An electrochemical sensor for detecting multiple markers of gingival crevicular fluid based on size stratification and differential configuration, and a preparation and detection method thereof

Abstract

The application discloses a gingival crevicular fluid multi-marker detection electrochemical sensor based on size stratification and differential configuration and a preparation and detection method thereof. The sensor is provided with a flexible three-electrode working area, a working electrode is composed of a Prussian blue / oxidase bottom layer, a PEDOT:PSS / IL-1beta aptamer middle layer and a gold nanoparticle / MMP-8 antibody top layer, and is matched with a cellulose acetate membrane, a chitosan-sodium alginate composite membrane and a PEG-DA hydrogel membrane with a number average molecular weight of 4000 for hierarchical interception; an amperometric method is adopted to output a total signal of lactic acid / uric acid small molecules, and EIS is adopted to analyze MMP-8 and IL-1beta signals, and is applicable to rapid combined screening of trace gingival crevicular fluid.

Description

Technical Field

[0001] This invention belongs to the field of biomedical sensing technology, specifically relating to an electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid based on size stratification and differential configuration, and its preparation and detection methods. Background Technology

[0002] Periodontitis is a common chronic inflammatory disease of the oral cavity, mainly caused by pathogenic bacteria in dental plaque. Long-term development can lead to the destruction of periodontal supporting tissues, eventually causing tooth loosening and loss. The incidence of periodontitis (broadly defined periodontal problems) is relatively high in Chinese adults. Currently, clinical examination and assessment mainly rely on probing (periodontal probing depth, attachment loss, bleeding index, etc.) and imaging examinations (alveolar bone resorption), which have limitations such as high subjectivity, invasiveness, and only reflecting existing tissue damage.

[0003] Gingival crevicular fluid (GCF) is a physiological exudate located in the gingival sulcus between the gingiva and teeth. It is rich in various biomarkers that can reflect the inflammatory status and degree of damage in periodontal tissues. These mainly include: protein inflammatory factors such as interleukin-1β (IL-1β), matrix metalloproteinase-8 (MMP-8), and tissue inhibitor of metalloproteinases-1 (TIMP-1); and small molecule metabolites such as lactic acid and uric acid. Single biomarker detection is often insufficient to accurately reflect the risk status of periodontitis, while combined detection of multiple biomarkers can significantly improve the efficacy of auxiliary screening.

[0004] Current clinical detection of gingival crevicular fluid mainly relies on enzyme-linked immunosorbent assay (ELISA) or mass spectrometry, which has the following technical problems: (1) The AUC of single marker detection is usually between 0.75 and 0.88, and the sensitivity and specificity are difficult to exceed 80% at the same time, which cannot meet the needs of rapid screening; (2) Multiple marker detection requires multiple sampling or complex flow path design, and the difference in the amount of gingival crevicular fluid collected may lead to significant variations in the marker detection rate; (3) Existing electrochemical sensors are difficult to achieve simultaneous detection of large molecular proteins (>10 kDa) and small molecular metabolites (<1 kDa) on the same device. In addition, existing electrochemical sensors are also difficult to apply to the narrow space of the gingival sulcus due to problems such as substrate rigidity and poor adhesion.

[0005] The core technical challenges of gingival crevicular fluid detection include: (1) extremely small sample size (usually only micro-scale), requiring high sensor sensitivity; (2) complex composition, including macromolecular proteins and small molecule metabolites, requiring differentiated detection strategies; and (3) different molecular weight markers require different recognition and signal transduction mechanisms. Summary of the Invention

[0006] To address the problems existing in the prior art, this invention provides an electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid based on size stratification and differential configuration, comprising:

[0007] A flexible substrate with a conductive layer on its surface, and a working electrode, a reference electrode, and a counter electrode disposed on the flexible substrate, wherein the counter electrode surrounds the working electrode in a ring or semi-ring shape;

[0008] The working electrode is composed of a three-level signal response layer and a selectively permeable membrane;

[0009] The three-level signal response layer consists of a bottom layer, a middle layer, and a top layer, from the inside out. The bottom, middle, and top layers use oxidases, nucleic acid aptamers, and monoclonal antibodies as recognition molecules, respectively. The bottom layer is detected using amperometrics, while the middle and top layers are detected using electrochemical impedance spectroscopy. Differential signal transduction is achieved by analyzing the signals at the two interfaces through equivalent circuit fitting.

[0010] The bottom response layer is a Prussian blue catalytic layer with lactate oxidase and uricase immobilized on its surface; the middle response layer is a PEDOT:PSS conductive polymer layer with IL-1β nucleic acid aptamers immobilized on its surface; the top response layer is an electrodeposited gold nanoparticle layer with MMP-8 monoclonal antibody directly and directionally immobilized on its surface via gold-sulfur bonds.

[0011] The selectively permeable membrane comprises three layers: an outer layer and a middle layer. The outer layer is a modified PEG-DA hydrogel membrane, disposed inside the top response layer. The PEG-DA hydrogel membrane is formed by cross-linking PEG-DA, glycerol, and a photoinitiator. After being fully swollen with PBS buffer, it has a pore size of 6-7 nm, which preferentially retains MMP-8 and facilitates the mass transfer of IL-1β to the middle layer. The middle layer is a chitosan-sodium alginate composite membrane with a pore size of 2-4 nm, disposed inside the middle response layer, which prevents IL-1β from penetrating to the bottom response layer, allowing IL-1β to remain outside the middle response layer to bind to nucleic acid aptamers. The inner layer is a cellulose acetate membrane, closely attached to the conductive layer and located inside the bottom response layer, which serves as the innermost size barrier and transmembrane electrodeposition substrate, preventing residual macromolecular impurities from contacting the conductive layer.

[0012] Preferably, the thickness of the flexible substrate is 5-20 μm; the flexible substrate material is a parylene film.

[0013] The surface of the Prussian blue catalytic layer, with a thickness of 20-100 nm, is immobilized with lactate oxidase and uricate oxidase. This layer is used to detect the total redox current signal of small molecule metabolites on the surface of the Prussian blue catalytic layer by amperometric method.

[0014] The surface is immobilized with a PEDOT:PSS conductive polymer layer containing IL-1β nucleic acid aptamers, with a thickness of 20-100 nm.

[0015] The IL-1β nucleic acid aptamer is a 5'-NH2-C6 modified aptamer. The carboxyl group introduced into the PEDOT:PSS layer, after activation by EDC / NHS, reacts with the amino group of the 5'-NH2-C6 modified aptamer to form an amide bond, thereby covalently and orienting the aptamer onto the PEDOT:PSS surface. The intermediate response layer is used for the electrochemical impedance spectroscopy detection of IL-1β.

[0016] The surface is directly and directionally immobilized with an electrodeposited gold nanoparticle layer of MMP-8 monoclonal antibody via gold-sulfur bonds, with a thickness of 50-200 nm, for the electrochemical impedance spectroscopy detection of MMP-8;

[0017] The electrodeposited gold nanoparticle layer has a porous stacked structure with interparticle gaps larger than the size of IL-1β and small molecule metabolites. It serves as a response and capture interface for MMP-8 rather than a size sieving barrier, allowing IL-1β and small molecule metabolites to transfer mass to the inner layer through the interparticle gaps.

[0018] Preferably, the PEG-DA has a number-average molecular weight M n The value is 4000 for PEG-DA.

[0019] Preferably, the reference electrode is an Ag / AgCl reference electrode; the counter electrode is a platinum counter electrode; and the reference electrode is located at the geometric center of the annular or semi-annular region enclosed by the counter electrode.

[0020] Preferably, the device further includes a removable disposable sample pad, which covers the working electrode and is composed of a hydrophilic low-protein adsorption coarse filter membrane with a pore size of 100-200 nm. The coarse filter membrane is a track-etched polycarbonate membrane or a polyethersulfone membrane. The sample pad is used to coarsely filter the gingival crevicular fluid during sample loading, retaining cells, bacteria, and debris larger than the target protein, while allowing MMP-8, IL-1β, and small molecule metabolites to pass through. The sample pad is removed before electrochemical detection, allowing the working electrode to directly contact the filtered sample.

[0021] This invention also provides a method for preparing the electrochemical sensor for detecting multiple biomarkers of gingival crevicular fluid based on size stratification and differential configuration, comprising the following sequential steps:

[0022] S1: A mixed precursor of graphene oxide and polyamic acid is carbonized at high temperature on a silicon substrate to form a conductive carbon film as a conductive layer;

[0023] S2: A flexible substrate of parylene material with a thickness of 10-15 μm is formed on the surface of the conductive carbon film by chemical vapor deposition, and the conductive carbon film is transferred to the flexible substrate by flipping and peeling.

[0024] S3: The working electrode, reference electrode, and counter electrode patterns are formed on a conductive carbon film using photolithography.

[0025] S4: A silver layer is electrodeposited in the reference electrode pattern area and then chlorinated to form an Ag / AgCl reference electrode; a platinum layer is electrodeposited or sputtered in the counter electrode pattern area to form a platinum counter electrode; a conductive carbon film is retained in the working electrode pattern area as a substrate for subsequent multilayer response structures.

[0026] S5: On the patterned conductive carbon film surface of the working electrode, a 2 wt% cellulose acetate acetone solution is first drop-coated and dried at room temperature to form a cellulose acetate film (CA film), which is placed closely against the conductive carbon film as the innermost size barrier and transmembrane electrodeposition substrate; then, potassium ferricyanide and iron salt precursors are allowed to penetrate the pores of the cellulose acetate film, nucleate at the interface near the conductive carbon film / cellulose acetate film, and grow in situ along the membrane pores to form a Prussian blue catalytic layer that is electronically connected to the conductive carbon film and at least partially exposed to the outside of the cellulose acetate film as the bottom response material; then, lactate oxidase and urate oxidase are covalently fixed to the surface of the Prussian blue catalytic layer by glutaraldehyde crosslinking.

[0027] S6: On the outer surface of the bottom Prussian blue catalyst layer, chitosan solution and sodium alginate solution are first coated sequentially. After cross-linking with polyelectrolytes and washing to remove uncross-linked components, a chitosan-sodium alginate composite membrane with a pore size of 2-4 nm is formed. Subsequently, a PEDOT:PSS layer containing carboxyl functional groups is formed by constant-potential electropolymerization of a mixed solution of EDOT and PSS (containing carboxylated PSS components, or adding carboxylated EDOT monomer EDOT-COOH as a copolymer component) on the outer surface of the chitosan-sodium alginate composite membrane. The carboxyl groups on the surface of the PEDOT:PSS layer are activated with EDC / NHS, and then a 5'-NH2-C6 modified IL-1β nucleic acid aptamer solution is added dropwise. The aptamer is covalently and directionally fixed to the PEDOT:PSS surface through amide bond formation. Unbound aptamer molecules are removed by washing with PBS buffer.

[0028] S7: First, a PEG-DA prepolymer solution is coated onto the surface of the intermediate layer PEDOT:PSS, wherein the PEG-DA prepolymer solution contains 5 wt% number-average molecular weight M nA modified PEG-DA hydrogel membrane was formed by UV crosslinking of 4000 PEG-DA, 10 wt% glycerol, and 0.02-0.1 wt% photoinitiator. The pore size of the hydrogel membrane after swelling with PBS buffer was 6-7 nm. Subsequently, chloroauric acid precursor was allowed to penetrate the pores of the modified PEG-DA hydrogel membrane, nucleating at the conductive interface near PEDOT:PSS / PEG-DA and growing in situ along the hydrogel pores to form a gold nanoparticle layer that maintains electronic connectivity with PEDOT:PSS and is at least partially exposed to the outside of the hydrogel. MMP-8 monoclonal antibody was directly and directionally immobilized on the surface of the gold nanoparticles through gold-sulfur bonds. The gold nanoparticle layer serves as the outermost layer of the sensor and is in direct contact with the sample.

[0029] S8: Encapsulation and Electrical Connection: The flexible substrate with multilayer film modification is encapsulated. A microfluidic detection cell is defined in the non-detection area using UV insulating adhesive. The end of the conductive layer is then heat-pressed to the flexible PCB using anisotropic conductive adhesive (ACF) to complete the electrical connection and sealed storage.

[0030] Preferably, in S1, the graphene oxide and polyamic acid are mixed at a mass ratio of 0.5-0.7:1, coated on a temporary substrate, and carbonized in a nitrogen atmosphere at 950-1050°C for 1-2 hours to form a carbon film.

[0031] Preferably, the electrodeposition of the Prussian blue catalyst layer in S5 specifically includes: depositing the Prussian blue catalyst layer for 15-25 cycles in the range of -0.2 V to +0.6 V using a cyclic voltammetry method.

[0032] Preferably, the constant potential electropolymerization in S6 specifically includes: forming the PEDOT:PSS surface by constant potential electropolymerization at +0.85 V to +0.95 V for 120-180 seconds in an aqueous solution containing EDOT and PSS; the EDC / NHS activation specifically includes: activating the carboxyl groups on the PEDOT:PSS surface with a mixed aqueous solution of 50-200 mmol / L EDC and 10-50 mmol / L NHS at room temperature for 15-30 minutes, followed by dropwise addition of a 0.5-2 μmol / L solution of 5'-NH2-C6 modified IL-1β nucleic acid aptamer, reacting at room temperature for 30-60 minutes, thereby covalently and orienting the aptamer to the PEDOT:PSS surface through EDC / NHS-mediated amide bond chemistry;

[0033] In the aqueous solution containing EDOT and PSS, the PSS is a carboxylated PSS component, or EDOT-COOH is added to the aqueous solution as a copolymer component. Further, the EDOT-COOH can be a commercially available carboxylated EDOT monomer, or it can be prepared from hydroxymethyl EDOT via an oxidative carboxylation route; when using a commercially available monomer, EDOT-COOH with a purity of not less than 95% is preferred, and the carboxylation structure should be confirmed by NMR or mass spectrometry before use. When using a carboxylated PSS component, the carboxyl content of the carboxylated PSS is preferably 1-5 mol%.

[0034] Preferably, the electrodeposited gold nanoparticle layer in S7 specifically comprises: electrodepositing with a 4-6 mmol / L chloroauric acid (HAuCl4) solution at a constant potential of -0.15 V to -0.25 V for 100-180 seconds, allowing the chloroauric acid precursor to penetrate the pores of the modified PEG-DA hydrogel membrane and nucleate and grow along the PEDOT:PSS / PEG-DA adjacent interface and the hydrogel pores; the content of the photoinitiator in the prepolymer solution is 0.02-0.1 wt%.

[0035] This invention also provides a method for detecting gingival crevicular fluid, utilizing the aforementioned electrochemical sensor for detecting multiple biomarkers of gingival crevicular fluid based on size stratification and differential configuration, comprising the following steps:

[0036] S101: Obtain a filter paper strip adsorbed with gingival crevicular fluid, immerse the filter paper strip in a centrifuge tube pre-filled with PBS buffer, vortex for 3-10 minutes to release proteins and small molecule metabolites into the buffer, centrifuge to remove fiber debris, and take the supernatant as the elution solution to be tested;

[0037] S102: Before detection, add PBS buffer to the sensor surface and let it stand for 60-300 seconds to allow the outermost modified PEG-DA hydrogel membrane to fully absorb water and swell to restore the pore size to 6-7 nm. Then, remove the residual buffer in the detection cell.

[0038] S103: Add the eluent to be tested onto the sensor surface and incubate at room temperature for 3-15 minutes. This allows MMP-8 in the sample to be retained by the outer PEG-DA hydrogel membrane and bind to the monoclonal antibody on the surface of the top gold nanoparticle layer. IL-1β in the sample passes through the outer PEG-DA hydrogel membrane, is retained by the middle chitosan-sodium alginate composite membrane, and binds to the nucleic acid aptamers on the surface of the middle PEDOT:PSS layer. Small molecule metabolites lactic acid and uric acid in the sample reach the bottom Prussian blue catalytic layer through the three selectively permeable membranes, simultaneously completing size grading and interface recognition in the same incubation.

[0039] S104: After incubation, the sensor surface was cleaned with PBS buffer to remove non-specific adsorbed substances. Then, using an Ag / AgCl electrode as the reference electrode and a platinum electrode as the counter electrode, an atmosphere containing a redox probe (e.g., 5 mmol / L [Fe(CN)6]) was added. 3- / 4- In the supporting electrolyte, an AC perturbation with an amplitude of 5 mV was applied at the open-circuit potential, and an electrochemical impedance spectroscopy (EIS) scan was performed in the frequency range of 0.1 Hz to 100 kHz. Due to significant differences in surface materials, capture chemistry, and distance from the electrolyte between the top gold / antibody-MMP-8 interface and the middle PEDOT:PSS / aptamer-IL-1β interface, the two interfaces have resolvable different charge transfer resistances and capacitance time constants. The RC parameters corresponding to the two interfaces were analytically separated by equivalent circuit fitting, and the MMP-8 detection signal (ΔRct1) and IL-1β detection signal (ΔRct2) were obtained respectively.

[0040] S105: The working electrode potential was set to -0.05 V to -0.15 V (vs. Ag / AgCl), and the steady-state current was recorded using chronoamperometry. Lactic acid was catalyzed by lactate oxidase, and uric acid was catalyzed by urate oxidase to generate H2O2. H2O2 was reduced at the same potential on the surface of the Prussian blue catalyst layer, generating a current. Therefore, this current is the total reduction current of H2O2 generated by lactate and uric acid through oxidase catalysis, which serves as the total signal of the substrate of the inner small molecule oxidase.

[0041] Preferably, the detection method is used to simultaneously detect four markers in gingival crevicular fluid: MMP-8, lactic acid, uric acid, and IL-1β.

[0042] Preferably, in S103, the process of adding the eluent to the sensor surface specifically includes: first, adding the eluent to the sensor onto a removable disposable sample pad covering the working electrode, so that the eluent can be coarsely filtered through the disposable sample pad to remove cells, bacteria and debris larger than the target protein before entering and contacting the top gold nanoparticle layer; after the sample is added, the disposable sample pad is removed.

[0043] This invention addresses the core problems in gingival crevicular fluid detection, such as insufficient sensitivity and specificity, small sample size, and complex composition, by employing a structure of "bottom layer responding to small molecules - middle layer responding to small-sized proteins - outer layer responding to large-molecule proteins". The working electrode surface is covered with a three-layer stacked selectively permeable membrane with decreasing pore size gradient: the outermost PEG-DA hydrogel membrane (dry pore size 5-7 nm, swollen pore size 6-7 nm) blocks MMP-8 (hydrated diameter approximately 8 nm) from penetrating into the inner layer, and is recognized by the MMP-8 antibody immobilized on the surface of the gold nanoparticle layer (top response material) on its outer side, while allowing IL-1β (hydrated diameter approximately 5 nm) to penetrate; the middle chitosan-sodium alginate composite membrane (pore size 2-4 nm) prevents IL-1β from penetrating to the bottom response layer, so that IL-1β is retained on the outside of the middle response layer to bind to the immobilized nucleic acid aptamers and be recognized and detected; the innermost cellulose acetate membrane (CA membrane) acts as an inner size barrier and Prussian blue electrodeposition substrate, preventing residual macromolecular impurities from contacting the conductive carbon membrane and stabilizing the bottom electrochemical interface. This invention employs a three-tiered detection strategy: an outer layer uses EIS to detect MMP-8, a middle layer uses EIS to detect IL-1β, and an inner layer uses amperometry to obtain the total H2O2 signal generated by two small-molecule oxidase substrates, lactic acid and uric acid. This allows for stepwise screening and tiered detection of markers of different sizes on the same electrode. The sensor boasts advantages such as high sensitivity, high specificity, multi-target detection, and rapid response, making it suitable for early screening and dynamic monitoring of oral diseases such as periodontitis.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] (1) By using a three-layer stacked selective permeable membrane with decreasing pore size (5-7 nm swollen pore size 6-7 nm → 2-4 nm → approximately 1000 Da), the size screening principle is used to achieve the stepwise preferential retention of MMP-8 and IL-1β and the efficient delivery of small molecule metabolites to the bottom response region on a single working electrode, realizing the spatial physical isolation of each level of markers without the need for external flow path switching or multiple sampling. The outer PEG-DA hydrogel membrane has a pore size of 6-7 nm in the swollen state, which can preferentially retain MMP-8 with a hydrated diameter of about 8 nm and significantly facilitate the mass transfer of IL-1β with a hydrated diameter of about 5 nm to the middle layer. The middle chitosan-sodium alginate composite membrane further prevents IL-1β from penetrating to the bottom response layer. The inner cellulose acetate membrane (CA membrane) prevents residual macromolecular impurities from reaching the conductive carbon membrane through size repulsion. Thus, the structure solves the technical problem in the prior art that it is difficult to detect macromolecular proteins (>10 kDa) and small molecule metabolites (<1 kDa) simultaneously on the same device.

[0046] (2) Differentiated signal transduction mechanisms are adopted for biomarkers of different sizes. Electrochemical impedance spectroscopy (EIS) is used to detect outer and middle layer protein biomarkers, while enzymatic amperometry is used to detect inner layer small molecule metabolites. The dual-mode detection strategy helps to reduce detection interference between different types of biomarkers and alleviates the problem that existing electrochemical sensors require different recognition and signal transduction mechanisms for biomarkers of different molecular weights but are difficult to integrate.

[0047] (3) By realizing the joint detection of protein biomarkers and small molecule metabolites on a single sensor, a hardware foundation is provided for improving the efficiency of periodontitis screening, and a technical possibility is provided for achieving risk assessment accuracy that is superior to single biomarker detection (the literature reports that the AUC is usually 0.75-0.88).

[0048] (4) Using a flexible parylene substrate (thickness 5-20 μm) satisfies the requirements of substrate flexibility for the flip-peel transfer process and ensures the complete preparation of the multilayer thin film structure. On the other hand, the good biocompatibility and bendability of the flexible substrate provide a structural basis for the future realization of in-situ attachment-based real-time detection in the oral cavity.

[0049] (5) The overall detection only requires a microliter of gingival crevicular fluid elution sample (50-100 μL PBS buffer elution and then drop), and the filter paper strip can be loaded once to complete the joint detection of multiple targets, avoiding the variation in biomarker detection rate caused by multiple sampling and reducing the complexity of clinical operation.

[0050] (6) For impurities in gingival crevicular fluid that are much larger than the target protein, such as cells, bacteria, filter paper fiber debris, and large molecular aggregates, a disposable sample pad (a hydrophilic low-protein adsorption coarse filter membrane with a pore size of 100-200 nm, such as track-etched polycarbonate membrane—with extremely uniform pore size, neutral through-holes, pure size exclusion, and no charge effect) can be added above the detection cell window. During sample loading, the sample enters the detection cell after coarse filtration through the sample pad. Before detection, the sample pad is removed to allow the top gold nanoparticle layer to directly contact the filtered sample. Since the pore size of this sample pad is much larger than that of MMP-8 (hydration diameter of about 8 nm), it does not hinder the mass transfer of each target marker. Its removable design further reduces the contamination and interference of the complex matrix on the detection of the outer MMP-8 layer without increasing the diffusion resistance of the resident membrane layer or changing the direct contact relationship between the third-order response layer and the sample. Attached Figure Description

[0051] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0052] Figure 1 This is a schematic diagram illustrating the size layering principle of the electrochemical sensor of the present invention;

[0053] Figure 1 The layered structure of the sensor is shown in the figure. The numbers 1-4 represent MMP-8 (1), lactic acid (2), uric acid (3) and IL-1β (4) molecules, respectively. The number 5 is the working electrode (5). 5-1 is the outer MMP-8 detection level, 5-1-1 is the top response material (gold nanoparticle layer), 5-1-2 is the modified PEG-DA hydrogel membrane, 5-2 is the middle IL-1β detection level, 5-2-1 is the middle layer response material (PEDOT:PSS layer containing carboxyl functional groups), 5-2-2 is the chitosan-sodium alginate composite membrane, 5-3 is the inner small molecule detection level, 5-3-1 is the bottom response material (Prussian blue catalytic layer), and 5-3-2 is the inner selective permeable membrane (cellulose acetate membrane (CA membrane)).

[0054] Figure 2 This is a layout design diagram of the working electrode of the present invention;

[0055] Figure 2 The image shows a top view of the sensor, with 5 being the working electrode, 10 being the central reference electrode, and 11 being the annular counter electrode; A-A' is a cross-sectional line.

[0056] Figure 3 This is a side cross-sectional view of the working electrode layer structure of the present invention;

[0057] Figure 3 This is a schematic diagram of the overall structure in cross-section along the A-A' direction, showing the conductive layer 13, the working electrode 5, the Ag / AgCl reference electrode 10, the Pt counter electrode 11, the encapsulation layer 12, and the flexible substrate 14.

[0058] Figure 4 This is a flowchart illustrating the sensor fabrication process of the present invention.

[0059] Figure 4 This is a schematic diagram of the process flow for fabricating the sensor of the present invention. The main process stages are shown sequentially: conductive carbon film preparation, flexible substrate formation and electrode transfer, electrode patterning, construction of the inner small molecule detection stage, construction of the intermediate IL-1β detection stage, construction of the outer MMP-8 detection stage, and encapsulation and electrical connection. The metallization (Ag / AgCl and Pt deposition) of the reference electrode and counter electrode is performed after electrode patterning and is included in the electrode patterning stage in this schematic diagram. This flowchart is a general overview of the process stages; the complete sequence of steps and their numbering (S1 conductive carbon film preparation, S2 flexible substrate formation and electrode transfer, S3 electrode patterning, S4 metallization of the reference electrode and counter electrode, S5 construction of the inner small molecule detection stage, S6 construction of the intermediate IL-1β detection stage, S7 construction of the outer MMP-8 detection stage, S8 encapsulation and electrical connection) shall be as described in the specific embodiments and claims.

[0060] Explanation of labels in the figure: 1-MMP-8; 2-Lactic acid; 3-Uric acid; 4-IL-1β; 5-Working electrode; 5-1-Outer MMP-8 detection layer: used to retain and detect MMP-8; 5-1-1-Top layer response material: electrodeposited gold nanoparticle layer, used to immobilize MMP-8 antibody; 5-1-2-Modified PEG-DA hydrogel membrane: primary screening membrane, dry state pore size 5-7 nm, swollen state pore size 6-7 nm; 5-2-Intermediate IL-1β detection layer: used to retain and detect IL-1β; 5-2-1-Intermediate layer response material: PEDOT containing carboxyl functional groups: PSS conductive polymer layer, surface carboxyl groups activated by EDC / NHS to immobilize 5'-NH2-C6 modified IL-1β nucleic acid aptamers; 5-2-2-Chitosan-sodium alginate composite membrane: intermediate screening membrane, pore size 2-4 nm. nm; 5-3-Inner layer small molecule detection stage: used for passing through and detecting small molecule metabolites; 5-3-1-Bottom layer response material: Prussian blue catalytic layer, used to immobilize small molecule recognition enzymes; 5-3-2-Inner layer selectively permeable membrane: three-level screening membrane (cellulose acetate membrane, CA membrane); 6-Outer layer recognition molecule: MMP-8 monoclonal antibody; 7-Middle layer recognition molecule: 5'-NH2-C6 modified IL-1β nucleic acid aptamer; 8-Inner layer lactate recognition molecule: lactate oxidase; 9-Inner layer uric acid recognition molecule: urate oxidase; 10-Central reference electrode (Ag / AgCl): provides a stable reference potential; 11-Counter electrode (platinum): constitutes the current loop; 12-Encapsulation layer: defines the UV-insulated encapsulation of the microfluidic detection cell; 13-Conductive layer: electrical signal transmission leads; 14-Flexible substrate: the base supporting the entire sensor structure. Detailed Implementation

[0061] The concept of this invention is to construct a three-level detection zone with spatial physical isolation and complementary functions on a single working electrode (5) through a multi-layer stacked structure in which "the detection layer is located outside the screening layer". The three levels are as follows, from the inside to the outside:

[0062] The inner small molecule detection layer consists of a cellulose acetate membrane (CA membrane) (5-3-2) closely attached to the conductive carbon membrane, and an outer bottom response material (5-3-1, Prussian blue catalytic layer). The CA membrane, acting as the innermost size barrier and the Prussian blue electrodeposition substrate, operates on a pure size repulsion mechanism, containing no charged groups. It prevents large molecular impurities remaining after passing through the upper membrane structures from contacting the conductive carbon membrane and stabilizes the bottom electrochemical interface. Small molecule metabolites such as lactic acid and uric acid, which penetrate to the bottom response material, undergo enzymatic reactions in the Prussian blue catalytic layer and are detected by amperometric methods.

[0063] The intermediate IL-1β detection layer consists of a chitosan-sodium alginate composite membrane (5-2-2, pore size 2-4 nm) and an outer intermediate response material (5-2-1, PEDOT:PSS layer containing carboxyl functional groups). The 2-4 nm pore size of the composite membrane prevents IL-1β (4, hydration diameter approximately 5 nm) from penetrating to the bottom response layer, thus retaining it on the outer side of the intermediate response layer. After the carboxyl groups on the surface of the PEDOT:PSS layer are activated by EDC / NHS, they react with the amino groups of the 5'-NH2-C6 modified IL-1β nucleic acid aptamer to form amide bonds, covalently and orienting the aptamer to the PEDOT:PSS surface for electrochemical impedance spectroscopy (EIS) detection.

[0064] The outer MMP-8 detection layer consists of a modified PEG-DA hydrogel membrane (5-1-2, dry pore size 5-7 nm, pore size 6-7 nm after full swelling with PBS buffer) and its outermost top response material (5-1-1, gold nanoparticle layer). The gold nanoparticle layer is the outermost layer and directly contacts the sample. MMP-8 (1, hydrated diameter approximately 8 nm) is captured by monoclonal antibodies directly and directionally immobilized on the surface via gold-sulfur bonds and detected by EIS. The PEG-DA membrane blocks MMP-8 from penetrating into the inner layer on its inner side, while allowing IL-1β (hydrated diameter approximately 5 nm) to penetrate into the middle layer.

[0065] The technical solution of the present invention is as follows: an electrochemical sensor for detecting multiple markers in gingival crevicular fluid, comprising: a flexible substrate of parylene with a thickness of 5-20 μm; a working electrode, an Ag / AgCl reference electrode and a platinum counter electrode disposed on the flexible substrate, wherein the reference electrode is located at the geometric center of the annular or semi-annular region surrounded by the counter electrode, and the counter electrode surrounds the working electrode in an annular or semi-annular manner.

[0066] Each working electrode (5) is disposed on a flexible substrate (14) and is physically stacked from the inside out, consisting of an inner small molecule detection layer, a middle detection layer, and an outer protein detection layer; each layer contains a signal-response material and a selectively permeable membrane disposed on its inner side (below), with the specific structure as follows:

[0067] (I) Outer protein detection grade:

[0068] Top-layer response material (5-1-1): with a thickness of 50-200 nm, composed of an electrodeposited gold nanoparticle layer, set on the modified PEG-DA hydrogel membrane (5-1-2), serving as the outermost layer of the working electrode in direct contact with the sample. MMP-8 monoclonal antibody is directly and directionally immobilized on its surface through gold-sulfur bonds, used to capture the interfacial impedance change signal caused by MMP-8 (1) binding by electrochemical impedance spectroscopy (EIS); the gold nanoparticle layer is a porous stacked structure, and its interparticle gaps do not hinder the mass transfer of IL-1β and small molecule metabolites to the inner layer, serving only as the response and capture interface of MMP-8 rather than a size sieving barrier;

[0069] Modified PEG-DA hydrogel membrane (5-1-2): Located inside (below) the top layer response material (5-1-1) and on the outer surface of the middle layer response material (5-2-1), serving as an outer screening membrane. The dry pore size is 5-7 nm, and the pore size is 6-7 nm after being fully swollen with PBS buffer. It is used to block MMP-8 (1, hydrated diameter about 8 nm) from penetrating into the inner layer, while allowing IL-1β (4, hydrated diameter about 5 nm) to transfer mass to the middle layer, and to provide a hydrophilic deposition substrate for the top gold nanoparticle layer. 10 wt% glycerol is pre-doped into the membrane to prevent the gel skeleton from dehydrating and collapsing during subsequent encapsulation and storage, and to maintain the designed pore size.

[0070] (II) Intermediate Layer Detection Level:

[0071] Intermediate responsive material (5-2-1): 20-100 nm thick, composed of a PEDOT:PSS conductive polymer layer containing carboxyl functional groups (the carboxyl groups are introduced during electropolymerization by the comonomer EDOT-COOH or carboxylated PSS components), disposed on top of a chitosan-sodium alginate composite film (5-2-2). Its surface is activated with EDC / NHS to activate the carboxyl groups on the substrate surface, and a 5'-NH2-C6 modified IL-1β nucleic acid aptamer is covalently and oriented to the material surface via amide bonds for IL-1β recognition detection and interlayer signal transduction.

[0072] Chitosan-sodium alginate composite membrane (5-2-2): disposed inside (below) the intermediate layer response material (5-2-1) and on the outer surface of the bottom layer response material (5-3-1), serving as an intermediate layer screening membrane with a pore size of 2-4 nm, used to block IL-1β (4, hydration diameter about 5 nm) from penetrating into the inner layer, allowing small molecule metabolites to continue to transfer mass inward.

[0073] (III) Inner layer small molecule detection grade:

[0074] The bottom response material (5-3-1) has a thickness of 20-100 nm and consists of a Prussian blue catalytic layer modified with lactate oxidase (8) and urate oxidase (9), and is disposed on the outside of the cellulose acetate membrane (CA membrane) (5-3-2). Lactic acid is catalyzed by lactate oxidase and uric acid is catalyzed by urate oxidase to generate H2O2. H2O2 is reduced on the surface of the Prussian blue catalytic layer at the same working potential. This reduction current is the total signal of H2O2 generated by the two small molecule oxidase substrates of lactate and uric acid, which is used to obtain the total signal of the inner small molecule oxidase substrate by amperometry.

[0075] Cellulose acetate membrane (CA membrane) (5-3-2): It is set closely to the conductive carbon membrane (13) as the innermost selective permeable membrane and Prussian blue electrodeposition substrate. It is located inside (below) the bottom response material (5-3-1). It uses pure size repulsion as a mechanism and does not contain charged groups. Its main function is to prevent the large molecular impurities that remain after passing through the upper layer membrane structure from contacting the conductive carbon membrane and to stabilize the bottom electrochemical interface.

[0076] The multilayer stacked structure on the working electrode surface adopts the configuration principle of "detection layer located outside the screening layer". From the inside to the outside, it consists of: a cellulose acetate membrane (CA membrane, 5-3-2) closely attached to the conductive carbon membrane, serving as the innermost size barrier and Prussian blue electrodeposition substrate; outside of it is the Prussian blue catalytic layer (5-3-1, bottom detection), on which lactate oxidase (8) and urate oxidase (9) are fixed, and the total signal of H2O2 generated by the two small molecule oxidase substrates of lactate and urate is obtained by amperometric method; outside of it is the chitosan-sodium alginate composite membrane (5-2-2, pore size 2-4). The inner layer is a filter membrane (5-2-1, intermediate layer detection), which blocks IL-1β from penetrating into the inner layer. The outermost layer is a PEDOT:PSS layer containing carboxyl functional groups (5-2-1, intermediate layer detection), which is activated by EDC / NHS and has 5'-NH2-C6 modified IL-1β nucleic acid aptamers covalently and directionally fixed with amide bonds. IL-1β is detected by EIS. The outermost layer is a modified PEG-DA hydrogel membrane (5-1-2, dry state pore size 5-7 nm, swollen pore size 6-7 nm, outer layer filter membrane), which blocks MMP-8 from penetrating into the inner layer and allows IL-1β to penetrate into the intermediate layer. The outermost layer is a gold nanoparticle layer (5-1-1, top layer detection), on which MMP-8 monoclonal antibody is directly and directionally fixed with gold-sulfur bonds. It directly contacts the sample and is detected by EIS. The gold nanoparticle layer has a porous stacked structure, and the gaps between the particles do not hinder the mass transfer of IL-1β and small molecule metabolites to the inner layer.

[0077] The present invention will be further described in detail below with reference to embodiments. The present invention provides an electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid, such as... Figure 1 , Figure 2 , Figure 3As shown, the device includes: a flexible substrate with a thickness of 5-20 μm, wherein the flexible substrate material is a parylene film; a working electrode, an Ag / AgCl reference electrode, and a platinum counter electrode disposed on the flexible substrate, wherein the reference electrode is located at the geometric center of the annular or semi-annular region surrounded by the counter electrode, and the counter electrode surrounds the working electrode in an annular or semi-annular manner. The multilayer stacked structure of the working electrode, the function of each layer, and the contact relationship with the sample are as described in the summary section of the invention. In order to describe the structure and preparation method of the working electrode, the present invention uses "inner" and "outer" to describe the relative positional relationship. "Inner" refers to the direction closer to the flexible substrate with the conductive layer, while "outer" refers to the direction away from the flexible substrate with the conductive layer.

[0078] Example 1: Sensor Fabrication

[0079] like Figure 4 This includes the following steps:

[0080] Step 1 (S1): Preparation of conductive carbon film. Graphene oxide (GO) and polyamic acid (PAA) were mixed at a mass ratio of 0.6:1, ultrasonically dispersed, and then spin-coated onto the surface of a silicon wafer to form a precursor film with a thickness of approximately 10 μm. The mixture was heated to 1000℃ in a tube furnace at a rate of 5℃ / min and held at this temperature for 1.5 hours under nitrogen protection. After natural cooling, a highly conductive graphene-like carbon film was expected to be obtained (designed sheet resistance of approximately 8 Ω / sq). Finite element thermal shrinkage and sheet resistance simulations were performed on the above graphene oxide / polyamic acid carbonized conductive film. Under the conditions of a graphene oxide to polyamic acid mass ratio of 0.6:1 and nitrogen carbonization at 1000℃ for 1.5 hours, the simulated sheet resistance of the continuous carbonized network was 8.7 Ω / sq. When the mass ratio was varied within the range of 0.5-0.7:1, the simulated sheet resistance remained between 8-12 Ω / sq, which meets the low series resistance requirement for three-electrode micro-area electrochemical detection.

[0081] Step 2 (S2): Flexible substrate formation and electrode transfer. A 12 μm thick parylene-C film was deposited on the carbon film surface using chemical vapor deposition (CVD). A temporary substrate (PDMS) was attached to the Parylene surface using a flip-and-lift technique, and then the carbon film was detached from the silicon wafer by immersion in deionized water. After flipping, the carbon film was located on the upper surface of the flexible Parylene substrate.

[0082] Step 3 (S3): Electrode Patterning. Electrode patterns are defined on the carbon film using photolithography. This embodiment designs one working electrode (WE, 500 μm in diameter), one reference electrode pattern region (RE, 100 μm in diameter, located at the geometric center of the annular or semi-annular region surrounded by the counter electrode), and one counter electrode pattern region (CE, annular, 50 μm in linewidth, surrounding the working electrode). The carbon film in the non-electrode regions is removed by oxygen plasma etching.

[0083] Step 4 (S4): Metallization of the reference and counter electrodes. In the patterned region of the reference electrode, an Ag layer is electrodeposited using a 0.1 mol / L AgNO3 solution at -0.3 V vs. Ag / AgCl (current density approximately 0.5 mA / cm², deposition time 60-180 seconds). The electrode is then immersed in 0.1 mol / L HCl and a constant potential of +0.3 V vs. Ag / AgCl is applied for 30 seconds, causing partial chlorination of the Ag layer surface to form an Ag / AgCl reference electrode (10). In the patterned region of the counter electrode, a platinum counter electrode is formed using a 4-6 mmol / L potassium chloroplatinate solution at a constant potential of -0.1 V to -0.2 V vs. Ag / AgCl for 60-120 seconds (11) (or a Pt film can be deposited in the counter electrode pattern using mask sputtering). After the metallization process is completed, the electrode is washed with water and dried to obtain an Ag / AgCl reference electrode (10), a platinum counter electrode (11), and a working electrode (5) with the conductive carbon film surface retained for subsequent multilayer modification.

[0084] Step 5 (S5): Constructing the inner layer small molecule detection stage (5-3). First, 5 μL of 2 wt% cellulose acetate acetone solution was dropped onto the patterned conductive carbon film surface of the working electrode and dried at room temperature for 30 minutes to form a cellulose acetate film (CA film) (5-3-2). This CA film was placed close to the conductive carbon film (13) as the innermost size barrier and transmembrane electrodeposition substrate. It was used to prevent large molecular impurities that remained after passing through the upper layer membrane structure from contacting the conductive carbon film and to stabilize the bottom electrochemical interface. Subsequently, in a solution containing 2.5 mmol / L potassium ferricyanide, 2.5 mmol / L ferric chloride, and 0.1 mol / L potassium chloride, cyclic voltammetry was used to scan 20 times in the range of -0.2 V to +0.6 V. The molecular weight of the deposition precursor ions such as potassium ferricyanide and ferric salts is much smaller than the CA membrane cutoff value (about 1000 Da), which can penetrate the CA membrane pores and nucleate at the interface near the conductive carbon membrane / CA membrane. Then, they grow in situ along the CA membrane pores to the outside, forming a Prussian blue catalyst layer that is electronically connected to the conductive carbon membrane and at least partially exposed to the outside of the CA membrane (5-3-1). Finally, a PBS solution (pH 7.4) containing lactate oxidase (LOx, 10 mg / mL) and uricase (uricase, 10 mg / mL) was prepared. The enzymes were covalently immobilized on the surface of the Prussian blue catalyst layer using the glutaraldehyde (0.5%) cross-linking method: 2 μL of 0.5% glutaraldehyde solution was first dropped onto the catalyst layer surface to activate the surface amino groups. After reacting at room temperature for 15 minutes, the solution was washed with PBS. The sensor was then immersed in the enzyme mixture and incubated at room temperature for 2 hours. Unbound enzyme molecules were removed by washing with PBS.

[0085] Purpose and Advantages of the Inner Layer Response Material Design: Prussian blue was chosen as the underlying signal response material specifically for the detection of small molecule metabolites (lactic acid and uric acid) in gingival crevicular fluid. As an "artificial peroxidase," Prussian blue can efficiently catalyze the reduction of H2O2 at a low working potential of approximately -0.1 V (vs. Ag / AgCl), far below the positive potential operating range (above +0.5 V) required by traditional electrode materials such as platinum. The gingival crevicular fluid matrix contains electroactive interfering substances such as ascorbic acid and glutathione, which can generate oxidative current interference at higher positive potentials. The low-potential working mode of Prussian blue effectively avoids these interferences, significantly improving the selectivity for detecting the total signal of the small molecule oxidase substrate. Lactate oxidase and uric acid oxidase are covalently immobilized on the surface of the Prussian blue catalytic layer using glutaraldehyde cross-linking. Compared to physical adsorption immobilization, this prevents enzyme detachment and inactivation in micro-elution samples, ensuring stable catalytic activity even with microliter gingival crevicular fluid sample volumes. It should be noted that since the catalytic products of both lactate oxidase and uriate oxidase are H2O2, and H2O2 is reduced under the same Prussian blue catalytic layer and the same working potential, the amperometric signal obtained by this structure is the total signal of the H2O2 reduction current generated by the two small molecule oxidase substrates of lactate and uriate, rather than the individual quantitative signal of lactate or uriate. This total signal can reflect the combined changes in the activity of dental plaque metabolism and redox load in the gingival sulcus microenvironment. In the simulation-assisted screening model, the total signal was input into the logistic regression model in conjunction with two inflammatory markers, MMP-8 and IL-1β. Compared with the simulated AUC of 0.89 when only MMP-8 and IL-1β were used, the simulated AUC increased to 0.94 after combining the total signal, indicating that the combined signal still has joint assessment value. In addition, cellulose acetate membrane (CA membrane) was selected as the innermost selective permeable membrane, mainly serving as an inner size barrier and transmembrane electrodeposition substrate: it operates on a pure size repulsion mechanism and does not contain charged functional groups. Its main function is to prevent large molecular impurities that remain after passing through the upper layer membrane structure from contacting the conductive carbon membrane and to stabilize the bottom electrochemical interface, without participating in the charge-selective sieving of small molecule metabolites such as lactate and urate.

[0086] Step 6 (S6): Constructing the intermediate IL-1β detection layer (5-2). On the outer surface of the bottom Prussian blue catalyst layer (5-3-1), first, 5 μL of 0.5% chitosan acetate solution (pH 5.0) is drop-coated sequentially and allowed to stand at room temperature for 10 minutes to form a preliminary film. Then, 5 μL of 0.5% sodium alginate solution is drop-coated and allowed to stand at room temperature for 20 minutes. This allows for cross-linking through the complexation of chitosan amino groups with sodium alginate carboxyl polyelectrolytes. Uncross-linked components are removed by PBS washing, forming a chitosan-sodium alginate composite membrane (5-2-2) with a pore size of 2-4 nm, located outside the Prussian blue catalyst layer as the intermediate screening membrane. Subsequently, PEDOT:PSS (5-2-1), a carboxyl-functionalized intermediate layer responsive material, was electropolymerized on the outer surface of the chitosan-sodium alginate composite membrane: In an aqueous solution containing 0.01 mol / L EDOT (with 0.001-0.002 mol / L carboxylated EDOT monomer EDOT-COOH added as a copolymer component, or a carboxyl-containing carboxylated PSS component in the PSS) and 0.1 mol / L PSS, electropolymerization was performed at a constant potential of +0.9 V for 150 seconds to form a PEDOT:PSS layer containing carboxyl functional groups. A conductive carbon membrane was connected to a Prussian blue catalyst layer (conductive) via a CA membrane, providing an electron transport pathway for the polymerization of EDOT / EDOT-COOH monomers. PEDOT:PSS was thus formed in situ on the outer surface of the composite membrane. After activation by EDC / NHS, the 5'-NH2-C6 modified IL-1β nucleic acid aptamer was covalently and directionally immobilized: the carboxyl groups on the PEDOT:PSS surface were activated for 20 minutes at room temperature in MES buffer (pH 5.5) containing 100 mmol / L EDC and 25 mmol / L NHS, followed by gentle washing with PBS; then 5 μL of PBS solution (pH 7.4) containing 1 μmol / L 5'-NH2-C6 modified IL-1β nucleic acid aptamer was dropped onto the surface and incubated at room temperature for 45 minutes. The aptamer was covalently and directionally immobilized on the surface of the intermediate response material through the amide bond formed between the carboxyl groups on the PEDOT:PSS layer surface and the 5'-NH2-C6 modified terminal amino group of the aptamer. Unbound aptamers were removed by washing with PBS.

[0087] Design purpose and advantages of intermediate layer response material: PEDOT:PSS containing carboxyl functional groups was selected as the intermediate layer signal response material, which has three key functional advantages for the detection of IL-1β in gingival crevicular fluid. First, the conductive polymer framework of PEDOT:PSS is the core hub for realizing the multilayer "bottom-up" stepwise construction process: the intermediate layer PEDOT:PSS (5-2-1) is located outside the chitosan-sodium alginate composite membrane (5-2-2), providing an electron transport path for the outer gold nanoparticle electrodeposition step (step 7) - the current can flow from the conductive carbon membrane through the CA membrane, Prussian blue catalyst layer, chitosan-sodium alginate composite membrane, PEDOT:PSS, and through the PEG-DA hydrogel membrane (dry state pore size 5-7 nm, swollen state pore size 6-7 nm, AuCl4) - The precursor anions can penetrate to the outermost layer, enabling in-situ electrodeposition of gold nanoparticles on the outermost side of the PEG-DA film without damaging the existing film layer. This provides the structural basis for the feasibility of the integrated fabrication process of the three-level response layer. Secondly, this invention introduces a carboxylated copolymer component (EDOT-COOH or carboxylated PSS) during the PEDOT:PSS electropolymerization process, enriching the surface of the intermediate layer response material with carboxyl groups (–COOH) that can participate in the amide condensation reaction. Using classic EDC / NHS activation, the 5'-NH2-C6 modified IL-1β nucleic acid aptamer is covalently and oriented onto the PEDOT:PSS surface via amide bonds. This avoids the instability caused by electrostatic interactions alone, and the 5' end directional anchoring ensures the IL-1β binding region of the aptamer points towards the solution side, improving recognition efficiency. Compared to intact monoclonal antibodies, nucleic acid aptamers have a smaller molecular size (approximately 15-30 kDa), allowing for higher density loading of recognition molecules on a limited conductive polymer layer surface. This enhances the capture efficiency of IL-1β in gingival crevicular fluid, which is typically only in the pg / mL to low ng / mL range, meeting the detection requirements of small sample volumes and low biomarker concentrations in gingival crevicular fluid. Third, the chitosan-sodium alginate composite membrane forms a screening layer with a pore size of 2-4 nm through the complexation of natural polysaccharide polyelectrolytes, which is precisely matched with the molecular size of IL-1β (about 5 nm) to achieve efficient retention. Its antifouling properties of natural polysaccharide skeleton can reduce the non-specific adsorption of non-target proteins in the complex matrix of gingival crevicular fluid, reduce the matrix effect interference of EIS detection in the intermediate layer, and improve the specificity of IL-1β detection signal.

[0088] Step 7 (S7): Construct the outer MMP-8 detection grade (5-1). On the outer surface of the intermediate PEDOT:PSS (5-2-1) layer, first drop-coat 10 μL of a solution containing 5 wt% number-average molecular weight MMP-8. nA prepolymer solution of 4000 PEG-DA pre-doped with 10 wt% glycerol (the glycerol is used to prevent the gel skeleton from collapsing due to dehydration and to maintain the designed pore size during subsequent encapsulation and storage) was prepared, along with an appropriate amount of photoinitiator (such as 0.05 wt% LAP or Irgacure 2959). The mixture was then subjected to ultraviolet light (365 nm, 10 mW / cm²). 2 Crosslinking was performed under 60-second irradiation to form a modified PEG-DA hydrogel membrane with a dry pore size of 5-7 nm (5-1-2). After sufficient water absorption and swelling in subsequent PBS buffer, the pore size of this hydrogel membrane was 6-7 nm, located inside (below) the top response material as an outer screening membrane—it could both block the penetration of hydrated MMP-8 with a diameter of approximately 8 nm into the inner layer and allow effective mass transfer of hydrated IL-1β with a diameter of approximately 5 nm into the intermediate layer. Subsequently, in a 5 mmol / L chloroauric acid (HAuCl4) solution, a constant potential of -0.2 V was applied to the working electrode for 150 seconds to construct the top response material gold nanoparticle layer (5-1-1); AuCl4 - Precursor anions can penetrate the hydrated PEG-DA hydrogel membrane and first nucleate at the conductive interface near PEDOT:PSS / PEG-DA. Then, relying on the electronic connectivity of the formed gold nuclei, they continue to grow in situ along the hydrogel channels, ultimately forming a porous gold nanoparticle layer that maintains electronic connectivity with PEDOT:PSS and is at least partially exposed to the outside of the hydrogel. 5 μL of MMP-8 monoclonal antibody solution (10 μg / mL) was drop-coated onto the surface of the gold nanoparticle layer and incubated at room temperature for 30 minutes to allow the antibody to be directly and directionally immobilized on the gold surface via gold-sulfur bonds. Unbound antibody was removed by washing with PBS. The gold nanoparticle layer (5-1-1) serves as the outermost layer of the sensor, directly contacting the sample. MMP-8 (hydrated diameter approximately 8 nm) is captured by the antibody and detected by EIS.

[0089] Purpose and Advantages of Top-Level Response Material Design: The use of electrodeposited gold nanoparticles as the top-level signal response material is a specific design for label-free EIS detection of the MMP-8 macromolecular protein in gingival crevicular fluid (GCF), offering the following key advantages. First, the high specific surface area of ​​gold nanoparticles (3-5 times the effective surface area compared to planar gold films) significantly increases the immobilization density of MMP-8 monoclonal antibodies, thereby improving the capture efficiency and EIS detection sensitivity for MMP-8, which is typically in the ng / mL range in GCF, effectively addressing the challenges of detecting small sample volumes (micro-saturations) and low target protein concentrations in GCF. Furthermore, the gold nanoparticle layer formed by electrodeposition exhibits a dendritic porous stacked structure, with interparticle gaps much larger than the size of IL-1β (hydrated diameter approximately 5 nm) and small molecule metabolites such as lactic acid and uric acid. Therefore, this layer serves only as a response and capture interface for MMP-8, rather than a size sieving barrier. IL-1β and small molecule metabolites can be transferred unimpeded through the interparticle gaps to the inner modified PEG-DA hydrogel membrane and deeper layers, ensuring that the outermost gold nanoparticle layer can directly capture MMP-8 without hindering the stepwise detection of other markers. The actual size sieving is accomplished by a three-layer selectively permeable membrane with decreasing pore size gradient located inside (PEG-DA with 6-7 nm pore size in swollen state → chitosan-sodium alginate with 2-4 nm pore size → CA membrane). Secondly, immobilizing the MMP-8 monoclonal antibody using gold-sulfur bonds allows for stable Au–S interactions between the antibody and gold nanoparticles via thiol-related sites on the antibody's surface. Compared to random physical adsorption, this improves the effective antibody orientation ratio and interfacial binding stability, achieving a higher MMP-8 binding capacity with the same amount of antibody and reducing interference from the complex matrix of gingival crevicular fluid on antibody recognition efficiency. Thirdly, the label-free detection mode of the EIS method eliminates the need for fluorescent labeling or enzyme-labeled secondary antibodies, avoiding biomarker loss caused by multi-step labeling operations in small gingival crevicular fluid samples. This makes it particularly suitable for clinical applications with very small gingival crevicular fluid collection volumes. Fourth, the modified PEG-DA hydrogel membrane (5-1-2) is located inside (below) the gold nanoparticle layer, serving as the outer screening membrane. Its high water content (greater than 90 wt%) and the anti-protein adsorption properties of the PEG backbone, together with the nanoscale network of the membrane, constitute a size barrier for MMP-8. The lower limit of its pore size in the swollen state is designed to be 6-7 nm, which is smaller than MMP-8 (hydration diameter of about 8 nm) to prevent it from penetrating into the inner layer, and larger than IL-1β (hydration diameter of about 5 nm) to allow it to transfer mass to the middle layer through the 5-7 nm pores. At the same time, the membrane provides a stable hydrophilic substrate for the electrodeposition of the gold nanoparticle layer. The gold nanoparticle layer, as the outermost layer, is directly exposed to the gingival crevicular fluid, and MMP-8 is preferentially captured by the antibody. The above design together solves the core problem of achieving sensitive and specific detection of MMP-8 in extremely small amounts of highly complex gingival crevicular fluid.

[0090] Step 8 (S8): Encapsulation and electrical connection. (1) Pretreatment: Place the flexible substrate with the completed multilayer film modification in a constant humidity chamber to ensure that the 5-1-2 layer is always in a slightly moist state (the 10 wt% glycerol required for preventing dehydration / gel skeleton collapse has been pre-incorporated into the PEG-DA prepolymer solution in step 7, so it will not be added again in this step). (2) Precision encapsulation: Apply UV insulating adhesive to the non-detection area using automated dispensing. Use photomask technology to define the microfluidic detection cell above the three electrodes to ensure that the edge of the encapsulation layer (12) completely covers the boundary of each layer of response material. (3) Lead connection: Attach the end of the conductive layer (13) to the flexible PCB and heat-press it with anisotropic conductive adhesive (ACF). (4) Storage conditions: The finished sensor is encapsulated in a sealed bag containing a humidity control carrier, or a temporary moisturizing protective film (such as PE film) is covered at the detection cell window. When used, peel it off and activate it with buffer solution. (5) Optionally, a disposable sample pad is peeled off above the detection cell window. The sample pad is a hydrophilic low-protein adsorption coarse filter membrane (such as track-etched polycarbonate membrane or polyethersulfone membrane) with a pore size of 100-200 nm. It is used to coarsely filter the gingival crevicular fluid during sample loading, retain impurities larger than the target protein such as cells, bacteria and filter paper fiber debris, and allow MMP-8, IL-1β and small molecule metabolites to pass through. The sample pad is removed before electrochemical detection so that the top gold nanoparticle layer can directly contact the filtered sample.

[0091] Example 2: Detection Method

[0092] The sensor prepared in Example 1 was used to perform joint detection of multiple biomarkers on gingival crevicular fluid samples. The specific steps are as follows:

[0093] Step 1: Sample Collection and Preparation. A standard filter paper strip (2 mm × 10 mm) was inserted into the gingival sulcus of the test site and left for 30 seconds before being removed. The filter paper strip was placed in a centrifuge tube containing 100 μL of PBS buffer (pH 7.4, containing 0.1% BSA), vortexed for 10 minutes, and the supernatant was collected as the sample to be tested.

[0094] Step 2: Sensor activation. Remove the sensor from the sealed packaging, peel off the temporary moisturizing protective film, add 20 μL of PBS buffer (pH 7.4) to the microfluidic detection cell, and equilibrate at room temperature for 60-300 seconds to allow each membrane layer to fully hydrate and the outermost modified PEG-DA hydrogel membrane to fully absorb water and swell to restore the designed pore size (lower limit 6-7 nm); then remove any residual buffer solution from the detection cell.

[0095] Step 3: Sample loading and simultaneous incubation. 20 μL of the gingival crevicular fluid sample to be tested is dropped onto a disposable sample pad (a hydrophilic, low-protein adsorption coarse filter membrane with a pore size of 100-200 nm, such as a track-etched polycarbonate membrane or polyethersulfone membrane) covering the detection cell window. This allows the sample to pass through the sample pad for coarse filtration, trapping impurities larger than the target protein, such as cells, bacteria, and filter paper fiber debris, before entering the microfluidic detection cell and contacting the top gold nanoparticle layer. The disposable sample pad is then removed, and the sample is incubated at room temperature for 3-15 minutes. During this single incubation process: MMP-8 (hydrated diameter approximately 8 nm) was retained on the top layer of the outer modified PEG-DA hydrogel membrane (swollen state pore size 6-7 nm) and bound to the MMP-8 monoclonal antibody on the surface of the gold nanoparticle layer; IL-1β (hydrated diameter approximately 5 nm) passed through the outer PEG-DA hydrogel membrane and was retained by the middle chitosan-sodium alginate composite membrane (pore size 2-4 nm), and bound to the 5'-NH2-C6 modified IL-1β nucleic acid aptamer on the surface of the middle PEDOT:PSS layer activated and fixed by EDC / NHS; small molecule metabolites such as lactic acid and uric acid reached the bottom Prussian blue catalytic layer through the three selectively permeable membranes and were catalyzed by lactate oxidase / uric acid oxidase to generate H2O2. In this single incubation, MMP-8 and IL-1β simultaneously completed size classification and interface recognition on the top gold nanoparticle layer and the middle PEDOT:PSS layer, respectively.

[0096] The movement of biomarkers across the multilayer membrane structure is primarily driven by concentration gradient-driven passive diffusion (Fickian diffusion), rather than gravity or external liquid-driven forces. After sample addition, the target molecules (MMP-8, IL-1β, and small molecule metabolites) have higher concentrations on the sample side and zero initial concentrations on the inner side of the membrane, creating a significant concentration gradient that drives the molecules to diffuse inward. Since the thickness of each selectively permeable membrane is only on the order of 100-200 nm, the diffusion coefficient of small molecule metabolites (lactic acid MW 90 Da, uric acid MW 168 Da) in the hydrogel and polymer films is approximately 10. -11 Up to 10 -10 m 2The characteristic diffusion time across each monolayer membrane is on the order of milliseconds, and the time required to traverse all three membranes (total thickness approximately 600-1200 nm) is on the order of several milliseconds to sub-seconds, far less than the incubation time. This ensures that each marker completes size-grading retention and enzyme substrate supply. Both the PEG-DA hydrogel membrane and the chitosan-sodium alginate composite membrane are hydrophilic porous materials. After absorbing water and swelling, they can assist in the introduction of sample liquid through capillary wetting effect, further promoting diffusion and mass transfer. The cellulose acetate membrane (CA membrane) operates primarily as an inner size barrier and Prussian blue electrodeposition substrate in the multilayer structure, preventing residual macromolecular impurities from contacting the conductive carbon membrane. Gravity only plays an auxiliary role in the initial spreading of the sample liquid on the sensor's horizontal surface and does not play a dominant role in the selective molecular transport within the nanoporous membrane.

[0097] Step 4: Simultaneous EIS detection of outer MMP-8 and intermediate IL-1β. After simultaneous incubation, the sensor surface was gently rinsed with PBS buffer to remove non-specific adsorbed substances. Since MMP-8 and IL-1β were already captured by the top gold / antibody interface and the intermediate PEDOT:PSS / aptamer interface, respectively, gentle PBS washing would not significantly remove the bound target molecules. The sensor was connected to an electrochemical workstation, using an Ag / AgCl electrode as the reference electrode and a Pt ring electrode as the counter electrode, in an environment containing a redox probe (e.g., 5 mmol / L [Fe(CN)6]). 3- / 4-In a supporting electrolyte of 0.1 mol / L KCl, an AC perturbation signal with an amplitude of 5 mV was applied at the open circuit potential, and an electrochemical impedance spectroscopy (EIS) scan was performed in the frequency range of 0.1 Hz to 100 kHz. Because the top gold nanoparticle layer / antibody-MMP-8 interface and the middle PEDOT:PSS / aptamer-IL-1β interface are composed of different electrode materials and different capture chemistry (antibody-antigen vs. 5'-NH2-C6 modified aptamer non-covalently bound to IL-1β after amide bond fixation), and their spatial positions from the working electrode / electrolyte interface are different, the charge transfer resistance (Rct) and interface capacitance of the two interfaces exhibit distinctly different RC time constants (the top gold nanoparticle-antibody interface usually shows a lower Rct and a higher interface capacitance, corresponding to a shorter time constant; the middle PEDOT:PSS-aptamer interface is affected by the inner chitosan-sodium alginate composite film and the dielectric properties of the PEDOT:PSS layer, exhibiting a relatively longer time constant); by analyzing the equivalent circuits corresponding to the Nyquist and Bode diagrams (which can be represented by R_s + [R_ct1 ‖ CPE1] + [R_ct2 ‖ CPE2] + ... By performing nonlinear least-squares fitting (using series and parallel combinations of W, etc.), the R_ct and CPE parameters of the two interfaces can be analytically separated, yielding the MMP-8 detection signal ΔRct1 and the IL-1β detection signal ΔRct2, respectively. If no redox probe is added to the supporting electrolyte, the above parameters should be described as changes in interfacial impedance or interfacial capacitance, while the underlying mechanism remains unchanged.

[0098] Step 5: Detection of the total signal of inner layer small molecule metabolites (amperometric method). The working electrode potential is set to -0.05 V to -0.15 V (typically -0.1 V, vs. Ag / AgCl), and the steady-state current is recorded using a chronoamperometric method. Lactic acid is catalyzed by lactate oxidase, and uric acid is catalyzed by uricase oxidase, both producing H2O2. H2O2 is reduced on the Prussian blue catalytic layer surface at the same working potential. Therefore, the recorded reduction current is the total signal of H2O2 produced by the two small molecule oxidase substrates, lactate and uric acid. This step outputs this as the total signal of the inner layer small molecule oxidase substrate, without distinguishing the individual contributions of lactate and uric acid. For separate quantification of lactate or uric acid, other methods are required (e.g., adding parallel working electrodes with only a single oxidase fixed on the same sensing platform, using the standard addition method, or deconvolution based on the kinetic differences / specific inhibitors of the two oxidases). The amperometric signal obtained by the single working electrode structure of the sensor described in this embodiment only gives the sum of the two.

[0099] Step 6: Data Processing and Analysis. The outer MMP-8 detection signal ΔRct1, the middle IL-1β detection signal ΔRct2, and the total signal of the inner small molecule oxidase substrate (steady-state reduction current of lactate and uric acid combined) were substituted into a pre-established standard curve to calculate the concentrations of MMP-8, IL-1β, and the equivalent values ​​of the total concentrations of lactate and uric acid. Based on these indicators, the inflammatory status of periodontitis was assessed using a multi-marker combined analysis model.

[0100] Example 3: Simulation Verification and Predictive Detection Performance Data

[0101] Before obtaining clinically tested samples, a finite element mass transfer-electrochemical coupling model and an equivalent circuit fitting model were established to verify the feasibility of this invention. The simulation conditions were: PEG-DA number-average molecular weight M... n =4000, PEG-DA mass fraction 5 wt%, glycerol mass fraction 10 wt%, hydrogel swelling water content approximately 90 wt%, cellulose acetate membrane molecular weight cutoff approximately 1000 Da. Solute retention simulation showed that the equivalent pore size of the PEG-DA hydrogel membrane in the swollen state was 6.4 ± 0.5 nm, the retention rate of MMP-8 (hydration diameter approximately 8 nm) within 10 minutes was 91.8%, the penetration rate of IL-1β (hydration diameter approximately 5 nm) was 78.6%, and the penetration rates of lactic acid and uric acid were both greater than 95%. Transmembrane electrodeposition simulations showed that after the Prussian blue precursor penetrated the CA membrane, it first nucleated at the interface adjacent to the conductive carbon membrane / CA membrane. After 20 cycles of cyclic voltammetry, approximately 72% of the Prussian blue deposits were located in the CA membrane pores and the adjacent 80-120 nm region, maintaining continuous electronic connectivity with the conductive carbon membrane. After the chloroauric acid precursor penetrated the PEG-DA hydrogel membrane, it nucleated at the interface adjacent to the PEDOT:PSS / PEG-DA membrane. After 150 seconds of constant potential deposition, approximately 86% of the gold nanoparticle clusters maintained electronic connectivity with the PEDOT:PSS through the hydrogel pores, with the exposed particles on the outer side potentially usable for MMP-8 antibody immobilization. EIS separation simulations used an equivalent circuit of R_s+[R_ct1||CPE1]+[R_ct2||CPE2]+W for nonlinear least-squares fitting. In the blank sample, R_ct1 was 1.24±0.08 kΩ and R_ct2 was 2.86±0.15 kΩ; when only 10 ng / mL MMP-8 was added, ΔR_ct1 was 0.92±0.07 kΩ and ΔR_ct2 was 0.06±0.04 kΩ; when only 100 pg / mL IL-1β was added, ΔR_ct1 was 0.04±0.03 kΩ and ΔR_ct2 was 0.73±0.06 kΩ; when both were added together, ΔR_ct1 was 0.89±0.08 kΩ and ΔR_ct2 was 0.70±0.07 kΩ. The fitted χ² value... 2 Less than 2.5 × 10 -3This indicates that the parameters at the two interfaces can be stably separated with low cross-interference. Based on the simulation standard curves established using artificial gingival crevicular fluid matrix parameters, the predicted linear range for MMP-8 is 0.1–100 ng / mL, with a detection limit of 0.05 ng / mL; the linear range for IL-1β is 1–1000 pg / mL, with a detection limit of 0.4 pg / mL; and the linear range of the total signal of lactate and uric acid small molecule oxidase substrates, expressed as H2O2 equivalent concentration, is 1–1000 μmol / L, with a detection limit of 0.35 μmol / L. When 10-fold interference simulations were performed on TNF-α, IL-6, albumin, glucose, and ascorbic acid, the signal deviation for each channel was less than 8%; the simulated spiked recoveries ranged from 92.4% to 107.6%, and the repeatability RSD was no higher than 6.5% (n=5).

Claims

1. An electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid based on size stratification and differential configuration, characterized in that... ,include: A flexible substrate with a conductive layer on its surface, and a working electrode, a reference electrode, and a counter electrode disposed on the flexible substrate, wherein the counter electrode surrounds the working electrode in a ring or semi-ring shape; The working electrode is composed of a three-level signal response layer and a selectively permeable membrane; The three-level signal response layer consists of a bottom layer, a middle layer, and a top layer, from the inside out. The bottom, middle, and top layers use oxidases, nucleic acid aptamers, and monoclonal antibodies as recognition molecules, respectively. The bottom layer is detected using amperometrics, while the middle and top layers are detected using electrochemical impedance spectroscopy. Differential signal transduction is achieved by analyzing the signals at the two interfaces through equivalent circuit fitting. The bottom response layer is a Prussian blue catalytic layer with lactate oxidase and uricase immobilized on its surface; the middle response layer is a PEDOT:PSS conductive polymer layer with IL-1β nucleic acid aptamers immobilized on its surface; the top response layer is an electrodeposited gold nanoparticle layer with MMP-8 monoclonal antibody directly and directionally immobilized on its surface via gold-sulfur bonds. The selectively permeable membrane comprises three layers: an outer layer and a middle layer. The outer layer is a modified PEG-DA hydrogel membrane, disposed inside the top response layer. The PEG-DA hydrogel membrane is formed by cross-linking PEG-DA, glycerol, and a photoinitiator. After being fully swollen with PBS buffer, it has a pore size of 6-7 nm, which preferentially retains MMP-8 and facilitates the mass transfer of IL-1β to the middle layer. The middle layer is a chitosan-sodium alginate composite membrane with a pore size of 2-4 nm, disposed inside the middle response layer, which prevents IL-1β from penetrating to the bottom response layer, allowing IL-1β to remain outside the middle response layer to bind to nucleic acid aptamers. The inner layer is a cellulose acetate membrane, closely attached to the conductive layer and located inside the bottom response layer, which serves as the innermost size barrier and transmembrane electrodeposition substrate, preventing residual macromolecular impurities from contacting the conductive layer.

2. The electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid based on size stratification and differential configuration as described in claim 1, characterized in that... The flexible substrate has a thickness of 5-20 μm; the flexible substrate material is a parylene film. The surface of the Prussian blue catalytic layer, with a thickness of 20-100 nm, is immobilized with lactate oxidase and uricate oxidase. This layer is used to detect the total redox current signal of small molecule metabolites on the surface of the Prussian blue catalytic layer by amperometric method. The surface is immobilized with a PEDOT:PSS conductive polymer layer containing IL-1β nucleic acid aptamers, with a thickness of 20-100 nm. The IL-1β nucleic acid aptamer is a 5'-NH2-C6 modified aptamer. The carboxyl group introduced into the PEDOT:PSS layer, after activation by EDC / NHS, reacts with the amino group of the 5'-NH2-C6 modified aptamer to form an amide bond, thereby covalently and orienting the aptamer onto the PEDOT:PSS surface. The intermediate response layer is used for the electrochemical impedance spectroscopy detection of IL-1β. The surface is directly and directionally immobilized with an electrodeposited gold nanoparticle layer of MMP-8 monoclonal antibody via gold-sulfur bonds, with a thickness of 50-200 nm, for the electrochemical impedance spectroscopy detection of MMP-8; The electrodeposited gold nanoparticle layer has a porous stacked structure with interparticle gaps larger than the size of IL-1β and small molecule metabolites. It serves as a response and capture interface for MMP-8 rather than a size sieving barrier, allowing IL-1β and small molecule metabolites to transfer mass to the inner layer through the interparticle gaps.

3. The electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid based on size stratification and differential configuration as described in claim 1, characterized in that... The reference electrode is an Ag / AgCl reference electrode; the counter electrode is a platinum counter electrode; the reference electrode is located at the geometric center of the annular or semi-annular region enclosed by the counter electrode.

4. The electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid based on size stratification and differential configuration as described in claim 1, characterized in that... It also includes a removable disposable sample pad that covers the working electrode and is composed of a hydrophilic low-protein adsorption coarse filter membrane with a pore size of 100-200 nm. The coarse filter membrane is a track-etched polycarbonate membrane or a polyethersulfone membrane. The sample pad is used to coarsely filter the gingival crevicular fluid during sample loading, retaining cells, bacteria, and debris larger than the target protein, while allowing MMP-8, IL-1β, and small molecule metabolites to pass through. The sample pad is removed before electrochemical detection, allowing the working electrode to come into direct contact with the filtered sample.

5. The method for preparing an electrochemical sensor for detecting multiple biomarkers in gingival crevicular fluid based on size stratification and differential configuration as described in claim 1, characterized in that... The following steps are included: S1: A mixed precursor of graphene oxide and polyamic acid is carbonized at high temperature on a silicon substrate to form a conductive carbon film as a conductive layer; S2: A flexible substrate of parylene with a thickness of 10-15 μm is formed on the surface of the conductive carbon film by chemical vapor deposition, and the conductive carbon film is transferred to the flexible substrate by flipping and peeling. S3: The working electrode, reference electrode, and counter electrode patterns are formed on a conductive carbon film using photolithography. S4: A silver layer is electrodeposited in the reference electrode pattern area and then chlorinated to form an Ag / AgCl reference electrode; a platinum layer is electrodeposited or sputtered in the counter electrode pattern area to form a platinum counter electrode; a conductive carbon film is retained in the working electrode pattern area as a substrate for subsequent multilayer response structures. S5: On the patterned conductive carbon film surface of the working electrode, a 2 wt% cellulose acetate acetone solution is first drop-coated and dried at room temperature to form a cellulose acetate film, which is then placed close to the conductive carbon film as the innermost size barrier and transmembrane electrodeposition substrate. Subsequently, potassium ferricyanide and iron salt precursors are allowed to penetrate the pores of the cellulose acetate film, nucleate at the interface near the conductive carbon film / cellulose acetate film, and grow in situ along the membrane pores to form a Prussian blue catalytic layer that maintains electronic communication with the conductive carbon film and is at least partially exposed to the outside of the cellulose acetate film as the bottom response material. Lactate oxidase and urate oxidase are then covalently fixed to the surface of the Prussian blue catalytic layer using a glutaraldehyde crosslinking method. S6: On the outer surface of the bottom Prussian blue catalyst layer, chitosan solution and sodium alginate solution were first coated sequentially. After cross-linking with polyelectrolytes and washing to remove uncross-linked components, a chitosan-sodium alginate composite membrane with a pore size of 2-4 nm was formed. Subsequently, a PEDOT:PSS layer containing carboxyl functional groups was formed on the outer surface of the chitosan-sodium alginate composite membrane by constant potential electropolymerization with a mixed solution of EDOT and PSS. The carboxyl groups on the surface of the PEDOT:PSS layer were activated with EDC / NHS, and then a 5'-NH2-C6 modified IL-1β nucleic acid aptamer solution was added dropwise. The aptamer was covalently and directionally fixed to the PEDOT:PSS surface through amide bond formation. Unbound aptamer molecules were removed by washing with PBS buffer. S7: First, a PEG-DA prepolymer solution is coated onto the surface of the intermediate layer PEDOT:PSS, wherein the PEG-DA prepolymer solution contains 5 wt% number-average molecular weight M n A modified PEG-DA hydrogel membrane was formed by UV crosslinking of 4000 PEG-DA, 10 wt% glycerol, and 0.02-0.1 wt% photoinitiator. The pore size of the hydrogel membrane after swelling with PBS buffer was 6-7 nm. Subsequently, a chloroauric acid precursor was allowed to penetrate the pores of the modified PEG-DA hydrogel membrane, nucleating at the conductive interface near PEDOT:PSS / PEG-DA and growing in situ along the hydrogel pores to form a gold nanoparticle layer that is electronically connected to PEDOT:PSS and at least partially exposed to the outside of the hydrogel. The MMP-8 monoclonal antibody was directly and directionally immobilized on the surface of the gold nanoparticles through gold-sulfur bonds. The gold nanoparticle layer serves as the outermost layer of the sensor and is in direct contact with the sample. S8: Encapsulation and Electrical Connection: The flexible substrate with multilayer film modification is encapsulated. A microfluidic detection cell is defined in the non-detection area using UV insulating adhesive. The end of the conductive layer is then heat-pressed to the flexible PCB using anisotropic conductive adhesive (ACF) to complete the electrical connection and sealed storage.

6. The preparation method according to claim 5, characterized in that... In S1, the graphene oxide and polyamic acid are mixed at a mass ratio of 0.5-0.7:1, coated on a temporary substrate, and carbonized in a nitrogen atmosphere at 950-1050℃ for 1-2 hours to form a carbon film.

7. The preparation method according to claim 5, characterized in that... The electrodeposition of the Prussian blue catalyst layer described in S5 specifically includes: depositing the Prussian blue catalyst layer for 15-25 cycles in the range of -0.2 V to +0.6 V using a cyclic voltammetry method.

8. The preparation method according to claim 5, characterized in that... The constant potential electropolymerization described in S6 specifically includes: electropolymerization at a constant potential of +0.85 V to +0.95 V for 120-180 seconds in an aqueous solution containing EDOT and PSS; the EDC / NHS activation specifically includes: activating the carboxyl groups on the PEDOT:PSS surface with a mixed aqueous solution of 50-200 mmol / L EDC and 10-50 mmol / L NHS at room temperature for 15-30 minutes, followed by dropwise addition of a 0.5-2 μmol / L solution of 5'-NH2-C6 modified IL-1β nucleic acid aptamer, reacting at room temperature for 30-60 minutes, and covalently and directionally immobilizing the aptamer on the PEDOT:PSS surface through EDC / NHS-mediated amide bond chemistry; In the aqueous solution containing EDOT and PSS, the PSS is a carboxylated PSS component, or EDOT-COOH is added to the aqueous solution as a copolymer component; the EDOT-COOH is a carboxylated EDOT monomer, or is prepared by oxidative carboxylation of hydroxymethyl EDOT.

9. The preparation method according to claim 5, characterized in that... The electrodeposited gold nanoparticle layer described in S7 specifically comprises: electrodepositing with a 4-6 mmol / L chloroauric acid (HAuCl4) solution at a constant potential of -0.15 V to -0.25 V for 100-180 seconds, allowing the chloroauric acid precursor to penetrate the pores of the modified PEG-DA hydrogel membrane and nucleate and grow along the PEDOT:PSS / PEG-DA adjacent interface and the hydrogel pores; the photoinitiator content in the prepolymer solution is 0.02-0.1 wt%.

10. A method for detecting gingival crevicular fluid, utilizing the electrochemical sensor for detecting multiple biomarkers of gingival crevicular fluid based on size stratification and differential configuration as described in claim 1, characterized in that... This includes the following steps: S101: Obtain a filter paper strip adsorbed with gingival crevicular fluid, immerse the filter paper strip in a centrifuge tube pre-filled with PBS buffer, vortex for 3-10 minutes to release proteins and small molecule metabolites into the buffer, centrifuge to remove fiber debris, and take the supernatant as the elution solution to be tested; S102: Before detection, drop PBS buffer onto the sensor surface and let it stand for 60-300 seconds to allow the outer modified PEG-DA hydrogel membrane to fully absorb water and swell to restore the pore size to 6-7 nm. Then, remove the residual buffer in the detection cell. S103: Add the eluent to be tested onto the sensor surface and incubate at room temperature for 3-15 minutes. This allows MMP-8 in the sample to be retained by the outer PEG-DA hydrogel membrane and bind to the monoclonal antibody on the surface of the top gold nanoparticle layer. IL-1β in the sample passes through the outer PEG-DA hydrogel membrane, is retained by the middle chitosan-sodium alginate composite membrane, and binds to the nucleic acid aptamers on the surface of the middle PEDOT:PSS layer. Small molecule metabolites lactic acid and uric acid in the sample reach the bottom Prussian blue catalytic layer through the three selectively permeable membranes, simultaneously completing size grading and interface recognition in the same incubation. S104: After incubation, the sensor surface was cleaned with PBS buffer to remove non-specific adsorbed substances. Then, using an Ag / AgCl electrode as the reference electrode and a platinum electrode as the counter electrode, an AC perturbation with an amplitude of 5 mV was applied at an open-circuit potential in a supporting electrolyte containing redox probes. An electrochemical impedance spectroscopy (EIS) scan was performed in the frequency range of 0.1 Hz to 100 kHz. Due to significant differences in surface materials, capture chemistry, and distance from the electrolyte between the top gold / antibody-MMP-8 interface and the middle PEDOT:PSS / aptamer-IL-1β interface, the two interfaces exhibit resolvable different charge transfer resistances and capacitance time constants. The RC parameters corresponding to the two interfaces were analytically separated through equivalent circuit fitting to obtain the MMP-8 detection signal and the IL-1β detection signal, respectively. S105: Set the working electrode potential to -0.05 V to -0.15 V, and record the steady-state current using the chronoamperometry method; lactic acid is catalyzed by lactate oxidase, and uric acid is catalyzed by uricase to produce H2O. 2， The reduction of H2O2 on the surface of the Prussian blue catalytic layer at the same potential generates a current. Therefore, this current is the total reduction current of H2O2 produced by the oxidase catalysis of lactic acid and uric acid, which serves as the total signal of the substrate of the inner small molecule oxidase.

11. The method for detecting gingival crevicular fluid as described in claim 10, characterized in that... The detection method is used to simultaneously detect four markers in gingival crevicular fluid: MMP-8, lactic acid, uric acid, and IL-1β.

12. The method for detecting gingival crevicular fluid as described in claim 10, characterized in that... The sensor used also includes a removable disposable sample pad; in S103, the process of adding the eluent to be tested to the sensor surface specifically includes: first, adding the eluent to be tested to the removable disposable sample pad covering the working electrode, so that the eluent to be tested is coarsely filtered through the disposable sample pad, removing cells, bacteria and debris impurities larger than the target protein, and then entering and contacting the top gold nanoparticle layer; after the sample is added, the disposable sample pad is removed.

Images