A Dual pH/O2 Sensing Film Based on Functionalized Electrospun Nanofibers for Real-Time Monitoring of Cellular Metabolism

Real-time monitoring of dissolved oxygen (DO) and pH is of great significance for understanding cellular metabolism. Herein, a dual optical pH/O2 sensing membrane was prepared by the electrospinning method. Cellulose acetate (CA) and poly(ε-caprolactone) (PCL) nanofiber membrane blended with platinum (II)-5,10,15,20-tetrakis-(2,3,4,5,6-pentafluorophenyl)-porphyrin (PtTFPP) was used as the DO sensing matrix, upon which electrospun nanofiber membrane of chitosan (CS) coupled with fluorescein 5-isothiocyanate (FITC) was used as the pH sensing matrix. The electrospun sensing film prepared from biocompatible biomaterials presented good response to a wide range of DO concentrations and physiological pH. We used it to monitor the exracellular acidification and oxygen consumption levels of cells and bacteria. This sensing film can provide a luminescence signal change as the DO and pH change in the growth microenvironment. Due to its advantages of good biocompatibility and high stability, we believe that the dual functional film has a high value in the field of biotechnology research.


Introduction
The monitoring of metabolite parameters associated with cellular growth is essential for biological and clinical studies [1]. During the processes of cellular respiration and metabolism, the consumption of oxygen and the production of metabolites will lead to a change in the concentration of dissolved oxygen (DO) and pH level in the culture microenvironment [2][3][4][5]. Therefore, sensors for real-time monitoring of DO and pH in the cellular growth microenvironment are of great significance.
Optical sensors based on reversible luminescence changes in specific indicators show high potential for environmental detection, due to the advantages of fast, simple and easyto-read signals [6]. These sensors composed with probes and polymer matrices can clearly reflect the environmental changes through luminescence variations. For the selection of biosensing matrix materials, biobased polymers from natural sources have inherent advantages, such as low cost, nontoxicity, biodegradability and eco-friendliness, which make them attractive in the field of biomonitoring.
Cellulose acetate (CA), the most abundant biobased material, has attracted the interest of scientists as an oxygen-sensitive matrix due to its high stability, simple molding process and easy modification [7][8][9]. However, its sensitivity to oxygen is still inadequate, because of its poor oxygen permeability [10]. Some attempts have been made to address this issue. For example, polymethyl methacrylate [7] and SiO 2 [11] were used to improve the response performance of CA [10], but both of them are nondegradable materials, which limited their application. Hence, searching for modified materials with biodegradability is of CA and PCL. The mixture was stirred for 3 h before electrospinning. A DC voltage of 18 kV was applied between the syringe tip and an aluminum collector at a distance of 15 cm for electronspinning. The flow rate of the solution was 0.6-1.0 mL/h at room temperature. The obtained CA&PCL film with PtTFPP was then treated in a 0.05 M NaOH aqueous solution for 7 days for deacetylation. The film was rinsed three times with DI water to remove residual NaOH, and after that it was placed in a vacuum oven to dry at room temperature for two days. The obtained film was called F1 (Scheme 1a).

Preparation of CS Electrospun Nanofibrous Matrix for pH Sensing (F2)
CS-FITC was dissolved in 70% aqueous acetic acid to prepare a 2 wt% solution. A DC voltage of 18 kV was applied. The flow rate of the solution for electrospinning was 1.0 mL/h at humidity below 40%. The obtained films were placed in a dessicator containing a mixed aqueous solution of 2.5% glutaraldehyde (GA) and 10% HCL. After the air inside the dessicator was exhausted, the films were cross-linked for 2 h by the reaction of GA vapor with the amino and/or hydroxyl groups of the CS compositions to increase the stability of the films when used in aqueous solution. Finally, these films (F2, Scheme 1b) were rinsed with DI water three times and dried in a vacuum oven at room temperature.

Preparation of CS Electrospun Nanofibrous Matrix on the Top of F1 for Dual Senor F3
Dual sensor F3 was prepared by electrospinning on the top of F1 film and crosslinked with GA. The procedure was similar to F2, except that the F3 film consists of both the F1 and F2 compositions (Scheme 1c).

Preparation of CS Electrospun Nanofibrous Matrix for pH Sensing (F2)
CS-FITC was dissolved in 70% aqueous acetic acid to prepare a 2 wt% solution. A DC voltage of 18 kV was applied. The flow rate of the solution for electrospinning was 1.0 mL/h at humidity below 40%. The obtained films were placed in a dessicator containing a mixed aqueous solution of 2.5% glutaraldehyde (GA) and 10% HCL. After the air inside the dessicator was exhausted, the films were cross-linked for 2 h by the reaction of GA vapor with the amino and/or hydroxyl groups of the CS compositions to increase the stability of the films when used in aqueous solution. Finally, these films (F2, Scheme 1b) were rinsed with DI water three times and dried in a vacuum oven at room temperature. The sensing films using only CA or PCL as the matrix were prepared by electrospinning of 10 wt% CA solution and 15 wt% PCL solution, respectively. Both of the films contained 0.05 wt% PtTFPP. The DC voltage and flow rates of electrospinning are consistent with those of F1.

Characterization of Nanofibrous Matrices
To characterize the fiber morphology, samples were sputter coated with platinum and analyzed with a scanning electron microscope (SEM, MIRA3, TESCAN, Brno, Czech). Contact angle was found with a contact angle tester (AST VCA Optima XE, Beijing, China) to study the hydrophilic and hydrophobic properties of the materials. The phosphorescence microscopy images of microfibrous sensing films were captured with a confocal laser scanning microscope (CLSM, TCS SP8, Leica, Germany).

Response to pH and DO
The sensing films were cut to dimensions of 1.4 cm × 1.4 cm and placed in a quartz cuvette filled with water or buffer. The fluorescence spectra were excited at 488 nm for pH responses and 405 nm for oxygen sensing and collected by using a JASCO FP-8600 fluorophotometer. Britton-Robison buffers (BR buffers) with various pH values were prepared to evaluate the pH responses of the sensing films. The film was washed at least three times to remove the residual solution when changing buffers. To ensure the accuracy of the experiment, titrations were performed twice to get the average value. DO was adjusted by bubbling mixed gas of N 2 and O 2 into the solution in the quartz cuvette for 5 min at a controlled flow rate of 100 SCCM to ensure an equilibrium was reached. NIH 3T3 mouse embryonic fibroblast cells were used for cell respiration studies. The cells were cultured with Dulbecco's modified Eagle's medium (DMEM) with 10% FBS and 1% antibiotics, and incubated at 37 • C in 5% CO 2 overnight. The cell density was detected by an automated cell counter (ACC, TC20, Bio-Rad, Hercules, CA, USA), then diluted by fresh medium to appropriate densities (10 6 cells/mL, 5 × 10 5 cells/mL, 2.5 × 10 5 cells/mL and 1.25 × 10 5 cells/mL) for further measurements. Escherichia coli (ATCC 25922) was cultured by shaking in liquid lysogeny broth (LB) broth at 37 • C overnight. After that, optical density at 600 nm (OD 600 nm ) was measured with a UV-Vis spectrophotometer to determine bacterial density. A value of 1.0 indicated a cell density of 5 × 10 8 colony-forming units per milliliter (CFU/mL). Then, the bacteria were diluted by LB medium to suitable densities (10 6 CFU/mL, 5 × 10 5 CFU/mL, 2.5 × 10 5 CFU/mL and 1.25 × 10 5 CFU/mL) for further studies.
The 200 µL diluted cell and bacterial suspensions with various cell densities were added to a 96-well plate covered with pre-prepared pH/O 2 responsive membrane F3 at the bottom, and a 70 µL mineral oil seal was used to isolate the cell medium from air to prevent potential oxygen exchange. The wells filled with only nutrient medium without cells and bacteria on the sensing me mbrane were used as a blank control group. The temperature of the microplate reader was set at 37 • C. The excitation wavelengths for pH probes and O 2 probes were 488 nm and 405 nm, respectively, and the emission intensities at 520 nm and 650 nm were measured with a microplate reader (Cytation 3, BioTek, Winooski, VT, USA) every 5 min.
2.6. In Vitro Biocompatibility Test 2.6.1. MTT Assay Membrane biocompatibility was evaluated through an MTT assay. Briefly, the disinfected and cleaned films (1 × 4 cm 2 ) were immersed in fresh medium for 3 days, then the extraction medium was treated using a 0.20 µm filter. Then, 8 × 10 3 NIH 3T3 cells per well were seeded in 96-well plates with 100 µL culture medium and incubated for 24 h. Diluted extraction medium (100 µL) of sensing films in a series of concentrations (100, 75, 50, 25 and 0%) was added. The NIH 3T3 cells were treated for 24, 48 and 72 h, then MTT reagent was added. After four hours of incubation, the medium was removed and 150 µL of DMSO was added to dissolve formazan purple crystals. The absorbance of the samples was recorded at 570 nm by a microplate reader.

Cell Proliferation and Spreading
Electrospun sensing films were sterilized by ethanol, then transferred to a 24-well plate. Then, 5 × 10 4 NIH 3T3 cells per well were seeded with complete medium on the samples. After incubation for 24 h, the samples were rinsed three times with PBS, then fixed with 2.5% GA for one hour. Then, the samples were dehydrated through an alcohol gradient and air dried for 24 h. Dry membranes with cells were sputter coated with platinum to evaluate the cell adhesion on the membranes by SEM.

Synthesis of CS-FITC Conjugates
The CS-FITC was synthesized by the reaction between the nucleophilic amino group of CS and the isothiocyanate groups of FITC ( Figure 1a) [19]. The CS-FITC solution showed a deep yellowish color (Figure 1b), while the solution under excitation showed a green emission (Figure 1c). cells and bacteria on the sensing me mbrane were used as a blank control group. The temperature of the microplate reader was set at 37 °C. The excitation wavelengths for pH probes and O2 probes were 488 nm and 405 nm, respectively, and the emission intensities at 520 nm and 650 nm were measured with a microplate reader (Cytation 3, BioTek, Winooski, VT, USA) every 5 min.

MTT Assay
Membrane biocompatibility was evaluated through an MTT assay. Briefly, the disinfected and cleaned films (1 × 4 cm 2 ) were immersed in fresh medium for 3 days, then the extraction medium was treated using a 0.20 µm filter. Then, 8 × 10 3 NIH 3T3 cells per well were seeded in 96-well plates with 100 µL culture medium and incubated for 24 h. Diluted extraction medium (100 µL) of sensing films in a series of concentrations (100, 75, 50, 25 and 0%) was added. The NIH 3T3 cells were treated for 24, 48 and 72 h, then MTT reagent was added. After four hours of incubation, the medium was removed and 150 µL of DMSO was added to dissolve formazan purple crystals. The absorbance of the samples was recorded at 570 nm by a microplate reader.

Cell Proliferation and Spreading
Electrospun sensing films were sterilized by ethanol, then transferred to a 24-well plate. Then, 5 × 10 4 NIH 3T3 cells per well were seeded with complete medium on the samples. After incubation for 24 h, the samples were rinsed three times with PBS, then fixed with 2.5% GA for one hour. Then, the samples were dehydrated through an alcohol gradient and air dried for 24 h. Dry membranes with cells were sputter coated with platinum to evaluate the cell adhesion on the membranes by SEM.

Synthesis of CS-FITC Conjugates
The CS-FITC was synthesized by the reaction between the nucleophilic amino group of CS and the isothiocyanate groups of FITC ( Figure 1a) [19]. The CS-FITC solution showed a deep yellowish color (Figure 1b), while the solution under excitation showed a green emission (Figure 1c).

Membrane Characterization
In our study, CA was selected as the supporting matrix for our primary study. In addition, the PCL, which has been proved to be an excellent host material for desired biodegradability properties and satisfactory oxygen permeability [20], was blended with CA material for electrospinning (Scheme 1a). The successful preparation of F1 fiber via electrospinning from DMF solution was evaluated by SEM. F1 was composed of fine nanofibers with an average fiber diameter of 409 ± 157 nm (Figure 2a,d). Additionally, regular, bead-free F2 nanofibers with an average diameter of 227 ± 128 nm (Figure 2b,e) were successfully prepared. Eventually, we obtained a two-layer electrospun membrane F3, and the morphology of F3 is shown in Figure 2c. It can also be seen intuitively in Figure 2c that F2 nanofibers' average diameter was lower than those of the F1 nanofibers. Other than the surface appearance, Figure S1 shows the SEM images of the cross-section morphologies of F3 film, whereby the two-layer structure can be clearly detected. The thickness of the F1 and F2 layer is about 5 µm and 15 µm, respectively. The increased surface-to-volume ratios are expected to prominently reduce barriers to oxygen diffusion. This structure is vital for applications in reporting localized DO and pH levels in the vicinity of cells, as these nanofibrous structures allow cells to interact closely with the matrix and the incorporated probes [20]. Furthermore, stress-strain curves of F1 membrane were studied. The composite materials of CA and PCL nanofibers showed a high flexibility, the elongation at break was up to 148% ( Figure S2). This property can broaden the application of the sensing films in other fields, such as the introduction of indicators on masks.
cessfully prepared. Eventually, we obtained a two-layer electrospun membrane F3, a the morphology of F3 is shown in Figure 2c. It can also be seen intuitively in Figure 2c t F2 nanofibers' average diameter was lower than those of the F1 nanofibers. Other than surface appearance, Figure S1 shows the SEM images of the cross-section morpholog of F3 film, whereby the two-layer structure can be clearly detected. The thickness of F1 and F2 layer is about 5 µm and 15 µm, respectively. The increased surface-to-volu ratios are expected to prominently reduce barriers to oxygen diffusion. This structur vital for applications in reporting localized DO and pH levels in the vicinity of cells these nanofibrous structures allow cells to interact closely with the matrix and the inc porated probes [20]. Furthermore, stress-strain curves of F1 membrane were studied. T composite materials of CA and PCL nanofibers showed a high flexibility, the elongat at break was up to 148% ( Figure S2). This property can broaden the application of sensing films in other fields, such as the introduction of indicators on masks. The optical images presented in Scheme 1d show the emissions of three nanofi membranes. The films give sufficient red and green emissions, respectively, when posed to ultraviolet light (365 nm). As PtTFPP probes were incorporated in the CA&P polymer fibers, red emission from the fibers was further observed under an excitat (Figure 3a). Similarly, since FITC probes were coupled to chitosan, green fluoresce from the fibers was observed (Figure 3b). The optical images presented in Scheme 1d show the emissions of three nanofiber membranes. The films give sufficient red and green emissions, respectively, when exposed to ultraviolet light (365 nm). As PtTFPP probes were incorporated in the CA&PCL polymer fibers, red emission from the fibers was further observed under an excitation (Figure 3a). Similarly, since FITC probes were coupled to chitosan, green fluorescence from the fibers was observed (Figure 3b).  The hydrophobicity/hydrophilicity of the four different films were determ water contact angle experiment (Figure 4). F4 is the original pure CA electro and showed strong hydrophobicity with a contact angle of 132.9° (Figure 4a). A ing with PCL, the mixed film F1 was obtained and the angle of the F1 fibers d The hydrophobicity/hydrophilicity of the four different films were determined by a water contact angle experiment (Figure 4). F4 is the original pure CA electrospun film, and showed strong hydrophobicity with a contact angle of 132.9 • (Figure 4a). After blending with PCL, the mixed film F1 was obtained and the angle of the F1 fibers decreased to 96.3 • (Figure 4c). The further interaction of CS made the water contact angle of the obtained F3 bilayer nanofibers decrease to 50.1 • , showing that the nanofibrous complex surface exhibits hydrophilicity.
water contact angle experiment (Figure 4). F4 is the and showed strong hydrophobicity with a contact ang ing with PCL, the mixed film F1 was obtained and the 96.3° (Figure 4c). The further interaction of CS made tained F3 bilayer nanofibers decrease to 50.1°, showing face exhibits hydrophilicity.

Oxygen Sensing Performance
The sufficient triplet-triplet energy transfer from to oxygen molecules results in the quenching of the e This kind of quenching behavior is influenced by m matrix or media and the microenvironment in whic quenching responses of PtTFPP in F3 were tested over emission spectra of the film in a series of diverse con Figure 5a under the excitation of 405 nm. Strong em absence of O2 was observed, and the emission decrea tions. The relative intensity change as a function of Stern-Volmer Equation (1).

Oxygen Sensing Performance
The sufficient triplet-triplet energy transfer from the oxygen probes (herein, PtTFPP) to oxygen molecules results in the quenching of the emission of the oxygen probes [21]. This kind of quenching behavior is influenced by molecules of probes, the structure of matrix or media and the microenvironment in which the probes are located [21]. The quenching responses of PtTFPP in F3 were tested over a range of DO concentrations. The emission spectra of the film in a series of diverse concentrations of DO are provided in Figure 5a under the excitation of 405 nm. Strong emission at 650 nm of PtTFPP in the absence of O 2 was observed, and the emission decreased with increasing DO concentrations. The relative intensity change as a function of oxygen concentration follows the Stern-Volmer Equation (1). showing that the sensitivity of the electrospun fabricated sensor would be improved due to the much higher surface area-to-volume ratio [22]. Moreover, composite sensing films of CA and PCL were demonstrated to have better sensing performance as compared with the pure CA film ( Figure S3). At the excitation wavelength of the PtTFPP (λ = 405 nm), the emission wavelength of FITC at 520 nm had no obvious changes under different DO concentrations, showing that the pH responses were not affected by oxygen.

pH Sensing Performance
Fluorescein has different chemical structures under acidic and basic conditions. The conjugated dianionic tautomer formed in a basic condition showed intense green emission under 488 nm excitation, while the nonconjugated lactone tautomer formed in an acidic condition showed weak or no emission under the 488 nm excitation [23] ( Figure S4). The pH response behavior of F3 was studied. As expected, the fluorescence intensity increased with the increase in pH (Figure 5c). The pKa value was determined according to the modified Henderson-Hasselbalch equation (2).
where Imax and Imin are the fluorescence intensities of the probe at pH = 9.0 and pH = 3.0, respectively, and Ix is the fluorescence intensity measured at 520 nm under the excitation of 488 nm. The pKa for the sensing film was calculated to be 6.07. The dynamic range for pH is from 5.0 to 8.0, indicating its suitableness for biological studies. The results of Figure S3 revealed that the electrospun fiber films as oxygen sensors were more sensitive than the smooth thin film. This is consistent with previous reports, showing that the sensitivity of the electrospun fabricated sensor would be improved due to the much higher surface area-to-volume ratio [22]. Moreover, composite sensing films of CA and PCL were demonstrated to have better sensing performance as compared with the pure CA film ( Figure S3). At the excitation wavelength of the PtTFPP (λ = 405 nm), the emission wavelength of FITC at 520 nm had no obvious changes under different DO concentrations, showing that the pH responses were not affected by oxygen.

pH Sensing Performance
Fluorescein has different chemical structures under acidic and basic conditions. The conjugated dianionic tautomer formed in a basic condition showed intense green emission under 488 nm excitation, while the nonconjugated lactone tautomer formed in an acidic condition showed weak or no emission under the 488 nm excitation [23] ( Figure S4). The pH response behavior of F3 was studied. As expected, the fluorescence intensity increased with the increase in pH (Figure 5c). The pKa value was determined according to the modified Henderson-Hasselbalch Equation (2).
where I max and I min are the fluorescence intensities of the probe at pH = 9.0 and pH = 3.0, respectively, and I x is the fluorescence intensity measured at 520 nm under the excitation of 488 nm. The pKa for the sensing film was calculated to be 6.07. The dynamic range for pH is from 5.0 to 8.0, indicating its suitableness for biological studies.
The red phosphorescence emitted by the PtTFPP at 650 nm was not influenced by pH. The inset diagram in Figure 5d shows the emission color changes at four different pH values. When the solution environment was basic (pH = 9.0), intense green emission was observed, whereas in acidic solution (pH = 3.0), the green fluorescence disappeared and only the red emission of the oxygen probe was visible. The emission spectrum of FITC and absorption spectrum of PtTFPP are shown in Figure S5. Their spectra overlap, but they are far away from each other in the two-layer film system, failing to reach the standard of a distance of less than 10 nm. Therefore, there is almost no FRET phenomenon between them, which can be ignored. This further shows that red and green channels are less likely to have interference with each other during the simultaneous detection of pH and O 2 , which ensures the effective use of the sensor film when it is used to obtain multiple biological signals in a complex environment.

Reversibility of the Sensing Film F3
The reversibility of the sensing film is vital for real-time biological application. The kinetic phosphorescent spectra of F3 by alternating cycles of saturated O 2 and saturated N 2 solution indicated that it exhibits reversible quenching and recovering properties (Figure 6a) and satisfies the demands of a continuous DO monitoring (Figure 6b). Combining synthetic polymers with natural polymers can make their respective advantages complementary to each other. As shown in Figure S6, the optical stability of F5 (pure PCL electrospun sensing films) is not good. Thus, the cooperation between CA and PCL can meet the requirements of improving both the sensitivity ( Figure S3) and the stability (Figure 6a,b) of the composite films. The resistance of the sensing film to mechanical deformation was also explored ( Figure S7). The sensing performance of the film barely changed after 10 times of folding. This also demonstrated the stability of the film. In Figure 6c,d, the fluorescence intensity between pH 4.0 and 8.0 is also reversible. This detection range can meet the basic needs of biological physiological monitoring. The red phosphorescence emitted by the PtTFPP at 650 nm was not influenced by pH. The inset diagram in Figure 5d shows the emission color changes at four different pH values. When the solution environment was basic (pH = 9.0), intense green emission was observed, whereas in acidic solution (pH = 3.0), the green fluorescence disappeared and only the red emission of the oxygen probe was visible. The emission spectrum of FITC and absorption spectrum of PtTFPP are shown in Figure S5. Their spectra overlap, but they are far away from each other in the two-layer film system, failing to reach the standard of a distance of less than 10 nm. Therefore, there is almost no FRET phenomenon between them, which can be ignored. This further shows that red and green channels are less likely to have interference with each other during the simultaneous detection of pH and O2, which ensures the effective use of the sensor film when it is used to obtain multiple biological signals in a complex environment.

Reversibility of the Sensing Film F3
The reversibility of the sensing film is vital for real-time biological application. The kinetic phosphorescent spectra of F3 by alternating cycles of saturated O2 and saturated N2 solution indicated that it exhibits reversible quenching and recovering properties (Figure 6a) and satisfies the demands of a continuous DO monitoring (Figure 6b). Combining synthetic polymers with natural polymers can make their respective advantages complementary to each other. As shown in Figure S6, the optical stability of F5 (pure PCL electrospun sensing films) is not good. Thus, the cooperation between CA and PCL can meet the requirements of improving both the sensitivity ( Figure S3) and the stability (Figure  6a,b) of the composite films. The resistance of the sensing film to mechanical deformation was also explored ( Figure S7). The sensing performance of the film barely changed after 10 times of folding. This also demonstrated the stability of the film. In Figure 6c,d, the fluorescence intensity between pH 4.0 and 8.0 is also reversible. This detection range can meet the basic needs of biological physiological monitoring.

Real-Time Monitoring of DO and pH during Cellular Respiration Process
Cells are characterized by vigorous metabolism and diversified metabolic pathways. When oxygen and nutrients are abundant, cells obtain energy mainly through aerobic respiration and the glycolysis pathway [24]. Therefore, the measurement of metabolites such as DO and pH during cellular metabolism can indicate the growth state of cells and bacteria [25]. Here, we used NIH 3T3 cells and E. coli as biological models to test the monitoring ability of the sensing film in the process of cell growth. Cells in culture medium with various concentrations were seeded into a 96-well microplate with F3 on the bottom. Mineral oil was used to seal the surface of the medium to avoid the exchange of oxygen in the medium with air. The emissions of oxygen probe and pH probe at 650 nm and 520 nm were recorded during cell growth with the excitation of 405 nm and 488 nm, respectively.
As illustrated in Figure 7a,e, the emission intensities of oxygen probes gradually increased with the prolongation of incubation time. The corresponding DO concentrations were calculated and are given in Figure 7b,f. This phenomenon can be mainly attributed to the depletion of oxygen in cellular respiration. Aerobic respiration is the main way for aerobic organisms to obtain energy through the complete oxidation of matrix. It is a biological oxidation process in which molecular oxygen acts as the final electron acceptor, and thus oxygen is consumed in this process ( Figure S8). In addition, high bacterial densities were accompanied by high rates of fluorescence change, due to rapid consumption of oxygen by massive bacterial respiration [26].
The acidification of the medium via the catabolism of carbohydrate is proved by the fluorescence changes of FITC, summarized in Figure 7c,d,g,h. With the increase in cell concentration and the prolonging of incubation time, the emission density at 520 nm decreased, indicating that the medium showed a tendency of acidification. Carbohydrates in the culture medium are consumed by cells and bacteria as carbon sources [24], and electrons supplied during metabolism are transferred to the membrane electron transport chain, where a series of redox reactions release protons, leading to acidification of the growth environment ( Figure S8).
Bacteria proliferate by a simple binary fission, and thus have an astonishing rate of proliferation and metabolism compared to cells [27]. The oxygen consumption and extracellular acidification manners of cells and bacteria are similar, when they are in an ultrahigh density (Figure 7a,e). However, as the density dropped, the differences between them became more and more pronounced. It seems that the cells need more time than bacteria to reach DO and pH equilibrium, revealing that the cells metabolize respiration at a relatively slow rate.
Some interference was also found in Figure 7a,e. The phosphorescence intensities of the O 2 probes decreased slightly during the experiment. This phenomenon is mainly attributed to the temperature instability of mixtures, which affected the oxygen probe of PtTFPP slightly [28]. Another interesting finding is that the variation trend of the pH was much slower than that of oxygen, which was based on the fact that the buffering capability of the culture medium retards the pH variations in the environment. Therefore, the F3 displayed the ability for simultaneously monitoring the DO and pH changes in microbial environments.
x FOR PEER REVIEW 11 of 14

In Vitro Biocompatibility Test
The MTT assay indicated that more than 90% of the NIH 3T3 cells were viable compared with the control group (Figure 8a), which also means all the membranes' extracts had no cytotoxicity. The above results were further confirmed, and the morphology of the cell attachment and spreading on the membranes after 24 h of culture were seen from the SEM images (Figure 8b). Most cells on the F3 membranes were intensively adhered to the surfaces and showed spindle-like shapes. These characteristics revealed the good biocompatibility of F3 nanofibrous membranes for NIH 3T3 proliferation. All of the above results indicate the outstanding potential of F3 membranes for biological application.
of the culture medium retards the pH variations in the environment. Therefore, the F3 displayed the ability for simultaneously monitoring the DO and pH changes in microbial environments.

In Vitro Biocompatibility Test
The MTT assay indicated that more than 90% of the NIH 3T3 cells were viable compared with the control group (Figure 8a), which also means all the membranes' extracts had no cytotoxicity. The above results were further confirmed, and the morphology of the cell attachment and spreading on the membranes after 24 h of culture were seen from the SEM images (Figure 8b). Most cells on the F3 membranes were intensively adhered to the surfaces and showed spindle-like shapes. These characteristics revealed the good biocompatibility of F3 nanofibrous membranes for NIH 3T3 proliferation. All of the above results indicate the outstanding potential of F3 membranes for biological application. The growth curve of E. coli was measured by the OD600 nm approach. Growth was characterized by a rapid initial exponential growth phase, and reached the platform stage in less than 3 h ( Figure S9). There was no significant difference in the growth curve between the experimental group with the sensing films and the control group, indicating the low toxicity of the material.

Conclusions
A thin film sensor based on composite materials has been successfully prepared by an electrospinning strategy. The film can be used to sense and image both pH and DO The growth curve of E. coli was measured by the OD 600 nm approach. Growth was characterized by a rapid initial exponential growth phase, and reached the platform stage in less than 3 h ( Figure S9). There was no significant difference in the growth curve between the experimental group with the sensing films and the control group, indicating the low toxicity of the material.

Conclusions
A thin film sensor based on composite materials has been successfully prepared by an electrospinning strategy. The film can be used to sense and image both pH and DO simultaneously without interference from each other's signals, which has the potential for biological applications. The unique chemical stability of the film ensures its suitability in biological environments. The sensing film was successfully applied to record the change in extracellular DO and pH during the process of cell growth, which can help understand the metabolism behavior of cells. Compared to sensors based on a single function, this multifunctional sensor design strategy is also expected to integrate simultaneous luminous imaging of multiple biological signals. These specific properties mean the sensing film is promising to achieve in situ dynamic monitoring in many fields such as environmental monitoring, biological manufacturing, wound care, etc., with great research and application value.
Supplementary Materials: The following are available online: Figure S1: SEM images of the film cross-section displaying the thickness of the two-layer membrane. Figure S2: The stress-strain behavior of F1 film. Figure S3: Stern-Volmer plots for the pure CA and CA&PCL in the form of electrospun films or smooth thin films. Figure S4: Structural changes in the fluorescein moieties in strong basic and acidic conditions. Figure S5: The absorption spectra of PtTFPP and emission spectra of FITC. Figure S6: Reversibility of the PCL electrospun film in O 2 -saturated and N 2 -saturated solutions. Figure S7: Response of the F1 sensing films before and after 10 times of folding. Figure S8: Scheme of aerobic respiration and glycolysis. Figure S9: OD 600 nm measurement of E. coli with and without CS film in neutral medium (pH = 7.0).