Ultrasonics - Sonochemistry 70 (2021) 105317 Contents lists available at ScienceDirect Ultrasonics - Sonochemistry journal homepage: www.elsevier.com/locate/ultson Sonochemical coating of Prussian Blue for the production of smart bacterial- T sensing hospital textiles Amparo Ferrer-Vilanovaa, , Yasmine Alonsob, Jiri Dietvorsta, Marta Pérez-Monteroc, ⁎ Rosalía Rodríguez-Rodríguezc, Kristina Ivanovad, Tzanko Tzanovd, Núria Viguése, Jordi Mase, Gonzalo Guiradob, Xavier Muñoz-Berbela a Instituto de Microelectrónica de Barcelona (IMB-CNM, CSIC), Carrer dels Til·lers s/n, Campus Universitat Autònoma de Barcelona, 08193, Cerdanyola del Vallès, Barcelona, Spain b Departament de Química, Universitat Autònoma de Barcelona, 08193, Cerdanyola del Vallès, Barcelona, Spain c Basic Sciences Department, Faculty of Medicine and Health Sciences, Universitat Internacional de Catalunya, 08195, Sant Cugat del Vallès, Barcelona, Spain d Universitat Politècnica de Catalunya, Edifici Gaia, Pg. Ernest Lluch/Rambla Sant Nebridi s/n. 08222, Terrassa, Barcelona, Spain e Departament de Genètica i Microbiologia, Universitat Autonòma de Barcelona, 08193, Cerdanyola del Vallès, Barcelona, Spain ARTICLE INFO ABSTRACT Keywords: In healthcare facilities, environmental microbes are responsible for numerous infections leading to patient’s Sonochemical coating health complications and even death. The detection of the pathogens present on contaminated surfaces is crucial, Smart textiles although not always possible with current microbial detection technologies requiring sample collection and Bacterial detection transfer to the laboratory. Based on a simple sonochemical coating process, smart hospital fabrics with the Prussian Blue capacity to detect live bacteria by a simple change of colour are presented here. Prussian Blue nanoparticles (PB- Metabolic indicators NPs) are sonochemically coated on polyester-cotton textiles in a single-step requiring 15 min. The presence of Hospital acquired infections PB-NPs confers the textile with an intensive blue colour and with bacterial-sensing capacity. Live bacteria in the textile metabolize PB-NPs and reduce them to colourless Prussian White (PW), enabling in situ detection of bacterial presence in less than 6 h with the bare eye (complete colour change requires 40 h). The smart textile is sensitive to both Gram-positive and Gram-negative bacteria, responsible for most nosocomial infections. The redox reaction is completely reversible and the textile recovers its initial blue colour by re-oxidation with en- vironmental oxygen, enabling its re-use. Due to its simplicity and versatility, the current technology can be employed in different types of materials for control and prevention of microbial infections in hospitals, in- dustries, schools and at home. 1. Introduction microbes on contaminated surfaces in situ and within a reasonable period of time. “Nosocomial” or “hospital-acquired infections” (HAI) are infections Microbial detection is currently possible in the laboratory with a that the patient acquires in the hospital or healthcare facility from number of dedicated techniques and instruments, the most common microbes present in the environment. The main means of transmission being cell culture imaging [3,4], or sequencing [5,6], polymerase chain include the direct physical contact of the patient with a contaminated reaction (PCR) [7] and enzyme-linked immunosorbent assay (ELISA) surface or element, either the skin (of the healthcare personnel or the [8]. With these techniques, samples are collected, transported to the own patient), aerosols, shared instruments, furniture or hospital fabrics. laboratory and analysed using tedious and long protocols, impossible to Recently, pathogenic bacteria have been identified in hospital textiles be implemented as a routing warning system. As a step forwards, a [1,2], but, in practice, there is no way of knowing if a textile in use is number of biosensors have been developed based on the previous sys- contaminated or not. Therefore, to minimize the impact of nosocomial tems where bacteria are identified through selective DNA strains (DNA- infections in hospitalized patients it is crucial to detect the presence of sensors) [9] or antigens/antibodies (immune-sensors) [10]. Portable Corresponding author. ⁎ E-mail addresses: amparo.ferrer@imb-cnm.csic.es (A. Ferrer-Vilanova), yasminee95@gmail.com (Y. Alonso), jiri.dietvorst@imb-cnm.csic.es (J. Dietvorst), mperez@uic.es (M. Pérez-Montero), rrodriguez@uic.es (R. Rodríguez-Rodríguez), kristina.ivanova@upc.edu (K. Ivanova), tzanko.tzanov@upc.edu (T. Tzanov), nuria.vigues@uab.cat (N. Vigués), jordi.mas@uab.cat (J. Mas), gonzalo.guirado@uab.cat (G. Guirado), xavier.munoz@imb-cnm.csic.es (X. Muñoz-Berbel). https://doi.org/10.1016/j.ultsonch.2020.105317 Received 8 April 2020; Received in revised form 2 July 2020; Accepted 23 August 2020 Available online 27 August 2020 1350-4177/ © 2020 Elsevier B.V. All rights reserved. A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 systems have been developed based on the latter [11,12], although they phosphate (KH2PO4) were from Panreac (Spain). Potassium chloride still requiring sample collection and sometimes concentration/amplifi- (KCl) and oxalic acid were purchased from Probus. All chemicals were cation before analysis. Hence, these technologies are still difficult to used as received. Aqueous solutions were prepared using de-ionized implement in situ. This is partially tackled by the use of metabolic in- water. dicators. Metabolic indicators are molecules that change their proper- ties when metabolized by microorganisms. The sensing strategy varies 2.2. Synthesis of PB-NPs depending on the indicator but they are mostly based on changes of pH (bacterial metabolism reduces pH) [13] or conductivity (bacterial me- Depending on the counter-ion used for charge compensation, two tabolism tends to increase medium conductivity) [14], ATP production forms of PB can be synthetized: the “soluble” PB (PBSol), which in- [15] or reduction of redox molecules by proteins and/or mediators from corporates potassium as counter-ion, and the “insoluble” PB (PBIns) the metabolic electron transport chain [16,17]. Although in most cases containing iron ions as counter ions. Both PB forms present important they are used as laboratory reagents, many kits containing these com- structural and functional differences, e.g. in the deposition kinetics. pounds for in situ identification of contaminated surfaces have been That is, while PBSol-NPs present slow deposition kinetics by the small developed [18]. The solution with the indicator is sprayed on the po- size of the particles, PBIns-NPs, containing aggregations of NPs, pre- tentially contaminated surface and an optical change is produced, cipitate quickly [28,32,33]. which can be associated to the presence of living bacteria. This optical PB-NPs were synthetized by a chemical reaction of ferric and hex- change is normally a colour appearance/disappearance (for pH, con- acyanoferrate ions in aqueous solution (i.e. water) until a dark blue ductivity or redox state) or fluorescence production (in the case of colloid is formed [33,34]. For PBSol synthesis, equimolar solutions ATP). Contaminated surfaces are thus identified using a simple and fast (30 mM) of FeCl2 and K3[Fe(CN)6] were used. Concretely, 25 mL of the protocol, with the main limitation of requiring dedicated personnel and ferricyanide solution were added dropwise to 25 mL of iron dichloride equipment to analyse the huge number of potentially risky surfaces and resulting in a dark blue solution corresponding to PBSol. On the other elements present in a hospital. hand, PBIns was synthesized by mixing 50 mL of 15 mM K3 [Fe(CN)6] In the current work, an equipment-free alternative for in situ de- with an excess of FeCl2 (30 mL of a 0.1 M solution). In this case, termination of bacterial contamination is presented based on the in- however, it was the iron dichloride solution which was added dropwise corporation of the sensing metabolic indicators in the surface of in- to the ferricyanide producing the formation of the dark blue precipitate terest, in this case hospital fabrics, through a single-step sonochemical corresponding to the PBIns formation. The precipitate was left to deposit coating process. Sonochemical coating of hospital textiles has been and cleaned twice with distilled water. As previously demonstrated, widely explored by some groups in the recent years. Gedanken’s and light can affect PB-NPs stability [35], hence synthetized nanoparticles Tzanov’s groups have successfully implemented metal nanoparticles were stored protected from light until used. (NPs), metal oxide-NPs and/or enzymes on textiles for the production of antibacterial hospital fabrics [19–22]. Also metal-oxides and dyes could 2.3. Sonochemical coating of polyester-cotton textiles with PB-NPs be co-deposited in a single sonochemical coating process [23]. In the present study, sonochemical coating is employed to incorporate iron- Polyester-cotton fabrics (KLOPMAN, Italy) from industrial produc- based complexes in the textile for the production of smart bacterial- tion of hospital pyjamas were used as model hospital textiles. The so- sensing hospital fabrics. Ferric hexacyanoferrate NPs, also known as nochemical coating process was performed employing an ultrasonic Prussian Blue (PB), has been selected since, apart from being a well- transducer (Ti-horn, 20 kHz, 750 W, Sonics and Materials CV334, USA), known textile dye [24], it presented ideal properties, such as: (i) a as illustrated in Fig. 1a. The power (21.5 W) and intensity (0.43 W/ sufficiently high redox potential, able to react with proteins and med- cm3) were determined calorimetrically by measuring the temperature iators from the bacterial electron transport chain [25]; (ii) high ex- increase in the ultrasonic vessel as detailed in [36]. Polyester-cotton tinction molar coefficients and an evident colour change after reduc- fragments (3 × 3 cm; approx. 0.18 g) were immersed in the ultrasonic tion, even visible to the naked eye [26,27]; (iii) biocompatibility pot already containing 50 mL of the PB-NPs aqueous suspension. In the [28–30]; and (iv) existing preliminary results supporting its use as optimization of the sonochemical coating process, PB at concentrations metabolic indicator by bacterial reduction [31]. The possibility of using from 0.03 to 0.3 mM and sonication times of 5, 15 and 30 min were the metabolic conversion from PB to its reduced form, Prussian White evaluated at a temperature of 20 °C and the amplitude was kept at 35%. (PW) as early warning system for contaminated surface identification is This amplitude was high enough to provide textiles with a homo- evaluated in this paper. The sonochemical implementation of PB-NPs is geneous distribution of NPs, resulting in colour homogeneity, but studied and optimized to obtain sensitive smart bacterial-sensing fab- without compromising the integrity of the textile. Higher amplitudes rics with a homogeneous colour distribution and through a simple were observed to produce textile damage. protocol. The sensing capacity of the final smart fabric is tested using Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus) samples, 2.4. Spectroelectrochemical analysis of the smart textiles used as model for Gram-negative and Gram-positive bacteria, respec- tively. The electrochromic behaviour of PB-NPs was studied after their deposition on polyester-cotton fabrics. Spectroelectrochemical analysis 2. Materials and methods was performed using a sandwich configuration spectroelectrochemical cell. Platinum electrodes were located in front of the PB-coated textiles, 2.1. Reagents with a solid hydrogel electrolyte layer in between, which allowed the study of solid materials without the need of solvents. In this case, 2- Potassium ferricyanide (K3[Fe(CN)6]), iron(II) chloride (FeCl2), hydroxyethyl cellulose was used as solid electrolyte. Cellulose-based hydrogen chloride (HCl 37%), 2-(N-morpholino)ethane-sulfonic acid electrolytes were prepared by dissolving 2-hydroxyethyl cellulose hydrate (MES hydrate), 2-hydroxyethyl cellulose with an average mo- (380 kDa) in 1 M KCl to a final concentration of 6.5% (w/v). Reagents lecular mass of 380 kDa and Dulbecco’s Modified Eagle’s Medium were mixed under magnetic stirring, increasing the temperature until (DMEM, supplemented with 10% fetal bovine serum, 1% penicillin/ boiling. The solution was then left to dry at room temperature. streptomycin solution) were purchased from Sigma-Aldrich (Spain). Electrochemical measurements were performed with a potentiostat PAR Iron (III) chloride anhydrous (FeCl3) was obtained from Fischer biologic, SA, VSP100 (BAS Inc. Tokyo) controlled by the EC-Lab V9.51 Scientific. Glucose, sodium borohydride (NaBH4), potassium phosphate sofware (BAS Inc., Tokyo). Simultaneous optical records were acquired dibasic trihydrate (K2HPO4·3H2O) and potassium dihydrogen with a Hamamatsu L10290 spectrophotometer (Hamamatsu, Japan) 2 A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 Fig. 1. Sonochemical coating process and characterization of PB-modified textiles. (a) Scheme of the sonochemical coating process and resulting textiles obtained after incorporation of PBSol or PBIns. (b) SEM image of the samples modified with PBIns 0.08 mM. (c) PB-modified textile reduced to PW after addition of NaBH4 0.5 M as reducing agent. (d) Colour stability of polyester-cotton textiles modified with PBIns 0.08 mM after 96 h submerged in (i) water, (ii) LB, 0.1 M phosphate buffered saline (PBS) at (iii) pH 6.2 and (iv) pH 7.2, and (v) 0.1 M MES buffer (pH 6.2), all supplemented with 0.1% glucose. Lost colour was expressed as percentage versus unmodified textiles colour (n = 3). controlled by the Bio-Kine 32 V4.46 software (BAS Inc. Tokyo). 2.7. Evaluation of bacterial-sensing activity of smart textiles incorporating PB-NPs 2.5. SEM measurements To study the bacterial-sensing capacity of the PB-modified textiles, The surface topology of the PB-modified textiles was studied by polyester-cotton fabrics (1 × 1 cm, approx. 0.02 g) modified with scanning electron microscopy (SEM) using an AURIGA® series en04 0.08 mM PBIns were incubated in bacterial suspensions (from 105 to 109 SEM (CarlZeiss). CFU mL−1) in (i) water, (ii) LB, 0.1 M phosphate buffered saline (PBS) at (iii) pH 6.2 and (iv) pH 7.2, and (v) 0.1 M MES buffer (pH 6.2), all of them supplemented with 0.1% glucose. Samples were always analysed 2.6. Bacterial cultures preparation during the first 15 days of preparation to ensure repeatability of the results. Colour changes were monitored in real time with a digital mi- Escherichia coli (ATCC 25922) (E. coli) and Staphylococcus aureus croscope camera DigiMicro 2.0 Scale for five days. Selected images (ATCC 29213) (S. aureus) were used as model Gram-negative and were analysed using the freeware ImageJ. The three RGB channels were Gram-positive bacteria, respectively. Both bacteria were incubated separated choosing the red one for quantification since it was the most aerobically in a Luria-Bertani (LB) broth overnight (18 h) at 37 °C with sensitive. The percentage of colour lost was calculated by comparing it shaking. After centrifugation (Eppendorf centrifuge AG R134A) at with the initial colour of the unmodified polyester-cotton textile. 2700xg for 10 min, the pellet was collected and re-suspended in 0.1 M MES (pH 6.2) adjusting bacterial concentration through optical density measurements in a spectrophotometer (Smartspec™ Plus spectro- 2.8. Cytotoxicity evaluation photometer, Bio-rad, California, US). An optical density of 0.1 A.U. was considered equivalent to 108 colony forming units per mL (CFU mL−1). PB cytotoxicity was determined in the human fibroblast cell line Suitable aliquots were cultured in agar plates and counted after in- MRC-5 (ATCC Line Bank, Virginia, USA) by using the colorimetric MTT cubation for adjustment of bacterial concentration. (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. 3 A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 Fig. 2. Spectroelectrochemical measurements on the modified textiles. (a) Set-up scheme used for the spectroelectrochemical measurements on the textiles. (b) CVs of PB-modified textiles developed between 0.4 and −0.3 V at different scan rates (1, 5, 10 and 20 mV s−1) and representation of current peak values obtained vs scan rate. (c) Comparative spectra obtained from textiles coated with PBIns freshly prepared and PBIns prepared 4 days before the coating process and corresponding spectra after apply −0.5 V during 15 min to each one of them. The protocol was based on a previous publication [37]. Briefly, MRC-5 suspension. High-intensity ultrasound is a simple, fast, efficient and cells were cultured in complete Dulbecco’s Modified Eagle’s Medium environmentally friendly approach for surface functionalization. Under (DMEM, supplemented with 10% foetal bovine serum, 1% penicillin/ ultrasound high-speed microjets are generated upon the implosive streptomycin solution). When confluent, cells were exposed to different collapse of bubbles formed in the liquid, the so-called cavitation phe- concentrations of PBSol and PBIns. After 24 h of incubation, wells were nomenon. These microjets project the NPs found in the vicinity of the washed with PBS, and the MTT solution (5 mg mL−1) was added and collapsing bubbles towards the textiles forming stable and homo- incubated for 1 h (37 °C, 5% CO2). After washing with PBS, the purple geneous coatings [22,36,38,39]. formazan generated by MTT metabolism of viable cells was solubilized In this work, polyester-cotton samples (3 × 3 cm) were immersed in with DMSO. The optical density of each well was determined at 570 nm aqueous suspensions of PB-NPs and sonochemically coated varying the in a spectrophotometer reader (BioTek® Synergy HT, Vermont, USA). sonication time (from 5 to 30 min), PB type (PBSol or PBIns) and con- Cell viability was expressed as percentage in relation to non-treated centration (from 0.03 to 30 mM) (Fig. S1). From the two forms of PB, cells. only PBIns was stably incorporated in the textile, while PBSol was almost completely removed after a soft cleaning process by rinsing with buffer 3. Results and discussion (Fig. 1a). Also, the ageing of the PBIns suspension influenced the so- nochemical coating. Freshly prepared samples presented more homo- 3.1. Sonochemical coating and characterization of the smart textiles geneity and colour intensity than their 4-days old counterparts (Fig. S2). Homogeneous and highly-coloured samples were obtained after PB was selected as bacterial-sensing probe in the development of the coating with freshly-prepared PB suspensions containing 0.08 mM PB smart textiles, based on previously reported results demonstrating its for at least 15 min of sonication. No relevant changes were observed potential as metabolic bacterial indicator [31]. In the sonochemical when longer sonication times were applied. SEM image of this sample coating of PB-NPs, high-intensity ultrasound is applied to the NPs revealed homogeneously distributed, although sometimes aggregated, 4 A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 NPs on the textile fibers (Fig. 1b). Accordingly, the single-step sono- chemical coating of freshly-prepared PB samples containing 0.08 mM for 15 min was chosen as optimal coating conditions and used in further experiments. The oxidative capacity of the coated textile was evaluated using NaBH4 as reducing agent. When in contact with 0.5 M NaBH4, textiles immediately lost the blue colour by the formation of PW (Fig. 1c), which confirmed the redox activity of the smart textile, recovering the initial colour when in contact with hydrogen peroxide. In dried con- ditions, coated samples were stable for more than four months with minimal decolouration. The stability was lower in solution and de- pended on the medium (Fig. 1d). First, the stability of PB-NPs in the textile was deeply influenced by the pH. It was clearly evidenced when comparing the samples incubated in the physiologic PBS buffer at pH Fig. 3. Scheme of the smart textile behaviour when in contact with live and dead bacteria. Living bacteria are represented in green colour and the dead one 7.2, which showed an important decrease in colour intensity, with those in red. (For interpretation of the references to colour in this figure legend, the incubated in PBS at pH 6.2, which maintained more than 70% of the reader is referred to the web version of this article.) initial colour for 4 days. This was in agreement with previous results concluding that hydrolysed ferric ions in the coordination sphere of PB could only be substituted at a pH lower than 6.4 [28,33,40–42]. Apart Fig. 4. Relation between bacterial-sensing capacity of the textiles and bacterial proliferation. Different parameters obtained after PB-NPs modified textiles were incubated with E. coli for 72 h, such as (a) textile loss of colour percentage; (b) E. coli concentration by cell counting and (c) absorbance values at 600 nm throughout the experiment. E. coli concentrations above 109 CFU mL−1 in 900 µL were incubated in different media (H2O, LB, 0.1 M PBS (at pH 6.2 and 7.2) and 0.1 M MES (pH 6.2)) and 0.1% of glucose was used (n = 3). A common legend for all the graphs is on top. 5 A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 Fig. 5. Antibacterial activity of PB-modified textiles and cytotoxicity. (a) Survival percentage of E. coli measured at different incubation times for polyester-cotton textiles with and without PB in MES 0.1 M medium at pH 6.2 and with 0.1% of glucose. As control, an inoculated solution in the same medium without textile sample was used (n = 3). (b) MTT cytotoxicity assay for PBSol (red circles) and PBIns (black squares) solutions at different concentrations. Concentrations are represented in a logarithmic scale and the one used in the textiles (0.08 mM PB) is highlighted in the graph with blue cross lines. Cell viability was determined after 24 h of incubation with both solutions (n = 3) and viability was expressed as percentage versus non-treated control samples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) from the pH, the medium composition also had a role in the stability of in the textile should confer it with bacterial-sensing activity through the PB-NPs. Coated samples were more stable in MES than in PBS, even mechanism illustrated in Fig. 3. Once the sonochemical coating of PB- when both buffers had the same pH (pH 6.2). The reason for that may NPs provides the fabrics with an intense blue colour, the attachment of be in the monovalent ions content, e.g. sodium and potassium, which is live bacteria would change the smart fabric colour by the metabolic much higher in PBS than in MES. The presence of such ions, which may reduction of PB to PW by proteins (e.g. cytochrome) and mediators (e.g. substitute iron as counterions in the PB structure, might reduce the ubiquinone) from the bacterial electron transport chain [44]. Since the stability of NPs incorporated in the textile. In fact, overnight incubation reaction is metabolic, dead bacteria would not be able to produce this of the textiles with KCl, LiCl or NaNO3 led to the dissociation of the PB change. If live bacteria died, for example by an antibacterial treatment, and colour disappearance (Fig. S3). LB medium, slightly acidic and with the atmospheric oxygen would re-oxidize PW to PB [45], and the smart monovalent ions in its composition, maintained around 70% of the textile would recover its initial blue colour. initial colour similarly to PBS at pH 6.2. On the other hand, distilled Bacterial-sensing capacity of PB-coated textiles was evaluated water, acidic on its own but without important concentration of spectroscopically (as % colour change) in distilled H2O, LB, PBS (at pH monovalent counter ions, presented low colour losses comparable to 6.2 and 7.2) and MES buffer (pH 6.2) containing between 108 and 109 MES. Therefore, both ionic composition and pH influenced the stability CFU mL−1 E. coli (Fig. 4a). Since bacterial proliferation depends on the of PB-NPs incorporated in the textile, but pH-mediated PB decomposi- medium composition, the evolution of this parameter over time was tion presented faster dissolution kinetics. determined in each medium through plating (Fig. 4b) and optical The reversibility of PB-NPs incorporated in the textile was studied density measurements at 600 nm (Fig. 4c). spectroelectrochemically to evaluate structural/compositional changes All modified textiles presented important colour changes over time, along reduction/oxidation cycles. A sandwich configuration with a stronger than those observed in the stability studies. Three different solid electrolyte, i.e. 2-hydroxyethyl cellulose was used to study the behaviours were observed depending on the media. First, in LB, bac- activity of the smart textiles without solvents (Fig. 2a). CVs (between teria almost doubled the initial concentration in the sample during the 0.4 to −0.3 V) at different scan rates (1, 5, 10 and 20 mV s−1) clearly experiment, enabling a fast metabolic reduction of PB resulting in a reported two peaks, corresponding to the reduction from PB to PW (at 60% colour loss after 72 h, the largest reported. Second, in buffer, ei- 0 V) and the subsequent re-oxidation of PW to PB at 0.13 V (Fig. 2b). ther PBS or MES, bacterial concentration slightly decreased over time, Peak intensities increased directly with the scan rate and the relation- which resulted in weaker colour changes of 40–45%. The higher colour ship between intensity of the anodic and cathodic peaks was approxi- change observed in the PBS sample at pH 7.2 was again attributed to mately 1, which corresponded to the behaviour of a thin film system the pH-mediated decomposition of PB. In fact, when set at the same pH, [43]. These samples were also very stable to electrolysis. After applying PBS and MES presented very similar colour change kinetics, as well as −0.5 V potential for 15 min, the sample only lost 20% of the initial analogous bacterial concentrations and bacterial proliferation rates. colour. Much more significant colour loss was observed in the case of Finally, textiles incubated in distilled water presented an initial colour samples prepared with old PB suspension, which lost more than 90% of loss of 30%, followed by a fast colour recovery by oxygen re-oxidation colour due to their low stability (Fig. 2c). It confirmed that freshly when the bacteria died (which was confirmed by cell counting). prepared PB-NPs in textiles were stable and reversible in solid state Regarding the latter, smart textile reported the presence/absence of when monovalent cations were not present in the medium. living bacteria on real time thanks to its reversibility. To confirm that PB-NPs detected bacterial activity without affecting microbial structure and/or function, the antibacterial activity of the 3.2. Bacterial-sensing activity of smart textiles smart textiles was evaluated. Antibacterial activity was determined by incubation of the samples in E. coli suspensions in MES (pH 6.2) As an already demonstrated metabolic bacterial indicator [31], PB 6 A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 Fig. 6. Bacterial-sensing activity of PB-modified textiles. Quantification of PB-modified textiles colour loss after incubation with E. coli and S. aureus in MES medium (pH 6.2) and 0.1% of glucose. Bacteria concentrations are shown in a common legend. As control, PB-modified textiles were submerged in MES solution, without bacteria. Colour loss was expressed as percentage versus unmodified-textiles colour (n = 3). In the schematic table, representative images from the textiles can be seen at different times of the experiment with E. coli inoculation, when they were in contact with air and after adding an oxidizing agent. supplemented with 0.1% glucose. Modified textiles, non-modified tex- mL−1, which was even detectable with the bare eye after 6 h of in- tiles (without PB-NPs) and control samples (bacterial suspension cubation. The complete colour loss required 48 h. In the case of S. without textile) were studied. The number of viable bacteria was re- aureus, the peptidoglycan layer typical in Gram-positive bacteria slows presented as a survival percentage from the initial concentration down the extracellular electron transfer to the PB-NPs [46], showing (Fig. 5a). No significant differences were observed between control, slower kinetics. After 90 h, when most of bacteria in the sample were non-modified and modified textiles, confirming the low or null anti- already dead as result of nutrients depletion, the textiles started to re- bacterial activity of the PB-NPs in the textile. cover their initial blue colour through oxygen re-oxidation. The com- On the other hand, because of the direct physical contact with skin, plete colour recovery required a reaction with an oxidizing agent, i.e. cytotoxicity assays of PB-NPs were performed to evaluate its potential hydrogen peroxide, as shown in the images, confirming the stability of risk for human health. Fig. 5b shows that both soluble and insoluble PB- PB-NPs in the textiles. NPs were cytotoxic when in high concentrations, with a decrease of the cell viability near to 50%. However, the concentration implemented in 4. Conclusions the textiles was not cytotoxic and should not represent a risk. Finally, the capacity to detect different bacterial types was eval- To prevent HAI contamination, a smart hospital fabric able to detect uated by comparing the response of the smart textile to the Gram-ne- the presence of living bacteria by changing its own colour is developed gative E. coli with the Gram-positive S. aureus bacteria. Experiments here based on the incorporation of sensing NPs by sonochemical were performed in MES (pH 6.2) with 0.1% of glucose at bacterial coating. PB-NPs are chosen as the most suitable metabolic indicator for concentrations of 107, 108 or 109 CFU mL−1. Results are summarized in being able to identify living bacteria by producing an intense and re- Fig. 6. In both cases, no detectable changes were observed in the control versible colour change when it is reduced by bacterial metabolism to sample (without bacteria) or after 5 days of incubation in bacterial PW. From the two forms of PB, only freshly-prepared PBIns-NPs can be suspension with a concentration below 107 CFU mL−1. Textiles in- sonochemically incorporated in textile with sufficient concentration cubated with 108 CFU mL−1 presented a small colour change that was and homogeneity. Optimal conditions require sonication of quickly reversed until it recovered the initial colour. More pronounced 0.08 mM PB-NPs for 15 min in a single-step ultrasonic process. PB-NPs and sustained changes were observed after incubation with 109 CFU in the textile are not antibacterial or cytotoxic, and maintain the 7 A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 bacterial-sensing capacity reported in solution both for Gram-positive [8] G.M. Wyatt, M.N. Langley, H.A. Lee, M.R.A. Morgan, Further studies on the feasi- and Gram-negative bacteria. In the latter case, a detectable colour bility of one-day Salmonella detection by enzyme-linked immunosorbent assay, Appl. Environ. Microbiol. 59 (5) (1993) 1383–1390. change is observed after 5–6 h of incubation in high bacterial con- [9] X. Mao, L. Yang, X.L. Su, Y. Li, A nanoparticle amplification based quartz crystal centrations (109 CFU mL−1), with a total colour change after 40 h. The microbalance DNA sensor for detection of Escherichia coli O157:H7, Biosens. textile recovers the colour when bacteria die by re-oxidation mediated Bioelectron. 21 (7) (2006) 1178–1185, https://doi.org/10.1016/j.bios.2005.04. 021. by environmental oxygen. Thus, these smart textiles allow the detection [10] K. Shirasu, The HSP90-SGT1 chaperone complex for NLR immune sensors, Annu. of live bacteria in situ, which are even detectable by the naked eye. Rev. Plant Biol. 60 (1) (2009) 139–164, https://doi.org/10.1146/annurev.arplant. Future efforts are focused on the improvement of the response time and 59.032607.092906. [11] K. Yamanaka, et al., Quantitative detection for porphyromonas gingivalis in tooth the sensitivity of the technology, either by chemical modification of the pocket and saliva by portable electrochemical DNA sensor linked with PCR, probe or by integration of transduction mechanisms. Also, the im- Electroanalysis 26 (12) (2014) 2686–2692, https://doi.org/10.1002/elan. plementation of the current sonochemical coating protocol in real 201400447. [12] A.N. Naimushin, S.D. Soelberg, D.U. Bartholomew, J.L. Elkind, C.E. Furlong, A production lines is now being explored. We envision the current portable surface plasmon resonance (SPR) sensor system with temperature reg- strategy of potential interest in the control of infections in hospitals, ulation, Sensors Actuators, B Chem. 96 (1–2) (2003) 253–260, https://doi.org/10. industries, schools or at home. 1016/S0925-4005(03)00533-1. [13] M.L. Pourciel-Gouzy, W. Sant, I. Humenyuk, L. Malaquin, X. Dollat, P. Temple- Boyer, Development of pH-ISFET sensors for the detection of bacterial activity, CRediT authorship contribution statement Sensors Actuators, B Chem. 103 (1–2) (2004) 247–251, https://doi.org/10.1016/j. snb.2004.04.056. Amparo Ferrer-Vilanova: Conceptualization, Formal analysis, [14] Y.-C. Lu, et al., Bacteria detection utilizing electrical conductivity, Biosens. Bioelectron. 23 (2008) 1856–1861, https://doi.org/10.1016/j.bios.2008.03.005. Investigation, Visualization, Writing - original draft. Yasmine Alonso: [15] R. Mempin, H. Tran, C. Chen, H. Gong, K. Kim Ho, S. Lu, Release of extracellular Validation, Investigation. Jiri Dietvorst: Investigation, Writing - re- ATP by bacteria during growth, BMC Microbiol. 13 (1) (2013) 1–13, https://doi. view & editing. Marta Pérez-Montero: Investigation, Resources. org/10.1186/1471-2180-13-301. [16] K. Richter, M. Schicklberger, J. Gescher, Dissimilatory reduction of extracellular Rosalía Rodríguez-Rodríguez: Investigation, Resources, Writing - re- electron acceptors in anaerobic respiration, Appl. Environ. Microbiol. 78 (4) (2012) view & editing. Kristina Ivanova: Validation, Investigation, Resources, 913–921, https://doi.org/10.1128/AEM.06803-11. Writing - review & editing. Tzanko Tzanov: Conceptualization, [17] K.L. Straub, M. Benz, B. Schink, Iron metabolism in anoxic environments at near neutral pH, FEMS Microbiol. Ecol. 34 (3) (2001) 181–186, https://doi.org/10. Validation, Project administration. Núria Vigués: Investigation, 1016/S0168-6496(00)00088-X. Resources. Jordi Mas: Validation, Resources. Gonzalo Guirado: [18] M.P. Buttner, P. Cruz, L.D. Stetzenbach, A.K. Klima-Comba, V.L. Stevens, Conceptualization, Methodology, Writing - review & editing. Xavier P.A. Emanuel, Evaluation of the biological sampling kit (BiSKit) for large-area surface sampling, Appl. Environ. Microbiol. 70 (12) (2004) 7040–7045, https://doi. Muñoz-Berbel: Conceptualization, Methodology, Writing - original org/10.1128/AEM.70.12.7040-7045.2004. draft, Writing - review & editing, Supervision, Funding acquisition. [19] M. Salat, P. Petkova, J. Hoyo, I. Perelshtein, A. Gedanken, T. Tzanov, Durable an- timicrobial cotton textiles coated sonochemically with ZnO nanoparticles em- Declaration of Competing Interest bedded in an in-situ enzymatically generated bioadhesive, Carbohydr. Polym., 189, (November) 2017, 198–203, 2018, 10.1016/j.carbpol.2018.02.033. [20] J. Beddow et al., Sonochemical coating of textile fabrics with antibacterial nano- The authors declare that they have no known competing financial particles, AIP Conf. Proc., vol. 1433, no. May 2014, pp. 400–403, 2012, doi: 10. 1063/1.3703213. interests or personal relationships that could have appeared to influ- [21] I. Perelshtein, et al., Making the hospital a safer place by sonochemical coating of all ence the work reported in this paper. its textiles with antibacterial nanoparticles, Ultrason. Sonochem. 25 (2015) 82–88, https://doi.org/10.1016/j.ultsonch.2014.12.012. [22] P. Petkova, A. Francesko, I. Perelshtein, A. Gedanken, T. Tzanov, Simultaneous Acknowledgements sonochemical-enzymatic coating of medical textiles with antibacterial ZnO nano- particles, Ultrason. Sonochem. 29 (2016) 244–250, https://doi.org/10.1016/j. This work was supported the European Commission through the ultsonch.2015.09.021. [23] I. Perelshtein, A. Lipovsky, N. Perkas, T. Tzanov, A. Gedanken, Sonochemical co- project PROTECT (H2020-NMBP-PILOT-720851). The authors thank deposition of antibacterial nanoparticles and dyes on textiles, Beilstein J. the Ministerio de Ciencia e Innovación of Spain for financial support Nanotechnol. 7 (1) (2016) 1–8, https://doi.org/10.3762/bjnano.7.1. though the projects CTQ 2015-65439-R and PID2019-106171RB-I00. [24] C. Gervais, et al., Why does Prussian blue fade? understanding the role(s) of the substrate, J. Anal. At. Spectrom. 28 (10) (2013) 1600–1609, https://doi.org/10. 1039/c3ja50025j. Appendix A. Supplementary data [25] P. Ertl, B. Unterladstaetter, K. Bayer, S.R. Mikkelsen, Ferricyanide reduction by Escherichia coli: Kinetics, mechanism, and application to the optimization of re- Supplementary data to this article can be found online at https:// combinant fermentations, Anal. Chem. 72 (20) (2000) 4949–4956, https://doi.org/ 10.1021/ac000358d. doi.org/10.1016/j.ultsonch.2020.105317. [26] D.R. Rosseinsky, L. Glasser, H.D.B. Jenkins, Thermodynamic clarification of the curious ferric/potassium ion exchange accompanying the electrochromic redox References reactions of Prussian Blue, iron(III) hexacyanoferrate(II), J. Am. Chem. Soc. 126 (33) (2004) 10472–10477, https://doi.org/10.1021/ja040055r. [27] P. Ding, G. Song, J. Zhou, Q. Song, Collection of rolling fingerprints by the elec- [1] L. Sehulster, R.Y.W. Chinn, Guidelines for environmental infection control in trochromism of Prussian blue, Dye. Pigment. 120 (2015) 169–174, https://doi.org/ health-care facilities, Morb. Mortal. Wkly. Rep. 52 (RR10) (2003) 1–42. 10.1016/j.dyepig.2015.04.019. [2] S. Fijan, S.Š. Turk, Hospital textiles, are they a possible vehicle for healthcare-as- [28] F. Ricci, G. Palleschi, Sensor and biosensor preparation, optimisation and applica- sociated infections? Int. J. Environ. Res. Public Health 9 (9) (2012) 3330–3343, tions of Prussian Blue modified electrodes, Biosens. Bioelectron. 21 (3) (2005) https://doi.org/10.3390/ijerph9093330. 389–407, https://doi.org/10.1016/j.bios.2004.12.001. [3] T. Zimmermann, J. Rietdorf, R. Pepperkok, Spectral imaging and its applications in [29] P.A. Fiorito, V.R. Gonçales, E.A. Ponzio, S.I. Córdoba De Torresi, Synthesis, char- live cell microscopy, FEBS Lett. 546 (1) (2003) 87–92, https://doi.org/10.1016/ acterization and immobilization of Prussian blue nanoparticles. A potential tool for S0014-5793(03)00521-0. biosensing devices, Chem. Commun. no. 3 (2005) 366–368, https://doi.org/10. [4] M. Arora, Cell culture media: a review, Mater. Methods 3 (2013) 1–29, https://doi. 1039/b412583e. org/10.13070/mm.en.3.175. [30] A.A. Karyakin, Advances of Prussian blue and its analogues in (bio)sensors, Curr. [5] B.S. Yadav, V. Ronda, D.P. Vashista, B. Sharma, Sequencing and computational Opin. Electrochem. 5 (1) (2017) 92–98, https://doi.org/10.1016/j.coelec.2017.07. approaches to identification and characterization of microbial organisms, Biomed. 006. Eng. Comput. Biol. 5 (2013), https://doi.org/10.4137/becb.s10886 p. BECB. [31] M.K. Jahn, S.B. Haderlein, R.U. Meckenstock, Reduction of Prussian Blue by the two S10886. iron-reducing microorganisms Geobacter metallireducens and Shewanella alga, [6] S.J. Park, S. Onizuka, M. Seki, Y. Suzuki, T. Iwata, K. Nakai, A systematic se- Environ. Microbiol. 8 (2) (2006) 362–367, https://doi.org/10.1111/j.1462-2920. quencing-based approach for microbial contaminant detection and functional in- 2005.00902.x. ference, BMC Biol. 17 (1) (2019) 72, https://doi.org/10.1186/s12915-019-0690-0. [32] D. Ellis, M. Eckhoff, V.D. Neff, Electrochromism in the mixed-valence hexacyanides. [7] C.W. Brasher, A. DePaola, D.D. Jones, A.K. Bej, Detection of microbial pathogens in 1. Voltammetric and spectral studies of the oxidation and reduction of thin films of shellfish with multiplex PCR, Curr. Microbiol. 37 (2) (1998) 101–107, https://doi. Prussian blue, J. Phys. Chem. 85 (9) (1981) 1225–1231, https://doi.org/10.1021/ org/10.1007/s002849900346. j150609a026. 8 A. Ferrer-Vilanova, et al. Ultrasonics - Sonochemistry 70 (2021) 105317 [33] A.A. Karyakin, Prussian Blue and its analogues: electrochemistry and analytical [40] M. Cheng, et al., In situ formation of pH-responsive Prussian blue for photoacoustic applications, Electroanal. 13 (10) (2001) 813–819, https://doi.org/10.1002/1521- imaging and photothermal therapy of cancer, RSC Adv. 7 (30) (2017) 4109(200106)13:10<813::AID-ELAN813>3.0.CO;2-Z. 18270–18276, https://doi.org/10.1039/C7RA01879G. [34] D. Davidson, L.A. Welo, The nature of prussian blue, J. Phys. Chem. 32 (8) (1928) [41] Hatamie, Amir, Zargar, Behrooz, A. Jalali, H. Ameri, Colorimetric assay for 4- 1191–1196, https://doi.org/10.1021/j150290a007. phenylthiosemicarbazide detection in environmental samples based on Prussian [35] C. Gervais, M. A. Languille, S. Reguer, C. Garnier, M. Gillet, Light and anoxia fading Blue nanoparticles formation ion, Iran. J. Chem. Chem. Eng, 36 (1), 2017. of Prussian blue dyed textiles, Herit. Sci., vol. 2, no. 1, 2014, doi: 10.1186/s40494- [42] B. Zargar, A. Hatamie, Prussian blue nanoparticles: A simple and fast optical sensor 014-0026-x. for colorimetric detection of hydralazine in pharmaceutical samples, Anal. Methods [36] P. Petkova, et al., Sonochemical coating of textiles with hybrid ZnO/chitosan an- 6 (15) (2014) 5951–5956, https://doi.org/10.1039/c4ay00618f. timicrobial nanoparticles, ACS Appl. Mater. Interfaces 6 (2) (2014) 1164–1172, [43] A.J. Bard, L.R. Faulkner, E. Swain, C. Robey, Electrochemical Methods. https://doi.org/10.1021/am404852d. Fundamentals and Applications, 1980. [37] A. Alonso, et al., Superparamagnetic Ag@Co-nanocomposites on granulated cation [44] F. Kracke, I. Vassilev, J.O. Krömer, Microbial electron transport and energy con- exchange polymeric matrices with enhanced antibacterial activity for the en- servation – The foundation for optimizing bioelectrochemical systems, Front. vironmentally safe purification of water, Adv. Funct. Mater. 23 (19) (2013) Microbiol., 6 (JUN), 1–18, 2015, doi: 10.3389/fmicb.2015.00575. 2450–2458, https://doi.org/10.1002/adfm.201202663. [45] K. Catterall, D. Robertson, P.R. Teasdale, D.T. Welsh, R. John, Evaluating use of [38] I. Perelshtein, et al., CuO-cotton nanocomposite: formation, morphology, and an- ferricyanide-mediated respiration bioassays to quantify stimulatory and inhibitory tibacterial activity, Surf. Coatings Technol. 204 (1–2) (2009) 54–57, https://doi. effects on Escherichia coli populations, Talanta 80 (5) (2010) 1980–1985, https:// org/10.1016/j.surfcoat.2009.06.028. doi.org/10.1016/j.talanta.2009.10.057. [39] T. Harifi, M. Montazer, A review on textile sonoprocessing: a special focus on so- [46] G. Pankratova, L. Hederstedt, L. Gorton, Extracellular electron transfer features of nosynthesis of nanomaterials on textile substrates, Ultrason. Sonochem. 23 (2015) Gram-positive bacteria, Anal. Chim. Acta 1076 (2019) 32–47, https://doi.org/10. 1–10, https://doi.org/10.1016/j.ultsonch.2014.08.022. 1016/j.aca.2019.05.007. 9