Fluorescent Polymeric Nanofibers as Ratiometric Multiplexed Skin Sensors of pH and Oxygen
Rémi Pelletier, Anila Hoskere Ashoka, Andrey S. Klymchenko
TL;DR
This paper introduces fluorescent nanofibers that can simultaneously monitor pH and oxygen levels in wounds, enabling remote tracking of healing progress.
Contribution
The development of dual FRET-based nanofibers for ratiometric and multiplexed sensing of pH and oxygen in biomedical contexts.
Findings
Nanofibers use dual FRET to amplify signals for pH and oxygen detection.
A multiplexing device was created to sense both parameters in a wound model.
The materials are compatible with smartphone-based detection for practical biomedical use.
Abstract
Monitoring wound healing requires the development of multiplexed sensors with remote readout. Here, we report fluorescent polymeric nanofibers and nanorods as ratiometric sensors of two important physiological parameters: pH and oxygen. These nanofibers operate by dual Forster resonance energy transfer (FRET) between large number of energy donor dyes (reference) and limited number of two distinct energy acceptors sensitive to these two analytes. This configuration ensures signal amplification of analyte‐sensitive energy acceptor by light‐harvesting principle. The oxygen sensor is based on encapsulation of cationic donor dyes (cyanine or rhodamine derivatives) with bulky hydrophobic counterions as FRET donors and Pt‐porphyrins as energy acceptors inside the nanofibers. In the pH sensor, nanofibers loaded with donor dyes are functionalized at the surface with the rhodamine‐derived energy…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIGURE 1
FIGURE 2
FIGURE 3
FIGURE 4
FIGURE 5
FIGURE 6
FIGURE 7
FIGURE 8
FIGURE 9
FIGURE 10| Fiber mat | Dye content | Polymer concentration in solvent (wt.%) | Solvent |
|---|---|---|---|
|
O2 sensor BlueCy/PtOEP | BlueCy 12wt.% versus polymer with PtOEP at 100:1 mol ratio of BlueCy to PtOEP | 22 | DMF |
|
O2 sensor R18/PtPTPB | R18‐F5‐TPB 20 wt.% versus polymer with PtPTPB at 100:1 mol ratio of BlueCy to PtOEP | 22 | DMF |
| Blank (large rods) | None; FITC grafted later | 30 | DMF |
|
pH sensor BlueCy/Rho‐pH | BlueCy 12 wt.% versus polymer; Rho‐pH grafted later | 22 | DMF |
- —HORIZON EUROPE European Research Council10.13039/100019180
- —SATT Conectus Alsace
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAnalytical Chemistry and Sensors · Carbon and Quantum Dots Applications · Luminescence and Fluorescent Materials
Introduction
1
Normal functioning and healing processes in living systems are controlled by some basic chemical components like protons and oxygen. In particular, wound healing, which involves a variety of medical cases, such as recovery after surgical operations, trauma or chronic wounds caused by diabetes, is strongly influenced by pH [1, 2] and oxygen concentration [3, 4, 5]. It has been well established that normal skin has a slightly acidic pH (5.5–6.0), whereas wounded areas are characterized by alkaline pH at the initial stages [6]. The normal healing processes is generally accompanied by a gradual pH decrease [1, 7]. This decrease is directly linked to enzymatic activity, cell proliferation and migration accompanying tissue remodeling during wound healing. On the other hand, the elevated pH values in wounds could be a signature of anomalies, linked to, for instance, bacterial infections or chronical wound cases [2, 8, 9]. Maintaining the pH values within an optimal range promotes faster and more effective healing, while any deviation can lead to infections, poor blood supply or chronic inflammation, etc. [8]. Oxygen is another key factor of wound healing [3, 4, 5, 10]. It is crucial in production of reactive oxygen species (ROS) during the inflammatory phase of wound healing, important against bacterial infections, and it contributes to reepithelialization by supporting the proliferation and migration of keratinocytes [11]. It also contributes to the hydroxylation of proline and lysine residues during synthesis of collagen, which provides structural integrity to the new tissue formed during the healing process [10, 12]. On the one hand, development of materials that control pH and oxygen levels can stimulate the wound healing processes [4, 5, 13, 14, 15, 16]. On the other hand, monitoring pH and oxygen is important in the wound assessment and management [17, 18, 19, 20, 21, 22, 23]. It is especially important as a measure to prevent bacterial infections and to estimate the efficacy of treatment of chronical wounds.
A variety of functional materials were developed for optical monitoring pH and oxygen [17, 18, 24, 25]. Recent advancements in optical pH sensing have introduced organic and inorganic materials in form solutions, thin films, immobilized nanoparticles, and responsive hydrogels [18]. They respond to pH changes by changing their absorption, luminescence, or refractive index. Here, one should mention fluorescent organic dyes that switch their emission intensity or color on the protonation or deprotonation [18, 26]. Classical example is fluorescein, which lights up on deprotonation. For instance, it was co‐encapsulated with reference red porphyrin dye into silica NPs to yield pH‐nanoprobes, which were further immobilized in agarose gel for monitoring bacterial growth [27]. Other important examples include rhodamine derivatives bearing piperazine units showing light up on the protonation [28] or amide derivatives that can undergo pH‐dependent spirolactam formation [29].
A variety of oxygen‐sensitive materials have been developed in form of nanoparticles, coatings and optical fibers [17]. In optical sensing modality, the key method for oxygen sensing is the use of phosphorescent dyes, which are quenched in their triplet state by the molecular oxygen. The examples include platinum(II) and palladium(II) benzoporphyrin complexes encapsulated into polymeric optodes [30], oxygen‐sensitive platinum(II) porphyrin derivative emitting in red in combination with reference green emitting naphthalimide in polyurethane thin films [31], or molecular dyads combining fluorescent donor with phosphorescent oxygen‐sensitive acceptor [32]. Previously, we developed a poly(methyl methacrylate‐co‐methacrylic acid) (PMMA‐MA) NPs, encapsulating rhodamine dye (reference) as FRET donor and oxygen‐sensitive Pt(II) phorphyrin as FRET acceptor [33]. Due to light‐harvesting properties of these NPs, it enabled efficient FRET with amplification of phosphorescence of the acceptor, allowing ratiometric oxygen sensing in cells with minimal phototoxicity. Using a similar FRET‐based concept, PMMA‐based NPs composed of BODIPY donor and Pt(II)photophyrin acceptor was developed and applied for ratiometric oxygen sensing inside spheroids [34]. Particularly challenging is to develop multifunctional sensors, which could combine detection of both pH and oxygen in one material. Previously, this was realized by combining three types of microparticles bearing oxygen‐sensitive platinum(II) porphyrin emitting in red, pH‐sensitive fluorescein isothiocyanate emitting in green and blue emitting reference dye diphenylanthracene. This composite material enabled three‐channel multiplexed detection of pH and oxygen using RGB camera, revealing high pH and hypoxia in a chronic wound [35].
Electrospun nanofibers are particularly attractive for preparation of sensors [36], because they present high surface area, biomimetic structure and mechanical properties, similar to the extracellular matrix and could eventually encapsulate or bear at their surface functional elements [37]. They have already been established as powerful materials for soft bioelectronics [38, 39], which includes, for instance, pressure sensors [40], wearable strain sensors [41], devices for human health assessment [42] and bioelectrical signal monitoring [43], etc. The development of optical sensors based on electrospun nanofibers is another rapidly developing research direction [44, 45, 46]. Previously, we showed that ultrabright fluorescent eletrospun nanofibers and nanorods can be prepared using cationic dyes with their bulky hydrophobic counterions [47, 48], which prevent dye self‐quenching [49, 50]. Electrospun nanofibers are particularly suitable for tissue engineering [37, 51, 52], which makes them promising materials for wound dressing [53, 54, 55].
The current challenge in the nanofiber sensors is to integrate FRET‐based principle, which would combine one common donor dye, serving as reference and two acceptors, one serving as pH probe, while another—as oxygen sensor. This configuration would enable the use of a single excitation of the donor and then simple RGB type detection at three channels for ratiometric multiplexed pH and oxygen sensing. However, in this system, the major difficulty is to achieve efficient FRET from nanofiber to both sensing acceptors. It will be particularly challenging to ensure efficient FRET to pH‐sensitive dye, because it is supposed to be grafted on the surface of nanofiber of 200–400 nm diameter, which creates very long distance from the encapsulated donor dyes far beyond the Förster radius. Our previous works on dye‐loaded polymeric nanoparticles [56] suggested that exciton migration within highly loaded dyes assembled with their bulky counterions enables energy transfer though distances far beyond the distances predicted by the Förster theory [57, 58]. This approach yielded nanoparticle‐based sensors for amplified sensing of nucleic acids [59, 60] and oxygen [33]. Here, we hypothesized that this exciton migration within reference donor dyes inside nanofibers would ensure efficient energy migration to the energy acceptors at the nanofiber surface, providing new type of FRET‐based nanofiber sensors.
In the present work, we report nanofiber‐based fluorescence ratiometric sensor operating by FRET principle for simultaneous detection of oxygen and pH. It is built from electrospun PMMA nanofibers encapsulating an energy donor based on blue cyanine dye (BlueCy) with bulky hydrophobic counterion (F5‐TPB). To provide oxygen‐sensing modality, they were encapsulated with phosphorescent Pt(II) phorphyrin dye, which provided oxygen‐sensitive phosphorescence in the red channel. On the other hand, to provide pH‐sensing modality, the surface of nanofibers was functionalized with pH‐sensitive rhodamine‐piperazine derivative. It was shown that sensitivity to pH required proper exposure of the dye away from the polymer matrix into the aqueous medium. Ultimately, combination of all three dyes yielded nanofibers presenting both oxygen and pH‐sensing capabilities, which operate in an orthogonal matter, using three detection channels, corresponding to standard RGB camera. This multiplexing nanofiber sensor installed on a commercial wound dressing was validated for pH and oxygen sensing on ex vivo tissue models. The developed nanosensor and the design principle will open the route to next generations of multiplexed nanofiber sensors for biomedical applications.
Results and Discussion
2
General Design Principle
2.1
Our nanosensor is designed based on polymeric nanofibers (Figure 1A), characterized by high surface area compared to films. To achieve a ratiometric response to the analyte, we exploit energy transfer between analyte‐insensitive energy donor (reference) and analyte‐sensitive energy acceptor at high donor/acceptor ratio. The donor dyes are cationic dyes with bulky hydrophobic counterion (F5‐TPB, Figure 1A), which allows high dye loading with minimized aggregation‐caused quenching (ACQ) and efficient energy transfer due to exciton migration (Figure 1B) [50, 56, 57]. In line with previous works on dye‐loaded polymeric nanoparticles, the present nanofiber system is expected to behave like light‐harvesting nanoantenna [57]. It allows efficient energy transfer from large number of donor dyes to relatively low number of the acceptor dyes (Figure 1B). This constitutes the basis for signal amplification of the analyte‐sensitive acceptor, which can help to improve sensitivity of the nanomaterial to the analyte [33, 60]. Moreover, efficient energy transfer is important in case of nanofibers in order to bring the excitation energy from the nanofiber interior to its surface located at distances far larger than the Forster radius (Figure 1B) [60]. In our design, the reference dye, being FRET donor is encapsulated inside nanofiber, whereas sensing dyes are either encapsulated inside the nanofiber in case of the oxygen sensor or grafted to its surface in case of pH sensor (Figure 1B). Single‐mode sensing nanofibers for detection of either pH or oxygen were designed first, followed by dual‐mode sensing device for multiplexed detection of pH and oxygen (Figure 1B).
Design of FRET‐based polymeric nanomaterials for sensing applications. (A) Scheme of preparation of luminescent nanofibers and nanorods for oxygen and pH sensing. Chemical structures of the polymer and dyes used. (B) Concept of energy transfer in the nanofibers for ratiometric detection of pH and oxygen.
Here, polymeric nanofibers were produced by electrospinning as bulk fiber mats that can be used as sensing surfaces. The fiber mats were then transformed into nanorods by ultrasonication [48], which enabled experiments in suspensions. Both fiber mats and nanorods were used as scaffolds for the nanosensor design (Figure 1).
Oxygen Sensing Nanofibers
2.2
In our oxygen sensor, two dyes coupled by FRET were encapsulated inside nanofibers during the electrospinning process: (i) oxygen‐insensitive (reference) donor dyes and (ii) phosphorescent dyes sensitive to oxygen through fluorescence quenching (Figure 1A). The latter were chosen from phosphorescent Pt‐porphyrin derivatives, known to be quenched in the presence of oxygen [17, 33]. Our primary choice of FRET pair was BlueCy/F5‐TPB donor and PtOEP O_2_‐sensitive acceptor (Figure 1A) at 100/1 molar ratio, which is optimized according to our previously reported oxygen‐nanosensor based on polymeric nanoparticles (NPs) [33]. Large excess of the donors vs acceptors is important to ensure signal amplification based on light harvesting, where a large number of donors collect efficiently the excitation energy and transfer to few sensor units producing so‐called antenna effect [57, 61]. The second FRET pair was R18/F5‐TPB donor with Pt‐TPTBP acceptor (Figure 1A), where both donor and acceptor are significantly red shifted with the latter emitting in near‐infrared region. For both donor dyes, bulky fluorinated counterion (F5‐TPB) was used, because according to our previous studies in polymeric NPs [49, 50] and nanofibers [47], it plays crucial role to ensure high dye loading with minimal aggregation‐caused quenching (ACQ) of the dyes. Moreover, it ensures efficient dye‐dye communication that is essential for efficient FRET to an acceptor dye [57, 59, 60].
The nanofibers were prepared by electrospinning and used in two forms: as fiber mats (Figure 2A–E) and nanorods (Figure 2F,G). The nanofiber mats containing both BlueCy/PtOEP and R18/Pt‐TPTBP pairs were successfully obtained and their fibrous structure was confined by epi‐fluorescence microscopy (Figure 2D). Then, they were converted into nanorods by ultrasonication, based on recently reported protocols [48]. Electron microscopy of the obtained sonicated samples confirmed the presence of nanorods of rather homogeneous diameter: 400 and 300 nm for BlueCy/PtOEP and R18/Pt‐TPTBP dye pairs, respectively (Figure 2F–I).
Electrospun nanofibers and nanorods from macroscopic to nanoscopic scale. (A) Samples of fiber mats obtained in 35 mm dishes under white light (B) Samples of fiber mats obtained in 35 mm dishes under UV lamp (365 nm) (C) Fiber mat loaded with R18 dye during electrospinning process. Cylindrical collector is 20 cm length with 10 cm diameter. (D,E) Respective epifluorescence images of BlueCy/PtOEP and R18/Pt‐TPTBP fibermats with 2× objective (scale bar 1 mm). Insert show higher magnification with 60x objective (scale bar 50 µm). (F,G) TEM micrographs of BlueCy/PtOEP (F) and R18/Pt‐TPTBP (G) nanorods. (H,I) Scale bar: 500 nm. Diameter distribution of BlueCy/PtOEP (H) and R18/Pt‐TPTBP (I) nanorods measured from TEM images.
The nanorods containing only BlueCy/F5‐TPB donor excited at 405 nm showed a characteristic single emission band of the dye centered around 480 nm (Figure S2). The presence of PtOEP at donor/acceptor ratio 100/1 led to appearance of intense narrow band at 650 nm (Figure S2). As a control corresponding to the absence of FRET, we mixed nanofibers containing only BlueCy/F5‐TPB with those containing only PtOEP. Under 405 nm excitation, the obtained emission spectra displayed mainly donor emission and only a very low emission of the acceptor (Figure S2). The other control nanorods containing only acceptor also showed poor emission of the acceptor (Figure S2). These results indicate that the direct excitation of the acceptor at 405 nm is poor. The excitation spectra of the nanofibers containing only acceptor confirmed relatively weak absorbance of the acceptor at 405 nm. Moreover, the molar concentration of the acceptor was 100‐fold lower compared to the donor, which all together explain almost negligible direct excitation of the acceptor at 405 nm excitation. Overall, these experiments suggest that the acceptor emission in nanorods containing both BlueCy/F5‐TPB and PtOEP originates from FRET. Nanorods encapsulating R18/F5‐TPB donor and Pt‐TPTBP acceptor also showed efficient FRET with strong emission of the Pt‐TPTBP, when the donor was excited (Figure S3).
The oxygen‐sensitive nanofibers encapsulating the two studied FRET pairs (BlueCy/PtOEP and R18/Pt‐TPTBP) were evaluated by measuring the emission spectra of corresponding nanorods suspensions in aqueous media with varying concentration of dissolved oxygen using an oxygen scavenger (Figure 3A,B). A good correlation was observed between the emission ratio acceptor/donor and the concentration in dissolved oxygen for the nanorods loaded with the two pairs tested. One should note that the response of both nanoprobes was not perfectly linear in the studied concentration range of dissolved oxygen (Figure 3C,D). Nevertheless, we expect that the oxygen quantification would be possible using nonlinear fitting of the calibration curve.
Fluorescence response of oxygen sensitive materials as nanorod suspensions in PBS. (A, B) Emission spectra at different concentration of dissolved oxygen of BlueCy/PtOEP and R18/Pt‐TPTBP pairs. (C, D) Variation with dissolved oxygen concentration of ratio between acceptor/donor maximum intensity R normalized by the acceptor/donor maximum intensity value for a dissolved O2 concentration of 0 mg/mL R0. (E,F) Reversibility of the oxygen sensing of the BlueCy/PtOEP and R18/Pt‐TPTBP pairs. Emission spectra of oxygen sensing nanorods after successive air and argon purging. (G,H) Ratio between acceptor/donor maximum intensity after successive air and argon purging. Excitation wavelegnth for BlueCy and R18 were respectively at 405 and 530 nm.
Next, we explored the reversibility of the spectral response of the nanosensors to oxygen by alternatively purging suspensions of oxygen‐sensitive nanorods in water with air and argon. For both BlueCy/PtOEP and R18/Pt‐TPTBP loaded nanorods, good reversibility of the acceptor/donor ratio response was observed (Figure 3G,H). Some deviations were observed for the first points in case of R18/Pt‐TPTBP nanoprobe, although its response stabilized for further cycles.
Then, we explored the sensing properties of the bulk fiber mats and their macroscopic optical properties (Figure 4). For this purpose, fiber mats containing different FRET pairs were conditioned in chambers respectively under ambient air and argon atmospheres and studied under a microscope with low magnifying objective (2×) to observe large areas. A fluorimeter connected to the microscope allowed us to record the spectra of the light emitted by the mats and validate the previous observations from suspensions of nanorods in water. The spectral data confirmed that both fiber mats showed ratiometric response to oxygen (Figure 4A). Next, the same samples were studied by a home‐built imager to obtain macroscopic images of the mats in both channels of the donors and the acceptors. From that, we compared the images of the two channels with ratiometric imaging by dividing acceptor channel by donor channel [5]. The obtained ratiometric images exhibited a significant variation in the color reflecting robust change in the donor/acceptor ratio (Figure 4B–F). In fact, the color change of BlueCy/PtOEP nanofiber sensor in response to oxygen was clearly visible by naked eyes under UV light excitation or using a standard smartphone RGB camera (Figure S4). Overall, the obtained results suggest that nanofibers loaded with BlueCy/PtOEP and R18/Pt‐TPTBP operate well as oxygen sensors, indicating that that the approach is quite generic and thus allows production of nanosensors with varied window of operation. The former system corresponds well to blue and red channels, respectively of the typical RGB camera, which will be important for simple detection protocols. On the other hand, the latter FRET couple, operating in the red to near‐infrared region would be more suitable for deep tissue imaging due to higher transparency of biological tissues to this wavelength region [62]. However, this is out of the scope of the skin applications, and therefore, further studies will be focused on the BlueCy/PtOEP FRET couple as oxygen sensor.
Ratiometric analysis of oxygen‐sensitive fiber mats at the macroscopic level. (A,D) Emission spectra recorded in air and argon atmosphere for fiber mats loaded with BlueCy/PtOEP (A) and R18/Pt‐TPTBP (D) pairs. Dashed rectangles represent bandpass filters used for ratiometric imaging displayed in (B,E). Ratiometric imaging of fiber mats loaded with BlueCy/PtOEP (B) and R18/Pt‐TPTBP (E) under air and argon atmospheres and LED irradiation. (B,E) Scale bar: 1 cm. Normalized ratiometric analysis is obtained from acceptor/donor ratio. (C,F) Ratio distributions obtained for ratiometric images of the fiber mats.
Nanofibers for pH Sensing: Design and Surface Functionalization
2.3
Then, we challenged our FRET‐based strategy for pH sensing. This modality is more complicated to implement as we need to place a pH sensitive unit at the nanofiber surface in order to achieve access to an aqueous medium. For this purpose, a methodology of surface modification of the nanofibers should be established. Moreover, to achieve ratiometric response to pH, FRET should be realized from the core of the nanofiber with encapsulated donor dyes to the pH‐sensitive acceptor at the surface. Here, we exploit excitation every transfer within encapsulated donors at high loading to achieve the transfer to the sufficiently long distance, as we previously shown for fluorescent nanoparticles [60].
First, the nanofiber production by the electrospinning method was optimized in order to tune nanofiber diameter. Initially, we needed thicker fibers of 1–2 µm range in order to observe surface functionalization directly by classical fluorescence microscopy. On the other hand, thinner fibers are more adapted for sensing application as it increases the surface to volume ratio. Moreover, thinner fibers would facilitate FRET due to reduced distance from the donors in the fiber core and acceptors at the surface.
To modify the surface of the polymeric nanofibers, we used hexamethylenediamine (HMDA) [63], which is expected to convert methyl esters of PMMA to amide and thus expose primary amine functions (Figure 5A). The latter were then modified with FITC as a model amino‐reactive pH‐sensitive dye. As an initial attempt, we performed such surface modifications by suspending nanofibers in aqueous mixtures with organic solvent (isopropanol). Even though the resulting materials were fluorescent, microscopy revealed homogeneous labeling including nanofiber core (Figure 5B, condition A), suggesting that FITC was deeply imbedded inside the nanofiber and not really located at its surface. Moreover, they showed poor sensitivity to external pH (Figure 5C). Our hypothesis is that the aqueous/organic mixture can produce polymer swelling, which leads to penetration of the HMDA and FITC inside the nanofiber and thus labeling of the nanofiber interior. For this reason, the major part of the pH‐sensitive FITC was shielded inside the polymer from the aqueous medium, preventing it from sensing pH.
Surface functionalization of nanofibers for grafting fluorescent dyes. (A) Procedures explored for nanofibers surface modification. (B) Fluorescence microscopy images of nanorods labeled with FITC in conditions A and B. Scale bar: 10 µm. (C) Normalized ratiometric pH titration of nanorods suspensions. Fluorescence intensity ratio (R) is measured by dividing the intensity recorded at 520/20 nm with excitation at 488/20 nm to that recorded at 520/20 nm with excitation at 400/20 nm. The ratio R is then divided to R0 and normalized to 1 at pH 2.
To prevent polymer swelling, the functionalization was carried out in aqueous buffer without any organic solvent. The resulting nanofibers exhibited a clear surface labeling without core labeling under fluorescence microscopy (Figure 5B, condition B). Most importantly, they were sensitive to pH as it was seen from the significant change in the intensity ratio as a function of pH (Figure 5C), which showed that surface localization of the dye is of primary importance for pH‐sensing by nanofibers. Nevertheless, the amplitude of pH response was lower than for fluorescein in solution (Figure 5C). We hypothesize that the PMMA nanofibers surface is a hydrophobic environment that may not allow the same accessibility and interactions with protons compared to bulk water. The second possibility is that even if the modified procedure in aqueous buffer allowed a functionalization of the fibers at the surface, the resolution of optical microscopy does not allow us to determine whether the fluorophore penetrated inside the polymer at distance <300 nm. Therefore, we replaced HMDA with a PEG_3_‐diamine, which is more polar than HMDA with a similar length in order to limit the reagent penetration inside the polymer. The obtained aminated nanofibers were successfully grafted at their surface with FITC (Figure S5). However, FITC‐nanofibers showed similar pH titration curve to that obtained for HDMA‐treated nanofibers (Figure 5C). This result suggests that in all these cases the modification took place at the nanofiber surface, but the sensitivity to pH is influenced by the apolar nature of PMMA surface. Moreover, these results showed that alternative diamines can be used for the surface functionalization of the PMMA nanofibers.
Synthesis of pH‐Sensitive Nanofibers
2.4
Then, to obtain pH‐sensing nanofibers operating by FRET we synthesized dye Rho‐pH (3) as pH‐sensitive acceptor. The latter is a N‐methylpiperazine derivative of rhodamine 1 already described in literature [28], operating by photoinduced‐electron transfer (PET) quenching mechanism. In acidic media, the methylpiperazine moieties are protonated, suppressing the PET quenching effect and lighting up the rhodamine (Figure S6). It was synthesized from rhodamine 1 in three steps, starting from preparation chloroanhydride of dye 1, then reaction with tBu‐protected N‐Methyl‐β‐alanine followed by tBu deprotection and transformation of the carboxylate into NHS ester (Figure S7). Using our HMDA‐based protocol, we obtained nanorods functionalized with this fluorescent pH‐sensitive dye, in line with fluorescence microscopy data (Figure 6A). Remarkably, the pH change from 8 to 2 resulted in a 60‐fold turn‐on and a hypsochromic shift of the emission maximum from 580 to 560 nm, suggesting good sensitivity of grafted dye to pH (Figure 7B). The titration in the pH range 2–8 revealed sigmoidal response of the nanoprobe, suggesting the optimal operation range for pH quantification between 7 and 3. Remarkably, the sigmoidal response to pH of the molecular probe Rho‐pH was slightly shifted to higher pH values compared to the nanoprobe, indicating that in the interface of nanoprobe provides some shielding effect for the sensing molecules from protons of bulk water, in line with data on FITC.
Imaging and pH sensitivity evaluation of nanofibers functionalized with pH sensitive dye Rho‐pH. (A) PMMA nanofiber modified on its surface with rhodamine‐based pH probe Rho‐pH in pH 2 buffer. Scale bar: 20 µm. (B) Fluorescence emission of a suspension of nanorods modified with Rho‐pH at pH 2 and 8. (C) pH‐titration of nanorods modified with Rho‐pH and molecular probe Rho‐pH in solution. Excitation wavelength: 520 nm.
Ratiometric pH sensing using fiber mats at the macroscopic scale. (A) Ratiometric macroscopic imaging of pH sensitive fiber mats at various pH values. Scale bar 5 mm. (B) Distribution of the acceptor/donor ratio obtained from the ratiometric image analysis. (C) Variation of mean values of the acceptor/donor ratio of the nanofiber sensor versus pH of the buffer applied. Two types of mean were used: max of the distribution (mean‐max ratio) or the mean of the entire distribution (mean ratio).
In the second step, nanofiber mats loaded with BlueCy/F5‐TPB donor were functionalized with the pH probe Rho‐pH. To prove that we do observe FRET from donor located inside 400‐nm thick fiber to the acceptor (pH probe) located at the surface, we measured fluorescence lifetime of the donor. The measured average lifetime decreased from 1.99 to 0.869 ns when pH probe was grafted (Figure S8), suggesting the FRET efficiency to the grafted acceptor of 56%. This high FRET efficiency through such long distances can be explained by exciton diffusion within donor dyes loaded at high concentration. Indeed, our previous works showed that at high donor dye loading the energy transfer can take place from 10 000 donor dyes to a single accepter within 60 nm particle, owing to the ultrafast exciton diffusion [57], while another study suggested that the exciton diffusion length in this type of nanomaterials can span to 70 nm, which was limited by the size of studied NPs [64].
As it was done for oxygen, we explored the pH sensing ability of the material at the level of the whole fiber mat. We recorded macroscopic fluorescence images of the mats at donor (blue) and acceptor (red) emission channels at various pH values of a buffer from 8 to 2 with 0.5‐unit steps. In the ratiometric images, the pseudo‐color reflecting acceptor/donor ratio changed gradually from blue two yellow–orange in the range from 7.5 to 4 (Figure 7A). According to quantitative image analysis presented in form of distribution of the ratio (Figure 7B) and mean ratio (Figure 7C), the ratio correlated well with pH in the range 4–7.5, suggesting that the nanoprobe could be used for measuring pH values with resolution down to 0.5 in the physiologically relevant pH range. Nevertheless, small heterogeneity of the ratio on the fiber mat surface was observed, which was probably related to varied acceptor functionalization level at different parts of the fiber mat. The color response of our nanofiber sensor to pH could be also seen by naked eye and by a RGB camera of a smartphone (Figure S4).
We also studied whether the fiber mat can detect a local variation of pH. For that, we added a drop of HCl on a pH sensitive mat under a microscope with a low magnifying objective in order to see the variation of fluorescence signal on a local area of the mat but not at the scale of a single nanofiber (Figure 8). We indeed measured an increase of the red vs blue signal centered at the place of the deposited acid. The observed pH gradient showed diffusion over 2 min timescale.
Two‐color imaging of pH distribution using fiber mats. (A) Top of pH sensitive fiber mat seen at 10× objective before and after been locally treated with 2 µL of 1.0 m HCl bottom corresponding ratiometric images. Scale bar 1 mm. (B) Intensity profile along white dashed line in BlueCy and Rhodamine channels. (C) ratiometric distribution of images.
Multiplexed pH/Oxygen‐Sensing Material
2.5
As we were able to sense pH and oxygen using nanofiber scaffold, we combined both sensing units in the same material. To this end, nanofibers loaded with both BlueCy/F5‐TPB and PtOEP were functionalized with the pH probe Rho‐pH. It is important that we used the combination of BlueCy/F5‐TPB, dye Rho‐pH and PtOEP, because BlueCy/F5‐TPB can be a FRET donor for the other two. Moreover, they contribute to blue (480 mm), yellow (560 nm) and red (650 nm), respectively (Figure S2), which correspond well to the spectral response of a typical RGB sensor (Figure S9). Fluorescence spectroscopy of nanorods in suspension was studied under excitation at 405 nm. The emission spectrum was composed of three bands corresponding to the signal of the three used dyes (Figure 9). Remarkably, the decrease in pH resulted in the rise of the 560‐nm band of pH probe without changes in the emission intensity of the oxygen sensor dye at 650 (Figure 9). Then, the suspension of nanorod sensors, purged sequentially with argon and ambient air, showed a good response to oxygen by the phosphorescent band of PtOEP (Figure 9). In the latter case, a slight increase of the Rho‐pH signal was observed after air purging. This effect might be due to a slight acidification of the media by the carbon dioxide present in ambient air.
Orthogonal luminescence response to pH and O2 of a suspension multiplexed sensing nanorods. (A) pH titration of oxygen and pH sensitive nanorods suspended in aqueous solutions. (B) Reponse to oxygen of the same material suspended in acidic media (pH 4). Excitation wavelength: 400 nm.
Overall, we developed all‐in‐one multiplexed sensing nanorods showing response to pH and oxygen in the orthogonal manner with minimal cross‐talk. As each emission band corresponds to red, green and blue region of the RGB camera, we expect that this material will enable multiplexed detection of pH and oxygen using the smartphone camera [65, 66].
Cellular Toxicity and Stability
2.6
With the perspective to apply the present sensing nanomaterials with cells and tissues, we performed cytotoxicity assays on HEK293T cell lines in presence of relatively high quantities of nanomaterials. We used the nanomaterials in form of nanorods suspensions in growth media and performed a MTT assay after 24 h incubation. No evidence of toxicity of these nanomaterials was observed, as none of the conditions gave statistically significant differences from the controls (Figure S10). To verify potential leakage of the dye components from the probe into biological medium, we incubated the multiplexed sensing nanorods in DMEM medium with 10% serum for 1 h at 37°C, separated them by centrifugation and re‐suspended them in the same medium. Comparison of their emission spectra with nanorods without incubation (Figure S11) suggested only very small changes in the FRET spectra of the nanorods. The emission of pH probe was systematically low because of neutral pH of the medium. Moreover, supernatant after centrifugation showed negligible emission, indicating rather low leakage of dye components into the serum‐containing medium. High stability against leakage in biological medium was previously observed for our nanoparticle‐based oxygen sensor that was fully functional even after long incubation inside live cells [33]. Remarkably, the same nanorod sensor after incubation in MilliQ water for 2 years at 4°C showed only minor changes in its three‐band emission (Figure S11, some changes in the oxygen content could explain this small variation), suggesting its excellent stability and long shelf life. The latter is an important characteristic of a medical device. In this case, it is particularly important that pH sensing unit was totally preserved (Figure S12) despite this long exposure to the aqueous medium (other components were encapsulated inside the polymer matrix and thus were well protected). Overall, these results suggest that nanosensors are expected to be non‐toxic and present good stability when applied to biological systems.
Sensing on a Skin Model
2.7
To demonstrate the potential of the developed nanomaterials as pH/oxygen sensors for wound healing, we modified commercial mainstream bandages in order to incorporate a transparent window supporting nanosensors (Figure 10A; Figure S13). These sensing bandages where then applied on a wound model (Figure 10B; Figure S13). For such purpose, we used raw pig skin supported on flesh, and we simulated wounds by making short cuts that were filled with aqueous media to modulate oxygen level with an oxygen scavenger or pH level using buffers (Figure 10A). First, we tested three individual nanofiber mat sensors: two oxygen sensors based on BlueCy/PtOEP and R18/Pt‐TPTBP FRET pairs and one pH sensor BlueCy/Rhro‐pH (Figure S14. In our first attempts, plain discs of fiber mats were used to produce sensing bandages. However, it was difficult to observe the optical response through such thick fiber mats (data not shown). Therefore, we modified the devices by applying and then peeling of nanofibers to obtain a thin layer of nanofibers adsorbed on the adhesive surface. Such a thinner deposit made the window transparent allowing a direct visualization of the wound (Figure 10B; Figure S14).
Validation of the multiplexed nanofiber sensor in a wound model. (A) Simplified scheme of developed sensing bandage and skin model sensing experiments. (B) Photo of a bandage modified with sensing fibers placed on a pork skin sample. The red circle highlights the imaging area. (C) Ratiometric images (acceptor/donor ratio) of the modified bandage with sensing nanofibers applied on a simulated wound in pig skin under 405 nm irradiation. Scale bar: 5 mm. Nanofibers used are simultaneously sensitive to pH and oxygen and are exposed to four different conditions: neutral (pH 7.4) in normoxia, acid (pH 4) in normoxia, neutral (pH 7.4) in hypoxia and acid (pH 4) in hypoxia. Excitation was 405 nm. Emission filters used were the following: 479/40 nm for BlueCy, 600/52 nm for Rho‐pH, and 690/50 PtOEP. (D) Corresponding distribution histograms of the ratiometric images.
Then, using home‐built imager, we recorded individual channels corresponding to reference (blue), pH probe (orange) and oxygen probe (red). The macroscopic ratiometric imaging of the skin samples covered with our sensing device was performed separately for the pH and oxygen sensors (Figure S14). Two types of experiments were made: (1) normoxia (cut filled with PBS) vs hypoxia (cut filled with sodium sulfite in PBS) and (2) neutral pH (cut filled with PBS) vs acidic pH (cut filled with acidic buffer, pH 4). The results revealed good ratiometric response of both oxygen sensors to varied oxygen level by changing the ratiometric fluorescence images and the quantitative analysis, presented as the ratio histograms (Figure S14). The pH sensor also showed clear change in the emission ratio with strong pseudo‐color switching (Figure S14).
Ultimately, we tested our all‐in‐one multiplexed nanofiber sensor for pH and oxygen using the same wound model. This sensor was deposited on a transparent disc and integrated into a commercial bandage, as explained above (Figure 10A), and then placed on the wound model (Figure 10B). The same four conditions were tested, where pH and oxygen concentration were varied within the wound model. Blue (reference), orange (pH probe) and red (oxygen probe) signals were recorded by the home‐built imager, yielding ratiometric images for pH and oxygen sensing modes. The multiplexed nanofiber sensor provided clear response to the changes in the oxygen and pH concentration by changing its pseudo‐color in the corresponding ratiometric images (Figure 10C) and distribution of the intensity ratio (Figure 10D). Importantly, the pH sensing modality reveled changes in pH independently of the level of oxygen and vice versa: oxygen sensing modality displayed response to oxygen in a similar way for the two different pH conditions (Figure 10C), which was in line with data obtained in suspensions (Figure 9). Thus, both analytes, important for wound monitoring, could be detected simultaneously and independently using our multiplexed nanofiber sensor with three‐channel (RGB) detection mode.
However, one should note that the ratiometric signal obtained for individual and dual‐mode nanofiber sensors was not perfectly homogeneous on the whole sensing surface which can be due to multiple factors: (a) an uneven distribution of sensing units within the nanofiber; (b) nanofiber aggregation; (c) uneven distribution of the nanofibers within the wound model and (d) complexity and heterogeneity on the skin.
Conclusion
3
In the present work, we developed bright fluorescent polymeric nanofibers and nanorods for ratiometric FRET‐based sensing of oxygen and pH. This strategy is shown to be highly versatile, as it is compatible with a large range of luminescent molecules such as cyanine, rhodamines and metallic complexes of porphyrins and phthalocyanines. Moreover, we explored two different approaches of fluorescence sensors encapsulating FRET donor (reference) dye: (1) encapsulating the FRET acceptor (sensor) dye inside the nanofibers in case of oxygen nanosensor and (2) covalent functionalization of the polymeric nanofibers at their surface with acceptor dyes in case of pH nanosensor. While the former approach is suitable for sensing dissolved gases, the latter extend the compatibility of the nanosensor to analytes in aqueous medium unable to diffuse inside the polymer matrix. Moreover, we combined both modalities in order to obtain multiplexed sensors of both oxygen and pH. All these materials in form of nanorod suspensions and nanofiber mats provide robust ratiometric response to oxygen and pH, according to florescence spectroscopy and three channel (RGB) microscopy and macroscopy. In case of pH sensing, the importance to exposure of pH probe at the nanofiber surface to the aqueous environment was demonstrated. The developed nanofiber sensors showed negligible cytotoxicity and proved to detect oxygen and pH levels on skin model of a wound. Multiplexed nanofiber sensor was shown to detect both pH and oxygen in an orthogonal way without cross‐talk using three channel (RGB) fluorescence detection. We expect that the developed advanced materials will be of particular interest as biosensing devices for wound healing applications, as they sense relevant parameters such as pH and oxygen levels in a biological matrix. Their emission in the three colors corresponding to the red, green and blue channels of RGB color cameras makes them compatible with simple detection using a smartphone. Based on the diversity offered by both co‐encapsulation and surface functionalization approaches, the present strategy can be extendable to other type of analytes and other spectral windows, which open a large range of potential biomedical applications.
Material and Methods
4
All of the chemicals and solvents were purchased from either Sigma–Aldrich or TCI or Carlo Erba and used as received without further purification. MilliQ‐water (Millipore) was used in all experiments. Nuclear magnetic resonance (NMR) spectra were recorded at 20°C on a Bruker Advance III 400 spectrometer. Mass spectra were obtained using an Agilent Q‐TOF 6520 mass spectrometer.
BlueCy F5‐TPB was prepared by following our previous reports. N‐methylpiperazine derivative of rhodamine 2 was synthetized by Buchwald‐Hartwig amination from fluorescein bis triflate as reported by Grim and Lavis [67] and then further modified as described in SI to provide activated pH probe Rho‐pH (3).
Nanofiber Preparation
4.1
Polymeric nanofibers were produced by electrospinning adapting the procedure we previously reported [47] to present fluorophores: The polymer solution for electrospinning was prepared by dissolving 666 mg of PMMA (30 wt.%, Mw = 120 000) in 2 mL of DMF overnight. The resulting viscous solution was mixed with desired concentrations of dyes in weight percentage regarding the mass of polymer (Table 1). The final concentration of the polymer in DMF was adjusted to 22–30 wt.%. Electrospinning was performed on a FLUIDNATEK LE‐10 electrospinning equipment (Bioinicia). Then, 1.5 mL of the above solution was loaded into a 5 mL syringe with an 18 G needle. A constant flow rate of 1.5 mL/h was maintained throughout with the help of a syringe pump. A voltage of 18 kV was applied to the syringe needle, and an aluminum foil was used as the collector. The distance between the needle tip and the collector was adjusted to 15 cm. The fiber mat obtained after the electrospinning was dried in an oven at 80°C for 2 h. It was further dried under high vacuum overnight (Table 1).
Nanorods Preparation
4.2
Portions of electrospun fibermats were collected and weighted (typically around 10 mg). The fibers were soaked with a methanol/water mixture (40:60vol%) for 5 min and then washed two times with water. Then, fibers suspended in water were sonicated (Branson 150 sonicator probe 4C15) for 2 × 6 min at 60% power in continuous mode in an ice bath to avoid over heating of the samples.
Fluorescence Spectroscopy
4.3
Fluorescence spectra of nanorods suspensions were recorded on an Edinburg FS5 spectrofluorometer. Fluorescence spectra of fibermats were measured using a Nikon Ti‐E inverted microscope with a CFI Plan Apochromat Lambda D 2X (NA = 0.1) objective, connected to a QEPro (OceanOptics) fluorimeter with a C‐mountSMA‐905 adaptor via a 200 µm VIS/NIR optical fiber (Ocean Optics). Excitation was performed with a SPECTRA X LIGHT ENGINE (Lumencore) source.
Fluorescence Lifetime Measurements
4.4
Time‐resolved fluorescence measurements were performed using time‐correlated single‐photon counting technique. Excitation pulses at 405 nm were generated by a supercontinuum laser (NKT Photonics SuperK Extreme) with 10 MHz repetition rate. The fluorescence decays were collected at 490 nm, using a polarizer set at magic angle and a 16 mm band‐pass monochromator (Jobin Yvon). The single‐photon events were detected with a microchannel plate photomultiplier R3809U Hamamatsu, coupled to a pulse pre‐amplifier HFAC (Becker‐Hickl GmbH) and recorded on a time‐correlated single photon counting board SPC‐130 (Becker‐Hickl GmbH). The instrumental response function (IRF) recorded with a polished aluminum reflector showed a full‐width at half‐maximum of 70 ps for 532 nm excitation. Time‐resolved exponential decays were fitted by using the global fit procedure of Igor Pro (Wavemetrics). The fitting function was a sum of exponential decays (up to 4 components) convolved with a normalized Gaussian curve of standard deviation 𝜎 standing for the temporal IRF and a Heavyside function. All emission decays were fitted using a weighting that corresponds to the standard deviation of the photon number squared root.
Fluorescence Imager Setup
4.5
A home‐built fluorescence imaging system was used to capture the fluorescence macroscale images of electrospun fiber mats. Imaging setup consists of a motorized filter wheel (ThorLabs FW102C) equipped with 479/40, 530/43, 600/52 and 650/90 band‐pass filters (BrightLine HC, AHF) to collect the emitted light. Scientific CMOS camera (ThorLabs sCMOS quantalux CS2100M‐USB) was used to capture the images. All the images were acquired from the camera at room temperature with 50 to 150 ms exposure time using ThorCam software. Excitation was performed using external light source such as 405 nm LED light or 530 nm LED array (ThorLabs). ImageJ software was utilized to analyze the images.
Nanomaterial Handling
4.6
Fiber mats were cut in small portions with a scissor, typically into squares of 1–2 cm length or in 12 mm discs using biopsy punches.
Surface Functionalization
4.7
Protocol A. Polymeric materials were suspended in a mixture of ^i^PrOH/H_2_O (50:50 vol.) containing 30 mg/mL of HMDA for 30 min at RT. Then, polymeric materials were washed with H_2_O until reaching neutral pH. Covalent grafting of fluorescent species was then performed for 2 h at room temperature in water/acetonitrile (50:50 vol.) mixture in presence of 1 mg/mL of a reactive dye. Protocol B. Polymeric materials were suspended in a 100 mm carbonate buffer (pH 9.4) containing 30 mg/mL of suitable amino derivative and shake at 55°C overnight. Then, polymeric materials were washed with H_2_O until reaching neutral pH. Covalent grafting of fluorescent species was then performed for 2 h at room temperature in 100 mM carbonate buffer (pH 9.4) in presence of 1 mg/mL of a reactive dye.
Fluorescence Microscopy
4.8
Microscopy imaging was performed with a Nikon Ti‐E inverted epi‐fluorescence microscope, equipped with following objectives: CFI Plan Apochromat Lambda D 2X (NA = 0.1), CFI Plan Fluor 10X (NA = 0.3), CFI Plan Apochromat Lambda D 20X (NA = 0.75), CFI Plan Apo × 60 oil (NA = 1.4), CFI Apo TIRF 100XC Oil (NA = 1.49) and a Hamamatsu Orca Flash 4 sCMOS camera. The images were recorded using NIS Elements.
Dissolved Oxygen Titration
4.9
To measure fluorescence responses of O_2_ sensing materials, a bottle of PBS with a magnetic stirring bar was kept at constant mixing rate below the speed that would generate a vortex, with an optical dissolved oxygen (DO) probe (Vernier) connected to a LabQuest2 interface (Vernier) plunged inside the buffer to measure dissolved O_2_ concentration. The pO_2_ was gradually reduced by stepwise addition of a freshly prepared solution of sodium sulfite at 1 mol/L in PBS. After each addition, sufficient was waited so the displayed DO concentration stabilized. The corresponding value was noted down and immediately, 950 µL of the resulting solution were added to 50 µL of a 1 mg/mL suspension of O_2_ sensing nanorods in PBS in a spectrometry cuvette, immediately sealed with a piece of parafilm. Right after, the fluorescence spectra of the resulting sample were measured by a spectrofluorometer. Obtained spectra were normalized to the maximum value of the donor emission band.
Oxygen Sensing Reversibility
4.10
Suspensions of oxygen sensitive nanorods at 50 µg/mL in PBS in semi‐micro spectrometry quartz cells were purged alternatively with 20 mL of ambient air or argon with a syringe through a 23G needle immersed in the solution. After purging, the cells were closed with a Teflon cap and the spectra immediately record. After air purges, cuvettes were shaken to resuspend nanorods accumulating on the wells at the edge of liquid level due to bubbling process. Obtained spectra are normalized to the maximum value of the donor emission band.
Sensing of pH
4.11
Sensing of pH was performed in buffers prepared from solution A (boric acid 0.2 m + citric acid 0.05 m) and solution B (tertiary sodium phosphate 0.1 M) as reported by Carmody [10] and pH was adjusted with pH meter (Mettler Toledo FiveEasy with InLab Expert Pro probe calibrated with Mettler calibrations standard at pH 4.01, 7.00, 9.21). Titration of FITC functionalized nanorods and pH+O_2_ sensitive nanorods were performed in polystyrene black 96 wells microplate lidless using a Tecan Spark plate reader. Spectra from fluorescence plate reader where corrected by using PMMA‐MA nanoparticules loaded with 12wt.% of BlueCy F5‐TPB respect to polymer, R18 F5‐TPB (1/1000 molar ratio respect to BlueCy) and PtOEP (1/100 molar ratio respect to BlueCy), formulated by nanoprecipitation in phosphate buffer 20 mm, pH 7.4. The fluorescence spectra of those particles were measured with both Edinburgh spectrofluorimeter (with spectral corrections) and Tecan Spark plate reader. The ratio of both normalized spectra was used as correction function.
Cell Culture
4.12
HEK293T (ATCC) were grown according to ATTC recommendations in DMEM media (Gibco Invitrogen) supplemented with 10% of decomplemented FBS, 200 mM of L‐glutamine, Sodium Pyruvate, phenol red and pen‐strep cocktail under 5% CO_2_, 37°C and controlled humidity.
Cytotoxicity
4.13
Cells were seeded in transparent TPP 96‐wells microplate at 5000 cells/wells. After 48 h to let cells recover from passage stress, growth medium was replaced by fresh medium containing 50 µg/mL of nanorods. Positive control was performed by replace half of the medium by DMSO and negative control was done with non‐treated cells. Each condition was represented by 8 wells each. After 24 h, media was replaced by fresh media supplemented with 1 mg/mL of MTT. Blank samples were prepared for each condition with cells, growth media and corresponding nanorods but without adding MTT to correct the background of phenol red and nanomaterials. After 4 h incubation, medium was removed and 100 µL of DMSO per wells was added to dissolve formazan crystals. After 10 min orbiting at 300 rpm, absorbance at 570 nm was measured with Tecan Spark plate reader. The values were normalized between 0 (positive control mean value) and 100 (negative control mean value) and statistic test performed with Graphpad Prism 5 for all pairs (Benett test) for P < 0.001.
Stability in Biological Media
4.14
A suspension of 3 colors nanorods (nanorods loaded with BlueCy and PtOEP and grafted with rhodamine Rho‐pH) was diluted in DMEM media (Gibco Invitrogen) supplemented with 10% of decomplemented FBS, 200 mm of L‐glutamine and Sodium Pyruvate without phenol red. The resulting suspension was incubated for 1 h at 37°C. Then, fluorescence emission spectra were measured (excitation at 405 nm) before and after centrifugation to detect eventual presence of leaked dyes. As the medium was naturally fluorescent, its emission spectrum acquired with the same parameters was subtracted.
Sensing Bandages and their Ex Vivo Application
4.15
Sensing bandages were produced by punching a 12 mm hole in brown mainstream bandages and punching 12 mm discs of a transparent silicon double sided tape (approx. 2 mm thick). The obtained discs were positioned in bandages holes and mounted together with a small piece of transparent adhesive on the exterior side. Then, punched 12 mm discs of nanofibers mat were applied on the silicon sticking side and gently pressed. Then, the nanofibers discs were peeled off to keep only the finest layer of fibers at the surface. For the wound model, we obtained raw pork knuckle from mainstream shop and cut in portions of approximate rectangles of 8 × 3 cm with 2 cm thickness. The skin was first cleaned with a compress conditioned with saline solution. A cut of approximatively 1 cm was made in the skin with a blade and the generated hole was filled with 100 µL of an aqueous solution to simulate wound condition. For normoxia and neutral condition, PBS was used. For hypoxia, PBS was supplemented with 100 mM of sodium sufite. For acidic conditions, the buffer at pH 4 was used.
Statistical Analysis
4.16
Nanorods size are measured by TEM and distributions are represented as histograms in Supporting Information and size are provided as mean ± SEM (n = 200) in Figure S1 legend. For MTT assays, data were normalized by subtracting the mean value measured for the positive control (DMSO treatment) and then divided by the mean value of the negative control (non‐treated) and multiply by 100 to represent a normalized viability between 0 and 100 expressed in percentage. MTT data were represented as bar plot showing mean± SEM of normalized viability (n = 8). P‐values were calculated with graphpad Prism 5 for all pairs (Benett test) for P < 0.001.
Conflicts of Interest
A.S.K. and R.P. deposited patent applications on the described technology: EP 25306012.3 and EP 25306011.5.
Supporting information
Supporting File: adhm70605‐sup‐0001‐SuppMat.docx.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1G. Power , Z. Moore , and T. O'Connor , “Measurement of p H, Exudate Composition and Temperature in Wound Healing: a Systematic Review,” Journal of Wound Care 26 (2017): 381–397, 10.12968/jowc.2017.26.7.381.28704150 · doi ↗ · pubmed ↗
- 2S. L. Percival , S. Mc Carty , J. A. Hunt , and E. J. Woods , “The Effects of p H on Wound Healing, Biofilms, and Antimicrobial Efficacy,” Wound Repair and Regeneration 22 (2014): 174–186, 10.1111/wrr.12125.24611980 · doi ↗ · pubmed ↗
- 3S. Schreml , R. M. Szeimies , L. Prantl , S. Karrer , M. Landthaler , and P. Babilas , “Oxygen in Acute and Chronic Wound Healing,” British Journal of Dermatology 163 (2010): 257–268, 10.1111/j.1365-2133.2010.09804.x.20394633 · doi ↗ · pubmed ↗
- 4Y. Li , R. Z. Fu , Z. G. Duan , C. H. Zhu , and D. D. Fan , “Artificial Nonenzymatic Antioxidant M Xene Nanosheet‐Anchored Injectable Hydrogel as a Mild Photothermal‐Controlled Oxygen Release Platform for Diabetic Wound Healing,” ACS Nano 16 (2022): 7486–7502, 10.1021/acsnano.1c 10575.35533294 · doi ↗ · pubmed ↗
- 5H. H. Chen , Y. H. Cheng , J. R. Tian , et al., “Dissolved Oxygen from Microalgae‐Gel Patch Promotes Chronic Wound Healing in Diabetes,” Science Advances 6 (2020): aba 4311.10.1126/sciadv.aba 4311 PMC 722876132440554 · doi ↗ · pubmed ↗
- 6L. A. Schneider , A. Korber , S. Grabbe , and J. Dissemond , “Influence of p H on Wound‐Healing: a New Perspective for Wound‐Therapy?,” Archives of Dermatological Research 298 (2007): 413–420, 10.1007/s 00403-006-0713-x.17091276 · doi ↗ · pubmed ↗
- 7S. Schreml , R. J. Meier , O. S. Wolfbeis , M. Landthaler , R.‐M. Szeimies , and P. Babilas , “2D Luminescence Imaging of p H in Vivo,” Proceedings of the National Academy of Sciences 108 (2011): 2432–2437, 10.1073/pnas.1006945108.PMC 303870021262842 · doi ↗ · pubmed ↗
- 8S. Schreml , R. J. Meier , M. Kirschbaum , et al., “Luminescent Dual Sensors Reveal Extracellular p H‐gradients and Hypoxia on Chronic Wounds that Disrupt Epidermal Repair,” Theranostics 4 (2014): 721–735.24883122 10.7150/thno.9052 PMC 4038754 · doi ↗ · pubmed ↗