nanomaterials Article Nanoplasmonic Paper-Based Platform for General Screening of Biomacromolecules Ferran Pujol-Vila 1, * , Andrew Tobias Aveling Jenkins 2 , Xavier Muñoz-Berbel 1 and Jordi Mas Gordi 3 1 Instituto de Microelectrónica de Barcelona (IMB-CNM, CSIC), Bellaterra, 08193 Barcelona, Spain; xavier.munoz@imb-cnm.csic.es 2 Department of Chemistry, University of Bath, Bath BA2 7AY, UK; chsataj@bath.ac.uk 3 Department of Genetics and Microbiology, Universitat Autònoma de Barcelona (UAB), Bellaterra, 08193 Barcelona, Spain; jordi.mas@uab.cat * Correspondence: ferran.pujol@imb-cnm.csic.es  Received: 30 October 2020; Accepted: 22 November 2020; Published: 25 November 2020  Abstract: Hygiene assessment in industrial and clinical environments is crucial in the prevention of health risks. Current technologies for routine cleanliness evaluation rely on the detection of specific biomolecules, thus requiring more than one test for broad-range screening. Herein, the modulation of the catalytic activity of gold nanoparticles (AuNPs) by biomacromolecules was employed to develop a nanoplasmonic platform for general hygiene screening. AuNPs were immobilized on cellulose paper by simple adsorption. When ferricyanide was dispensed onto the paper, the AuNPs catalysed the ferricyanide’s dissociation, releasing free cyanide ions that dissolved them. The AuNP dissolution produced an intense colour shift detectable with the naked eye. When biomacromolecules (e.g., proteins and polysaccharides) were present, they spontaneously attached to AuNPs, forming a biomolecular corona (biocorona), reducing their catalytic activity until complete suppression when the NPs were fully covered by molecules. The concentration-dependent decrease in the catalytic activity was here used to quantify biomacromolecules and complex samples such as milk, eggs, soy sauce and yeast extract (in 20 min), with detection limits comparable to those of standard methods, i.e., 0.25 µg mL−1 for albumin. This nano-enabled technology may be applied as a broad-range (unspecific) alert system for inexpensive cleanliness evaluation, with potential applications in sensitive sectors including productive industries and hospitals. Keywords: nanoplasmonic platform; gold nanoparticles; biomolecular coronas; nanocatalytic activity; hygiene screening 1. Introduction Hygiene management is critical in environments such as in the food, beverage and pharma industries; clinical settings; and public spaces such as schools, universities or kindergartens, where contamination is becoming a risk for public health. Apart from intensive cleaning and disinfection, many sectors have implemented routine monitoring protocols to ensure the sufficient cleanliness of surfaces and to avoid biological risks. Hygiene levels in healthcare and industrial settings are currently evaluated by means of technologies detecting specific biomolecules as a proxy for organic matter and/or microbial contamination. Adenosine triphosphate (ATP) and protein testing are among the most widespread methods [1,2]. The ATP bioluminescence assay uses the firefly enzyme luciferase to convert ATP into adenosine monophosphate (AMP) with light emission, which is monitored by a luminometer. This approach allows the rapid and sensitive detection of microbial contamination, since ATP is the universal energy transfer molecule of living cells, but is unable to detect non-metabolic Nanomaterials 2020, 10, 2335; doi:10.3390/nano10122335 www.mdpi.com/journal/nanomaterials Nanomaterials 2020, 10, 2335 2 of 13 contaminants, e.g., viruses. Protein detection at trace levels is achieved thanks to colorimetric assays based on Bradford or Lowry reagents, among others. Although total protein may be considered a good biological indicator, a host of other biomolecules such as lipids and sugars may promote microbial growth in contaminated environments. Unfortunately, their assays are specific to concrete molecules or families and may not reflect the complexity of biological contamination, which may require the combination of various tests. This becomes particularly important considering the current threats of virus epidemics as well as the harmful effects of secreted bio-products such as microbial toxins [3,4]. Thus, the whole risk is impossible to attain with the selective tests developed to date, and new strategies with wider detection spectra are required to obtain a realistic picture of potential contaminants [5]. Nanomaterials, with their unique sensing [6] and catalytic properties [7] widely surpassing those of their bulk counterparts, are positioned as a very powerful alternative to conventional technologies. In this regard, gold nanoparticles (AuNPs) are taking a preferential position for their unmet catalytic activity, being able to catalyse glucose oxidation or ferricyanide dissociation [8,9], as well as their plasmonic properties that confer their suspensions a characteristic red colour. This plasmonic nature has been widely used in lateral flow analysis for the fast detection of biomarkers in a very low-cost and simple paper-based format [10]. Additionally highly reported, the plasmonic peak position shifts after NP aggregation [11]. This property has been employed in biosensing too, for example, in the detection of the proteins through ionically induced aggregation [12]. This detection scheme is based on the fact that, due to their high surface energy, nanosized materials are prone to attracting biomacromolecules, e.g., proteins, that attach on their surface until completely covering them, forming what is called a biomolecular corona or biocorona [13–16]. This process is mainly driven by electrostatic, van der Waals and hydrogen bonding [17,18]. Protein-coated AuNPs are sterically stabilized and insensitive to changes in the ionic strength of the medium, while naked AuNPs aggregate, changing their optical properties [19]. This assay allows the fast and simple detection of protein content in samples at trace levels. However, aggregation-based methods are restricted to suspended NPs, being unsuitable for detection in solid-state substrates. The formation of the biocorona has also been reported to reduce the catalytic activity of NPs in the case of proteins, polysacharides [9] and deoxyribonucleic acid (DNA) [20,21]. On the other hand, the presence of small organic molecules such as citrate did not affect nanocatalysis, while polyethylene glycol (PEG) showed a molecular weight-dependent behaviour, i.e., large PEG molecules (>400 Da) inhibited the catalytic activity of AuNPs, but smaller PEG chains did not [9]. Based on that, the modulation of the ferricyanide-dissociation catalytic activity of AuNPs by the formation of the biocorona was here used in the production of a cost-effective nanoplasmonic platform for general hygiene screening. AuNPs were immobilized by drop-casting on cellulose paper, acting as an eco-friendly and low-cost support material, able to pump the sample by capillarity without the need for external instruments [22–24]. When incorporating ferricyanide, it is dissociated by AuNPs releasing cyanide ions that dissolve the AuNPs, which lose their characteristic red colour. The presence of biomacromolecules reduces ferricyanide dissociation and NP dissolution in a concentration-dependent manner. This mechanism was evaluated in albumin and agar solutions, as examples of proteins and polysaccharides, and complex samples such as milk, eggs, soy sauce and yeast extract. 2. Materials and Methods 2.1. Chemical Reagents Tetrachloroauric acid, sodium citrate tribasic dihydrate, potassium ferrocyanide trihydrate, casein, fructose, sucrose, glycine and L-lysine were purchased from Sigma-Aldrich (St. Louis, MO, USA). Potassium ferricyanide, bovine serum albumin (BSA), sodium chloride and agar were purchased from ITW Reagents (Glenview, IL, USA). BSA-Alexa FluorTM 647 conjugate was purchased from Thermo Fisher Scientific (Waltham, MA, USA). All the chemicals were of analytical grade and were used as Nanomaterials 2020, 10, 2335 3 of 13 received without any purification steps. All the solutions were prepared in deionizedwater unless otherwise stated. 2.2. Synthesis and Processing of Gold Nanoparticles Gold nanoparticles (AuNPs) were produced by the Turkevich–Frens method [25], which is based on the single-phase aqueous reduction of tetrachloroauric acid by sodium citrate, allowing the size- and shape-controlled synthesis of citrate-stabilized AuNPs. A 2.2 mM sodium citrate solution (150 mL) was heated until boiling (15 min) in a thermal magnetic stirrer (600 rpm). Then, 1 mL of 25 mM tetrachloroauric acid was added to the solution. After 5 min of boiling, citrate-stabilized AuNPs of ~10 nm in diameter at a concentration of ~3 × 1012 NPs mL−1 were obtained and stored at 4 ◦ C until use. The experiments were carried out within a period of 15 days after synthesis since long-term storage could have compromised the stability of the NPs. All the experiments were carried out at room temperature using a concentration of ~2.3 × 1012 NPs mL−1 . The adsorption isotherm of BSA for citrate-stabilized AuNPs was determined with BSA-Alexa FluorTM 647 conjugate. AuNPs were exposed to different concentrations of the fluorescent conjugate for 10 min, followed by centrifugation (13,000 rpm/50 min) and protein quantification in the supernatant. When necessary, unbound proteins were removed by centrifugation (13,000 rpm/50 min) and washing the pellet 3 times with Milli-Q water. 2.3. UV-Vis Spectroscopy and Fluorimetry Optical measurements were performed in transparent 96-well plates by placing 200 µL of sample per well. A Varioskan Flash microplate reader (Thermo Fisher Scientific, Waltham, MA, USA) was used for spectrum acquisition within a range of 350 to 700 nm. The water baseline was always subtracted from the measured spectra. The fluorometric determination of BSA-Alexa FluorTM 647 was performed in black 96-well plates with excitation at 650 nm and emission at 668 nm. All the experiments were carried out in triplicate, and averaged values with the corresponding standard deviations were calculated. 2.4. Ferricyanide Dissociation Activity The catalytic ferricyanide dissociation by AuNPs and subsequent gold dissolution were monitored by UV-vis spectroscopy using a Varioskan Flash microplate reader (Waltham, MA, USA) since the reaction produces a decrease in the plasmonic band at 522 nm until disappearance when cyanidation is completed. 2.5. Transmission Electron Microscopy (TEM), Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray Spectroscopy (EDX) NPs were imaged with a JEM-2011 transmission electron microscope (ELECMI, Zaragoza, Spain))operating at an accelerating voltage of 200 kV, from the Servei de Microscòpia at Universitat Autònoma de Barcelona (UAB, Bellaterra, Spain). The samples were prepared by drop-casting 5 µL of sample onto carbon-coated copper TEM grids and dried at room temperature. Visualization was performed on various parts of the grids to obtain a global picture of the sample. Size distribution analysis was performed by measuring at least 150 NPs using the free software ImageJ. The plasmonic paper was imaged with a MERLIN FE-SEM operating at an accelerating voltage of 1 kV. EDX analysis was performed with a coupled Oxford LINCA X-Max detector (Oxford Instruments, Abingdon, UK). 2.6. Dynamic Light Scattering and Zeta Potential A Malvern Zetasizer Nano ZS instrument was employed for dynamic light scattering (DLS) (Malvern Panalytical, Malvern, UK) and zeta potential measurements, with a light source wavelength of 532 nm and a scattering angle of 173◦ for the detection of DLS. Non-adsorbed BSA was removed since free proteins could interfere with the proper characterization of the NPs. All the experiments were carried out in triplicate, and averaged values with the corresponding standard deviations were calculated. Nanomaterials 2020, 10, 2335 4 of 13 Nanomaterials 2020, 10, x FOR PEER REVIEW 4 of 13 2.7. were carriedPaper Plasmonic out inFabrication triplicate, and averaged values with the corresponding standard deviations were calculated. Citrate-stabilized AuNPs were embedded within cellulose matrices (absorbent paper discs, 0.7 mm in thickness) byPaper 2.7. Plasmonic Fabricationa volume of 100 µL of AuNP suspension and drying at 60 ◦ C until drop-casting complete dehydration (30 Citrate-stabilized min). AuNPs After were four inoculation/drying embedded cycles, within cellulose matrices paper paper (absorbent matrices containing discs, 0.7 × 10 ~9.2mm 11 NPs were obtained, presenting a deep purple colour owing to the plasmonic nature of in thickness) by drop-casting a volume of 100 µL of AuNP suspension and drying at 60 °C until AuNPs. The dehydration complete paper strips(30 were then min). stored After four at 4 ◦ C until use. For inoculation/drying sample cycles, paperanalysis, matricesvolumes of~9.2 containing 100 µL of sample × 10 solutions 11 NPs werewere dropped obtained, onto the presenting papers. a deep purple colour owing to the plasmonic nature of AuNPs. The paper strips were then stored at 4 °C until use. For sample analysis, volumes of 100 µL of sample 2.8. solutions Colour Analysis were dropped onto the papers. For colour analysis, images were acquired with a phone camera (13 megapixels) and analyzed using 2.8. Colour Analysis the free software ImageJ (National Institutes of Health (NIH), Bethesda, MD, USA), as reported [26]. For colour analysis, images were acquired with a phone camera (13 megapixels) and analyzed Colour images (RGB) were split into the three primary colour channels (red, green and blue). The green using the free software ImageJ (National Institutes of Health (NIH), Bethesda, MD, USA), as reported channel image, corresponding to the complementary colour of red/purple, was selected for the analysis [26]. Colour images (RGB) were split into the three primary colour channels (red, green and blue). of the plasmonic colour. The images were converted into grey-scale, and the grey magnitude was The green channel image, corresponding to the complementary colour of red/purple, was selected found for to thebe inversely analysis proportional of the to theThe plasmonic colour. purple images colour were intensity. The grey-scale, converted into average grey magnitude and the grey of twomagnitude perpendicular lines to was found crossing the center be inversely of the to proportional paper discs was the purple obtained, colour intensity.and The the corresponding average grey standard deviation magnitude of twowas calculated.lines perpendicular Allcrossing the experiments the center were of thecarried out in paper discs wastriplicate, obtained, and and averaged the corresponding values standard deviation with the corresponding was calculated. standard deviationsAllwere the experiments calculated. were carried out in triplicate, and averaged values with the corresponding standard deviations were calculated. 3. Results and Discussion 3. Results and Discussion The impact exerted by biomolecular coronas on catalytic ferricyanide dissociation by AuNPs The impactusing was investigated exertedcitrate-stabilized by biomolecular coronas AuNPson ofcatalytic 10 nm ferricyanide in diameterdissociation (illustrated byin AuNPs Figure 1a). was investigated using citrate-stabilized AuNPs of 10 nm in diameter (illustrated in Figure 1a). Citrate prevented AuNP aggregation by electrostatic forces, although it could be easily replaced Citrate prevented AuNP aggregation by electrostatic forces, although it could be easily replaced by by other ligands with stronger affinities, e.g., biomolecules, in the biocorona formation process. other ligands with stronger affinities, e.g., biomolecules, in the biocorona formation process. Concerning ConcerningNPNP size, both size, thethe both catalytic catalyticactivity activityand andthe thecapacity capacity for biocoronaformation for biocorona formationwereweretaken taken into account. into account. The catalytic activity is inversely proportional to the NP diameter, and hence, smaller NPs The catalytic activity is inversely proportional to the NP diameter, and hence, smaller present NPs higher presentcatalytic activities higher catalytic [27]. However, activities small-sized [27]. However, NPs may small-sized NPs present dimensions may present similar to, dimensions or even below, similar to, orthose of some even below, common those of somebiomacromolecules such as proteins, common biomacromolecules being unable such as proteins, to develop a being unable full to develop a[28]. biocorona full biocorona Regarding [28]. theRegarding the latter, latter, a size of 10 anm size of 10 was nm wasfor selected selected for this study. this study. Figure 1. Protein corona formation. (a) Illustration of the bovine serum albumin (BSA) corona and BSA molecules approximated by an equilateral triangular prism (not drawn to scale). (b) Adsorption isotherm of BSA onto gold nanoparticles (AuNPs) (left y-axis), and estimated percentage of AuNP surface coverage as a function of the BSA concentration (right y-axis) (error bars represent the standard deviation, n = 3). Nanomaterials 2020, 10, 2335 5 of 13 3.1. Biocorona Formation and Characterization Bovine serum albumin (BSA), with a size of 66.5 kDa, was used as a model molecule for biocorona formation. Citrate-stabilized AuNPs (~2.3 × 1012 NPs mL−1 ) were exposed to fluorescently labelled BSA (BSA-Alexa Fluor) at concentrations ranging from 0.1 to 10 µg mL−1 for 10 min, which was enough to reach adsorption equilibrium (Figure S1). The amount of adsorbed proteins was determined by fluorescence analysis after the interpolation of the fluorescent values with the calibration curve obtained by measuring BSA solutions at different concentrations (Figure S2). The adsorption isotherm was then obtained by plotting the amount of adsorbed BSA against the total BSA concentration in the solution (Figure 1b, left y-axis). As shown, the amount of adsorbed BSA increased proportionally with the BSA concentration in the solution up to 1 µg mL−1 , and complete saturation was reached at 10 µg mL−1 . The experimental data were complemented with a theoretical analysis considering the native conformation of the BSA molecules as equilateral triangular prisms with 8 nm sides and 3.5 nm heights, with triangular cross-sections of 32 nm2 (Figure 1a), based on previous publications [29]. The theoretical number of adsorbed proteins to complete a monolayer was estimated by dividing the AuNP surface by the cross-sectional area of BSA, corresponding to ~10 BSA molecules per NP. Extrapolating to the experimental data, where solutions containing 2.3 × 1012 NPs mL−1 were used, a concentration of adsorbed proteins of ~2.5 µg mL−1 should be enough to produce the complete coverage of the NPs (Figure 1b, right y-axis). According to the above experimental results, significant coverage of 70% was observed after incubation with 2.5 µg mL−1 , although it required a higher concentration of 10 µg mL−1 to reach 98% coverage. The coverage values were confirmed through hydrodynamic diameter analysis and colloidal stability evaluation (Figure S3). Thus, it may be concluded that, under these experimental conditions, a full BSA corona can be obtained after 10 min of incubation with 10 µg mL−1 samples. 3.2. Impact of the Biocorona on Ferricyanide Dissociation by AuNPs AuNPs present the capacity to catalytically dissociate ferricyanide molecules [8], which was employed to detect biomacromolecules through the mechanism illustrated below: [Fe(CN)6 ]3− = [Fe(CN)6−n ](3−n)− + nCN− (1) 4Au + 8CN− + O2 + 2H2 O = 4[Au(CN)2 ]− + 4OH− (2) First, as shown in Equation (1) (ferricyanide dissociation), the AuNPs dissociated ferricyanide, releasing cyanide ions in the vicinity of the NPs. Then, the free cyanide ions led to the cyanidation of AuNPs as shown in Equation (2) (Elsner equation), producing dicyanoaurate ions as reported in [30–32]. The final consequence of the catalytic dissociation of ferricyanide was, therefore, the dissolution and disappearance of AuNPs from the medium, as illustrated in Figure 2. Figure 2a shows the decrease in the AuNPs’ size, from 10 ± 2 to 5 ± 2 nm, after 24 h of incubation in 1 mM ferricyanide (a ferricyanide/gold molar ratio of 8.3) with an NP volume reduction of 80%. It is worth noting that the average final size of the AuNPs was probably overestimated since NPs below 1.5 nm could not be observed and the completely dissolved ones were not considered in this calculation. The NP dissolution entailed a progressive decrease in the plasmonic band of the AuNPs at a wavelength of 522 nm, which completely vanished after 24 h of incubation (Figure 2b). Since no shifts were observed in this band, aggregation processes were disregarded. A second absorbance band at 420 nm was obtained, which corresponded to the yellow-coloured ferricyanide. This band also disappeared over time, although not completely with the excess of ferricyanide in the mixture. The full process could be visually monitored (Figure 2b, inset), where the initial orange solution, resulting from the combination of the red AuNPs with the yellow ferricyanide, progressively became yellow with NP dissolution. Nanomaterials 2020, 10, x FOR PEER REVIEW 6 of 13 combination of the red AuNPs with the yellow ferricyanide, progressively became yellow with NP Nanomaterials 2020, 10, 2335 6 of 13 dissolution. Figure2.2.Ferricyanide Figure Ferricyanidedissociation dissociationby byAuNPs. AuNPs.(a)(a)Scheme Schemerepresenting representingferricyanide ferricyanidedissociation dissociationbyby AuNPs, subsequent gold cyanidation and representative TEM images of AuNPs AuNPs, subsequent gold cyanidation and representative TEM images of AuNPs before before and after and reaction after with 1 mMwith reaction ferricyanide. Inset, size distribution 1 mM ferricyanide. histograms (the Inset, size distribution average (the histograms size was obtained average by counting size was obtained atby least 150 NPs). counting (b) Time at least evolution 150 NPs). of the (b) Time UV-vis spectra evolution of AuNPs of the UV-vis withofferricyanide spectra AuNPs with(dashed line ferricyanide corresponds to untreated NPs). Inset, images showing the colour evolution of the AuNP (dashed line corresponds to untreated NPs). Inset, images showing the colour evolution of the AuNPsuspension. suspension. By contrast, the AuNPs previously coated with a full BSA protein corona (10 min in 10 µg mL−1 of BSA)Bydid not catalyse contrast, the AuNPsferricyanide previously dissociation coated with andaremained stable, exhibiting full BSA protein corona (10 plasmonic min in 10 µg bands mL−1 similar to those of uncoated AuNPs (Figure 3a) and to BSA-coated AuNPs of BSA) did not catalyse ferricyanide dissociation and remained stable, exhibiting plasmonic bandsafter centrifugation and resuspension similar to thosein protein-free of uncoatedmedia AuNPs with ferricyanide (Figure 3a) and (Figure S4). When to BSA-coated the protein AuNPs corona was not after centrifugation and completely formed, the modulation of the catalytic activity was observed, resuspension in protein-free media with ferricyanide (Figure S4). When the protein corona which is illustrated in was not Figure 3b. BSA concentrations below 0.25 −1 completely formed, the modulation of µgthemL did not catalytic modify activity wasthe catalytic activity observed, which isofillustrated the AuNPs,in and the plasmonic peak disappeared after incubation. In the range Figure 3b. BSA concentrations below 0.25 µg mL did not modify the catalytic activity of the −1 between 0.25 and 5 mL−1 , µgAuNPs, aand reduction in the catalytic the plasmonic activity proportional peak disappeared to the BSA after incubation. In concentration the range betweenwas found, until 0.25 and complete 5 µg mL−1, a suppression above 10 µg mL −1 . This may have derived from a reduction in the NPs’ surface available reduction in the catalytic activity proportional to the BSA concentration was found, until complete for catalysis, as suppression already above 10 µgreported mL−1. This formay PEGhave molecules derived[19]. from Considering a reduction inprevious the NPs’theoretical data, surface available afor minimum ofas20% coverage (corresponding to the 0.25 −1 mL concentration) catalysis, already reported for PEG molecules µgConsidering [19]. previous was necessary theoretical data,toa produce a detectable reduction in catalytic activity, which progressively minimum of 20% coverage (corresponding to the 0.25 µg mL concentration) was necessary to −1 and linearly decreased until complete produce coverage a detectableat 10 µg mL−1in reduction (full protein catalytic coronawhich activity, formation). Similar results progressively were obtained and linearly decreased when until agar was used as a model polysaccharide molecule instead of proteins (Figure complete coverage at 10 µg mL (full protein corona formation). Similar results were obtained when −1 3c). Thus, polysaccharide coronas, agar was as protein used asorapreviously reported DNAmolecule model polysaccharide ones, modulated instead the of catalytic proteins activity (Figure of3c). AuNPs. Thus, This provided a simple, fast and sensitive way to detect general biological contamination polysaccharide coronas, as protein or previously reported DNA ones, modulated the catalytic activity at trace levels. According of AuNPs. Thistoprovided the above results,fast a simple, theandbiomacromolecules’ sensitive way to detectadsorption generalonto the AuNPs’ biological surface contamination modulates their at trace levels. catalytic activity towards ferricyanide dissociation, and this mechanism may be useful for cleanliness evaluation. Owing to its unpecific nature, this approach may serve as a broad-range alert system for hygiene monitoring in sensitive environments. It is worth noting, however, that the nanocatalytic response may be influenced by both the size and the binding affinity of biomolecules. Therefore, it is difficult to predict the outcome in the case of concrete biomolecules. In this respect, the interference of some typical small-sized biomolecules including sugars (fructose and sucrose) and of aminoacids (glycine and L-lysine) was studied (Figure 4). The catalytic activity of the AuNPs was not affected by the small molecules at the tested concentrations, whereas casein (a typical milk protein (~22 kDa) used as a positive control) induced a decrease at concentrations similar to those observed for BSA. Nanomaterials 2020, 10, 2335 7 of 13 Nanomaterials 2020, 10, x FOR PEER REVIEW 7 of 13 Figure 3. Modulation of ferricyanide dissociation by biocoronas. (a) Time evolution of the Figure 3. Modulation of ferricyanide dissociation by biocoronas. (a) Time evolution of the UV-vis UV-vis spectra of BSA-coated AuNPs with ferricyanide (dashed line corresponds to untreated NPs). spectra of BSA-coated AuNPs with ferricyanide (dashed line corresponds to untreated NPs). Inset, Inset, images showing the colour evolution of the AuNP suspension. (b) Absorbance variation at 522 nm images showing the colour evolution of the AuNP suspension. (b) Absorbance variation at 522 nm of of AuNPs (with respect to untreated NPs) exposed to BSA concentrations ranging from 0.01 to 100 µg AuNPs (with respect to untreated NPs) exposed to BSA concentrations ranging from 0.01 to 100 µg mL−1 after reaction with ferricyanide. The concentration ranges corresponding to low, medium and mL−1 afterinhibition maximum reaction ofwith the ferricyanide. Theare catalytic activity concentration ranges indicated by (i), corresponding (ii) and to low, (iii), respectively. (c) medium and Absorbance variation at 522 nm of AuNPs (with respect to untreated NPs) exposed to agar concentrations ranging(c) maximum inhibition of the catalytic activity are indicated by (i), (ii) and (iii), respectively. Absorbance from 10 to 5000 variation µg mL−1atafter522reaction nm of withAuNPs (with respect ferricyanide. to untreated ranges The concentration NPs) exposed to agar corresponding concentrations ranging from 10 to 5000 µg mL −1 after reaction with ferricyanide. The concentration to low, medium and maximum inhibition of the catalytic activity are indicated by (i), (ii) and (iii), ranges corresponding respectively. (error bars to low, medium represent and maximum the standard deviation,inhibition n = 3). of the catalytic activity are indicated by (i), (ii) and (iii), respectively. (error bars represent the standard deviation, n = 3). 3.3. Nanoplasmonic Paper Platform for General Screening of Biomacromolecules According to the above results, the biomacromolecules’ adsorption onto the AuNPs’ surface The AuNP catalysis modulation by biocoronas enabled a sensitive determination of the modulates their catalytic activity towards ferricyanide dissociation, and this mechanism may be biomacromolecules in solution. To simplify the assay and the management of residues, AuNPs were useful for cleanliness evaluation. Owing to its unpecific nature, this approach may serve as a broad- deposited on a cellulose matrix (absorbent paper discs) by drop-casting to produce deeply coloured range alert system for hygiene monitoring in sensitive environments. It is worth noting, however, nanoplasmonic platforms with the advantages of being low-cost and presenting capillarity, thus that the nanocatalytic response may be influenced by both the size and the binding affinity of avoiding the use of external pumps for sample/reagent introduction. The optimal amount of NPs in the biomolecules. Therefore, it is difficult to predict the outcome in the case of concrete biomolecules. In paper was defined by taking into account two main factors: (i) allowing visual analysis by obtaining this respect, the interference of some typical small-sized biomolecules including sugars (fructose and deeply coloured papers and (ii) avoiding NP aggregation, since aggregated AuNPs lose their plamonic sucrose) and of aminoacids (glycine and L-lysine) was studied (Figure 4). The catalytic activity of the colour and may lose their catalytic activity as well. Considering this, the use of four inoculation cycles, AuNPs was not affected by the small molecules at the tested concentrations, whereas casein (a typical obtaining papers containing ~9.2 × 1011 NPs, was found to be optimal for providing vivid colours milk protein (~22 kDa) used as a positive control) induced a decrease at concentrations similar to without NP aggregation. At higher concentrations of NPs, the papers presented dark colours due to those observed for BSA. aggregation (data not shown). Nanomaterials Nanomaterials2020, 2020,10, 10,x2335 FOR PEER REVIEW 8 8ofof13 13 Figure 4. Modulation of the ferricyanide dissociation by small-sized biomolecules. Absorbance Figure 4. Modulation of the ferricyanide dissociation by small-sized biomolecules. Absorbance variation at 522 nm of AuNPs (with respect to untreated NPs) exposed to casein, fructose, sucrose, variation at 522 nm of AuNPs (with respect to untreated NPs) exposed to casein, fructose, sucrose, glycine and L-lysine after reaction with ferricyanide (error bars represent the standard deviation, n = 3). glycine and L-lysine after reaction with ferricyanide (error bars represent the standard deviation, n = 3). BSA was used as a model molecule to validate the platform’s performance. The ferricyanide concentration was increased to 50 mM to avoid diffusion limitations in the cellulose matrix and to 3.3. Nanoplasmonic Paper Platform for General Screening of Biomacromolecules ensure fast kinetics for the reaction within 20 min. The nanoplasmonic platform responded to protein The AuNPsimilarly concentration catalysisto modulation by biocoronas in solution (Figure 5a). Thatenabled a sensitive is, the paper determination completely discoloured of after the biomacromolecules ferricyanide incubation in solution. in samples To simplify containing the below assay and 0.25the mL−1 of BSA of µg management due residues, to AuNP AuNPs were dissolution. deposited The colouronintensity a cellulose matrix (absorbent progressively increased paper discs) in the by between range drop-casting 0.25 toandproduce 5 µg mL −1 , while deeply coloured 10 µg −1 nanoplasmonic mL samples almost platforms with the maintained theadvantages initial colour of of being low-cost the paper. SEMand andpresenting EDX studies capillarity, confirmedthus that avoiding the colour in these papers after ferricyanide incubation was due to the presence of AuNPs (Figure in the use of external pumps for sample/reagent introduction. The optimal amount of NPs 5b, the paperS5was Figures and defined S6), which by were takingnot into account present twouncoloured in the main factors:paper (i) samples, allowingi.e.,visual those analysis incubatedby obtaining in 0.1 µg mL −1 BSA. deeply coloured papers The visual and (ii) avoiding observations NP aggregation, were corroborated sinceanalysis by colour aggregated using AuNPs lose the ImageJ their plamonic colour and may lose their catalytic activity as well. Considering software. After splitting the images in the RGB channels, the green one was selected for being the most this, the use of four inoculation sensitive tocycles, obtaining the purple colour.papers containing The degree ~9.2 × 1011was of colour-shift NPs,calculated was found to be with theoptimal followingfor providing expression: vivid colours without NP aggregation. At higher concentrations of NPs, the papers presented dark colours due to aggregationColour (data not change shown).(%) = (GT − GP ) × 100/(GP − GN ) (3) BSA was used as a model molecule to validate the platform’s performance. The ferricyanide where GT is the concentration grey was magnitude increased of the to 50 mMtestto nanoplasmonic avoid diffusionpaper, GP corresponds limitations to the positive in the cellulose matrix and control to paper (blank nanoplasmonic paper without biomolecules, 0% change) and ensure fast kinetics for the reaction within 20 min. The nanoplasmonic platform responded to protein G N to the negative control paper (pristinesimilarly concentration paper without NPs, 100% to in solution change). (Figure 5a). That is, the paper completely discoloured after The changes ferricyanide reported incubation completely in samples coincidedbelow containing with the 0.25visual µg mL interpretation, −1 of BSA due validating to AuNPthe possibility dissolution. of the visual analysis of the samples without the need for external instrumentation The colour intensity progressively increased in the range between 0.25 and 5 µg mL , while 10 (Figure −1 5c). In any µg case, the response range (0.25–5 µg mL −1 ) and limit of detection (0.25 µg mL−1 , empirically determined) mL samples almost maintained the initial colour of the paper. SEM and EDX studies confirmed that −1 of the the nanoplasmonic colour in these papers platform were in agreement after ferricyanide incubation withwas other due colorimetric to the presence protein tests based of AuNPs (Figureson non-functionalized gold nanoparticles [12,33] or chemical reagents, e.g., 5b, S5 and S6), which were not present in the uncoloured paper samples, i.e., those incubated in 0.1the bicinchoninic acid (BCA) assay, µg mLthe Bradford −1 BSA. The assay visualorobservations the Lowry assay were (detection corroborated = 10–20 limits by colour mL−1 ) [34], µg analysis which using thevalidated ImageJ software. After splitting the images in the RGB channels, the green one was selected for being also the nanoplasmonic platform for protein detection. On the other hand, good sensitivities have the been sensitive most reported for other to the etching-based purple colour. The methods degreeapplied to the detection of colour-shift of diverse was calculated analytes with [35–37]. the following expression: Colour change (%) = (GT − GP) × 100/(GP − GN) (3) where GT is the grey magnitude of the test nanoplasmonic paper, GP corresponds to the positive control paper (blank nanoplasmonic paper without biomolecules, 0% change) and GN to the negative control paper (pristine paper without NPs, 100% change). Nanomaterials 2020, 10, 2335 9 of 13 Nanomaterials 2020, 10, x FOR PEER REVIEW 9 of 13 Figure 5. Nanoplasmonic Figure 5. Nanoplasmonicplatform platform for for biomacromolecule biomacromolecule screening.screening. (a) Colour (a) Colour response response of theof the nanoplasmonic nanoplasmonic paper to biocorona paper formation. to biocorona formation.(b) (b)EDX EDXanalysis analysisofofpapers papersexposed exposed to to ferricyanide ferricyanide with BSAwith concentrations ranging from 0from to 10 µg10mL −1 . −1 (c) Graph BSA concentrations ranging 0 to µg mL . (c) Graphandandimages showingthe images showing the colour colour shift of shift nanoplasmonic of nanoplasmonicpapers exposed papers to ferricyanide exposed to ferricyanidewith with BSA concentrations BSA concentrations ranging ranging fromfrom 0.1 to0.1 25 to µg25 µg mL−1 mL −1 for 20 min. (d) Graph and images showing the colour shift of nanoplasmonic papers exposed to for 20 min. (d) Graph and images showing the colour shift of nanoplasmonic papers exposed ferricyanide with to ferricyanide with milk milk concentrations concentrations ranging rangingfrom from 0.01 to 25 0.01 µg µg to 25 mLmL −1 20 −1 for formin. Colour 20 min. shift isshift is Colour defined as the degree of change (expressed as percentage) with respect to the defined as the degree of change (expressed as percentage) with respect to the positive control positive control (blank(blank nanoplasmonic paper, 0%) and negative control (pristine paper, 100%) (error bars represent the nanoplasmonic paper, 0%) and negative control (pristine paper, 100%) (error bars represent the standard standard deviation, n = 3). deviation, n = 3). The changes reported completely coincided with the visual interpretation, validating the As a step forward, the platform was applied to detect a complex sample. To this end, milk was possibility of the visual analysis of the samples without the need for external instrumentation (Figure selected for containing biomolecules of different natures, e.g., proteins, sugars and lipids, and 5c). In any case, the response range (0.25–5 µg mL−1) and limit of detection (0.25 µg mL−1, empirically being a well-known determined) of theallergenic agent nanoplasmonic causing platform severe were health problems in agreement with other[38]. As shown colorimetric in Figure protein tests 5d, the nanoplasmonic platform presented based on non-functionalized good sensitivity gold nanoparticles [12,33]for whole milk or chemical detection, reagents, with e.g., the a detection limit bicinchoninic of 0.025 −1 (empirically determined), and a large dynamic linear range of two mLassay, decades, acid µg (BCA) the Bradford assay or the Lowry assay (detection limits = 10–20 µg mL−1) [34], which up to 2.5 µg −1 validated mL . The the nanoplasmonic platform forinprotein fate of the nanoplatform detection. the presence or On the other absence hand, good sensitivities of biomolecules is illustrated in have Figure 6. also These been reported results for other validated theetching-based methods applied use of this technology in theto the detection of determination ofdiverse analytes from contamination [35–37]. complex matrices such as milk. Additionally, the nanoplasmonic paper’s response to different food As a step forward, the platform was applied to detect a complex sample. To this end, milk was samples of different natures was studied to add data supporting the previous conclusion. In this case, selected for containing biomolecules of different natures, e.g., proteins, sugars and lipids, and being chicken eggs, soy sauce and yeast extract (all dissolved in water) were tested with the nanoplatform as a well-known allergenic agent causing severe health problems [38]. As shown in Figure 5d, the before. As depictedplatform nanoplasmonic in the graphs (Figure presented good7), all the three sensitivity samples for whole milkformed biocoronas, detection, which prevented with a detection limit AuNP dissolution at high concentrations (10 −4 % (v/v), 10−4 % (v/v) and 10−3 % (w/v) for the eggs, of 0.025 µg mL (empirically determined), and a large dynamic linear range of two decades, up to 2.5 −1 soy sauce µg mLand yeast −1. The fate extract, respectively). of the nanoplatform At lower in the presence concentrations, however, catalytic or absence of biomolecules ferricyanide is illustrated in dissociation Figure 6.and subsequent These gold dissolution results validated weretechnology the use of this not avoided, and in the thus, colourof determination change was detected. contamination Thesefrom complex matrices observations suchthe confirmed as milk. Additionally, suitability the nanoplasmonic of the proposed paper’sapproach nanoplasmonic response for to different the detection food samples of complex food of different samples natures and was studied its potential to add data for general supporting hygiene the previous conclusion. In this screening. case, chicken eggs, soy sauce and yeast extract (all dissolved in water) were tested with the nanoplatform as before. As depicted in the graphs (Figure 7), all the three samples formed biocoronas, which prevented AuNP dissolution at high concentrations (10−4% (v/v), 10−4% (v/v) and 10−3% (w/v) ferricyanide dissociation and subsequent gold dissolution were not avoided, and thus, colour change Nanomaterials 2020, 10, x FOR PEER REVIEW 10 of 13 was detected. These observations confirmed the suitability of the proposed nanoplasmonic approach for thefordetection the eggs, ofsoycomplex sauce and food samples yeast extract,and its potential respectively). for general At lower hygiene however, concentrations, screening. catalytic ferricyanide dissociation and subsequent gold dissolution were not avoided, and thus, colour change was detected. Nanomaterials 2020, 10,These 2335 observations confirmed the suitability of the proposed nanoplasmonic approach 10 of 13 for the detection of complex food samples and its potential for general hygiene screening. Figure 6. Illustration showing the colour evolution of the nanoplamonic paper in the presence or absence of biomacromolecules forming a biocorona. Figure 6. Illustration Figure showing 6. Illustration showingthe thecolour colour evolution ofthe evolution of thenanoplamonic nanoplamonic paper paper in presence in the the presence or or absence absence of biomacromolecules of biomacromolecules formingaabiocorona. forming biocorona. Figure 7. Colour Colour responseofofthe response thenanoplasmonic nanoplasmonic paper paperto different complex samples. ColourColour shift is shift Figure Figure 7.7.Colour response of the nanoplasmonic paper toto different different complex complex samples. samples. Colour shift is defined is defined as the degree as degree the degree of change of change (expressed as (expressed percentage) with as percentage) respect to the with respect positive control (blank to the positive control defined as the of change (expressed as percentage) with respect to the positive control (blank nanoplasmonic (blank nanoplasmonic paper, 0%)0%) paper, andand negative control negative (pristine control paper, (pristine 100%) paper, (error(error 100%) bars bars represent the the represent nanoplasmonic paper, 0%) standard deviation, n = 3).and negative control (pristine paper, 100%) (error bars represent the standard deviation, n = 3). standard deviation, n = 3). 4. Conclusions 4. Conclusions 4. Conclusions We have developed a low-cost nanoplasmonic platform for sensitive and broad-range hygiene We have developed a low-cost nanoplasmonic platform for sensitive and broad-range hygiene screening We have based on the developed modulation of the catalytic activity for of AuNPsand towards ferricyanide screening based on thea low-cost modulation nanoplasmonic of the catalytic platform activity sensitive of AuNPs broad-range towards hygiene ferricyanide dissociation screening based via on biocoronas. the Our modulation results of indicate the that catalyticthe surface activity coverage of AuNPs of AuNPs towards with proteins ferricyanide dissociation via biocoronas. and polysaccharides, as Our usedOur model results indicate that biomacromolecules, the surface modulates coverage ferricyanide of AuNPs dissociation (and with dissociation proteins via biocoronas. and polysaccharides, usedresults indicate as model that the biomacromolecules,surface coverage of AuNPs with proteins subsequent gold dissolution) until complete suppression when modulates a full corona ferricyanide is formed. dissociation In the and (andpolysaccharides, subsequent used as model biomacromolecules, modulateswhen ferricyanide dissociation (and platform, wheregold AuNPs dissolution) are embedded untilin acomplete suppression cellulose matrix, AuNP dissolution a full due corona is formed. to the catalytic subsequent In the gold platform, dissolution) where AuNPs until complete arechanges. embedded suppression in a cellulose when a full corona is formed.due to the In activity results in intense colour The formation of thematrix, AuNP biocorona dissolution modulates this catalytic the platform, where activity, AuNPs resulting in are embedded samples with in differenta cellulose colour matrix, intensities AuNP dissolution proportional to catalytic activity results in intense colour changes. The formation of the biocorona modulates thisthe due initial to the catalytic biomolecule activity catalyticresults in intense concentrations, activity, resultingcolour even appreciable changes. with the in samples The bare with formation of the eye. The platform, different colour biocorona initially intensities modulates validated proportional this with albumin, to thecatalytic isinitial activity, resulting capable of in samples providing a with rapid different (20 min) colour intensities concentration-dependent proportional response biomolecule concentrations, even appreciable with the bare eye. The platform, initially validated totothe initial complex biomolecule matrices concentrations, with albumin, iseven appreciable capable with the of providing bare(20 a rapid eye. Theconcentration-dependent min) platform, initially validated with albumin, response to complex is capable of providing a rapid (20 min) concentration-dependent response matrices containing proteins, sugars and lipids, such as milk, eggs, soy sauce and yeast extract, to complex matrices with analytical performance comparable to that of standard methods. Considering its unspecific nature, this approach may serve as an alert system for hygiene monitoring in sensitive environments. Hence, broad-range cleanliness evaluation in a user-friendly and inexpensive fashion may be facilitated by this nano-enabled technology. Nanomaterials 2020, 10, 2335 11 of 13 Supplementary Materials: The following are available online at http://www.mdpi.com/2079-4991/10/12/2335/s1. Figure S1: Fluorescence kinetics of samples containing AuNPs and BSA-Alexa Fluor® , Figure S2: Calibration curve for BSA-Alexa Fluor® , Figure S3: Characterization of protein-coated AuNPs, Figure S4: UV-vis spectra of protein-coated AuNPs centrifuged and resuspended in protein-free media after reaction with 1 mM ferricyanide, Figure S5: SEM images of the plasmonic paper, Figure S6: EDX spectra of plasmonic papers exposed to ferricyanide with BSA. Author Contributions: Conceptualization, F.P.-V., A.T.A.J., X.M.-B. and J.M.G.; methodology, F.P.-V.; validation, F.P.-V.; formal analysis, F.P.-V.; data curation, F.P.-V.; writing—original draft preparation, F.P.-V. and X.M.-B.; writing—review and editing, F.P.-V., A.T.A.J., X.M.-B. and J.M.G.; visualization, F.P.-V., A.T.A.J., X.M.-B. and J.M.G.; supervision, A.T.A.J., X.M.-B. and J.M.G.; project administration, J.M.; funding acquisition, A.T.A.J., X.M.-B. and J.M.G. All authors have read and agreed to the published version of the manuscript. Funding: The work described in this paper was financed by projects CTQ2014-54553-C3-2-R and RTC-2016-5766-2 from the Ministerio de Economía y Competitividad of the Spanish Government to J.M. The work was partially supported by FEDER funds. Conflicts of Interest: The authors declare no conflict of interest. References 1. Messina, G.; Ceriale, E.; Nante, N.; Manzi, P. Effectiveness of ATP bioluminescence to assess hospital cleaning: A review. Eur. J. Public Health 2014, 24. [CrossRef] 2. Sanna, T.; Dallolio, L.; Raggi, A.; Mazzetti, M.; Lorusso, G.; Zanni, A.; Farruggia, P.; Leoni, E. ATP bioluminescence assay for evaluating cleaning practices in operating theatres: Applicability and limitations. BMC Infect Dis. 2018, 18, 583. [CrossRef] [PubMed] 3. Morse, S.S.; Mazet, J.A.; Woolhouse, M.; Parrish, C.R.; Carroll, D.; Karesh, W.B.; Zambrana-Torrelio, C.; Lipkin, W.I.; Daszak, P. Prediction and prevention of the next pandemic zoonosis. Lancet 2012, 380, 1956–1965. [CrossRef] 4. Rudkin, J.K.; McLoughlin, R.M.; Preston, A.; Massey, R.C. Bacterial toxins: Offensive, defensive, or something else altogether? PLoS Pathog. 2017, 13. [CrossRef] [PubMed] 5. Moore, G.; Griffith, C.; Fielding, L. A Comparison of Traditional and Recently Developed Methods for Monitoring Surface Hygiene within the Food Industry: A Laboratory Study. Int. J. Environ. Health Res. 2002, 12, 317–329. [CrossRef] 6. Saha, K.; Agasti, S.S.; Kim, C.; Li, X.; Rotello, V.M. Gold nanoparticles in chemical and biological sensing. Chem. Rev. 2012, 112, 2739–2779. [CrossRef] 7. Comotti, M.; Pina, C.D.; Matarrese, R.; Rossi, M. The catalytic activity of “naked” gold particles. Angew. Chem. Int. Ed. 2004, 43, 5812–5815. [CrossRef] 8. Zhai, J.; Zhai, Y.; Dong, S. Direct dissolution of Au nanoparticles induced by potassium ferricyanide. Colloids Surf. A Physicochem. Eng. Asp. 2009, 335, 207–210. [CrossRef] 9. Lang, N.J.; Liu, B.; Liu, J. Characterization of glucose oxidation by gold nanoparticles using nanoceria. J. Colloid Interface Sci. 2014, 428, 78–83. [CrossRef] 10. Quesada-González, D.; Merkoçi, A. Nanoparticle-based lateral flow biosensors. Biosens. Bioelectron. 2015, 73, 47–63. [CrossRef] 11. Pamies, R.; Cifre, J.G.H.; Espín, V.F.; Collado-González, M.; Baños, F.G.D.; de la Torre, J.G. Aggregation behaviour of gold nanoparticles in saline aqueous media. J. Nanopart Res. 2014, 16, 2376. [CrossRef] 12. Ho, Y.T.; Poinard, B.; Yeo, E.L.L.; Kah, J.C.Y. An instantaneous colorimetric protein assay based on spontaneous formation of a protein corona on gold nanoparticles. Analyst 2015, 140, 1026–1036. [CrossRef] [PubMed] 13. Monopoli, M.P.; Åberg, C.; Salvati, A.; Dawson, K.A. Biomolecular coronas provide the biological identity of nanosized materials. Nat. Nanotechnol. 2012, 7, 779–786. [CrossRef] [PubMed] 14. Casals, E.; Pfaller, T.; Duschl, A.; Oostingh, G.J.; Puntes, V. Time evolution of the nanoparticle protein corona. ACS Nano 2010, 4, 3623–3632. [CrossRef] [PubMed] 15. García Vence, M.; Chantada-Vázquez, M.D.P.; Vázquez-Estévez, S.; Manuel Cameselle-Teijeiro, J.; Bravo, S.B.; Núñez, C. Potential clinical applications of the personalized, disease-specific protein corona on nanoparticles. Clin. Chim. Acta 2020, 501, 102–111. [CrossRef] [PubMed] 16. Cai, R.; Chen, C. The Crown and the Scepter: Roles of the Protein Corona in Nanomedicine. Adv. Mater. Weinh. 2019, 31, e1805740. [CrossRef] [PubMed] Nanomaterials 2020, 10, 2335 12 of 13 17. Sasidharan, A.; Riviere, J.E.; Monteiro-Riviere, N.A. Gold and silver nanoparticle interactions with human proteins: Impact and implications in biocorona formation. J. Mater. Chem. B 2015, 3, 2075–2082. [CrossRef] 18. Shemetov, A.A.; Nabiev, I.; Sukhanova, A. Molecular Interaction of Proteins and Peptides with Nanoparticles. ACS Nano 2012, 6, 4585–4602. [CrossRef] 19. Lundqvist, M.; Stigler, J.; Cedervall, T.; Berggård, T.; Flanagan, M.B.; Lynch, I.; Elia, G.; Dawson, K. The evolution of the protein corona around nanoparticles: A test study. ACS Nano 2011, 5, 7503–7509. [CrossRef] 20. Zheng, X.; Liu, Q.; Jing, C.; Li, Y.; Li, D.; Luo, W.; Wen, Y.; He, Y.; Huang, Q.; Long, Y.-T.; et al. Catalytic Gold Nanoparticles for Nanoplasmonic Detection of DNA Hybridization. Angew. Chem. Int. Ed. 2011, 50, 11994–11998. [CrossRef] 21. Zhou, P.; Jia, S.; Pan, D.; Wang, L.; Gao, J.; Lu, J.; Shi, J.; Tang, Z.; Liu, H. Reversible Regulation of Catalytic Activity of Gold Nanoparticles with DNA Nanomachines. Sci. Rep. 2015, 5, 1–7. [CrossRef] [PubMed] 22. Lu, Y.; Ozcan, S. Green nanomaterials: On track for a sustainable future. Nano Today 2015, 10, 417–420. [CrossRef] 23. Meredith, N.A.; Quinn, C.; Cate, D.M.; Reilly, T.H.; Volckens, J.; Henry, C.S. Paper-based analytical devices for environmental analysis. Analyst 2016, 141, 1874–1887. [CrossRef] 24. Sanahuja, D.; Giménez-Gómez, P.; Vigués, N.; Ackermann, T.N.; Guerrero-Navarro, A.E.; Pujol-Vila, F.; Sacristán, J.; Santamaria, N.; Sánchez-Contreras, M.; Díaz-González, M.; et al. Microbial trench-based optofluidic system for reagentless determination of phenolic compounds. Lab Chip 2015, 15, 1717–1726. [CrossRef] [PubMed] 25. Bastús, N.G.; Comenge, J.; Puntes, V. Kinetically controlled seeded growth synthesis of citrate-stabilized gold nanoparticles of up to 200 nm: Size focusing versus Ostwald ripening. Langmuir 2011, 27, 11098–11105. [CrossRef] 26. Pujol-Vila, F.; Vigués, N.; Guerrero-Navarro, A.; Jiménez, S.; Gómez, D.; Fernández, M.; Bori, J.; Vallès, B.; Riva, M.C.; Muñoz-Berbel, X.; et al. Paper-based chromatic toxicity bioassay by analysis of bacterial ferricyanide reduction. Anal. Chim. Acta 2016, 910, 60–67. [CrossRef] 27. Hvolbæk, B.; Janssens, T.V.W.; Clausen, B.S.; Falsig, H.; Christensen, C.H.; Nørskov, J.K. Catalytic activity of Au nanoparticles. Nano Today 2007, 2, 14–18. [CrossRef] 28. Piella, J.; Bastús, N.G.; Puntes, V. Size-Dependent Protein–Nanoparticle Interactions in Citrate-Stabilized Gold Nanoparticles: The Emergence of the Protein Corona. Bioconjugate Chem. 2017, 28, 88–97. [CrossRef] 29. He, X.M.; Carter, D.C. Atomic structure and chemistry of human serum albumin. Nature 1992, 358, 209–215. [CrossRef] 30. Gökelma, M.; Birich, A.; Stopic, S.; Friedrich, B. A Review on Alternative Gold Recovery Re-agents to Cyanide. MSCE 2016, 04, 8–17. [CrossRef] 31. La Brooy, S.R.; Linge, H.G.; Walker, G.S. Review of gold extraction from ores. Miner. Eng. 1994, 7, 1213–1241. [CrossRef] 32. Xie, F.; Dreisinger, D.B. Use of ferricyanide for gold and silver cyanidation. Trans. Nonferrous Met. Soc. China 2009, 19, 714–718. [CrossRef] 33. Lai, T.-S.; Chang, T.-C.; Wang, S.-C. Gold nanoparticle-based colorimetric methods to determine protein contents in artificial urine using membrane micro-concentrators and mobile phone camera. Sens. Actuators B Chem. 2017, 239, 9–16. [CrossRef] 34. Noble, J.E.; Knight, A.E.; Reason, A.J.; Di Matola, A.; Bailey, M.J.A. A Comparison of Protein Quantitation Assays for Biopharmaceutical Applications. Mol. Biotechnol. 2007, 37, 99–111. [CrossRef] [PubMed] 35. Nitinaivinij, K.; Parnklang, T.; Thammacharoen, C.; Ekgasit, S.; Wongravee, K. Colorimetric determination of hydrogen peroxide by morphological decomposition of silver nanoprisms coupled with chromaticity analysis. Anal. Methods 2014, 6, 9816–9824. [CrossRef] 36. Liang, J.; Yao, C.; Li, X.; Wu, Z.; Huang, C.; Fu, Q.; Lan, C.; Cao, D.; Tang, Y. Silver nanoprism etching-based plasmonic ELISA for the high sensitive detection of prostate-specific antigen. Biosens. Bioelectron. 2015, 69, 128–134. [CrossRef] Nanomaterials 2020, 10, 2335 13 of 13 37. Gu, Y.; Kong, S.; Diao, X.; Guo, Y.; Zhang, K.; He, H. Mechanistic study on the facet etching effect of silver nanoprisms in the presence of halide ions and their application in the colorimetric sensing of metformin hydrochloride. New J. Chem. 2016, 40, 7557–7563. [CrossRef] 38. Monaci, L.; Tregoat, V.; van Hengel, A.J.; Anklam, E. Milk allergens, their characteristics and their detection in food: A review. Eur. Food Res. Technol. 2006, 223, 149–179. [CrossRef] Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).