Developing a Silk Fibroin Composite Film to Scavenge and Probe H2O2 Associated with UV-Excitable Blue Fluorescence

A silk fibroin composite film that can simultaneously scavenge and probe H2O2 in situ was developed for possibly examining local concentrations of H2O2 for biomedical applications. A multi-functional composite film (GDES) that consists of graphene oxide (G), a photothermally responsive element that was blended with polydopamine (PDA, D)/horseradish peroxidase (HRP, E) (or DE complex), and then GDE microaggregates were coated with silk fibroin (SF, S), a tyrosine-containing protein. At 37 °C, the H2O2-scavenging ability of a GDES film in solution at approximately 7.5 × 10−3 μmol H2O2/mg film was the highest compared with those of S and GS films. The intensities of UV-excitable blue fluorescence of a GDES film linearly increased with increasing H2O2 concentrations from 4.0 μM to 80 μM at 37 °C. Interestingly, after a GDES film scavenged H2O2, the UV-excitable blue fluorescent film could be qualitatively monitored by eye, making the film an eye-probe H2O2 sensor. A GDES film enabled to heat H2O2-containing samples to 37 °C or higher by the absorption of near-IR irradiation at 808 nm. The good biocompatibility of a GDES film was examined according to the requirements of ISO-10993-5. Accordingly, a GDES film was developed herein to scavenge and eye-probe H2O2 in situ and so it has potential for biomedical applications.


Introduction
H 2 O 2 , a reactive oxygen species (ROS), participates in numerous physiological and pathological conditions, including stem cell proliferation and differentiation, anti-bacterial defense, wounds, cancer, and ageing [1,2]. The concentration of H 2 O 2 in vivo affects cellular responses. For example, at a low concentration (10 −2~1 0 −1 µM), H 2 O 2 promotes the proliferation of stem cells while at a medium to high concentration (10 0~1 0 2 µM), it arrests cell growth and consequently causes the apoptosis of cells, as in the ageing process [1,2]. Moreover, the production of H 2 O 2 is one of three important To prepare GDE microaggregates, 1 mg GO that was dispersed in 1.5 mL DI water was added to dark red DE suspensions and vigorously stirred, causing DE complexes to adhere to the GO surface, forming GDE microaggregates. The GDE microaggregates, containing suspensions, were centrifuged and washed several times to remove non-adhering DE complexes. The particle sizes of

Fabrication of GDES Films
Graphene oxide (GO) was synthesized using a modified Hummer's method [24]. Briefly, 1 g graphite flakes and 0.5 g NaNO 3 were mixed with concentrated (98%) sulfuric acid, and 3 g KMnO 4 was slowly added to the mixture at 0 • C. The mixture was vigorously stirred for 24 h at 35 • C. The reaction was quenched by adding DI water and H 2 O 2, and also to remove those oxidation agents and the GO containing suspension was combined with HCl. Finally, GO/solvent suspensions were washed, centrifuged several times to remove solvents, and GO in the bottom containing residual solvents was evaporated off at 50 • C to produce dry GO.
To synthesize PDA-HRP (or DE) complexes, a modified method of Dai et al. [16] was adopted to synthesize PDA-HRP (or DE) complexes. Briefly, 0.4 mg HRP and 0.4 mg DA (1:1 in wt.) were dissolved in PBS (Phosphate Buffered Saline), and 0.3% H 2 O 2 was added to the DA/HRP mixture to promote the polymerization of DA. The polymerization of PDA in the mixture rapidly changed the color of the mixture from none (or transparent) to dark red. The reaction was continued for 0.5-1 h to prevent over-polymerization of PDA. Since HRP is an enzyme with a large molecular weight (~44 KDa), the PDA was most likely located on the surface of HRP (Figure 1a).
To prepare GDE microaggregates, 1 mg GO that was dispersed in 1.5 mL DI water was added to dark red DE suspensions and vigorously stirred, causing DE complexes to adhere to the GO surface, forming GDE microaggregates. The GDE microaggregates, containing suspensions, were centrifuged and washed several times to remove non-adhering DE complexes. The particle sizes of the GDE aggregates were measured and checked using a laser particle size analyzer (Particulate Systems, Nano-Plus, Norcross, GA, USA). SF (MW~127 kDa) solution was prepared as described in early reports that were published by the authors' laboratory [25]. Briefly, B. mori cocoons were boiled with 0.02% Na 2 CO 3 to degum to obtain SF. SF was dissolved in 9.3 M LiBr solution, and then dialyzed in DI water to remove Li + to produce 5~10 wt% of SF solution for further applications. To prepare GDES films, 60 mg of the aforementioned SF solution was added to GDE micro-aggregate suspension and stirred to cover GDE microaggregates completely with SF, producing GDES microaggregates in suspension. Aliquots of the suspensions were extracted and evaporated at 45 • C to produce a dry GDES film with size of 12 mm (diameter) × 0.04 mm (thickness) and about 12.32 mg (Figure 1a) with the ratios of compositions of G:D:E:S of 1.62%:0.62%:0.39%:97.4%, respectively. The films were further treated by varying percentages of ethanol (e.g., 75%) for about 15 min to induce β-sheets of SF (e.g.,~45%) to stabilize the SF polymers in solutions [26].

Characterizations of Components and Films Using ATR-FTIR, SEM and TEM
Attenuated total reflectance Fourier transform infrared spectra (ATR-FTIR) of various samples (i.e., GO, PDA, HRP, SF, DE, GDE, and GDES) were examined using an ATR-FTIR spectrometer (IRAffinity-1, Shimadzu, Japan). The samples were scanned from 4000 cm −1 to 600 cm −1 at a resolution Sensors 2020, 20, 366 6 of 19 of 4 cm −1 and the spectra that were obtained from the IRsolution software were baseline-corrected and smoothed.
Squares of the GDES films with an area of 1.5 cm 2 were cut and their surface morphologies were examined via scanning electron microscopy (SEM) (JSM-7600F, JEOL, Tokyo, Japan) following procedures that was previously employed by the authors' group [26]. Briefly, the films were immersed in 80 µM H 2 O 2 , so-called H 2 O 2 -treated GDES, and in PBS, GDES film, respectively, for 30 min at 37 • C, dried and coated with platinum for SEM. The details of the procedure can be found elsewhere [25]. The morphology of GO was observed by transmission electron microscopy (TEM, JEM-2000EXII, JEOL, Tokyo, Japan) according to a procedure that was previously used by the authors' group [26,27]. The particle sizes of the GO and GDE microaggregates were determined using a zeta potential/nanoparticle analyzer (Nano-Plus Particulate System, Norcross, GA, USA). The images of the surfaces of GDE microaggregates were taken by a tapping mode with a Si cantilever (App Nano, ACSTA-50, Mount View, CA, USA) by an atomic force microscope (AFM) (Bruker, Dimension Icon, Billerica, MA, USA) to examine the roughness of the surfaces equipped with built-in software (Nanoscope IIIa, Digital Instrument, Santa Barbara, CA, USA) [26].
To determine the E and D contents of the GDE microaggregates, the Bradford protein assay [28] for HRP quantification and the Arnow assay [29] for determining the catechol contents of PDA were carried out on the supernatant after the GDE suspension had been centrifuged. For the Bradford protein assay, 1.6 mL of the Bradford reagent was added to 0.4 mL of supernatant and allowed to react for 10 min; then absorbance at 595 nm was measured using a UV/VIS spectrophotometer (Multiskan, Thermo Scientific, Waltham, MA USA). To perform the Arnow assay, 1 mL of supernatant was mixed with 1 mL of 0.5 M HCl, 1 mL of a mixed sodium nitrite/sodium molybdate solution, 1 mL of 1 M NaOH and 1 mL of DI water, and the absorbance at 510 nm was measured.

H 2 O 2 Scavenging Assay
An H 2 O 2 solution with a concentration of 64 µM was applied to the as-prepared GDES and GS films to measure the H 2 O 2 -scavenging abilities of the films at various temperatures of 15 • C, 25 • C, or 37 • C. The residual H 2 O 2 in the solution after reaction with the films was pipetted out for analysis by the 1,10-phenanthroline/FeCl 2 method [30,31]. In the absence of H 2 O 2 , 1,10-phenanthroline molecules form chelates with ferrous ions (Fe 2+ ), and these chelates exhibit a significant absorption peak at 510 nm. In contrast, in the presence of H 2 O 2 , the Fenton reaction takes place, transforming the ferrous ions (Fe 2+ ) of FeCl 2 to ferric ions (Fe 3+ ). Accordingly, the chelate of 1,10-phenanthroline molecules/Fe 2+ does not form and no absorption peak at 510 nm is detected. The degrees of absorption at 510 nm by the pipetted supernatant solutions on the SF, EDG and EDGS films were determined using a UV/VIS spectrophotometer. The H 2 O 2 -scavenging abilities of the films were determined by subtracting the absorption value at 510 nm by the highest concentration of H 2 O 2 on the calibration line from the absorption value at 510 nm by those pipetted supernatant solutions.
To determine the H 2 O 2 -scavenging ability of the formed dityrosine and the auxiliary photothermal conversion ability of PDA, the graphene oxide/silk fibroin (GS) film was made in a manner similar to the control film, which was fabricated from GO and SF.  (Figure S1)). The data were further analyzed using Origin 8 (Origin Lab, Northampton, MA, USA).

Photothermal Conversion of GDES Films
Carbonaceous materials such as graphene oxide and carbon nanotubes are effective agents of photothermal conversion [8]. They convert energy that is absorbed in the form of infrared light into heat; PDA materials reportedly have the same property [20]. The photothermal conversion effects of GDES and GS films in 2 mL H 2 O were assessed by irradiating them using a 2W 808 nm NIR [20], and continuously measuring the temperatures of the solutions using a K-type thermocouple with resolution of 0.1 • C at temperature range of this study for 10 min.

Assessment of In Vitro Biocompatibility
The in vitro biocompatibility of GDES film was determined according to standards of ISO 10993-5 and 10993-12 [32]. L929 fibroblasts were purchased from the Bio-resource Collection and Research Center (BCRC, Hsin-Chu, Taiwan) and cultured in Dulbecco's modified Eagle's medium that contained 10% horse serum at 37 • C in a 5% CO 2 incubator. To obtain various dilutions of extraction supernatant solutions on the GDES films, the films were immersed into a culture medium, for 24 h, which was then diluted with a fresh culture medium to make different diluted ratios of mediums. According to requirements of ISO-10993-5, the L929 fibroblasts were then incubated with the various extraction of the various diluted-ratio mediums. for another 24 h, and an MTS assay ((3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2-tetrazolium) and MW. of 487.5 Da, abcam plc, Cambridge, UK), a colorimetric method, was performed to quantify cell proliferation and viability, according to the manufacturer's instructions. The cell viability of each group was compared with that of the control group in which no extraction medium was used.

Statistical Analysis
A Student's t test was conducted to analyze the statistical significance of the variations in the H 2 O 2 -scavenging ability among SF, GS, and GDES films at each working temperature, and among GDES films at various working temperatures. A confidence level of 95% was used to determine statistical significance. Data are presented as mean ± SD from triplicate measurements.

Fabricating GDES Films and Analysis of Their Compositions
The lateral size of GO that was prepared using the modified Hummer's method was in the range of 200~500 nm, as determined from TEM micrographs ( Figure 1c). Also, the lateral size of GDE (Table 1) aggregates around several µm (e.g., 3-5 µm) were shown ( Figure 1c). In addition, the sizes of GO and GDE aggregates were 279 ± 8 nm and 1692 ± 13 nm (n = 3), respectively, as determined using a zeta/nanoparticle analyzer. The size of GDE microaggregates was around 1.7 µM, possibly because DE adhered to the GO surfaces, forming aggregates, perhaps because of extensive hydrogen bonding and π-π interactions between PDA and GO [7,33,34]. However, the possibly forming aggregates by GO during adhesion of DE to GO surfaces could not be ruled out although the suspensions were prepared in highly stirring conditions. To determine the E and D contents of GDE microaggregates, the Bradford protein assay for HRP and the Arnow assay for determining the catechol content of PDA were conducted [28,29]. The results revealed that the amounts of E and D in GDE microaggregates were 240 µg and 380 µg per mg GO, respectively. Approximately 60% of the initial amount of E and 95% of that of D that was used in preparing the DE complexes adhered to the GO surfaces. Interestingly, the mass of DE in 1 mg GDE microaggregates (with a size of 1.7 µm) herein was around 620 µg, which was similar to those obtained elsewhere [35,36]. For instance, Xu and Lai obtained 600 µg and 300 µg of enzyme per mg GO, respectively [35,36]. Since the chemical structures of PDA and GO contains many aromatic rings and abundant hydrophilic moieties, including carboxyl groups and hydroxyl groups [7,33,34], 620 µg of DE complexes might have bonded to GO per gm, producing GDE microaggregates, because of extensive hydrogen bonding and π-π interactions between PDA and GO [7,33,34]. The AFM image for GDE microaggregates was shown ( Figure 1d) which was similar to those of image presented by TEM (Figure 1c). The blending of E and D was monitored using a UV spectrophotometer at about 480 nm ( Figure 1e(A,B)). Most of the changes in absorption might were associated with the degree of polymerization of DA to dark red intermediates of PDA polymers [37]. The size of the DE complexes increased with the blending time of E and DA because polymerization of PDA continued. The blending time of E and DA was thus adjusted to 0.5~1 h to avoid the production of large aggregates of DE complexes (Figure 1a). The size of GDE microaggregates increased to 1.70 ± 0.01 µm (n = 3), which was approximately six times that of GO (~280 nm) (Figure 1c). Since the molecular weight of HRP is about 44 kDa, which substantially exceeds that of DA, the PDA polymers might be located on the shell or outside layer of HRP (Figure 1a), protecting E from the harmful environment. To produce a GDES film, the GDE microaggregate suspension was homogeneously mixed with SF solution by strongly stirred, and then casted and dried. To produce a stable GDES film in water, the film was further treated in ethanol to induce β-sheets of SF (e.g., around 44% β-sheets induced after immersed at 60~85% of ethanol) [26] to avoid SF to be resolved in water.

ATR-FTIR Spectroscopic and SEM Analyses of GDES Films
ATR-FTIR absorbed spectra were obtained to qualitatively examine the functional groups of the components in GDES films ( Figure 2). The characteristic peaks of GO at 1723 cm −1 , 1619 cm −1 , and 1040 cm −1 were attributed to the stretching of the C=O, C=C, and C-O bonds of GO, respectively [38,39]. In the spectra of E (or HRP), the characteristic peaks were amide I, II (1526 cm −1 ) and III [40], which differed slightly from those associated with the amide bonds (of OR in) SF. In the spectra of PDA, the characteristic peaks at 1603 cm −1 , 1499 cm −1 , and 1283 cm −1 corresponded to the C=C stretching, and C=N stretching of the indole ring, and the C-O stretching of the primary amine, respectively [41]. The characteristic peaks of SF were at 1642 cm −1 , 1515 cm −1 , and 1229 cm −1 , corresponding to amide I, amide II, and amide III, respectively ( Figure 2) [26,42]. The characteristic peaks of the GDE microaggregates corresponded to amides I and II, and the C-O bond (~1040 cm −1 ), which were associated with E and G, respectively. Interestingly, the characteristic peak of GO at 1723 cm −1 was much lower in the spectra of DE and GDE owing to the partial reduction of GO to rGO [43,44]. The characteristic peaks of a GDES film corresponded to amides I, II and III, and the C-O bond, which were associated with SF, and G. respectively. Notably, most of characteristic peaks in SF were observed in the spectra of GDES films, possibly because the GDES films contained large amount of SF especially after ethanol treatment. cm −1 was much lower in the spectra of DE and GDE owing to the partial reduction of GO to rGO [43,44]. The characteristic peaks of a GDES film corresponded to amides I, II and III, and the C-O bond, which were associated with SF, and G. respectively. Notably, most of characteristic peaks in SF were observed in the spectra of GDES films, possibly because the GDES films contained large amount of SF especially after ethanol treatment. SEM micrographs of the surfaces of GDES films before and after they were used to scavenge H2O2 are displayed (Figure 3a,b). The GDES film that had not scavenged H2O2 was relatively smooth (Figure 3a) while that of the film that had scavenged H2O2 was very rough and wrinkled (Figure 3b), possibly as a result of the formation many dityrosine bonds in S, induced by H2O2/HRP, which would have caused polymerization and the de-arrangement of polymers of S in the GDES film, producing the wrinkled surface ( Figure 3b). However, the data to quantitate and prove many dityrosine bonds formations in the GDES films induced by H2O2/HRP was not able to be obtained in this study while the indirect evidence for those formations could be evaluated by the intensity of UV-excitable blue fluorescence of GDES films ( Figure 3a) and reports elsewhere [20]. The SEM image of cross-section of the GDES films after they were used to scavenge H2O2 are shown (Figure 3c,d). Figure 3d showed high magnitude of micrograph at the bottom area of center region in Figure 3c. According to those Figure 3b,d, the morphologies of cross-section of GDES films after they scavenged H2O2 were rough and wrinkled which might be associated with the surfaces of the film (Figure 3b). Interestingly, using EDS to analyze the elements of surfaces (C, N and O) of the two aforementioned films (Figure 3a,b) revealed no difference in their elemental contents. For example, for GDES films that had not and had SEM micrographs of the surfaces of GDES films before and after they were used to scavenge H 2 O 2 are displayed (Figure 3a,b). The GDES film that had not scavenged H 2 O 2 was relatively smooth (Figure 3a) while that of the film that had scavenged H 2 O 2 was very rough and wrinkled (Figure 3b), possibly as a result of the formation many dityrosine bonds in S, induced by H 2 O 2 /HRP, which would have caused polymerization and the de-arrangement of polymers of S in the GDES film, producing the wrinkled surface ( Figure 3b). However, the data to quantitate and prove many dityrosine bonds formations in the GDES films induced by H 2 O 2 /HRP was not able to be obtained in this study while the indirect evidence for those formations could be evaluated by the intensity of UV-excitable blue fluorescence of GDES films ( Figure 3a) and reports elsewhere [20]. The SEM image of cross-section of the GDES films after they were used to scavenge H 2 O 2 are shown (Figure 3c,d). Figure 3d showed high magnitude of micrograph at the bottom area of center region in Figure 3c. According to those Figure 3b,d, the morphologies of cross-section of GDES films after they scavenged H 2 O 2 were rough and wrinkled which might be associated with the surfaces of the film (Figure 3b). Interestingly, using EDS to analyze the elements of surfaces (C, N and O) of the two aforementioned films (Figure 3a,b) revealed no difference in their elemental contents. For example, for GDES films that had not and had scavenged H 2 O 2 the C, N, and O contents were 56.9 ± 0.2, 19.7 ± 0.6 and 23.5 ± 0.7% (n = 3); and 56.8 ± 2.4%, 20.3 ± 1.7 and 22.8 ± 0.7 (n = 3), respectively. Since EDS analysis for the image of film surfaces was semi-quantitative, the data for C, N, O analysis of the films for pre-and post-scavenged of H 2 O 2 might qualitatively not be influenced.
Sensors 2020, 20, x FOR PEER REVIEW 9 of 18 scavenged H2O2 the C, N, and O contents were 56.9 ± 0.2, 19.7 ± 0.6 and 23.5 ± 0.7% (n = 3); and 56.8 ± 2.4%, 20.3 ± 1.7 and 22.8 ± 0.7 (n = 3), respectively. Since EDS analysis for the image of film surfaces was semi-quantitative, the data for C, N, O analysis of the films for pre-and post-scavenged of H2O2 might qualitatively not be influenced.

H2O2-Scavenging by GDES Films
Since the GDES films might be applied for sensing and scavenging H2O2 in outdoor at low temperature environments, the temperature might be one of the factors to affect the performance of the films. To quantify whether the scavenging abilities of GDES, GS and SF films were influenced by temperature, they were immersed in H2O2 solutions for 30 min at various temperatures. Scavenging ability was determined as the initial concentration of H2O2 in solution subtracted the residual H2O2 concentration in solution, as determined using 1,10-phenanthroline/FeCl2 method. Figure 4a presents the H2O2 scavenging ability of GDES films at various temperatures. It was highest, 73.5 ± 8.7% or 9.39 × 10 −2 μmol H2O2 at 37 °C (n = 3) compared with the performance at low temperatures. The effect of temperature on H2O2 scavenging ability was consistent with reports that the activity of HRP is highest at approximately 35 °C [45,46]. Since the H2O2 scavenging ability of a GDES film was better in at 37 °C than that at low temperature, the photothermal properties of the film in order to raise working temperature to have a good sensing and scavenging ability were further investigated in the later section (Section 3.5). However, the factors which caused low H2O2 scavenging ability of the film at 25 °C needed to be further studied. Interestingly, SF films had an H2O2 scavenging ability of ~30% (Figure 4b), partly because they highly facilitated the self-decomposition of H2O2 into H2O and oxygen within 30 min because 1,10-phenanthroline/FeCl2 method needed to take around 30 min for determining the concentration of residual H2O2 in solution. Interestingly, the presence of SF in H2O2

H 2 O 2 -Scavenging by GDES Films
Since the GDES films might be applied for sensing and scavenging H 2 O 2 in outdoor at low temperature environments, the temperature might be one of the factors to affect the performance of the films. To quantify whether the scavenging abilities of GDES, GS and SF films were influenced by temperature, they were immersed in H 2 O 2 solutions for 30 min at various temperatures. Scavenging ability was determined as the initial concentration of H 2 O 2 in solution subtracted the residual H 2 O 2 concentration in solution, as determined using 1,10-phenanthroline/FeCl 2 method. Figure 4a presents the H 2 O 2 scavenging ability of GDES films at various temperatures. It was highest, 73.5 ± 8.7% or 9.39 × 10 −2 µmol H 2 O 2 at 37 • C (n = 3) compared with the performance at low temperatures. The effect of temperature on H 2 O 2 scavenging ability was consistent with reports that the activity of HRP is highest at approximately 35 • C [45,46]. Since the H 2 O 2 scavenging ability of a GDES film was better in at 37 • C than that at low temperature, the photothermal properties of the film in order to raise working temperature to have a good sensing and scavenging ability were further investigated in the later section (Section 3.5). However, the factors which caused low H 2 O 2 scavenging ability of the film at 25 • C needed to be further studied. Interestingly, SF films had an H 2 O 2 scavenging ability of 30% (Figure 4b), partly because they highly facilitated the self-decomposition of H 2 O 2 into H 2 O and oxygen within 30 min because 1,10-phenanthroline/FeCl 2 method needed to take around 30 (Figure 4b) [47]. However, the H 2 O 2 -scavenging ability of GS films exceeded that of SF (Figure 4b). Hence, whether GO plays a role in H 2 O 2 scavenging in a GS film must be further investigated. Since the H 2 O 2 scavenging ability of PDA was reported [19], the greater H 2 O 2 -scavenging ability of the GDES film might arise from the synergistic effects of scavenging H 2 O 2 by its reduction to H 2 O by HRP and the radical character of the catechol/quinone structure in the PDA structure. Figure 1b displays (Figure 1b).
Sensors 2020, 20, x FOR PEER REVIEW 11 of 18 solution increased the decomposition of H2O2 to H2O by approximately 12~30% although this increase depended on both the concentrations of SF (1 or 2%) and the mixing time (for example.10 min herein) of the SF and H2O2 solutions. The entrapment of H2O2 by the hydroxyl group, amine, and carboxyl group in SF may play a role in its scavenging (Figure 4b) [47]. However, the H2O2scavenging ability of GS films exceeded that of SF (Figure 4b). Hence, whether GO plays a role in H2O2 scavenging in a GS film must be further investigated. Since the H2O2 scavenging ability of PDA was reported [19], the greater H2O2-scavenging ability of the GDES film might arise from the synergistic effects of scavenging H2O2 by its reduction to H2O by HRP and the radical character of the catechol/quinone structure in the PDA structure. Figure 1b displays the assumed mechanisms of H2O2 scavenging by GDES film. H2O2 molecules in solution are assumed to diffuse through void spaces among SF polymers to the inner components of the GDES film, where they are reduced by E to H2O. The amount of H2O2 that is reduced by a GDES film can be used to evaluate the ability of the film to scavenge H2O2 (Figure 1b).
(a)  . H2O2 scavenging ability tested by using 1, 10-phenanthroline agent to measure the residual H2O2 after immersed GDES films fully scavenged H2O2 in the solution. The 0% scavenging ability was defined as the measurement VIS values using the agent to examine the residual H 2O2 without adding any film to scavenge H2O2, (a) H2O2 scavenging ability for GDES films at various temperatures, and (b) H2O2 scavenging ability for various films at 37 °C. The ability of GDES films was the highest compared with that for SF and GS films at 37 °C. (Note: * p < 0.05, ** p < 0.01, *** p < 0.001; Data are mean ± SD, n = 3).

Use of UV-Excitable Blue Fluorescence of GDES Films to Probe H2O2
The reduction reactions of H2O2 in aqueous solution by E in a GDES film may trigger simultaneous oxidation reactions of tyrosine in S in the film to produce tyrosyl radicals, ultimately forming dityrosine bonds in S, which emit blue fluorescence under UV irradiation (Supplementary Figure S1 and Figure 5a), such that the GDES film can serve as a naked eye-probe of H2O2 [23]. The absorbance of blue fluorescence (at 425 nm) verse the reaction times would reach stable at 30 min till the time of the end observation (e.g., 60 min) and the values at 30 min were chosen in this study ( Figure S1). The linearity between the intensity of the blue fluorescence of the dityrosine bonds of SF that was leak from GDES films and the initial concentration of H2O2 in aqueous solution was determined (Figure 5b, n = 3). The normalized intensity of blue fluorescence was linearly correlated with the initial concentration of H2O2 in aqueous solution from 4.0 to 80 μM (R 2 = 0.984%, Figure 5b, n = 3) with a detection limit of 4 μM although the deviations of data at H2O2 concentration of 40 μM were needed to further be investigated. H2O2 concentration. In aqueous solution has frequently been determined from the fluorescence intensity or absorbance using fluorescent dyes or chromogenic reagents, such as Amplex ® red [14] and 3,5,3′,5′-tetramethylbenzidine (TMB) [15]. However, neither dye is biocompatible, so each may have cytotoxic effects at the site of application such as a wound, limiting its biomedical application in situ. Also, none of those chromogenic reagents can locally scavenge sufficient H2O2. Although blue fluorescence of dityrosine bonds of SF, induced by UV, wasbeen reported [23], this investigation is the first to determine the H2O2-scavenging ability of GDES films, and quantify the linearity between the intensity of their blue fluorescence and the H2O2 concentration in aqueous solution. Although similar UV-excitable fluorescence system using SF, HRP and H2O2 in liquid state was reported by the author's group [48], there were two major differences between two studies; 1. The enzyme, HRP, was immobilized in PDA and coated by SF to produce a

Use of UV-Excitable Blue Fluorescence of GDES Films to Probe H 2 O 2
The reduction reactions of H 2 O 2 in aqueous solution by E in a GDES film may trigger simultaneous oxidation reactions of tyrosine in S in the film to produce tyrosyl radicals, ultimately forming dityrosine bonds in S, which emit blue fluorescence under UV irradiation (Supplementary Figure S1 and Figure 5a), such that the GDES film can serve as a naked eye-probe of H 2 O 2 [23]. The absorbance of blue fluorescence (at 425 nm) verse the reaction times would reach stable at 30 min till the time of the end observation (e.g., 60 min) and the values at 30 min were chosen in this study ( Figure S1). The linearity between the intensity of the blue fluorescence of the dityrosine bonds of SF that was leak from GDES films and the initial concentration of H 2 O 2 in aqueous solution was determined (Figure 5b, n = 3). The normalized intensity of blue fluorescence was linearly correlated with the initial concentration of H 2 O 2 in aqueous solution from 4.0 to 80 µM (R 2 = 0.984%, Figure 5b, n = 3) with a detection limit of 4 µM although the deviations of data at H 2 O 2 concentration of 40 µM were needed to further be investigated. H 2 O 2 concentration. In aqueous solution has frequently been determined from the fluorescence intensity or absorbance using fluorescent dyes or chromogenic reagents, such as Amplex ® red [14] and 3,5,3 ,5 -tetramethylbenzidine (TMB) [15]. However, neither dye is biocompatible, so each may have cytotoxic effects at the site of application such as a wound, limiting its biomedical application in situ. Also, none of those chromogenic reagents can locally scavenge sufficient H 2 O 2 . Although blue fluorescence of dityrosine bonds of SF, induced by UV, wasbeen reported [23], this investigation is the first to determine the H 2 O 2 -scavenging ability of GDES films, and quantify the linearity between the intensity of their blue fluorescence and the H 2 O 2 concentration in aqueous solution. Although similar UV-excitable fluorescence system using SF, HRP and H 2 O 2 in liquid state was reported by the author's group [48], there were two major differences between two studies; 1. The enzyme, HRP, was immobilized in PDA and coated by SF to produce a film in this study while all reactants in the early system were reacted in a liquid state, and 2. the concentrations of H 2 O 2 herein was in uM level while the early one was mM, revealing there were 10 3 differences [49]. The detection range of H 2 O 2 concentrations herein was suitable for detecting harmful concentrations in vivo (10 0~1 0 2 µM) [1,49,50]. However, the mechanisms of complex electron transfers among HRP, PDA, and SF in a GDES film that are involved in scavenging and probing H 2 O 2 in aqueous solution were not investigated herein. These mechanisms will be investigated in the near future. However, without the presence of HRP in the film, for instance, S and GS films, it is hardly to stimulate the rates of the reduction reactions of H 2 O 2 to H 2 O which might result in producing small amounts of dityrosine bonds formations in those films. There is hardly or not able to detect the intensity of UV-excitable blue fluorescence for the S and GS films.
Sensors 2020, 20, x FOR PEER REVIEW 13 of 18 film in this study while all reactants in the early system were reacted in a liquid state, and 2. the concentrations of H2O2 herein was in uM level while the early one was mM, revealing there were 10 3 differences [49]. The detection range of H2O2 concentrations herein was suitable for detecting harmful concentrations in vivo (10 0~1 0 2 μM) [1,49,50]. However, the mechanisms of complex electron transfers among HRP, PDA, and SF in a GDES film that are involved in scavenging and probing H2O2 in aqueous solution were not investigated herein. These mechanisms will be investigated in the near future. However, without the presence of HRP in the film, for instance, S and GS films, it is hardly to stimulate the rates of the reduction reactions of H2O2 to H2O which might result in producing small amounts of dityrosine bonds formations in those films. There is hardly or not able to detect the intensity of UV-excitable blue fluorescence for the S and GS films.

Photothermal Responses of GDES Films
Since the GDES films might be applied for scavenging and sensing H2O2 in outdoor at low temperature environments, raising the working temperature might affect the performance of the films. The photothermal property of GO is widely documented [8] Figure 6 plots the photothermal responses of GDES films during 10 min of NIR irradiation. The temperature of the GDES-containing

Photothermal Responses of GDES Films
Since the GDES films might be applied for scavenging and sensing H 2 O 2 in outdoor at low temperature environments, raising the working temperature might affect the performance of the films. The photothermal property of GO is widely documented [8]. Figure 6 plots the photothermal responses of GDES films during 10 min of NIR irradiation. The temperature of the GDES-containing solution increased from 22 • C to 52 • C whereas that of the GS-containing solution increased from 24 • C to 48 • C. The temperatures of the solutions that contained GDES films and GS films significantly exceed that of the solution without those films under irradiation by NIR, revealing that the GO in the system exhibited a favorable photothermal response whereas the PDA in the GDES films made a minor contribution to it. These findings probably follow from the low PDA content in the GDES films because the proportions of GO and PDA in the GDES films were about 1.6 wt% and 0.6 wt%, respectively. Notably, the photothermal response of a GDES film can be used to heat samples in situ under NIR irradiation, enabling the working temperature to be optimally adjusted for the enzyme, E, in the film to enhance scavenge locally H 2 O 2 . Therefore, the property of scavenging H 2 O 2 locally of the films could be carried out in outdoor at low temperature. The sensing local H 2 O 2 by the films might be improved at the working temperature near 37 • C because the activity of HRP is highest at approximately 35 • C [47,48].
Sensors 2020, 20, x FOR PEER REVIEW 14 of 18 solution increased from 22 °C to 52 °C whereas that of the GS-containing solution increased from 24 °C to 48 °C. The temperatures of the solutions that contained GDES films and GS films significantly exceed that of the solution without those films under irradiation by NIR, revealing that the GO in the system exhibited a favorable photothermal response whereas the PDA in the GDES films made a minor contribution to it. These findings probably follow from the low PDA content in the GDES films because the proportions of GO and PDA in the GDES films were about 1.6 wt% and 0.6 wt%, respectively. Notably, the photothermal response of a GDES film can be used to heat samples in situ under NIR irradiation, enabling the working temperature to be optimally adjusted for the enzyme, E, in the film to enhance scavenge locally H2O2. Therefore, the property of scavenging H2O2 locally of the films could be carried out in outdoor at low temperature. The sensing local H2O2 by the films might be improved at the working temperature near 37 °C because the activity of HRP is highest at approximately 35 °C [47,48]. Figure 6. Photothermal responses of GDES or GS films were carried out by heating the films in water by 2W and at 808 nm NIR for 10 min. The temperature of aqueous increased shortly for GDES or GS films. The temperature of water is about the same for 10 min heating. (Data are mean ± SD, n = 3).

In Vitro Biocompatibility of GDES Films
The viabilities of L929 fibroblasts that were incubated with various extraction solutions which were taken from cultural mediums in which GDES films had been immersed for 24 h were determined and shown in Figure 7 (n = 3). The viabilities of L929 fibroblasts that were incubated in the group of 100% extraction solution remained around 90%, which was slightly lower than that of the other groups. According to ISO 10993-5, the GDES films were therefore non-toxic biomaterials and suitable for use in in vitro and further in vivo studies. Since these films mainly comprise SF and PDA, their biocompatibility might be reasonable [19,26]. Although GO was reported to be a cytotoxic material because it has negative surface charges with the possible production of reactive oxygen species (ROS) on its surface [51,52], its surface might have been fully covered by DE complexes and S in this study. Therefore, the cytotoxic factors on the GO surface might have been suppressed by other components of the GDES film, and the cytotoxicity of GO and the film might have been negligible (Figure 7). . Photothermal responses of GDES or GS films were carried out by heating the films in water by 2W and at 808 nm NIR for 10 min. The temperature of aqueous increased shortly for GDES or GS films. The temperature of water is about the same for 10 min heating. (Data are mean ± SD, n = 3).

In Vitro Biocompatibility of GDES Films
The viabilities of L929 fibroblasts that were incubated with various extraction solutions which were taken from cultural mediums in which GDES films had been immersed for 24 h were determined and shown in Figure 7 (n = 3). The viabilities of L929 fibroblasts that were incubated in the group of 100% extraction solution remained around 90%, which was slightly lower than that of the other groups. According to ISO 10993-5, the GDES films were therefore non-toxic biomaterials and suitable for use in in vitro and further in vivo studies. Since these films mainly comprise SF and PDA, their biocompatibility might be reasonable [19,26]. Although GO was reported to be a cytotoxic material because it has negative surface charges with the possible production of reactive oxygen species (ROS) on its surface [51,52], its surface might have been fully covered by DE complexes and S in this study. Therefore, the cytotoxic factors on the GO surface might have been suppressed by other components of the GDES film, and the cytotoxicity of GO and the film might have been negligible (Figure 7). Sensors 2020, 20, x FOR PEER REVIEW 15 of 18 Figure 7. Cell viability of L929 fibroblasts were performed to examine the biocompatibility of GDES films which were incubated with various percentages (e.g., 25~100%) of extract medium from GDES films/incubated medium according to the ISO-10993-5. Accordingly, the GDES films were biocompatible. (Data are mean ± SD, n = 3).

Conclusions
In this work, photothermally responsive GDES films that can scavenge and probe H2O2 by UVexcitable blue fluorescence was developed. The TEM micrographs and AFM topographies for GDE aggregates (about 2 μM) were shown. In addition, the rough and wriggle surfaces of SEM micrographs for the GDES films after they scavenged H2O2 were shown (Figure 3b). The H2O2scavenging abilities of the GDES film were influenced by temperature with the highest value of 9.39 × 10 −2 μmol H2O2 per film at 37 °C (Figure 4a). The normalized blue fluorescent intensities after GDES films scavenged H2O2 were linearly increased with increasing H2O2 concentration from the range of 4.0 to 80 μM (Figure 5b). In addition, the GDES film exhibited a photothermal response and good biocompatibility (Figure 7). Since GDES films could be used to heat samples in situ under NIR irradiation, and enabled raising the working temperature thereof to 37 °C for enhancing scavenging H2O2 locally. Therefore, the property of scavenging H2O2 locally of the films could be carried out in outdoor at low temperature. In addition, GDES films may have the potential for locally scavenging associated with probing H2O2 under various pathological conditions, such as those that pertain to unhealing wound management.

Conflicts of Interest:
The authors declare no conflict of interest. Figure 7. Cell viability of L929 fibroblasts were performed to examine the biocompatibility of GDES films which were incubated with various percentages (e.g., 25~100%) of extract medium from GDES films/incubated medium according to the ISO-10993-5. Accordingly, the GDES films were biocompatible. (Data are mean ± SD, n = 3).