Chem. Soc., 2016, 18,395–400.

18 A. Reinhard, W. J. Sandborn, H. Melhem, L. Bolotine,M. Chamaillard and L. Peyrin-Biroulet, Expert Rev. Clin.Immunol., 2015, 11, 637–657.

19 S. S. Lucky, K. C. Soo and Y. Zhang, Chem. Rev., 2015, 115,1990–2042.

20 W. Fan, P. Huang and X. Chen, Chem. Soc. Rev., 2016, 45,6488–6519.

21 K. Y. Choi, G. Liu, S. Lee and X. Chen, Nanoscale, 2012, 4,330–342.

22 S. Perni, P. Prokopovich, J. Pratten, I. P. Parkin andM. Wilson, Photochem. Photobiol., 2011, 10, 712–720.

23 Z. Zhen, W. Tang, C. Guo, H. Chen, X. Lin, G. Liu, B. Fei,X. Chen, B. Xu and J. Xie, ACS Nano, 2013, 7, 6988–6996.

24 H. Park, J. Lee, S. Jeong, S. Jeong, B. N. Im, M. K. Kim,S. G. Yang and K. Na, Adv. Healthcare Mater., 2016, 5, 3139–3147.

25 H. Ren, J. Liu, Y. Li, H. Wang, S. Ge, A. Yuan, Y. Hu andJ. Wu, Acta Biomater., 2017, 59, 269–282.

26 K. A. Carter, S. Shao, M. I. Hoopes, D. Luo, B. Ahsan,V. M. Grigoryants, W. Song, H. Huang, G. Zhang andR. K. Pandey, Nat. Commun., 2014, 5, 3546.

27 C. R. Di, G. Bealle, J. Kolosnjajtabi, A. Espinosa, A. S. Ak,C. Menager and C. Wilhelm, ACS Nano, 2015, 9, 2904–2916.

28 J. Liu, G. Yang, W. Zhu, Z. Dong, Y. Yang, Y. Chao andZ. Liu, Biomaterials, 2017, 146, 40–48.

29 R. Hao, J. Liu, F. Su, S. Ge, A. Yuan, W. Dai, J. Wu andY. Hu, ACS Appl. Mater. Interfaces, 2017, 9, 3463–3473.

30 K. K. Ng and Z. Gang, Chem. Rev., 2015, 115, 11012–11042.31 W. D. Jang, N. Nishiyama, G. D. Zhang, A. Harada,

D. L. Jiang, S. Kawauchi, Y. Morimoto, M. Kikuchi,H. Koyama and T. Aida, Angew. Chem., Int. Ed., 2005, 44,419–423.

32 N. Nishiyama, Y. Morimoto, W. D. Jang and K. Kataoka,Adv. Drug Delivery Rev., 2009, 61, 327–338.

33 K. Lu, C. He and W. Lin, J. Am. Chem. Soc., 2015, 137,7600–7603.

34 J. E. Squires, Science, 2002, 295, 1002–1005.35 J. Fuchs and J. Thiele, Free Radicals Biol. Med., 1998, 24,

835–847.36 C. I. Castro and J. C. Briceno, Artif. Organs, 2010, 34, 622–

634.37 K. Schaffer and C. T. Taylor, FEBS J., 2015, 282, 2260–2266.38 J. Xue, M. He, Y. Liang, A. Crawford, P. Coates, D. Chen,

R. Shi and L. Zhang, J. Mater. Chem. B, 2014, 2, 6867–6877.39 X. Xie, C. Mao, X. Liu, Y. Zhang, Z. Cui, X. Yang,

K. W. K. Yeung, H. Pan, P. K. Chu and S. Wu, ACS Appl.Mater. Interfaces, 2017, 9, 26417–26428.

40 Y. Cheng, H. Cheng, C. Jiang, X. Qiu, K. Wang, W. Huan,A. Yuan, J. Wu and Y. Hu, Nat. Commun., 2015, 6, 8785.

41 G. Song, C. Ji, C. Liang, X. Song, X. Yi, Z. Dong, K. Yangand Z. Liu, Biomaterials, 2017, 112, 257–263.

42 C. Liu, H. Dong, N. Wu, Y. Cao and X. Zhang, ACS Appl.Mater. Interfaces, 2018, 10, 6991–7002.

43 L. Bourré, S. Thibaut, A. Briffaud, N. Rousset, S. Eléouet,Y. Lajat and T. Patrice, J. Photochem. Photobiol., B, 2002, 67,23–31.

44 Y. Zhang, X. Liu, Z. Li, S. Zhu, X. Yuan, Z. Cui, X. Yang,P. K. Chu and S. Wu, ACS Appl. Mater. Interfaces, 2018, 10,1266–1277.

Paper Biomaterials Science

2470 | Biomater. Sci., 2018, 6, 2460–2471 This journal is © The Royal Society of Chemistry 2018

45 S. H. Hsu, H. J. Tseng and Y. C. Lin, Biomaterials, 2010, 31,6796–6808.

46 C. M. Tilmaciu, M. Mathieu, J. P. Lavigne, K. Toupet,G. Guerrero, A. Ponche, J. Amalric, D. Noel andP. H. Mutin, Acta Biomater., 2015, 15, 266–277.

47 H. Ren, J. Liu, Y. Li, H. Wang, S. Ge, A. Yuan, Y. Hu andJ. Wu, Acta Biomater., 2017, 59, 269–282.

48 W. Zhang, S. Shi, Y. Wang, S. Yu, W. Zhu, X. Zhang,D. Zhang, B. Yang, X. Wang and J. Wang, Nanoscale, 2016,22, 11642–11648.

49 R. Chen, M. D. Willcox, K. K. Ho, D. Smyth and N. Kumar,Biomaterials, 2016, 85, 142–151.

50 L. Feng, M. Gao, D. Tao, Q. Chen, H. Wang, Z. Dong,M. Chen and Z. Liu, Adv. Funct. Mater., 2016, 26, 2207–2217.

51 L. Feng, C. Liang, Z. Dong, D. Tao, T. E. Barnhart, W. Cai,M. Chen and L. Zhuang, ACS Nano, 2017, 11, 927–937.

52 C. Jiang, H. Cheng, A. Yuan, X. Tang, J. Wu and Y. Hu, ActaBiomater., 2015, 14, 61–69.

53 F. Wang and J. Liu, Small, 2014, 10, 3927–3931.54 F. Wang and J. Liu, J. Am. Chem. Soc., 2015, 137, 11736–

11742.55 X. B. Chen, Y. C. Li, P. D. Hodgson and C. Wen,

Acta Biomater., 2009, 5, 2290–2302.56 T. Nakamura, H. Akita, Y. Yamada, H. Hatakeyama and

H. Harashima, Acc. Chem. Res., 2012, 45, 1113–1121.

57 V. A. Yavorskii, N. K. Pogorelaya, N. A. Bogdanova andE. A. Lukyanetz, Neurophysiology, 2011, 43(3), 201–204.

58 Y. Yuan, J. Liu and B. Liu, Angew. Chem., Int. Ed., 2014, 53,7163–7168.

59 A. Pace, P. Pierro, S. Buscemi, N. Vivona and E. L. Clennan,J. Org. Chem., 2007, 72, 2644–26466.

60 A. N. Macpherson, T. G. Truscott and P. H. Turner, J. Chem.Soc., Faraday Trans., 1994, 90, 1065–1072.

61 J. Kim and J. L. Gilbert, Acta Biomater., 2016, 30, 368–377.62 Y. Tu, M. Lv, P. Xiu, T. Huynh, M. Zhang, M. Castelli,

Z. Liu, Q. Huang, C. Fan and H. Fang, Nat. Nanotechnol.,2013, 8, 594–601.

63 G. Jin, H. Qin, H. Cao, S. Qian, Y. Zhao, X. Peng, X. Zhang,X. Liu and P. K. Chu, Biomaterials, 2014, 35, 7699–7713.

64 P. Linda, H. Sonja, R. Marcus, S. Gabor, A. Volker,S. Reinhard and K. S. Lips, Acta Biomater., 2014, 10, 439–449.

65 C. Mao, Y. Xiang, X. Liu, Z. Cui, X. Yang, Z. Li, S. Zhu,Y. Zheng, K. W. Yeung and S. Wu, ACS Nano, 2018, 12,1747–1759.

66 J. Li, L. Tan, X. Liu, Z. Cui, X. Yang, K. W. Yeung, P. K. Chuand S. Wu, ACS Nano, 2017, 11, 11250–11263.

67 V. Brinkmann, U. Reichard, C. Goosmann, B. Fauler,Y. Uhlemann, D. S. Weiss, Y. Weinrauch and A. Zychlinsky,Science, 2004, 303, 1532–1535.

68 C. Nathan, Nat. Rev. Immunol., 2006, 6, 173–182.

Biomaterials Science Paper

This journal is © The Royal Society of Chemistry 2018 Biomater. Sci., 2018, 6, 2460–2471 | 2471

S-1

Supporting Information

Construction of Perfluorohexane/IR780@Liposome Coating on Ti for Rapid

Bacteria Killing under Permeable Near Infrared Light

Xiuhua Wang a, Lei Tan a, Xiangmei Liu *a, Zhenduo Cui b, Xianjin Yang b, Kelvin W. K. Yeung c, Paul K. Chu d, Shuilin Wu *ab

a Hubei Collaborative Innovation Center for Advanced Organic Chemical Materials, Ministry-of-Education Key Laboratory for the Green Preparation and Application of Functional Materials, Hubei Key Laboratory of Polymer Materials, School of Materials Science & Engineering, Hubei University, Wuhan 430062, China.b School of Materials Science & Engineering, Tianjin University, Tianjin 300072, China. E-mail: shuilin.wu@gmail.com; shuilinwu@tju.edu.cn (S.L. Wu); liuxiangmei1978@163.comc Department of Orthopaedics & Traumatology, Li KaShing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong 999077, China.d Department of Physics and Department of Materials Science and Engineering, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong 999077, China.

Electronic Supplementary Material (ESI) for Biomaterials Science.This journal is © The Royal Society of Chemistry 2018

mailto:shuilin.wu@gmail.commailto:sxwsl1976@163.commailto:liuxiangmei1978@163.com

S-2

Figure S1. Structure and design of the oxygen self-enriched PDAT system. The photosensitizer and perfluorohexen are co-encapsulated by liposome and dispersed uniformly inside the liposome.

S-3

Figure S2. GC-MS spectra of PFH and Lip(IR780+PFH).

S-4

Figure S3. The zeta potential of Lip, Lip(IR780) and Lip(IR780+PFH) (mean ± SD, n = 3, ***P < 0.001).

S-5

Figure S4. The photostability of different liposome. The UV-Vis absorption spectrum of IR780, Lip(IR780) and Lip(IR780+PFH) measured right after the dilution process or 1 day of storage (under light or darkness).

S-6

Figure S5. Postulated reaction mechanism for chemisorption of liposome by AHT.

S-7

Figure S6. The morphologies of Ti, AHT, Lip-Ti, Lip(IR780)-Ti, Lip(IR780+PFH)-Ti observed by AFM.

S-8

Figure S7. EDS spectra of Lip-Ti, Lip(IR780)-Ti, and Lip(IR780+PFH)-Ti.

S-9

Figure S8. XPS spectra of different liposome modified AHT.

S-10

Figure S9. The corresponding oxygen concentration of Lip, Lip(IR780), Lip(IR780+PFH) and deoxygenated water when the oxygen concentration reached the equilibrium of (a) (n=3, *P < 0.05, **P < 0.01.); Oxygen concentration changes in real time of different samples on Ti plates in deoxygenated water (b).

S-11

Figurre S10. Representative images of the plate samples showing colonies of S. aureus and E. coli incubated with Lip(PFH) under the irradiation of NIR for 15 min or in the dark compared with the control group.

S-12

Figurre S11. Representative images of the plate samples showing colonies of E. coli and S. aureus incubated with different samples under irradiation of NIR for 15 min or in darkness;

S-13

Figure S12. Inflammatory cell ratios calculated from bone tissue H&E staining data. Inflammatory cell ratios = [(total area of lymphocyte, monocytes and neutrophil)/area of tissue] × 100%, n=3 animals per group, *P < 0.05, **P < 0.01.

S-14

Table S1. Surface roughness of different samples which is analyzed by AFM software.

MP-Ti AHT Lip-Ti Lip (IR780)-Ti Lip (IR780+PFH)-Ti

Root mean square (nm) 20.817 91.212 10.134 5.667 7.646

Roughness average (nm) 14.307 71.405 8.139 4.518 5.913

Button 1: