Jaworskietal.NanoscaleResearchLetters (2018) 13:116 https://doi.org/10.1186/s11671-018-2533-2 NANO EXPRESS Open Access Graphene Oxide-Based Nanocomposites Decorated with Silver Nanoparticles as an Antibacterial Agent Sławomir Jaworski1, Mateusz Wierzbicki1, Ewa Sawosz1, Anna Jung2, Grzegorz Gielerak2, Joanna Biernat3,4, Henryk Jaremek3, Witold Łojkowski5, Bartosz Woźniak5, Jacek Wojnarowicz5, Leszek Stobiński6, Artur Małolepszy6, Marta Mazurkiewicz-Pawlicka6, Maciej Łojkowski7, Natalia Kurantowicz1 and André Chwalibog8* Abstract One ofthe most promising methods against drug-resistant bacteria can be surface-modified materials withbiocidal nanoparticles and nanocomposites. Herein, we present a nanocompositewith silver nanoparticles(Ag-NPs) onthe surfaceofgrapheneoxide(GO)asanovelmultifunctionalantibacterialandantifungalmaterial.Ultrasonictechnologies havebeenusedasaneffectivemethodofcoatingpolyurethanefoils.Toxicityongram-negativebacteria(Escherichia coli),gram-positivebacteria(StaphylococcusaureusandStaphylococcusepidermidis),andpathogenicyeast(Candida albicans)wasevaluatedbyanalysisofcellmorphology,assessmentofcellviabilityusingthePrestoBlueassay,analysis ofcellmembraneintegrityusingthelactatedehydrogenaseassay,andreactiveoxygenspeciesproduction.Compared toAg-NPsandGO,whichhavebeenwidelyusedasantibacterialagents,ournanocompositeshowsmuchhigher antimicrobialefficiencytowardbacteriaandyeastcells. Keywords:Grapheneoxide,Silvernanoparticles,Antimicrobialproperties Background relatedbacteria,whicharethenabletopassontheresist- The development of antibiotics has played a significant ance genes to subsequent generations. Thus, the emer- role in controlling the number of bacterial infections. gence of antibiotic-resistant bacteria represents a serious However, the improper use and the overuse of antibiotics problem that could be overcome by the development of have led to the development of multidrug resistance in novel antimicrobial agents. Antibacterial agents are very many bacterial species. Some strains have become importantinthetextileindustry,waterdisinfection,medi- resistant to practically all of the commonly available cine, and food packaging. Nanoparticles and nanomater- agents: beta-lactams, tetracyclines, and aminoglycosides ials can be used as an alternative to antibiotics [4]. The [1]. The major resistant pathogens are methicillin- mechanismofantibacterial activity of nanoparticlesvaries resistant Staphylococcus aureus, vancomycin-resistant among the different types of nanoparticle. While some Enterococcus, and extended-spectrum β-lactamase- proposed mechanisms relate to the physiochemical struc- producing Klebsiella pneumoniae and Escherichia coli [2, tureofthenanoparticles,othersrelatetotheincreasedre- 3].Bacteria,withtheirverylargepopulationsandfastpro- lease of antibacterial ions from nanoparticle surfaces. liferation time, are able to rapidly develop mechanisms of Multiple simultaneous mechanisms of action against mi- antibioticresistancewhenasubsetofthebacteriapopula- crobes would require a variety of synchronous DNA mu- tion survives antibiotic treatment. Moreover, antibiotic- tations in the same microbial cell for the development of resistant bacteria are able to transfer copies of DNA that resistance; therefore, it is difficult for bacterial cells to be- code for a mechanism of resistance to other distantly come resistant to nanoparticles and nanomaterials. Anti- microbial nanomaterials, such as silver, copper, fullerenes, *Correspondence:[email protected] andsingle-walledcarbonnanotubes,mayofferseveralad- 8DepartmentofVeterinaryandAnimalSciences,UniversityofCopenhagen, vantages due to their unique physicochemical properties Groennegaardsvej3,1870Frederiksberg,Denmark and high surface areas [5–8]. The exact mechanisms of Fulllistofauthorinformationisavailableattheendofthearticle ©TheAuthor(s).2018OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.0 InternationalLicense(http://creativecommons.org/licenses/by/4.0/),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinkto theCreativeCommonslicense,andindicateifchangesweremade. Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page2of17 nanoparticle (NP) toxicity against various bacteria are not GO sheets via covalent and non-covalent interactions. completelyunderstood.Accordingtothecurrentresearch, The strong antibacterial activity of GO has been the major processes underlyingthe antibacterialeffects of reported. The antibacterial activity of GO has been NPs are disruption of the bacterial cell membrane, metal assigned to membrane stress induced by sharp edges of ionrelease,generationofROS,penetrationofthebacterial graphene oxide nanosheets, which may result in physical cellmembrane,andinductionofintracellularantibacterial damage to cell membranes, leading to the loss of effects, including interactions with DNA and proteins [9, bacterial membrane integrity [21]. Recently, graphene- 10].NPsareabletoattachtothemembraneofbacteriaby functionalized antimicrobial nanoparticles have been electrostatic interaction and disrupt the integrity of the used as promising antibacterial materials [22, 23]. bacterial membrane. The positive charge of the surface of Nanocomposites can overcome the limitations of the in- the NPs is essential for the adhesion. The positive charge dividual components. For example, antibacterial nano- enables electrostatic addition between NPs and negatively materials attached to the graphene substrate are more charged cell membrane of the microorganisms [11]. The stable and well dispersed [24]. These nanocomposites electrostatic connection between NPs with the sulfur- couldcontainmetals,metaloxides,andpolymers. containing proteins present on the surface of bacterial One of the most promising methods against drug- cells causes irreversible changes in cell wall structure resistant bacteria can be surface-modified materials with resultingin damages ofcellwalland membrane[12].The biocidal nanoparticles. Ultrasonic technologies have bacterial membrane is crucial, irrespective of the meta- been confirmed as an effective method of coating bolicstatusofthecell,asitprovidesselectivepermeability various materials with antibacterial and fungicidal sub- for cellular homeostasis and metabolic energy transduc- stances [25–28]. Many researchers classify the ultra- tion. The second antibacterial and antifungal activity of sound method as a “green technology” [29, 30]. The NPsisduetotheirabilitytoproduceROSandfreeradical method is based on the use of cavitation phenomena, species [13]. Increased level of ROS induced hyperoxida- which is the formation, growth, and collapse of cavita- tionoflipids,proteins,andDNA[14]. tion bubbles in the liquid medium [31, 32]. Imploding Moreover, the structures of many types of NPs are bubbles generate immense amounts of energy in micro- suitable for carrying antimicrobial agents [15, 16]. regions up to 5000 K and pressure up to 2000 atm Carriers can help to protect the drugs from resistance by within a short period of time [33, 34]. Consequently, target bacteria. A nanoparticle-based drug delivery sys- shock waves and so-called microjets directed toward the tem can help to target antibiotics to an infection site solid surface are generated [35]. Located in a liquid and thereby minimize systemic side effects. Other ad- medium, NPs are driven up by the implosion effect and vantages include improved solubility of hydrophobic jetstreamsathighspeed(>100m/s)onthesolidsurface drugs, prolonged systemic circulation time and drug and form a layer [36]. Acoustic cavitation can also lead half-life,andsustaineddrugrelease[4]. to change in the physical properties of sonicated objects, Recently, it has been demonstrated that graphene, a e.g.,resizingofGOflakes[37,38]. new allotrope of carbon, has antibacterial activity. We achieved promising results in our previous studies Graphene is a material made of carbon atoms that are with Salmonella enterica and Listeria monocytogenes bonded together in a repeating pattern of hexagons. A treated with pristine graphene, GO, and reduced GO unique feature ofgraphene flakes istheratio ofitsthick- [20]. Of the different types of graphene, GO was also ness to the surface. The surface of graphene is covered found to have the highest antibacterial activity at a low with an electron cloud, which probably predisposes this concentration. Bacterial cells were distributed over the material to be an electron donor and gives it the ability entire surface of the GO. In this study, we hypothesized to make special bonds. The edges of graphene have that GO decorated with silver nanoparticles (GO-Ag) other bonds (characteristic for diamond sp3 type bonds), will havestronger toxic influence on microbial cells than and these places may have different physicochemical bare GO or silver nanoparticles (Ag-NPs). Because it has characteristics [17]. These characteristics suggest that two active sides (surface and edges), GO oxide can graphene can be exposed to plastic adhesion to different attach Ag-NPs to the edges and adhere to the cell sur- intercellular structures, including bacterial cells [18–20]. face. The antibacterial activity of graphene-based nano- In addition, because it has two active sides (surface and composites may be due to the disruption of the cell edges), graphene can attach biological molecules to its membrane and oxidative stress. The objective of this edges and adhere to the cell surface. An oxidized form study was to evaluate the antimicrobial activity of GO- of graphene, graphene oxide (GO), is easily dispersible based nanocomposites decorated with Ag-NPs in com- in water and other organic solvents due to the presence parison to bare GO and Ag-NPs using gram-negative ofthe oxygen functionalities. Theoxygenated groups en- bacteria (Escherichia coli), gram-positive bacteria able the straightforward chemical functionalization of (Staphylococcus aureus and Staphylococcus epidermidis), Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page3of17 and pathogenic yeast (Candida albicans) using an in lefttodry.After itwasdried,thespectrum wascollected vitro model. The investigation consisted of structural intherange400–4000cm−1. analysis of nanocomposites using X-ray diffraction, Average particle size and zeta potential measurements Raman spectroscopy transmission, FT-IR, electron mi- werecarriedoutusingZetasizerNano-ZSZEN3600pro- croscopy (TEM), scanning electron microscopy (SEM) duced by Malvern Instruments Ltd. (Malvern, UK) using and atomic force microscopy (AFM), evaluation of mi- the dynamic light scattering (DLS) mode and laser Dop- crobial cell morphology, assessment of cell viability by pler electrophoresis, respectively, at room temperature PrestoBlue™ assay, investigation of cell membrane integ- (23°C). rity by lactate dehydrogenase assay (LDH), and assess- mentofreactive oxygenspecies(ROS)production. TEM/SEM/AFMAnalysisofNanomaterials The morphology of powders and foils was determined Methods using the transmission electron microscope (TEM; Synthesis,Modification,andCharacterizationofGraphene JEM-1220 JEOL, Tokyo, Japan, accelerating voltage of Oxide 80 kV) and scanning electron microscope (SEM; Zeiss, In this study, a commercially available graphite powder Ultra Plus, Oberkochen, Germany). Samples for TEM (Acros Organics, New Jersey, USA) was oxidized by the observations were prepared by placing droplets of hydro- modified Hummers method [39]. Ten grams of graphite colloidsontoTEMgrids(Formvaron3mm200MeshCu powder were mixed with 230 mL of concentrated sul- Grids, Agar Scientific, Stansted, UK). Immediately after furic acid (98%) below 10 °C. Then, 4.7 g of sodium ni- air-drying the droplets, the grids were inserted into the trate and 30 g of potassium permanganate were added microscope. gradually to the sulfuric acid and graphite mixture while For SEM analysis, samples were coated with a thin maintaining the temperature below 10 °C. Then, the carbon layer using the sputter coater (SCD 005/CEA mixture was heated to 30 °C and stirred for 2 h. In the 035, BAL-TEC, Pfäffikon, Switzerland). An internal la- next step, 100 mL of water was added, and the mixture boratory measurement procedure was applied (P5.10, temperature reached ~100 °C. Finally, the mixture was edition 6of26.08.2015). treated with 10 mL of hydrogen peroxide. For purifica- AFM (atomic force microscopy) imaging was carried tion, the slurry was filtrated and washed with deionized out using Asylum Research MFP3D Bio software water untilthepHofthefiltrate reached6.5. (version: Asylum Research MFP3D 15.106.09). Surface X-ray diffraction patterns of GO were gathered at topography imaging and detection of GO on the tested roomtemperaturewithintherangeof2thetaanglefrom foil surfaces were carried out using two imaging modes, 10° to 100° with the step of 0.02° using the X-ray powder AC mode for phase contrast imaging and lateral force diffractometer (CuKα1) (X’Pert PRO, PANalytical, microscopy (LFM) for GO detection since GO reduces Almelo,Netherlands). friction forces[40]. The analysis of carbon, hydrogen, nitrogen, and sulfur content by weight in GO was carried out using theVario ELIIIapparatusproducedbyElementarAnalysensysteme PreparationofPolyurethaneFoilsCoatedwithGOand GmbH (Langenselbold, Germany). Prior to performing Ag-NPs measurements of chemical analyses of GO, the samples For covering polyurethane foils, suspensions of Ag-NPs were subject to 24-h desorption in a desorption station (HydroSilver1000, Amepox, Łódź, Poland) and GO were (VcPrep 061, Micromeritics, Norcross, GA, USA) under used. Suspensions of GO, Ag-NPs, and GO-Ag (GO vacuum (0.05 mbar) at 50 °C. Oxygen content was calcu- (200 μg/mL), Ag-NPs (100 μg/mL), GO (200 μg/mL)+ lated by subtracting the determined contents of carbon, Ag-NPs (100 μg/mL)) were prepared in deionized water hydrogen,nitrogen,andsulfurfrom100%weight. (conductance 0.09 μS/cm, deionizer: HLP 20UV, Hydro- Raman spectroscopy was performed using an inVia lab, Staszyn, Poland). The suspensions were used with- Ramanmicroscope (Renishaw, UK). Graphene oxide was outadditionalpurification andfiltration. analyzed with the 514-nm laser wavelength with the 5% Ultrasonic coating of polyurethane foils (15×15×0. of its initial power. The spectra were collected from five 05 mm) took place in a glass flask with a volume of different spots on the sample. The exposure time was 50 ml. Foil samples were fastened on a stand (Teflon) 10sandtwoscanswere collected. andsubsequently immersedinthepreparedsuspensions. FT-IR measurements were performed using Nicolet The coating process was performed using an ultrasonic iS10 spectrometer(ThermoFisher Scientific,USA)inat- horn (Ti horn, Ø13 mm, 60% efficiency, 20 kHz, Sonics tenuated total reflectance mode on a diamond crystal. & Materials, Inc., Newtown, CT, USA) placed square to Five microliters of graphene oxide water suspension was the foil samples present in the suspension. The process dripped on the surface of the diamond crystal and it was temperature was 30±1 °C. The covered samples were Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page4of17 flushed in deionized water and dried in a laminar cham- without nanoparticles were used as control group. The ber andsubsequentlypacked insterilepackages. bacteria and yeast growth under the foils was measured after24hofincubationat37°C. SurfaceFreeEnergy Wettability tests were carried out using the Data Physics ViabilityAssay OCA– 20goniometer(DataPhysics InstrumentsGmbH, Cell viability was evaluated using the PrestoBlue™ Cell Filderstadt,Germany).Surfacefree energy(SFE) was cal- Viability Assay (Life Technologies, Taastrup, Denmark). culated using the Owens, Wendt, Rabel, and Kaelble PrestoBlue™ reagent is quickly reduced by metabolically (OWRK) method using two test liquids: deionized water active cells, providing a quantitative measure of viability anddiiodomethane [41]. and cytotoxicity. Bacterial and yeast cells were cultured onto foils coated with GO, Ag-NPs, and GO-Ag located BacterialandYeastCultivationandPreparation on inserts inserted into 6-well plates (200 μL MH broth Staphylococcus aureus (ATCC 25923) and Staphylococ- with 5×103 cells per foil) and incubated for 24 h. In the cus epidermidis (ATCC 14990), Escherichia coli (ATCC next step, 90 μL of each sample was transferred to 96- 25922), and Candida albicans (90028) were obtained well plates and 10 μL of PrestoBlue™ reagent was added from LGC Standards (Lomianki, Poland). The strains to each well and incubated for an additional 2 h at 37 ° were stored as spore suspensions in 20% (v/v) glycerol at C. The optical density of each well was recorded at −20 °C. Prior to their use in experiments, the strains 570 nm on an enzyme-linked immunosorbent assay were defrostedandthe glycerolwasremovedbywashing (ELISA) reader (Infinite M200, Tecan, Durham, NC, the bacterial cells with distilled water. The bacteria and USA). Cell viability was expressed as the percentage yeast were then grown on the following nutrient media: (OD −OD )/(OD −OD )×100%, where tryptic soy agarfor S.aureus andE.coli,brain heart agar test blank control blank OD is the optical density of cells exposed to tested for S. epidermidis, and Sabouraud’s agar for C. albicans test foils, OD is the optical density of the control sam- (Merck Millipore, Darmstadt, Germany). The bacteria control ple, and OD is the optical density of wells without and yeast grown on agar plates were harvested by gently blank bacterialandyeastcells. washing the plates with sterile distilled saline solution. To calculate the number of bacteria in the cell suspen- sion, the optical density of the suspensions at 600 nm MembraneIntegrity (OD ) was measured using a spectrophotometer An LDH test (In Vitro Toxicology Assay Kit, lactic de- 600 (Helios Epsilon, Unicam, Milwaukee, WI, USA). Calibra- hydrogenase based, Sigma-Aldrich, Hamburg, Germany) tion curves for each of the microorganisms were pre- was used to evaluate cell membrane integrity. The pared by performing serial tenfold dilutions (up to 10−5) resulting reduced NAD (NADH+) was utilized in the of bacterial and yeast suspensions of known optical stoichiometric conversion of a tetrazolium dye. When density. One milliliter of each dilution was spread on cell-free aliquots of the medium from cultures were petri dishes containing the nutrient medium. After 24 h assayed, the amount of LDH activity could be used as an of incubation at 37 °C, the number of colonies formed indicator of membrane integrity. If the membrane was onthepetridishes wasenumerated.Based ontheresults damaged, intracellular LDH molecules were released of the enumerations (conducted in triplicate), the into the culture medium. Bacterial and yeast cells were density of the original bacterial suspension in colony- cultured on foils (GO, Ag-NPs, and GO-Ag) located on formingunits(CFU)/mLwascalculated. inserts inserted in 6-well plates (200 μL MH broth with 5×103 cells per foil) and incubated for 24 h. Cells AntimicrobialAssay cultured on foil without nanoparticles were used as a The inoculum for the antibacterial assay was prepared control. After this time, the samples were transferred to from actively growing organisms (logarithmic phase). microcentrifuge tubes and centrifuged at 1200 rpm for The inoculums of all microorganisms were prepared 5 min. One hundred microliters of supernatant were from an overnight culture grown aerobically in Mueller– transferred to 96-well plates, and 100 μL of the LDH Hinton (MH) broth at 37 °C. The bacterial and yeast assay mixture was added to each well. The plate was concentration was determined by measuring optical covered and incubated for 30 min at room temperature. density at 600 nm (OD ). Briefly, bacterial and yeast The optical density of each well was recorded at 450 nm 600 suspensions were prepared from overnight cultures and on an ELISA reader (Infinite M200, Tecan, Männedorf, adjusted to 106 CFU/ml. Inoculum was inoculated Switzerland). LDH leakagewas expressedasthe percent- evenly onto the surface of MH agar in petri dishes by age {(OD −OD )−(OD −OD )/(OD test blank control blank control swabbing. Sterile foils coated with GO, Ag-NPs, and −OD )}×100%, where OD is the optical density of blank test GO-Ag were deposited onto the agar surface. Foils cells exposed to tested foils, OD is the optical control Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page5of17 density of the control sample, and OD is the optical ofgraphite,sodium nitrate,andprobablyareducedform blank density ofwellswithoutcells. ofgrapheneoxide. Raman spectroscopy can give information about the structural features of graphene oxide. The D band is at- SEMAnalysisofMicroorganisms tributed to the structural disorder, while the G band Prior to SEM analysis, samples of bacteria and yeast in- comes from the bond stretching of carbon sp2 atoms cubated on foils with GO-Ag and untreated bacteria [42]. The additional bands (including D′, 2D, and D+G) were prepared. Briefly, a drop of bacterial and yeast cul- ture (106 CFU/ml) was incubated on foils with GO-Ag arise from the defects present in the graphitic structure of the carbon material. I /I ratio (calculated from the nanocomposite, or untreated bacteria was deposited on D G intensity of D and G bands) can be used to characterize the surface of a sterile cover glass and incubated for the disorder of the graphitic structure in carbon 24 h at 37 °C inside an empty petri dish. All samples materials. As can be seen in Fig. 2, GO has a highly were dried and covered with gold. Finally, the samples disordered structure due to many functional groups in were imaged with SEM (FEI Quanta 200, Tokyo, Japan) the structure formed during oxidation of graphite atanaccelerationvoltageof15kV. powder. The position of the D band is 1351 cm−1 and theGband1590cm−1;theI /I ratiois1.15. D G ROSProduction The FT-IR spectrum of graphene oxide collected in the ROS production was evaluated using DCFDA, Cellular ATRmoderevealedthatGOhasalotoffunctionalgroups Reactive Oxygen Species Detection Assay Kit (Abcam, presentinthestructure.Themostnotablepeakcanbeob- Cambridge, UK). DCFDA uses the cell permeant reagent served at ~3500 cm−1, which is assigned mainly to water 2′,7′–dichlorofluorescein diacetate, a fluorogenic dye and hydroxyl groups (Fig. 3). Avery intensivepeakaround that measures hydroxyl, peroxyl, and other ROS activ- 1080 cm−1 can also be attributed to hydroxyl groups. The ities within the cell. After diffusion into the cell, DCFDA peak around 1600 cm−1 usually is assigned to C=C bonds is deacetylated by cellular esterases to a non-fluorescent present in graphitic carbon. However, our previous XPS compound, which is later oxidized by ROS into 2′,7′– studies show that there is a few C=C bonds in graphene dichlorofluorescein (DCF). Bacterial and yeast cells were oxide [43]; hence, we attribute this peak to mostly water cultured on foils (GO, Ag-NPs, and GO-Ag) located on still present in the graphene oxide. Other peaks observed inserts inserted in 6-well plates (200 μL MH broth with ontheFT-IRspectrumshowthatGOisrichingroupscon- 5×103 cells per foil) and incubated for 24 h. Cells taining C=O bonds (mainly carboxyl groups), peak around cultured on foil without nanoparticles were used as a 1720 cm−1, epoxy (C–O–C) with the visible peak around control. After this time, the samples were transferred to 1200cm−1,andC–Hbonds(peakaround2800cm−1).The microcentrifuge tubes and centrifuged at 1200 rpm for FT-IR analysis is in good agreement with XPS measure- 5 min. One hundred microliters of supernatant were ments performed for graphene oxide where also hydroxyl, transferred to 96-well plates, and 100 μL of diluted carboxyl, epoxy, and carbonyl groups were identified [44]. DCFDA was added to each well and incubated for an GO and GO after a 10-min ultrasonic homogenization additional 45 min at 37 °C in the dark. DCF production were compared (Figs. 4 and 5), and similarly, Ag-NPs with was measured by fluorescence spectroscopy with an ex- Ag-NPs after a 10-min ultrasonic homogenization (Figs. 4 citation wavelength at 485 nm and an emission wave- and 6) were compared. In order to avoid changes to com- length at 535 nm on an ELISA reader (Infinite M200, pound morphology, all compounds were rapidly cooled Tecan,Durham,NC,USA). down with liquid nitrogen and dried in a lyophilizer. Figure 5a, b presentsGO flakes, whileFig. 5c,d shows the Results effectsof ultrasounds on GO flakes, which undergo partial CharacteristicsofGOandAg-NPs foldingandfragmentation.Figure6alsoshowsasimilaref- The chemical analysis revealed the presence of nitrogen, fectforAg-NPs,whereachangeofmaterialmorphologyis carbon, sulfur,hydrogen, andoxygen(Table1). visible.Figure6a,bdisplaysdriedpoly(vinylalcohol),which The phase analysis of the GO sample (Fig. 1) revealed was used for stabilizing the water suspension of Ag-NPs. the presence of impurities coming from trace quantities The destructive effect of ultrasonic homogenization was notableasthepolyvinylalcoholstructurewasbrokendown intolongheterogeneouspartswithsmallsphericalopenings Table1Resultsofchemicalanalysesofgrapheneoxide (Fig.6c,d). samples Sample Element(%wt) AverageSizeandZetaPotential N C S H O The results of average particle/agglomerate size in water Grapheneoxide 0.042 48.41 0.390 1.963 49.195 suspensions are presented in Table 2. Analyses of Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page6of17 Fig.1X-RaydiffractionpatternsofGOpowders.ThephaseanalysisoftheGOsamplerevealedthepresenceofimpuritiescomingfromtrace quantitiesofgraphite,sodiumnitrate,andprobablyareducedformofgrapheneoxide average size were carried out for concentrated suspen- suspension stabilization. The large standard deviation of sions which were not subject to ultrasonic the Ag-NP sample homogenized by ultrasound resulted homogenization (as received) and for diluted suspen- from the presence of both loose Ag-NPs and Ag-NPs sions. Diluted suspensions before the test were subject driven into the poly(vinyl alcohol) in the suspension. In to ultrasonic homogenization, with homogenization the case of GO suspension, the average particle size of parameters identical to those used during the ultrasonic the sample subject to ultrasonic homogenization was coating of the foil with nanomaterial layers (Ag-NPs, 263 nm and was ca 7.7 times smaller than the average GO). For Ag-NP suspension, the action of the particle size of the sample that was not subject to ultrasounds caused an increase in the average particle homogenization. The obtained results are convergent size from 80 to 218 nm. The main cause of the increase with the SEM tests (Fig. 5), which show the destructive in the average particle size after ultrasonic effect of ultrasounds on GO flakes. The decrease of the homogenization in the suspension (apart from the average GO particle size was caused by defragmentation process of Ag-NP agglomeration) was that Ag-NPs were or folding of the GO flakes. However, it should be driven into the poly(vinyl alcohol) that was used for emphasized that the results of the average particle size of Fig.2RamanspectrumofgrapheneoxidewithproposeddeconvolutionoftheD,G,D′,2D,andD+Gbands.GOhasahighlydisorderedstructure duetomanyfunctionalgroupsinthestructureformedduringoxidationofgraphitepowder.ThepositionoftheDbandis1351cm−1andtheGband 1590cm−1;theI /I ratiois1.15 D G Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page7of17 Fig.3FT-IR(ATR,attenuatedtotalreflectance)spectrumofgrapheneoxidewithproposedassignmentoffunctionalgroupspresentinGO.The mostnotablepeakswereobservedat~3500cm−1,(attributedtowaterandhydroxylgroups),~1080cm−1(hydroxylgroups),~1600cm−1(assignedto C=Cbondspresentingraphiticcarbon).OtherpeaksobservedontheFT-IRspectrumshowthatGOisrichingroupscontainingC=O(mainlycarboxyl groups),peakaround1720cm−1,epoxy(C–O–C)withthevisiblepeakaround1200cm−1,andC–Hbonds(peakaround2800cm−1) GO suspension samples involve an error related to the sample of Ag-NP suspension was stabilized sterically by nanomaterial shape. The results obtained by the DLS preserving Ag-NP distances through poly(vinyl alcohol) method are a hydrodynamic average that is calculated addition, which prevented agglomeration/aggregation of basedontheshapeofaspherethathasthesamediffusion Ag-NPs. The zeta potential of the GO suspension sam- coefficient as the measured particles; however, the shape ple, in turn, was −41 mV, which gave a moderate elec- ofGOwasflakes,whichwasconfirmedbySEMimages. trostatic stability to the sample. A moderate electrostatic Test results of the zeta potential analysis of samples stability of a sample is characterized by slow sedimenta- are provided in Table 3. The zeta potential of Ag-NPs in tion with virtually negligible change of particle size in a water suspension was merely −5.9 mV, which resulted theperiodofdeclaredfitnessofthesuspension.Thezeta in a lack of electrostatic stability of the sample. The potential result of the mixture of Ag-NPs and GO was Fig.4TEMimagesofagglomeratedGOflakes(a),GOflakesafterultrasonictreatment(b),agglomeratedAg-NPs(c),Ag-NPsafterultrasonictreatment (d),andGO-Ag(e,f).ThedecreaseoftheaverageGOparticlesizeafterultrasonictreatmentwascausedbydefragmentationorfoldingoftheGO flakes.ThedecreaseoftheaverageAgsizeafterultrasonictreatmentwascausedbydefragmentationofAgagglomerates.Note:Arrowspoint toAg-NPs/agglomerates Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page8of17 Fig.5ComparisonofmorphologyoflyophilizedGOflakes(a,b)andGOflakesafterultrasonictreatment(c,d)usingscanningelectronmicroscopy. ThedecreaseoftheaverageGOparticlesizeafterultrasonictreatmentwascausedbydefragmentationorfoldingoftheGOflakes −7.1 mV, which potentially means that during the ac- FoilCharacteristics tion of the ultrasounds, the GO flakes were coated by In order to determine the morphology of the created poly(vinyl alcohol) and Ag-NPs. The obtained zeta layers, four types of foil samples were compared (Fig. 7): potential result of the mixture of Ag-NPs and GO pure polyurethane foil (A, B), GO-coated polyurethane sample in the water suspension implied that electro- foil (C, D), Ag-NP-coated polyurethane foil (E, F), and static stability was not present. GO-Ag mixture-coated polyurethane foil(G,H). Fig.6ComparisonofmorphologyoflyophilizedAg-NPmixture(a,b)andAg-NPmixtureafterultrasonictreatment(c,d)usingscanningelectronmicroscopy Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page9of17 Table2Testresultsofaverageparticle/agglomeratesizein those in Fig. 8e, f. Figure 8c, d shows GO-Ag-NP-coated suspensions film.Figure8eshowsthesurfaceofthefoilalmostentirely Sample SizebyDLS,Z-average, coveredwithGOflakes;thephasecontrastimagehelpsto diameter[nm]±SD observe these two phases, the darker area is GO and the Ag-NPs(asreceived) 80±1 lighterareaispolymerfoil.Itwasnoticedthatthemorph- Ag-NPs(100μg/mL,afterUS) 218±93 ology of the foils has changed after Ag-NP coating comparing to the not-coated foils. The GO-Ag-NP GO(asreceived) 2030±36 coating differs from the previous one because it GO(200μg/mL,afterUS) 263±8 contains also small amounts of GO flakes seen as GO(200μg/mL)-Ag(100μg/mL) 251±10 small black spots on the image, as it was mentioned DLSdynamiclightscattering,Z-averageharmonicintensityaveragedparticle earlier. Figure 8f depicts magnification of one GO diameter,SDstandarddeviation,USultrasoundmethod,Ag-NPs(asreceived) HydroSilver1000(Amepox,Łódź,Poland)materialwasnotprocessedAg-NPs flake made in LFM. The reduced friction confirms (100μg/mL,afterUS)asabovebutwasprocessedunderUSundersimilar that it is, in fact, a GO flake. conditionstothefoilsintheprecedingsections,GO(asreceived)materialnot processed,GO(200μg/mL,afterUS)asabovebutprocessedunderUSunder Thepolarcomponent for the GO-coated foil increased similarconditionstothefoilsintheprecedingsections,GO(200μg/mL)-Ag in relation to pure foil, from 2.3±0.6 to 68.9±2.8 mJ/ (100μg/mL)materialsmixedtogetherandprocessedunderUSundersimilar m2, while the dispersion component decreased from 34. conditionstothefoilsintheprecedingsections 4±1.3 to 8.2±1.2 mJ/m2. SFE increased from 36.7±1.4 Figure 7a, b shows an uncoated polyurethane foil with to 77.0±3.4 mJ/m2. A similar effect was not observed a smoothsurface with singleimpurities.In Fig.4c, d, the on foil surfaces coated with Ag-NPs and GO-Ag mix- broken-down GO flakes deposited on the polyurethane ture. SFE of foil samples coated with Ag-NPs and GO- foilsurfacearenoticeable. Figure7e,fshowsthefoilsur- Agmixturedoesnot differstatistically(Table4). face coated with Ag-NPs on which grid structures com- posed of polyvinyl alcohol and Ag-NPs are observable. AntibacterialProperties Figure 7g, h presents a mixture of GO-Ag composition, The antibacterial activity of the different foils coated which was mixed under the influence of ultrasounds and with GO, Ag-NPs, and GO-Ag were tested with E. coli, depositedonthefoilsurface. S. aureus, S. epidermidis, and C. albicans. Results showed that after co-incubation with bacteria at 37 °C AFMAnalysisandSurfaceFreeEnergy for 24 h, foils inhibited the growth of all tested microor- AFM and LFM were used to complement the informa- ganisms but to various degrees. The maximum antibac- tion about the surface morphology investigated by SEM. terial effect against all tested microorganisms was with The investigation confirmed evolution of surface foil coated with the GO-Ag nanocomposite. The bacter- morphology by sonication of the polyvinyl alcohol with ial growth of the cells treated with foils coated with GO Ag-NP, GO, and GO-Ag NP solutions on the foils. Pure and Ag-NPs was slightly lower than that of cells in the polyurethane foil was used as a reference foil in relation control group whereas the growth of bacterial cells to foils coated by the ultrasonic method. The images in treated with foils coated with GO-Ag was greatly inhib- Fig. 8 are the AFM phase contrast images made in AC; ited, 88.6% of E. coli, 79.6% of S. aureus, 76.5% of S. epi- additionally, cross sections of the GO flakes are at- dermidis,and77.5%ofC.albicans (Figs.9,10,and11). tached under the corresponding images. Figure 8a is an image of pure polyurethane film; Fig. 7b depicts the MembraneIntegrity Ag-NP-coated film, where characteristic and homoge- In cases where the cell membrane is damaged, intracel- neous grid structures are observable, being similar to lular LDH molecules could be released into the culture medium. The LDH level outside the cells demonstrates the cell membrane integrity. Foils coated with GO, Ag- Table3Testresultsofzetapotential NPs, and GO-Ag disrupted cell membrane functionality Sample ZPbyLDE[mV]±SD and integrity with significant differences between control Ag-NPs(100μg/mL,afterUS) −5.9±0,4 groups and the Ag-NPs and GO-Ag groups (Fig. 12). GO(200μg/mL,afterUS) −41.0±3 The highest disruption of cell membranes was observed GO(200μg/mL)-Ag(100μg/mL,afterUS) −7.1±0,5 in the GO-Ag groups, 66.3% of E. coli, 59.4% of S. aur- ZPzetapotential,LDElaserDopplerelectrophoresis,SDstandarddeviation,US eus,54.8%ofS.epidermidis,and48.5% ofC.albicans. ultrasoundmethod,Ag-NPs(asreceived)HydroSilver1000,Amepox,Poland,material wasnotprocessedAg-NPs(100μg/mL,afterUS)asabovebutprocessedunderUS ROSProduction undersimilarconditionstothefoilsintheprecedingsections,GO(asreceived) materialnotprocessed,GO(200μg/ml,afterUS)asabovebutprocessedunderUS Low levels (or optimum levels) of ROS play an import- undersimilarconditionstothefoilsintheprecedingsections,GO(200μg/mL)-Ag ant role in signaling pathways. However, when ROS pro- (100μg/mL)materialsmixedtogetherandprocessedunderUSundersimilar conditionstothefoilsintheprecedingsections duction increases and overwhelms the cellular Jaworskietal.NanoscaleResearchLetters (2018) 13:116 Page10of17 Fig.7Scanningelectronmicroscopyimagesofa,bnon-coatedpolyurethanefoilwithasmoothsurfacewithsingleimpurities;c,dfoilcoated withGO,thebroken-downGOflakesdepositedonthepolyurethanefoilsurface;e,ffoilcoatedwithAg-NPsonwhichgridstructurescomposed ofpolyvinylalcoholandAg-NPsareobserved;andg,hfoilcoatedwithGO-Ag,whichwasmixedundertheinfluenceofultrasoundsanddeposited onthefoilsurface antioxidantcapacity,itcaninducemacromoleculardam- an end, and new antibacterial agents are needed. In re- age (by reactingwith DNA, proteins, and lipids) and dis- cent years, studies have reported nanoparticles as a rupt thiol redox circuits. Foils coated with Ag-NPs and promising alternative to antibacterial reagents because GO-Ag (P<0.05) increased the ROS production of all of their antibacterial activity in several biomedical appli- tested microorganisms compared to the control group. cations [19, 45]. Nanoparticles canbe an effective way to Foils coated with GO only increased the ROS produc- control many pathogenic and antibiotic-resistant micro- tion ofC.albicans.ThehighestROSproduction wasob- organisms. Among many metal nanoparticles, Ag-NPs servedintheGO-Aggroup (Fig. 13). havebeen intensely studied because of the distinct prop- erties of their antibacterial activity [7]. Ag-NPs have Discussion been proved effective against over 650 microorganisms The discovery of antibiotics, natural products produced including bacteria (both gram-positive and negative), by microorganisms that are able to prevent the growth fungi, and viruses; however, the precise mechanism of of bacteria and thus cure infectious diseases, revolution- antimicrobial action is not understood completely [46]. ized medical therapy; however, the overuse and misuse Ag-NP exposure to microorganisms could cause adhe- of antibiotics have been key factors contributing to anti- sion of nanoparticles to the peptidoglycan and the cell biotic resistance. Now, theeraofantibiotics iscomingto membrane [47], penetration inside the cell [48],