Implantable and bioresorbable label-free biosensors: leveraging natural and synthetic receptors in nanoporous silica optical platforms

Implantable and Bioresorbable LabelFree Biosensors: Leveraging Natural and Synthetic Receptors in Nanoporous Silica Optical Platforms Giuseppe Barillaro Information Engineering Dpt University of Pisa [email protected] Photo credit: Beckman Institute, University of Illinois and Tufts University Credits to: Prof. J. Rogers’s Group J.-K. Chang, et al. , Proc. Natl. Acad. Sci. USA 2017, 114, E5522. Bioresorbable sensing systems Why we do need bioresorbable, implantable systems? Corsi et al, Nature Review Electronics Engineering (2025), accepted Physical sensors Chemical sensors Encapsulation Dielectric Conductor Semiconductor Substrate PLGA PVA Silicon Silicon ZnO Germanium Magnesium Zinc Molibdenum Cellulose SiO2 Collagen PLGA SiO2 Architecture and challenges in bioresorbable sensors S. W. Hwang, et al., Nano Lett. 2015, 15, 2801. C. M. Boutry, et al. Nat. Biomed. Eng. 2019, 3, 47. G. A. Salvatore, et al., Adv. Funct. Mater. 27, (2017). H. S. Kim et al.. Adv. Healthc. Mater. 2018, 7. 58 Figure 4.7: Characterization of PSi and PSiOx scaffolds. (a) Interformetric reflectance spectroscopy of PSi and PSiOx samples. Left: typical reflectance spectra of the fabricated PSi and PSiOx samples. Middle: porosity and pore thickness evaluated from reflectance spectroscopy measurements. Average porosity 74.9%, average thickness 3.89 μ m. Right: Variation in EOT and MaxFFT resulting from the oxidation process. (b) Characterization of the morphology of PSi samples through SEM. Left: representative SEM image in top-view of a PSi sample, magnification 150000X. The porosity calculated from the analysis of the image is around 70.9%. Middle: distribution of the equivalent radii of the pores evaluated through the analysis of the SEM image in top-view. Right: representative SEM image in cross-section of PSi sample, magnification 35000X. The thickness of the porous layer results to be around 4.7 μ m. (c) Characterization of the morphology of PSiOx samples through SEM. Left: representative SEM image in top-view of a PSiOx sample, magnification 150000X. The porosity calculated from the analysis of the image is around 62.9%. Middle: distribution of the equivalent radii of the pores evaluated through the analysis of the SEM image in top-view. Right: representative SEM image in cross-section of PSi sample, magnification 35000X. (d) Characterization of the surface chemistry of PSi and PSiOx samples through FTIR (left) and zoom of the main absorption regions of PSi (middle and right). 1 cm + _ 1 cm Why nanostructured porous silica membranes? 58 Figure 4.7: Characterization of PSi and PSiOx scaffolds. (a) Interformetric reflectance spectroscopy of PSi and PSiOx samples. Left: typical reflectance spectra of the fabricated PSi and PSiOx samples. Middle: porosity and pore thickness evaluated from reflectance spectroscopy measurements. Average porosity 74.9%, average thickness 3.89 μ m. Right: Variation in EOT and MaxFFT resulting from the oxidation process. (b) Characterization of the morphology of PSi samples through SEM. Left: representative SEM image in top-view of a PSi sample, magnification 150000X. The porosity calculated from the analysis of the image is around 70.9%. Middle: distribution of the equivalent radii of the pores evaluated through the analysis of the SEM image in top-view. Right: representative SEM image in cross-section of PSi sample, magnification 35000X. The thickness of the porous layer results to be around 4.7 μ m. (c) Characterization of the morphology of PSiOx samples through SEM. Left: representative SEM image in top-view of a PSiOx sample, magnification 150000X. The porosity calculated from the analysis of the image is around 62.9%. Middle: distribution of the equivalent radii of the pores evaluated through the analysis of the SEM image in top-view. Right: representative SEM image in cross-section of PSi sample, magnification 35000X. (d) Characterization of the surface chemistry of PSi and PSiOx samples through FTIR (left) and zoom of the main absorption regions of PSi (middle and right). www.advancedsciencenews.com www.advancedscience.com Figure 3. Long-term operation of the bioresorbable pH sensor. a) Normalized photoluminescence intensity (red, at 580 nm) and mass of the porous scaffold (black) measured over ≈100 h on apH sensor in buffer at pH 7.4 and 37 °C. b) Calibration curve (photoluminescence intensity at 580 nm versus pH value) of thepH sensor in (a) measured over the pH range 7.4–4 at t= 0, 50, and 95 h (n= 3 full cycles). c) Sensitivity value of the pH sensor in (b). d) Mass percentage of nPSiO2scaffolds oxidized in different conditions measured over time until full dissolution. The raw data are best-fitted using a piecewise linear model (dotted traces). e) Dissolution rate of the Si O2capping layer and silicon skeleton of the nPSiO2scaffolds in (d) (n= 4 samples). f) Full dissolution time of the nPSiO2scaffolds in (d) (n= 3 samples). g) Time-resolved normalized photoluminescence intensity (at 580 nm) of pH sensors on native Si in a flow cell and on PLGA foil under synthetic skin. h) Degradation rate (after 110 h of stable operation) of the photoluminescence of pH sensors on native Si in flow cell and on PLGA foil under synthetic skin (n= 3 samples). i) Photoluminescence stability and degradation times of pH sensors on native Si in flow cell and on PLGA foil under synthetic skin (n= 3 samples). Data are presented as mean (± s.d). of a bare nPSiO2scaffold results in a signal whose intensity is proportional to the mass of the porous scaffold (Figure S13, Supporting Information). Given that fluorescence emission of the sensor is stable for atime of 100 h, at least, one can assum e that the polymer stack isalso stable. Thus, changes occurring in the intensity of the FFT signal with time can be directly linked to the dissolution of the porous scaffold in the pH sensor.[54] We found that dissolution of the nPSiO2scaffold occurs in ≈50 h (Figure 3a). Remarkably, fluorescence stability (Figure 3a) and sensi ng properties (Figure 3b) of the polymer stack ar e not affected by the scaffold dissolution. The pH sensor reduces to a 6-nanometer-thick polymer brush with 5-µm-long wire bristles after dissolution of the silica scaffold occurring in ≈2days, further lessening invasiveness and improving adaptability of the sensor to tissue, yet retaining its mechanical integrity and in turn PL and sensing properties for at least 2more days (Figure S14, Supporting Information). The dissolution time achi eved for the porous scaffold in the pH sensor (Figure3a) was consistent with that of a bare nPSiO2 scaffold oxidized at 1000 °C for 5 min (Figure 3d). No statistical differenceswereobserved between dissolution timesof polymercoated and bare scaffolds (FigureS15, Supporting Information). Control experiments on dissolution were carried out on bare nPSiO2scaffoldsoxidized in different conditions, in buffer at pH 7.4 and 37 °C (Figure S16a–c, Supporting Information). A progressive reduction of both fringe contrast and the effecti ve optical thickness (EOT) values of the reflectance spectrum was observed with time (Figure S16d–f, Supporting Information). The FFT peak intensityretr i eved from thereflectancespectrum of the bare nPSiO2scaffold (i.e., without any polymer coating) wasused to estimate the silicon content in the scaffold over time. The dissolution kinetics of nPSiO2was best-fitted with a piece-wise linear model with two slopes, from which it was possible to extrapolate the dissolution rates of the SiO2capping layer (first slope) Adv.Sci . 2022, 2202062 © 2022 The Authors. Advanced Science published by Wiley-VCH GmbH 2202062 (6 of 10) Tunable pore size to filter out unwanted guests in vivo 50 nm Complex optical structures can be easily engineered Very high specific surface M. Corsi, A. Paghi, et al., Advanced Science 2202062 (2022) Biofunctionalization approaches of nanostructured silica membranes Layer-by-layer polyelectrolytes covalent chemistry molecularlyimprinted polymers (MIPs) ---- nanoparticles biomolecules T. Di Giulio et al., Small Methods 2401315 (2025) T. Di Giulio et al., ACS Appl Mater Interf , 17, 12663 (2025) T. Di Giulio et al, ACS Appl Mater Interf 16, 43752 (2024) E. Mazzotta et al, Small 19 , 2302274 (2023) E. Maurina et al, Adv Sensor Research 2400090 (2024) A. Paghi et al, Adv Materials Technologies 8, 2201600 (2023) M. Corsi et al., Adv Science 9, 2202062 (2022) S. Mariani et al, ACS Appl Mater Interf 11, 43731 (2019) S. Mariani et al, Nature Communications 9, 5256 (2018) M. Corsi et al., Sci. Adv. 11, eads0265 (2025) T. Di Giulio et al, ACS Appl Mater Interf 16, 43752 (2024) S. Mariani et al, ACS Sens. 3, 595 (2018) S. Mariani et al, Anal. Chem. 88, 8502 (2016) S. Mariani, ACS Sens. 1, 1471 (2016) In-vivo tracking of doxorubicin concentration n=9 mice n=9 mice Corsi et al., Sci. Adv. 11, eads0265 (2025) In-vivo biocompatibility over 3 months Corsi et al., Sci. Adv. 11, eads0265 (2025) Skin histology and blood analysis at 3 months Corsi et al., Sci. Adv. 11, eads0265 (2025) Wearable electronic patch for wireless readout M. Corsi, A. Paghi, et al., Advanced Science 2202062 (2022)Corsi et al., Sci. Adv. 11, eads0265 (2025) MIPs in nanoporous silica as synthetic receptors for HHb www.advancedsciencenews.com www.small-journal.com Figure 1. Polypyrrole (PPy) vapor-phase deposition on nanostructured porous silicon oxide (PSiO2)scaf folds.a)Sketchof the vapor-phasepolymerization process, starting from as-made PSi (1), which is thermally oxidized to PSiO2(2), exposed to the oxidizing agent FeCl3(3), and finally coated with PPy in achamber saturated with pyrrole vapors (4) for different time intervals. b) Effective optical thickness changes (EOT-EOTPSi O2)of PSiO2 scaffolds coated with PPy through vapor-phase and liquid-phase polymerization, for different timeintervals; theEOT value of thebarePSiO2(EOTPSiO2) scaffold is used as reference (n = 3 samples for each tested time). c) SEM top-view images of as-made PSi and of PPy-coat ed PSi O 2scaffolds after 5 h liquid-phase polymerization and 8 hvapor-phase polymerization. Insets report pore size distribution. Scale bar is 200 nm. d)Porosity value ofthe top surface of as-made PSi and PPy-coated PSi O 2scaffolds after 5 h liquid-phase polymerization and 8 h vapor-phase polymerization (n = 3 samples), estimated from SEM top-view images. e) SEM cross-section view images of full thickness (left) and pore bottom (right) of a PSi O 2scaffold after PPy vapor-phase deposition for 8h. Scal e bare is 1 µm(left) and 200 nm(right). f)DetailedC1s andN1sXPSsignals of PPy depositedon PSiO2scaffolds by vapor-phase deposition for 8 h. Spectra are fitted and charging corrected. All data are presented as mean (± s.d). vapors within thenanostructure and, in turn, their adsor pti on on the nanostructure inner surface, as widely reported for gas diffusion through nanoporous membranes.[91–93] This leads to the vapor-phase deposition of PPy films with higher reliability and uniformity within the PSiO2scaffolds compared to liquid phase for a given polymerization time, consistently with data shown in Figure1b. Furthermore, liquid-phase polymerization isprone to the formation of oligomers and (nano)aggregates due to partial dissolution of the oxidant agent upon immersion of the FeCl3decorated PSiO2scaffold in the monomer solution. Diffusion of oligomers and (nano)aggregates within the nanopores isfurther hindered, contributing in preventing polymer deposition on the inner pore surface.[94,95] Figure 1c shows SEM top-views of PSiO2scaffolds as-m ade and coated with PPyfor 5h in liquid phaseand 8h in vapor phase. The scaffold coated with PPy for 8 h in vapor phase closely resembles the as-m ade scaffold, with si mi lar aver age size and size distribution of pores (inset in Figure1c and FigureS1, Supporting Information). Aporosity reducti on of only 10% is achi eved for the 8 h coated scaffold with respect to the as-m ade scaffold, which corresponds to an aver age reducti on of the pore diameter of ≈4 nm (Figure 1d). These data are consistent with the homogeneous deposition of a few nm-thick polymer on the inner surface of the pores. Conversely, liquid-phase deposition for 5 h leads to the formation of a PPy layer on top of the PSiO2 scaffold with low porosity (10%) almost occluding the pores inlet (Figure 1c,d). This is consistent with the smaller EOT values achi eved for liquid-phase PPy polymerization, given that most of thepolymer isdeposited on top of thePSiO2scaffold thusreducing the materials coating the inner surface of the pores. Figure 1e shows aSEM cross-section view of the PSiO2scaffold af ter vapor -phase deposition of PPy for 8h. Columnar pores with sam e morphology of that of the as-m ade scaffold ar e visible (compare Figure 1e; Figure S3b,c, Supporting Information), Sm al l 2023, 2302274 © 2023 The Authors. Small published by Wiley-VCH GmbH 2302274 (3 of 9) 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202302274, Wiley Online Library on [29/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License E. Mazzotta et al., Small, 2302274 (2023) MIP synthesis www.advancedsciencenews.com www.small-journal.com Figure 2. Vapor-phase synthesis of PPy-based molecularly imprinted polymer (MIP) for human hemoglobin (HHb) on PSiO2scaffold. a) Sket ch of the imprinting protocol consisting of (1) preliminary PSiO2silanization by APTESand (2) linking with glutaraldehyde for (3) HHb coupling, followed by (4) PPy vapor-phase polymerization for 8 h and (5) final washing treatment for HHb removal aiming at obtaining the imprinted cavities within the PPy film. b) Effective optical thickness changes (EOT-EOTPSiO2)achievedfor eachfunctionalization stepin(a);the EOTvalue ofbarePSiO2(EOTPSiO 2) scaffold is used as reference (n = 3 samples). Data are presented as mean (± s.d.). c) Detailed N 1s XPS signal of MIPfor HHb deposited on PSiO2scaffolds by 8 h vapor-phase polymerization. Si gnal is fitted and charging corrected. with no cloggi ng signs over depth. The presence of the polymer in the pores over depth is corroborated by SEM -EDX an al yses carried out on the cross-section of PSiO2scaffolds after PPy deposition (FigureS4, Supporting Information). Nitrogen distribution over the pore depth is selected as the fingerprint of PPy deposition. FigureS4 (Supporting Information) reveals that vapor - phase polymerization for 8 h results in a homogeneous polymer distribution within the porous scaffold. Conversely, a PPy mass decreasing over depth resultsfrom EDX analysis after deposition for 5 h in liquid phase. With the ai m of gai ni ng adeeper understanding of vapor - phase deposition of PPy, we car r i ed out XPS an al ysi s of PPycoated scaffolds prepared under both vaporand liquid-phase polymerization (Figure 1f; Tabl e S1, Supporting Information). The attenuation of the Si 2p si gnal upon polymer deposition is indicative of the thickness of the PPy deposited, accor di n g to a standard uniform overlayer model[96–98];the variability of the Si 2p signal over different measurement points gives an indication of the spatial homogeneity of the coating polymer. Upon vaporphase polymerization a steady increase of the Si 2p signal attenuation is recorded from 1 to 8 h (up to 35%), confi rmi ng an increase of the PPy layer thickness that agr ees with spectroscopic data in Figure1b. Conversely, aconspicuous attenuati on of theSi 2p signal (97%) isobserved already after 5 h of liquid-phasepolymerization. This can be ascribed to the formation of a thick PPy film on top of the PSiO2scaffold, in agr eem en t with SEM data of Figure 1c, which results in anonuniform polymer deposition within the pores, evidenced by EDX data in Figure S4 (SupportingInformation).Thisisfurther confirmed by thehigh variability (percentage relative standard deviation - RSD% - 39.9%) associated with the Si 2p si gnal on differ ent points of the PSiO2scaffolds af ter liquid-phase deposition for 5 h (Tabl e S1, Supporting Information). On the other hand, low var i abi li ty of the Si 2p signal (RSD% = 3.5%) measured after vapor-phase polymerization for 8 h is consistent with the deposition of a homogeneous PPy film. Surfacechemistry of thePPy film deposited by 8h vapor-phase synthesis on PSiO2scaffold is further investigated by XPSanalysis. Figure1f reportsfitted spectra of C 1s and N 1ssignals. Binding energy and chemical assi gn m ent of each component ar e in good agr eem ent with the literature,[99–101] thus highlighting that the chemical structure of vapor -phase grown PPy on PSiO2scaffolds is si m i lar to that of PPy deposited on conventional planar substrates from liquid-phase.[100,101] Thepresenceof components at 287.8 eV in C 1s and 402.0 eV in N 1s, assi gned to C≐N+functionality, reveals that the film isin an oxidized state. This feature is leveraged to promote el ectr ostati c interactions with the template in imprinting procedures. We next investigated the vapor-phase synthesis of MIP receptors for the protein HHb within the PSiO2scaffolds. The protocol is sketched in Figure 2a. The imprinting procedure leverages the covalent coupling of the target protein HHb to the PSiO2surface before PPy deposition. HHb an chor i n g to the PSiO2surface is car r i ed out through si lani zati on with (3am i nopr opyl)tr i ethoxysi lane (APTES) to expose amino groups followed by glutaraldehyde binding (Figure2a1–4). Vapor -phase deposition of PPy for 8 h isthen performed as already described (Figure 1a) and eventually HHb is removed through a washi ng Sm al l 2023, 2302274 © 2023 The Authors. Small published byWiley-VCH GmbH 2302274 (4 of 9) 16136829, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202302274, Wiley Online Library on [29/05/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License E. Mazzotta et al., Small, 2302274 (2023) E. Mazzotta et al., Small, 2302274 (2023) PPy-based MIP HHb detection in human plasma Summary •Nanoporous silica membranes have high potential for implantable and bioresorbable chemical sensors with high selectivity and specificity •Functinalization of the inner surface of the nanopores can be carried out using natural or siynthetic receptors •In-vivo studies over trhree moths clearly demonstrate biocompatibility and fully elimination of the sensor components Acknoledgments Collaborators: -Dr. L. Dahne, @Surflay Nanotec, Berlin -Prof. D. Giuliani and group, @University of Modena e Reggio Emilia -Prof. E. Mazzotta and group, @University of Salento -People @ab medica s.r.l. Positions open for talented and motivated PhD students and postdocs. If interested, please send your CV and motivation letter to: [email protected] Acknoledgments D1.2 Project Public Launch WP1 Project Management, Dissemination, Exploitation, Communication This project is funded by the European Union Horizon Europe programme under grant agreement No 101046946 This project has received funding from the European Union’s Horizon 2020 research and innovation programm e under grant agreement No 952071 D1.2 - Project Quality Handbook WP1 - Project coordination and technical management Partner: PNO Authors: Carolina Salas, Antonio M . Ortiz, Jeanett Bolther Version: v. final Date: 31.08.2020 www.resorb-project.eu